UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Candidate gene and high throughput genetic analysis of habituation in Caenorhabditis elegans Giles, Andrew Christopher 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2012_spring_giles_andrew.pdf [ 5.38MB ]
Metadata
JSON: 24-1.0072619.json
JSON-LD: 24-1.0072619-ld.json
RDF/XML (Pretty): 24-1.0072619-rdf.xml
RDF/JSON: 24-1.0072619-rdf.json
Turtle: 24-1.0072619-turtle.txt
N-Triples: 24-1.0072619-rdf-ntriples.txt
Original Record: 24-1.0072619-source.json
Full Text
24-1.0072619-fulltext.txt
Citation
24-1.0072619.ris

Full Text

CANDIDATE GENE AND HIGH THROUGHPUT GENETIC ANALYSIS OF HABITUATION IN CAENORHABDITIS ELEGANS  by Andrew Christopher Giles B.Sc.H., Queen’s University, 2004  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY  in  The Faculty of Graduate Studies (Neuroscience)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  March 2012  © Andrew Christopher Giles, 2012  Abstract  The goal of my dissertation was to identify genes that are important for habituation (a decrease in response to a repeated stimulus) with the hope of bringing us closer to understanding the cellular and molecular mechanism that mediates this basic form of learning. To accomplish this I studied habituation of the tap withdrawal response in Caenorhabditis elegans; an organism with a tractable nervous system, well characterized habituation and availability of genetic tools and resources that make it easy to investigate the mechanisms of behaviour. Two approaches were taken. The first was a candidate gene approach where I investigated mutations in genes important for dopamine neurotransmission. A previous study showed that dopamine deficient and dopamine receptor mutants have abnormal habituation and the dopamine receptor is expressed within the tap sensory neurons. Investigating this effect more closely, I found that short-term tap habituation in C. elegans was dependent on the presence of E. coli (their food) and that this food-dependent modulation of habituation was dopamine dependent. The second approach involved characterizing habituation of a large set of C. elegans strains with known mutations in genes predicted to function in the nervous system. Many of these mutants had not previously been characterized. To accomplish this task, it was necessary to improve the speed and detail with which habituation can be assayed. In collaboration with the Kerr Lab at Janelia Farm Research Campus, we developed a high throughput C. elegans behavioural tracking system called the Multi-Worm Tracker. Using this tracking system, I examined many mutants and discovered hundreds of novel phenotypic variants for habituation in C. elegans. The genes affected by these mutations can now be investigated in more detail in order to identify the role that they play in the molecular and cellular mechanism of habituation.  ii  Preface  The ‘Introduction’ chapter contains some parts that have been adapted with permission from articles that were first published in International Review of Neurobiology (Giles et al., 2006)1 and Neurobiology of Learning and Memory (Giles and Rankin, 2009)2. Many of these parts have been rewritten to fit into the scope of my dissertation. I was the primary author of these publications. Chapter 2 is adapted with permission from a research article first published in Neuron as a collaboration between the Rankin Lab of the University of British Columbia, Canada and the Schafer Lab first at University of California San Diego, USA and then at Cambridge University, United Kingdom (Kindt et al., 2007)3. My contribution to the work included the food-dependent nature of the dopaminemediated effect on habituation and the dat-1 mutant behavioural phenotype. I made the initial discovery of these two findings (with the help of Jean Hsu and Lee Lau, who helped run experiments) and then helped design supporting experiments together with Katie Kindt and Kathleen Quast. The behavioural experiments were conducted by me, Katie Kindt, Kathleen Quast, Dan Hendrey and Ian Nicastro; Katie and Kathleen conducted all imaging experiments. Statistical analyses were performed with the aid of Subhajyoti De. These experiments will be presented in the results section of chapter 2. I would like to be clear that Katie Kindt had already discovered that dopamine affects habituation and the calcium response decrement before I discovered the food-dependent effect, even though the order that they are presented in my dissertation may not be suggestive of this. I also contributed ideas and feedback for experiments throughout the original article, however, to a lesser capacity; hence, the rest of the results are described when necessary in the discussion section of chapter 2. Katie Kindt wrote the 1  “Reprinted from International Review of Neurobiology, Vol. 69, Giles AC, Rose JK, Rankin CH, Investigations of learning and memory in Caenorhabditis elegans, pp. 37-71, Copyright (2006), with permission from Elsevier.” 2 “Reprinted from Neurobiology of Learning and Memory, Vol. 92, Giles AC, Rankin CH, Behavioral and genetic characterization of habituation using Caenorhabditis elegans, pp. 139-46, Copyright (2009), with permission from Elsevier.” 3 “Reprinted from Neuron, Vol. 55, Kindt KS, Quast KB, Giles AC, De S, Hendrey D, Nicastro I, Rankin CH, Schafer WR, Dopamine mediates context-dependent modulation of sensory plasticity in C. elegans, pp. 662-76, Copyright (2007), with permission from Elsevier.”  iii  majority of the first draft of the original publication. I wrote a few paragraphs and edited the first draft. Further drafts were edited by the entire author list. I have re-written and/or re-structured the content in order to fit it into the scope of my dissertation; however, any sections similar or identical to the original publication should be credited as described above. Chapter 3 is adapted with permission from a research article that was first published in Nature Methods (Swierczek et al., 2011)4. The article was the result of collaboration between the Rankin Lab of the University of British Columbia, Canada and the Kerr Lab at Janelia Farm Research Campus, Howard Hughes Medical Institute, Ashburn, USA. My contribution to the work included testing and debugging early prototypes of the Multi-Worm Tracker, helping to design and conduct experiments for validation, and designing, supervising (the experiments were run by an undergraduate student, Po Liu under my direction) and analyzing the small scale tap habituation screen. The engineering of the Multi-Worm Tracker was conducted by Nicholas Swierczek and Rex Kerr. Rex Kerr also designed and conducted some experiments and performed statistical analysis. Rex wrote the first draft of the original manuscript except for the sections concerning the tap habituation screen, which I initially drafted. I edited the first draft, with Rex editing the portions that I had contributed. Further drafts were edited by the entire author list. Much of this chapter is a reproduction of the published article; however, some parts have been re-written and/or re-structured in order to fit it into the scope of my dissertation (particularly the Introduction and Discussion). Any sections similar or identical to the original publication should be credited as described above. Chapter 4 was also conducted in collaboration between the Rankin Lab of the University of British Columbia, Canada and the Kerr Lab at Janelia Farm Research Campus, Howard Hughes Medical Institute, USA. My contribution to the work involved designing, conducting and analyzing all experiments. Nicholas Swierczek and Rex Kerr provided technical support for the Multi-Worm Tracker. Rex Kerr aided  4  “Reprinted from Nature Methods, Vol. 8, Swierczek NA, Giles AC, Rankin CH, Kerr RA, High-throughput behavioral analysis in C. elegans, pp. 592-8, Copyright (2011), with permission from Nature Publishing Group.”  iv  in designing and analyzing some experiments. Rex Kerr and Kasper Podgorski aided in data analysis. I have written all parts of this chapter with feedback from those mentioned above as well as Kurt Haas and Catharine Rankin.  v  Table of contents Abstract ......................................................................................................................................................... ii Preface ......................................................................................................................................................... iii Table of contents.......................................................................................................................................... vi List of tables ................................................................................................................................................. ix List of figures ................................................................................................................................................. x Acknowledgements ..................................................................................................................................... xii Dedication ................................................................................................................................................. xvii 1. Introduction ..............................................................................................................................................1 1.1. Background ........................................................................................................................................2 1.1.1. Short-term habituation in model organisms...............................................................................8 1.1.2. Habituation in C. elegans ..........................................................................................................29 1.1.3. Summary of mechanism for habituation ..................................................................................41 1.2. Objectives .........................................................................................................................................48 2. Role of dopamine in habituation ............................................................................................................49 2.1. Introduction .....................................................................................................................................49 2.2. Results ..............................................................................................................................................51 2.2.1. Food-dependent modulation of habituation is mediated by dopamine ..................................51 2.2.2. Modulation of a neural correlate of habituation by dopamine is food dependent .................62 2.3. Discussion .........................................................................................................................................66 2.3.1. Downstream signalling of the DOP-1 receptor .........................................................................66 2.3.2. The presence of food may activate dopamine neurons through mechanosensory receptors.68 2.3.3. Dopamine neurons may act in a positive feedback loop with mechanosensory neurons of the tap circuit ............................................................................................................................................68 2.3.4. Implications for dopamine release............................................................................................69 2.3.5. Novel molecular and cellular mechanism for habituation ........................................................70 2.4. Methods ...........................................................................................................................................71 2.4.1. C. elegans strains and genotypic analysis .................................................................................71 2.4.2. Habituation assay ......................................................................................................................72 2.4.3. In vivo calcium imaging .............................................................................................................73 2.4.4. Statistical analysis......................................................................................................................74 vi  3. Development and validation of the Multi-Worm Tracker ......................................................................76 3.1. Introduction .....................................................................................................................................76 3.2. Results ..............................................................................................................................................78 3.2.1. Analysis of locomotion ..............................................................................................................78 3.2.2. Tap habituation .........................................................................................................................81 3.3. Discussion .........................................................................................................................................88 3.4. Methods ...........................................................................................................................................90 3.4.1. Strains ........................................................................................................................................90 3.4.2. Apparatus and image acquisition of behavioural recordings ...................................................90 3.4.3. Movement experiments ............................................................................................................91 3.4.4. Tap habituation experiments ....................................................................................................92 4. Characterization of a nervous system-biased mutant library .................................................................94 4.1. Introduction .....................................................................................................................................94 4.2. Results ..............................................................................................................................................95 4.2.1. Phenotypic characterization .....................................................................................................96 4.2.2. Correlations .............................................................................................................................116 4.2.3. Covariance within wild-type phenotypic profiles enhances better distinction ......................126 4.2.4. Confirmation of novel habituation genes ...............................................................................130 4.2.5. Phenotypic profiles predict genetic interactions ....................................................................131 4.3. Discussion .......................................................................................................................................139 4.4. Supplemental information .............................................................................................................143 4.5. Methods .........................................................................................................................................145 4.5.1. C. elegans strains.....................................................................................................................145 4.5.2. C. elegans maintenance ..........................................................................................................146 4.5.3. Behavioural testing .................................................................................................................147 4.5.4. Apparatus and image acquisition of behavioural recordings .................................................148 4.5.5. Data analysis............................................................................................................................149 5. General discussion ................................................................................................................................153 5.1. Parametrics of habituation of the tap withdrawal response in C. elegans....................................157 5.1.1. Parameters that had little or no effect on habituation...........................................................157 5.1.2. Habituation parameters ..........................................................................................................158 vii  5.2. Importance of not trading off detailed behavioural analysis for high throughput ........................162 5.3. Contributions to habituation..........................................................................................................164 5.4. Future directions ............................................................................................................................168 References.................................................................................................................................................170 Appendix ...................................................................................................................................................190  viii  List of tables  Table 1.1 Summary of cellular mechanisms of habituation ........................................................................43 Table 1.2 Summary of molecular components for the mechanism of habituation ....................................44 Table 3.1 Tap habituation screen ................................................................................................................86 Table 4.1 Correlation coefficients .............................................................................................................118 Table 4.2 Transformation equations to normalize mutant distributions .................................................152  ix  List of figures  Figure 1.1 Neural circuit of the tap withdrawal response ..........................................................................35 Figure 1.2 Anatomy of tap withdrawal response with molecules implicated in habituation .....................37 Figure 2.1 Manual Assay: dopamine deficient effect on habituation is not apparent in the absence of E. coli ...............................................................................................................................................................52 Figure 2.2 Single-Worm Tracker: habituation is dependent on the presence of E. coli and is dopamine receptor dependent ....................................................................................................................................56 Figure 2.3 Multi-Worm Tracker: habituation is dependent on the presence of E. coli and is dopamine receptor dependent ....................................................................................................................................58 Figure 2.4 Dopamine transport mutation slows habituation in the absence of food ................................60 Figure 2.5 Modulation by dopamine of a neural correlate of habituation in sensory neurons is dependent on the presence of food ..............................................................................................................................63 Figure 3.1 Locomotion ................................................................................................................................79 Figure 3.2 Analysis of tap habituation.........................................................................................................83 Figure 3.3 Tap habituation screen ..............................................................................................................84 Figure 4.1 Distribution and summary of phenotypes. ................................................................................97 Figure 4.2 Novel mutants with midline length and area of body size phenotypes ..................................100 Figure 4.3 Replication of mutants with body size phenotypes .................................................................101 Figure 4.4 Mutants with spontaneous reversal phenotypes ....................................................................104 Figure 4.5 Speed-related spontaneous locomotion phenotypes..............................................................106 Figure 4.6 Novel mutants with spontaneous locomotion phenotypes.....................................................107 Figure 4.7 Mutant with rapid deceleration during spontaneous locomotion ..........................................109 Figure 4.8 Initial response and habituation of tap-induced reversal probability .....................................110 Figure 4.9 Initial response and habituation of tap-induced reversal distance .........................................111 Figure 4.10 Initial response and habituation of tap-induced reversal duration .......................................112 Figure 4.11 Replication of mechanosensory and tap habituation mutants..............................................114 Figure 4.12 Habituation of eat-4 mutants ................................................................................................117 Figure 4.13 Covariance of phenotypes......................................................................................................122 Figure 4.14 Effect of inter-stimulus interval on habituation.....................................................................125 Figure 4.15 Examples of phenotypic profiles ............................................................................................127  x  Figure 4.16 Mahalanobis distances to the mean of wild-type distribution in 14 dimensional phenotypic space .........................................................................................................................................................129 Figure 4.17 Habituation of various eat-16 and goa-1 mutants ................................................................132 Figure 4.18 Average-linkage hierarchical clustering dendrogram ............................................................134 Figure 4.19 Distribution of Mahalanobis distances between wild-type replicates and mutant strains ...137 Figure 4.20 t-SNE clustering highlights 6 interactions supported by other studies..................................138 Figure 5.1 Putative molecular components involved in habituation of the tap withdrawal response in C. elegans ......................................................................................................................................................154  xi  Acknowledgements I have been so fortunate to have wonderfully supportive people behind me throughout my entire graduate student career and I would like to take this opportunity to thank them all dearly. Without their help my dissertation would not be the same. The most influential of these people was my supervisor Dr. Catharine Rankin. When I first joined the lab, Cathy provided me with the guidance I needed to learn about the neurobiology of learning and memory. Often travelling to conferences together, I was able to experience the world of science whether this was in Los Angeles, San Diego, Washington DC, Madison, New York or even Tuscany! I have always enjoyed our scientific discussion as well as talks about life. Cathy has always done her best to give me the best career advice and opportunity that she can. As I gained knowledge and skills Cathy gradually gave me the opportunity, resources and flexibility to become a successful and independent scientist, which is something I will cherish for the rest of my life. Cathy, thank you so much! I must thank my undergraduate supervisor Dr. Richard Beninger for introducing me to Cathy. Without his insight I may never have discovered her lab during my search for graduate schools. I would also like to thank Rick for seeding my mind with the thrill of science. Every member of the Rankin Lab had an impact on my experience in graduate school. I have to thank Jackie Rose and Celia Ebrahimi for first introducing me to life as a graduate student in the Rankin Lab. I would particularly like to thank Jackie for giving me tremendous guidance during my first years in the lab and always went to an extra effort to include and introduce me to her friends and contacts throughout the research community. Susan Rai was a fellow incoming graduate student in the Rankin lab who had previously worked in the lab during her undergraduate degrees. She was helpful in my orientation to the lab, UBC, and Vancouver. The other incoming graduate student when I joined the lab was Justin Davis. Together we tackled the challenges of moving across the country from Ontario, living in a new city on graduate student wages, first year neuroscience courses, TAing, working in a C. elegans xii  lab, writing scientific reviews, and going to conferences. It was incredibly helpful to have a friend and colleague to support and share these experiences. Over the next few years in the lab, a number of other students joined the lab that I feel became the core group of students during my tenure in the lab. This included Tiffany Timbers, Mike Butterfield, Conny Lin, Evan Ardiel and Lee Lau. They have become some of my best friends and colleagues. Their help, support, criticism, advice, debates, ideas, feedback and fun helped shape the science and personality behind this dissertation. I would particularly like to acknowledge Tiffany and Evan, with whom I have collaborated, travelled, roomed and shared so many great experiences. Some of our deep scientific, artistic, political, or just plain fun discussions and adventures during days and nights in and out of the lab rank in the top of my life experiences. Tiffany and Evan, thank you so much! Although, my time with newer students in the lab, Ricardo Bortolon, Andrea McEwan and Tahereh Bozorgmehr, has been much shorter I have valued their ideas and feedback as well. There have also been dozens of undergraduate volunteers, directed studies and honours project students, and undergraduate research assistants during my time in the lab and I would like to thank them all particularly those who worked with me directly including Hannah, Dave, Alex, Ravinder, Jean, Jake, Serena, Nadia, Jay, Josh, Andy, Sepehr, Aileen, Helena, Maria, Po and Angela. I had the opportunity to discuss and debate science with many of the other graduate students in the neuroscience program, some of whom I have learned a lot from or benefited from their feedback and discussions. This includes Lasse Dissing-Oleson, Simon Chen, Kasper Podgorski, Ted Dobie, Kevin She, Derek Dunfield, Sesath Hewapathirane, and Leon French. I would particularly like to acknowledge Lasse. After bonding over fire and water, whiskey, music, sports (particularly football (soccer and American) and Olympic curling too ;), politics, traveling and of course a love of science, he has become one of my best friends. Lasse, thank you so much!  xiii  I benefited greatly from having an excellent supervisory committee made up of Peter Reiner (chair), Stan Floresco, Doug Allan and Anthony Phillips. All of whom gave me much needed constructive criticism, valued advice and feedback that challenged me to be a better academic and scientist. A few other faculty members had an impact on my graduate school experience and dissertation. Kurt Haas provided motivation and valued discussion for both technical and theoretical problems. Christian Nass provided me with the opportunity to work on a side project that was not part of my dissertation, but broadened my knowledge and experience. I believe this increased perspective has benefited me in constructing ideas in my dissertation. Don Moerman has been incredibly valuable as another faculty member in the C. elegans community in Vancouver. In his lab, I learned the basics of molecular biology and genetics in C. elegans with particular help from Nicholas Dube, Adam Warner, Ryan Viveros, Jaryn Perkins, Barbara Meissner, Jay Mayden, and Mark Edgley. I would particularly like to acknowledge Nick with whom I have become good friends. Thank you for introducing me to the world of molecular biology, hockey pools, Canucks games and poker. Don was also the external examiner on my comprehensive examination committee and in doing so provided valuable feedback on my research proposal that benefited the project greatly. The C. elegans community in Vancouver, known as VanWorm, is made up of more than ten research labs from institutes across the lower mainland of British Columbia and is incredibly collaborative and supportive to each other. I have learned so much attending and presenting at the monthly VanWorm meetings shaping the research that I conducted for my dissertation. All of the projects in my dissertation have been parts of bigger collaborations. Help from my collaborators has been critical for the success of this dissertation. The first collaboration was with William Schafer’s lab and involved visiting his lab to conduct experiments. I thank Bill for this great opportunity. During this collaboration, I worked very closely with Katie Kindt who was the primary author of the collaboration. I learned a lot from my work with Katie that I applied to the rest of my  xiv  project thereafter. Other members in the lab helped me feel comfortable and welcome and aided me in my research while I was there, including Kathleen, James, Marina, Yoshinori, Ithai, Robyn, and Marios. The other collaboration that contributed to my dissertation was with Rex Kerr’s Lab at Janelia Farm Research Campus. It involved visiting Rex’s lab for many months at a time. While there I learned a phenomenal amount about science from Rex and I will treasure the experience for the rest of my life. Others in or visiting his lab also helped on various levels with the projects that are presented in my dissertation including Nicholas Swierczek, Wafa Amir, Victoria Butler, Sarah Moorehead, and Lori Kurminsky. The research in my dissertation was funded by a PhD tuition award from the University of British Columbia, Post-Graduate Scholarship from the Natural Sciences and Engineering Research Council (NSERC) of Canada, a Junior Research Fellowship from Michael Smith Health Research Foundation, the Janelia Farm Research Campus Visiting Scientist Program. Funding from the National Institute on Drug Abuse to William Schafer, from Janelia Farms Research Campus, Howard Hughes Medical Institute to Rex Kerr, and from NSERC to Catharine Rankin also supported the projects in the dissertation. I am so lucky to have had an amazing set of friends who have supported me over the course of my doctoral research. Of particular note are some of my closest friends: Steve Nace, Cam Gross, Brad Blackwell, Nick Gariepy, Kevin Eugene, Matt Kirkey, Julia Armstrong, Jamey Trewartha, Drew Gaudet, Jake Mossop, and Melissa Nicholson. I would like to take a moment to remember Jared Stanley (19792005), whose memory will always remind me of my first year in Vancouver as a graduate student, a happy, exciting, fun-filled time of my life and of the dangers of adventure in the wilderness. The support that is dearest to my heart is that from my loving partner (or co-person ;), and soon to be wife, Kelsey Lindeman. Kelsey supported my doctoral work in so many ways including living through a number of long distance relationships during the collaborative visits needed to other labs and late all-nighters during experiments and long writing sessions. She even helped me maintain my worms xv  during a visit to Janelia Farm, a task only a completely committed partner would volunteer for! Although not an expert in my field, Kelsey often helped me bounce research ideas around and was always available for practice presentations, brainstorming, questions and discussion. Waiting for me to finish my PhD degree, so that we could leave Vancouver together to travel the world has been an incredible feat of patience, commitment and companionship. Kelsey, your love and support means so much to me. Thank you so much and Vomyka! Finally, the longest, most constant and unwavering support has come from my family. My mom, Josephine, my dad, Richard, and my brother, Stephen provided emotional (and financial, very much appreciated!!) support and have always been an ear for me to discuss life, the universe and everything including my research, which at times was probably an arduous task. My brother has always been enthusiastic and interested in my work and has been there to support me in times of need. I appreciate the naïve but intelligent feedback that they all offered to support this dissertation. They are the best family in the world - Mom, Dad, Steve, I love you guys! Thank you all.  xvi  Dedication  To My Dad  xvii  1. Introduction Organisms need to be able to modify their behaviour in response to experience in order to survive the constantly changing environment in which they live. This process is known as learning. For learning to be advantageous, a process must also exist to store modified behaviours so that during future interaction with the environment the learned behaviour can be re-used. The storage and recall of this information is encompassed by the term memory. Behaviour is controlled by the nervous system, so for an organism to alter its behaviour in response to experience with the environment via learning and memory, the organism must alter its nervous system to allow for these changes. Many organisms, from simple single-cell bacteria to highly complex mammals, have the capacity for both learning and memory. Many of the behavioural changes that occur during learning and memory have been characterized and categorized across this long list of organisms; however, the molecular modifications that take place within the nervous system to mediate the altered behaviour are poorly understood. The goal of this dissertation was to identify genes that play a role in this process in order to gain a better understanding of some of these molecular changes. Learning is an incredibly broad concept ranging from relatively simple changes in response to a single low-dimensional stimulus to complex changes in behavioural repertoires brought about by multifaceted interactions with many high-dimensional stimuli. In my research dissertation, I have focused on a simple but fundamental form of learning known as habituation. Habituation is a decrement in response to a repeated stimulus. I describe it as fundamental for two reasons: 1) habituation is highly conserved throughout the animal kingdom suggesting that it may be critical for survival. 2) Habituation is essential for most other types of learning. Many behavioural protocols developed to study learning begin with habituation sessions to both the experimenter and the experimental environment or apparatus (Deacon, 2006). Nervous systems of animals have a finite amount of resources. For this reason, animals that have been habituated to the irrelevant stimuli of an experimental paradigm prior 1  to testing perform better during learning tasks than animals that have not been habituated (Deacon, 2006). Hence, experimental manipulations that disrupt habituation can indirectly affect other types of learning. For these reasons, understanding more about habituation will have broad implications to the study of learning and neurobiology. Habituation is arguably far less simple than it may first appear, something I hope my dissertation will illustrate; however, it is clearly less complex than many learning problems that exist in nature. I believe studying a simple behaviour will be easier and quicker for the discovery of neural mechanisms and that problems solved doing this will be useful for the investigation of more complex behaviours. Following this same philosophy, my dissertation focuses on habituation in the model organism Caenorhabditis elegans, an animal with a relatively simple nervous system compared to the rest of the animal kingdom. In addition to this, the C. elegans research community has developed many genetic tools that make it easy to perform cellular and molecular manipulations, making it ideal for the study of cellular and molecular mechanisms of habituation.  1.1. Background The first biological description of habituation, although it was simply described as learning at the time, was by George and Elizabeth Peckham during their study of spiders (Peckham and Peckham, 1887). Before this quantitative example, however, it is apparent that habituation had been observed by humans for millennia. Both George Humphrey (Humphrey, 1933) and more recently Richard Thompson (Thompson, 2009) provided an excellent example described in one of AEsop’s Fables, The Fox and the Lion, whose origin is likely before the 5th century BC: “The fox had never seen a lion before, so when she happened to meet the lion for the first time she all but died of fright. The second time she saw him, she was still afraid, but not as much as before. The third time, the fox was bold enough to go right up to the lion and speak to him.” (Gibbs, 2002) 2  In the first scientific description of habituation, the Peckhams found that wild garden spiders reacted to the sound of a tuning fork by dropping out of their webs on a line of silk and then after a delay, climbed back up the silk and relaxed in the centre of their webs. With repeated presentation of this acoustic stimulus, the spiders gradually dropped a smaller distance out of their webs and spent less time before climbing back, until finally, they ignored the tuning fork completely. Since this detailed description of a decrease in response following repeated stimulation, similar examples have been identified and characterized in hundreds of organisms. Unfortunately, most of these reports were scattered across different fields of zoology and hidden in papers discussing various topics, usually not with a primary goal of investigating the response decrement. In a seminal work in 1943, Donald Harris extensively surveyed the literature and collected quantitative examples of response decrements from seven diverse phyla within the animal kingdom, Coelenterata, Echinodermata, Platyhelminthes, Annelida, Mollusca, Anthropoda, and Chordata, as well as Protista, which is no longer considered a phylum within the animal kingdom but a separate kingdom of its own (Harris, 1943). Although this was likely not an exhaustive list, acknowledged by Harris himself, it was an incredible feat and for the first time illustrated the conservation of this phenomena throughout most of the stimulusresponse relationships described in the animal kingdom. When a feature of biology is so broadly shared throughout nature, it can be taken as an indication of its importance for survival. Harris suggested that these response decrements, which had until then been called by many different terms including “negative adaptation,” “acclimatization,” “extinction,” “stimulatory inactivation”, were all a similar phenomenon and suggested a common term, “habituation” (Harris, 1943). Harris was not the first to use this term – it is not clear who the first was (Thompson, 2009) – however, his reasons for choosing it were not arbitrary. Unlike the other options, “Habituation” does not inherently speculate a mechanism for the phenomenon, it was not commonly used to describe other behavioural phenomena, and although not the most commonly used term, it had been used often to describe the phenomena in question and its common definition was remarkably similar to the 3  description of the phenomenon. Since this reasoning was presented, few scientists have deviated from the term “habituation” to describe a response decrement with repeated stimuli; there are a few exceptions that are of importance to this dissertation, but will be addressed when necessary. Discussing the nature of habituation in his concluding paragraphs, Harris believed that it was unlikely that a single mechanism was responsible for every example of habituation that he had found since the organisms that he presented were themselves so diverse, but he suggested that the mechanisms would likely be similar and generalizable. Today, this hope is even more likely with the discovery that the fundamental elements that comprise nervous systems, such as neurotransmitters, receptors, and signalling cascades, are highly conserved throughout the animal kingdom. There are many examples of processes that work similarly between very diverse animals. For example, synaptic vesicle docking required for neurotransmitter release is remarkably conserved from C. elegans to mammals, using homologous proteins, such as SNB-1 (synaptobrevin), unc-64 (syntaxin) and RIC-4 (SNAP-25) (Richmond, 2007). Much enthusiasm was stimulated in the field of learning by Harris’s review and over the next 20 years, many studies were conducted to identify the behavioural characteristics of habituation. Richard Thompson and Alden Spencer distilled these data to describe nine behavioural features of habituation common to all animals (Thompson and Spencer, 1966), in order to distinguish it from other similar response decrements that should not be considered learning, such as sensory adaptation and fatigue. Also, by fully understanding the parameters that control habituation, it is much easier to hypothesize and test putative mechanisms. Recently, these characteristics were reviewed and revised to reflect the new data that has been collected since Thompson’s and Spencer’s original work (Rankin et al., 2009). In doing this a tenth characteristic was added. The ten characteristics that define habituation follow (Rankin et al., 2009):  4  1) “Repeated application of a stimulus results in a progressive decrease in some parameter of a response to an asymptotic level. This change may include decreases in frequency and/or magnitude of the response. In many cases, the decrement is exponential, but it may also be linear; in addition, a response may show facilitation prior to decrementing because of (or presumably derived from) a simultaneous process of sensitization.” This characteristic reflects the basic definition of habituation as a response decrement after repeated stimulation; however, it also adds to the description that the decrement is not always exponential and even often increases before decrementing. It is believed that this is caused by sensitization, another non-associative form of learning that affects the entire state of an organism by arousing them and facilitating other stimulus-response relationships. In some cases, it also increases the response that it elicits as well. The competition within a stimulus-response relationship between habituation and sensitization is described by dual process theory (Groves and Thompson, 1970). In cases where sensitization is not observed, such as in the tap withdrawal response in C. elegans, the underlying reason may be that the animal started in a sensitized state. Unfortunately, this hypothesis is very difficult to test. 2) “If the stimulus is withheld after response decrement, the response recovers at least partially over the observation time (‘‘spontaneous recovery”).” This excludes response decrements caused by injury or other non-recoverable decrements that should not be considered forms of learning. This is an important piece in the definition of learning because the goal of learning is to change behavioural output in order to more appropriately respond to a changing environment. If a stimulus changes from being inert to predictive, it is crucial to be able to recover the response in order to detect this change.  5  3) “After multiple series of stimulus repetitions and spontaneous recoveries, the response decrement becomes successively more rapid and/or more pronounced (this phenomenon can be called potentiation of habituation).” This can potentially differentiate habituation from other forms of passive response decrement such as sensory adaptation and fatigue because sensory adaptation and fatigue should decrement at the same rate regardless of previous exposure. 4) “Other things being equal, more frequent stimulation results in more rapid and/or more pronounced response decrement, and more rapid spontaneous recovery (if the decrement has reached asymptotic levels).” Although an inter-stimulus interval dependence of the decrement does not differentiate habituation from other non-learning decrements, an inter-stimulus interval dependent spontaneous recovery does. For decrements that solely recover in a time-dependent fashion, a deeper decrement, like one caused by more frequent stimulation should take longer to recover; this is not the case for habituation. 5) “Within a stimulus modality, the less intense the stimulus, the more rapid and/or more pronounced the behavioural response decrement. Very intense stimuli may yield no significant observable response decrement.” Again, this helps to differentiate passive decrements from learned decrements. More intense adapting or fatiguing stimuli will cause the opposite effect, more rapid and deeper decreases in response. 6) “The effects of repeated stimulation may continue to accumulate even after the response has reached an asymptotic level (which may or may not be zero, or no response). This effect of stimulation beyond asymptotic levels can alter subsequent behaviour, for example, by delaying the onset of spontaneous recovery.” 7)  “Within the same stimulus modality, the response decrement shows some stimulus specificity. To test for stimulus specificity/stimulus generalization, a second, novel stimulus is presented 6  and a comparison is made between the changes in the responses to the habituated stimulus and the novel stimulus. In many paradigms (e.g. developmental studies of language acquisition) this test has been improperly termed a dishabituation test rather than a stimulus generalization test, its proper name.” If a decrement to a given stimulus does not show any specificity, it is an indication that the decrement maybe caused by fatigue instead of habituation. 8) “Presentation of a different stimulus results in an increase of the decremented response to the original stimulus. This phenomenon is termed ‘‘dishabituation.” It is important to note that the proper test for dishabituation is an increase in response to the original stimulus and not an increase in response to the dishabituating stimulus (see point #7 above). Indeed, the dishabituating stimulus by itself need not even trigger the response on its own.” Fatigue cannot be dishabituated if the site of fatigue is downstream of any dishabituating mechanism; hence this is a good test to discriminate habituation from fatigue. 9) “Upon repeated application of the dishabituating stimulus, the amount of dishabituation produced decreases (this phenomenon can be called habituation of dishabituation).” This point simply emphasizes that most stimulus-response relationships habituate, even dishabituating stimuli whose effects are to enhance another response; this should be taken into consideration when designing experiments that test dishabituation. 10) “Some stimulus repetition protocols may result in properties of the response decrement (e.g. more rapid re-habituation than baseline, smaller initial responses than baseline, smaller mean responses than baseline, less frequent responses than baseline) that last hours, days or weeks. This persistence of aspects of habituation is termed long-term habituation.” If long-term habituation is observed, in addition to normal short-term habituation, it is a good indication that the observed decrement is habituation because passive decrements such as sensory adaptation 7  and fatigue cannot produce this effect. In extreme cases, injury could be mistaken as long-term habituation, so this should be considered when designing and interpreting long-term habituation experiments. Taking all the characteristics of habituation together, it is important to note that examples of habituation that have been studied may not share all of these characteristics, however, when a response decrement is found to exhibit enough of these features to distinguish it from passive decrements such as injury/damage, sensory adaptation, and fatigue, it can then be studied as a model example of habituation in the hopes that other parameters and mechanisms discovered will generalize throughout the animal kingdom. 1.1.1. Short-term habituation in model organisms Despite the fact that habituation has been behaviourally characterized, the cellular and molecular mechanisms responsible for habituation are not fully understood. The recent focus in the field of habituation has been to investigate model organisms that exhibit well characterized examples of habituation in an attempt to identify the underlying mechanisms. Organisms, such nematodes, molluscs, fruit flies, zebra fish, songbirds and rodents, are well suited to this task because a number of tools have been developed for each model that enable subtle and precise cellular and/or molecular manipulations. Since the focus of this dissertation will be short-term habituation, I will review evidence for the mechanism of short-term habituation from these model systems. 1.1.1.1. Mollusc In Aplysia californica, habituation is studied using the gill withdrawal reflex (Pinsker et al., 1970). If a mechanical stimulus is applied to the skin of the siphon of Aplysia, the animal responds by retracting its gill for protection. With repeated stimulation, the amplitude and the duration of the withdrawal gradually decreased. Using a semi-intact preparation of the animal, where a small portion of the siphon, the abdominal ganglion (where the siphon sensory neurons synapse onto the gill motor neurons) and 8  the gill are dissected, electrophysiological recordings from the motor neurons are possible while mechanically stimulating the siphon and measuring gill withdrawal (Castellucci et al., 1970; Kupfermann et al., 1970). Repeated stimulation of the siphon led to simultaneous decreases in gill withdrawal and excitatory post-synaptic potential (EPSP), and spiking in the motor neurons. Repeated electrical stimulation of the motor neurons that caused gill withdrawal magnitudes similar to siphon stimulation were not decreased after habituation, suggesting that the mechanism of habituation is upstream of motor neuron excitability (Kupfermann et al., 1970). This decrement is likely the result of synaptic depression because direct stimulation of the sensory neurons using an electrode caused a similar reduction in EPSPs in the motor neuron, suggesting the decrement was occurring at the synapses between sensory and motor neurons (Castellucci et al., 1970). High concentration of calcium perfused onto the ganglion, which is thought to inhibit interneuron activation, did not affect the observed decrement, suggesting that the depression is homosynaptic between the sensory neurons and motor neurons (Castellucci et al., 1970). The way in which the Aplysia is dissected can influence the contribution of a particular identified neuron to the withdrawal response of the preparation. The contribution of the motor neuron measured in these experiments (L7) to the gill withdrawal behaviour is thought to be approximately 40% (Kupfermann et al., 1974). A more recent preparation isolates a motor neuron (LGD1) responsible for 84% of the dissected response behavioural response. Experiments using this dissection supported the previous findings by showing that EPSPs in the motor neuron decreased with repeated siphon stimulation, while action potentials in the sensory neuron stayed constant (Cohen et al., 1997). Furthermore, using the same preparation, motor neuron PSPs evoked by directly injecting current into a siphon sensory neuron (LE) were significantly smaller after repeated mechanical stimulation of the siphon within the receptive field of that specific sensory neuron; repeated stimulation outside of the receptive field of the sensory neuron had no effect on the motor neuron PSPs evoked by electrical stimulation of that sensory neuron (Frost et al., 1997). This provides  9  strong evidence that depression of the sensory neuron-to-motor neuron synapses is a major part of the mechanism for habituation of the gill withdrawal reflex in Aplysia. Quantal analysis showed that the spontaneous mini-EPSPs in the motor neurons did not change in size after repeated stimulation of the siphon, suggesting that post-synaptic changes were not responsible for this synaptic depression (Castellucci and Kandel, 1974). Application of glutamate to sensory and motor neuron co-cultures that evokes motor neuron EPSPs provide further support; repeated glutamate stimulation at behaviourally relevant inter-stimulus intervals do not decrement the EPSPs (Armitage and Siegelbaum, 1998), although it should be noted that at very high stimulation frequencies (> 5 Hz), a decrement is observed (Antzoulatos et al., 2003). The nature of the presynaptic mechanism that causes this synaptic depression is unclear, but there are some hypotheses. Although it was initially suggested that this presynaptic mechanism was inactivation of the calcium currents necessary for neurotransmitter release at terminals (Klein et al., 1980), recent studies have found that the method of manipulation of these early studies using TEA, a potassium blocker that broadens the action potential, was very artificial and could have mislead the conclusions (Gover et al., 2002). More recent studies show that calcium influx measured by calcium sensitive fluorescent dye is not altered at presynaptic terminals by repeated stimulation using normal duration action potentials (Armitage and Siegelbaum, 1998). Interestingly though, calcium is necessary for the synaptic depression to occur. Superfusing the ganglia with saline with nanomolar concentrations of calcium during the repeated stimulation protocol completely blocked depression once the calcium concentration was returned to normal levels (Gover et al., 2002). A simple and popular hypothesis in the literature for synaptic depression of the siphon sensory neuron is depletion of the readily releasable pool of synaptic vesicles. A mathematical model (Gingrich and Byrne, 1985) suggested this as a possibility but only if some activity dependent mechanism that facilitates transfer of vesicles from a reserve pool to a readily releasable state exists in order to explain the inter-stimulus interval dependent spontaneous recovery (Byrne, 1982). A calcium dependent 10  facilitation of vesicle mobilization via the protease calpain has been reported for serotonin-mediated facilitation of depressed synapses (Khoutorsky and Spira, 2005), which could potentially be activated by calcium influx during repeated stimulation of the siphon sensory neuron; however, no evidence has been found to show that this does in fact occur. There is also evidence from an electron microscopy study that shows vesicle depletion from the readily releasable pool of presynaptic active zones in the sensory neuron after homosynaptic depression using two short-term depression protocols: 35 stimuli at a 30 s inter-stimulus interval and 250 stimuli at a 10 s inter-stimulus interval (Bailey and Chen, 1988). However, due to the nature of electron microscopy, the sample size for each group was only two animals. This is a concern because of the large variability between individuals that is observed in the actual behaviour. Also, short-term habituation often reaches asymptotic levels after only 5 stimuli and recovers more quickly than the time needed to prepare electron microscopy samples (Castellucci and Kandel, 1974; Byrne, 1982; Gover et al., 2002). Although vesicle depletion remains an attractive explanation for the synaptic depression that mediates habituation of the gill withdrawal reflex, it is still unclear how much it contributes to the mechanism. Abrams and colleagues have attempted to explain the mechanism of depression by creating a model dependent on a release-independent silencing of individual vesicle release sites as opposed to vesicle depletion (Gover et al., 2002). This successfully models some of the results where the vesicle depletion model fails, such as the fact that synapses with different release probabilities depress at the same rate (Jiang and Abrams, 1998; Gover et al., 2002). Again, no direct evidence supports or challenges this hypothesis yet, so it is unclear the extent of its contribution to the mechanism of habituation. 1.1.1.2. Fruit fly In Drosophila melanogaster, the molecular mechanism of habituation has been studied using parametric manipulation, pharmacological treatment and mutant analysis for a number of different behaviours, which can be grouped into three categories based on the nature of the stimulus that evokes 11  them: visual, chemical, and mechanical. I will begin by introducing each of the behaviours and experimental manipulations specific to those behaviours that have yielded insight into mechanism and then discuss what has been learned from mutant and pharmacological analyses because most of the mutants have been tested in multiple habituation protocols. 1.1.1.2.1. Visual stimuli Visual stimulation can trigger either a jump and flight response or a landing response depending on the state of the fly, conditions of the experiment and the nature of the visual stimulus. The jump response is initiated by a large approaching object (usually simulated by a sudden ‘lights-off’ condition) that evokes a jump and flight reflex. This response is also triggered by a mechanical stimulus (usually a puff of air, which also presumably predicts a large approaching object). However, the characterization of habituation has been done using the visual stimulus. The neural circuitry and physiology for this behaviour have been thoroughly characterized (Trimarchi and Schneiderman, 1995a), making it an attractive model to investigate mechanisms of habituation. This response is usually referred to as the giant fibre response because the major neural output to the motorneurons responsible for jump and flight is the giant fibre, which projects from the brain of the fly to the thoracicoabdominal ganglion. Repeated visual stimulation leads to a gradual decrease in the probability of jumps, which is immediately recoverable by a flash of light or puff of air, indicating dishabituation (Engel and Wu, 1996). To model this behaviour and investigate physiological changes, the afferent inputs to the giant fibre can be stimulated by low voltage electrode stimulation of the eyes; action potentials in the muscle responsible for jump and flight can be recorded. Similar to the behaviour, muscle activation decrements with repeated electrical stimulation and can be immediately recovered by the same dishabituating stimuli (Engel and Wu, 1996). The site of plasticity is believed to be in the afferent input to the giant fibre because higher voltage electrical stimulation than used in the habituation paradigm, which directly activates the giant fibre neurons, does not cause a decrement in the response unless very high frequencies are used, which are likely behaviourally irrelevant, and even then the decrement is small 12  and does not occur until after hundreds of stimuli presentations. Also, when stimulating the afferents to the giant fiber, muscle recordings are always performed in the right tergotrochantral muscle, the leg extensor used for jumping, and the left dorsal longitudinal muscle, the wing depressor used for flying. Decrements observed in these contralateral muscles occur synchronously suggesting the plasticity occurs before the bifurcation of the giant fibre as it enters the thorax. It is unclear exactly where the afferent stimulation begins; however, it is at least one chemical synapse away from the giant fibre because afferent stimulation has a response latency that is ~2.5 ms longer than direct stimulation of the giant fibre. Visual stimulation takes another 15 ms longer than afferent stimulation, so a portion of the circuit is not represented by this model, which might explain why visually evoked responses habituate with a slightly different range of frequencies compared to the electrically-evoked responses. The other visually evoked response in Drosophila is the landing response (Fischbach, 1981). When a fly in flight approaches an object, the animal conducts a stereotyped response in order to initiate a landing on the object. The fly extends its middle and hind legs and reaches up and in front of its head with its anterior legs. When the anterior legs make contact with the object, the fly stops beating its wings and curls its body to grasp the surface with all of its legs. This behaviour can be simulated by suspending a fly in the air by tethering its back to copper wire, which causes the animal to simulate flight. When an object, such as a piece of paper, approaches the fly, it initiates the leg extension and reaching behaviour. If the object touches the legs of the fly it will grasp the object and stop beating its wings. The leg extension and reaching can also be evoked by simulating an approaching object on a screen in the visual field of the fly, in which case the fly only performs the leg extension behaviour. Repeated visual stimulation at a 2 s inter-stimulus interval causes decrement of the probability of the response as well as an increase in the latency of the response (Fischbach, 1981; Wittekind and Spatz, 1988; Rees and Spatz, 1989). The response decrement is likely caused by habituation and not sensory adaptation or fatigue for the following reasons. Although, the latency of the response decreases, the duration of the response was constant during the stimulation (Wittekind and Spatz, 1988). If flies are 13  stimulated while they are not flying, they do not perform a landing response thereby not inducing any motor activity that could potentially cause motor fatigue. Repeated stimulation in this state still led to smaller landing responses than naïve flies after flies were returned to flight (Wittekind and Spatz, 1988). Weak stimuli habituate faster than intense stimuli (Wittekind and Spatz, 1988). Finally, allowing the flies to land causes dishabituation of the decremented landing response (Fischbach, 1981; Wittekind and Spatz, 1988). Insights into the site of landing response habituation have been gained from generalization studies where the landing-evoking stimulus is repeatedly presented in one region of the visual field and then the flies are tested in response to the same stimulus presented in a different region to see how much plasticity is retained across the visual field (Fischbach and Bausenwein, 1988). Habituation only partially generalizes between different regions of the visual field. Interestingly, along the vertical axis, this is graded. The region right next to the habituated region generalizes to a greater extent than the furthest region, but along the horizontal axis the generalization is not graded. Close and far regions to the habituated one generalize the same amount. Also, the more the habituated region overlaps with the test region, the greater the generalization. Taken together, this suggests that habituation is occurring downstream of both receptor neurons and motion detection neurons due to the nature of the landing-evoking stimulus, but is upstream of neurons that converge information about neighbouring regions of the visual field. The neural circuitry involved is not well understood, but Fischbach and Bausenwein suggest it must be at least as downstream as the lobula plate, which is known to be activated by moving stimuli, but rule out the lobula plate, since genetic ablation of these cells has no effect on the landing responses (Fischbach and Bausenwein, 1988). Instead they hypothesize the next brain region downstream in the visual system of the fly, the lobula itself, is the likely site of habituation of the landing response; however, this has not yet been tested. Mathematical modeling of the landing response using sensory-motor gating networks predicts that habituation must be occurring at multiple sites, with major local contributions just after sensory processing and minor global components further 14  downstream but prior to the generation of the ritualistic response (Ogmen and Moussa, 1993), which seems to fit well with the experimental observations detailed above. 1.1.1.2.2. Chemosensory stimuli Drosophila melanogaster also respond to chemical stimuli of various kinds. When a chemosensory stimulus (a smell or a taste) is not paired with reinforcement or punishment, the response habituates. Many olfactory stimuli, such as a 4 sec presentation of 10% benzaldehyde, evoke a similar jump response as described above for the giant fibre response (McKenna et al., 1989). Repeated stimulation causes the probability of response to gradually decrement; this decrement can be rapidly dishabituated by centrifuging the animals (Boynton and Tully, 1992). Despite its similarity to the response to the visually evoked stimuli, it is thought that at least the cellular pathway for the chemosensory response may be different because the sensory afferents are different and the jumping response is not driven by the giant fibre (Trimarchi and Schneiderman, 1995b) as it is with visual stimuli; however, the possibility remains that intermediate cells are shared. Within the first 30 s of exposure to chemical odorants, flies increase their locomotion. Repeated stimulation causes a decrement in the odour-evoked change in speed. Mechanical stimulation after the decrement immediately causes dishabituation of the response (Cho et al., 2004b). Habituation generalization has been observed between odorants detected by different primary olfactory neurons, suggesting that the site of plasticity is further downstream in the fly’s brain. One brain structure that is both known to receive olfactory information and be important for memory in the fly is the mushroom bodies. Ablation of the mushroom bodies impaired habituation of this olfactory induced startle without disrupting the animal’s sensitivity to the initial exposure (Cho et al., 2004b) suggesting that a portion of the plasticity may be occurring in this brain structure. Olfactory habituation can also be measured using a choice test. When a population of flies are placed in a Y-maze that has an aversive chemical, such as carbon dioxide or ethyl butyrate, in one arm of  15  the maze and a control substance in the other, fewer flies explore the arm with the aversive chemical than the control arm. Exposure to the aversive odorant for 30 minutes prior to testing causes fewer animals to avoid the chemical (Das et al., 2011). The decrement spontaneously recovers with a half-life of approximately 20 minutes. Unlike olfactory-induced locomotion habituation, this olfactory preference habituation is stimulus specific; for instance, pre-exposure to carbon dioxide does not change the preference to ethyl butyrate and vice versa. Mutant analysis that will be explained in further detail later has localized the site of plasticity to the central nervous system of the fly suggesting that the observed decrement is habituation, not the result of sensory adaptation. Soluble chemicals are tasted by sensory organs on the tarsus of the forelegs. In response to a sucrose solution, the fly extends its proboscis in an attempt to drink the liquid. If the stimulus is removed quickly, the fly is unable to drink any liquid, avoiding reinforcement or satiety, making it a good stimulus to test habituation. As expected, repeated stimulation at an inter-stimulus interval of one minute led to a decrement in response probability; habituation was also induced by a single exposure to very high sucrose concentration lasting as long as ten minutes (Duerr and Quinn, 1982). Some of the neural plasticity that mediates this behaviour must occur in the central nervous system because repeated stimulation to one leg generalizes to the contra-lateral leg (Duerr and Quinn, 1982; Bouhouche et al., 1993). The circuits involved in both the habituation and generalization are unknown; however, it is clear that the protocerebral bridge of the fly nervous system is important because mutants that lack this structure are unable to generalize proboscis extension habituation (Bouhouche et al., 1993). 1.1.1.2.3. Mechanosensory stimuli Habituation of two mechanosensory reflexes have been studied in decapitated fly preparations because the circuitry involved in the behaviour is far less complex without the brain and experimental conditions can be more easily controlled. The first is the leg extension reflex, a proprioceptive response induced when the mesothoracic tibia is deflected from its resting position (Jin et al., 1998). Repeated flexion causes a decrement in the electrophysiological response in the extensor muscles responsible for 16  the behaviour and follows many of the same characteristics that rule habituation, such as spontaneous recovery, inter-stimulus interval dependent habituation rates and dishabituation by a mechanical stimulus to the abdomen (Jin et al., 1998). The other reflex shown to habituate in this reduced preparation is a cleaning reflex induced by a puff of air to the thoracic bristles (Corfas and Dudai, 1989). Decapitated flies respond to the movement of one of their bristles by sweeping their foreleg or third leg across their body. Repeated stimulation causes the decrement of both response probability and magnitude and can be dishabituated by a train of puffs to the dorsal thorax. Individual bristles could be targeted, no generalization of habituation occurred between bristles providing further evidence the decrement was not the result of fatigue. 1.1.1.2.4. Genetic and pharmacological investigations Different wild-types isolates have been found to habituate differently in some habituation protocols, implicating the influence of genetic factors. For example, the Canton-S wild-type line habituates more slowly than the Berlin wild-type line in the visually evoked landing response (Rees and Spatz, 1989). Differences in behaviour seen in mutant strains of Drosophila in many of the habituation protocols support the roles of several genes in habituation. The most widely studied of these include manipulation of cyclic-AMP metabolism and signalling. rutabaga mutants that have decreased activity of the cAMP synthesis enzyme adenylyl cyclase habituated more slowly than wild-type in the visually evoked giant fibre response (Engel and Wu, 1996), both the olfactory evoked jump (Asztalos et al., 2007a) and locomotory startle (Cho et al., 2004a), and the proboscis extension response (Duerr and Quinn, 1982). rutabaga mutants did not habituate at all to carbon dioxide or ethyl butyrate aversion in y-maze (Das et al., 2011). This was not the case for the landing response; rutabaga mutants habituate more rapidly (Wittekind and Spatz, 1988; Rees and Spatz, 1989). dunce mutants that have decreased phosphodiesterase activity, which degrades cAMP in cells, habituated more rapidly in the visually evoked behaviours (Wittekind and Spatz, 1988; Rees and Spatz, 1989; Engel and Wu, 1996), but habituated more slowly to the chemically evoked behaviours (Duerr and Quinn, 1982; Asztalos et al., 17  2007a). Confusing the issue further, three pharmacological treatments (caffeine, theobromine and theophylline) that impair phosphodiesterases, all caused flies to habituate more slowly than controls in the landing response, while forskolin, a drug that stimulates adenylyl cyclases, increasing the production of cAMP, had no effect on habituation (Wittekind and Spatz, 1988). It is difficult to conclude much from this other than proper cAMP regulation is critical for normal habituation. It also suggests that although some circuits may use similar mechanisms for habituation, others may use alternative mechanisms. The effect of adenylyl cyclase in habituation has been used to further dissect the details of the circuitry and mechanism of habituation to aversive odorants (Das et al., 2011). The neural circuit that mediates this behaviour is partially characterized. Olfactory sensory neurons (OSN) project to the glomeruli of the antennae lobe to make connections onto various classes of neurons. One class are the glomeruli-specific projection neurons (PN) that are often specific to a particular odour and send axons to deeper structures of the central nervous system such as the lateral horn and the mushroom bodies. Another class is the multi-glomerular local interneurons (LN) that have intra- and inter-glomerular inhibitory input onto PNs. Many of the other neuronal types have not yet been characterized. Conditional expression of wild-type rutabaga selectively in the LNs in rutabaga mutants rescued the mutant phenotype (no habituation) so that flies habituated like wild-type. Selective expression in OSNs, PNs or neurons in the mushroom bodies did not change the mutant effect. A dominant, temperature sensitive dynamin selectively expressed in the LNs that inhibits neurotransmission at restrictive temperatures blocked the behavioural expression of habituation when the switched to the restrictive temperature during either the habituation period or the test. Together these data suggests that adenylyl cyclase signaling and neurotransmission in the LNs are required for olfactory avoidance habituation. The LNs express glutamic acid decarboxylase (GAD1), an enzyme required for GABA synthesis, and have connections onto the PNs. Expression of wild-type rutabaga in GAD1-positive neurons also rescued the rutabaga phenotype. The projection neurons express a GABAA-type receptor (Rdl). 18  Selective knockdown of Rdl using RNAi targeted specifically to the PNs activated by carbon dioxide blocked habituation to carbon dioxide without affecting habituation to ethyl butyrate and vice-versa. Das et al. hypothesize that potentiation of the inhibitory connections between LNs and PNs is responsible for olfactory habituation. Further supporting this hypothesis, artificial depolarization of the LNs by selective expression of a temperature sensitive TRPA receptor mimicked habituation when the receptor was activated during exposure to aversive chemicals; however, this was not stimulus specific (i.e. response decrement was observed during both carbon dioxide and ethyl butyrate stimulation). This is not surprising because LNs are multi-glomerular, synapsing onto many types of PNs, but it begs the question: how is stimulus specificity mediated if LN-PN synaptic potentiation is the mechanism underlying olfactory habituation? Das et al. reveals that others have found that presynaptic potentiation of GABA neurons can be mediated by hetero-, post-synaptic NMDA receptors acting as coincidence detectors and sending a retrograde signal to potentiate the GABA neurons (Nugent et al., 2007; Castillo et al., 2011). The NR1 subunit of the NMDA receptor is broadly expressed in the insect brain, but selective RNAi knockdown of NR1 in the carbon dioxide sensitive PNs blocked carbon dioxide habituation without disrupting ethyl burtyrate habituation and vice-versa. In addition to this, RNAi knockdown of a vesicular glutamate transporter required for the loading of glutamate into synaptic vesicles for neurotransmission also blocked habituation by targeting the RNAi to either LNs or GAD1-expressing neurons. This suggests that glutamate is co-released with GABA from the LNs to activate NMDA receptors on the PNs and that this is necessary for odor-specific habituation. It supports the hypothesis that NMDA receptors act as a coincidence detector to mediate the odor-specificity of habituation. To complete this mechanism, a retrograde signal is needed from the PNs to the LNs that potentiates the LN terminals. This has not yet been identified, but is an extremely exciting avenue for future research. The work by Das et al. is important for two reasons. First, it is one of the most developed mechanisms for habituation that has been identified in the field so far. Second, it supports an alternate 19  type of cellular mechanism for habituation. The majority of habituation research has focused on a potential mechanism explained by the depression of excitatory synapses such as those described earlier in the Aplysia section. The experiments performed by Das et al. provide support for another potential mechanism for habituation, the potentiation of inhibitory synapses. This supports the conclusions from work on tail-flip escape habituation in crayfish that suggested that descending inhibitory interneurons in the central nervous system of crayfish play a role in the cellular mechanism of habituation (Krasne, 1969; Shirinyan et al., 2006). The second most studied category of molecules in habituation of Drosophila is potassium channels. They are good candidates for potential roles in neural plasticity because they are involved in neural excitability and regulate synaptic transmission. Four potassium channel genes have been investigated in habituation of the giant fibre response. Slowpoke is a calcium-activated potassium channel subunit. Mutants homozygous for either of two null alleles of this gene cause slow habituation of the giant fibre response, which models visually evoked jump (Engel and Wu, 1998), and one of these alleles also has slow olfactory evoked jump (Joiner et al., 2007). The function of Shaker, a pore-forming alpha subunit of a voltage-gated potassium channel, is less clear. Mutants that are homozygous for an allele that alters gating kinetics and voltage sensitivity of the channel have more rapid habituation of the giant fibre response. Similarly, another allele, which is thought to affect alternative splicing and dramatically reduces function in neurons but not in muscle, also causes more rapid habituation of the giant fibre response. However, mutants that are homozygous for a null allele habituate at a similar rate as control flies (Engel and Wu, 1998). In the olfactory jump response, similar mutations all cause flies to habituate more slowly (Joiner et al., 2007). Furthermore, two alleles of Hyperkinetic, a beta-subunit that is thought to interact with Shaker containing channels, both cause slow habituation of the giant fibre response; one of these alleles is amorphic (null) (Engel and Wu, 1998). Interestingly, the small amount of habituation that is observed is only seen in the dorsal longitudinal muscle required for flight and not the contralateral tergotrochanteral muscle that extends 20  the legs during a jump. This suggests that habituation in the giant fibre afferents are eliminated in these Hyperkinetic mutants and allows some habituation to occur further downstream of the giant fibre but only in the flight circuitry after it bifurcates from the jump motor neurons. The hypomorphic allele of Hyperkinetic , as well as a second hypomorph, both showed slow habituation of the olfactory evoked jump as well (Joiner et al., 2007). Finally, a gene that encodes another pore-forming subunit, ether a go-go, causes extremely rapid habituation for both the giant fibre response (Engel and Wu, 1998) and olfactory jump (Joiner et al., 2007). It is unclear if any of these potassium channel subunits are working together, either directly in the same channel or in different channels within the same pathway or mechanism. However, Hyperkinetic is epistatic to the other genes because in most instances double and triple mutants with alleles of Shaker and ether a go-go, which both cause rapid habituation phenotypes, lead to slower rates similar to those caused by single Hyperkinetic mutations (Engel and Wu, 1998), suggesting that its effect on habituation is furthest downstream than Shaker and ether a go-go. More experimentation is needed to fully understand how these potassium channels exert their effect within the mechanism for habituation; however it is obvious that potassium currents play a critical role. A well-known process that can modify cellular components is phosphorylation, which is positively and negatively catalyzed by kinases and phosphatases, respectively. It is not surprising then that these modifiers of cellular plasticity also play a role in behavioural plasticity; several mutations in kinase and phosphatase genes cause habituation abnormalities. One kinase that has been implicated in many forms of learning and memory is calcium-calmodulin kinase II (CaMKII) (Griffith, 2004). A point mutation in Drosophila CamKII that results in a constitutively active calcium-independent variant, specifically expressed in the sensory neurons involved in the leg extension reflex of decapitated flies, results in almost no decrement of the muscle response during repeated stimulation (Jin et al., 1998). Sensory  21  neurons expressing a CaMKII inhibitory peptide had smaller initial responses to stimulation that immediately sensitized to normal levels with repeated stimulation and never habituated. These data support a role for CaMKII in the molecular mechanism of habituation that occurs in the sensory neurons of the leg extension reflex. A genetic screen for transposon insertion mutants with habituation deficits in olfactory startle identified a mutation in Shaggy, a glycogen synthase kinase-3 homolog (Wolf et al., 2007). Precise excision of the transposon reverted the habituation phenotype back to wild-type, while imprecise excision, which removed part of the coding sequence, maintained the habituation deficit. Also, one other independent transposition allele of Shaggy phenocopied these effects. The protein products of precision and imprecision excision were purified and their kinase activity was found to be similar, suggesting that alleles with behavioural deficits are not caused by decreased kinase activity but more likely by a decrease in expression level. Over expression of Shaggy in neurons supported this hypothesis because it caused more rapid habituation. In an attempt to isolate the site of plasticity, over expression was targeted to more discrete sets of neurons in the known olfactory circuit (primary sensory neurons, secondary projection neurons, and the mushroom bodies); however, none increased habituation in the same way as ubiquitous over expression, suggesting that the effect of Shaggy is further downstream (although the projection neurons cannot be ruled out because over expression in the projection neurons caused lethality). Phosphorylation of the serine-9 residue of Shaggy down regulates its activity, thereby activating downstream signals (Jope and Johnson, 2004); habituated flies had significantly higher levels of Shaggy in the phosphorylated state than naive controls (Wolf et al., 2007). Although, the role of Shaggy has yet to be tested in other habituation protocols, this evidence supports an important role in the molecular mechanism of habituation in Drosophila olfactory startle. A natural variation in the foraging gene (Sokolowski, 1980) that encodes a cGMP-dependent protein kinase, leads to a difference in activity level of the enzyme (Osborne et al., 1997). Flies with the allelic variant that causes higher kinase activity habituate more slowly than flies with the lower activity 22  variant; this has been observed for both the giant fibre response (Engel et al., 2000) and the proboscis extension response (Scheiner et al., 2004). Two other alleles caused by P-element transposon insertion/excision had even more dramatic phenotypic differences. One foraging mutant, with a Pelement inserted into the 5’-end of one of the three splice variants, habituated more rapidly than all other mutants. When the P-element was precisely excised, the resulting mutant showed almost no habituation (Engel et al., 2000). Unfortunately, the kinase activity of these two alleles has not been assayed. Taken together these data suggest a strong role for cGMP-dependent protein kinase in the rate of habituation. The role of protein phosphatase 1 at 87B in habituation of the landing response was investigated using a non-lethal allele (Su-var(3)601) crossed into the Berlin wild-type strain (Asztalos et al., 1993). Both heterozygotes and homozygotes for this allele habituated more rapidly than control lines. Suvar(3)601 heterozygotes had significantly less protein phosphatase activity than controls. This suggests that protein phosphatase 1 at 87B is important for the normal rate of habituation of the landing response in Drosophila. A mutation that disrupts the period gene, a transcription factor, causes flies to habituate to the electrically evoked giant fibre response at much longer inter-stimulus intervals than observed in control flies (Megighian et al., 2001). However, at frequencies that cause habituation in controls, no difference is observed in the mutants. An interesting finding is that this is only observed when flies were housed in constant light conditions; period mutants and control flies showed similar habituation if housed in a 12 hour light-dark cycle. Seeing as the period gene is important for regulating circadian rhythm, circadian rhythm might play a role in modulating habituation. In the majority of situations discussed so far, loss of function of the gene leads to slower habituation; however, a few genes cause the opposite effect. Loss of function mutations in amnesiac, fickle, synapsin, and caki cause flies to habituate more rapidly. amnesiac encodes a neuropeptide  23  precursor (Feany and Quinn, 1995) and affects habituation to landing evoking visual stimuli (Wittekind and Spatz, 1988; Rees and Spatz, 1989). fickle encodes a tyrosine kinase (Baba et al., 1999) and affects olfactory jump habituation (Asztalos et al., 2007b). synapsin encodes a synaptic vesicle phosphoprotein that regulates vesicle release (Cesca et al., 2010) and affects olfactory jump habituation (Godenschwege et al., 2004). caki, the human CASK homolog, encodes a membrane-associated guanylate kinase that is expressed in large portions of the Drosophila brain as well as the neuromuscular junctions (Martin and Ollo, 1996; Lu et al., 2003). caki affects habituation of the electrically-evoked giant fibre response (Zordan et al., 2005). The caki mutant deficit was observed both when giant fibre afferents were stimulated and when the giant fibre was activated directly, suggesting the effect is likely occurring at the neuromuscular junction rather than the brain (Zordan et al., 2005). Seeing as habituation is usually not observed in wild-type flies when the giant fibre is directly stimulated, a loss of caki expression may lead to premature synaptic transmission failure at the neuromuscular junction causing the apparent increase in habituation. amnesiac, fickle and synapsin have not been explored further in terms of habituation, so it is difficult to draw any conclusions other than these genes may play a role in the mechanism of habituation. All together 15 genetic components have been identified to influence Drosophila habituation across six different behavioural protocols. This is a step forward in our understanding of the molecular mechanisms of habituation; however, much more investigation is needed before these components and others can be grouped together into fully understood mechanisms. 1.1.1.3. Zebrafish Habituation has recently been described in Danio rerio, Zebrafish. An acoustic stimulus startles the fish causing it to move a significantly greater distance than when no stimulus is presented; repeated stimulation causes the distance moved to decrement, and this decrement can be dishabituated by a light stimulus (Best et al., 2008). Using pharmacological agents, Best et al. found that the habituated level of animals treated with the acetylcholine esterase inhibitor, donepezil, showed slower habituation. 24  This effect was blocked by co-administration of the nicotinic receptor antagonist, Mecamylamine. They also found that the NMDA-type glutamate receptor antagonist, memantine, slowed habituation of the acoustic startle. It should be noted that both donepezil and memantine caused much larger initial responses as well. More recently, using a high speed video analysis that was capable of actually identifying the stereotyped acoustic startle response, known as a C-start, the glutamate effect was replicated; three different NMDA receptor antagonists all slowed habituation, but also all increased sensitivity to the initial response (Wolman et al., 2011). Because the drugs increased sensitivity to the stimuli, the habituation rate effects are difficult to interpret because one of the characteristics of habituation is that more intense stimuli habituate more slowly (Thompson and Spencer, 1966). However, this work suggests glutamate and acetylcholine neurotransmission may be involved in the mechanisms of habituation. A number of other good candidates were identified in a small molecule screen using automated video analysis of the C-start startle habituation (Wolman et al., 2011). Most of the drugs that caused slower habituation also increased sensitivity except for hydrastine, a GABA-A receptor antagonist, butaclamol, a dopamine receptor antagonist, and SU-9516, an inhibitor of cyclic dependent kinases. Additional results using several other drugs supported the hypothesis that the GABAergic and dopaminergic effects were not caused by off target interactions; two positive allosteric modulators of GABA-A receptors as well as a D3 receptor agonist caused more rapid habituation, suggesting that these neurotransmitters are involved in the mechanism for C-start habituation. Two other cyclic dependent kinase inhibitors were tested as well, but both had opposite effects to SU-9516, suggesting more investigation is needed to confirm the role of these kinases in habituation. However, this highly quantitative high throughput approach seems very promising for such a task. 1.1.1.4. Song birds In their natural environment, white-crowned sparrows (Zonotrichia leucophrys nuttalli) perform a battery of behaviours in response to a (likely novel) male birdsong of the local dialect presented within 25  their nesting territory. These behaviours were fully described in a set of field studies (Patterson and Petrinovich, 1979; Petrinovich and Patterson, 1979, 1980) and can be divided into sex-specific behaviours, which include, for example, their own bird song, flight, and fluttering. Repeated presentation of the ‘intruder’ birdsong had no effect on many of these evoked behaviours (i.e. they continued to response at the same levels as the initial trial). However, their own bird song, female chinks (short, rapid notes covering a range of frequencies), and male flutters significantly decremented in direct response to the stimulus. Songs during the silence between stimuli, which initially increased from baseline after the first stimulus, decremented as well (Petrinovich and Patterson, 1979). In contrast to other forms of short-term habituation, more frequent stimulation did not lead to more rapid decrement (Patterson and Petrinovich, 1979) and no spontaneous recovery was observed over the time course that was tested, other than male flutters. Although, the recovery was significant it was very minimal (Patterson and Petrinovich, 1979). Perhaps the stimulus protocol that they used was invoking long-term memory mechanisms, or complex stimuli like birdsong follow different habituation characteristics than the simple stimuli that are commonly used to study habituation. It was obvious the decrement was not the result of sensory adaptation or fatigue because if the birdsong stimulus was contingent on the reply song instead of evenly distributed, the behaviours did not decrement to the same extent (Petrinovich and Patterson, 1980). In the lab, song birds are studied using zebra finch (Taeniopygia guttata). Unfortunately, their behavioural response to song playback is to pause motionless and silent often for 10 minute (Stripling et al., 2003). This makes it very difficult to study short-term habituation on the same time scale as other models. However, repeated song presentation leads to long-term memory for habituation, so that 24 hours later, their freezing time is significantly shorter than naïve birds, who have not experience the birdsong stimulus previously (Stripling et al., 2003). Shorter time points have not been tested. A neural correlate of short-term habituation to birdsong has been identified in the auditory forebrain using multi-unit electrical recordings (Chew et al., 1995; Stripling et al., 1997) in a region that 26  is thought to be important for song perception, the caudomedial neostriatum (NCM) (Mello et al., 1992; Mello and Clayton, 1994). Neurons in the NCM are responsive to simple tones, conspecific and heterospecific songs. Interestingly, only the activation by songs (both conspecific and heterospecific) causes the neural responses to decrement, while simple tones do not (Chew et al., 1996; Stripling et al., 1997). Perhaps this suggests that simple stimuli habituate at a different level in the nervous system, such as the sensory neurons as seen in Aplysia and C. elegans (Castellucci et al., 1970; Wicks and Rankin, 1997; Rankin and Wicks, 2000; Kindt et al., 2007), while compound/complex stimuli habituate more centrally in the nervous system. Birdsong activates the transcript of the immediate early gene zif-268/egr-1/NGFI-A/Krox-24 (ZENK) in the auditory lobe of Zebra finch, most prominently in NCM (Mello et al., 1992; Kruse et al., 2000). Repeated stimulation leads to higher levels of expression peaking with 180 presentations over the course of 30 minutes (an inter-stimulus interval of 10 seconds)(Kruse et al., 2000), after which the response habituates back to baseline levels after 3 hours of stimulation (Mello and Clayton, 1995). The time course for this gene expression is roughly 30 minutes, which means it probably does not contribute to the mechanism of short-term habituation. An extensive investigation of changes in RNA and protein expression has been conducted using this long-term habituation paradigm (Dong et al., 2009; Warren et al., 2010; Gunaratne et al., 2011). This large scale genomic and proteomic studies has identified potential genes, particularly regulatory micro-RNAs, and proteins important for long-term memory for habituation. Although, this might not directly relate to the mechanism of short-term habituation because the time course of short-term habituation may be too brief for changes in gene expression to occur. With some modifications to assay more short-term changes, large scale proteomic assays may be applied to the study of short-term habituation as well. 1.1.1.5. Rodents Similar to song birds, neural correlates in central neurons have been identified for auditory, tactile and olfactory stimuli in rats (Best et al., 2005). However, much more work has been targeted towards 27  short-term habituation. A brief loud acoustic sound or a sudden, but gentle, tactile stimulus to the head, like a puff of air, causes rodents to startle by contracting their body muscles. Repeated presentation of these stimuli causes a response decrement characteristic of habituation (Davis and Wagner, 1969). The most direct neural circuit for both the auditory- and tactile-evoked reflexes have been traced and have been found to converge. Sensory afferents synapse onto projection neurons in the cochlear nucleus/root (auditory) or principal nucleus V (tactile), which then send axons to synapse onto giant neurons of the pontine caudal reticular formation (PnC). These in turn project down the spinal cord onto motor neurons that drive the muscle contractions across the body (Davis et al., 1982a). Neural correlates of the behavioural decrement have been found in the giant neurons of the PnC. However, electrical stimulation of these cells, which elicits a similar startle response, does not habituate (Lingenhohl and Friauf, 1994). In contrast, stimulation of the projection neurons in the cochlear nucleus does lead to startle responses that habituate with repeated activation (Davis et al., 1982b), suggesting that the site of plasticity may be at the synapse between these two neuronal populations. Direct stimulation of the fibres from either the auditory or tactile nuclei leads to detectable excitatory post-synaptic currents and potentials (EPSC/P) in the giant neurons that decrement with repeated stimulation (Weber et al., 2002; Schmid et al., 2003; Simons-Weidenmaier et al., 2006). Although cross-modal paired-pulse potentiation can be observed (Schmid et al., 2003), the decrement caused by repeated stimulation does not generalize (Simons-Weidenmaier et al., 2006). Glutamate can be uncaged in the PcN in a way that creates similar EPSCs as those evoked by cochlear nucleus stimulation. Repeated glutamate uncaging does not cause a decrement in EPSCs. Furthermore, when synaptic depression is induced by cochlear nucleus stimulation, uncaging of glutamate still evokes a similar response. Taken together, this suggests the mechanisms causing this synaptic depression and the habituation with which it correlates are presynaptic (Simons-Weidenmaier et al., 2006). Presentation of a novel olfactory stimulus to a rat causes a drop in heart rate that occurs within 510 seconds of stimulus onset and lasts 20-30 seconds. This is known as the heart rate orienting 28  response. For odours, the minimal neural circuitry from sensory afferents to autonomic efferents is mapped from the olfactory bulb to the anterior piriform cortex to the amygdala, which has input onto the vagal projections to the heart (Wilson, 2009). While olfactory receptors in the nose and mitral cells in the olfactory bulb do decrement slightly, they stay quite responsive during repeated odour presentation. In contrast, pyramidal neurons in the anterior piriform cortex that receive synaptic input from olfactory mitral cells, rapidly inactivate with repeated odour stimulation(Wilson, 1998). In vitro investigations in brain slices found that group III metabotropic glutamate receptor antagonists block mitral cell-piriform pyramidal cell homosynaptic depression (Best and Wilson, 2004). When this was translated back to in vivo experiments, the metabotropic glutamate receptor blocker reduced odourevoked heart rate orienting response habituation (Best et al., 2005) when injected into the piriform cortex. This is compelling evidence to suggest that metabotropic glutamate receptors participate in the mechanism of olfactory habituation in rats. Metabotropic glutamate receptors were initially thought to play a role in this homosynaptic depression (Weber et al., 2002) that was hypothesized to mediate the acoustic startle response habituation as well. However, more recent experiments present strong evidence to rule this out and present a valid explanation for the original artifact (Schmid et al., 2010). Finally, although mutant mice with the expression of the vesicular acetylcholine transporter knocked down have impaired long-term memory for habituation of the acoustic startle, they had completely normal short-term habituation (Schmid et al., 2011). 1.1.2. Habituation in C. elegans C. elegans is a soil dwelling, bacteria eating nematode (roundworm) that is 1 mm in length. In the lab, it is cultivated on lawns of E. coli prepared on agar-filled Petri plates. This organism has been studied extensively over the past 40 years to understand how genes regulate development and cell biology. The highly detailed level of characterization, particularly of its neurobiology, makes C. elegans an excellent model to explore the molecular mechanism of plasticity in the nervous system. Some of the advantages of using C. elegans as a model include the following. The worm has a small nervous 29  system comprised of 302 neurons in the adult hermaphrodite. These neurons are all identified, meaning that their development and anatomical location have been mapped and are invariant between individuals. This allows researchers to attribute behaviours to specific neurons that can then be studied between animals. The C. elegans neural connectome has also been mapped using serial section electron microscopy. There are fewer than 10,000 synapses in an adult worm. Despite this simplicity, many of the components that comprise the nervous system of C. elegans are similar to more complex organisms such as mammals. These components include neurotransmitters, such as glutamate and dopamine, their respective receptors, and intracellular signalling cascades. In fact, over 5,000 genes in C. elegans are orthologous to human genes (Remm et al., 2001). This suggests that molecular mechanisms for plasticity discovered in C. elegans may be conserved in other organisms. C. elegans is ideal for the study of cellular and molecular mechanisms because of the ease with which genetic manipulations can be performed. For example, mutation can easily be introduced to the genome (Barstead and Moerman, 2006) and transgenes can be introduced into C. elegans, allowing for temporal and spatial control of gene expression. This technique can also be used to introduce exogenous proteins such as fluorescent proteins (Chalfie et al., 1994), which are useful for neuroanatomical studies, and subcellular localization of proteins. Recent advances in genetically encoded fluorescent protein variants, such as Cameleon , which is sensitive to intracellular calcium, allow experimenters to monitor cell physiology (Miyawaki et al., 1997). C. elegans is transparent, making it possible to visualize these fluorescent proteins in vivo while worms are awake and behaving (Suzuki et al., 2003). The worm’s transparency also allows individual neurons to be easily ablated using lasers, a useful tool for dissecting neural circuits (Chalfie et al., 1985). Furthermore, C. elegans short life-cycle and ability to survive in clonal populations makes them an excellent genetic model (Brenner, 1974), which has led to the development of many tools and resources such as a fully sequenced genome (The C. elegans Sequencing Consortium, 1998), thousands of identified mutants (a consortium also exists that accepts requests for the generation of new knockout mutants), and RNAi libraries capable of knocking down gene expression 30  (e.g. Kamath and Ahringer, 2003). Finally, and most importantly for the study of habituation, C. elegans exhibits well described and measurable behaviours that change with experience. These advantages make C. elegans an excellent model in which to understand how the molecular components of nervous systems mediate habituation. Response decrements to repeated stimuli have been described for a number of C. elegans behaviours: nose touch response (Hart et al., 1999; Ezcurra et al., 2011), chemical attraction (such as benzaldehyde and isoamyl alcohol) (Colbert and Bargmann, 1995), chemical avoidance (such as osmotic and copper avoidance)(Hilliard et al., 2005). However, the first documented example of habituation and the most fully characterized is the tap withdrawal response. Habituation in C. elegans was first observed for a reversal response (swimming backwards for a short distance) elicited by a mechanical tap to the Petri plate in which the worms live in the laboratory; this response was named the tap withdrawal response (Rankin et al., 1990). The magnitude of this response was measured by quantifying the distance that the worm moved backwards during the reversal. When C. elegans were stimulated 40 times (once every 10 seconds), their response on the last stimulus was significantly smaller than their response on the first. In order to confirm that this decrement was caused by habituation and not some form of sensory adaptation or motor fatigue, Rankin et al. tested to see if the response could be immediately recovered by administering a dishabituating stimulus. In this case they used an electrode to deliver a shock through the agar on either side of a swimming worm; electric shock has been used as a dishabituating stimulus previously for other animals (Pinsker et al., 1970). The response to tap increased significantly immediately after the shock, indicating that the decrement after the repeated stimulation was in fact caused by habituation (Rankin et al., 1990). The main objective of using C. elegans as a model for the study of biological phenomena is to use the ease and power of genetic and molecular manipulation available to efficiently dissect the underlying mechanism. Although habituation has been studied behaviourally for over a century, the study of habituation in C. elegans is relatively new. Therefore, in order to elucidate the mechanisms of 31  habituation using C. elegans, it is important to first understand the behaviour. A number of parametric studies tackled this preliminary step by carefully characterizing the behavioural nature of habituation of the tap withdrawal response in C. elegans. These include the role of development and life-span (Chiba and Rankin, 1990; Beck and Rankin, 1993; Rankin et al., 2000), inter-stimulus interval (ISI), level of habituation, number of stimuli, missed stimuli (Rankin and Broster, 1992) and switching between ISIs (Broster and Rankin, 1994). The role of ISI has particularly been important for developing hypotheses regarding the molecular mechanisms of habituation. By measuring the response to each stimulus during the habituation protocol, Rankin and Broster found that the rate of habituation followed a negative exponential curve with a rapid initial decrease in response that gradually approached an asymptotic level (Rankin and Broster, 1992). This is consistent with habituation curves observed in other animals (Thompson and Spencer, 1966). The rate of habituation is dependent on the ISI; short ISIs cause a more rapid decrement in response as well as a lower asymptotic level than long ISIs (Rankin and Broster, 1992), which is consistent with other animals such as rats (Askew, 1970). ISI also affects the rate of spontaneous recovery (Rankin and Broster, 1992). Rankin and Broster showed that the spontaneous recovery from a habituated level is dependent on the ISI (the time between successive stimuli). Habituation to short ISIs (such as 10 seconds) led to rapid recovery in approximately 10 minutes, while habituation to long ISIs (such as 60 seconds) led to much slower recovery, after 30 minutes animals habituated at a 60 second ISI had only recovered to about 20% of their initial response, and later studies show that worms have recovered to approximately 80% after an hour, but are back to naïve levels after 24 hours (Beck and Rankin, 1995; Beck and Rankin, 1997). A search of the literature revealed that ISI dependent spontaneous recovery was observed in other studies of habituation; e.g. a single post-habituation trial showed that rats had smaller responses after habituation at long compared to short ISIs (Davis, 1970a, b). This focus on ISI dependent recovery provided two important outcomes (Rankin and Broster, 1992). First, ISI-dependent recovery provided an alternative method for distinguishing habituation from fatigue. After habituation, animals stimulated 32  at a higher frequency (shorter ISI) recovered more rapidly than animals stimulated at a lower frequency (long ISI). This is opposite to what would be expected after a decrement caused by fatigue; animals stimulated with high frequency stimulation who are responding at lower levels than those caused by low frequency stimulation would take longer to recover than animals fatigued to a lesser extent by low frequency stimulation. This technique for distinguishing between habituation and fatigue has some advantages over the use of dishabituation because the relationship between habituation and dishabituation is not currently known, and in genetic/mechanistic studies of habituation there needs to be a way of demonstrating habituation in a system that does not dishabituate. The second important outcome of the ISI dependence of short-term habituation and spontaneous recovery was a prediction of mechanism. Previously, short-term habituation was assumed to be mediated by a single underlying mechanism. However, with the discovery that habituation and spontaneous recovery are different at various ISIs, Rankin and Broster hypothesized that habituation is mediated by multiple mechanisms that are differentially activated by distinct stimulation parameters (Rankin and Broster, 1992). Some mechanisms will be common to all stimulation parameters, while others will be activated by specific frequencies of stimulation. This was the first clue that the apparent simplicity of habituation may be misleading, and that habituation may in fact be mediated by complex sets of molecular mechanisms. The first step in identifying these molecular mechanisms was to establish the location of plasticity within the worm’s nervous system (Wicks and Rankin, 1995). To do this, Wicks and Rankin first identified the tap withdrawal circuit by using the known neuroanatomy and connectivity that had been described using serial electron microscopy of sections through the entire body of C. elegans (White et al., 1986) in combination with the circuit analysis that had already been carried out for the response to anterior and posterior body touch (Chalfie et al., 1985). Wicks and Rankin methodically ablated individual neurons and combinations of neurons identified as suitable candidates by these earlier 33  studies, and assessed how these ablations affected the worm’s response to tap. Using this approach, they identified 7 mechanosensory neurons (3 anterior, 2 posterior and 2 full body), and 10 interneurons responsible for the tap response (Figure 1.1; Figure 1.2). Once this circuit was identified, experiments could be generated to pinpoint the specific site of plasticity within the circuit (Wicks and Rankin, 1997). Wicks and Rankin found that there was incomplete overlap between the neurons responsible for the tap withdrawal response and those responsible for spontaneous reversals. The command interneurons are important for both behaviours, for example, laser ablation of the AVA interneurons disrupts both spontaneous reversals and tap evoked reversals (Chalfie et al., 1985; Wicks and Rankin, 1995; Gray et al., 2005). The mechanosensory neurons (ALMs, AVM and PLMs) are exclusive to the tap circuitry, for example, ablation of these neurons only disrupted tap evoked reversals (Wicks and Rankin, 1995, 1997) and not spontaneous reversals. If tap habituation affected spontaneous reversals then the site of plasticity must be located in part of the overlapping circuit, but if tap habituation had no effect on spontaneous reversals then the site of plasticity must be located in part of the circuit that does not overlap. The latter was found to be the case; habituation of the tap withdrawal response had no effect on spontaneous reversals. The neural circuit that mediates noxious heat avoidance may also only partially overlap with the tap circuit. Although the circuit has not yet been fully described, the tap sensory neurons have not been reported to affect any thermosensory behaviour despite thorough investigation (Mori and Ohshima, 1995; Biron et al., 2008; Kuhara et al., 2008; Beverly et al., 2011) suggesting the heat avoidance circuit utilizes different sensory neurons. Tap habituation had no effect on the heat avoidance response (Wicks and Rankin, 1997). Taken together, these data suggest that the major site of plasticity for short-term habituation is located in the portion of the tap circuit that does not overlap with spontaneous reversal or heat avoidance circuits, the seven mechanosensory neurons and their synaptic connections onto the interneurons.  34  Figure 1.1 Neural circuit of the tap withdrawal response The neural circuit that mediates the tap withdrawal response is thought to be an integration of the anterior touch circuit, which drives a backward locomotion (observed as a reversal response), and the posterior touch circuit, which drives forward locomotion (observed as an acceleration), and a few other neurons which help to bias the circuit towards a backwards locomotion when both sides of the circuit are stimulated at once (which occurs during a tap stimulus) (Wicks and Rankin, 1995). Neurons implicated in the anterior touch circuit are shaded in blue. Neurons implicated in the posterior touch circuit are shaded in red. Additional neurons implicated in the tap response are shaded in grey. Dopamine has been implicated in the plasticity of the circuit (Sanyal et al., 2004), so the dopamine neurons have also been included and are yellow diamonds. Sensory neurons are indicated as squares. Interneurons are indicated as circles. The set of motor neurons that carry out the forward or backwards locomotion programs are represented by triangles. Synaptic connectivity between neurons are represented by joining lines; dashed lines represent electrical synapses and solid arrow lines represent chemical synapses (White et al., 1986). Green lines are hypothesized to be excitatory and red lines are hypothesized to be inhibitory; connections where the nature of the chemical synapse is unclear are shown in black. Almost all interneurons have reciprocal synaptic connections with all other interneurons shown; these have been omitted from the figure for clarity. There are no direct synaptic connections from the dopamine neurons to the neurons in the tap circuit, so the yellow pie-shaped gradients represent extrasynaptic signaling by dopamine. A simplified circuit and representative subcircuit are represented in Figure 1.2.  35  36  Figure 1.2 Anatomy of tap withdrawal response with molecules implicated in habituation Both sides of the tap withdrawal circuit (anterior and posterior) are almost mirror images and are thought to undergo the plasticity similarly since both sides of the circuit habituate, albeit not at the same rate (Wicks and Rankin, 1996). This suggests that the tap withdrawal circuit can be simplified (a) as sensory neurons that have excitatory synapses onto the interneurons that promote locomotion in the appropriate direction and inhibitory synapses onto the interneurons that drive locomotion in the opposite direction. A sub-circuit of the entire tap withdrawal circuit (b) that represents this simplified model includes the anterior touch sensory neuron (ALML) which has excitatory chemical synapses onto an interneuron that promotes backwards locomotion (AVDR) and inhibitory chemical synapses onto an interneuron that promotes forward locomotion (PVCL) (White et al., 1986). ALML also expresses the dopamine receptor gene dop-1 (Sanyal et al., 2004) suggesting that it receives extrasynaptic signals from the closely neighbouring dopaminergic neuron (CEPVL). The lower panel details the neuroanatomy of this sub-circuit. (c) The head of C. elegans is shown emphasizing the location of the nerve ring and ventral nerve cord (light brown), where most synaptic connections are found in the worm. ALML (green) has a sensory process along the left anterior body wall and projects into the nerve ring. AVDR (blue) has a cell body in the head (circle), receives input in the nerve ring and projects down the ventral nervous cord to drive motor neurons along the body for backwards locomotion. PVCL (red) lies along a similar path, although its cell body is found in the tail and it drives motor neurons responsible for forward locomotion. CEPVL (yellow) has a cell body (circle) in the head, projects a sensory process to the amphid sensillum (nose of the worm) and an axon that projects into the nerve ring. (d) A closer view looking through the nerve ring from the top of the worm. (e) A section of the nerve ring to show how the four neurons of this sub-circuit connect and putative molecular components that have been implicated in habituation of the tap withdrawal response. dop-1 encodes a dopamine receptor, kht-1 encodes a potassium channel subunit, mps-1 encodes a potassium channel accessory subunit and eat-4 encodes a vesicular glutamate transporter; ?s represent some of the unknown components. 37  38  Three of these sensory neurons are activated by anterior touch, while two others are stimulated by posterior touch, suggesting that the tap stimulus is activating both pools of neurons and the information is being integrated with an apparent bias to the anterior neurons in order to lead to a reversal response (similar to the anterior touch withdrawal response, but much shorter). The other two sensory neurons (PVDs) are thought to be involved in setting this bias. By ablating either the anterior (ALMs and AVM) or posterior (PLMs) portion of the circuit, each part could be individually studied. When this was done, Wicks and Rankin found that the anterior head touch portion of the circuit habituated more rapidly than the posterior, tail touch portion of the circuit (Wicks and Rankin, 1996). The rate of habituation of the intact circuit reflects an integration of the different rates of habituation of the head and tail touch sub-circuits. In order to study the anterior and posterior circuits in isolation, gentle body touch to either the anterior or posterior body wall needs to be used instead of tap, which is very laborious. Heroically, Kitamura et al. performed a circuit analysis for habituation of anterior body touch by ablating each class of the anterior sensory neurons (ALM and AVM) and each of pair of the interneurons (AVD, PVC, AVA, AVB) (Kitamura et al., 2001). Interestingly, only ablation of AVD neurons affected the rate of habituation suggesting that AVD is an important site of plasticity for the anterior portion of the tap withdrawal circuit. Taken together with the site of plasticity results from generalization studies already described (Wicks and Rankin, 1997), this suggests that the synapses between the mechanosensory neurons and AVD may be a major site of plasticity during habituation (Figure 1.2). With a potential site of plasticity identified, a candidate gene approach could now be used to investigate the molecular mechanisms of habituation. The mechanosensory neurons in the circuit are thought to be glutamatergic due to the presence of a vesicular glutamate transporter called EAT-4 (homologous to the mammalian VGlut1 transporter), which is responsible for loading synaptic vesicles with glutamate ready for neurotransmission (Lee et al., 1999). Rankin and Wicks hypothesized that 39  glutamate neurotransmission from the mechanosensory neurons onto the interneurons would be a good candidate for a possible mechanism for habituation (Rankin and Wicks, 2000). They tested shortterm habituation in worms carrying a mutation in eat-4 (the gene that encodes the transporter) and found that at both a 10 second ISI and a 60 second ISI, the mutants habituated significantly more rapidly than wild-type animals. After 40 stimuli, however, wild-type worms habituate to almost the same asymptotic level as eat-4 mutants. This suggests that glutamate neurotransmission does play a role in the mechanisms for habituation (Figure 1.2). Initially, it was a concern that the eat-4 mutation was causing some form of sensory adaptation or fatigue as opposed to more rapid habituation. This concern was magnified by the fact that eat-4 mutants did not show dishabituation, the traditional test for adaptation/fatigue. However, using ISI-dependent spontaneous recover to distinguish habituation from adaptation and fatigue, Rankin and Wicks found that these mutants showed more rapid spontaneous recovery following habituation at a 10 second ISI than at a 60 second ISI indicating that the rapid decrement observed in eat-4 mutants was in fact caused by habituation. Therefore, the lack of dishabituation observed in the eat-4 mutants can be instead attributed to a critical role of the glutamate vesicular transporter in the mechanism of dishabituation. Knowing the neural circuit for the tap response allowed for the identification and testing of additional candidate genes expressed in the neurons of the circuit. One such candidate, a gene for a D1like dopamine receptor homolog called dop-1 is expressed in the mechanosensory neurons of the tap withdrawal circuit (Tsalik et al., 2003; Sanyal et al., 2004). Dopamine is a neurotransmitter that often acts as a neuromodulator, so an obvious hypothesis was that dopamine might modulate the tap withdrawal circuit thereby playing a role in habituation of the tap withdrawal response. Dopaminedeficient mutants (cat-2) and dop-1 receptor mutants both habituate rapidly, implicating a role of dopamine in habituation (Sanyal et al., 2004)(Figure 1.2).  40  Using a similar rationale, a potassium channel accessory subunit with auto-phosphorylation capabilities known as MPS-1 was found to be expressed in the mechanosensory neurons of C. elegans (Bianchi et al., 2003), making it a possible candidate for mechanosensory function and plasticity (Cai et al., 2009). To investigate its role, Cai et al. screened a number of potassium channel subunits to find a potential partner for MPS-1 and found that KHT-1, a homolog to the human Kv3.1 potassium channel subunit, co-expressed, co-localized and co-immunoprecipitated with MPS-1 in vitro and in cultured C. elegans embryonic mechanosensory neurons, strongly suggesting they form a channel together in vivo. Null mutants for both mps-1 and kht-1 genes had significantly smaller touch- and tap-evoked responses compared to wild-type animals. Transgenic expression in the mps-1 null mutants of either wild-type mps-1 or a mutant version of mps-1, which has a point mutation (D178N) that inactivates its kinase activity, restored the normal mechanosensory responses. Interestingly, the tap habituation of these kinase-inactive transgenics was significantly slower than the wild-type expressing transgenics. This suggests that the potassium channels formed by KHT-1 and MPS-1 complexes are necessary for normal response to tap, and the kinase activity of MPS-1 is necessary for normal rates of tap habituation in C. elegans (Figure 1.2). Electrophysiological experiments in cultured mechanosensory neurons implicated KHT-1 and MPS-1 containing potassium channels in mediating voltage-gated potassium current that repolarize neurons after activation; kht-1 and mps-1 mutants had impaired repolarizing currents. The authors suggest that this may explain the effect on the initial behavioural response to tap in the single mutants, but it remains unclear how the currents mediated by this potassium channel regulates habituation of the tap response. 1.1.3. Summary of mechanism for habituation The clues that C. elegans, Aplysia, Drosophila, Zebrafish, birds and rodents have revealed about the mechanisms of habituation are invaluable. Taken together, it is easy to speculate that habituation often involves some type of synaptic plasticity (Table 1.1) at some point in the circuitry between receptor and effector. Some of the molecules that might mediate these synaptic changes include 41  neurotransmitter receptors, second messenger signalling molecules, kinases and phosphatases, ion channels and transcription factors (Table 1.2). However, all the molecular components have not yet been identified and how these components work together in a functional mechanism to explain both the observed cellular and behavioural plasticity of habituation is still an unanswered question of great importance in the field of neuroscience.  42  Table 1.1 Summary of cellular mechanisms of habituation Mechanism Depression of Excitatory Synapses  Potentiation of Inhibitory Synapses  Organism Aplysia  Behaviour Gill Withdrawal Reflex  C. elegans Rodent  Tap Withdrawal Response Acoustic/Tactile Startle Response  Crayfish Drosophila  Tail-flip Escape Response Olfactory Avoidance  Evidence (Castellucci et al., 1970; Castellucci and Kandel, 1974; Bailey and Chen, 1988; Gover et al., 2002) (Rankin and Wicks, 2000) (Davis et al., 1982b; Lingenhohl and Friauf, 1994; SimonsWeidenmaier et al., 2006) (Krasne, 1969) (Das et al., 2011)  43  Table 1.2 Summary of molecular components for the mechanism of habituation Component  Organism  Neurotransmitters/Receptors: VGlut Drosophila (vesicular glutamate transporter)  Behaviour  Olfactory Avoidance  eat-4 (vesicular glutamate transporter) amnesiac (neuropeptide precursor) NR1 (NMDA receptor subunit) NMDA-type glutamate receptor  C. elegans  Tap Withdrawal Response  Drosophila  Landing response  Drosophila  Olfactory Avoidance  Zebrafish  Acoustic startle  Metabotropic glutamate receptor  Rat  Rdl (GABA-A receptor subunit) GABA-A receptor subunit  Drosophila  Heart rate orienting response Olfactory Avoidance  Zebrafish  Acoustic startle  Nicotinic-type acetylcholine receptors  Zebrafish  Acoustic startle  dop-1 (dopamine receptor)  C. elegans  Tap Withdrawal Response  D3 dopamine receptor  Zebrafish  Acoustic startle  Effect  Knockout in LNs or GAD1-expressing neurons blocks habituation Knockdown increases habituation Mutation causes rapid habituation Knockdown of NR1 in PNs stops habituations Pharmacological blockers slowed habituation Pharmacological antagonists block habituation Knockdown of Rdl in PNs stops habituations Pharmacological blockers slowed habituation; positive modulators caused rapid habituation Pharmacological blockers slowed habituation Knockdown increases habituation Pharmacological blockers slowed habituation; positive modulators caused rapid habituation  Evidence  (Das et al., 2011)  (Rankin and Wicks, 2000) (Wittekind and Spatz, 1988; Rees and Spatz, 1989) (Das et al., 2011)  (Best et al., 2008; Wolman et al., 2011) (Best and Wilson, 2004; Best et al., 2005) (Das et al., 2011)  (Wolman et al., 2011)  (Best et al., 2008)  (Sanyal et al., 2004)  (Wolman et al., 2011)  44  Component Cell Signaling: calcium  rutabega (adenylyl cyclase) -produces cAMP  Organism  Behaviour  Effect  Evidence  Aplysia  Gill Withdrawal Response  (Gover et al., 2002)  Drosophila  Giant Fibre Response Olfactory Jump Response Locomotory Startle Proboscis Extension Response Olfactory Avoidance  Removal of extracellular calcium decreases habituation Mutation slows habituation  Landing Response dunce (phosphodiesterase) -degrades cAMP  Drosophila  Giant Fiber Response  Landing Response  Proboscis Extension Response Olfactory Jump Response  (Engel and Wu, 1996) (Asztalos et al., 2007a) (Cho et al., 2004b) (Duerr and Quinn, 1982)  Knockout in LNs or GAD1-expressing neurons blocks habituation Mutation makes habituation faster Mutation that decreases enzyme activity habituated rapidly Mutation that decreases enzyme activity habituate rapidly Mutation that decreases enzyme activity slows habituation Mutation that decreases enzyme activity slows habituation  (Das et al., 2011)  (Wittekind and Spatz, 1988; Rees and Spatz, 1989) (Engel and Wu, 1996)  (Wittekind and Spatz, 1988; Rees and Spatz, 1989) (Duerr and Quinn, 1982)  (Asztalos et al., 2007a)  45  Component  Organism  Behaviour  Kinases/Phosphatases/Phosphoproteins: CaMKII Drosophila Leg extension (calcium-calmodulin reflex dependent kinase II) Shaggy Drosophila Olfactory startle (glycogen synthase kinase 3)  Effect  Pp1-87B (protein phosphatase 1)  Drosophila  Landing response  fickle (tyrosine kinase) synapsin (synaptic vesicle phosphoprotein) caki (membrane-associated guanylate kinase)  Drosophila  Olfactory Jump Response Olfactory Jump Response  Constitutively activate mutation blocks habituation Low expression mutation caused slow habituation; over expression caused rapid habituation Mutation that increases kinase activity caused slow habituation Mutation that increases kinase activity caused slow habituation Mutation that decreases phosphatase activity caused rapid habituation Mutation causes rapid habituation Mutation causes rapid habituation  Giant Fiber Response  Mutation causes rapid habituation  foraging (cGMP-dependent protein kinase)  Drosophila  Giant Fiber Response  Proboscis Extension Response  Drosophila  Drosophila  Evidence  (Jin et al., 1998)  (Zordan et al., 2005)  (Wolf et al., 2007)  (Engel et al., 2000)  (Scheiner et al., 2004)  (Asztalos et al., 1993)  (Asztalos et al., 2007b) (Godenschwege et al., 2004)  46  Component Potassium Channels: Slowpoke (calcium-activated potassium channel subunit) Shaker (voltage-gated potassium channel alpha subunit)  Hyperkinetic (voltage-gated potassium channel beta subunit) Ether a go go (pore-forming potassium channel subunit) mps-1 (kht-1-containingpotassium channel accessory subunit) Transcription Factors: ZENK (transcription factor)  period (transcription factor)  Organism  Behaviour  Drosophila  Giant Fibre Response Olfactory Jump Response Giant Fibre Response  Drosophila  Olfactory Jump Response Drosophila  Drosophila  C. elegans  Giant Fibre Response Olfactory Jump Response Giant Fibre Response Olfactory Jump Response Tap Withdrawal Response  Song bird  Presentation of bird song  Drosophila  Giant Fiber Response  Effect  Knockout slows habituation  Evidence  (Engel and Wu, 1998) (Joiner et al., 2007)  Altered function mutation causes rapid habituation Altered function mutation causes slow habituation Loss of function causes slow habituation  (Engel and Wu, 1998)  Loss of function causes rapid habituation  (Engel and Wu, 1998)  Mutation that blocks kinase activity slows habituation  (Cai et al., 2009)  ZENK expression increases after repeated presentation Mutation alters interstimulus interval parameters of habituation  (Joiner et al., 2007)  (Engel and Wu, 1998) (Joiner et al., 2007)  (Joiner et al., 2007)  (Kruse et al., 2000)  (Megighian et al., 2001)  47  1.2. Objectives The objectives of my dissertation included three major research goals. The first was to investigate the mechanism of habituation using a candidate gene approach. I did this by investigating the role of the dopaminergic nervous system of C. elegans in habituation to extend published evidence that suggested that disruptions in dopamine neurotransmission result in abnormal habituation in C. elegans (Sanyal et al., 2004). In collaboration with the Schafer Lab, I found that dopamine’s modulation of habituation is food-dependent and identified potential downstream intracellular signalling molecules through which dopamine exerts its effect (Kindt et al., 2007). The second goal of my dissertation was to take a high-throughput approach to screen through many mutant with the hope of identifying novel molecules that may be involved in the mechanism of habituation. In order to accomplish this task, however, a new behavioural tracking system needed to be developed to significantly decrease the time that it takes to assay habituation in C. elegans. In collaboration with the Kerr Lab, I helped develop and validate a system capable of the throughput needed for such a task (Swierczek et al., 2011). The third objective of my dissertation was to use this novel behavioural tracking system to screen a set of mutants. However, the system was even better than I had initially anticipated, so instead of simply screening the mutants, I characterized the habituation (as well as a few other phenotypes) of 522 mutants. For two of the novel mutants, which affected the eat-16 and goa-1 genes, I analyzed further alleles of these genes to support their role in habituation. Finally, I conducted phenotypic profiling using this rich dataset of 522 mutants in order to predict genetic interactions.  48  2. Role of dopamine in habituation 2.1. Introduction In order to further investigate the molecular mechanisms involved in the neural plasticity that underlies habituation of the tap withdrawal response in C. elegans, we sought to explore molecules whose expression within the neural circuit mediating the response highlights them as potential candidates. The neural circuit that controls the tap withdrawal response in C. elegans is primarily composed of five mechanosensory neurons, four pairs of command interneurons and pools of motorneurons (Wicks and Rankin, 1995). A D1-like dopamine receptor homolog, dop-1, is expressed in the mechanosensory neurons of the tap withdrawal circuit (Sanyal et al., 2004) making it a potential candidate for influencing learning and memory. There are many examples of dopamine playing important roles in neural and behavioural plasticity. For example, dopamine is implicated in simple appetitive conditioning (Young et al., 2005), novelty detection (Redgrave and Gurney, 2006), goaldirected behaviours (Grace et al., 2007), and incentive learning and memory (Phillips et al., 2008). In C. elegans, the dopaminergic nervous system has been well characterized. It consists of 8 dopamine producing neurons (CEPDL, CEPDR, CEPVL, CEPVR; ADEL, ADER; and PDEL, PDER). The enzyme that catalyzes the rate-limiting step of dopamine synthesis is encoded by the gene cat-2 (Lints and Emmons, 1999). Mutations that knock out this gene decrease the levels of dopamine in the organism to approximately 40% of wild-type levels (Sanyal et al., 2004). Other genes important for dopamine biosynthesis in C. elegans include bas-1, aromatic amino acid decarboxylase, and cat-4, an enzyme that synthesizes a cofactor necessary for the function of CAT-2; however, neither are specific to dopamine synthesis, disrupting the serotonin and tyramine pathways as well. A non-specific monoamine vesicular transporter, cat-1, loads dopamine and other aminergic neurotransmitters into synaptic vesicles. Dopamine release is thought to be both synaptic and hormonal, allowing dopamine to affect cell targets that are not necessarily adjacent to the releasing neurons (Chase et al., 2004). Four  49  dopamine receptors have been identified in C. elegans: dop-1, mentioned above, dop-2, a D2-like dopamine receptor homolog, expressed on the dopaminergic neurons suggesting a role for autoregulation (Suo et al., 2003), and dop-3 and dop-4, which are not expressed on neurons in the tap withdrawal circuit (Chase et al., 2004; Sugiura et al., 2005). Two other genes, T02E9.3 and C24A8.1 (named dop-5 and dop-6; www.wormbase.org) are paralogous to these four receptors(Tsalik et al., 2003); however, since they are also paralogous to the C. elegans serotonin, tyramine and octopamine receptors (Tsalik et al., 2003) and dop-5 partially phenocopies some serotonin receptor mutants (CarrePierrat et al., 2006), it is unclear of their specificity. Finally, the transporter responsible for reuptake of dopamine from the extracellular volume after neurotransmitter release is encoded by the gene dat-1 (Jayanthi et al., 1998). When this gene is disrupted, dopamine neurotransmission is increased because dopamine is not cleared from the synapse properly leading to longer, more sustained neurotransmission. dat-1 mutants show behaviours consistent with hyperdopaminergic function (McDonald et al., 2007). In other organisms, some neural plasticity occurs to compensate for dopamine transporter disruptions, however, it is not known to what extent this occurs in C. elegans (Nass and Blakely, 2003). The role of dopamine has been studied in a number of behaviours in C. elegans, for example, locomotion, including slowing in response to food (Sawin et al., 2000), and the transition between swimming and crawling (Vidal-Gadea et al., 2011), foraging strategy (Hills et al., 2004), and egg laying behaviours (Schafer and Kenyon, 1995). Dopamine receptor (dop-1) and dopamine deficient (cat-2) mutants have more rapid habituation to the tap withdrawal response than wild-type worms (Sanyal et al., 2004). Dopamine is probably acting on the mechanosensory neurons because expression of the wild-type DOP-1 receptor, specifically in the mechanosensory neurons of dop-1 mutants, reverts the habituation rate back to wild-type (Sanyal et al., 2004). Dopamine supplement of the cat-2 dopamine deficit mutants specifically in adulthood prior to testing also rescued the wild-type habituation rate suggesting the effect is not indirectly caused by  50  developmental plasticity. Taken together, this suggests a strong role for dopamine and the dop-1 receptor in the mechanism of habituation. In this series of experiments our goal was to elucidate the cellular and molecular mechanism by which dopamine alters tap habituation by characterizing the behavioural parameters necessary for the effect, testing candidate genes downstream of dopamine receptor signalling and exploring the physiology of the mechanosensory neurons as habituation occurs. We discovered that the dopamine effect is dependent on the presence of food. Wild-type animals habituated at different rates in the absence or presence of E. coli (their food) and the dopamine effect was only apparent when worms were tested in the presence of food. A neural correlate of habituation in the mechanosensory neurons was also dependent on the presence of food and this dependence was also mediated by dopamine.  2.2. Results 2.2.1. Food-dependent modulation of habituation is mediated by dopamine In order to further explore the mechanism for dopamine’s modulation of habituation, we attempted to replicate the cat-2 and dop-1 mutant effect on habituation observed previously (Sanyal et al., 2004). We stimulated individual animals with 30 taps to the side of the Petri plate at a 10 second inter-stimulus interval and recorded the percent of animals that responded to each of the 30 stimuli. We did not observe any differences between the mutants and wild-type worms (Figure 2.1). After comparing the details of our behavioural protocol with those of the Schafer Lab, who conducted the original experiments (Sanyal et al., 2004), we identified four differences: 1) the stimulus, 2) the behavioural scoring, 3) the wild-type strain, and 3) the presence of food during the habituation test.  51  Figure 2.1 Manual Assay: dopamine deficient effect on habituation is not apparent in the absence of E. coli Wild-type (N2) and mutants C. elegans were presented with 30 tap stimuli at a 10 s inter-stimulus interval. Worms were tested individually in the absence of E. coli using manual video recording and behavioural scoring by human inspection. Probability of response (+/- standard error of proportion) was measured for each stimulus. (a) Neither the initial response nor habituation of dopamine deficient mutants (cat-2) was statistically different from wild-type (n.s.). (b) Neither the initial response nor habituation of dopamine receptor mutants (dop-1) was statistically different from wild-type (n.s.). (c) The initial response of dopamine transporter mutants (dat-1) was not significantly different compared to wild-type (n.s.), but dat-1 mutants habituated significantly more slowly (*p < 0.05; i.e. the rank sum of the last 29 responses was significantly higher than wild-type; see methods for details).  52  53  To generate the tap stimulus, we used a magnetic relay to drive a metal rod into the side of the Petri plate, while the Schafer lab used a plastic arm lifted by an electric motor and dropped under the force of gravity to tap the top of the plate. Although these stimuli are similar, it is possible that differences in the stimulus properties led to different rates of habituation. Stimulus intensity is known to be an important factor in the rate of habituation (Thompson and Spencer, 1966; Rankin et al., 2009) (Timber, Giles et al., unpublished). Response to tap was scored by human inspection in our lab, while the Schafer lab used custom designed computer-vision software to detect responses. We found some differences between these two methods of analysis. For example, a pause in forward locomotion after a stimulus was scored as a response by human inspection, but this was not always detected by the automated software. These cases were rare, however, and did not account for the discrepancy between our results. Both labs maintain their own line of the N2 wild-type strain. It is unclear exactly how long each of these strains had diverged from the N2 strain stored at the Caenorhabditis Genetics Centre, but it was likely in the order of years. It is possible that the strains had genetically drifted from one another causing the observed difference. Finally, most of the behaviours that dopamine has been found to modulate are either food-related or food-dependent, such as the basal slowing response. Wild-type animals slow their locomotion when entering a bacterial lawn and dopamine deficient mutants do not slow their locomotion on the lawn and move the same speed as wild-type off of the lawn (Sawin et al., 2000). The Schafer lab tested their animals in the presence of food, while we tested our worms in the absence of food. We hypothesized that the stimulus, analysis method, and genetic variation between different wild-type strains were not the major cause of the difference between our results and those observed by the Schafer lab. Instead we hypothesized that the presence of food is likely the important parameter for dopamine’s effect on habituation. To test this hypothesis we conducted experiments using the same apparatus (Schafer Lab) with the same wild-type animals (Schafer Lab), but varied the presence of E.coli during habituation. We found that wild-type worms habituated significantly more slowly in the 54  presence of food than in the absence (Figure 2.2). We further hypothesized that this food-dependent modulation of habituation is mediated by dopamine. dop-1 mutants habituated more rapidly than wildtype in the presence of food (Figure 2.2); however, in the absence of food, dop-1 mutants habituated at the same rate as wild-type (Figure 2.2). We have recently replicated this finding with a different apparatus (the Multi-Worm Tracker, described in chapters 3 and 4) using a more recently diverged N2 stock (Figure 2.3) providing further evidence that the difference was not the apparatus or wild-type strain. This supports our hypothesis that habituation is modulated by the presence of food and that this modulation is dependent on dopamine neurotransmission through the dop-1 receptor. A loss of function mutation in the dopamine transporter, dat-1, responsible for the reuptake of dopamine after vesicular release, is thought to increase the levels of extrasynaptic dopamine since dat-1 mutants show behaviours consistent with hyperdopaminergic function (McDonald et al., 2007). To further evaluate dopamine’s role in the modulation of habituation by food, we tested dat-1 mutants. We found that dat-1 mutants habituated similarly to wild-type in the presence of food, but in the absence of food habituated significantly more slowly (Figure 2.1; Figure 2.4). This suggests that excess dopamine neurotransmission when dopamine is usually present (in the presence of food) does not alter habituation further; however, excess dopamine when dopamine is usually absent (the absence of food) causes a phenotype similar to when animals are in the presence of food. Recently the habituation of dat-1 mutants was tested using the Multi-Worm Tracker. We found that dat-1 mutants habituated more slowly in the presence of food (Figure 2.3). It is unclear why we were unable to detect this initially. One possibility is that the properties of the stimulus in the Schafer lab lead to slower rates of habituation compared to those generated by the Multi-Worm Tracker, which might create a ceiling effect, masking the dat-1 phenotype in the presence of food. Regardless, the evidence suggests that habituation of the tap withdrawal response is modulated by dopamine neurotransmission.  55  Figure 2.2 Single-Worm Tracker: habituation is dependent on the presence of E. coli and is dopamine receptor dependent Wild-type (N2) and dopamine receptor (dop-1) mutant C. elegans were presented with 30 tap stimuli at a 10 s inter-stimulus interval in the presence (on food) and absence (off food) of E. coli. Behaviour was recorded using a single-worm tracker and scored by automated computer algorithms. Probability of response (+/- standard error of proportion) was measured for each stimulus. (a) The initial response of wild-type (N2) worms was not significantly different in the presence or absence of food (n.s.), but worms habituated significantly more rapidly in the absence of food (*p < 0.05; i.e. the rank sum of the last 29 responses was significantly lower off food; see methods for details). (b) In the presence of food, the initial response of dopamine receptor mutants (dop-1) was not significantly different to wild-type (n.s.), but dop-1 mutants habituated significantly faster (*p < 0.05; i.e. the rank sum of the last 29 responses was significantly lower than wild-type; see methods for details). (c) In the absence of food, the neither the initial response nor habituation of dopamine receptor mutants (dop-1) was statistically distinguishable from wild-type (n.s.).  56  57  Figure 2.3 Multi-Worm Tracker: habituation is dependent on the presence of E. coli and is dopamine receptor dependent Wild-type (N2) and mutant C. elegans were presented with 30 tap stimuli at a 10 s inter-stimulus interval. Behaviour was recorded and scored using the Multi-Worm Tracker. Probability of response (+/- standard error of proportion) was measured for each stimulus. (a) The initial response of wild-type (N2) worms was significantly lower in the absence of E. coli (on food; *p < 0.05) than in the presence of E. coli (off food), and wild-type worms habituated significantly more rapidly in the absence of E. coli (#p < 0.05; i.e. the rank sum of the last 29 responses was significantly lower off food using raw data or when standardizing to the initial response; see methods for details). (b) C. elegans were tested in the presence of E. coli. Dopamine deficient mutants (cat-2) had significantly smaller initial responses probability than wild-type (^p < 0.05), while dopamine receptor mutants (dop-1) and dopamine transporter mutants (dat-1) had significantly larger initial response probabilities than wild-type (#,*p < 0.05). cat-2 and dop-1 mutants habituated significantly more rapidly than wild-type (^^,##p < 0.05; i.e. the rank sum of the last 29 responses was significantly lower than wild-type using raw data or when standardizing to the initial response; see methods for details). dat-1 mutants habituated significantly more slowly than wild-type (**p < 0.05; i.e. the rank sum of the last 29 responses was significantly lower than wild-type using raw data or when standardizing to the initial response; see methods for details).  58  59  Figure 2.4 Dopamine transport mutation slows habituation in the absence of food Wild-type (N2) and dopamine transporter (dat-1) mutant C. elegans were presented with 30 tap stimuli at a 10 s inter-stimulus interval in the presence (on food) and absence (off food) of E. coli. Behaviour was recorded using a single-worm tracker and scored by automated computer algorithms. Probability of response (+/- standard error of proportion) was measured for each stimulus. (a) Neither the initial response nor habituation of dopamine transporter mutants (dat-1) was significantly different from wild-type in the presence of food (n.s.). (b) In the absence of food, the initial response of dat-1 mutants was not significantly different from wild-type (n.s.), but dat-1 mutants habituated significantly more rapidly than wild-type (*p < 0.05; i.e. the rank sum of the last 29 responses was significantly higher than wild-type; see methods for details).  60  61  2.2.2. Modulation of a neural correlate of habituation by dopamine is food dependent In C. elegans, neural excitation propagates throughout the extensions and cell bodies of neurons by graded potentials amplified by voltage-gated calcium channels, which results in large influxes of calcium within neurons during excitation (Goodman et al., 1998). Using in vivo, fluorescent resonant energy transfer (FRET) microscopy of transgenic strains expressing the calcium-sensitive protein cameleon (YC3.12) behind a promoter specific to the mechanosensory neurons (mec-4p), it is possible to measure these calcium transients in the cell bodies of ALM or PLM neurons in response to mechanical stimuli (Suzuki et al., 2003). The canonical tap stimulus was not possible to administer while conducting this imaging because the tap would bounce the cell out of the field of view or the plane of focus. Instead, a flame-blunted glass capillary was used to vibrate against the cuticle of the worm in order to provide mechanosensory stimulation, simulating the tap stimulus. The anterior or posterior body was stimulated when recording from ALM or PLM, respectively, due to the extent of their receptor fields (Chalfie et al., 1985; Suzuki et al., 2003). Worms were stimulated at a 10 second inter-stimulus interval. Image quality deteriorated variably after the 13th stimulus due to photobleaching of the sample and sample movement during the experiment. Much of the decrement observed in our behaviour experiments occurs during the first 15 stimuli and the food- and dopamine-related effects can be detected by this point, so 13 stimuli were given per experiment to ensure image quality during experiments. The calcium response was measured as the peak increase in the FRET ratio from the baseline before each stimulus. The calcium response within the ALM neuron decreased with repeated stimulation (Figure 2.5), as reported previously (Suzuki et al., 2003).  62  Figure 2.5 Modulation by dopamine of a neural correlate of habituation in sensory neurons is dependent on the presence of food Transgenic worms expressing the genetically encoded calcium indicator, Cameleon, were stimulated 13 times with a vibrating blunted glass needle on the anterior body at a 10 s inter-stimulus interval while FRET imaging of the ALM sensory neuron was recorded. The mean (+/- SEM) change in the FRET ratio in response to stimulation for the first stimulus and following stimuli (binned by 3 stimuli) was measured. (a) Transgenic worms with either a wild-type dopamine receptor gene (control) or a mutant dopamine receptor gene (dop-1) were tested in the presence of food (On food). The initial response was not significantly different between dop-1 mutants and controls, but dop-1 mutants had significantly smaller responses during later responses suggesting that their response decrements more rapidly than controls (*p < 0.05; **p < 0.05 after a Bonferroni correction for five comparisons). (b) Transgenic worms with either a wild-type dopamine receptor gene (control) or a mutant dopamine receptor gene (dop-1) were tested in the absence of food (Off food). None of the responses were significantly different between dop-1 mutants and controls. (c) Transgenic worms with either a wildtype dopamine transporter gene (control) or a mutant dopamine receptor gene (dat-1) were tested in the presence of food (On food). The initial response was not significantly different between dat-1 mutants and controls, but dat-1 mutants had significantly larger responses during later responses suggesting that their response decrements more slowly than controls (*p < 0.05; **p < 0.05 after a Bonferroni correction for five comparisons).  63  64  To determine how the DOP-1 receptor might act in the touch neurons to modify their functional properties, the Schafer lab tested the calcium response decrement for dop-1 mutants. The initial response of dop-1 mutants was similar to control worms; however, the calcium transients of dop-1 mutants decremented significantly more rapidly than control (Figure 2.5) (Kindt et al., 2007), mimicking the behavioural effect that we had observed. We were interested in knowing whether, like the behavioural effect, this dopamine-dependent neural correlate of habituation was sensitive to the presence of food. The previous experiments were conducted with a small amount of E. coli in the buffer in which the worms were immersed during imaging. When E. coli was not added to the buffer and worms were rinsed prior to experimentation to ensure an absence of food, dop-1 mutants significantly indistinguishable from control worms (Figure 2.5). Together these results paralleled the food-dependent fast habituation phenotype observed at the level of behaviour, and indicate that the effect of dopamine signalling on touch habituation occurs at a step between touch sensation and cell body calcium entry. Seeing as previous electrophysiological data indicated that the mechanotransduction apparatus does not undergo sensory adaptation (O'Hagan et al., 2005), these results are consistent with the possibility that dopamine modulates habituationinduced changes in sensory neuron excitability when food is present. In principle, the tap habituation phenotype of dop-1 could also involve changes in the responses of the posterior touch receptor neurons (PLMs), which are also activated by tap and trigger accelerations rather than reversals. When PLM neurons were imaged during repeated stimulation, the rate at which touch induced calcium transients were attenuated in PLM was not altered in dop-1 mutants compared to controls (data not shown)(Kindt et al., 2007). These results suggest that dopamine may be required specifically to modulate anterior, but not posterior touch neurons, and are consistent with previous ablation data that indicates the anterior and posterior avoidance responses may habituate through different mechanisms (Wicks and Rankin, 1996).  65  The dopamine transporter mutants, dat-1, which habituated more slowly than wild-type animals suggesting they are in a hyperdopaminergic state, were also tested with calcium imaging. Repeated stimulation of dat-1 mutants led to a slower decrement of the calcium response (Figure 2.5) in the anterior mechanosensory neuron ALM. This further supports the hypothesis that dopamine modulates the experience-dependent decrease in neural excitability of the mechanosensory neurons.  2.3. Discussion We have demonstrated that the rate of habituation of the tap withdrawal response in C. elegans is dependent on the presence of food. Worms habituate more rapidly when food was absent. This dependence is mediated by dopamine neurotransmission via the DOP-1 dopamine receptor expressed in the mechanosensory neurons of the tap circuit because dop-1 mutants and dopamine deficient mutants, cat-2, habituated more rapidly than wild-type animals only in the presence of food, extending previous results (Sanyal et al., 2004). This was supported by the observation that dat-1 hyperdopaminergic mutants habituated more slowly than wild-type. Similarly, the decrement of calcium responses evoked by repeated mechanical stimulation was more rapid than wild-type for dop-1 mutants and was slower for dat-1 mutants, suggesting that dopamine modulates the neural excitability of the anterior mechanosensory neurons of the tap withdrawal circuit, a neural correlate of habituation, extending previous results (Suzuki et al., 2003). This work identifies a novel characteristic of habituation that had not previously been described, food-dependence, and identifies a novel mechanism that contributes to habituation of the tap withdrawal response in C. elegans, dopamine modulated depression of neuron excitability of the sensory neurons ALMs. 2.3.1. Downstream signalling of the DOP-1 receptor To further investigate the molecular mechanism responsible for dopamine’s modulation of the mechanosensory neuron excitability through the dop-1 receptor, an extensive mutant analysis of potential downstream targets was performed (Kindt et al., 2007). DOP-1 is a G-protein coupled 66  receptor. Hence, the habituation of three G-protein alpha subunit mutants were tested, gsa-1 (GS), goa1, (GO), and egl-30 (Gq). egl-30 mutants phenocopied dop-1, habituating to 30 taps more rapidly that wild-type. Expression of the wild-type egl-30 using the mechanosensory neuron specific promoter (mec-4p) in the egl-30 mutant background partially restored the habituation phenotype to wild-type. Together this suggests that EGL-30 is the G-protein alpha subunit that couples to DOP-1 in the mechanosensory neurons. One of the putative downstream effectors of EGL-30 is phospholipase C beta, encoded in C. elegans by the gene egl-8 (Lackner et al., 1999). Like both dop-1 and egl-30 mutants, egl-8 mutants habituated more rapidly than wild-type. When activated, EGL-8 catalyzes the hydrolysis of phosphatidylinositol 4, 5-bisphosphate (PIP2) into the signalling molecules diacyl glycerol (DAG) and inositol trisphosphate (IP3). In many systems, DAG activates protein kinase C (PKC) (Carr et al., 2002; Yang and Kazanietz, 2003), for which there are three variants expressed in the nervous system in C. elegans, PKC-1, PKC-2 and TPA-1. Mutants for all three PKCs were tested for habituation of the tap withdrawal response and only pkc-1 mutants phenocopied dop-1, egl-30 and egl-8 (Kindt et al., 2007). Diacyl glycerol kinase, encoded by dgk-1, is thought to inhibit DAG signalling (Miller et al., 1999) (Nurrish et al., 1999). dgk-1 mutants had the opposite phenotype to egl-8 and pkc-1, habituating more slowly that wild-type (Kindt et al., 2007). Taken together, this suggests a putative signalling cascade through which dopamine mediates its modulation of habituation. Dopamine activates the DOP-1 receptor coupled to EGL-30. Once activated, EGL-30 in turn activates EGL-8, which produces DAG to trigger PKC1. Members of this potential signalling cascade were tested to see if they also had a similar phenotype for the excitability of the mechanosensory neurons. The calcium transients of egl-30 and pkc-1 mutants were measured in response to repeated stimulation. In both cases, the calcium responses decremented more rapidly than wild-type, similar to the dop-1 mutants (Kindt et al., 2007).  67  This provides further support that dopamine is signalling through this pathway to modulate habituation by affecting the neural excitability of the mechanosensory neurons. 2.3.2. The presence of food may activate dopamine neurons through mechanosensory receptors The dopamine neurons contain microtubules, suggestive of mechanosensory function. It has been suggested that the neurons directly sense bacteria by the texture a bacterial lawn produces against the surface of the animal (Sawin et al., 2000). TRP-4, a mechanosensensitive transient receptor potential channel (Li et al., 2006; Kang et al., 2010) is expressed in the anterior dopamine neurons (CEPs and ADEs) and when mutated leads to similar locomotion and habituation phenotypes seen in dopamine deficient mutants, cat-2 (Li et al., 2006; Kindt et al., 2007), suggesting that it may be the bacterial mechanosensor. By expressing the genetically encoded calcium sensor, Cameleon, in the dopamine neurons using the cat-1p promoter, calcium responses in the CEP neurons were observed in response to a vibrating glass capillary against the amphid sensilla at the tip of the worm’s nose, which is where the dendrites of the CEP neurons project (Figure 1.2)(Kindt et al., 2007). No calcium responses were observed in trp-4 mutants, supporting the mechanosensory nature of the TRP-4 channel in the dopamine neurons. It is difficult to acutely administer E. coli to the worm’s amphid during imaging because the worm needs to be submerged in buffer, so more direct evidence for TRP-4’s role as a mechanosensory of the texture of bacteria could not be achieved; however, it is hypothesized that this is the case (Kindt et al., 2007). 2.3.3. Dopamine neurons may act in a positive feedback loop with mechanosensory neurons of the tap circuit Tap stimuli may also activate dopamine neurons. Glass capillary vibration against the anterior body causes calcium responses in the CEP cell bodies; however, unlike the amphid stimulation, trp-4 mutants have normal responses compared to wild-type (Kindt et al., 2007). Synaptic connections exist from ALMs (two of the anterior sensory neurons) to CEPs (one set of dopamine neurons)(White et al., 68  1986) and a mutation in the gene mec-4, a component of the mechanosensory transduction machinery in the tap sensory neurons (Ernstrom and Chalfie, 2002), completely blocks the CEP calcium response when the anterior body is stimulated (Kindt et al., 2007). This suggests that mechanosensory stimuli to the body (not the amphid) activate the dopamine neurons via the mechanosensory neurons of the tap circuit, setting up a potential positive feedback loop, since dopamine release then modulates these mechanosensory neurons. It remains possible that tap activates the dopamine neurons directly since tap is a non-localized stimulus, unlike the vibrating glass capillary that was used for these experiments. Unfortunately with the current technology this is impossible to test because the calcium imaging is not amenable to the tap stimulus, as described earlier. Regardless, at least a portion of dopamine neuron activation is the result of stimulation of the mechanosensory neurons of the tap circuit. 2.3.4. Implications for dopamine release No direct evidence has been reported to explain how dopamine neuron activation by E. coli and mechanosensory stimulation leads to dopamine release at presynaptic terminals. However, our results suggest some hypotheses. In the absence of bacteria, there is likely little or no dopamine being released because when food is absent, dopamine deficient mutants have wild-type phenotypes for fooddependent behaviours such as locomotion (Sawin et al., 2000) and tap habituation (Sanyal et al., 2004)(this chapter). In contrast, dopamine neuron activation by stimulation of the tap sensory neurons potentially causes small but sub-threshold release of dopamine in the absence of food, because if dopamine was not released at all, then disrupting reuptake transporters should have no effect. This is not the case, dat-1 mutants habituated more slowly than wild-type in the absence of food, suggesting dopamine may be released during tap stimulation in the absence of food. This may be evoked by the synaptic connections between the tap sensory neurons and the dopamine neurons (White et al., 1986). The presence of bacteria is hypothesized to stimulate the dopamine neurons to release dopamine because this by itself is enough to change the dopamine-dependent locomotion (Sawin et al., 2000). Further experiments are needed to investigate whether activation of the dopamine neurons by tap 69  stimulation through the tap sensory neurons enhances this release further or if the food-dependent release causes a ceiling effect. An important question in the field of associative learning is what is encoded by dopamine (Young et al., 2005). In mammals, it was initially thought to be the signal for reward or positive reinforcement (Wise and Rompre, 1989). This was strongly supported by that fact that dopamine is released in the nucleus accumbens coincidentally with rewarding stimuli (Young, 1993). However, this hypothesis has been challenged by more recent studies that find that dopamine is also released during the presentation of aversive stimuli (Joseph et al., 2003) suggesting that dopamine cannot simply be the reward signal. More recent experiments investigating dopamine’s role in latent inhibition suggest that dopamine may instead signal salience (Young et al., 2005). Since dopamine signals the presence of food in C. elegans, an apparently rewarding stimulus for the worm, a preliminary conclusion might be that dopamine signals reward, supporting the more traditional view of the dopamine code. However, the results from this dissertation, that dopamine modulates non-associative learning by slowing habituation, thereby acting to maintain responses implies that dopamine could be a signal for salience in the worm. Further work is needed to confirm this hypothesis; however, it suggests a conserved role for dopamine across species in order to mediate simple forms of learning. 2.3.5. Novel molecular and cellular mechanism for habituation The data presented here provides evidence that dopamine release activates DOP-1 receptors coupled to the G-protein EGL-30. EGL-30 activates phospholipase C beta (EGL-8), which produces DAG to activate the kinase PKC-1. At this point, it’s unclear what the downstream targets of PKC-1 phosphorylation are to enhance neuron excitability, thereby modulating habituation; however, phosphorylation events are known to have this capability (Cai et al., 2009). To our knowledge, this is the first signalling cascade that has been identified in a mechanism for habituation. A number of molecules have been identified in other systems (Table 1.2), but none have found the downstream effectors to this extent. 70  Previous studies pointed to the synapse as the potential site of plasticity for habituation. For example, habituation of the gill withdrawal reflex in Aplysia results at least in part from decreased vesicular release from presynaptic sensory neurons onto post-synaptic motor neurons (Castellucci et al., 1970; Castellucci and Kandel, 1974) either from a depletion of vesicles in the readily releasable pool (Bailey and Chen, 1988) or silencing of release cites (Gover et al., 2002). The experiments presented in this chapter are the first example to our knowledge that part of the mechanism for habituation involves modulation of the excitability of neurons. Other forms of neural plasticity have been found to modulate neural excitability. Serotonin enhances the excitability of sensory neurons in Aplysia during sensitization of the gill withdrawal reflex after tail shock (Marinesco and Carew, 2002). In mammals, D1 receptors enhance the excitability of pyramidal neurons (Yang and Seamans, 1996; Gulledge and Jaffe, 1998; Henze et al., 2000; Lavin and Grace, 2001; Wang and O'Donnell, 2001; Chen et al., 2004). A potential mechanism for this involves the enhancement of L-type calcium currents (Yang and Seamans, 1996; Seamans et al., 1997; Yang et al., 1999). Voltage-gated calcium channels are thought to be important for neural excitability in C. elegans neurons (Goodman et al., 1998; Suzuki et al., 2003) and L-type channels have been shown to be regulated by PKC phosphorylation (Young and Yang, 2004; Toselli and Taglietti, 2005), suggesting that the signalling cascade that we’ve identified for modulation of neural excitability of the tap sensory neurons and habituation in C. elegans, may be conserved in other systems.  2.4. Methods 2.4.1. C. elegans strains and genotypic analysis Strains were maintained as described by (Brenner, 1974), at 20oC, on NGM agar with OP50 E. coli as a food source. The wild-type reference strain used was N2. The mutant alleles used were: CB1112 cat-2(e1112) and LX636 dop-1(vs101) which were obtained from the C. elegans Gene Knockout Consortium, RM2702 dat-1(ok157), which was a gift from Janet Deurr, and the Cameleon strain (ljEx123[lin-15(+);mec-4p ::YC3.12]), which had been created in the Schafer Lab (Suzuki et al., 2003). The 71  following Cameleon mutant strains were also created in the Schafer Lab (Kindt et al., 2007): AQ1312 dop-1(vs101) ; ljEx123[lin-15(+) ;mec-4p ::YC3.12], AQ1425 cat-2(e1112);ljEx123[lin-15(+);mec4p ::YC3.12], and dat-1(ok157) ; ljEx123[lin-15(+);mec-4p ::YC3.12]. 2.4.2. Habituation assay Habituation of the tap withdrawal response assays were performed using three different approaches. The first was by manual tracking and human scoring. It was performed similar to previous studies (Broster and Rankin, 1994; Rankin and Wicks, 2000). Single 4 day old worms were transferred to a Petri plate filled with 10 ml of NGM agar and recorded using a video camera (Panasonic Digital 5100) attached to a VCR (Panasonic AG1960) and monitor (NEC) under a stereomicroscope (Wild M3Z, Wild Leitz Canada). A time-date generator (Panasonic WJ-810) was used to superimpose a digital stopwatch and time - date display on the video. The worm was kept within the field of view by manually moving the plate with a micromanipulator (Marzhauser, model MM33). Six minutes after the worm was transferred to the plate, 30 tap stimuli were administered at a 10 second inter-stimulus interval. Taps were delivered to the side of the plate using a metal rod driven by an electromagnetic relay controlled by a voltage generator (Grass Instruments, S88 stimulator). A tap response was assessed by reviewing the video and determining if an animal paused or reversed within one second of a tap stimulus. Worms that were already reversing when the tap occurred were ignored for that stimulus. The second approach used a single worm automated tracker developed by the Schafer Lab and was performed similar to a previous study (Sanyal et al., 2004). Young egg-laying adults, 18-22 hours post L4 stage at 20°C were used. To quantify the reversals, animals were tracked using a Zeiss Stemi 2000-c Stereomicroscope with a Cohu High Performance CCD video camera with a capture rate of 4 frames/second. A computer-controlled tracker (Parker Automation, SMC-1N) placed the worm in view during the 5 minute recording of the habituation training. An automated tapper was made using a Lego Mindstorms Robotics kit. Lego Mindstorm software was used to program the robot to tap the plate every 10 second 30 times to achieve a 10 second inter-stimulus interval. This Lego tapper utilized an 8.2 72  g lever arm with a rubber-coated tip that used the stretch of rubber bands to dissipate 6 mJ of energy to the culture plate. Prior to the assays, animals were placed onto freshly seeded OP50 plates (25 µl OP50 culture spread onto plate and thoroughly dried for at least 1 hour) and allowed to recover for at least 5 minutes on the new plate before the assay. Grayscale images were processed using image processing software (Cronin et al., 2006; Huang et al., 2006). Images were thinned into a skeleton set of 30 backbone points with defined head and tail coordinates. A reversal was defined when the distance the first backbone point traveled was greater than 5% of the worm’s body length. Matlab scripts were used to analyze the raw data. For reversal analysis, the start of a reversal response to a tap was scored in the time interval, 0.5 sec before the tap to 3 sec after for a non-reversing worm. The end of a reversal is marked by a change in direction or no movement for at least 2 frames. Individual data points were discarded if the worm was already reversing before the tap stimulus. From this reversal data, the reversal frequency was calculated. Reversal frequency is defined as the number of animals reversing out of the number of valid animals at each point or tap stimulus. The final approach was using the Multi-Worm Tracker, as described in chapter 4 (Swierczek et al., 2011), using animals that were 80-90 hours after egg-lay at 20°C. 2.4.3. In vivo calcium imaging Optical recordings were performed on a Zeiss Axioskop 2 upright compound microscope equipped with a Dual View beam splitter (Optical Insights), and a Uniblitz Shutter (Vincent Associates). Fluorescence images were acquired using MetaVue 6.2 (Universal Imaging). Acquisitions were taken at 100Hz for touch cell habituation recordings with 4x4 binning using a 63x Zeiss Achroplan water immersion objective. Filter/dichroic pairs were: excitation, 420/40; excitation dichroic 455; CFP emission, 480/30; emission dichroic 505; YFP emission, 535/30 (Chroma). Individual adult worms (~24 hours past L4) were glued with Nexaband S/C cyanoacrylate glue to pads comprised of 2% agarose in 73  extracellular saline (145 mM NaCl, 5 mM KCl, 1 mM CaCl2, 5 mM MgCl2, 10 mM HEPES, 20 mM dglucose, pH 7.2). The stimulator was a needle with 50 m diameter made of a drawn glass capillary (1.2 mm OD, 0.69 mm ID borosilicate glass) with the tip rounded to 10-15 m. The stimulator was positioned using a motorized stage (Polytec/PI M-111.1DG microtranslation stage with C-862 Mercury II controller) to stimulate. Stimulation of the anterior body was accomplished by positioning the glass capillary at a 45° to the animal’s body halfway between the large bulb of the pharynx and the cell body of ALM. The glass tip was moved toward the animal so that it depressed the cuticle by 10 m and vibrated for 1 second. To synchronize stimulations with optical recordings, a white LED was triggered at the onset of mechanical stimulation. Image analysis was performed using a custom program written in Java, and parameterized using scripts written in Matlab R13 (The Mathworks) as described in (Hilliard et al., 2005). 2.4.4. Statistical analysis For all tap withdrawal response habituation experiments, the response probability was calculated as the number of worms that responded to a tap divided by the total that were valid for that tap. Invalid worms were worms that were not recognized properly by the imaging software, or worms that were already reversing when the tap stimulus was given. Within group variation was measured using the standard error of the mean for a proportion. Initial responses between two groups were compared using Z-tests for proportions and deemed to be significantly different if the probability that the initial responses were the same was less than 5% (i.e. p < 0.05). Habituation was assessed using the MannWhitney rank sum test on the response probabilities for the following 29 stimuli. That is the probabilities for the last 29 responses for two groups were ranked. The sum of these ranks was calculated for each group. Rank sums were compared to the distribution of rank sums expected from two groups sampled from the same population (the null hypothesis). The groups were considered to have significantly different habituation if the observed rank sums deviated from what is expected 95% of the time from the null hypothesis (i.e. p < 0.05). It is important to note that this test is powerful for identifying a difference in habituation, but cannot distinguish whether the difference is caused by a 74  change in rate or asymptotic level. Thus, the terms slow/shallow habituation were used interchangeably, as were fast/rapid/deep habituation. For cases when the initial response between groups was significantly different (such as Figure 2.3), data was standardized to the initial response and the Mann-Whitney test for habituation was re-analyzed to confirm the habituation effect was not completely explained by changes in response sensitivity in general. For calcium response experiments, the mean response was calculated as the average peak increase in FRET ratio after a stimulus from the baseline prior to the stimulus. Responses after the initial response were averaged or binned across three stimuli at a time (2-4, 5-7, 8-10, and 11-13) to reduce noise. Within group variation was measured using the standard error of the mean. Group differences were evaluated using the Mann-Whitney rank sum test for stimulus/binned stimuli. Since this involved five comparisons per experiment (initial, 2-4, 5-7, 8-10, and 11-13), a Bonferroni correction was applied.  75  3. Development and validation of the Multi-Worm Tracker 3.1. Introduction One of the major advantages of working with C. elegans as a model organism to study a biological phenomenon is the speed with which genetic manipulations can be performed and assessed. There are many examples of where this advantage has been used successfully to discover novel genes and pathways for important biological phenomena. For example, the paralyzed mutant unc-13 that was initially identified in a forward genetics screen by Sydney Brenner (Brenner, 1974) and then mapped (Brenner, 1974; Rose and Baillie, 1980). Further analysis for aldicarb resistance led to its possible role at synapses (Nguyen et al., 1995). With the genomic location identified the gene was sequenced, cloned and GFP fusion constructs were generated determining that unc-13 is localized to presynaptic zones (Kohn et al., 2000). Electrophysiological recordings provided evidence for its role as part of the priming complex of vesicle release at the neuromuscular junction of C. elegans (Richmond et al., 1999) and it homologs (Dunc13 and Munc13) have been identified to be critical for vesicular priming in Drosophila and mammals (Aravamudan et al., 1999; Augustin et al., 1999). Most of the cases where novel genes have been identified have benefited from phenotypes that can easily and rapidly be scored by immediate human inspection (in the example above, severe paralysis). The major bottleneck in the study of habituation is the behavioural assay because it unfolds over a relatively long period of time (minutes), it is difficult to score easily and accurately without close observation, and it manifests stochastically making it necessary to assess large number of individuals to fully characterize the behaviour. Testing one worm at a time and scoring the behaviour by human inspection (Rankin and Broster, 1992; Rankin and Wicks, 2000) can take in the order of 40 person hours of work to assess enough individuals to roughly phenotype the habituation characteristics of a single genetic manipulation, such as a mutant strain. Using an automated behavioural tracking and scoring system that follows a single worm at a time only reduces this to about 10 hours. This makes it very  76  difficult to take advantage of the rapid molecular manipulations that are available to C. elegans, such as forward or reverse screens. For example, a forward genetic screen for habituation in C. elegans was attempted by repeatedly tapping mutants and searching for ones that were still responding to taps long after wild-type worms typically stopped responding. This seemed like a great idea, and in fact a number of mutants were identified, one of which, named hab-1 was more carefully characterized (Xu et al., 2002). However, these mutants have never been mapped to specific genes making it impossible to draw any mechanistic conclusions. It’s difficult to explain why this approach was not successful, but it is likely due to the stochastic nature of habituation. A system capable of tracking and analyzing many worms at once, while maintaining a high level of detail in the behavioural analysis, is necessary to relieve the behavioural bottleneck so that high throughput techniques can be used for the study of habituation in C. elegans. Other systems have been designed to track and analyze the behaviour of multiple worms (Roussel et al., 2007; Ramot et al., 2008; Tsechpenakis et al., 2008). These systems require video to be recorded and then image analysis of the video. In order to accurately score the tap withdrawal response, high speed high resolution video is needed. Because storing high-speed high-resolution video is impractical, these systems use resolutions (~75 μm/pixel) and frame rates (3–5 frames/s) that make it difficult to accurately analyze tap evoked reversals. The Kerr lab has engineered a novel system named the Multi-Worm Tracker (http://sourceforge.net/projects/mwt) (Swierczek et al., 2011), which is capable of monitoring many worms on a single plate while processing the image data in real time, so that no video storage is necessary, using commodity computer hardware and a high-speed, high-resolution camera. This has the potential to measure tap habituation at an unprecedented scale that should enable genetic screens for defects in this complex behaviour. The goal of the research in this chapter was to validate this system and then use it to conduct a small scale reverse genetic screen as proof of principle. Spontaneous movement and habituation of the tap withdrawal response were analyzed across multiple wild-type and 77  mutant strains. Screening a list of 33 diverse mutants was successful at identifying two novel habituation mutants, CX20 and NM1815, one of which has a known mutation in the tomosyn homolog, tom-1.  3.2. Results 3.2.1. Analysis of locomotion C. elegans modulates its speed in response to a variety of stimuli including both acute insults and chronic environmental conditions such as the presence of food. We first monitored basal movement rates of adult wild-type hermaphrodites on food. The process of putting a plate on the tracker involves temporarily removing the lid, and presumably changes temperature and humidity, induces mechanical vibration, and increases light levels. We observed, both by eye and with the MWT, that this process increased the worms’ movement speed, which then descended to a basal level over 10 min (Figure 3.1). To assess the reproducibility of these results, we recorded from seven additional plates of wildtype worms (Figure 3.1) with an average of 27 worms tracked per time point per plate. Every plate showed approximately the same trend; Monte Carlo sampling of tracks suggested that plate-to-plate variability was ~60% larger than expected if all worms on a plate were independent (i.e. not influenced by unidentified variables between plates that are difficult to control such as small differences in humidity, volume of media and percent of agar). For this reason, we ran all experiments on several plates to average out this variability.  78  Figure 3.1 Locomotion (a) Worm speeds over time immediately after being placed on the tracker and observed for 1 hour of spontaneous exploration on an agar-filled Petri plate seeded with a thin lawn of E. coli. Instantaneous vector speed (positive is forward locomotion; negative backward locomotion) of four representative individual worms (lines of various shades of purple/blue). The identity of individual worms cannot be maintained during collisions with other worms or if the worm leaves the region of interest, thus four worms were selected to show representative behaviour for the entire hour of observation. Average scalar speed (i.e. forward and backward locomotion are both positive) of all worms on two representative plates (yellow and pink diamonds; ~30 tracked worms per plate). The mean speed of eight plates (orange circles). (b) Mean scalar worm speed for the indicated mutants and strains after being placed on the tracker (mean of four plates, ≥15 worms per plate). (c) Mean scalar worm speed for different Caenorhabditis species (mean of four plates, ≥20 worms per plate).  79  80  To explore which sensory modalities were responsible for the initial elevation in movement speed, we recorded mec-4 mutants, defective in the gentle touch response (including the tap response, which is an integration of the anterior gentle touch and posterior gentle touch), and che-2 mutants with defective sensory cilia that eliminates chemotaxis. mec-4 mutants responded similarly to wild-type worms; che-2 mutants increased speed but returned to baseline much faster (Figure 3.1). Unexpectedly, a line of the wild-type strain (N2) recently obtained from the C. elegans Genomic Center also showed notable differences from the Kerr lab’s wild-type line (hereafter called XJ1 to distinguish it). To determine whether the che-2 phenotype was a result of a chemosensory defect or was due to genetic drift in strains, we tested two additional chemosensory mutants, osm-6 and tax-2. Both of these strains phenocopied che-2 mutant behaviour (Figure 3.1), indicating that a major portion of the initial increase in speed requires properly functioning ciliated sensory neurons. We also investigated whether the difference between wild-type strains was typical of variation found in the wild. We examined an extreme case by tracking other Caenorhabditis species, namely briggsae, remanei and brenneri. All three species showed similarly elevated movement rates, but C. briggsae had a markedly slower return to baseline (Figure 3.1). Collectively, these tests demonstrated that the MWT is broadly suited for studying baseline locomotion in Caenorhabditis. 3.2.2. Tap habituation Worms respond to non-localized vibration (tap to the Petri plate) by executing an escape response (reversal) and habituate to repeated stimuli by lowering the magnitude and probability of response (Rankin et al., 1990; Rankin and Broster, 1992; Sanyal et al., 2004; Kindt et al., 2007). To deliver tap stimuli automatically, we constructed a solenoid tapper that drove a metal rod into the side of the plate at intervals specified in the MWT and wrote a reversal-detection plug-in. Careful manual annotation disagreed with the plug-in results in 5% of cases for which reversals were small; as the manual annotator had to revisit the data several times to catch his or her own errors, we concluded that automatic annotation is likely more reliable than routine manual annotation. 81  When we applied taps with a 10 second inter-stimulus interval, initially nearly all worms responded with large reversals, but with repeated taps both the probability and size of responses decreased (Figure 3.2); these results were consistent across six independent wild-type plates (strain XJ1). As expected, the mechanosensory mutant mec-4 had low reversal probabilities (Figure 3.2) and small reversal sizes (Figure 3.2) in response to tap, which is consistent with the involvement of mec-4 in mechanosensory transduction in the sensory neurons of the tap withdrawal circuit (AVM, ALMs, and PLMs; Figure 1.1) (Chalfie and Au, 1989; O'Hagan et al., 2005). In contrast, all chemosensory mutants responded to tap (Figure 3.2) and showed a surprising range of phenotypes. che-2 worms showed a dramatically increased habituation rate, and none of the chemosensory mutants matched wild-type reversal distances; tax-2 mutants had larger responses, whereas osm-6 mutant responses were as small as basal reversals shown by mec-4 worms. Sensory deprivation during development is reported to cause a smaller response to tap (Rose et al., 2005). Consistent with this, osm-6 mutants lack functional sensory cilia (Perkins et al., 1986) and che-2 mutants encode a gene expressed in many ciliated sensory neurons but not in touch neurons (Fujiwara et al., 1999). However, this does not explain the wide diversity of phenotypes in the chemosensory mutants; additional work is required to understand this. We then used the MWT to conduct a pilot screen for new tap habituation mutants. We chose 33 strains with mutations in genes with a diversity of predicted functions (Figure 3.3; Table 3.1). Although we quantified a variety of parameters, to simplify analysis we focused on only the probability of reversal on initial and final taps. We recorded from three to four plates of each strain, averaging 136 worms per tap, plus six to eight plates of wild-type worms each day as a control. Of the 33 strains, four had movement defects that were too severe to properly score response to tap, twenty were not significantly different from wild-type worms (P > 0.05), six had defects in initial (non-habituated) response, and three were identified as specifically tap habituation defective, CX20 (adp-1), MT8943 (bas-1;cat-4), and NM1815 (tom-1) (Figure 3.3; Table 3.1).  82  Figure 3.2 Analysis of tap habituation Probability of reversing (a) or reversal distance (b) after a tap are plotted against the number of tap stimuli. Data for six plates (~30 worms per plate) of wild-type (XJ1) worms were plotted in different shades of gray. Reversal probability (c) and distance (d) for mechanosensory and chemosensory mutants plotted against the number of tap stimuli. Error bars are standard error of the mean; n = 3 plates for mutants and N2, 6 plates for XJ1, ≥10 worms per plate.  83  Figure 3.3 Tap habituation screen Probability of response for initial response (stimulus 1) (a) and habituated response (average for stimuli 28–30) (b) of various mutants and wild-type controls; Z-score was normalized by wild-type distribution. Rejected mutants were those with abnormal initial response (P < 0.05; Bonferroni correction was not applied here to decrease the chance of false negatives since habituation-specific mutants were of particular interest to us, not mutants that had altered sensitivity to tap). Probability of reversal after tap plotted for the loss-of-habituation mutant CX20 (adp-1) (g) and the hyper-habituation mutant NM1815 (tom-1) (h). Error bars are standard error of the mean; n = 4 plates with ~30 tracked worms each.  84  85  Table 3.1 Tap habituation screen Strain  Gene  Description  CX20 MT8943 JD21  adp-1 bas-1; cat-4 cca-1  OS122 VC855  cfi-1 cle-1  CB3241  clr-1  RB812  fax-1  RB1816 VC2138  gpa-16 kcnl-2  VC2209 RB1593 RB1416  lgc-46 klp-15 nhr-83  RB921  nhr-84  VB674  nnt-1  RB2462  oct-1  RB2164  pde-1  RB1231  pde-4  VC282  pdl-1  RB809  ptl-1  RB1638 RB1537 VC14 RB699  rab-18 rab-19 rap-2 rgs-9  Unknown locus Enzymes for monoamine biosynthesis T-type calcium channel alpha subunit DNA-binding protein Vertebrate type XV/XVIII collagen homolog Receptor tyrosine phosphatase Nuclear hormone receptor G protein alpha subunit Calcium-activated potassium channel Ligand-gated ion channel Kinesin-like protein Nuclear hormone receptor family Nuclear hormone receptor family Nicotinamide nucleotide transhydrogenase Organic cation transporter cGMP phosphodiesterase cAMP phosphodiesterase cGMP phosphodiesterase Microtubule-associated protein small GTPase small GTPase RAS-like GTPase Regulator of G-protein signalling  Phenotype Response to Average Response Initial Tap to Taps 28-30 wt wt  high low  wt  wt  wt wt  wt wt  wt  wt  low  low  wt wt  wt wt  n/a high wt  n/a high wt  wt  wt  wt  wt  wt  wt  wt  wt  low  wt  low  wt  wt  wt  high wt wt wt  wt wt wt wt  wt = not significantly different from wild-type high = significantly higher tap response(s) than wild-type (P < 0.05) low = significantly lower tap response(s) than wild-type (P < 0.05) n/a = not assessed due to severe locomotion defects  86  Strain  Gene  RB976 RB2027 VC380  rhgf-1 src-1 tag-68  RB1182 NM1815 LY130  tba-1 tom-1 twk-20  CB246 CB1220 VC1392 LY140  unc-64 unc-82 zip-5 --  Description  Rho-like GTPase Tyrosine kinase TGFbeta receptor signalling family Tubulin (alpha) Tomosyn ortholog TWiK family of potassium channels Syntaxin Serine/threonine kinase bZip transcription factor Voltage-gated potassium channel  Phenotype Response to Initial Tap  Average Response to Taps 28-30  wt n/a wt  wt n/a wt  n/a wt wt  n/a low wt  n/a wt low wt  n/a wt wt wt  wt = not significantly different from wild-type high = significantly higher tap response(s) than wild-type (P < 0.05) low = significantly lower tap response(s) than wild-type (P < 0.05) n/a = not assessed due to severe locomotion defects -- = gene is not mapped  87  3.3. Discussion We have shown that the MWT is capable of assessing spontaneous locomotion and tap habituation by assessing two wild-type strains (N2 and XJ1), three wild-type strains of other Caenorhabditis species (C. briggsae, C. remanei, and C. brenneri), and a set of sensory mutants (osm-6, tax-2, che-2, and mec-4). We were then able to rapidly assess the habituation of a list of candidate mutants, identifying 3 novel habituation mutants. MT8943, which has a low habituation phenotype, is known to have mutations in bas-1 and cat-4 (genes important for the biosynthesis of dopamine and serotonin). This is consistent with previous studies that have found dopamine metabolism and receptor mutants to have similar habituation phenotypes (Sanyal et al., 2004; Kindt et al., 2007)(chapter 2). This validates the MWT as a system that can identify previously described phenotypes. CX20 is a strain that was isolated during a screen for chemosensory and salt adaptation (Colbert and Bargmann, 1995; Jansen et al., 2002). Adaptation, like habituation, is also defined as a response decrement after repeated or extended exposure to a stimulus. Psychologist have traditionally differentiated these phenomena; after habituation, a response can fully or partially be recovered after the presentation of a novel dishabituating stimulus, as seen with an electric shock for the tap withdrawal response in C. elegans (Chiba and Rankin, 1990) and the gill withdrawal reflex in Aplysia (Pinsker et al., 1970), while response decrement following adaptation can only recovery spontaneously at a rate proportional to the time since the last stimulus presentation. Dishabituation was not attempted in the odorant and salt adaptation assays (Colbert and Bargmann, 1995; Jansen et al., 2002), so it is unclear whether the adaptation and habituation assays that are reported are the same phenomena occurring in different sensory circuits, or if the mechanisms that mediate adaptation and habituation share some aspects but not others. Unfortunately, the locus/loci that cause these affects in CX20 have not been mapped, so further insight into the mechanism is not possible at this point. Finding this locus will be of great interest because adaptation and habituation are currently thought to be very 88  different processes, hence the reason for many of the characteristics that separate the two phenomena (Thompson and Spencer, 1966; Rankin et al., 2009). Finding a common link between them might provide a better understanding of the relationship that they share. NM1815 has a mutation in tom-1 (a homolog of Tomosyn) which has been shown to play a role in negatively regulating both synaptic and dense core vesicle priming in C. elegans (Gracheva et al., 2006; Gracheva et al., 2007b) resulting in changes in neurotransmission (Gracheva et al., 2007a). eat-4, another synaptic vesicle protein thought to load glutamate into vesicles in the mechanosensory neurons of the tap withdrawal response has a similar phenotype (Rankin and Wicks, 2000). Together, these results implicate neurotransmitter release as a potential site of plasticity for the mechanism of habituation, which is consistent with other models of cellular plasticity for habituation. It is believed that modulation of vesicular release at the synapse between the sensory and motor neurons is the major site of plasticity for habituation of the gill withdrawal reflex in Aplysia (Gover and Abrams, 2009). This seems distinct from the neural excitability mechanism meditated by dopamine in the mechanosensory neurons of the tap circuit (Kindt et al., 2007), which supports the hypothesis that multiple mechanisms of cellular plasticity likely exist to fully explain habituation in its entirety, as suggested by the evidence provided by parametric analysis of tap habituation in C. elegans (Rankin and Broster, 1992; Broster and Rankin, 1994). High-throughput behavioural analysis has many advantages. One can run many controls along with experiments and test many conditions instead of being restricted to one carefully defined protocol without knowing which aspects are important and which are irrelevant. One can test large numbers of strains or screen for new mutants instead of relying on previous research and intuition to select a small number of candidate genes for analysis. Using a habituation protocol of 30 taps with an inter-stimulus interval of 10s, we were able to reliably measure the habituation of 29 of these strains and 15 wild-type replicates in 76 hours of person work (32 hours of strain maintenance and experimental preparation; 44 hours of tracking) over the course of 14 days (Figure 6). As a comparison, a total of 26 strains (covering 89  17 genes) have been tested for habituation of the tap withdrawal response in the literature to date (spanning the last 11 years). Assuming a sample size of 40 worms per strain (although experiments using the MWT presented here averaged sample sizes of 136), previous methods to assay tap habituation behaviour would have taken roughly 580 hours of tracking plus 290 hours of maintenance and preparation (totalling 870 hours across 145 days). The behavioural scoring and analysis, which is fully automated and included in the time given for the MWT approach, is not automated for some of the previous methods, suggesting 870 hours might be a significant underestimate. This demonstrates the strength of using the MWT to measure complex behaviour in a high-throughput manner. This highthroughput, data-rich tool is now ready to be partnered with the already advanced genetic techniques available to the C. elegans research community in order to provide a deeper understanding of habituation.  3.4. Methods 3.4.1. Strains All assays were conducted with the wild-type Bristol isolate of C. elegans (N2) unless otherwise stated. The mutant strains CB1033 (che-2), PR811 (osm-6), RB2464 (tax-2) and CB1611 (mec-4) as well as the strains of wild-type Caenorhabditis species N2 (C. elegans, Bristol isolate) HK104 (C. briggsae), PB4641 (C. remanei) and PB2801 (C. brenneri) were obtained from the Caenorhabditis Genomic Center. These strains were maintained unfrozen for less than 30 generations before testing. The strain XJ1 is descended from N2 but was maintained unfrozen for at least several years (hundreds of generations). The 33 strains tested for tap habituation defects (Table 3.1) were also obtained from the Caenorhabditis Genomic Center. 3.4.2. Apparatus and image acquisition of behavioural recordings Experiments were performed on a custom-built stage with a window (5 cm diameter) to allow diffuse lighting of the plate from 15 cm below using with a 4 inch × 4.9 inch light plate (Schott A08925 90  with ACE I illuminator), a custom-built plate holder to secure 5 cm Petri plates in the window, and a custom-built tapper (http://sourceforge.net/projects/mwt)(Swierczek et al., 2011), which generated the mechanical stimuli. A Dalsa Falcon 4M30 camera (8 bits; 2,352 × 1,728 pixels, 31 Hz) used with a Rodenstock 60 mm f-number 4.0 Rodagon lens was held above and focused so that the entire surface of the plate was projected onto the CCD of the camera at a resolution of 24.3 µm per pixel. Lighting was adjusted using the power supply and the lens aperture so that the lighting power was as close to max as possible (lower power sometimes caused a flicker that disrupted tracking) and background pixel intensity was 200-210, which resulted in wild-type worm image intensity to be 70-100. Data from the camera was input into a computer with 3 GHz Intel Core 2 Duo processors and 4 GB of RAM running the Windows XP operating system using National Instruments PCIe-1427 CameraLink capture card. Image data was analyzed in real-time using the MWT software (http://sourceforge.net/projects/mwt)(Swierczek et al., 2011) to identify and save the position and shape of worms to disk. The following data acquisition settings were used: Object Contrast = 15%, Fill Hysteresis = 20%, Maximum Object Size = 300 pixels, Minimum Object Size = 80 pixels (for very small strains this was reduced to 50 pixels to improve tracking), Object Size Hysteresis = 50%. The MWT software was also used in combination with a PCI-bus compatible counter/timer board (#PCI-CTR05, Measurement Computing, Norton, MA, USA) with a solid state relay (#17M6585, Newark, Chicago, IL, USA) powered by a 24 V, 3.5 A, linear regulated AC-DC power supply (#A24MT350, Acopian, Easton, PA, USA) to control the custom-built tapper; which was positioned 0.5 mm from the plate with a stroke distance of 5 mm. 3.4.3. Movement experiments Baseline recordings on food were performed on 5-cm nematode growth medium (NGM) plates seeded 24 h earlier with 50 μl of OP50 bacteria. Worms were transferred to the plate at 80–90 h of age and allowed to acclimate for 4 h before being tracked for 60 min. Unless otherwise specified, all  91  analysis was performed with the Choreography options --shadowless -M 3 -t 20 -S --plugin Reoutline::exp --plugin Respine. To assess whether plate-to-plate differences in movement rates were a result of the statistics of behaviour of individual worms, we pooled all worms from eight plates and randomly sampled them (with replacement) to generate data for 1,000 virtual plates. The observed differences between real plates and their mean was 60% larger than the difference between virtual plates and their mean (20.7 μm/s versus 12.9 μm/s); five of eight real plates had difference scores that were statistical outliers (P < 0.01) of the virtual plates’ distribution. Thus, the observed variability between plates was not solely due to statistical sampling. Other likely causes include different environmental conditions on the plates, subtle perturbations during recording or behavioural states that last longer than identity is maintained (205 s on average here). Fortunately, plate-to-plate variability does not obscure the advantage of multiworm tracking: Monte Carlo generation of virtual plates with ten tracked worms produced a dataset with slightly larger plate-to-plate variability than was observed for real plates. 3.4.4. Tap habituation experiments Tap habituation assays were performed on food after a minimum of 6 h recovery from transfer; worms were tracked for 10 min to allow them to approach steady-state behaviour and then were tapped 20 or 30 times at a 10 second inter-stimulus interval. A reversal was scored when a worm was still or moving forward at the time of the tap and moved backwards within 1 second of the impact; the reversal was considered to be complete when the worm began a pause or forward motion lasting for more than 0.2 seconds. Cases where the worm was already reversing were removed from analysis. For reversal probability computations, reversal distances of less than 30 μm were considered ‘no reversal’, and we converted counts of reversals into probabilities by dividing the number of worms reversing by the total number valid for each stimulus.  92  Validation was performed by comparing with manually annotated reversals. Out of 148 responses scored by hand as either yes, no or already-reversing, the tracker disagreed with eight (5% error rate). All discrepancies but one were due to small disagreements over the timing or size of a response. Putative tap habituation mutants were prepared using a higher throughput method: five gravid adults were placed on a plate seeded with 50 μl E. coli 24 hours before. The adults were left to lay eggs and then removed from the plate 3 hours later, leaving ~60–80 eggs. The plate was tracked once the worms reached adulthood (80 ± 3 hours later, at 20°C). The recording protocol consisted of 100 s of baseline followed by 30 taps applied to the side of the plate at a 10 second inter-stimulus interval. The data were analyzed with Choreography’s reversal detection plugin: --plugin MeasureReversal::tap::dt=1::collect=0.5. For each mutant strain and wild-type replicate, three or four plates were tested (except CB1220, two plates); on average, 34 worms per plate were suitable for analysis. A wild-type distribution for all wild-type plates tested on the same day was created for the initial and habituated (28th–30th) responses. All strains and wild-type replicates were standardized to the distribution for the day they were tested. Monte Carlo sampling was used to generate the null hypothesis distribution given this scheme, and the resulting P values were re-interpreted as an effective Z-score for plotting. Values outside ± 3.13 were considered to be significantly different than the wildtype distribution (P < 0.05, two-tailed, with Bonferroni correction for 29 comparisons, the number of testable strains in the screen). CX20 and NM1815 were confirmed using an unpaired two-tailed t-test between the raw reversal probabilities of the last stimulus for each plate versus the wild-type replicate that was tested most closely chronologically.  93  4. Characterization of a nervous system-biased mutant library 4.1. Introduction With the successful development of the Multi-Worm Tracker (MWT, described in chapter 3) to rapidly and accurately collect detailed high-throughput behavioural data, I sought to use this technology to characterize the habituation of a set of 522 strains of C. elegans with known mutations. Because the MWT is capable of measuring many phenotypes simultaneously, a number of non-habituation phenotypes were also collected in order to improve the catalog of phenomics for C. elegans. Phenomics is the collection of phenotypes expressed by an organism of a given genotype and is in need of tools like the MWT because discovery and characterization of complex phenotypes such as whole-animal morphology and behavioural plasticity have been particularly slow due to the time and human effort needed for accurate and rapid assessment (Houle et al., 2010). The study of the molecular and cellular functions of genes can vastly benefit from the identification of robust, rapidly assessable behavioural phenotypes. Once a phenotype of this nature has been discovered, it can be used to identify gene function. For example, identification of the rover versus sitter behavioural phenotypes in Drosophila melanogaster with natural polymorphisms in the foraging gene, which encodes cyclic-GMP dependent protein kinase (PKG), led to a better understanding of the cellular function of PKG in the nervous system (Kaun and Sokolowski, 2009). Increasing the size and detail of the behavioural phenotypic catalogue of genomes is critical to exploit the full potential of the rapidly accumulating genomic data. With a large and detailed set of phenotypic data, it is also possible to do phenotypic profiling in order to identify potential genetic interactions, which will help our understanding of how genes work together in networks and pathways. I have collected extensive behavioural and morphological data for a mutant library of Caenorhabditis elegans. Using the MWT, 15 minutes of behaviour was recorded for hundreds of individuals for each of 522 strains of C. elegans with known mutations. Wild-type animals typically  94  explore their environment (in the lab this is an agar-filled Petri-plate seeded with a thin lawn of the bacteria E. coli) in bouts of forward locomotion and periods of foraging, interspersed with occasional spontaneous reversals that lead to a change in direction (Zhao et al., 2003). The experiments included ten minutes of spontaneous behaviour in order to characterize it within our mutant library. Five minutes of repeated mechanical stimuli (taps) were administered to assess their tap withdrawal response. From these experiments, 14 variables were quantified to describe four morphological and behavioural phenotypes for each strain: body size, spontaneous locomotion, mechanosensory responsiveness and habituation. Of the 7308 phenotypes measured across the 522 mutant strains, 1490 were statistically distinguishable from wild-type. Multi-variable analysis further distinguished all but six of the mutant strains from wild-type. Each mutant strains that was tested had a known mutation (mostly insertion/deletions or null point mutations) in an identified gene, suggesting that the mutated gene may contribute to the biological mechanism underlying the phenotype. Secondary mutant analysis was conducted to further support the identification of two novel habituation genes, goa-1, a Goalpha subunit of a heterotrimeric G-protein, and eat-16, a regulator of G-protein signaling. Taking advantage of this rich data set, a similarity metric was developed to identify mutants with similar phenotypic profiles in order to predict 1187 genetic interactions, which is approximately 0.87% of all possible pairwise interactions between the 522 mutants.  4.2. Results Five hundred twenty two strains of C. elegans, each with a known mutation in a gene predicted to function in the nervous system (Supplemental Table 4.2), were obtained from the Caenorhabditis Genetics Center (CGC) and tested in order to phenotype body size, spontaneous locomotion, initial mechanosensory responsiveness and habituation. Most strains were generated by the C. elegans Knockout Concortium using a mixture of trimethylpsoralen and ultraviolet radiation exposure, which 95  often leads to insertion and/or deletion alleles, thereby having a higher chance of null function (Barstead and Moerman, 2006). Some strains were collected by the Caenorhabditis Genetics Center donated from labs around the world. The method of mutation for these mutants varied, but most often used the mutagen ethyl methanesulfonate, which often creates point mutations (Brenner, 1974). Confirmed nulls were chosen from these when possible. For each strain, the behaviour of 57.4 +/- 25.4 (mean +/- standard deviation) age-synchronized (86.6 +/- 4.1 hours after egg-lay) worms were simultaneously observed on a Petri plate during ten minutes of spontaneous behaviour and five minutes of stimulus-evoked behaviour (30 mechanical taps to the side of the plate at a 10 second inter-stimulus interval). Four plates of worms were observed per strain; this was considered one replicate (a total of 174.6 +/- 101.4 worms were valid for analysis per replicate). The N2 wild-type strain was used for comparison; fifty replicates of N2 were tested across multiple days and apparatuses to account for variability caused by these factors (five MWTs were used; temperature, humidity, circadian time, and some plate conditions could not be fully controlled). 4.2.1. Phenotypic characterization Phenotypic metrics for all wild-type replicates (Supplemental Table 4.1) and mutant strains (Supplemental Table 4.2, including Z-scores after being standardized to the wild-type distribution: Supplemental Table 4.3) are included in supplemental information. Visual representation of all mutant phenotypes are also included in supplemental information (Supplemental Figure Set 4.1). The distribution of wild-type replicates and mutant strains for each of the 14 phenotypes are represented in Figure 4.1, and the 1490 phenotypes significantly different from wild-type (N2) are summarized.  96  Figure 4.1 Distribution and summary of phenotypes. Distributions of 50 wild-type (N2) replicates (green) and 522 different mutant strain (purple) for 14 morphological and behavioural variables (a, midline length; b, body area; c, reversals frequency during spontaneous locomotion; d, baseline speed during spontaneous locomotion; e, deceleration rate of spontaneous locomotion after being placed into the tracker; f, reversal probability of the initial response to tap; g, reversal probability after 30 tap stimuli (habituated level); h, rate of habituation for reversal probability; i, reversal distance of the initial response to tap; j, reversal distance after 30 tap stimuli (habituated level); k, rate of habituation for reversal distance; l, reversal duration of the initial response to tap; m, reversal duration after 30 tap stimuli (habituated level); n, rate of habituation for reversal duration. Mutant distribution is shaded in blue for values significantly lower than wild-type and red for strains significantly higher than wild-type. The blue number represents the number of strains that were significantly lower and the red number indicates the number of strains that were significantly higher than the wild-type distribution for each phenotype. The green line shows the Gaussian distribution used to model the wild-type distribution.  97  98  4.2.1.1. Body Size Body size was characterized by measuring the length of the body midline and the area (2D projection) of the worm. Wild-type worms were 1.007 +/- 0.034 mm in length and occupied an area of 0.11 +/- 0.0062 mm2 (Figure 4.1a; Figure 4.1b; Figure 4.2), which is similar to other reports (Brenner, 1974). 179 and 142 mutant strains had significantly smaller midline length and body area than wildtype, respectively, while none were significantly larger for either measure (Figure 4.1a; Figure 4.1b). Some of these observations replicate previous findings. These include strains with mutations in the TGF-beta pathway, important for controlling development and body size of C. elegans: CB491 (contains a mutation in the gene sma-3)(Brenner, 1974; Savage et al., 1996), CB1482 (sma-6)(Krishna et al., 1999), and NU3 (dbl-1)(Morita et al., 1999; Suzuki et al., 1999) (Figure 4.3). Both midline length and area significantly correlated with age (described in detail later), therefore, some of these body size effects may be indirectly caused by delays in the development or aging of the strains. In many cases, however, the size phenotype was more extreme than would be predicted by developmental delay because all of the worms that were tested were laying eggs suggesting that they were at least as old as young adults, yet many strains were significantly smaller than young adult wild-types (midline length = 0.80 +/- 0.40 mm; area = 0.074 +/- 0.008 mm2 for eight replicates between 68 and 73 hours old after egg-lay). Some novel discoveries included RB1329 (C56G3.1), RB2526 (nmur-2), and RB911 (fshr-1), which all had significantly smaller body size phenotypes than wild-type (Figure 4.2). These three mutants have mutations in various neuropeptide receptors. C56G3.1 is an unnamed and uncharacterized gene, nmur2 is homologous to the human neuromedin U receptor (Lindemans et al., 2009), and fshr-1 is homologous to the human follicle-stimulating hormone receptor, thyrotropin receptor, and luteinizing hormone receptor (Cho et al., 2007). Taken together, these data suggest a role for neuropeptide signaling in the regulation of body size in C. elegans.  99  Figure 4.2 Novel mutants with midline length and area of body size phenotypes Mean (“-”) +/- standard error of the mean (error bars) for midline length (a) and area (b) for each experimental plate and the weighted average across all plates of the same strain (diamond). One of the 50 wild-type (N2) replicates is represented in green (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). RB1329 (C56G3.1), RB2526 (nmur-2), and RB911 (fshr-1) have significantly smaller midline length and area (all in blue). No strains had significantly larger midline length or area, however, VC619 (inx-9) was the mutant with the largest midline length and LX475 (rgs-5) had the largest area (both in red). The numbered magenta diamonds indicate the phenotypes measured in the phenotypic profiles shown in Figure 4.15: <1> midline length, and <2> body area. 100  Figure 4.3 Replication of mutants with body size phenotypes Mean (“-”) +/- standard error of the mean (error bars) for midline length (a) and area (b) for each experimental plate and the weighted average across all plates of the same strain (diamond). One of the 50 wild-type (N2) replicates is represented in green (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). CB491 (sma-3), CB1492 (sma-6), and NU3 (dbl-1) had significantly smaller midline length and body area (all in blue).  101  4.2.1.2. Spontaneous Locomotion C. elegans respond to stimuli and navigate chemical or temperature gradients using reversals (periods of crawling backward) followed by turning (Hedgecock and Russell, 1975; Chalfie et al., 1985; Kaplan and Horvitz, 1993; Pierce-Shimomura et al., 1999). Even in a relatively neutral environment worms perform spontaneous reversals, possibly in response to changes in the micro-environment that are undetectable to human observers (Zheng et al., 1999; Brockie et al., 2001). During the 10 minutes of spontaneous behaviour, the frequency of spontaneous reversals was calculated. Wild-type worms reversed 4.61 +/- 0.51 times per minute (Figure 4.1c; Figure 4.4). Previously reported spontaneous reversal frequencies vary (Zheng et al., 1999; Shingai, 2000; Zhao et al., 2003; Gray et al., 2005) because the reversal rate is modulated by many intrinsic and extrinsic factors including but not limited to temperature, humidity, agar dryness, and the presence of E. coli (Zhao et al., 2003); however, the study that was closest to the conditions that we used reported similar reversal frequencies (Gray et al., 2005). Twenty-three mutant strains had a significantly lower frequency of spontaneous reversals compared to wild-type (Figure 4.1c). Some of the mutants with low spontaneous reversal frequencies also had a low baseline speed (described below), such as CB55 (unc-2) and CB251 (unc-36) (Figure 4.4). These were some of the first uncoordinated mutants identified (Brenner, 1974), and were qualitatively described as slow moving and uncoordinated. Hence, it is not surprising that these mutants have low reversal frequencies. Many of the mutants with significantly lower spontaneous reversal rates had normal or high baseline speeds, for example, MT1212, MT1443, and IK105 which have mutations in egl19 (an L-type calcium channel subunit), egl-10 (a regulator of G-protein signaling) and pkc-1 (protein kinase C), respectively (Figure 4.4). Seventy-five mutants had a significantly higher rate of reversals than wild-type worms (Figure 4.1c). Some of these strains were also uncoordinated mutants, CB5 (unc-7) and BC96 (unc-22) (Figure 4.4). These mutants have previously been qualitatively described with 'twitcher' and 'kinker' phenotypes, describing rapid changes in directions of movement. Visual inspection of the behavioural recordings confirmed that these high rates of reversal were caused by uncoordinated 102  locomotion. Examples where high reversal rates were a novel phenotype included RB907, which has a mutation in cgef-2, a CDC-42 guanine nucleotide exchange factor (Figure 4.4). Guanine nucleotide exchange factors regulate the activity of small GTPases. Interestingly, a strain (RB942) with a mutation in cdc-42, the GTPase that CGEF-2 is thought to regulate, also had significantly higher reversal rates (Figure 4.4). Mutants with disruptions in completely uncharacterized genes also showed this phenotype, such as RB1978 (ZK1248.15) (Figure 4.4). Previous reports indicated that overexpression of the wild-type glr-1 gene as well as expression of a constitutively active allele causes a significant increase in spontaneous reversal rates (Zheng et al., 1999; Chao et al., 2005). glr-1 encodes a non-NMDA glutamate receptor subunit (Hart et al., 1995; Maricq et al., 1995; Mellem et al., 2002). We found that the glr-1 null mutant VM3109 had significantly lower frequencies than wild-type (Figure 4.4), which is consistent with the observations that glr-1 mutants spend more time traveling forward before initiating a reversal than wild-type (Brockie et al., 2001). Other reports of glr-1 loss of function mutants did not find this phenotype (Hart et al., 1995; Maricq et al., 1995; Chao et al., 2005). One of the instances was using the same allele (Maricq et al., 1995). There are two possible explanations for this. First, mechanical stimulation inhibits spontaneous reversals (Zhao et al., 2003); in our assay, worms are not handled prior to testing, so mechanical stimulation is very limited; for other assays this may not be the case, resulting in a masking of the effect. Second, glr-1 mutants were tested in the absence of food in the other studies, which also causes a dramatic reduction in spontaneous reversal frequency (Shingai, 2000), potentially masking the effect. In our assay, worms were tested in the presence of food. Our finding is consistent with the role of glr-1 in the command interneurons for mediating spontaneous reversals (Zheng et al., 1999; Brockie et al., 2001; Chao et al., 2005).  103  Figure 4.4 Mutants with spontaneous reversal phenotypes Mean (“-”) reversal rate during the 600 second observation of spontaneous behaviour for each experimental plate and the weighted average across all plates of the same strain (diamond). One of the 50 wild-type (N2) replicates is represented in green (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). MT1212 (egl-19), MT1443 (egl-10), IK105 (pkc-1) and VM3109 (glr-1) had significantly lower reversal rates (all in blue). RB942 (cgef-2), RB907 (cdc-42) and RB1978 (ZK1248.15) had significantly higher reversal rates (all in red). The numbered magenta diamonds indicate the phenotypes measured in the phenotypic profiles shown in Figure 4.15: <3> spontaneous reversal rate.  104  Baseline speed and deceleration rate were also measured to phenotype locomotion in our mutant library. During the first 600 seconds of behavioural recording, the speed of worms was measured every 10 seconds (averaged over an interval of 1 second). The speed was initially high and gradually decelerated to a stable level, so an exponential curve was fit to the data and the baseline speed was measured as the value on this exponential curve at 600 seconds (Figure 4.5). Deceleration to baseline was measured as the time when the exponential curve decreased to the speed half-way between the peak and baseline value (Figure 4.5). The baseline speed of wild-type (N2) worms was 0.0593 +/- 0.0100 mm/s (Figure 4.1d; Figure 4.5). Although, the baseline speed of worms varies greatly for the same reasons that spontaneous reversal rate varies, our results are similar to those previous measurements (Ramot et al., 2008). Thirty-four mutants had significantly faster baseline speeds, while 12 mutants were significantly slower compared to wild-type worms (Figure 4.1d). NM1278 (rbf-1) had significantly slower baseline speed than wild-type (Figure 4.5), which has previously been shown to be required for basal locomotion (Staunton et al., 2001). Many of the phenotypes detected were novel. For example, RB2266 (skr-5) was significantly slower than wild-type (Figure 4.6). skr-5 is homologous to a yeast E3-ubiquitin ligase. Interestingly, 4 of the fastest mutants that we found have mutations in uncharacterized genes: RB2344 (Y4C6A.1; Figure 4.6), RB1553 (K02E10.1), VC1076 (C14F11.2), and RB2171 (C49C3.21) (Supplemental Figure Set 4.1).  105  Figure 4.5 Speed-related spontaneous locomotion phenotypes Mean speed of worms during the 600 second observation of spontaneous behaviour for each plate (fine lines) and the weighted average across all plates of the same strain (diamonds) fit with an exponential curve. One of the 50 wild-type (N2) replicates is represented in green (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). NM1278 (rbf1) has significantly slower baseline speed than wild-type (blue). The numbered magenta diamonds indicate the phenotypes measured in the phenotypic profiles shown in Figure 4.15: <4> baseline speed, and <5> deceleration rate.  106  Figure 4.6 Novel mutants with spontaneous locomotion phenotypes Mean speed of worms during the 600 second observation of spontaneous behaviour for each plate (fine lines) and the weighted average across all plates of the same strain (symbols) fit with an exponential curve. One of the 50 wild-type (N2) replicates is represented in green diamonds (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). RB2266 (skr-5) had significantly slower baseline speed than wild-type (blue circles); RB2344 (C49C3.21) had significantly faster baseline speed than wild-type (red squares).  107  Deceleration to baseline has not been previously used as a phenotypic metric for locomotion. Wild-type worms decelerated half-way to baseline after 88.43 +/- 10.30 seconds (Figure 4.1e; Figure 4.5). One-hundred-forty-four mutants decelerated significantly more rapidly than wild-type, while 46 mutants decelerated slower (Figure 4.1e). Confirming the usefulness of this metric for further defining spontaneous locomotion, some mutants with a severe phenotype were not significantly different from wild-type on any of the other 14 phenotypic measures; for example, RB1068, which has a mutation in nra-1, a gene that encodes a putative calcium-dependent phospholipid binding protein, decelerated significantly faster than wild-type (Figure 4.7). nra-1 is expressed in the nervous system and muscle, particularly co-localizing with puncta formed by the nicotinic acetylcholine receptor subunit LEV-1 at the neuromuscular junction(Gottschalk et al., 2005). LEV-1 containing receptors are necessary for proper locomotion (Lewis et al., 1980; Fleming et al., 1997). Using this novel phenotype to further analyze the function of nra-1 may help further our understanding of neurotransmission at the neuromuscular junction and of acetylcholine receptor associated proteins. 4.2.1.3. Mechanosensation We assessed responsiveness to mechanical stimulation by measuring the response to the initial tap using three metrics: the probability of reversal, the distance travelled during the reversal and the duration of the reversal. Similar to previous reports (Wicks and Rankin, 1995) 91.51 +/- 2.68 percent of wild-type worms reversed in response to the initial tap (Figure 4.1f; Figure 4.8). The reversals of the animals that responded were 0.536 +/- 0.049 mm in length (Figure 4.1g; Figure 4.9) and lasted 2.43 +/0.13 seconds (Figure 4.1h; Figure 4.10). This is also similar to the reversal magnitude described previously (Wicks and Rankin, 1995), however, this value varies greatly in the literature and may depend on a variety of environmental variables and properties of the tapping stimulus (e.g. intensity; Timbers, Giles et al., unpublished).  108  Figure 4.7 Mutant with rapid deceleration during spontaneous locomotion Mean speed of worms during the 600 second observation of spontaneous behaviour for each plate (fine lines) and the weighted average across all plates of the same strain (diamonds) fit with an exponential curve. One of the 50 wild-type (N2) replicates is represented in green (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). RB1086 (nra1) had significantly faster deceleration to baseline speed than wild-type (purple).  109  Figure 4.8 Initial response and habituation of tap-induced reversal probability Reversal probability for each plate (fine lines) +/- standard error of proportion (error bars) and all plates combined (symbol) with fitted exponential (thick lines) in response to each of 30 mechanical tap stimuli. One of the 50 wild-type (N2) replicates is represented in green (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). KB6 (rle-1) had a significantly smaller initial response (yellow circle). RB874 (M110.7) had a significantly lower habituation level and significantly more rapid habituation (blue circles). VC1179 (F08B12.1) had a significantly higher habituation level and significantly slower habituation rate (red squares). The numbered magenta diamonds indicate the phenotypes measured in the phenotypic profiles shown in Figure 4.15: <6> initial response, <7> habituated level, and <8> habituation rate for reversal probability.  110  Figure 4.9 Initial response and habituation of tap-induced reversal distance Mean (fine lines) +/- standard error of the mean (error bars) reversal distance for each plate and the weighted average across all plates (symbol) with fitted exponential (thick lines) in response to each of 30 mechanical tap stimuli to the side of the Petri plate. One of the 50 wild-type (N2) replicates is represented in green (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). VC1000 (C04E6.5) had a significantly larger initial response (yellow circle). KG524 (gsa-1) had significantly more rapid habituation rate (blue circles); MT1083 (egl-8) had significantly higher habituation level (red squares). The numbered magenta diamonds indicate the phenotypes measured in the phenotypic profiles shown in Figure 4.15: <9> initial response, <10> habituated level, and <11> habituation rate for reversal distance.  111  Figure 4.10 Initial response and habituation of tap-induced reversal duration Mean (fine lines) +/- standard error of the mean (error bars) reversal duration for each plate and the weighted average across all plates (symbol) with fitted exponential (thick lines) in response to each of 30 mechanical tap stimuli to the side of the Petri plate. One of the 50 wild-type (N2) replicates is represented in green (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). NL792 (gpc-1) had significantly more rapid habituation (blue circles); DG1856 (goa-1) had significantly higher habituation level (red squares). The numbered magenta diamonds indicate the phenotypes measured in the phenotypic profiles shown in Figure 4.15: <12> initial response, <13> habituated level, and <14> habituation rate for reversal duration.  112  Fifty-seven, 103 and 95 mutants were less responsive measured by reversal probability, distance and duration, respectively (Figure 4.1f; Figure 4.1g; Figure 4.1h). The MWT clearly detected previously identified mechanosensory defective mutant CB75 (mec-2; Figure 4.11) (Chalfie and Sulston, 1981; Huang et al., 1995). Novel mechanosensory defective mutants were also identified, such as KB6 (Figure 4.8), which has a mutation in rle-1, an E3 ubiquitin ligase that has been implicated in longevity (Li et al., 2007). More importantly, this examination of the behavioural characterization allowed me to identify 24 and 99 mutants with significantly larger reversal distance and duration in response to tap, respectively (Figure 4.1). For example, VC1000 (C04E6.5) had significantly larger response to tap measured by reversal distance (Figure 4.9). This is of particular interest because very few genes have been discovered that when disrupted, positively regulate mechanosensation, so some of these novel mutants may reveal deeper insights into the transduction of mechanical stimuli. No mutants had significantly higher reversal probability; however, this may be due to a ceiling effect since the proportion of wild-type worms that respond to a tap is close to 100%. 4.2.1.4. Habituation Responses to the habituation training (30 taps) were also measured using reversal probability, distance and duration. An exponential curve was fitted to the data for each measure. Habituation level was defined as the value on this curve at the final stimulus, and habituation rate was defined as the number of stimuli needed to reach the mid-point between the initial response and the habituated level. Thus, six metrics were measured to phenotype tap habituation across our mutant library ([probability, distance, duration] X [level, rate]). Wild-type worms habituated to a reversal probability of 39.8 +/- 5.3 percent with a habituated reversal distance of 0.0645 +/- 0.0203 mm and duration of 0.448 +/- 0.128 seconds (Figure 4.1i, Figure 4.8; Figure 4.1j, Figure 4.9; Figure 4.1k, Figure 4.10). Half-habituation occurred at 6.28+/-1.26 stimuli for reversal probability, 5.63 +/- 0.90 stimuli for reversal distance and 5.38 +/- 0.78 stimuli for reversal duration (Figure 4.1l, Figure 4.8; Figure 4.1m, Figure 4.9; Figure 4.1n, Figure 4.10). 113  Figure 4.11 Replication of mechanosensory and tap habituation mutants Reversal probability for each plate (fine lines) +/- standard error of proportion (error bars) and all plates combined (diamonds) with fitted exponential (thick lines) in response to each of 30 mechanical tap stimuli. One of the 50 wild-type (N2) replicates is represented in green (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). CB75 (mec-2) had a significantly smaller initial response than wild-type (purple). CB1112 (cat-2) had a significantly lower habituation level than wild-type (blue). KP1182 (acy-1) had significantly higher habituation level than wild-type (red).  114  Focusing first on the reversal probability metric, 24 strains habituated to significantly deeper levels than wild-type, 143 strains habituated to significantly more shallow levels, 4 mutants habituated significantly more rapidly and 73 mutants habituated significantly more slowly (Figure 4.1i; Figure 4.1l). Reversal probability habituation correlated significantly with age (described in detail later), so mutant effects could be indirect by affecting rates of development and/or aging. Confirming previous reports and results from chapter 2 of this dissertation, a mutation in the enzyme that catalyzes the rate limiting step of dopamine biosynthesis (CAT-2) caused more rapid tap withdrawal habituation (Sanyal et al., 2004; Kindt et al., 2007) (Figure 4.11). KP1182 (acy-1) had significantly more shallow habituation (Figure 4.11). acy-1 encodes adenylyl cyclase. In Drosophila melanogaster, rutabaga mutants, which have a mutation in adenylyl cyclase, have shallow habituation in five different behavioural protocols (Engel and Wu, 2009). This supports the hypothesis that the mechanisms of habituation may be conserved across phylogeny, and the prediction that mutations identified in C. elegans may inform us about the function of habituation genes in other organisms. For example, two of the most extreme habituation mutants had mutations in human homologs. RB874 (M110.7) habituated significantly more rapidly and more deeply than wild-type, and VC1179 (F08B12.1) habituated more slowly and to a higher level (Figure 4.8). M110.7 is homologous to human neuropathy target esterase and F08B12.1 is a homolog of to human CD118 antigen. For reversal distance, no strains had significantly deeper habituation, likely because of a floor effect caused by very deep habituation of wild-type worms (Figure 4.9). Forty-two mutants habituated to a significantly higher level than wild-type, 8 habituated significantly more rapidly and 48 habituated significantly more slowly (Figure 4.1j; Figure 4.1m). The rate of reversal distance habituation correlated with age (described in detail later), so mutant effects for reversal distance habituation rate could be indirect by affecting rates of development and/or aging. Confirming previous work, a mutation in a vesicular glutamate transporter (EAT-4), necessary for loading synaptic vesicles with glutamate in the mechanosensory neurons of the circuit that mediates the tap response caused more rapid habituation 115  than wild-type (Rankin and Wicks, 2000) (Figure 4.12). Some examples from within the mutant library: KG524, which has a gain-of-function mutation in gsa-1 (GS alpha subunit), habituated significantly more rapidly than wild-type and MT1083 (egl-8) habituated to a significantly higher level (Figure 4.9). egl-8 encodes the signaling enzyme phospholipase C beta. Reversal duration has not previously been used as a metric to assay tap habituation. No strains had significantly deeper habituation of response duration than wild-type. Sixty-one strains habituated to significantly more shallow levels, 15 habituated at a significantly faster rate, and 49 habituated significantly more slowly (Figure 4.1k; Figure 4.1n). The two examples with the most extreme phenotypes had mutations in G-protein subunits: DG1856 (goa-1, GO alpha subunit) habituated to a more shallow level than wild-type and NL792 (gpc-1, a G gamma subunit) habituated more rapidly (Figure 4.10). 4.2.2. Correlations Correlations between the 14 different phenotypic measures suggest common or interrelated cellular and molecular processes. The correlation coefficients for all comparisons between the 14 metrics are shown for the 50 wild-type replicates and the 522 mutant strains (Table 4.1). As expected, strong a correlation was found between midline length and body area (Figure 4.13a). Similarly, each of the reversal distance and reversal duration comparisons (initial response, habituated level and habituation rate) also demonstrated very strong correlations (Figure 4.13b; Figure 4.13c; Figure 4.13d), suggesting that factors determining these measurements overlap. In contrast, the same comparisons between reversal probability and reversal distance did not correlate (Figure 4.13e; Figure 4.13f) suggesting that different cellular mechanisms may mediate habituation of response probability and response distance. For both reversal probability and reversal distance, there was no relationship between the level of habituation and rate of habituation (Figure 4.13g; Figure 4.13h), which suggests that different mechanisms are responsible for each of these variables as well.  116  Figure 4.12 Habituation of eat-4 mutants Mean (fine lines) +/- standard error of the mean (error bars) reversal distance for each plate and the weighted average across all plates (diamonds) with fitted exponential (thick lines) in response to each of 30 mechanical tap stimuli. One of the 50 wild-type (N2) replicates is represented in green (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes). MT6308 (eat-4) had significantly more rapid habituation than wild-type (blue).  117  Table 4.1 Correlation coefficients Metric  Midline_Length  Body_Area  Spon_Rev  Baseline_Speed  Midline_Length  1.000000  0.947950  -0.241287  0.462282  (wild-type)  (1.000000)  (0.885107)  (0.301000)  (-0.155801)  Body_Area  0.947950  1.000000  -0.251509  0.405310  (wild-type)  (0.885107)  (1.000000)  (0.441860)  (-0.232921)  Spon_Rev  -0.241287  -0.251509  1.000000  -0.387094  (wild-type)  (0.301000)  (0.441860)  (1.000000)  (-0.247111)  Baseline_Speed  0.462282  0.405310  -0.387094  1.000000  (wild-type)  (-0.155801)  (-0.232921)  (-0.247111)  (1.000000)  Decel_Rate  0.346113  0.389264  -0.381499  0.267938  (wild-type)  (0.248656)  (-0.024868)  (-0.616183)  (0.031929)  Rprob_Init  0.158759  0.130238  -0.036289  0.204348  (wild-type)  (0.311651)  (0.424612)  (0.449128)  (-0.215131)  Rprob_Hab  -0.132402  -0.141565  0.234905  -0.024820  (wild-type)  (-0.651009)  (-0.558876)  (0.222504)  (0.269207)  Rprob_Rate  -0.011415  -0.072044  -0.065659  0.094417  (wild-type)  (-0.456499)  (-0.369408)  (0.075004)  (0.434991)  Rdist_Init  0.378741  0.351463  -0.186005  0.498509  (wild-type)  (0.372611)  (0.184640)  (-0.097034)  (0.422017)  Rdist_Hab  -0.260930  -0.291451  0.352905  -0.104902  (wild-type)  (0.094008)  (0.009926)  (-0.446805)  (0.024348)  Rdist_Rate  -0.192312  -0.207388  0.161685  -0.225834  (wild-type)  (-0.570002)  (-0.558031)  (-0.206471)  (0.029736)  Rdur_Init  0.137982  0.138585  -0.087802  0.115830  (wild-type)  (0.394217)  (0.291140)  (0.057736)  (-0.341444)  Rdur_Hab  -0.313262  -0.332953  0.386217  -0.240863  (wild-type)  (-0.063117)  (-0.097682)  (-0.406006)  (-0.030363)  Rdur_Rate  -0.070033  -0.087219  0.057839  -0.184785  (wild-type)  (-0.403035)  (-0.456256)  (-0.326595)  (0.120817)  118  Metric  Decel_Rate  Rprob_Init  Rprob_Hab  Rprob_Rate  Midline_Length  0.346113  0.158759  -0.132402  -0.011415  (wild-type)  (0.248656)  (0.311651)  (-0.651009)  (-0.456499)  Body_Area  0.389264  0.130238  -0.141565  -0.072044  (wild-type)  (-0.024868)  (0.424612)  (-0.558876)  (-0.369408)  Spon_Rev  -0.381499  -0.036289  0.234905  -0.065659  (wild-type)  (-0.616183)  (0.449128)  (0.222504)  (0.075004)  Baseline_Speed  0.267938  0.204348  -0.024820  0.094417  (wild-type)  (0.031929)  (-0.215131)  (0.269207)  (0.434991)  Decel_Rate  1.000000  -0.048284  -0.192130  -0.066425  (wild-type)  (1.000000)  (-0.303430)  (-0.463927)  (-0.130058)  Rprob_Init  -0.048284  1.000000  0.577494  -0.019768  (wild-type)  (-0.303430)  (1.000000)  (0.080911)  (-0.319541)  Rprob_Hab  -0.192130  0.577494  1.000000  0.186938  (wild-type)  (-0.463927)  (0.080911)  (1.000000)  (0.462487)  Rprob_Rate  -0.066425  -0.019768  0.186938  1.000000  (wild-type)  (-0.130058)  (-0.319541)  (0.462487)  (1.000000)  Rdist_Init  0.257731  0.520425  0.449715  0.141308  (wild-type)  (0.277795)  (-0.053846)  (0.002635)  (0.138030)  Rdist_Hab  -0.196196  -0.102154  0.120360  0.003275  (wild-type)  (0.157138)  (-0.304500)  (-0.381831)  (-0.359130)  Rdist_Rate  -0.028048  -0.444728  -0.186363  0.106544  (wild-type)  (-0.094393)  (-0.220280)  (0.504486)  (0.235258)  Rdur_Init  0.142907  0.473676  0.422437  0.094000  (wild-type)  (0.327943)  (0.120016)  (-0.191779)  (-0.146585)  Rdur_Hab  -0.224159  -0.208916  0.022750  0.014976  (wild-type)  (0.083742)  (-0.313945)  (-0.261866)  (-0.297740)  Rdur_Rate  -0.021775  -0.434855  -0.215444  0.109075  (wild-type)  (0.072413)  (-0.148267)  (0.345434)  (0.112865)  119  Metric  Rdist_Init  Rdist_Hab  Rdist_Rate  Rdur_Init  Midline_Length  0.378741  -0.260930  -0.192312  0.137982  (wild-type)  (0.372611)  (0.094008)  (-0.570002)  (0.394217)  Body_Area  0.351463  -0.291451  -0.207388  0.138585  (wild-type)  (0.184640)  (0.009926)  (-0.558031)  (0.291140)  Spon_Rev  -0.186005  0.352905  0.161685  -0.087802  (wild-type)  (-0.097034)  (-0.446805)  (-0.206471)  (0.057736)  Baseline_Speed  0.498509  -0.104902  -0.225834  0.115830  (wild-type)  (0.422017)  (0.024348)  (0.029736)  (-0.341444)  Decel_Rate  0.257731  -0.196196  -0.028048  0.142907  (wild-type)  (0.277795)  (0.157138)  (-0.094393)  (0.327943)  Rprob_Init  0.520425  -0.102154  -0.444728  0.473676  (wild-type)  (-0.053846)  (-0.304500)  (-0.220280)  (0.120016)  Rprob_Hab  0.449715  0.120360  -0.186363  0.422437  (wild-type)  (0.002635)  (-0.381831)  (0.504486)  (-0.191779)  Rprob_Rate  0.141308  0.003275  0.106544  0.094000  (wild-type)  (0.138030)  (-0.359130)  (0.235258)  (-0.146585)  Rdist_Init  1.000000  -0.031427  -0.382294  0.809726  (wild-type)  (1.000000)  (0.148710)  (-0.093989)  (0.416146)  Rdist_Hab  -0.031427  1.000000  0.131741  0.030015  (wild-type)  (0.148710)  (1.000000)  (-0.121443)  (0.035565)  Rdist_Rate  -0.382294  0.131741  1.000000  -0.401748  (wild-type)  (-0.093989)  (-0.121443)  (1.000000)  (-0.073265)  Rdur_Init  0.809726  0.030015  -0.401748  1.000000  (wild-type)  (0.416146)  (0.035565)  (-0.073265)  (1.000000)  Rdur_Hab  -0.164999  0.936484  0.227971  -0.008130  (wild-type)  (0.002930)  (0.928971)  (0.002872)  (0.047427)  Rdur_Rate  -0.365967  -0.045317  0.803426  -0.454539  (wild-type)  (0.026040)  (-0.107820)  (0.784145)  (-0.244299)  120  Metric  Rdur_Hab  Rdur_Rate  Midline_Length  -0.313262  -0.070033  (wild-type)  (-0.063117)  (-0.403035)  Body_Area  -0.332953  -0.087219  (wild-type)  (-0.097682)  (-0.456256)  Spon_Rev  0.386217  0.057839  (wild-type)  (-0.406006)  (-0.326595)  Baseline_Speed  -0.240863  -0.184785  (wild-type)  (-0.030363)  (0.120817)  Decel_Rate  -0.224159  -0.021775  (wild-type)  (0.083742)  (0.072413)  Rprob_Init  -0.208916  -0.434855  (wild-type)  (-0.313945)  (-0.148267)  Rprob_Hab  0.022750  -0.215444  (wild-type)  (-0.261866)  (0.345434)  Rprob_Rate  0.014976  0.109075  (wild-type)  (-0.297740)  (0.112865)  Rdist_Init  -0.164999  -0.365967  (wild-type)  (0.002930)  (0.026040)  Rdist_Hab  0.936484  -0.045317  (wild-type)  (0.928971)  (-0.107820)  Rdist_Rate  0.227971  0.803426  (wild-type)  (0.002872)  (0.784145)  Rdur_Init  -0.008130  -0.454539  (wild-type)  (0.047427)  (-0.244299)  Rdur_Hab  1.000000  0.057144  (wild-type)  (1.000000)  (-0.087786)  Rdur_Rate  0.057144  1.000000  (wild-type)  (-0.087786)  (1.000000)  121  Figure 4.13 Covariance of phenotypes. Scatter plots of 50 wild-type (N2) replicates (green) and 522 mutant strains (purple) for (a) midline length versus body area, (b) initial response of reversal distance versus initial response of reversal duration, (c) habituated level of reversal distance versus habituated level of reversal duration, (d) habituation rate for reversal distance versus habituation rate for reversal duration, (e) initial response of reversal probability versus initial response of reversal distance, (f) habituated level of reversal probability versus habituated level of reversal distance, (g) habituated level of reversal probability versus habituation rate for reversal probability, and (h) habituated level of reversal distance versus habituation rate for reversal distance.  122  123  Inter-stimulus interval is another parameter of habituation that has been hypothesized to be mediated by multiple mechanisms (Rankin and Broster, 1992); however, its effect on reversal probability and distance has not been separately analyzed. Wild-type (N2) worms were habituated with 30 taps at either a 10 or 60 second inter-stimulus interval (Figure 4.14). Reversal probability and reversal distance habituated to significantly (p < 1.0x10-13 and p < 1.0x10-5, respectively) lower levels at the shorter inter-stimulus interval, which is consistent with previous reports (Rankin and Broster, 1992). Interestingly, reversal probability was affected significantly greater than reversal distance (p < 1.0x10-14). The lack of correlation between reversal probability and distance observed in the mutant library and the differential effect of inter-stimulus interval on reversal probability and distance habituation provides further evidence that inter-stimulus interval is mediated by different molecular mechanisms. The age of the wild-type worms was measured with an accuracy of plus or minus 1.5 hours. Wildtype worms were tested between 77 and 94 hours after egg-lay (mean = 86.6 hours; standard deviation = 4.1 hours). Within this range, significant correlations were found between age and some of the tested variables. Midline length (r = 0.92), body area (r = 0.77), reversal probability habituated level (r = -0.70), reversal probability habituation rate (r = -0.49), and reversal distance habituation rate (r = -0.51). This suggests that in some cases, the mutant phenotype for these variables may not be caused by a direct effect on the variable but by an indirect effect on the rate of development or aging.  124  Figure 4.14 Effect of inter-stimulus interval on habituation Mean +/- standard deviation (between plates) of reversal probability (a) and reversal distance (b) for wild-type (N2) C. elegans in response to 30 tap stimuli presented at a 10 s (blue) or 60 s (red) interstimulus interval. Twenty plates (~50 worms per plate) were tested for each group. Stimulation at 10 s inter-stimulus interval caused deeper habituation than stimulation at a 60 s interval. Initial responses for both reversal probability and distance were not significantly different. Habituated responses (response to the 30th stimulus) were significantly different for both probability and distance (p < 1.0 x 1013  and p < 1.0 x 10-5, respectively). The effect of inter-stimulus interval had a larger effect on reversal  probability than reversal distance. Difference scores calculated between a 10 and 60 s inter-stimulus intervals for percent habituation were significantly larger for reversal probability compared to reversal distance (p < 1.0 x 10-14).  125  4.2.3. Covariance within wild-type phenotypic profiles enhances better distinction Analyses of the 14 morphological and behavioural variables individually, led to the discovery of 1490 phenotypes distinguishable from wild-type (Figure 4.1). These phenotypes were distributed across 418 of the 522 mutants (80.1%), with 104 mutants that could not be statistically separated from wildtype. However, by analyzing the covariance observed within the phenotypic profiles of the 50 wild-type replicates, mutants were further distinguishable from wild-type based on their entire phenotypic profiles. Example phenotypic profiles for a wild-type replicate, DG1856 (goa-1) and NL792 (gpc-1) are shown in Figure 4.15. A multi-dimensional Gaussian was fit to the data and the Mahalanobis distance between the wild-type mean and each of the mutant strains was calculated in order to find strains that were significantly different; distances to each wild-type replicate were also calculated (Figure 4.16). Mahalanobis distance calculates the linear distance between two points within multi-dimensional space with covariance. All but six of the mutant strains (98.9%) had significantly larger distances than the wild-type distribution. A major obstacle in understanding gene function is the detection of observable phenotypes in mutant strains. Our high-throughput collection of phenotypic profiles of morphological and behavioural data has addressed this problem and has the potential to aid functional genetic studies of hundreds of uncharacterized genes, many of which have mammalian and human homologs.  126  Figure 4.15 Examples of phenotypic profiles In previous figures, phenotypic variables are highlighted by numbered magenta diamonds: <1> midline length (Figure 4.2a), <2> body Area (Figure 4.2b), <3> spontaneous reversal rate (Figure 4.4), <4> baseline speed (Figure 4.5), <5> deceleration rate (Figure 4.5), reversal probability (Figure 4.8): <6> initial response, <7> habituated level, and <8> habituation rate, reversal distance (Figure 4.9): <9> initial response, <10> habituated level, and <11> habituation rate, reversal duration (Figure 4.10): <12> initial response, <13> habituated level, and <14> habituation rate. Z-scores were calculated for each variable of each wild-type replicate and mutant strain (standardized to the wild-type distribution). Example phenotypic profiles are shown for (a) an N2 replicate (the replicate with the least sum of squared differences from the wild-type mean across all 14 phenotypes), (b) DG1856 (goa-1) and (c) NL792 (gpc1).  127  128  Figure 4.16 Mahalanobis distances to the mean of wild-type distribution in 14 dimensional phenotypic space Frequency distribution of Mahalanobis distances adjusted based on the covariance observed within the 14 dimensional Gaussian distribution of the 50 wild-type (N2) replicates. Distances from the N2 mean to each of the 50 N2 replicates (green) and each of the 522 mutant strains (purple; not all distances are shown, distribution continues to decay until a maximum distance of 6819.7).  129  One wild-type replicate that was not included when constructing the Gaussian was used as control. The Mahalanobis distance between it and the wild-type mean was not significantly larger than the wild-type distribution. A descendent of the CGC N2 strain (designated XJ1) had been separately maintained in the Kerr laboratory using standard C. elegans maintenance protocols for over 2 years. XJ1 had likely been passaged a minimum of 50 generations since being isolated from N2. The Mahalonobis distance between XJ1 and the mean of the N2 Gaussian was significantly different from the N2. This is an important consideration for geneticists because the XJ1 strain has been maintained in the same way as many lab strains throughout the research community, yet its phenotypic profile has significantly drifted from the canonical N2. A strain known as the "ancestral N2" that has been stored in liquid nitrogen and passaged only five generations since the N2 strain was first studied by Sydney Brenner (1974) was also tested. The Mahalanobis distance between the “ancestral N2” and the mean of the canonical N2 was not significantly different from the canonical N2 distribution. Hence, with proper care by freezing strains and regularly thawing from the original stocks, the CGC has managed to avoid significant genetic drift within the morphological and behavioural variables that were measured of the N2 strain for over thirty years. This is a significant feat considering the Kerr Lab descendent (XJ1) had significantly drifted in only two years of bench top maintenance. 4.2.4. Confirmation of novel habituation genes It is possible that mutants in our library have background mutations in addition to the known mutation. Hence, some of the observed mutant phenotypes may be caused by a mutation other than the one that is known. Thus, it is important to use alternate manipulations to confirm the role of a gene in a phenotype of interest. Because of my interest in habituation I chose to further investigate some of the strongest habituation phenotypes that we identified. The two most extreme phenotypes were found in DG1856 and KG571 mutants (Figure 4.17), which have mutations in goa-1, a GO alpha subunit, and eat-16, a regulator of G-protein signaling (RGS), respectively. To confirm that these genes were responsible for 130  the habituation phenotype, strains with different alleles of goa-1 and eat-16 were tested. Two additional mutants for each gene phenocopied DG1856 and KG571 (Figure 4.17). These data support the hypothesis that both goa-1 and eat-16 are novel critical genes for habituation of the reversal distance of the tap withdrawal response. 4.2.5. Phenotypic profiles predict genetic interactions To predict genetic interactions the assumption that mutation in genes that positively interact in a genetic pathway will cause similar phenotypes. This assumption has been successfully used in the past to identify genetic interactions (recent example: (Green et al., 2011)). For the mutants characterized in this chapter, similarity between mutant strains was assessed by calculating the Mahalanobis distance between each mutant strain in the 14 dimension phenotypic space (Supplemental Table 4.4). To visualize these relationships, we used two clustering algorithms, average-linkage agglomerative hierarchical clustering (Figure 4.18) and t-SNE clustering (Supplemental Movie 4.1). The former method iteratively compares strains to form progressively larger clusters ordered according to the average distance between clusters. The latter method collapses the data from the 14 variable space in which our phenotypic profiles exist into 3 dimensions while attempting to maintain pair-wise distances to better visualize multi-dimensional groupings. These provide good visual representations as strains that are closer together along a branch of the hierarchical clustering dendrogram or closer in 3-dimensional space in the t-SNE clustering have more similar phenotypic profiles than mutant pairs that are further apart; however, these methods necessarily reduce information available in the original 14-dimensional space. Thus, it is difficult to understand how meaningful these distances/clusters are without more information.  131  Figure 4.17 Habituation of various eat-16 and goa-1 mutants Tap withdrawal response habituation of 3 different eat-16 (a, c, e) and goa-1 (b, d, f) mutants (all in purple) and wild-type (N2) replicates (green). Mean (fine lines) +/- SEM (error bars) reversal distance for each plate and the weighted average across all plates (diamonds) with fitted exponential (thick lines) in response to each of 30 mechanical tap stimuli to the side of the Petri plate.  132  133  Figure 4.18 Average-linkage hierarchical clustering dendrogram Each row represents the morphological and behavioural phenotypic profile of a mutant strain. Phenotypes are coloured based on their Z-scores (standardized to the wild-type distribution): 1) midline length, 2) body area, 3) spontaneous reversal rate, 4) baseline speed, 5) deceleration rate, 6) reversal probability: initial response, 7) reversal probability: habituated level, 8) reversal probability: habituation rate, 9) reversal distance: initial response, 10) reversal distance: habituated level, 11) reversal distance: habituation rate, 12) reversal duration: initial response, 13) reversal duration: habituated level, 14) reversal duration: habituation rate. Mutant strains are ordered vertically based on average-linkage agglomerative hierarchical clustering using the pair-wise Mahalanobis distances between each strain in the phenotypic space. The structure of clustering is represented by the dendrogram. *Strain name and genotype are visible in a higher resolution image available in the supplemental information (Supplemental Figure 4.1). Genes that have previously been shown to interact and were found to be significantly similar based on the current analysis are indicated along the right side of the figure: eat-16 and goa-1 (red), unc-6 and unc-40 (green), unc-6 and unc-71 (green), snb-1 and unc-13 (magenta), unc-2 and unc-36 (cyan), and sma-3 and sma-6 (yellow).  134  135  By comparing Mahalanobis distances between mutants to the distribution of Mahalanobis distance observed between wild-type replicates (50 N2 and 152 XJ1 replicates), 78054 pairs of mutants (57.4% of all possible mutant pairs) were found to be significantly different from each other (Figure 4.19). However, mutant pairs that are not significantly different are likely not good predicted interactions; a more stringent analysis is needed. To identify mutants that were significantly similar, the probability that two random mutants would have a smaller distance between them than the observed distance was measured. To generate random mutants, the mutant population was modelled by normalizing and fitting a multi-dimensional Gaussian to the 522 mutant phenotypic profiles. Using this approach, 1187 mutant pairs were found to be significantly similar. Our hypothesis is that the genes mutated in pairs of strains with significantly similar phenotypic profiles may genetically interact. To test this hypothesis, a list of published genetic interactions in C. elegans (obtained from Raymond Lee at www.wormbase.org) was searched and six of these predicted interactions (Figure 4.18; Figure 4.20; Supplemental Movie 4.2) have already been confirmed experimentally: unc-71 and unc-6 (Huang et al., 2003), unc-36 and unc-2 (Schafer et al., 1996), sma-3 and sma-6 (Wang et al., 2002), unc-40 and unc-6 (Davies, 1994), snb-1 and unc-13 (Yook et al., 2001), and eat-16 and goa-1 (Guan and Han, 1999; HajduCronin et al., 1999). This supports the hypothesis that these significantly similar phenotypic profiles predict novel genetic interactions.  136  Figure 4.19 Distribution of Mahalanobis distances between wild-type replicates and mutant strains Frequency distribution of Mahalanobis distances adjusted based on the covariance observed within the 14 dimensional Gaussian distribution of the 50 wild-type (N2) replicates. Distances between each of the 50 N2 replicates (dark green), the 152 XJ1 replicates (light green), and the 522 mutant strains (purple; not all distances shown, continues to decay until a maximum distance of 16,591).  137  Figure 4.20 t-SNE clustering highlights 6 interactions supported by other studies This is a still frame from (Supplemental Movie 4.2) and represents a plot of 522 mutants after dimensional reduction from 14 dimensional phenotype space to three dimensions using t-SNE clustering. The labels include genes that were predicted to interact based on our phenotypic profiling and for which evidence already exists suggesting a potential interaction. Interactions include: eat-16 and goa-1 (red), unc-6 and unc-40 (green), unc-6 and unc-71 (green), snb-1 and unc-13 (magenta), unc-2 and unc-36 (cyan), and sma-3 and sma-6 (yellow).  138  4.3. Discussion A problematic result for any geneticist is seeing no obvious phenotype after a genetic manipulation. Without strong replicable phenotypes, it is difficult to investigate the biological role of a gene. Unfortunately this is often the case when studying a novel gene. There are a number of reasons why no phenotype might be observed, including the presence of redundant genes or the possibility that the organism has evolved to a point where the gene is no longer of use, but selective pressure has not yet removed it from the genome. Although these are possibilities, a more likely explanation for the lack of phenotype is that the resolution of our phenotypic observations is poor. This is particularly true for genes that function to regulate the plasticity of the nervous system. The work presented in this chapter helps to address this problem. By taking extensive, high resolution and detailed measurements of the morphology and behaviour of hundreds of individuals with known mutations using the Multi-Worm Tracker, 1490 instances of statistically distinguishable phenotypes compared to wild-type have been identified. By taking into account covariance of phenotypes, 516 strains are now distinguishable from wild-type. Discovering such strong phenotypes will be of great use for understanding the cellular function of these many uncharacterized genes. Many efforts are currently being undertaken to identify which genes work together in pathways and networks to carry out biological function (Byrne et al., 2007; Green et al., 2011; Horn et al., 2011). The work in this chapter extends these efforts by predicting genetic interactions based on similar behavioural phenotypic profiles. Hopefully the list of 1187 candidates drives hypothesis testing in future studies. However, these candidates are predictions and should be treated as such because there are a number of reasons that pairs within the list might be incorrect. The predicted interactions are based on a key assumption: strains with mutations in genes that function in similar pathways will have similar phenotypes. There are cases where this hypothesis has been supported; for example, sma-3 and sma-6 function in a TGF beta like pathway and mutations in either gene cause worms to have a small body size (Wang et al., 2002). However, it may be possible that different pathways can cause 139  converging phenotypes. The mutant model created to identify significant interactions is based on 522 mutants out of thousands of potential mutants. It is likely that the model does not yet fully represent mutations throughout the entire genome, potentially changing the threshold for how similar phenotypic profiles need to be to be considered significantly similar. Only six of the predicted interactons are confirmed. Once more of these interactions have been tested, it will be easier to pick a threshold that best predicts interaction. Using similar reasoning for the opposite case, gene pairs that are absent from our list of predictions should not be excluded from further investigation because there are several reasons why interacting genes might not be detected using the method of phenotypic profiling described. Background mutations may alter the phenotypic profile. Genes that work together in one cell/tissue to produce one phenotype might work separately in other cells/tissues for other phenotypes. In this case, mutants of interacting genes might have different phenotypic profiles. Allelic variation may cause phenotypic variation. Gene pairs that interact may have opposing function, such as suppressors or negative regulators. In this case, mutants of interacting genes might have opposite phenotypic profiles. It is not clear how to define an opposite phenotypic profile. Steps were taken to minimize these possibilities. For example, null or deletion alleles were preferentially tested to avoid allelic variation. Also, all pair-wise Mahalanobis distances and similarity metric probabilities calculated using our mutant model have been disseminated (Supplemental Table 4.4; Supplemental Table 4.5), so that others can adjust the statistical stringency themselves. The fact that some of the interactions have already been empirically confirmed by previous research supports the potential usefulness of these predictions. Probably the biggest contribution of this study is to our understanding of habituation. Evidence exists for three potential cellular mechanism, decreased excitability of neurons (Kindt et al., 2007), depression of excitatory synapses (Castellucci and Kandel, 1974; Bailey and Chen, 1988; Gover et al.,  140  2002; Weber et al., 2002), and potentiation of inhibitory synapse (Krasne, 1969; Shirinyan et al., 2006; Das et al., 2011). It is not clear whether this is a complete list of ways in which neural circuits can change in order to mediated habituation, however, the work conducted in this chapter not only supports the idea of multiple mechanisms for habituation but also implicates multiple mechanisms within a single behavioural circuit. This is evidenced by the fact that habituation of reversal probability and distance, as well as the level of habituation and the rate of habituation are all affected by independent sets of mutants (i.e. no correlations were observed, which was not the case between reversal distance and duration habituation). It is unclear which of the above (or possible novel) mechanisms explains each distinct aspect of habituation of the tap withdrawal response, but it seems multiple mechanisms will be needed to fully explain habituation of the tap withdrawal response in C. elegans. Perhaps this points to the importance of habituation. If one mechanism is disrupted, many others are available to maintain the important for of behavioural plasticity. In other organisms, usually only a single behavioural metric is used to assay habituation because behavioural scoring is so time consuming. This makes it difficult to predict whether multiple mechanisms are responsible for habituation to other behavioural responses, but seeing as habituation is so highly conserved throughout the animal kingdom, it seems a reasonable hypothesis. The molecular mechanisms that mediate these proposed (and/or yet to be described) cellular changes that take place continue to be a mystery. In the introduction of this dissertation, I introduced a list of molecules that are known to play a role in short-term habituation across many behaviours in a wide variety of model organisms. However, seeing as only a few molecules have been identified to play a role in the same behaviour of one organism, it is difficult to understand how they work together to form a mechanism or set of mechanisms that mediate habituation. The results of this chapter increase the list of potential habituation genes by ~5 times. 259 mutants had one or more habituation phenotypes. These mutants have all been identified in one species using one behaviour, which greatly simplifies the problem of understanding how they work together to produce habituation. 141  Mutations in two genes, goa-1 and eat-16, were discovered to cause an extremely strong habituation phenotype. In fact, these mutants almost completely blocked habituation. Genetic disruptions that block habituation so completely have never before been identified. We hypothesize that these genes play a fundamental role in the molecular mechanism of habituation. Interestingly, these genes have also been found to disrupt chemosensory adaptation in C. elegans (Matsuki et al., 2006), a similar behavioural decrement in response to prolonged exposure to an attractive chemical. These are the second and third genes that have been identified to disrupt both mechanosensory habituation and chemosensory adaptation, the first being adp-1 (Swierczek et al., 2011) that was identified during the pilot screen when first developing the MWT (presented in chapter 3). adp-1 has not yet been mapped, so it cannot help with the understanding of the mechanism of habituation by itself. However, taken together with the goa-1 and eat-16 findings, it suggests that the behavioural plasticity caused by repeated or long-lasting stimulation in both the chemosensory and mechanosensory circuit of C. elegans may be controlled by similar mechanisms, bringing into question the distinction between “adaptation” and “habituation.” This supports the idea that the mechanisms of habituation, like the behaviour itself, are highly conserved throughout the nervous system, and potentially throughout phylogeny. It is clear that more experiments are needed to confirm this, but with the list of novel habituation mutants that have been generated, it offers many candidates for others in the habituation community to test. Although 1187 predicted genetic interactions from the 522 genes in this work, it is unlikely that any pathways have been resolved due to the relatively small number of genes tested compared to the size of the genome (~20,000 genes). However, with the success of this project, it will encourage characterizations of much larger libraries, perhaps all the mutant strains available at the CGC, or using high-throughput genetic manipulations, such as genome-wide RNAi (e.g. Kamath and Ahringer, 2003). Undertakings of this scale are very realistic with the use of the MWT and would identify further genetic interactions and possibly entire pathways. By publishing the entire library of phenotypic profiles along 142  with similarity metrics and predicted interactions, others can use this rich dataset to generate exciting hypotheses to explore the mechanism of habituation even after the conclusion of my dissertation.  4.4. Supplemental information Summaries of the vast amount of data collected in this study are available as supplementary material. All supplemental information can be found in an electronic format (“.zip” file) on the Accompanying Material (DVD). Within the “.zip” file there is a folder for each supplemental item. There are five supplemental tables, two supplemental movies, one supplemental figure set and one supplemental figure. All supplemental tables include 3 files. A space-delimited “.txt” file, a “.xlsx” spreadsheet file, and a “.pdf” file. All three files contain the same information. The multiple formats aim to improve accessibility, as different formats may be useful for different purposes (i.e. data mining/meta-analysis, searching and sorting, or readability). Further descriptions follow. Supplemental Table 4.1 Raw data of 14 phenotypes for reference wild-type strain (N2) The values in this table represent the means for each of the 50 wild-type replicates across the 14 phenotypes analyzed in this chapter. Supplemental Table 4.2 Raw data of 14 phenotypes for mutant strains The values in this table represent the means for each of the 522 mutant strains across the 14 phenotypes analyzed in this chapter. Supplemental Table 4.3 Z-scores of 14 phenotypes for mutant strains The values in this table represent the Z-scores (standardized to the distribution of the reference wild-type strain, N2) for each of the 522 mutant strains across the 14 phenotypes analyzed in this chapter. Supplemental Table 4.4 Pair-wise Mahalanobis distances between mutant strains The values in this table represent the Mahalanobis distances between each mutant strain calculated using the covariance measured within the distribution of the reference wild-type (N2). 143  Supplemental Table 4.5 Similarity metric for pair-wise interactions between mutant strains A model was generated using the distribution observed across the 522 mutants that were characterized. The values in this table represent the probability that a random sample from this model would fall as close to the given mutant as was observed. All pair-wise comparisons are reported so that statistical stringency can be chosen independently. A critical value of (p < 1/522) generates a list of 1187 predicted interactions. Supplemental Figure Set 4.1 Phenotypic profile of mutant strains This folder contains a set of 522 figures, one for each mutant strain. All are “.png” image files. In each figure, the mutant is represented in purple, while a wild-type replicate (least-square differences across the 14 phenotypes compared to the wild-type mean) is represented in green. (Upper left panel) Phenotypic profile: Z-scores of mutant across all 14 phenotypes (standardized to the wild-type distribution). (Upper middle panels) Mean (“-”) +/- standard error of the mean (errorbars) body size measured by midline length and area for each experimental plate and the weighted average across all plates of the same strain. (Upper right panel) Mean (“-”) reversal rate during the 600 second observation of spontaneous behaviour for each experimental plate and the weighted average across all plates of the same strain. (Middle left panel) Mean speed of worms during the 600 second observation of spontaneous behaviour for each plate (fine lines) and the weighted average across all plates (symbols) with fitted exponential (thick lines). (Middle right panel) Reversal probability for each plate (fine lines) +/- standard error of proportion (errorbars) and all plates combined (symbols) with fitted exponential (thick lines) in response to each of 30 mechanical tap stimuli to the side of the Petri plate. (Lower panels) Mean (fine lines) +/- standard error of the mean (errorbars) reversal distance (left panel) and duration (right panel) for each plate and the weighted average across all plates (symbol) with fitted exponential (thick lines) in response to each of 30 mechanical tap stimuli.  144  Supplemental Figure 4.1 High-resolution copy of hierarchical clustering This is a high-resolution replica of Figure 4.18. Higher resolution allows one to see the strain name and genotype of individual mutants and makes it easier to browse the structure of the dendrogram. Supplemental Movie 4.1 t-SNE clustering labeling all mutants This is an animated “.gif” file that can be viewed using any standard web browser (e.g. Mozilla Firefox version 8.0). The image represents a rotating plot of 522 mutants after dimensional reduction from 14 dimensional phenotype space to three dimensions using t-SNE clustering. Labels indicate the gene that is known to be mutated for each strain. Supplemental Movie 4.2 t-SNE clustering highlighting 6 interactions supported by other studies This is an animated “.gif” file that can be viewed using any standard web browser (e.g. Mozilla Firefox version 8.0). The image represents a rotating plot of 522 mutants after dimensional reduction from 14 dimensional phenotype space to three dimensions using t-SNE clustering. The labels include genes that were predicted to interact based on our phenotypic profiling and for which evidence already exists suggesting a potential interaction. Interactions include: eat-16 and goa-1 (red), unc-6 and unc-40 (green), unc-6 and unc-71 (green), snb-1 and unc-13 (magenta), unc-2 and unc-36 (cyan), and sma-3 and sma-6 (yellow).  4.5. Methods 4.5.1. C. elegans strains The strains included in our study were chosen based on a list of 2072 genes with predicted neural function based on domain structure (Sieburth et al., 2005). We cross-referenced this list with the list of strains available at the Caenorhabditis Genetics Center (CGC) as of May 2009 for a collection of 145  approximately 700 mutant strains (many of which were produced by the C. elegans Knockout Consortiums), which were ordered from the CGC. When multiple mutations in the same gene were available, we preferentially choose a confirmed null allele. If a null was not available or known, an insertion/deletion mutant was chosen over a point mutation to increase the chance of testing complete knock-outs or severe loss-of-function mutations. Occasionally, a point mutation was chosen instead of the null or deletion allele because the null or deletion was already report to be lethal, too sick or paralyzed to be tested behaviourally; in a few cases, gain-of-function mutations were tested instead of loss-of-function alleles for the same reason. 522 of these strains were found to be suitable for our highthroughput method; the others were excluded because of severe locomotion, egg-laying/brood size deficits, or low survival rates during standard lab maintenance. Three strains were used as positive controls for various phenotypes: CB1112, MT6308, and VM3108 (a gift from AV Maricq). Three wild-type strains were used, N2, XJ1, and “ancestral N2”; all were lines of the N2 strain described by Brenner (1974). It is unclear exactly when XJ1 was attained from the CGC, but it was maintained in the Kerr Lab at Janelia Farm Research Campus using standard C. elegans maintenance protocols for at least 2 years before the behavioural tests presented here. During this time it diverged from the original stock by 50-100 generations. This strain was tested on each apparatus, each day of the mutant characterization. During the course of experiments, a recent acquisition of the N2 strain from the CGC was tested and was found to be significantly (p < 0.0019) different from XJ1. This prompted us to re-order a fresh stock of the N2 strain from the CGC and test 50 replicates within 4 generations of its receipt; this strain is referred to as N2 in the rest of the paper. The “ancestral N2” strain is thought to have diverged from the strain Brenner originally characterized by only 6 generations. 4.5.2. C. elegans maintenance C. elegans were cultured on Agar filled 5 cm Petri plates seeded with OP50 E. coli lawns; this was maintained by transferring to a fresh plate every two weeks by cutting a piece of agar from the old plate and placing it on the fresh plate. If plates became contaminated by mold or foreign bacteria, the strain 146  was cleaned by placing 5-10 gravid adults into a drop (~5 µl) of 5%-hypochlorite:1-M-NaOH (1:1) (cleaning) solution positioned close to the edge of an agar plate away from the bacterial lawn, killing the contamination and the adult worms but preserving the un-hatched eggs inside the adults. Once the newly hatched larvae crawled to the bacterial lawn (12-24 hours later), the section of the plate where the cleaning solution had been placed was cut away to avoid the spread of any residual contamination. 4.5.3. Behavioural testing Petri plates (5 cm diameter) were filled with 12 ml of sterilized agar, allowed to set, and then placed in a plastic storage container at room temperature for 2-3 weeks to dry before use. 12-24 hours before plates were to be used 50 µl of OP50 E. coli was spread evenly onto each plate using a sterile glass rod and allowed to dry at room temperature with the lid on. Five gravid adult worms were placed on each E. coli seeded plate for 3-4 hours to allow the collection of 60-80 age-synchronized eggs. After removal of the five adults, plates were then placed into a plastic storage container (50-60% full) in a 20°C incubator for 77 to 94 hours (mean = 86.6; standard deviation = 4.1). At this age, wild-type worms have reached adulthood and are in their reproductive prime. There were many eggs on the plate; at the earliest time point a few of the eggs had hatched into L1 larvae, at the latest time point, most eggs had hatched and some larvae had developed to the L2 stage. Plates were removed from the incubator and plastic storage container, and transferred to the behavioural testing room to equilibrate to the changed environmental conditions before testing (approximately 30-60 min). Lids were removed from plates and plates were secured right-side up on the stage of the apparatus; a modified Petri plate lid, which had the top cut away and replaced with a 48 X 60 mm cover-glass (Gold Seal, #12-519-21D, Fisher Scientific, Ottawa, Ontario) fixed to the lid with wax, was place onto the plate. Behavioural recording began immediately with 10 minutes of spontaneous behaviour followed by the administration of 30 taps to the side of the plate generated by a solenoid at 10 second intervals; recordings was terminated 10 seconds after the final tap.  147  4.5.4. Apparatus and image acquisition of behavioural recordings Experiments were performed on a custom-built stage with a window (5 cm diameter) to allow diffuse lighting of the plate from 15 cm below using with a 4 inch × 4.9 inch light plate (Schott A08925 with ACE I illuminator), a custom-built plate holder to secure 5 cm Petri plates in the window, and a custom-built tapper (http://sourceforge.net/projects/mwt) (Swierczek et al., 2011), which generated the mechanical stimuli. A Dalsa Falcon 4M30 camera (8 bits; 2,352 × 1,728 pixels, 31 Hz) used with a Rodenstock 60 mm f-number 4.0 Rodagon lens and lens adaptors was held above and focused so that the entire surface of the plate was projected onto the CCD of the camera at a resolution of 24.3 µm per pixel. Lighting was adjusted using the power supply and the lens aperture so that the lighting power was as close to max as possible (lower power sometimes caused a flicker that disrupted tracking) and background pixel intensity was 200-210, which resulted in wild-type worm image intensity to be 70-100. Data from the camera was input into a computer with 3 GHz Intel Core 2 Duo processors and 4 GB of RAM running the Windows XP operating system using National Instruments PCIe-1427 CameraLink capture card. Image data was analyzed in real-time using the MWT software (http://sourceforge.net/projects/mwt) (Swierczek et al., 2011) to identify and save the position and shape of worms to disk. The following data acquisition settings were used: Object Contrast = 15%, Fill Hysteresis = 20%, Maximum Object Size = 300 pixels, Minimum Object Size = 80 pixels (for very small strains this was reduced to 50 pixels to improve tracking), Object Size Hysteresis = 50%. The MWT software was also used in combination with a PCI-bus compatible counter/timer board (#PCI-CTR05, Measurement Computing, Norton, MA, USA) with a solid state relay (#17M6585, Newark, Chicago, IL, USA) powered by a 24 V, 3.5 A, linear regulated AC-DC power supply (#A24MT350, Acopian, Easton, PA, USA) to control the custom-built tapper; which was positioned 0.5 mm from the plate with a stroke distance of 5 mm.  148  4.5.5. Data analysis To calculate the 14 metrics, preliminary analyses were performed using Choreography (version 1.3.0.r1020; a custom-written java program that is part of the MWT software package; http://sourceforge.net/projects/mwt) (Swierczek et al., 2011) on a computer with Intel Core i7-930 processor and 12 GB of RAM running the Ubuntu operating system (version 10.04 LTS) and the Java SE Runtime Environment (version 1.6.0_20-b02; "-Xm6G" java argument was used in order to allocate enough memory for Choreography to process some of the files). The following arguments were used during data analysis with Choreography: "--shadowless -t 20 -M 2" to ignore animals until their entire body was recognized by the MWT and remove objects that exist for less than 20 seconds or moved less than 2 body lengths, which helps eliminate the occasional imaging artifact; and "-S --plugin Reoutline -plugin Respine" to improve the direction detection, thereby more accurately identifying reversal detection. The window for calculating instantaneous speed was set to 0.1 seconds (“-s 0.1”). The "output" and "trigger" functions were used to calculate the number of animals valid for analysis, the mean and standard error of the mean for midline-length and area, and the mean speed (averaged over 1 second) at 10 seconds intervals during the first 600 seconds of the behavioural recording. The "MeasureReversal" plugin was used to calculate the number, distance and duration of reversals during the first 600 seconds as well as tap-induced reversals (beginning within 1 second of a stimulus). Custom-written Matlab (version: r2008b) and Octave (version 3.2.3, with miscellaneous, statistics, optim, and optim-intercept packages installed) were used to conduct the rest of the analysis. For midline-length and body-area, the weighted average across the 4 plates tested was calculated; for spontaneous reversal frequency, the number of spontaneous reversals during the first 600 seconds was divided by the number of worm-minutes (valid worms multiplied by length of observation (600 seconds)) and then a weighted average was found across the 4 plates tested; for baseline speed and deceleration rate, the weighted average of the mean speed at 10 second intervals during the first 600 seconds was calculated, an exponential curve (y = a+b*(1-e^(-x/c))) was fit using a least squares method, 149  the value of y at x = 600 was the baseline speed, the value of x at y = (midpoint between the max speed and the baseline speed) was the deceleration metric; for reversal probability, the number of animals that reversed divided by the total number valid for analysis was calculated across all plates, the value for the first stimulus was the initial response and then the habituated level and rate were calculated by fitting an exponential as with the speed metrics; for distance and duration, weighted averages were calculated for each stimulus, the value for the first stimulus was the initial response, habituated level and rate was calculated by fitting an exponential as with the speed metrics. Correlation coefficients were calculated using the corrcoef() function. Wild-type correlations were considered significant based on Pearson’s Correlation Coefficient Table for n = 50 using a critical pvalue = 0.05 (corrected with a Bonferroni correction for 210 comparisons; p < 0.001). For the inter-stimulus interval experiment, 20 plates were prepared and tested for each group as described (4 on each of the 5 trackers). Mean and standard deviations for reversal probability and reversal distance were calculated for each of the 30 stimuli. T-tests were performed to compare between different inter-stimulus intervals for the initial and final stimuli for each measure. To test if inter-stimulus interval had more of an effect on probability versus distance, the habituation percentage [(initial response - final response)/initial response] was calculated for probability and distance at both inter-stimulus intervals. Standardized difference scores (between inter-stimulus intervals) were calculated for each metric (percent habituation with 60 s inter-stimulus interval/average percent habituation during 10 s inter-stimulus interval). A t-test was used to compare the standardized difference scores. Mahalanobis distance using the mean and covariance of the 50 wild-type (N2) replicates dataset was calculated using Matlab or Octave functions for each wild-type replicate to find outliers (p < 1/50; none were found) and then for each mutant to find significantly different mutants (p < 1/522). The same was done for XJ1 replicates using the XJ1 mean and covariance to remove outliers (p < 1/161; 9  150  were removed). The mean and standard deviation of the 50 wild-type (N2) replicates was used to calculated Z-scores for all mutant strains. Any Z-scores less than -3.7 or more than 3.7 were deemed to be significantly different from wild-type (p < 0.05/522; adjusted using a Bonferroni correction to avoid false positives). Assuming that covariance between the 14 measures is similar between strains, Mahalanobis distances were also calculated between each N2 replicate (1225 comparisons), each XJ1 replicate (11476 comparisons), and each mutant strain (135981 comparisons) using the covariance of the N2 replicates. Hierarchical clustering was performed using a custom written script. Briefly, each mutant was initially considered a cluster. Iteratively, clusters with the smallest Mahalanobis distance between them were combined (new distances were calculated before the next iteration as the average distance to the newly formed cluster). This was repeated until all strains were part of a single cluster. A dendrogram representing the order in which clusters agglomerated was constructed using custom written scripts. Three dimensional visualization of data was performed using t-distributed stochastic neighbour embedding (t-SNE) (van der Maaten and Hinton, 2008) on the pair-wise Mahalanobis distance matrix of the mutants using a perplexity of 15; this was performed using Matlab. The mutant model was constructed by fitting a 14 dimensional Gaussian distribution to the mutant data; i.e. each measure was normalized (see Table 4.2 for transformations) and standardized. Principal Component Analysis was performed to identify independent axis of variation along which to sample. Random samples (106) were generated from this model, rotated and transformed back to the original variable space. A probability distribution was calculated for each mutant for Mahalanobis distances to each random sample from the mutant model and was used to estimate the probability that two mutant strains were significantly similar (p < 1/522).  151  Table 4.2 Transformation equations to normalize mutant distributions Metric  Transformation  Midline length Body area Spontaneous reversal rate Baseline speed Deceleration rate Reversal probability: initial response Reversal probability: habituated level Reversal probability: habituation rate Reversal distance: initial response Reversal distance: habituated level Reversal distance: habituation rate Reversal duration: initial response Reversal duration: habituated level Reversal duration: habituation rate  y = (-log(x/1.1))^(1/4) y = x^2 y = x^(1/2) y = x^(1/4) y = x^(1/4) y = -log((1.001-x)/1.001) y = (-log(x))^(1/4) y = x^(1/2) y = (-log(x+0.001))^(1/4) y = x^(1/4) y = -log(x/30) y=x y = x^(1/4) y = -log(x/30)  152  5. General discussion The goal of my dissertation research was to gain a better understanding of the cellular and molecular mechanisms underlying habituation. I have accomplished this using two approaches: investigation of a candidate gene (dop-1) using available techniques, and a high-throughput genetic approach that first required the development of a new behavioural tracking system. Both approaches were successful in revealing new insights into the mechanisms of habituation. In chapter 2, using a candidate gene approach, I presented a novel cellular mechanism for habituation, a food-dependent dopamine mediated modulation of the neural excitability of the mechanosensory neurons of the tap withdrawal circuit. Working with the Schafer Lab, we identified the molecular signalling cascade that triggers these changes. Extracellular dopamine neurotransmission activates the DOP-1 dopamine receptor on the mechanosensory neurons which are coupled to the EGL30 G-protein alpha subunit. EGL-30 in turn activates phospholipase C beta (EGL-8), which creates DAG in an enzymatic reaction in the cell. DAG then activates PKC-1. The intracellular targets of PKC-1 are unknown, however, this is the most complete molecular mechanism ever identified as part of the underlying process of habituation (Figure 5.1).  153  Figure 5.1 Putative molecular components involved in habituation of the tap withdrawal response in C. elegans Updated diagram of Figure 1.2e, which summarizes all the molecular components implicated in the mechanism of habituation for the tap withdrawal response in C. elegans after the research conducted in this dissertation. Green arrows indicate activation of downstream components. Red lines with balls indicates inhibition of downstream components (purple lines/labels indicate that the site of action of these genes has not been identified yet, so although they have been place in ALML here to summarize the findings of this dissertation, it is possible that they are working in another neuron(s) in the circuit). Yellow arrows indicate the production, release and re-uptake of the dopamine. trp-4 encodes a transient receptor potential channel. cat-2 encodes tyrosine hydroxylase, which metabolizes dopamine (DA). dat-1 encodes a dopamine re-uptake transporter. dop-1 encodes a dopamine receptor. egl-30 encodes a G-protein alpha q subunit. egl-8 encodes phospholipase C beta, which metabolizes PIP2 into the second messengers diacyl glycerol (DAG) and IP3 (not shown). dgk-1 encodes diacyl glycerol kinase, which inhibits production of DAG. pkc-1 encodes a protein kinase C. egl-19 encodes an L-type voltage-gated calcium (Ca2+) channel thought to be important for amplifying the depolarization of ALML after sensory transduction is trigger by a mechanical stimulus (touch/tap). eat-16 encodes a regulator of G-protein signaling. goa-1 encodes a G-protein alpha o subunit. kht-1 encodes a potassium channel subunit. mps-1 encodes a potassium channel accessory subunit. tom-1 encodes tomosyn, which inhibits synaptic and dense core vesicle release. eat-4 encodes a vesicular glutamate transporter. ?s represent some of the unknown components.  154  155  In chapter 3, I report the results of a project in collaboration with the Kerr lab that led to the development of the Multi-Worm Tracker (MWT), which has vastly increased the rate by which C. elegans tap habituation can be assayed. As proof of principle, a pilot screen for tap habituation mutants was conducted. A mutant with a null allele of the tomosyn homolog tom-1, important for vesicular release (Gracheva et al., 2006; Gracheva et al., 2007b, a), showed a striking habituation phenotype (rapid and deep habituation) supporting the idea that proper neurotransmission is critical for habituation. Vesicular release has been a proposed site of plasticity for habituation for many years because of evidence in Aplysia (Bailey and Chen, 1988). The tom-1 mutant data suggests that elements of C. elegans habituation are likely mediated by changes in vesicle dynamics as well, providing further insight into the molecular mechanism of habituation in C. elegans (Figure 5.1). Seeing as this mechanism shows similarity to the mechanism identified in Aplysia, novel mechanisms discovered in C. elegans may also generalize to other systems. In chapter 4, I describe a highly detailed characterization of a large set of strains with known mutations in genes with predicted neural function. This was achieved using the MWT. I identified an unprecedented number of novel habituation mutants. For perspective, habituation phenotypes of 26 C. elegans strains (with mutations covering 17 genes) had been published during the 10 years prior to the development of the MWT. During our pilot screen in chapter 3, another 29 strains were phenotyped (3 of which had novel habituation effects). In chapter 4, 522 strains were phenotyped, of which 259 had abnormal habituation. In the rest of the animal kingdom, roughly 50 molecules have been identified to play a role in habituation, however, only a few molecules have been identified in any given animal/behavioural model. Identifying over 260 mutants in one species using one behaviour greatly simplifies the problem of understanding how they work together to produce molecular mechanisms for habituation. Two of the mutants were followed up with second allele analysis to confirm that goa-1 and eat-16 are important for habituation. It is unclear at this point where or how they are acting in the tap withdrawal circuit, however, one hypothesis might be that they are working in the mechanosensory 156  neurons (e.g. ALML) to inhibit the downstream signaling of dop-1 (Figure 5.1) since these mutants have the opposite phenotype to the dopamine signaling mutants and goa-1/eat-16 have been found to antagonistically interact with egl-30/egl-8 in locomotion and egglaying behaviours in C. elegans (HajduCronin et al., 1999).  5.1. Parametrics of habituation of the tap withdrawal response in C. elegans One of the major findings from chapter 2 of this dissertation was the fact that the presence of E. coli significantly changed the rate of habituation. We were able to quantitatively show this and identify a mechanism by which we hypothesize it is working (Kindt et al., 2007). In developing the MWT, one of the major advantages of this data-rich, yet high throughput technology was that we could perform many parametric experiments. We were able to replicate the food dependence. I have not experimentally explore all of the parameters that I think might contribute to habituation. However, I believe it may be of use to discuss them, so that others can explore them in more detail, or at least keep them in mind when designing experiments and/or interpreting results from current or future studies. In discussing these parameters, it should be noted that all these observations were made for habituation at a 10 second inter-stimulus interval. It is possible that other habituation protocols may be affected differently. For instance, different inter-stimulus intervals, or patterns of stimulation that lead to intermediate or long-term memory for habituation might be regulated by different factors. The utility of the MWT is that these parameters can rapidly be tested in the future. 5.1.1. Parameters that had little or no effect on habituation Some of the variables that affected neither the initial response nor habituation included circadian rhythm, previous mechanical stimulation and the number of worms on the test plate. Habituation was tested at 3 hours intervals throughout the day and no effect was observed, suggesting that circadian rhythm is not a factor for habituation. While conducting the experiments in chapter 3 and 4, care was taken to avoid excess mechanosensory stimulation immediately prior to testing. However, pilot 157  experiments suggested that C. elegans recover quite rapidly from small amounts of mechanical stimulation. The most extreme example was when worms were transferred by platinum pick to a fresh plate of E. coli immediately before testing. No difference was observed with initial responses or habituation in any of these cases. The fact that transferring to fresh plates had no effect also ruled out any effects resulting from having L1 and L2 larvae and un-hatched eggs on the plate during testing, since results were the same when worms were tested on the plates on which they developed or after transfer to fresh plates. Similarly, the number of adult worms tested per plate did not affect the initial response or habituation (to a maximum of ~150/plate, at which point MWT processing slowed, impacting frame rate). With more worms on the plate, there are more worm to worm collisions. The tracker ignores worms that are touching and object persistence was filtered at 20 seconds to be valid for analysis. For this reason, the effect of males was not assessed because the presence of males made it difficult to maintain a high level of persistent tracking because males constantly attempt to mate with the other worms on the plate causing much more frequent collisions. Certain sex-related phenotypes may be missed if only hermaphrodites are studied. However, previous results suggested that wild-type males and hermaphrodites habituate similarly (Mah and Rankin, 1992), suggesting the risk of missing sex-related effects is low. A more important consideration is that some mutant strains cannot be maintained as hermaphrodite colonies because they spontaneously produce male offspring at a much higher rate than wild-type. If this is not controlled, observed phenotypes may be the result of poor tracking due to the presence of males as opposed to a direct effect of the mutation on habituation. 5.1.2. Habituation parameters A number of variables had an effect on either the tap withdrawal response itself or habituation or both. These include environmental conditions (such as temperature and humidity), the age and recipe of the agar-filled Petri plates, the stimulus characteristics of the tapper (such as stimulus intensity), and 158  both the age of the worms being tested and the age of their parents when they were conceived. I collaborated with another student in the lab, Tiffany Timbers, to do a detailed analysis of worm age and tap stimulus intensity on habituation (Timbers, Giles et al., unpublished). The effect of the others parameters are reported from anecdotal observations. In this section I will briefly describe the conditions that affect tap habituation. 5.1.2.1. Environmental conditions The effect of cultivation temperature has not been directly tested. However, cultivation temperature affects the rate of development and aging in C. elegans (Byerly et al., 1976). As habituation is dependent on age (chapter 4 and explained further below), it likely indirectly affects habituation by regulating the rate of aging, however, this needs to be directly tested to be confirmed. Temperature at the time of testing had a drastic effect on the tap withdrawal response itself. I was unable to evoke reliable tap withdrawal responses at temperatures below 18°C, where the probability of reversal to an initial tap stimulus was 50-60%. The tap withdrawal response could be evoked at high frequencies at temperatures between 19-23°C, which was the temperature range used for all experiments presented in this dissertation. Humidity had a large effect on habituation. When C. elegans are cultured on Petri plates, the plates are generally kept inverted with their lids on. Because the agar has a lot of moisture and the airspace in the 5-cm diameter plates is relatively small, the environment in which the worm is cultured is extremely high (~90%). The MWT was designed to record from an upright plate. If the lid is removed during experiments and the room is 40-50% humidity, then the amount of time the lid has been off of the plate decreases both the likelihood and the size of tap responses. Unfortunately, when the lid is kept on and the plate is upright, evaporation often fogs the lid. To avoid this we modified lids using cover glass treated with an anti-fogging agent. However, Petri plates are not constructed to be airtight, so for longer experiments, loss of moisture may become a problem that needs to be addressed, or at least acknowledged when results are interpreted. 159  Related to humidity is the amount of moisture in the agar. The amount of time since the agar was prepared and poured into the plate, the cooling conditions, and storage condition of the plate before use all contribute to the moisture of the plate at the time of testing. Newly poured plates cause higher frequency and larger tap withdrawal responses than older plates. Other parameters related to plate preparation that were tested included the volume of media in the plate (7-12 ml), percent agar (1.7-2.0 %) in the media as well as the salt concentration (3 – 5 grams NaCl per litre). These variables did not appear to have any direct effect. However, plates with less volume did dry out more rapidly, accelerating the effect of plate-age. 5.1.2.2. Stimulus characteristics Stimulus-related variables did affect habituation. For example, different solenoids, which were used to generate the tap, would elicit slight differences in initial response and habituation. This might be caused by a number of changes in stimulus character. The tap causes a vibration that propagates through the agar of the Petri plate to stimulate the worm. The amplitude of the vibration (intensity), as well as the frequency and duration may affect the response and habituation. Investigating stimulus intensity in more detail, Tiffany Timbers and I found that as it increases, the response probability habituation becomes slower and decrements less (Timbers, Giles et al., unpublished). Stimulus intensity is an important parameter for habituation, as many animals show decreased habituation to stimuli of increased intensity (Thompson and Spencer, 1966). 5.1.2.3. Aging C. elegans develop rapidly, and begin laying eggs approximately 70 hours after they themselves had been laid. They continue to lay eggs until approximately 120 hours after they had been laid. I tested worms between 77 and 94 hours of age and found that habituation of reversal probability of the tap response was negatively correlated with age (chapter 4). This extended the finding that old worms (168 - 288 hours old) habituate more rapidly than worms in their reproductive prime (96 hours)(Beck and Rankin, 1993). Working with a fellow graduate student (Tiffany Timbers), we investigated this effect 160  in more detail by testing worms that were 72 -120 hours old and found that habituation of reversal probability was almost absent in young adult worms (72 hours after egg-lay). As they aged, reversal probability habituation decremented to a deeper level at older ages (Timbers, Giles, et al., unpublished). By depolarizing the mechanosensory neurons of the tap circuit using optogenetics instead of using the tap stimulus, we found no age-dependent changes in habituation, suggesting that the age-dependent effect on habituation was upstream of whole cell depolarization of the sensory neurons. We hypothesized that younger animals were more sensitive to tap stimuli than older animals since the changes in habituation mimicked those observed when varying stimulus intensity. This hypothesis was supported when we tested both young and old worms using various stimulus intensities. The habituation of young animals was more sensitive to intensity manipulations than older animals (Timbers, Giles, et al., unpublished). There could be a number of explanations for this age-dependent intensity change. For example, the cuticle of the animal might thicken with age, the expression of the mechanosensory transduction machinery might decrease with age, or perhaps the expression of voltage-gated calcium channels responsible for amplifying the graded potential induced by transduction might decrease. Another possibility might be related to the dopamine-mediated effect on habituation (Sanyal et al., 2004; Kindt et al., 2007) (chapter 2). The habituation of dopamine deficient mutants and DOP-1 receptor mutants were similar to the habituation that we observed in older wild-type adults. The dopamine-mediated mechanism was hypothesized to affect neural excitability, which is upstream of whole cell depolarization, the putative site of plasticity for the age-dependent effect. Dopamine neurons are known to degenerate in other organisms (Sulzer, 2007). I hypothesize that the dopamine neurons gradually lose function between the ages of 72-120 hours after egg-lay in C. elegans. This loss of dopaminergic neurotransmission negatively modulates the neural excitability of the mechanosensory neurons leading to more rapid habituation. This hypothesis needs to be directly tested; however, if it is true, then C. elegans tap habituation could be a strong model for studying neural degeneration of 161  dopaminergic neurons and could have implications to Parkinson’s Disease, which is caused by the degradation of dopamine neurons in the substantia nigra (Sulzer, 2007). This would complement other C. elegans models that have been developed for this purpose (Nass and Blakely, 2003; Vanduyn et al., 2010). Finally, the last parameter that was qualitatively observed during my work with the MWT was the age of the parents of the worms that were tested. I noticed much more variable results for reversal probability habituation and the initial reversal distance when parents were older than their reproductive prime. I hypothesize that this is because the developmental age that eggs are laid becomes much more variable in older parents. This means the offspring are much less age-synchronized, and because habituation in C. elegans is age-dependent (Timbers, Giles, et al., unpublished) this might lead to more variable results when working with older parents. Since most of these parameters have been qualitatively described instead of quantitatively presented (with the exception of stimulus intensity and worm age; Timber, Giles et al. unpublished), it is important to be cautious of these findings. However, until they have been more thoroughly characterized, taking them into consideration will be important when designing and interpreting the results of experiments investigating habituation. The parameters described here are also not likely to be an exhaustive list, so other parameters that may vary during experimentation should be carefully investigated in the future. This is not a difficult task with the high throughput potential of the MWT.  5.2. Importance of not trading off detailed behavioural analysis for high throughput When we first began developing the MWT, our goal was a system that could quickly screen a mutant library using a relatively low detail behavioural correlate since detailed behavioural algorithms are difficult to design and successfully program. This technology could then be applied using the same approach as used in most genetic screens: a primary screen identifies a large number of potential hits; secondary, tertiary and often more screens, re-test the positives to remove false positives and increase 162  the chance of actually identifying a mutant of interest. Once enough re-screening has been done, the positive hits are fully characterized using a higher detail but slower throughput assay, such as manual scoring or a single worm tracking system. The low detail variable chosen was the average instantaneous speed because it is a very easy variable to measure and when worms respond to a tap by reversing, their speed backwards is much faster than their basal forward speed. Thus the tap withdrawal response can be indirectly assessed by measuring the change in average speed of a population of worms in response to tap. We found that this change in speed decremented with repeated stimulation in wild-type animals. However, the variance was large, suggesting that we would only be able to identify mutants with extreme phenotypes. Also, when we attempted to replicate the phenotypes of previously reported mutants, such as eat-4, cat-2 and dop-1, we were unable to reliably distinguish them from wild-type. Other screens for habituation have had similar problems. For example, a forward mutant screen for tap habituation mutants in C. elegans identified a number of mutants with habituation deficits (Xu et al., 2002). One of the mutants, hab-1, was fully characterized, however, its phenotype could not be reliably scored to map the mutation to a specific gene. Another example comes from a high-throughput behavioural tracking system for Drosophila that was used to attempt a reverse genetic screen for habituation (Sharma et al., 2009). They measured the change in speed of flies in response to the presentation of odours, presumably a similar response to olfactory jump and/or olfactory startle induced locomotion. They tested 150 P-insertion lines and identified 32 potential hits after their primary screen. After re-screening these 32 mutants against appropriate wild-type controls, only one mutant was not ruled out as a false positive. However, after a third round of testing, even that mutant was eliminated as no significant differences from wild-type were observed. These two cases of habituation screens suggest that sacrificing detailed behavioural analysis for higher throughput is not a productive approach.  163  Despite the increased development time involved in creating a behavioural tracker that was capable of both high detail behavioural data acquisition and high throughput, it gave us the ability to fully characterize an entire mutant library, something others have not been able to do. In addition to this, the high detail has yielded important mechanistic information. By being able to measure reversal probability, reversal distance and reversal duration, we identified mutants that are specific to each metric. Not surprisingly, distance and duration were highly correlated, suggesting that the mechanisms that mediate these variables are the same. More importantly, reversal probability did not correlate with distance/duration suggesting that separate mechanisms control the decision to respond to tap versus the decision about how big a response to make. This is an important point because it provides further support for the hypothesis of multiple mechanisms of habituation. This idea was first supported by evidence that different inter-stimulus intervals have different recovery times, an observation that led to the hypothesis that different inter-stimulus intervals are mediated by different mechanisms (Rankin and Broster, 1992). It is unclear how many parallel mechanisms might be involved in fully controlling all the parameters that affect habituation, but it is suggestive that a number of mechanisms are responsible as opposed to a single all-encompassing mechanism.  5.3. Contributions to habituation The focus of my dissertation was to expand our knowledge of habituation. I have achieved this goal in a number of ways. First of all, I have participated in the development of a new tool, the MWT, for the study of habituation and I believe it will revolutionize the field. It has increased the speed with which habituation can be assayed to an unprecedented level without sacrificing detail. In fact, the system is actually capable of higher detail analysis than other slower systems. For perspective, before I began the research for my dissertation, habituation of approximately 50 mutants and pharmacological treatments had been tested over the previous 40 years across all model organisms in attempts to explore the mechanism of habituation. Using the MWT, I have characterized 551 mutants (29 in chapter 3, and 522 in chapter 4) and identified aspects of their habituation that is distinguishable from wild-type 164  in 262 strains in just two years since the development of the MWT was completed. Already, fellow lab members have being using it to expand this list further. Habituation analysis using the MWT is not restricted to mutant analysis. Although, I have not reported any results in my dissertation, the MWT is amenable to RNAi knockdown experiments as well. Genome-wide RNAi libraries exist for C. elegans (eg. Kamath and Ahringer, 2003) making the potential for a genome-wide strategy realistic with the speed with which the MWT can assay habituation. In addition to RNAi knowndown, Evan Ardiel and I have adapted the MWT for population-scale optogenetics (Giles, Ardiel et al., unpublished) creating the possibility for a more detailed analysis of the circuitry by temporally and spatially activating and inactivating neurons while worms are behaving. This is important because although the circuit has been analyzed in detail for the initial tap response (Wicks and Rankin, 1995), more neurons may be important to mediate the plasticity of the response. A good example of this is the dopamine neurons. Wicks and Rankin (1995) did not identify the dopamine neurons as part of the circuit. However, my work in chapter 2 of this dissertation together with other published research (Sanyal et al., 2004; Kindt et al., 2007) provide strong evidence that the dopamine neurons are important for habituation of the tap withdrawal response. It is unclear how many other neurons also contribute to the plasticity of the circuit that mediates habituation, but using an optogenetic approach in tandem with the MWT could solve this problem very quickly. Further contributions to the field of habituation include the discovery of novel parameters important for habituation in C. elegans. The first was the presence of food, which I presented in chapter 2. Research investigating habituation to food or food-related cues had been pursued previously (Review: Epstein et al., 2009). However, this is the first to show that the presence of food affects habituation to a stimulus that is unrelated to food. I found that worms habituate more slowly in the presence of E. coli (their food source) than in the absence. Research in other animals is necessary to find if this parameter is conserved throughout the kingdom like the canonical characteristics of habituation (Thompson and Spencer, 1966). 165  The other parameter for which I presented evidence is the age of the worm (presented in chapter 4). I found that as wild-type C. elegans age from 77 to 94 hours after egg-lay their reversal probability habituation becomes deeper and more rapid. This is consistent with the trend that was previously observed in much older animals (Beck and Rankin, 1993). Studies in other organisms have also found that short-term habituation changes with age (Rattan and Peretz, 1981). The effect is not completely generalizable because in some organisms the opposite trend is found (Fois et al., 1991; Minois and Le Bourg, 1997; Richardson et al., 2011), or no effect has been observed (Le Bourg, 1983; Ellwanger et al., 2003). It has been speculated that habituation may be a useful model for human aging (Kastenbaum, 1980). Perhaps the mechanisms that regulate these age-related effects generalize within the organisms that share the same type of aging effect on habituation. I have also qualitatively described a number of other parameters that I observed during my experimentation; however, more detailed study is necessary before any conclusions can be drawn. I have made contributions to understanding the mechanisms of habituation. I helped identify a role for dopamine in the mechanism for food-dependent modulation of habituation in C. elegans. The evidence suggests that dopamine signals through DOP-1 receptors expressed on the tap sensory neurons activating the G-protein EGL-30, which goes on to trigger an intracellular cascade of phospholipase C beta (EGL-8), and PKC-1 (Kindt et al., 2007) (Figure 5.1). These same molecules disrupt the decrement of the calcium response in the tap sensory neurons after repeated stimulation, a neural correlate of habituation. Thus, it seems possible that the dopamine signaling modulates the tap sensory neurons by affecting their neural excitability, although it is unclear how the signalling cascade leads to changes in neural excitability at this point. Up to this point, the only cellular mechanisms of habituation with experimental evidence to support them were synaptic in nature, such as depression of excitatory synapses (Castellucci and Kandel, 1974; Bailey and Chen, 1988; Gover et al., 2002; Weber et al., 2002) and potentiation of  166  inhibitory synapses (Krasne, 1969; Shirinyan et al., 2006; Das et al., 2011). The fact that changes in neural excitability are potentially an additional cellular mechanism of habituation was an important discovery (Kindt et al., 2007) and suggests that nervous systems have developed multiple ways to mediate habituation. Complementing this finding is the discovery that different sets of genes are necessary for habituation of reversal probability versus reversal distance. Until now, analysis of habituation in other model organisms usually focused on a single behavioural metric. Evidence from chapter 4 of my dissertation suggests that different behavioural metrics may be mediated by different mechanisms because there was no correlation or pattern that relates reversal probability habituation mutants with reversal distance mutants. In contrast, reversal distance mutants and reversal duration mutants were affected by a similar set of mutants. Research in other organisms is needed to determine whether this is a conserved characteristic of habituation. The hypothesis that reversal probability and reversal distance habituation are controlled by different mechanisms further extends the idea of multiple mechanisms of habituation. The prediction that habituation is mediated by multiple mechanisms first developed from the fact that animals recovery from habituation differently depending on the inter-stimulus intervals used during repeated stimulation (Thompson and Spencer, 1966; Broster and Rankin, 1994). Rankin and Broster hypothesized that this difference suggests that there are multiple mechanisms for habituation, some that control habituation at short inter-stimulus intervals and some that control habituation with long inter-stimulus intervals. In chapter 4, I found that there is a much larger difference between the habituated level of groups stimulated using long versus short inter-stimulus intervals in reversal probability compared to reversal distance. This supports the hypothesis that different inter-stimulus intervals mediate different mechanisms of habituation seeing as reversal probability and reversal distance habituation may be controlled by different sets of genes.  167  Finally, my most important contribution to habituation is the complete list of mutants that I have characterized for the various measures of habituation. It is the largest set of mutants to be so fully described for habituation and more importantly they are all described for the same response within the same organism. Much more work is needed to continue this research in order to piece the genes together into a pathway; however, this contribution moves us a large step closer to understanding the mechanism of habituation in C. elegans.  5.4. Future directions Within the vast number of novel habituation mutants identified in chapter 4 of this dissertation, it is clear that habituation researchers have a lot of work before this important biological phenomenon is fully understood. The most obvious route forward is to further investigate the genes mutated in these strains (and their protein products) as potential players in the various mechanisms involved. This can be done by testing secondary alleles or performing RNAi knockdown manipulations for the mutations identified in this work. An analysis of the anatomical expression of these genes using techniques such as GFP fusion and fluorescent microscopy will aid in hypothesizing sites of action. Cell specific transgenetics can then be used to test these hypotheses. Genes or proteins that have similar effects and are expressed in similar locations would be good candidates to work in the same pathway. To test whether two molecules work in the same pathway double mutant analysis can be used. In cases where two genes are thought to physically interact, coimmunoprecipitation experiments can be used. More advanced proteomics approaches might be able to identify complexes with further candidates whose role in habituation can then be tested. It is likely that some of the identified genes have multiple functions throughout the nervous system and in other tissues. To address potential pleiotropy, inducible genetics could be used to show a gene is specifically functioning during habituation and not during development, for instance.  168  Larger mutant libraries can be tested. Over ten thousand mutant strains are available at the Caenorhabditis Genetics Center with mutant alleles in the majority of genes. Genome-wide RNAi libraries could be characterized as well. My dissertation has focused on short-term habituation. Other aspects of habituation need to be investigated in as much depth as I have done here. These include short-term habituation at different inter-stimulus intervals, dishabituation, spontaneous recovery, context habituation, and intermediate and long-term memory for habituation. It will be interesting to see which mutants have phenotypes across multiple types of habituation and which ones are specific to a single protocol. Finally, help is needed from others in the field of habituation to test which of these genes are conserved in other model systems. It is important to investigate whether habituation in various sensory modalities and/or species use a common set of cellular and molecular mechanisms in order to improve our understanding of the most fundamental form of learning, especially because habituation impacts many other behaviours.  169  References Antzoulatos EG, Cleary LJ, Eskin A, Baxter DA, Byrne JH (2003) Desensitization of postsynaptic glutamate receptors contributes to high-frequency homosynaptic depression of aplysia sensorimotor connections. Learning & memory 10:309-313. Aravamudan B, Fergestad T, Davis WS, Rodesch CK, Broadie K (1999) Drosophila UNC-13 is essential for synaptic transmission. Nature neuroscience 2:965-971. Armitage BA, Siegelbaum SA (1998) Presynaptic induction and expression of homosynaptic depression at Aplysia sensorimotor neuron synapses. The Journal of neuroscience : the official journal of the Society for Neuroscience 18:8770-8779. Askew HR (1970) Effects of Stimulus Intensity and Intertrial Interval on Habituation of Head-Shake Response in Rat. J Comp Physiol Psych 72:492-&. Asztalos Z, Arora N, Tully T (2007a) Olfactory jump reflex habituation in Drosophila and effects of classical conditioning mutations. J Neurogenet 21:1-18. Asztalos Z, Baba K, Yamamoto D, Tully T (2007b) The fickle mutation of a cytoplasmic tyrosine kinase effects sensitization but not dishabituation in Drosophila melanogaster. J Neurogenet 21:59-71. Asztalos Z, von Wegerer J, Wustmann G, Dombradi V, Gausz J, Spatz HC, Friedrich P (1993) Protein phosphatase 1-deficient mutant Drosophila is affected in habituation and associative learning. J Neurosci 13:924-930. Augustin I, Rosenmund C, Sudhof TC, Brose N (1999) Munc13-1 is essential for fusion competence of glutamatergic synaptic vesicles. Nature 400:457-461. Baba K, Takeshita A, Majima K, Ueda R, Kondo S, Juni N, Yamamoto D (1999) The Drosophila Bruton's tyrosine kinase (Btk) homolog is required for adult survival and male genital formation. Mol Cell Biol 19:4405-4413. Bailey CH, Chen M (1988) Morphological basis of short-term habituation in Aplysia. The Journal of neuroscience : the official journal of the Society for Neuroscience 8:2452-2459. Barstead RJ, Moerman DG (2006) C. elegans deletion mutant screening. Methods Mol Biol 351:51-58. Beck CD, Rankin CH (1995) Heat shock disrupts long-term memory consolidation in Caenorhabditis elegans. Learn Mem 2:161-177. Beck CDO, Rankin CH (1993) Effects of Aging on Habituation in the Nematode Caenorhabditis-Elegans. Behav Process 28:145-163. 170  Beck CDO, Rankin CH (1997) Long-term habituation is produced by distributed training at long ISIs and not by massed training or short ISIs in Caenorhabditis elegans. Anim Learn Behav 25:446-457. Best AR, Wilson DA (2004) Coordinate synaptic mechanisms contributing to olfactory cortical adaptation. J Neurosci 24:652-660. Best AR, Thompson JV, Fletcher ML, Wilson DA (2005) Cortical metabotropic glutamate receptors contribute to habituation of a simple odor-evoked behavior. J Neurosci 25:2513-2517. Best JD, Berghmans S, Hunt JJ, Clarke SC, Fleming A, Goldsmith P, Roach AG (2008) Non-associative learning in larval zebrafish. Neuropsychopharmacology 33:1206-1215. Beverly M, Anbil S, Sengupta P (2011) Degeneracy and Neuromodulation among Thermosensory Neurons Contribute to Robust Thermosensory Behaviors in Caenorhabditis elegans. Journal of Neuroscience 31:11718-11727. Bianchi L, Kwok SM, Driscoll M, Sesti F (2003) A potassium channel-MiRP complex controls neurosensory function in Caenorhabditis elegans. J Biol Chem 278:12415-12424. Biron D, Wasserman S, Thomas JH, Samuel ADT, Sengupta P (2008) An olfactory neuron responds stochastically to temperature and modulates Caenorhabditis elegans thermotactic behavior. P Natl Acad Sci USA 105:11002-11007. Bouhouche A, Vaysse G, Corbiere M (1993) Immunocytochemical and learning studies of a Drosophila melanogaster neurological mutant, no-bridgeKS49 as an approach to the possible role of the central complex. Journal of neurogenetics 9:105-121. Boynton S, Tully T (1992) latheo, a new gene involved in associative learning and memory in Drosophila melanogaster, identified from P element mutagenesis. Genetics 131:655-672. Brenner S (1974) The genetics of Caenorhabditis elegans. Genetics 77:71-94. Brockie PJ, Mellem JE, Hills T, Madsen DM, Maricq AV (2001) The C. elegans glutamate receptor subunit NMR-1 is required for slow NMDA-activated currents that regulate reversal frequency during locomotion. Neuron 31:617-630. Broster BS, Rankin CH (1994) Effects of changing interstimulus interval during habituation in Caenorhabditis elegans. Behav Neurosci 108:1019-1029. Byerly L, Cassada RC, Russell RL (1976) The life cycle of the nematode Caenorhabditis elegans. I. Wildtype growth and reproduction. Developmental biology 51:23-33.  171  Byrne AB, Weirauch MT, Wong V, Koeva M, Dixon SJ, Stuart JM, Roy PJ (2007) A global analysis of genetic interactions in Caenorhabditis elegans. J Biol 6:8. Byrne JH (1982) Analysis of synaptic depression contributing to habituation of gill-withdrawal reflex in Aplysia californica. J Neurophysiol 48:431-438. Cai SQ, Wang Y, Park KH, Tong X, Pan Z, Sesti F (2009) Auto-phosphorylation of a voltage-gated K+ channel controls non-associative learning. EMBO J 28:1601-1611. Carr DB, Cooper DC, Ulrich SL, Spruston N, Surmeier DJ (2002) Serotonin receptor activation inhibits sodium current and dendritic excitability in prefrontal cortex via a protein kinase C-dependent mechanism. The Journal of neuroscience : the official journal of the Society for Neuroscience 22:6846-6855. Carre-Pierrat M, Baillie D, Johnsen R, Hyde R, Hart A, Granger L, Segalat L (2006) Characterization of the Caenorhabditis elegans G protein-coupled serotonin receptors. Invert Neurosci 6:189-205. Castellucci V, Pinsker H, Kupfermann I, Kandel ER (1970) Neuronal mechanisms of habituation and dishabituation of the gill-withdrawal reflex in Aplysia. Science 167:1745-1748. Castellucci VF, Kandel ER (1974) A quantal analysis of the synaptic depression underlying habituation of the gill-withdrawal reflex in Aplysia. P Natl Acad Sci USA 71:5004-5008. Castillo PE, Chiu CQ, Carroll RC (2011) Long-term plasticity at inhibitory synapses. Curr Opin Neurobiol 21:328-338. Cesca F, Baldelli P, Valtorta F, Benfenati F (2010) The synapsins: key actors of synapse function and plasticity. Prog Neurobiol 91:313-348. Chalfie M, Sulston J (1981) Developmental genetics of the mechanosensory neurons of Caenorhabditis elegans. Developmental biology 82:358-370. Chalfie M, Au M (1989) Genetic control of differentiation of the Caenorhabditis elegans touch receptor neurons. Science 243:1027-1033. Chalfie M, Tu Y, Euskirchen G, Ward WW, Prasher DC (1994) Green fluorescent protein as a marker for gene expression. Science 263:802-805. Chalfie M, Sulston JE, White JG, Southgate E, Thomson JN, Brenner S (1985) The Neural Circuit for Touch Sensitivity in Caenorhabditis-Elegans. Journal of Neuroscience 5:956-964. Chao MY, Larkins-Ford J, Tucey TM, Hart AC (2005) lin-12 Notch functions in the adult nervous system of C. elegans. BMC neuroscience 6:45. 172  Chase DL, Pepper JS, Koelle MR (2004) Mechanism of extrasynaptic dopamine signaling in Caenorhabditis elegans. Nature neuroscience 7:1096-1103. Chen G, Greengard P, Yan Z (2004) Potentiation of NMDA receptor currents by dopamine D1 receptors in prefrontal cortex. P Natl Acad Sci USA 101:2596-2600. Chew SJ, Vicario DS, Nottebohm F (1996) A large-capacity memory system that recognizes the calls and songs of individual birds. P Natl Acad Sci USA 93:1950-1955. Chew SJ, Mello C, Nottebohm F, Jarvis E, Vicario DS (1995) Decrements in Auditory Responses to a Repeated Conspecific Song Are Long-Lasting and Require 2 Periods of Protein-Synthesis in the Songbird Forebrain. P Natl Acad Sci USA 92:3406-3410. Chiba CM, Rankin CH (1990) A developmental analysis of spontaneous and reflexive reversals in the nematode Caenorhabditis elegans. J Neurobiol 21:543-554. Cho S, Rogers KW, Fay DS (2007) The C. elegans glycopeptide hormone receptor ortholog, FSHR-1, regulates germline differentiation and survival. Current biology : CB 17:203-212. Cho W, Heberlein U, Wolf FW (2004a) Habituation of an odorant-induced startle response in Drosophila. Genes, brain, and behavior 3:127-137. Cho W, Heberlein U, Wolf FW (2004b) Habituation of an odorant-induced startle response in Drosophila. Genes Brain Behav 3:127-137. Cohen TE, Kaplan SW, Kandel ER, Hawkins RD (1997) A simplified preparation for relating cellular events to behavior: mechanisms contributing to habituation, dishabituation, and sensitization of the Aplysia gill-withdrawal reflex. J Neurosci 17:2886-2899. Colbert HA, Bargmann CI (1995) Odorant-specific adaptation pathways generate olfactory plasticity in C. elegans. Neuron 14:803-812. Consortium TCeS (1998) Genome sequence of the nematode C. elegans: a platform for investigating biology. Science 282:2012-2018. Corfas G, Dudai Y (1989) Habituation and dishabituation of a cleaning reflex in normal and mutant Drosophila. The Journal of neuroscience : the official journal of the Society for Neuroscience 9:56-62. Cronin CJ, Feng Z, Schafer WR (2006) Automated imaging of C. elegans behavior. Methods Mol Biol 351:241-251.  173  Das S, Sadanandappa MK, Dervan A, Larkin A, Lee JA, Sudhakaran IP, Priya R, Heidari R, Holohan EE, Pimentel A, Gandhi A, Ito K, Sanyal S, Wang JW, Rodrigues V, Ramaswami M (2011) Plasticity of local GABAergic interneurons drives olfactory habituation. P Natl Acad Sci USA 108:E646-654. Davies AM (1994) Neural development. Chemoattractants for navigating axons. Current biology : CB 4:1142-1145. Davis M (1970a) Interstimulus Interval and Startle Response Habituation with a Control for Total Time during Training. Psychon Sci 20:39-41. Davis M (1970b) Effects of Interstimulus Interval Length and Variability on Startle-Response Habituation in Rat. J Comp Physiol Psych 72:177-&. Davis M, Wagner AR (1969) Habituation of Startle Response under Incremental Sequence of Stimulus Intensities. J Comp Physiol Psych 67:486-&. Davis M, Gendelman DS, Tischler MD, Gendelman PM (1982a) A Primary Acoustic Startle Circuit - Lesion and Stimulation Studies. Journal of Neuroscience 2:791-805. Davis M, Parisi T, Gendelman DS, Tischler M, Kehne JH (1982b) Habituation and Sensitization of Startle Reflexes Elicited Electrically from the Brain-Stem. Science 218:688-690. Deacon RM (2006) Housing, husbandry and handling of rodents for behavioral experiments. Nat Protoc 1:936-946. Dong S, Replogle KL, Hasadsri L, Imai BS, Yau PM, Rodriguez-Zas S, Southey BR, Sweedler JV, Clayton DF (2009) Discrete molecular states in the brain accompany changing responses to a vocal signal. Proc Natl Acad Sci U S A 106:11364-11369. Duerr JS, Quinn WG (1982) Three Drosophila mutations that block associative learning also affect habituation and sensitization. P Natl Acad Sci USA 79:3646-3650. Ellwanger J, Geyer MA, Braff DL (2003) The relationship of age to prepulse inhibition and habituation of the acoustic startle response. Biol Psychol 62:175-195. Engel JE, Wu CF (1996) Altered habituation of an identified escape circuit in Drosophila memory mutants. J Neurosci 16:3486-3499. Engel JE, Wu CF (1998) Genetic dissection of functional contributions of specific potassium channel subunits in habituation of an escape circuit in Drosophila. J Neurosci 18:2254-2267. Engel JE, Wu CF (2009) Neurogenetic approaches to habituation and dishabituation in Drosophila. Neurobiol Learn Mem 92:166-175. 174  Engel JE, Xie XJ, Sokolowski MB, Wu CF (2000) A cGMP-dependent protein kinase gene, foraging, modifies habituation-like response decrement of the giant fiber escape circuit in Drosophila. Learn Mem 7:341-352. Epstein LH, Temple JL, Roemmich JN, Bouton ME (2009) Habituation as a determinant of human food intake. Psychological review 116:384-407. Ernstrom GG, Chalfie M (2002) Genetics of sensory mechanotransduction. Annu Rev Genet 36:411-453. Ezcurra M, Tanizawa Y, Swoboda P, Schafer WR (2011) Food sensitizes C. elegans avoidance behaviours through acute dopamine signalling. Embo Journal 30:1110-1122. Feany MB, Quinn WG (1995) A neuropeptide gene defined by the Drosophila memory mutant amnesiac. Science 268:869-873. Fischbach K-F (1981) Habituation and sensitization of the landing response of Drosophila melanogaster. Naturwissenschaften 68:332. Fischbach K-F, Bausenwein B (1988) Habituation and sensitization of the landing response of Drosophila melanogaster: II. Receptive field size of habituating units. In: Modulation of Synaptic Transmission and Plasticity in Nervous Systems (Hertting G, Spatz HC, eds), pp 369-385. Berlin Heidelberg: Springer-Verlag. Fleming JT, Squire MD, Barnes TM, Tornoe C, Matsuda K, Ahnn J, Fire A, Sulston JE, Barnard EA, Sattelle DB, Lewis JA (1997) Caenorhabditis elegans levamisole resistance genes lev-1, unc-29, and unc38 encode functional nicotinic acetylcholine receptor subunits. The Journal of neuroscience : the official journal of the Society for Neuroscience 17:5843-5857. Fois C, Medioni J, Le Bourg E (1991) Habituation of the proboscis extension response as a function of age in Drosophila melanogaster. Gerontology 37:187-192. Frost L, Kaplan SW, Cohen TE, Henzi V, Kandel ER, Hawkins RD (1997) A simplified preparation for relating cellular events to behavior: contribution of LE and unidentified siphon sensory neurons to mediation and habituation of the Aplysia gill- and siphon-withdrawal reflex. J Neurosci 17:2900-2913. Fujiwara M, Ishihara T, Katsura I (1999) A novel WD40 protein, CHE-2, acts cell-autonomously in the formation of C. elegans sensory cilia. Development 126:4839-4848. Gibbs L (2002) Fable 269. The Fox and the Lion. In: Aesop's Fables. A new translation by Laura Gibbs. Oxford: Oxford University Press (World's Classics). Giles AC, Rankin CH (2009) Behavioral and genetic characterization of habituation using Caenorhabditis elegans. Neurobiol Learn Mem 92:139-146. 175  Giles AC, Rose JK, Rankin CH (2006) Investigations of learning and memory in Caenorhabditis elegans. Int Rev Neurobiol 69:37-71. Gingrich KJ, Byrne JH (1985) Simulation of synaptic depression, posttetanic potentiation, and presynaptic facilitation of synaptic potentials from sensory neurons mediating gill-withdrawal reflex in Aplysia. Journal of neurophysiology 53:652-669. Godenschwege TA et al. (2004) Flies lacking all synapsins are unexpectedly healthy but are impaired in complex behaviour. Eur J Neurosci 20:611-622. Goodman MB, Hall DH, Avery L, Lockery SR (1998) Active currents regulate sensitivity and dynamic range in C. elegans neurons. Neuron 20:763-772. Gottschalk A, Almedom RB, Schedletzky T, Anderson SD, Yates JR, 3rd, Schafer WR (2005) Identification and characterization of novel nicotinic receptor-associated proteins in Caenorhabditis elegans. The EMBO journal 24:2566-2578. Gover TD, Abrams TW (2009) Insights into a molecular switch that gates sensory neuron synapses during habituation in Aplysia. Neurobiol Learn Mem 92:155-165. Gover TD, Jiang XY, Abrams TW (2002) Persistent, exocytosis-independent silencing of release sites underlies homosynaptic depression at sensory synapses in Aplysia. The Journal of neuroscience : the official journal of the Society for Neuroscience 22:1942-1955. Grace AA, Floresco SB, Goto Y, Lodge DJ (2007) Regulation of firing of dopaminergic neurons and control of goal-directed behaviors. Trends Neurosci 30:220-227. Gracheva EO, Burdina AO, Touroutine D, Berthelot-Grosjean M, Parekh H, Richmond JE (2007a) Tomosyn negatively regulates both synaptic transmitter and neuropeptide release at the C. elegans neuromuscular junction. J Physiol 585:705-709. Gracheva EO, Burdina AO, Touroutine D, Berthelot-Grosjean M, Parekh H, Richmond JE (2007b) Tomosyn negatively regulates CAPS-dependent peptide release at Caenorhabditis elegans synapses. The Journal of neuroscience : the official journal of the Society for Neuroscience 27:10176-10184. Gracheva EO, Burdina AO, Holgado AM, Berthelot-Grosjean M, Ackley BD, Hadwiger G, Nonet ML, Weimer RM, Richmond JE (2006) Tomosyn inhibits synaptic vesicle priming in Caenorhabditis elegans. PLoS Biol 4:e261. Gray JM, Hill JJ, Bargmann CI (2005) A circuit for navigation in Caenorhabditis elegans. P Natl Acad Sci USA 102:3184-3191.  176  Green RA, Kao HL, Audhya A, Arur S, Mayers JR, Fridolfsson HN, Schulman M, Schloissnig S, Niessen S, Laband K, Wang S, Starr DA, Hyman AA, Schedl T, Desai A, Piano F, Gunsalus KC, Oegema K (2011) A High-Resolution C. elegans Essential Gene Network Based on Phenotypic Profiling of a Complex Tissue. Cell 145:470-482. Griffith LC (2004) Calcium/calmodulin-dependent protein kinase II: an unforgettable kinase. The Journal of neuroscience : the official journal of the Society for Neuroscience 24:8391-8393. Groves PM, Thompson RF (1970) Habituation: A dual-process theory. Psychological Review 77:419-450. Guan KL, Han M (1999) A G-protein signaling network mediated by an RGS protein. Genes & development 13:1763-1767. Gulledge AT, Jaffe DB (1998) Dopamine decreases the excitability of layer V pyramidal cells in the rat prefrontal cortex. The Journal of neuroscience : the official journal of the Society for Neuroscience 18:9139-9151. Gunaratne PH, Lin YC, Benham AL, Drnevich J, Coarfa C, Tennakoon JB, Creighton CJ, Kim JH, Milosavljevic A, Watson M, Griffiths-Jones S, Clayton DF (2011) Song exposure regulates known and novel microRNAs in the zebra finch auditory forebrain. Bmc Genomics 12. Hajdu-Cronin YM, Chen WJ, Patikoglou G, Koelle MR, Sternberg PW (1999) Antagonism between G(o)alpha and G(q)alpha in Caenorhabditis elegans: the RGS protein EAT-16 is necessary for G(o)alpha signaling and regulates G(q)alpha activity. Genes & development 13:1780-1793. Harris JD (1943) Habituatroy response decremtn in the intact organism. Psychological Bulletin 40:385422. Hart AC, Sims S, Kaplan JM (1995) Synaptic code for sensory modalities revealed by C. elegans GLR-1 glutamate receptor. Nature 378:82-85. Hart AC, Kass J, Shapiro JE, Kaplan JM (1999) Distinct signaling pathways mediate touch and osmosensory responses in a polymodal sensory neuron. Journal of Neuroscience 19:1952-1958. Hedgecock EM, Russell RL (1975) Normal and mutant thermotaxis in the nematode Caenorhabditis elegans. P Natl Acad Sci USA 72:4061-4065. Henze DA, Gonzalez-Burgos GR, Urban NN, Lewis DA, Barrionuevo G (2000) Dopamine increases excitability of pyramidal neurons in primate prefrontal cortex. Journal of neurophysiology 84:2799-2809. Hilliard MA, Apicella AJ, Kerr R, Suzuki H, Bazzicalupo P, Schafer WR (2005) In vivo imaging of C. elegans ASH neurons: cellular response and adaptation to chemical repellents. EMBO J 24:63-72.  177  Hills T, Brockie PJ, Maricq AV (2004) Dopamine and glutamate control area-restricted search behavior in Caenorhabditis elegans. The Journal of neuroscience : the official journal of the Society for Neuroscience 24:1217-1225. Horn T, Sandmann T, Fischer B, Axelsson E, Huber W, Boutros M (2011) Mapping of signaling networks through synthetic genetic interaction analysis by RNAi. Nature methods 8:341-346. Houle D, Govindaraju DR, Omholt S (2010) Phenomics: the next challenge. Nature reviews Genetics 11:855-866. Huang KM, Cosman P, Schafer WR (2006) Machine vision based detection of omega bends and reversals in C. elegans. Journal of neuroscience methods 158:323-336. Huang M, Gu G, Ferguson EL, Chalfie M (1995) A stomatin-like protein necessary for mechanosensation in C. elegans. Nature 378:292-295. Huang X, Huang P, Robinson MK, Stern MJ, Jin Y (2003) UNC-71, a disintegrin and metalloprotease (ADAM) protein, regulates motor axon guidance and sex myoblast migration in C. elegans. Development 130:3147-3161. Humphrey G (1933) The Nature of Learning in its Relation to the Living System. New York: Harcourt, Brace and Company. Jansen G, Weinkove D, Plasterk RH (2002) The G-protein gamma subunit gpc-1 of the nematode C.elegans is involved in taste adaptation. The EMBO journal 21:986-994. Jayanthi LD, Apparsundaram S, Malone MD, Ward E, Miller DM, Eppler M, Blakely RD (1998) The Caenorhabditis elegans gene T23G5.5 encodes an antidepressant- and cocaine-sensitive dopamine transporter. Mol Pharmacol 54:601-609. Jiang XY, Abrams TW (1998) Use-dependent decline of paired-pulse facilitation at Aplysia sensory neuron synapses suggests a distinct vesicle pool or release mechanism. The Journal of neuroscience : the official journal of the Society for Neuroscience 18:10310-10319. Jin P, Griffith LC, Murphey RK (1998) Presynaptic calcium/calmodulin-dependent protein kinase II regulates habituation of a simple reflex in adult Drosophila. The Journal of neuroscience : the official journal of the Society for Neuroscience 18:8955-8964. Joiner MA, Asztalos Z, Jones CJ, Tully T, Wu CF (2007) Effects of mutant Drosophila K+ channel subunits on habituation of the olfactory jump response. J Neurogenet 21:45-58. Jope RS, Johnson GV (2004) The glamour and gloom of glycogen synthase kinase-3. Trends Biochem Sci 29:95-102.  178  Joseph MH, Datla K, Young AM (2003) The interpretation of the measurement of nucleus accumbens dopamine by in vivo dialysis: the kick, the craving or the cognition? Neurosci Biobehav Rev 27:527-541. Kamath RS, Ahringer J (2003) Genome-wide RNAi screening in Caenorhabditis elegans. Methods 30:313321. Kang L, Gao J, Schafer WR, Xie Z, Xu XZ (2010) C. elegans TRP family protein TRP-4 is a pore-forming subunit of a native mechanotransduction channel. Neuron 67:381-391. Kaplan JM, Horvitz HR (1993) A dual mechanosensory and chemosensory neuron in Caenorhabditis elegans. Proc Natl Acad Sci U S A 90:2227-2231. Kastenbaum RJ (1980) Habituation as a model of human aging. Int J Aging Hum Dev 12:159-170. Kaun KR, Sokolowski MB (2009) cGMP-dependent protein kinase: linking foraging to energy homeostasis. Genome 52:1-7. Khoutorsky A, Spira ME (2005) Calcium-activated proteases are critical for refilling depleted vesicle stores in cultured sensory-motor synapses of Aplysia. Learning & memory 12:414-422. Kindt KS, Quast KB, Giles AC, De S, Hendrey D, Nicastro I, Rankin CH, Schafer WR (2007) Dopamine mediates context-dependent modulation of sensory plasticity in C. elegans. Neuron 55:662-676. Kitamura K, Amano S, Hosono R (2001) Contribution of neurons to habituation to mechanical stimulation in Caenorhabditis elegans. Journal of Neurobiology 46:29-40. Klein M, Shapiro E, Kandel ER (1980) Synaptic plasticity and the modulation of the Ca2+ current. The Journal of experimental biology 89:117-157. Kohn RE, Duerr JS, McManus JR, Duke A, Rakow TL, Maruyama H, Moulder G, Maruyama IN, Barstead RJ, Rand JB (2000) Expression of multiple UNC-13 proteins in the Caenorhabditis elegans nervous system. Mol Biol Cell 11:3441-3452. Krasne FB (1969) Excitation and habituation of the crayfish escape reflex: the depolarizing response in lateral giant fibres of the isolated abdomen. The Journal of experimental biology 50:29-46. Krishna S, Maduzia LL, Padgett RW (1999) Specificity of TGFbeta signaling is conferred by distinct type I receptors and their associated SMAD proteins in Caenorhabditis elegans. Development 126:251260. Kruse AA, Stripling R, Clayton DF (2000) Minimal experience required for immediate-early gene induction in zebra finch neostriatum. Neurobiology of Learning and Memory 74:179-184. 179  Kuhara A, Okumura M, Kimata T, Tanizawa Y, Takano R, Kimura KD, Inada H, Matsumoto K, Mori I (2008) Temperature sensing by an olfactory neuron in a circuit controlling behavior of C-elegans. Science 320:803-807. Kupfermann I, Carew TJ, Kandel ER (1974) Local, reflex, and central commands controlling gill and siphon movements in Aplysia. Journal of neurophysiology 37:996-1019. Kupfermann I, Castellucci V, Pinsker H, Kandel E (1970) Neuronal correlates of habituation and dishabituation of the gill-withdrawal reflex in Aplysia. Science 167:1743-1745. Lackner MR, Nurrish SJ, Kaplan JM (1999) Facilitation of synaptic transmission by EGL-30 Gqalpha and EGL-8 PLCbeta: DAG binding to UNC-13 is required to stimulate acetylcholine release. Neuron 24:335-346. Lavin A, Grace AA (2001) Stimulation of D1-type dopamine receptors enhances excitability in prefrontal cortical pyramidal neurons in a state-dependent manner. Neuroscience 104:335-346. Le Bourg E (1983) Aging and habituation of the tarsal response in Drosophila melanogaster. Gerontology 29:388-393. Lee RYN, Sawin ER, Chalfie M, Horvitz HR, Avery L (1999) EAT-4, a homolog of a mammalian sodiumdependent inorganic phosphate cotransporter, is necessary for glutamatergic neurotransmission in Caenorhabditis elegans. Journal of Neuroscience 19:159-167. Lewis JA, Wu CH, Berg H, Levine JH (1980) The genetics of levamisole resistance in the nematode Caenorhabditis elegans. Genetics 95:905-928. Li W, Feng Z, Sternberg PW, Xu XZ (2006) A C. elegans stretch receptor neuron revealed by a mechanosensitive TRP channel homologue. Nature 440:684-687. Li W, Gao B, Lee SM, Bennett K, Fang D (2007) RLE-1, an E3 ubiquitin ligase, regulates C. elegans aging by catalyzing DAF-16 polyubiquitination. Dev Cell 12:235-246. Lindemans M, Janssen T, Husson SJ, Meelkop E, Temmerman L, Clynen E, Mertens I, Schoofs L (2009) A neuromedin-pyrokinin-like neuropeptide signaling system in Caenorhabditis elegans. Biochem Biophys Res Commun 379:760-764. Lingenhohl K, Friauf E (1994) Giant-Neurons in the Rat Reticular-Formation - a Sensorimotor Interface in the Elementary Acoustic Startle Circuit. Journal of Neuroscience 14:1176-1194. Lints R, Emmons SW (1999) Patterning of dopaminergic neurotransmitter identity among Caenorhabditis elegans ray sensory neurons by a TGFbeta family signaling pathway and a Hox gene. Development 126:5819-5831.  180  Lu CS, Hodge JJ, Mehren J, Sun XX, Griffith LC (2003) Regulation of the Ca2+/CaM-responsive pool of CaMKII by scaffold-dependent autophosphorylation. Neuron 40:1185-1197. Mah KB, Rankin CH (1992) An analysis of behavioral plasticity in male Caenorhabditis elegans. Behav Neural Biol 58:211-221. Maricq AV, Peckol E, Driscoll M, Bargmann CI (1995) Mechanosensory signalling in C. elegans mediated by the GLR-1 glutamate receptor. Nature 378:78-81. Marinesco S, Carew TJ (2002) Serotonin release evoked by tail nerve stimulation in the CNS of aplysia: characterization and relationship to heterosynaptic plasticity. The Journal of neuroscience : the official journal of the Society for Neuroscience 22:2299-2312. Martin JR, Ollo R (1996) A new Drosophila Ca2+/calmodulin-dependent protein kinase (Caki) is localized in the central nervous system and implicated in walking speed. The EMBO journal 15:1865-1876. Matsuki M, Kunitomo H, Iino Y (2006) Goalpha regulates olfactory adaptation by antagonizing GqalphaDAG signaling in Caenorhabditis elegans. P Natl Acad Sci USA 103:1112-1117. McDonald PW, Hardie SL, Jessen TN, Carvelli L, Matthies DS, Blakely RD (2007) Vigorous motor activity in Caenorhabditis elegans requires efficient clearance of dopamine mediated by synaptic localization of the dopamine transporter DAT-1. The Journal of neuroscience : the official journal of the Society for Neuroscience 27:14216-14227. McKenna M, Monte P, Helfand SL, Woodard C, Carlson J (1989) A simple chemosensory response in Drosophila and the isolation of acj mutants in which it is affected. Proc Natl Acad Sci U S A 86:8118-8122. Megighian A, Zordan M, Costa R (2001) Giant neuron pathway neurophysiological activity in per(0) mutants of Drosophila melanogaster. J Neurogenet 15:221-231. Mellem JE, Brockie PJ, Zheng Y, Madsen DM, Maricq AV (2002) Decoding of polymodal sensory stimuli by postsynaptic glutamate receptors in C. elegans. Neuron 36:933-944. Mello CV, Clayton DF (1994) Song-Induced Zenk Gene-Expression in Auditory Pathways of Songbird Brain and Its Relation to the Song Control-System. Journal of Neuroscience 14:6652-6666. Mello CV, Clayton DF (1995) Differential Induction of the Zenk Gene in the Avian Forebrain and Song Control-Circuit after Metrazole-Induced Depolarization. Journal of Neurobiology 26:145-161. Mello CV, Vicario DS, Clayton DF (1992) Song Presentation Induces Gene-Expression in the Songbird Forebrain. P Natl Acad Sci USA 89:6818-6822.  181  Miller KG, Emerson MD, Rand JB (1999) Goalpha and diacylglycerol kinase negatively regulate the Gqalpha pathway in C. elegans. Neuron 24:323-333. Minois N, Le Bourg E (1997) Hypergravity and aging in Drosophila melanogaster. 9. Conditioned suppression and habituation of the proboscis extension response. Aging (Milano) 9:281-291. Miyawaki A, Llopis J, Heim R, McCaffery JM, Adams JA, Ikura M, Tsien RY (1997) Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature 388:882-887. Mori I, Ohshima Y (1995) Neural Regulation of Thermotaxis in Caenorhabditis-Elegans. Nature 376:344348. Morita K, Chow KL, Ueno N (1999) Regulation of body length and male tail ray pattern formation of Caenorhabditis elegans by a member of TGF-beta family. Development 126:1337-1347. Nass R, Blakely RD (2003) The Caenorhabditis elegans dopaminergic system: opportunities for insights into dopamine transport and neurodegeneration. Annu Rev Pharmacol Toxicol 43:521-544. Nguyen M, Alfonso A, Johnson CD, Rand JB (1995) Caenorhabditis elegans mutants resistant to inhibitors of acetylcholinesterase. Genetics 140:527-535. Nugent FS, Penick EC, Kauer JA (2007) Opioids block long-term potentiation of inhibitory synapses. Nature 446:1086-1090. Nurrish S, Segalat L, Kaplan JM (1999) Serotonin inhibition of synaptic transmission: Galpha(0) decreases the abundance of UNC-13 at release sites. Neuron 24:231-242. O'Hagan R, Chalfie M, Goodman MB (2005) The MEC-4 DEG/ENaC channel of Caenorhabditis elegans touch receptor neurons transduces mechanical signals. Nature neuroscience 8:43-50. Ogmen H, Moussa M (1993) A neural model for nonassociative learning in a prototypical sensory-motor scheme: the landing reaction in flies. Biol Cybern 68:351-361. Osborne KA, Robichon A, Burgess E, Butland S, Shaw RA, Coulthard A, Pereira HS, Greenspan RJ, Sokolowski MB (1997) Natural behavior polymorphism due to a cGMP-dependent protein kinase of Drosophila. Science 277:834-836. Patterson TL, Petrinovich L (1979) Field Studies of Habituation .2. Effect of Massed StimulusPresentation. J Comp Physiol Psych 93:351-359. Peckham GW, Peckham EG (1887) Some observations on the mental powers of spiders. Journal of Morphology 1:383-419.  182  Perkins LA, Hedgecock EM, Thomson JN, Culotti JG (1986) Mutant sensory cilia in the nematode Caenorhabditis elegans. Developmental biology 117:456-487. Petrinovich L, Patterson TL (1979) Field Studies of Habituation .1. Effect of Reproductive Condition, Number of Trials, and Different Delay Intervals on Responses of the White-Crowned Sparrow. J Comp Physiol Psych 93:337-350. Petrinovich L, Patterson TL (1980) Field Studies of Habituation - .3. Playback Contingent on the Response of the White-Crowned Sparrow. Anim Behav 28:742-751. Phillips AG, Vacca G, Ahn S (2008) A top-down perspective on dopamine, motivation and memory. Pharmacol Biochem Behav 90:236-249. Pierce-Shimomura JT, Morse TM, Lockery SR (1999) The fundamental role of pirouettes in Caenorhabditis elegans chemotaxis. The Journal of neuroscience : the official journal of the Society for Neuroscience 19:9557-9569. Pinsker H, Kupfermann I, Castellucci V, Kandel E (1970) Habituation and dishabituation of the gillwithdrawal reflex in Aplysia. Science 167:1740-1742. Ramot D, Johnson BE, Berry TL, Jr., Carnell L, Goodman MB (2008) The Parallel Worm Tracker: a platform for measuring average speed and drug-induced paralysis in nematodes. PloS one 3:e2208. Rankin CH, Broster BS (1992) Factors affecting habituation and recovery from habituation in the nematode Caenorhabditis elegans. Behav Neurosci 106:239-249. Rankin CH, Wicks SR (2000) Mutations of the caenorhabditis elegans brain-specific inorganic phosphate transporter eat-4 affect habituation of the tap-withdrawal response without affecting the response itself. J Neurosci 20:4337-4344. Rankin CH, Beck CD, Chiba CM (1990) Caenorhabditis elegans: a new model system for the study of learning and memory. Behav Brain Res 37:89-92. Rankin CH, Gannon T, Wicks SR (2000) Developmental analysis of habituation in the Nematode C. elegans. Dev Psychobiol 36:261-270. Rankin CH, Abrams T, Barry RJ, Bhatnagar S, Clayton DF, Colombo J, Coppola G, Geyer MA, Glanzman DL, Marsland S, McSweeney FK, Wilson DA, Wu CF, Thompson RF (2009) Habituation revisited: an updated and revised description of the behavioral characteristics of habituation. Neurobiol Learn Mem 92:135-138. Rattan KS, Peretz B (1981) Age-dependent behavioral changes and physiological changes in identified neurons in Aplysia californica. Journal of neurobiology 12:469-478. 183  Redgrave P, Gurney K (2006) The short-latency dopamine signal: a role in discovering novel actions? Nat Rev Neurosci 7:967-975. Rees CT, Spatz HC (1989) Habituation of the landing response of Drosophila wild-type and mutants defective in olfactory learning. J Neurogenet 5:105-118. Remm M, Storm CE, Sonnhammer EL (2001) Automatic clustering of orthologs and in-paralogs from pairwise species comparisons. J Mol Biol 314:1041-1052. Richardson C, Bucks RS, Hogan AM (2011) Effects of aging on habituation to novelty: an ERP study. Int J Psychophysiol 79:97-105. Richmond J (2007) Synaptic function. In: Wormbook (Community TCeR, ed): Wormbook. Richmond JE, Davis WS, Jorgensen EM (1999) UNC-13 is required for synaptic vesicle fusion in C. elegans. Nature neuroscience 2:959-964. Rose AM, Baillie DL (1980) Genetic organization of the region around UNC-15 (I), a gene affecting paramyosin in Caenorhabditis elegans. Genetics 96:639-648. Rose JK, Sangha S, Rai S, Norman KR, Rankin CH (2005) Decreased sensory stimulation reduces behavioral responding, retards development, and alters neuronal connectivity in Caenorhabditis elegans. J Neurosci 25:7159-7168. Roussel N, Morton CA, Finger FP, Roysam B (2007) A computational model for C. elegans locomotory behavior: application to multiworm tracking. IEEE Trans Biomed Eng 54:1786-1797. Sanyal S, Wintle RF, Kindt KS, Nuttley WM, Arvan R, Fitzmaurice P, Bigras E, Merz DC, Hebert TE, van der Kooy D, Schafer WR, Culotti JG, Van Tol HHM (2004) Dopamine modulates the plasticity of mechanosensory responses in Caenorhabditis elegans. Embo Journal 23:473-482. Savage C, Das P, Finelli AL, Townsend SR, Sun CY, Baird SE, Padgett RW (1996) Caenorhabditis elegans genes sma-2, sma-3, and sma-4 define a conserved family of transforming growth factor beta pathway components. P Natl Acad Sci USA 93:790-794. Sawin ER, Ranganathan R, Horvitz HR (2000) C. elegans locomotory rate is modulated by the environment through a dopaminergic pathway and by experience through a serotonergic pathway. Neuron 26:619-631. Schafer WR, Kenyon CJ (1995) A calcium-channel homologue required for adaptation to dopamine and serotonin in Caenorhabditis elegans. Nature 375:73-78.  184  Schafer WR, Sanchez BM, Kenyon CJ (1996) Genes affecting sensitivity to serotonin in Caenorhabditis elegans. Genetics 143:1219-1230. Scheiner R, Sokolowski MB, Erber J (2004) Activity of cGMP-dependent protein kinase (PKG) affects sucrose responsiveness and habituation in Drosophila melanogaster. Learn Mem 11:303-311. Schmid S, Simons NS, Schnitzler HU (2003) Cellular mechanisms of the trigeminally evoked startle response. Eur J Neurosci 17:1438-1444. Schmid S, Brown T, Simons-Weidenmaier N, Weber M, Fendt M (2010) Group III metabotropic glutamate receptors inhibit startle-mediating giant neurons in the caudal pontine reticular nucleus but do not mediate synaptic depression/short-term habituation of startle. J Neurosci 30:10422-10430. Schmid S, Azzopardi E, De Jaeger X, Prado MA, Prado VF (2011) VAChT knock-down mice show normal prepulse inhibition but disrupted long-term habituation. Genes Brain Behav. Seamans JK, Gorelova NA, Yang CR (1997) Contributions of voltage-gated Ca2+ channels in the proximal versus distal dendrites to synaptic integration in prefrontal cortical neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 17:5936-5948. Sharma P, Keane J, O'Kane CJ, Asztalos Z (2009) Automated measurement of Drosophila jump reflex habituation and its use for mutant screening. J Neurosci Methods 182:43-48. Shingai R (2000) Durations and frequencies of free locomotion in wild type and GABAergic mutants of Caenorhabditis elegans. Neurosci Res 38:71-83. Shirinyan D, Teshiba T, Taylor K, O'Neill P, Lee SC, Krasne FB (2006) Rostral ganglia are required for induction but not expression of crayfish escape reflex habituation: role of higher centers in reprogramming low-level circuits. Journal of neurophysiology 95:2721-2724. Sieburth D, Ch'ng Q, Dybbs M, Tavazoie M, Kennedy S, Wang D, Dupuy D, Rual JF, Hill DE, Vidal M, Ruvkun G, Kaplan JM (2005) Systematic analysis of genes required for synapse structure and function. Nature 436:510-517. Simons-Weidenmaier NS, Weber M, Plappert CF, Pilz PK, Schmid S (2006) Synaptic depression and shortterm habituation are located in the sensory part of the mammalian startle pathway. BMC Neurosci 7:38. Sokolowski MB (1980) Foraging strategies of Drosophila melanogaster: a chromosomal analysis. Behav Genet 10:291-302.  185  Staunton J, Ganetzky B, Nonet ML (2001) Rabphilin potentiates soluble N-ethylmaleimide sensitive factor attachment protein receptor function independently of rab3. The Journal of neuroscience : the official journal of the Society for Neuroscience 21:9255-9264. Stripling R, Volman SF, Clayton DF (1997) Response modulation in the zebra finch neostriatum: Relationship to nuclear gene regulation. Journal of Neuroscience 17:3883-3893. Stripling R, Milewski L, Kruse AA, Clayton DF (2003) Rapidly learned song-discrimination without behavioral reinforcement in adult male zebra finches (Taeniopygia guttata). Neurobiology of Learning and Memory 79:41-50. Sugiura M, Fuke S, Suo S, Sasagawa N, Van Tol HH, Ishiura S (2005) Characterization of a novel D2-like dopamine receptor with a truncated splice variant and a D1-like dopamine receptor unique to invertebrates from Caenorhabditis elegans. Journal of neurochemistry 94:1146-1157. Sulzer D (2007) Multiple hit hypotheses for dopamine neuron loss in Parkinson's disease. Trends Neurosci 30:244-250. Suo S, Sasagawa N, Ishiura S (2003) Cloning and characterization of a Caenorhabditis elegans D2-like dopamine receptor. Journal of neurochemistry 86:869-878. Suzuki H, Kerr R, Bianchi L, Frokjaer-Jensen C, Slone D, Xue J, Gerstbrein B, Driscoll M, Schafer WR (2003) In vivo imaging of C. elegans mechanosensory neurons demonstrates a specific role for the MEC-4 channel in the process of gentle touch sensation. Neuron 39:1005-1017. Suzuki Y, Yandell MD, Roy PJ, Krishna S, Savage-Dunn C, Ross RM, Padgett RW, Wood WB (1999) A BMP homolog acts as a dose-dependent regulator of body size and male tail patterning in Caenorhabditis elegans. Development 126:241-250. Swierczek NA, Giles AC, Rankin CH, Kerr RA (2011) High-throughput behavioral analysis in C. elegans. Nature methods 8:592-598. Thompson RF (2009) Habituation: a history. Neurobiol Learn Mem 92:127-134. Thompson RF, Spencer WA (1966) Habituation: a model phenomenon for the study of neuronal substrates of behavior. Psychol Rev 73:16-43. Toselli M, Taglietti V (2005) L-type calcium channel gating is modulated by bradykinin with a PKCdependent mechanism in NG108-15 cells. Eur Biophys J 34:217-229. Trimarchi JR, Schneiderman AM (1995a) Flight initiations in Drosophila melanogaster are mediated by several distinct motor patterns. Journal of comparative physiology A, Sensory, neural, and behavioral physiology 176:355-364.  186  Trimarchi JR, Schneiderman AM (1995b) Different neural pathways coordinate Drosophila flight initiations evoked by visual and olfactory stimuli. J Exp Biol 198:1099-1104. Tsalik EL, Niacaris T, Wenick AS, Pau K, Avery L, Hobert O (2003) LIM homeobox gene-dependent expression of biogenic amine receptors in restricted regions of the C-elegans nervous system. Developmental Biology 263:81-102. Tsechpenakis G, Bianchi L, Metaxas D, Driscoll M (2008) A novel computational approach for simultaneous tracking and feature extraction of C. elegans populations in fluid environments. IEEE Trans Biomed Eng 55:1539-1549. van der Maaten L, Hinton G (2008) Visualizing data using t-SNE. Journal of Machine Learning Research 9:2579-2605. Vanduyn N, Settivari R, Wong G, Nass R (2010) SKN-1/Nrf2 inhibits dopamine neuron degeneration in a Caenorhabditis elegans model of methylmercury toxicity. Toxicol Sci 118:613-624. Vidal-Gadea A, Topper S, Young L, Crisp A, Kressin L, Elbel E, Maples T, Brauner M, Erbguth K, Axelrod A, Gottschalk A, Siegel D, Pierce-Shimomura JT (2011) Caenorhabditis elegans selects distinct crawling and swimming gaits via dopamine and serotonin. P Natl Acad Sci USA 108:1750417509. Wang J, O'Donnell P (2001) D(1) dopamine receptors potentiate nmda-mediated excitability increase in layer V prefrontal cortical pyramidal neurons. Cereb Cortex 11:452-462. Wang J, Tokarz R, Savage-Dunn C (2002) The expression of TGFbeta signal transducers in the hypodermis regulates body size in C. elegans. Development 129:4989-4998. Warren WC et al. (2010) The genome of a songbird. Nature 464:757-762. Weber M, Schnitzler HU, Schmid S (2002) Synaptic plasticity in the acoustic startle pathway: the neuronal basis for short-term habituation? Eur J Neurosci 16:1325-1332. White JG, Southgate E, Thomson JN, Brenner S (1986) The Structure of the Nervous-System of the Nematode Caenorhabditis-Elegans. Philos T Roy Soc B 314:1-340. Wicks SR, Rankin CH (1995) Integration of mechanosensory stimuli in Caenorhabditis elegans. J Neurosci 15:2434-2444. Wicks SR, Rankin CH (1996) The integration of antagonistic reflexes revealed by laser ablation of identified neurons determines habituation kinetics of the Caenorhabditis elegans tap withdrawal response. J Comp Physiol A 179:675-685.  187  Wicks SR, Rankin CH (1997) Effects of tap withdrawal response habituation on other withdrawal behaviors: the localization of habituation in the nematode Caenorhabditis elegans. Behav Neurosci 111:342-353. Wilson DA (1998) Habituation of odor responses in the rat anterior piriform cortex. J Neurophysiol 79:1425-1440. Wilson DA (2009) Olfaction as a model system for the neurobiology of mammalian short-term habituation. Neurobiol Learn Mem 92:199-205. Wise RA, Rompre PP (1989) Brain dopamine and reward. Annu Rev Psychol 40:191-225. Wittekind WC, Spatz HC (1988) Habituation of the landing response of Drosophila. In: Modulation of Synaptic Transmission and Plasticity in Nervous Systems (Hertting G, Spatz HC, eds), pp 351-368. Berlin Heidelberg: Springer-Verlag. Wolf FW, Eddison M, Lee S, Cho W, Heberlein U (2007) GSK-3/Shaggy regulates olfactory habituation in Drosophila. Proc Natl Acad Sci U S A 104:4653-4657. Wolman MA, Jain RA, Liss L, Granato M (2011) Chemical modulation of memory formation in larval zebrafish. P Natl Acad Sci USA 108:15468-15473. Xu X, Sassa T, Kunoh K, Hosono R (2002) A mutant exhibiting abnormal habituation behavior in Caenorhabditis elegans. Journal of neurogenetics 16:29-44. Yang C, Kazanietz MG (2003) Divergence and complexities in DAG signaling: looking beyond PKC. Trends Pharmacol Sci 24:602-608. Yang CR, Seamans JK (1996) Dopamine D1 receptor actions in layers V-VI rat prefrontal cortex neurons in vitro: modulation of dendritic-somatic signal integration. The Journal of neuroscience : the official journal of the Society for Neuroscience 16:1922-1935. Yang CR, Seamans JK, Gorelova N (1999) Developing a neuronal model for the pathophysiology of schizophrenia based on the nature of electrophysiological actions of dopamine in the prefrontal cortex. Neuropsychopharmacology 21:161-194. Yook KJ, Proulx SR, Jorgensen EM (2001) Rules of nonallelic noncomplementation at the synapse in Caenorhabditis elegans. Genetics 158:209-220. Young AM (1993) Intracerebral microdialysis in the study of physiology and behaviour. Rev Neurosci 4:373-395. Young AM, Moran PM, Joseph MH (2005) The role of dopamine in conditioning and latent inhibition: what, when, where and how? Neurosci Biobehav Rev 29:963-976. 188  Young CE, Yang CR (2004) Dopamine D1/D5 receptor modulates state-dependent switching of somadendritic Ca2+ potentials via differential protein kinase A and C activation in rat prefrontal cortical neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 24:8-23. Zhao B, Khare P, Feldman L, Dent JA (2003) Reversal frequency in Caenorhabditis elegans represents an integrated response to the state of the animal and its environment. The Journal of neuroscience : the official journal of the Society for Neuroscience 23:5319-5328. Zheng Y, Brockie PJ, Mellem JE, Madsen DM, Maricq AV (1999) Neuronal control of locomotion in C. elegans is modified by a dominant mutation in the GLR-1 ionotropic glutamate receptor. Neuron 24:347-361. Zordan MA, Massironi M, Ducato MG, Te Kronnie G, Costa R, Reggiani C, Chagneau C, Martin JR, Megighian A (2005) Drosophila CAKI/CMG protein, a homolog of human CASK, is essential for regulation of neurotransmitter vesicle release. J Neurophysiol 94:1074-1083.  189  Appendix The supplemental information in chapter 4 can be found in an electronic format (“.zip” file) on the Accompanying Material (DVD). Within the “.zip” file there is a folder for each supplemental item. There are five supplemental tables, two supplemental movies, one supplemental figure set and one supplemental figure. For details refer to section 4.4.  190  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0072619/manifest

Comment

Related Items