UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Modulation of habituation kinetics and behavioural shifts by members of the heterotrimeric G-protein… McEwan, Andrea 2013

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

Item Metadata

Download

Media
24-ubc_2013_spring_mcewan_andrea.pdf [ 1.8MB ]
Metadata
JSON: 24-1.0073828.json
JSON-LD: 24-1.0073828-ld.json
RDF/XML (Pretty): 24-1.0073828-rdf.xml
RDF/JSON: 24-1.0073828-rdf.json
Turtle: 24-1.0073828-turtle.txt
N-Triples: 24-1.0073828-rdf-ntriples.txt
Original Record: 24-1.0073828-source.json
Full Text
24-1.0073828-fulltext.txt
Citation
24-1.0073828.ris

Full Text

Modulation of Habituation Kinetics and behavioural shifts by members of the heterotrimeric G-protein signaling pathways by Andrea McEwan  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in The Faculty of Graduate Studies (Cell and Developmental Biology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April (2013)  © Andrea McEwan, 2013  Abstract Despite its apparent simplicity, the soil-dwelling nematode Caenorhabditis elegans has a surprisingly large capacity to learn and remember. Previous characterization of C. elegans genome and neuronal circuit makes this worm an ideal choice for studying behavior and the mechanisms that underlie it. Through careful behavioral and genetic studies, nematodes have been shown to form many different types of memory, including short-term non-associative memory called habituation. Habituation is the defined as the decrement in response after repeated, irrelevant stimuli. In the first part of this thesis, detailed analyses showed that as one response type, reversals, decreased other responses, accelerations, decelerations and pauses became more prevalent. An earlier large-scale screen of mutant strains of C.elegans showed that several genes related to heterotrimeric G-protein family of signaling pathways exhibited striking defects in habituation. To follow-up on those findings, in the second part of my thesis I investigated the role of heterotrimeric G-protein signaling pathway and showed that Gαi and Gαq signaling pathways share a broad role regulating habituation whereas the Gαs pathway modulates the rate of habituation. The analyses of all the behaviours nematodes preform in response to habituation training showed that heterotrimeric G-protein signaling pathways play a role in regulating the shift in behaviour during habituation training. Together these data add to our understanding of the mechanisms underlying habituation of the tap response in C. elegans.  ii  Preface A version of the abstract and introduction along is in press in a book entitled Mechanosensory Learning and Memory by Randolf Menzel and Paul Benjamin (McEwan, A., Rankin, C. "Mechanosensory Learning and Memory." In Invertebrate learning and memory, by R., Benjamin, P. Mezel, 89-108. Elsevier, 2013). Figure 1.2 is adapted from the same publication Figure 1.6.5, showing a proposed mechanism of reversal distance habituation is from Giles et al. (in preparation). Additionally, Figure 1.6.6 is adapted from Bastiani and Mendel (2006). Finally, habituation of worms on food and off of food (section 2.4.2) was collected by Blaine Menon and re-analyzed using my behavioural analysis program.  iii  Table of Contents Abstract ........................................................................................................................................... ii Preface............................................................................................................................................ iii Table of Contents ........................................................................................................................... iv List of Tables ................................................................................................................................ vii List of Figures .............................................................................................................................. viii Acknowledgements ........................................................................................................................ ix 1 Introduction .................................................................................................................................. 1 1.1 C.elegans ............................................................................................................................... 1 1.2 Circuitry Underlying Tap Habituation .................................................................................. 2 1.3 Neurotransmitters Involved in Tap Habituation.................................................................... 5 1.4 Locus of Plasticity in Tap Habituation .................................................................................. 5 1.5 High-throughput behavioural screens ................................................................................... 7 1.6 Heterotrimeric G-protein signaling ....................................................................................... 8 1.6.1 Heterotrimeric G-protein signaling in mammalian systems ........................................... 9 1.6.2 Heterotrimeric G-protein signaling in C.elegans ......................................................... 10 1.6.3 Gαi signaling in C. elegans .......................................................................................... 11 1.6.4 Gαq signaling pathway in C.elegans ............................................................................ 13 1.6.5 Gαs signaling pathway in C.elegans ............................................................................ 13 1.6.6 Heterotrimeric G-protein signaling in habituation ....................................................... 15 2 Experiment 1: Modulation of tap-induced behaviour ................................................................ 19 2.1 Introduction ......................................................................................................................... 19 2.2 Methods ............................................................................................................................... 21 2.2.1 Strain Maintenance ....................................................................................................... 21 2.2.2 Behavioural assay ......................................................................................................... 21 2.2.3 Behavioural data analysis ............................................................................................. 22 2.2.4 Statistics ........................................................................................................................ 23 2.3 Results ................................................................................................................................. 24 2.3.1 Characterization of non-reversal responses induced by tap ......................................... 24 2.3.2 Reconstruction of the habituation curve for reversal probability ................................. 25 2.3.3 Correlation between probability of reversals, accelerations, decelerations and pauses and no responses. ................................................................................................................... 27 2.3.4 Correlation between magnitude of reversals, forward accelerations and pauses ......... 28 2.4 Discussion ........................................................................................................................... 29  iv  2.4.1 The decrement in reversal probability is associated with the increment of other tapelicited behaviours ................................................................................................................. 29 2.4.2 Correlation analysis between tap-induced behaviours ................................................. 30 2.4.3 Nature of habituation for reversal probability and the C.elegans startle response ....... 34 2.4.4 The magnitude of the reversals and accelerations decrements over repeated taps ....... 35 2.4.5 Mechanisms of changes in tap-induced behaviour ....................................................... 35 2.5 Conclusion........................................................................................................................... 37 3 Experiment 2a: Role of G protein pathways in habituation ....................................................... 38 3.1 Introduction ......................................................................................................................... 38 3.2 Methods ............................................................................................................................... 40 3.2.1 Animals ......................................................................................................................... 40 3.2.2 Behavioural assay ......................................................................................................... 41 3.2.3 Behavioural data analysis ............................................................................................. 41 3.2.4 Statistics ........................................................................................................................ 43 3.3 Results ................................................................................................................................. 43 3.4 Habitation kinetics of Gαq signaling pathway mutants ...................................................... 44 3.4.1 Habituation kinetics of egl-30/GNAQ for reversal distance ........................................ 45 3.4.2 Habituation kinetics of egl-30/GNAQ for reversal probability .................................... 46 3.4.3 Habituation kinetics of genes downstream of egl-30. .................................................. 49 3.4.4 Reversal distance of genes downstream of egl-30/GNAQ ........................................... 50 3.4.5 Reversal probability of genes downstream of egl-30/GNAQ ...................................... 51 3.5 Habitation kinetics of Gαi signaling pathway mutants ....................................................... 53 3.5.1 Habituation kinetics of goa-1/GNAO for reversal distance ......................................... 53 3.5.2 Habituation kinetics of goa-1/GNAQ for reversal probability ..................................... 54 3.5.3 Habituation kinetics of genes downstream of goa-1/GNAQ for reversal probability .. 55 3.5.4 Reversal distance of genes downstream of goa-1/GNAO ............................................ 56 3.5.5 Reversal probability of genes downstream of goa-1/GNAO ........................................ 57 3.6 Habitation kinetics of Gαs signaling pathway mutants ....................................................... 58 3.6.1 Habituation kinetics of gsa-1/GNAS for reversal distance .......................................... 59 3.6.2 Habituation kinetics of gsa-1/GNAS for reversal probability ...................................... 60 3.6.3 Habituation kinetics of genes downstream of gsa-1/GNAS ......................................... 60 3.6.4 Reversal distance of genes downstream of gsa-1/GNAS ............................................. 61 3.6.5 Reversal probability of genes downstream of gsa-1/GNAS ........................................ 62 3.7 Discussion ........................................................................................................................... 64 3.7.1 DAG levels and Gαq signaling pathway regulate habituation ..................................... 64 v  3.7.2 Gαi interacts with Gαq to regulate habituation rate an asymptotic level ..................... 66 3.7.3 Gαs specifically regulates habituation rate ................................................................... 66 3.7.4 PKA regulates some aspects of habituation ................................................................. 67 3.8 Experiment 2b: Non-reversal responses elicited by tap in Heterotrimeric G-protein mutants ................................................................................................................................................... 68 3.8.1 Non-reversal tap responses in Gαq ............................................................................... 68 3.8.2 Non-reversal tap responses in Gαi ................................................................................ 71 3.8.3 Non-reversal tap responses in Gαs ............................................................................... 74 3.8.4 Strains predicted to interact in heterotrimeric G-protein signaling .............................. 75 2.8.5 Caveats of non-reversal behavioural measurements .................................................... 77 3.9 Discussion ........................................................................................................................... 78 3.9.1 Gαq and Gαi shift their behaviour from pauses to reversals ........................................ 78 3.9.2 Gαs mutants shift behaviour from reversals to accelerations ....................................... 79 3.9.3 Heterotrimeric G-protein signaling pathways .............................................................. 80 3.9.4 Genes have a similar phenotype to members of the heterotrimeric G-protein signaling pathway.................................................................................................................................. 80 4 Conclusion ................................................................................................................................. 82 4.1 Contributions to research on non-associative learning ....................................................... 85 4.2 Potential genes acting upstream or downstream of heterotrimeric G-protein signaling pathways .................................................................................................................................... 85 4.3 Future directions.................................................................................................................. 86 Bibliography ................................................................................................................................. 88 Appendices .................................................................................................................................... 92  vi  List of Tables Table 1.6.2 Conserved α subunits in C.elegans ......................................................................... 10 Table 2.3.3.1 Correlation coefficients between the rate of behavioural types ........................... 27 Table 2.3.3.2 Correlation coefficients between asymptote of behavioural types ...................... 28 Table 2.3.4 Correlation coefficients for magnitude of accelerations, reversals, decelerations and pauses ......................................................................................................................................... 29 Table 2.4.2.1 Correlation coefficient between accelerations and reversals for response rate and asymptotic level in wild-type animals ....................................................................................... 31 Table 3.4.1 Five alleles of egl-30 that were tested for habituation defects ............................... 44 Table 3.4.3 Members of the Gαq signaling pathway that were tested for habituation defects .. 50 Table 3.5 Alleles of goa-1 (from the Gαi signaling pathway) tested for habituation defects.... 53 Table 3.5.3 Members of the Gαi signaling pathway that were tested for habituation defects... 56 Table 3.6 Alleles of gsa-1 (from the Gαs signaling pathway) that were tested for habituation defects ........................................................................................................................................ 59 Table 3.6.3 Members of the Gαs signaling pathway that were tested for habituation defects .. 61 Table A.1 p-values of Pearson correlation coefficients ............................................................. 92 Table B.1 Stains tested............................................................................................................... 94 Table B.2 p-values of strains tested for defects in reversal probability and non-reversal behaviours .................................................................................................................................. 95 Table B.3 p-values of heterotrimeric G-protein-associated genes tested for defects in reversal probability and non-reversal behaviours .................................................................................... 96 Table B.4 p-values of all strains tested for reversal distance ..................................................... 97  vii  List of Figures Figure 1.2 Tap withdrawal circuit .............................................................................................. 4 Figure 1.6.5 Heterotrimeric G-protein signaling ....................................................................... 15 Figure 1.6.6 Hypothetical pathway for regulation of habituation distance (asymptote) .......... 18 Figure 2.2.1 Tap-induced behaviours in wild-type nematodes .................................................. 25 Figure 2.3.2 Reconstruction of the reversal curve ..................................................................... 26 Figure 2.3.4 Magnitude of accelerations.................................................................................... 29 Figure 2.4.2 Context-dependent shift in behaviour ................................................................... 34 Figure 3.4.1: Reversal distance habituation of egl-30 ............................................................... 45 Figure 3.4.2.1 Reversal probability habituation of egl-30 ......................................................... 46 Figure 3.4.2.2 Reversal probability habituation of egl-30(ep271) (96h vs 120h). .................... 47 Figure 3.4.2.3 Reversal probability habituation of egl-30(js126) (96h vs 120h). ..................... 48 Figure 3.4.2.4 Reversal probability habituation of egl-30(n686) (96h vs 120h). ...................... 48 Figure 3.4.4 Reversal distance habituation of Gαq associated genes ........................................ 51 Figure 3.4.5 Reversal probability habituation of Gαq associated genes .................................... 52 Figure 3.5.1 Reversal distance habituation of goa-1 ................................................................. 54 Figure 3.5.2 Reversal probability habituation of goa-1 ............................................................. 55 Figure 3.5.4 Reversal probability habituation of Gαi-associated genes .................................... 57 Figure 3.5.5 Reversal distance habituation of Gαi-associated genes ........................................ 58 Figure 3.6.1 Reversal distance habituation of gsa-1 .................................................................. 59 Figure 3.6.2 Reversal probability habituation Gαs associated genes ........................................ 60 Figure 3.6.4 Reversal distance habituation Gαs associated genes ............................................. 62 Figure 3.6.5 Reversal probability habituation Gαs signaling pathway ...................................... 64 Figure 3.8.1.1 Behavioural profile of Gαq mutants ................................................................... 70 Figure 3.8.1.2 Behavioural profile of Gαq mutants (120h) ....................................................... 73 Figure 3.8.2 Behavioural profile of Gαi mutants ....................................................................... 75 Figure 3.8.4 Behavioural profile of Gαi mutants ....................................................................... 77  viii  Acknowledgements I’d like to thank Dr. Catharine Rankin for all of the advice, help and guidance and for accepting me into her lab. I would also like to thank my committee members Shernaz Bamji and Michael Gordon for agreeing to be on my committee and their advice on the direction of this study. I’d like to thank Blaine Menon for letting me apply my analysis to his data in section 2.4.2 (Figure 2.4.2). I would also like to acknowledge and thank Andy Giffen and Ben Huang for technical support in Python. Finally, I’d like to thank Dr. Eric Aamodt, Dept. of Biochmistry and Molecular Biology at LSU and members of the Rankin lab for their discussions on habituation and making the lab a great place to work!  ix  1 Introduction Habituation is defined as the decrement of response to repeated, irrelevant stimuli. Habituation is apparent in all organisms studied. Because of this, it is thought to be the foundation of selective attention and important for survival. Although we have a great deal of understanding of the behavioural features of habituation we know surprisingly little about the cellular mechanisms underlying this simple form of non-associative learning. The purpose of this thesis is to use behavioural and genetic analyses to investigate the role of heterotrimeric gproteins in tap habituation in the genetic model system C. elegans.  1.1 C.elegans Caenorhabditis elegans is a 1 mm long soil-dwelling nematode selected by Sydney Brenner in 1965 in order to study animal behaviour and development (in particular, neural development (Wood 1988). C.elegans have become a popular model organism because of their rapid lifecycle, large number of progeny and ease of cultivation in a laboratory setting. Additionally, because C.elegans are self-fertilizing hermaphrodites, each progeny represents a genetic clone. The first C. elegans behaviour in which learning was studied in detail was a mechanosensory response: the “tap withdrawal response” (Rankin 1990). The tap withdrawal response is the worm equivalent of a startle response. This response is elicited by a tap to the side of the Petri-dish on which worms are cultured. The tap to the Petri dish sends vibrations through the agar that are sensed by C.elegans through mechanosensory neurons. In response to the tap the adult worms typically react by initiating a reversal (crawling backwards), sometimes switching direction before moving forward again. 1  In 1990, Rankin et al. found that with repeated taps to the Petri dish C. elegans would reverse the direction of movement with progressively shorter distances. A total of 30 taps were delivered with a fixed interstimulus interval (ISI) of 10 seconds between taps. To rule out the possibility that the worms were responding less because of motor or sensory fatigue, they applied a mild electric shock to the agar and found that worms would reverse with a much larger distance (a process called dishabituation). This demonstrated that the decrement in response to repeated taps was not due to motor or sensory fatigue but rather fit the description of habituation. This was the first evidence that C.elegans were capable of learning. While this classical protocol for testing short-term habituation in C.elegans has evolved over the two decades it has been studied, the basic framework has remained constant. Current studies using the machine scoring of the data separate these into two discrete measurements: “frequency/probability” habituation and “distance/magnitude” habituation. In the first measurement, the proportion of worms that respond to the tap is determined (regardless of the distance they respond) while in the second measurement the distance analysis the distance is calculated for only those worms that respond to the tap (Swierczek et al, 2011).  1.2 Circuitry Underlying Tap Habituation Serial section EM experiments over two decades resulted in a complete map of C.elegans neuronal circuitry including information on the location of synapses (Sulston et al., 1983). The head and tail touch circuits were defined through laser ablation studies (Chalfie et al., 1985); the sensory neurons and interneurons that were required to produce a behavioural response after a either a head touch or a tail touch were identified. The body touch circuit includes sensory neurons ALML/R, PLML/R, and AVM, the interneurons, AVAL/R, AVBL/R, AVDL/R, and possibly PVC and/or RIM. AVA, AVB and AVD are each bilateral interneurons responsible for 2  transmitting sensory information to motorneurons. Ablation of the precursor cell that differentiates into the command interneurons AVA, AVD, AVE and RIM resulted in worms that were unable to initiate backwards movement (Chalfie et al., 1985). In their laser ablation analysis of the tap withdrawal circuit Wicks and Rankin (Wicks, 1996) began with the touch circuit and systematically ablated sensory neurons (3 types) and interneurons (5 types) either singly or in combination to test for alteration in the behavioural responses to tap (Wicks., 1995). The results indicated that the tap withdrawal circuit was mediated by 5 sensory neurons (ALML, ALMR, PLML, PLMR, and AVM) 4 pairs of command interneurons (AVA, AVB, AVD, PVC,) along with PVDL, PVDR and DVA (involved in harshtouch and proprioception, respectively). A general feature found in the study was that ablations in the anterior touch cells resulted in a decrease in reversal frequency (and an increase in the number of forward accelerations) while ablations in the posterior touch cells increased the frequency of reversals. This suggested that the tap withdrawal response was composed of two antagonistic sub-circuits: one that drove accelerations and one that drove reversals. By ablating members of one circuit tap-induced behaviour was biased towards the output of the remaining circuit. Although the tap withdrawal response was originally defined as containing 8 sensory neurons (ALML, ALMR, AVM, PLML, PLMR, PVDL, PVDR, and DVA), 4 pairs of command interneurons (AVA, AVB, AVD, PVC) there may be other members of the circuit. Recent evidence from Pigott et al. suggests that another sub-circuit may also be involved in the tap withdrawal response (Piggott et al., 2011). Through laser ablation studies and imaging techniques, this group identified a second sub-circuit, which included the pair of RIM neurons and an interneuron named AIB that governed reversal behaviour. This introduces the possibility  3  that the tap withdrawal response is governed by three sub-circuits: a circuit driving forward behaviour and two redundant circuits driving reversal behaviour (Figure 1.2). Through modeling and electrophysiology data from a larger related species (Ascaris), Wicks et al. were able to make hypotheses about the polarity of neurons in the circuit. Generally speaking, the behaviour produced by various cell ablations suggested that the gap junctions within the circuit were excitatory and the synaptic connections were inhibitory (Chalfie et al., 1985; Wicks, 1996). This meant the role of gap junctions between command interneurons and motor neurons is to drive forward or backwards movement and the role of (presumably inhibitory) chemical synapses between sensory neurons and interneurons is to modulate the response. Wicks et al.’s model largely supported this view as sensory neurons ALM, AVM and PLM were hypothesized to be inhibitory while the synapses between AVD and AVA are likely to be excitatory (Wicks., 1996).  Figure 1.2 Schematic of the tap-withdrawal circuit. Mechanosensory neurons are represented as octogons, command interneurons are shown as white circles and pools of motorneurons are represented as triangles. Arrows represent chemical synapses and dashed lines represent gap-junctions.  4  1.3 Neurotransmitters Involved in Tap Habituation C.elegans express many different neurotransmitter types including dopamine, serotonin, octopamine, GABA, and glutamate (Horvitz et al., 1982; Loer, 1993; Hart et al., 1995). Two major lines of evidence suggested that synaptic transmission from the touch cells to the command interneurons is glutamatergic. First, three classes of glutamate receptor subunits are expressed singly or in combination in the command interneurons AVA, AVB, AVD and PVC. These are AMPA type glutamate receptor subunits glr-1 (Hart et al., 1995; Maricq et al., 1995) and glr-2, expressed in all aforementioned interneurons; NMDA-type glutamate receptor subunits, nmr-1 and nmr-2, expressed in interneurons AVA, AVD, AVE, RIM, and PVC (Brockie et al., 2001) as well as glutamate gated chloride channel subunits such as avr-14 that is expressed on the tap mechanosensory neurons and may serve as an inhibitory autoreceptor. The second line of evidence came from the analysis of a gene called eat-4, which is a C. elegans homologue of mammalian VGlut1, a vesicular glutamate transporter. Examination of the expression pattern of eat-4 showed that it is expressed in a number of neurons including the mechanosensory neurons (Lee et al, 1999). Thus, it is very probable that the synaptic connections between sensory cells and command interneurons are glutamatergic.  1.4 Locus of Plasticity in Tap Habituation In general, habituation is likely encoded through changes in synaptic strength and/or changes in cell excitability; however, the locus in which these changes may occur is unknown. One way of narrowing down potential sites is to examine circuits that govern other behaviours and determine whether habituation training also affects aspects of the other behaviour. If this is the case, then the locus of plasticity for habituation can be said to be overlapping with the circuit  5  governing the second behaviour. Conversely, if habituation training does not alter the second behaviour, the site of plasticity is likely outside of the circuit governing the second behaviour. One type of behaviour that was used to assess the site of plasticity is the rate of spontaneous reversals. In order to do determine whether the neurons that govern spontaneous reversal overlap with the site of plasticity for tap habituation, Wicks et al. examined spontaneous reversals in animals that had undergone habituation training (Wicks, 1997). They found that habituated worms did not show a decrease in either the frequency or magnitude of spontaneous reversals, suggesting that the site of plasticity in habituation and the neurons governing spontaneous reversal are non-overlapping (Wicks, 1997). Laser ablation of AVA resulted in a significant decrease in the magnitude of reversal in response to tap and the frequency of spontaneous reversals (Bargmann et al., 1995). Conversely, ablation of the sensory neurons ALM, PLM and AVM and the interneuron AVD did not affect the rate of spontaneous reversals (Wicks, 1995). In 2011, Zhen et al. (2011) investigated the role of the command interneurons and motorneurons in directional movement). Zhen et al. (2011) measured the calcium transients in the command interneurons in freely moving worms (Kawano et al., 2011). Backwards movement corresponded with an increase in calcium transients in AVA and AVE; however, there was no change in calcium transients in AVD interneuron suggesting it is only involved in stimulusevoked reversals rather than spontaneous reversals (Kawano et al., 2011). It is interesting to note that during the original touch circuit studies Chalfie found AVD played an important role however he found no role for AVE in stimulus evoked responses (Chalfie et al. 1985).  6  Taken together, both studies suggest that the site of habituation is in the mechanosensory neurons and/or AVD, most likely encoded in alterations in the synaptic connection between the sensory neurons and the command interneurons.  1.5 High-throughput behavioural screens Behavioural studies have been greatly facilitated by the development of worm trackers (Geng et al., 2004; Huang et al., 2007; Ramot et al., 2008; Simonetta et al., 2007). The MWT comprises a camera, a platform on which C.elegans plates are mounted and a push-solenoid positioned next to the Petri-dish. In order to deliver a tap-stimulus automatically to the plate, a current is applied to the solenoid in pre-determined intervals. A skeletonized cartoon of each animal is generated from which multiple morphological and behavioural properties are collected. Once collected, the behavioural properties of the worm can be extracted and compiled into meaningful data. In a typical MWT experiment a plate of age-synchronized worms is mounted on the MWT platform. Naive behaviour is recorded for 10 minutes, after which, the solenoid automatically delivers the tap stimulus for the duration of the experiment. The MWT quantifies a plethora of behaviours including basal speed, the rate and size of spontaneous reversals, and habituation of reversal probability, distance, duration and speed after a tap (Swierczek et al., 2011). Although habituation was originally measured as a combination of reversal probability and distance, with the MWT these two variables are measured independently. For each of these variables, two measurements describing the shape of the curve can be calculated. These are the rate of habituation, defined as the half-life of the curve, and the habituated level (or asymptotic level) of the habituation curve. Using the MWT, massive datasets of behavioural data can be 7  collected relatively quickly. A recent behaviour screen (Giles et al., 2011) resulted in the characterization of over 500 C.elegans strains with mutations in nervous system genes. The analysis of these 500+ worm strains resulted in the identification hundreds of mutations that affected one or more aspect of habituation. Analysis of various measures of habituation across these strains led Giles et al to hypothesize that tap habituation was composed of at least four genetically independent features, these are: the rate of response decrement for the reversal probability measure, the final level of habituation for the reversal probability measure, the rate of response decrement for the reversal distance measure and the final level of habituation for the reversal distance measure. Interestingly, two genes exhibited particularly strong phenotypes: goa-1 and eat-16. goa-1 and eat-16 are members of the Gαi signaling pathway, which is in turn a member of the heterotrimeric G-protein signaling pathways. Giles et al. (in preparation) proposed that goa-1 and eat-16 have a fundamental role in regulating the final level for habituation of reversal distance through the MAP kinase signaling cascade.  1.6 Heterotrimeric G-protein signaling Heterotrimeric G proteins are members of a set of signaling pathways which transduce signals from the cellular membrane to intracellular substrates (Reviewed in Oldham, 2008). Heterotrimeric G proteins are composed of three subunits: α, β and γ. In its inactive conformation, these subunits form a complex together along with an ADP molecule making a bond with the alpha subunit. The αβγ-ADP complex associates with a heterotrimeric G protein coupled receptor (GPCR). When a ligand binds to the receptor, it causes a conformational change in the α subunit and exchanges its ADP for and ATP. With the exchange of ADP for ATP, the α subunit dissociates from the β and γ subunits and initiates downstream signaling (Oldham, 2008). Each α subunit encodes endogenous GTPase activity. After a downstream 8  response has been initiated, the α subunit will cleave its ATP to an ADP and re-associate with the βγ complex. The rate of activation or deactivation of heterotrimeric G-protein signaling pathways can be modulated through the action of guanine exchange factors (GEFs), GTPase activating proteins (GAPs) and regulators of G-proteins (RGS). Guanine exchange factors catalyze the rapid exchange of ADP to ATP on the α subunit so that it may be activate more quickly and often. Conversely, GAPs activate the α subunit’s endogenous GTPase activity, causing the protein to become inactive and re-associate with the βγ complex. RGSs activate GAPs (Oldham 2008).  1.6.1 Heterotrimeric G-protein signaling in mammalian systems In mammalian systems, the α subunits are separated into four families of signaling pathways: Gαi, Gαq, Gαs, and Gα12/13. Each pathway signals downstream effectors and has wide-reaching molecular functions. In mammals, Gαi’s role is to inhibit the adenylyl cyclase, an enzyme important for translating extracellular signals into intracellular responses. Adenylyl cyclase catalyzes a reaction to convert ATP to the second messanger, cyclic AMP (cAMP). cAMP then activaties cAMP-binding proteins, including kinases, transcription factors and ionchannels (Oldham 2008). Physiologically, the Gαi signaling pathway is associated with the depression of neural activity. Conversely, the Gαs signaling cascade results in the activation of adenylyl cyclase and is associated with the increase in neural activity. The Gαq signaling cascade governs the conversion of phospholipid phosphatidylinositol 4,5-bisphosphate (PIP2) into the second messangers diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP3). IP3 then binds to IP3 receptors on the sarcoplasmic reticulum and endoplasmic reticulum causing them to release calcium into the cytosol. Additionally, DAG and calcium work together to activate protein kinase C (PKC), a protein associated with neuronal excitation potentially involved in 9  memory formation. Finally, the Gα12/13 signaling cascade is thought to activate the Rho family of GTPases. Rho GTPases are associated with the regulation of cell shape and motility through changes in the cytoskeleton.  1.6.2 Heterotrimeric G-protein signaling in C.elegans The C.elegans genome encodes 21 α, 2β and 2γ subunits (Jansen et al., 1999; Cuppen et al., 2003). The majority of α subunits in C.elegans cannot be classified as members of any of the four major subfamilies and may be involved in C.elegans-specific signaling. Many of the C.elegans specific α subunits are Gαi-like, however, do not share sufficient sequence similarity to be classified as part of the Gαi signaling pathway. These α subunits are found in chemosensory neurons and contribute to the worm’s ability to detect and respond to a wide range of chemosensory and gustatory cues. Up to seven Gαi-like divergent α subunits can be found in a single chemosensory neuron which, given the relatively simplicity of the C.elegans nervous system, may be an answer to detecting many different chemical cues with very few neurons (Jansen et al., 1999). There are however, four C.elegans alpha subunits that correspond each of the four major subfamilies of heterotrimeric G-protein signaling pathways based on sequence similarity. These are shown in Table 1.6.2. Table 1.6.2 Conserved α subunits found in C.elegans  Name goa-1 egl-30 gsa-1 gpa-16  Description α subunit, Gαi α subunit, Gαq α subunit, Gαs α subunit, Gα12  Expression Pan neuronal Pan neuronal Pan neuronal Head, ventral cord, tail  10  Since expression patterns show that gpa-16 has limited expression in neurons, this thesis will focus on Gαi, Gαq and Gαs, whose corresponding alpha subunits are widely expressed in neurons. Gαi, Gαq and Gαs signaling cascades play important and diverse roles in C.elegans development and cell function. These functions include involvement in locomotion, egg-laying, response to volatile anesthetics, spindle orientation, neuronal migration and embryo viability (among other things). Through studies on egg-laying rate and locomotion the effect of each pathway on neuronal function has been partially elucidated (Miller, 1999 et al.; Schade et al., 2005).  1.6.3 Gαi signaling in C. elegans Gαi/goa-1 encodes a protein that has over 80% identity to the mammalian Gαi (Lochrie et al., 1991). In the classical signaling scheme found in mammal systems, Gαi detects signals from acetylcholine, dopamine, serotonin and several hormones to activate downstream responses. Through studies in locomotion, defecation and egg-laying studies, goa-1 has been shown to act downstream of serotonin, dopamine, octopamine, acetylcholine and FRMFamide signaling in C.elegans. Next, through studies with the cholinesterase inhibitor aldicarb, goa-1 was shown to be involved in inhibiting neurotransmitter release. The level of acetylcholine release can be assessed with a worm’s sensitivity to the aldicarb. Aldicarb is a cholinesterase inhibitor that prevents the degregation of acetylcholine in the synaptic cleft. Increased sensitivity to aldicarb suggests that excess acetylcholine is released into the synaptic cleft whereas decreased sensitivity suggests decreased acetylcholine neurotransmission. Loss-of-function mutations of goa-1 result in hyperactive locomotion and hypersensitivity to aldicarb. This suggested that excess acetylcholine is being released in these 11  mutants and led to the hypothesis that goa-1’s function is to inhibit synaptic release. (Mendel et al., 1995; Ségalat et al., 1995; Miller et al., 1999; Nurrish et al., 1999). In order to identify genes acting in the Gαi/goa-1 signaling pathway, a series of forward genetic screens and RNAi screens were performed starting over a decade ago. Two studies in particular contributed to improving our understanding of the Gαi signaling pathway. In the first study Miller et al. (1999) identified goa-1 and dgk-1 as negative regulators of egl-30 through aldicarb sensitivity screens. In these experiments, Miller et al.found that ric-8, a suppressor of goa-1 showed reduced sensitivity to aldicarb suggesting that mutants have reduced neurotransmitter release. The study identified mutants that suppressed the ric-8 phenotype and thereby found genes that enhanced neurotransmitter release. dgk-1 was identified as a suppressor of ric-8 and double mutants of dgk-1;egl-30 show near normal levels of aldicarb sensitivity. Further epistatic analysis with goa-1demonstrated Gαi signaling pathway acted upstream to antagonize Gαq (Miller et al. 1999). In the second study Hajdu-Cronin et al. (1999) developed an inducible transgene of constituatively active goa-1 and used mutagenesis to find suppressors of activated goa-1. Worms with constituatilvey active forms of goa-1 were lethargic, egg-laying defective and showed reduced pharyngeal pumping. Hajdu-Cronin et al. found that two genes, dgk-1 and eat16 could suppress the lethargic phenotype of activated goa-1. The expression pattern of eat-16 was found largely overlap that of goa-1. Additionally, a PLCβ-IP3 assay was performed to determine whether eat-16 could interfere with the production of IP3. Recall that the Gαq pathway results in the hydrolysis of PIP3 to DAG and IP3. By cotransfecting eat-16 with the M1 receptor (a receptor specifically coupled to Gαq in mammalian cells) in COS-7 cells, this study  12  demostrated that, in the presence of eat-16, less IP3 is produce. This showed that goa-1, dgk-1 and eat-16 interact genetically and that eat-16 supresses the Gαq signaling pathway in vitro. These studies, along with others, resulted in the development of a model whereby goa-1 along with eat-16 and dgk-1 act together to reduce neurotransmitter release (indicated by aldicarb sensitivity) through their interaction with the Gαq signaling pathway (Miller et al.; Hajdu-Cronin et al.1999; Lackner et al., 1999; Reynolds et al., 2005; Nurrish et al.1999; Chase et al. 2001; Schade et al. 2005). Given that goa-1 negatively regulates egl-30, the next question was to determine what egl-30/Gαq does.  1.6.4 Gαq signaling pathway in C.elegans Since goa-1 negatively inhibits both acetylcholine transmission and egl-30 function, it is a reasonable hypothesis that the Gαq/egl-30 pathway is involved in neurotransmitter release. With respect to acetylcholine release, Gαq/egl-30 activates egl-8/PLCβ, an enzyme which catalyzes the production of two second messengers: IP3 and DAG (Mendel et al., 1995; Ségalat et al., 1995; Miller et al., 1999; Nurrish et al., 1999). Through aldicarb-sensitivity assays, unc-13, a gene that facilitates vesicle fusion (or priming) at the synapse along with egl-30 and egl-8 were shown to lack sensitivity to aldicarb (Richmond et al., 1999; Richmond et al., 2001). Increasing DAG levels through genetic or pharmaceutical techniques rendered egl-30 and egl-8 (but not unc-13) sensitive to aldicarb. This suggests that acetylcholine release is governed by DAG levels and is dependent on unc-13. Taken together, the Gαq/egl-30 pathway is proposed to be a critical component for synaptic priming and neurotransmitter release (Reynolds et al., 2005).  1.6.5 Gαs signaling pathway in C.elegans In vertebrates, the Gαs signaling pathway governs the formation of cAMP by activating adenylyl cyclase. cAMP is a second messenger that has been shown to activate a number of 13  proteins including protein kinase A (PKA) (Sunahara et al., 1996; Walsh et al., 1994). The Gαs pathway has also been shown to activate calcium channel in skeletal muscle cells and inhibit sodium channels in cardiac tissue (reviewed in Wickman, 1995a; Wickman, 1995b) In C.elegans, what is known of the Gαs signaling pathway largely parallels the vertebrate results. gsa-1, the alpha subunit of the Gαs signaling pathway activates acy-1, which is orthologous to adenylyl cyclase 9. gsa-1 is required for worm viability and proper locomotion. Reduction of function mutants have sluggish movement and delayed egg laying (and like egl30/Gαq, loss-of-function mutants are completely paralyzed) whereas gain-of-function mutants have the opposite effect (Korswagen et al., 1997; Schade et al., 2005). Like egl-30/Gαq, gain-offunction mutations in gsa-1 and acy-1 result in hypersensitivity to aldicarb indicating that the Gαs signaling pathway has a role in acetylcholine release. Interestingly, reduction-of-function mutations in gsa-1 maintained the same level of aldicarb sensitivity compared to wild-type controls despite that fact that loss of gsa-1 resulted in paralysis. This showed that the Gαs signaling pathway is not required to maintain steady-state levels of neurotransmitter release but is critical for locomotion. To understand these seemingly paradoxical results, Reynolds et al. (2005) carried out a series of epistatic experiments involving members of the Gαq and Gαs signaling pathway. They found that Gαs function is largely dependent on proper signaling from the Gαq pathway. Indeed, hyperactivation of the Gαs pathway in an egl-30/Gαq mutant background only resulted in a small (though significant) suppression of the paralysis phenotype. This suggests that the action of the Gαs pathway is largely dependent on Gαq function. Using the reverse experiment whereby Gαq was hyperactivated in an acy1/Gαs mutant background, Reynolds et al. demonstrated that the locomotion rate in Gαs mutants could be improved to one  14  third of control levels when the Gαq signaling was hyperactivated. This led to their theory that main role of the Gαs signaling pathway is to act as a modulator of the Gαq signaling pathway. Through important studies completed between 1999 and 2005, a model of how the Gαi, Gαq and Gαs signaling pathways interact was proposed. In general, Gαi and Gαs modulate the activity of Gαq (Figure 1.6.5).  Figure 1.6.5 Schematic of the Gαi, Gαq and Gαs signaling pathway and its regulation of acetylcholine release. Gαi (unc-43, goa-1, and dgk-1) acts upstream of Gαq (egl-30, egl-8) to inhibit neurotransmitter release. Gαs (gsa-1, acy1 and pde-4_ feeds into Gαq downstream of second messenger DAG to facilitate synaptic priming and neurotransmitter release. Figure adapted from (Bastiani and Mendel, 2006)  1.6.6 Heterotrimeric G-protein signaling in habituation The Gαq signaling pathway was first identified as regulator of habituation in the Kindt et al.2007 study which found that C.elegans stay responsive to tap for longer if they are in the  presence of food; when habituation training occurs off food the probability of a reversal response  15  to tap drops off rapidly compared to when they are in the presence of food. Kindt et al. used a candidate gene approach to find which genes mediated the food-effect of habituation and found that a dopamine receptor (a heterotrimeric G-protein coupled receptor or GPCR) named dop-1 is required for the modulation of habituation on food (Kindt et al., 2007). In classical heterotrimeric G-protein signaling, once a ligand has bound to a GPCR, the α subunit from the αβγ-GPCR complex will dissociate and begin downstream signaling. To determine which α-subunit is associated with dop-1, the habituation kinetics of goa-1/Gαi, egl30/Gαq, and gsa-1/Gαs were assessed. egl-30/Gαq and other members from the Gαq signaling pathway most closely resembled the habituation phenotype found in dop-1 mutants suggesting that activation of dop-1 resulted in downstream signaling from the Gαq cascade (Kindt et al. 2007). Habituation is proposed to be the behavioural consequence of changes in synaptic strength or cell excitability of neurons within the tap-withdrawal circuit (Bailey et al., 1988); Groves and Thompson, 1970; Kindt et al. 2007). Cell excitability was assessed using in vivo calcium imaging, a technique whereby calcium levels are visualized using fluorescence of a genetically encoded calcium sensor called cameleon. By monitoring calcium transients of ALM and PLM in dop-1 and wild-type animals undergoing habituation training (on food), they showed that calcium transients detected in ALM (but not PLM) decremented at a faster rate in dop-1 mutants compared to wild-type. Since ALM cell excitability is reduced in dop-1 mutants, this suggests that dopamine transmission acts on ALM to increase cell excitability. Additionally, calcium transients of ALM were imaged in egl-30 and egl-8 mutants (genes in the dop-1 pathway). Again, these mutants had a more rapid decrement in calcium levels, which is consistent with dop-1 and supports the behavioural data (Kindt et al., 2007).  16  Recently, a large-scale behavioural screen for genes involved in habituation was complete by Giles et al. (in preparation) using the MWT in conjunction with an automated tapping system. In this study, over 500 strains of worms were analyzed for defects in habituation of the tap-withdrawal response along with anomalies in a handful of spontaneous and physiological attributes such as locomotion speed and body size. Cluster analysis was applied in order to find strains of worms with exceptionally similar phenotypes in multiple categories. From the resultant clusters, Giles et al. (in preparation) proposed genetic pathways that may regulate aspects of habituation kinetics. Giles et al. (in preparation) found that two genes from the Gαi signaling pathway, goa-1 and eat-16 exhibited slow habituation for both magnitude and probability of response. Based on the clustering analysis, bioinformatics and literature searches, Giles et al. hypothesized that goa-1 and eat-16 act as upstream regulators to govern the asymptotic level of habituation for response distance (Figure 1.5.6.). While further research is needed to test the accuracy of the hypothesis, this is the first large-scale habituation screen performed in C.elegans and represents a massive increase in knowledge about habituation in nematodes.  17  Figure 1.6.6 Proposed mechanism regulating the asymptotic level of reversal distance during habituation training according to cluster analysis and bioinformatics(Giles et al., in preparation)  With new technology, not only are large-scale habituation screens possible but more detailed behavioural data can be collected. In 1996, Wicks and Rankin found that worms performed two behaviours in response to tap: reversals and accelerations. They examined the frequency of accelerations or reversals in response to tap and found that as habituation training progressed, the proportion of accelerations increased until reaching an asymptotic level at about 30 stimuli. This study was scored by hand and consequently only the most extreme behaviours were measured, and only a small number of worms were tested. In the research reported here I used the MWT to further examine the alternative behaviours in response to tap in order to get a more complete characterization of the range of responses to tap over the course of short-term habituation in C.elegans 18  2 Experiment 1: Modulation of tap-induced behaviour 2.1 Introduction Typical habituation studies measure the decrement of behaviour over the course of training; however, because of the integrative nature of neuronal cells, it is possible that, while the behaviour of interest is decrementing, another type of behaviour is increasing in frequency or magnitude. As described, the tap-withdrawal circuit in C.elegans consists of at least two sub-circuits: one that drives forward locomotion and another that regulates reversal locomotion. The circuit analysis performed by Wicks et al. (1996a) indicated that the tap stimulus excites mechanosensory receptors in both the anterior and posterior of the worm which trigger movement in opposite directions, therefore, the resultant habituation curve is actually the summation of at least two competing sub-circuits: one driving forward behaviour (accelerations) one driving reversals. In order to gain insight into the habituation kinetics of the sub-circuits they identified, Wicks et al. (1996a) laser ablated members of the anterior mechanosensory neurons and the posterior mechanosensory neurons in C.elegans and characterized the magnitude of accelerations in ALM- and ALM- and AVM- worms over 40 tap stimuli (both at 10s and 60s ISI). The results showed that, for 10s and 60s ISI, the magnitude of accelerations decremented after repeated taps. In PLM- animals the magnitude of reversals decremented over repeated taps. These results suggest that over the course of habituation the relative strength of both the accelerations and the reversals is changing. However because the two sub-circuits change at different rates the relative input from each component changes over the course of habituation. Interestingly, acceleration of ALM- AVM- at 10s ISI exhibited an initial facilitation whereas worms trained at 60s ISI did not. Wicks et al. (1996a) postulated that the initial  19  facilitation of acceleration represented increased competition from the forward circuit and explained why worms trained at 10s ISI show a much faster initial rate of habituation compared to those trained at 60s ISI (the stronger input from the facilitation of the accelerations leads to smaller reversals). Next, Wicks et al (1996a) examined the behavioural patterns of intact worms. Animals were given 40 tap stimuli and the proportion of accelerations performed in response to tap was measured. Wicks et al. demonstrated that the probability of accelerations increased over the course of 30 taps, after which, the proportion of accelerations stabilized. This was hypothesized to be due to the more rapid decrement of the head circuit and slower decrement of the tail circuit. Over the course of habituation, more rapid decrement of reversals and slower decrement of accelerations allowed more accelerations to appear. Taken together, these studies demonstrate that the rate of habituation of the forward circuit and the reverse circuit differ. The amount of competition between each circuit varies over the course of habituation and alters the kinetics of habituation accordingly (Wicks,, 1996a). The Wicks et al. (1996a) study eloquently showed how the combination of laser ablation and behavioural studies could be used to investigate the relative strength of the circuits mediating behaviour. One caveat of these studies is, since the process of analyzing locomotion data by hand was labour intensive, thus only a limited number of parameters could be tested (in this case, accelerations and reversals). Additionally, when the type of behavioural output of the worm was limited to to accelerations, reversals or no-responses, it meant that the potential behaviours produced while there is low to moderate competition between sub-circuits were not accounted for. The purpose of the first set of experiments was to investigate all of the behaviours that occurred after taps in a habituation series in order learn more about the actual behaviour during habituation.  20  The multi-worm tracker (MWT) is a computer vision system that allows the analysis of data from up to 100 animals simultaneously (Swierczek et al., 2011). The MWT has previously been used to characterize dozens of behavioural characteristics in C. elegans, including speed, frequency and magnitude of spontaneous reversals. For this study I developed a Matlab program that measured a variety of behaviours (including accelerations, decelerations, pauses etc.) in order to get a more complete picture of how a worm responses change over the course of repeated taps.  2.2 Methods 2.2.1 Strain Maintenance N2 (wild-type) worms were used to assess responses to tap. Worms were cultured at 20ºC on NGM plates seeded with OP50, a non-virulent strain of E.coli (Brenner 1974).  2.2.2 Behavioural assay MWT plates were made by pouring 12 ml of NGM into 5cm petri-dishes. Plates were allowed to dry for 3 days. The day before colonies were made, 50 µl of OP50 was pipetted to the MWT plates, spread to form a uniform layer of bacteria and allowed to dry for up to 24h. Worms were synchronized by allowing 5 N2 worms to lay eggs on E. coli seeded plates for 4-5 hours. This resulted in approximately 60-80 worms per plate. When worms were 4 days old 4-6 plates (containing 60-80 worms) of each strain were tested on the multi-worm tracker. For each experiment on the MWT, worms were tracked for 10 minutes before the first stimulus was delivered to the plate (unless otherwise specified). Taps were delivered to the side of the petri dish via a push-type solenoid that fired every 10 seconds for 30 stimuli. In age-experiments, strains were tested at 96h and 120h. First, both control and mutant strains were tested at 96h. 60-80 worms were picked off their original plate and transferred to a 21  new plate that had been seeded up to 24h beforehand. These were then stored in a 20ºC incubator for 24h and retested.  2.2.3 Behavioural data analysis For each habituation experiment .blob files generated by the MWT were analyzed with choreography version 1.3.0. To collect locomotion data, the “Trigger” argument was used to find the average speed and bias of individual worms 1 second before each tap stimulus and 1 second afterwards. Data was outputted as .trig files with 4 columns: speed, standard deviation of speed, bias and instance of tap. A Python script was used to concatenate all .dat files for each plate. Worm bias (the direction the worm is moving) was multiplied by its speed. Since the average bias was calculated as an average, when bias was not 1 (representing forward), 0 (representing no movement) or -1 (representing reversal movement), the bias was changed from its value to -1. This was done to remove instances where worms moved in more than one direction during the time-frame their speed was measured. The Python script generated .err files that were used to calculate the standard-deviation between the speeds of the worms on the plate. They also generated .dat files which took the values of reversal probability and reversal distance calculated by the MeasureReversal plugin for choreography and .txt files which contained the speed data for all worms tracked during an experiment. The speed data was used to determine the proportion of accelerations, decelerations, pauses and “no responses”. Each behavioural type was calculated using the following formulas: num_swmthr = (nansum(acc<0.5 & acc./A>-0.5 & acc~=0 & B>=2*sem & A>=2*sem)) + num_accel = nansum(acc>2 & B>2*sem & A>=0); num_pause = nansum(B<=0.5 & B>=0.5*-sem & A>=2*sem); num_decel = nansum(acc2>2 & A>2*sem & B>2*sem); See below for a description of the variables: num_swmthr=number of worms that do not respond to tap 22  num_accel=number of worms that accelerate to tap num_pause=number of worms that pause to tap num_decel=number of worms that decelerate to tap A=average speed 1-0 seconds before a tap stimuli B=average speed 0-1 seconds after a tap stimuli acc=B-A acc2=A-B s.e.m.=standard error of the mean for a worm’s speed The proportion of behavioural type was calculated by finding the sum of the worms meeting one of the four conditions and dividing that by the total population of worms. One caveat of this analysis is that it relies on worm speed. The sum of all the behavioural types measured should equal approximately 100%, however, in worms with extreme alteration in locomotion speed, the sum of behaviour was often less than 90%. To ensure behaviours were not being under-reported due to locomotion effects, the sum of the behavioural types was added and only strains that resulted in a sum of 100% + 10% were considered. Reversal curves using the speed data were also generated to ensure that they approximated that calculated by the MeasureReversal plugin.  2.2.4 Statistics For each strain, 4-6 plates with 60-80 worms were analyzed. The mean for each plate was calculated and the s.e.m. was found using n=number of plates tested/strain. For each reversal curve, a line of best fit was found using the “power1” function in Matlab (based on the model y=axb). Using the best-fit line, the reversal rate, initial response and asymptotic level were found. The rate was calculated by finding the half-life of the line of best fit. The initial response and asymptotic level were found by finding the y-value at the first and last tap. Z-scores were calculated by comparing the mean and standard error of 12 control groups against all the strains tested. Z-scores were converted to p-values using the Normsdist function in excel.  23  The Pearson correlation coefficients were found between all behaviours for rate and asymptotic level. The Holm-Bonferroni method was use to correct for multiple comparisons in table 2.4.2.1. Specifically,p-values that were less than 0.05/n-k (where n is the number of comparison and k is the rank of the p-value in order of smallest to largest) were considered significant.  2.3 Results 2.3.1 Characterization of non-reversal responses induced by tap With the introduction of the MWT (Swierczek et al., 2011), tap habituation can now be easily scored in terms of the probability to reverse and the magnitude of reversals (Giles et al.in prep). In this study, additional measures were considered. To fully understand worm behaviour after tap, forward accelerations, tap-induced pauses, decelerations and no-responses were characterized for wild-type worms (N2). This analysis showed that as the worms habituated the probability of reversals to tap decreases and worms showed increases in probability of pauses, accelerations or decelerations in response to tap. The probability that a worm would have no response to a tap also increased but, overall, contributed only a small proportion to total worm behaviour. The rate in which pauses and decelerations and forward accelerations increased followed exponential curves (Figure 2.2.1).  24  N2 behaviour 1 0.9  Percent response  0.8 0.7  accelerations  0.6  pauses  0.5  no response  0.4  reversals  0.3  decelerations  0.2 0.1 0 -0.1  1  3  5  7  9  11 13 15 17 19 21 23 25 27 29  Figure 2.2.1 Tap-elicited behaviours exhibited by N2 for response probability. N2 shows a decrement in reversals coupled with an increment in pauses, decelerations, accelerations and no responses  These results show that during habituation training worms shift their behaviour from reversing to an alternate response of accelerating, decelerating or pausing.  2.3.2 Reconstruction of the habituation curve for reversal probability The tap withdrawal circuit is made up of a forward circuit that coordinates forward movement and a reverse circuit that regulates reversal behaviour. The tap is thought to activate both of these circuits at the same time. Wicks and Rankin (1996a) examined the habituation curves of worms with laser ablation of the sensory neurons of either the forward sub-circuit or the reversal sub-circuit. They found that the reversal habituation curve could be reconstructed by subtracting the acceleration curves of the reversal sensory neuron ablated animals from the reversal curves of forward sensory neuron ablated animals. These data supported the hypothesis  25  that the forward sub-circuit competes with the reversal sub-circuit to generate the tap response (Wicks, 1996a). To test whether this same approach could be used to study intact worms the contributions of the probability of accelerations, pauses, decelerations and “no responses” to the shape of the reversal probability habituation curve were calculated. To do this, each behavioural type was subtracted from 1 (where 1 is a flat line simulating a 100% reversal response to tap over 30 taps; Figure 4). Subtracting the proportion of accelerations was not sufficient to regenerate the observed reversal curve suggesting that accelerations alone are not enough to account for nonreversal responses to tap. The best approximation of the reversal curve to the actual reversal curve occurred when the proportion of accelerations pauses and decelerations were subtracted from 1 (Figure 2.3.2).  Reconstruction of the reversal curve 1 0.9  reversals  Percent response  0.8  1-accel  0.7  1-(accel +pause)  0.6 0.5  1-(accel+pause+decel)  0.4  1-all  0.3 0.2 0.1 0 1  3  5  7  9 11 13 15 17 19 21 23 25 27 29  Figure 2.3.2: Reconstruction of reversal curve from non-reversal behaviour. To reconstruct the reversal curve (in black), different non-reversal behaviour were subtracted from 1. The closest reconstruction to the actual reversal curve was found when the sum of accelerations, pauses, decelerations and no-responses were subtracted from 1.  26  These results show that reversal probability, accelerations, deceleration, pauses and noresponses account for approximately 100% of the behaviour exhibited by worms after a tap. Thus, subtracting all non-reversal behaviours from 1 resulted in a curve that closely approximates the experimental curve for reversal probability.  2.3.3 Correlation between probability of reversals, accelerations, decelerations and pauses and no responses. What is the relationship between these five major types of behaviour in response to tap: reversals, accelerations, decelerations and pauses? In order to gain insight into this question, the correlation coefficients between these behaviours were calculated. The correlation coefficients for 12 separate experiments of N2 worm (each consisting of 4-6 plates, each plate holding ~ 80 worms) were calculated between all permutations of behaviour for rate, initial response and asymptotic level. Significance was set based on the Holm-Bonferroni correction for multiple comparisons The rate of change of each type of behaviour was obtained by calculating a curve of best fit and determining the half-life of the curve. Correlation analyses indicated that there were low to moderations correlations between the different types of behaviour studied. Interestingly, there was no correlation between the rate of change for acceleration and the rate of change in reversal response. This suggests that rate for acceleration and reversal probability are regulated independently of each other. No correlation coefficients were considered significant. Table 2.3.3.1 Correlation coefficients between the rate of behavioural types Rate (probability) accelerations no responses pauses decelerations 1 -0.192 0.230 -0.171 accelerations -0.192 1 -0.550 -0.385 no responses 0.230 -0.550 1 0.586 pauses -0.171 -0.385 0.586 1 decelerations 0.086 -0.157 0.574 0.606 reversals  reversals 0.086 -0.157 0.574 0.606 1  27  Next, the correlation coefficient for the asymptotic level for each behavioural type was calculated. The results show that pauses and accelerations are highly correlated (correlation coefficient=0.867 (p<0.01)) while asymptotic levels for reversals and accelerations are not correlated nor are asymptotic levels for reversals and no-responses (0.052 and 0.025, respectively).  Table 2.3.3.2 Correlation coefficients between asymptote of behavioural types Asymptotic level (probability) accelerations no responses pauses decelerations 1 -0.400 -0.867 0.326 accelerations -0.400 1 0.309 -0.010 no responses -0.867* 0.309 1 -0.134 pauses 0.326 -0.010 -0.134 1 decelerations -0.052 -0.025 -0.231 -0.676 reversals  reversals -0.052 -0.025 -0.231 -0.676* 1  2.3.4 Correlation between magnitude of reversals, forward accelerations and pauses The correlation coefficient for the magnitude of reversals and the magnitude of acceleration was also calculated. The magnitude of accelerations showed a slight facilitation before decrementing back to its initial response (Figure 2.3.4). In 1996, Wicks ablated members of the reverse and forward sub-circuit (PLM and ALM & AVM, respectively) and found that the forward circuit habituated at a different rate than the reversal circuit. ALM & AVM ablated worms showed an initial facilitation of accelerations which subsequently decremented. Taken together the data from these ablations suggested that the final habituation curve for reversal distance is the result of the integration of the habituation kinetics of the forward and reversal subcircuits (Wicks, 1996a).  28  Acceleration Magnitude 0.45  Change in speed (mm/ms)  0.4 0.35 0.3 0.25  N2 acceleration magnitude  0.2 0.15 0.1 0.05 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29  Figure 2.3.4 Change in acceleration magnitude during habituation training. Error bars represent s.e.m.  Correlation analysis between reversal distance and acceleration magnitude show that the reversal distance habituation rate is moderately correlated to the change in acceleration magnitude (correlation coefficient =0.376). This is also true for the asymptotic level (correlation coefficient=0.350). However, neither correlation coefficients were significant. Table 2.3.4 Correlation coefficients for magnitude of accelerations, reversals, decelerations and pauses.  Acceleration rate Acceleration asymptote  Reversal distance rate 0.375 0.208  Reversal distance asymptote 0.208 0.350  2.4 Discussion 2.4.1 The decrement in reversal probability is associated with the increment of other tap-elicited behaviours The tap withdrawal response is produced by the integration of information from two competing sub-circuits, the head touch reversal circuit and the tail touch forward movement circuit (Wicks, 1996). The rate of increase in accelerations, pauses, decelerations and no 29  responses over the 30 stimuli was quantified to gain insights into how each sub-circuit changed over the course of habituation. Since both the forward and reverse circuits are recruited during a tap stimulus and since worms are biased towards reversals after a tap, it is hypothesized that the reverse sub-circuit drives movement more strongly than the forward sub-circuit and introduces the possibility that the decrement in reversal response corresponds to an increment in other behaviours. Reconstructing the reversal curve with data from accelerations, pauses and decelerations demonstrated that accelerations alone do not account for the majority of the nonreversal behaviour; pauses, decelerations and accelerations together tend to account for nonreversal behaviour. The proportion of no-responses during habituation training was relatively small compared to the other types of behaviour. Taken together, these data show that, as reversal behaviour decreases, other types of behaviour increase.  2.4.2 Correlation analysis between tap-induced behaviours The relationship between accelerations, pauses, no responses and decelerations were examined through correlation analysis. While the probability of non-reversal types of behaviour are known to increase while reversal probability decreases, it is unclear whether the mechanisms that cause their increase are regulated independently or whether they represent one behavioural state on a continuum of non-reversal movements. In order to address this, correlation coefficients for rate, initial response and asymptotic level were calculated for each type of behaviour. This was done by fitting each curve to a positive or negative exponential curve and then finding the half-life (rate) and the final y-value for the last tap (asymptotic level). Correlation analysis demonstrated that the rate of change in acceleration was not correlated to that of reversals ( in terms of response probability). Similarly, the asymptotic level for acceleration and reversals was not correlated. Taken together, this suggests that the forward sub-circuit and the reversal sub30  circuit are modulated independent of each other in the regulation of both response rate and response level (significance was set using the Holm-Bonferroni correction for multiple comparisions). If acceleration and reversal responses are regulated independently of one another, the other non-reversal behavioural types may be more closely associated with either accelerations or reversal behaviours. Comparing the correlation coefficients of accelerations and reversals with the other non-reversal behaviour types show that there is a greater association between reversals and decelerations compared to accelerations and decelerations. Table 2.4.2.1 Correlation coefficient between accelerations and reversals for response rate and asymptotic level in wild-type animals Rate Asymptotic level accelerations reversals accelerations reversals 1 0.086 1 -0.052 accelerations accelerations 0.192 -0.157 -0.400 -0.025 no responses no responses 0.230 -0.574 -0.867* -0.231 pauses pauses 0.171 -0.606 0.326 -0.676* decelerations decelerations 0.086 1 -0.052 1 reversals reversals  For response probability asymptotic level of wild-type worms, pauses were more associated with accelerations than with reversals ( (p<0.01)). Similarly, no-responses had a higher correlation coefficient with accelerations compared to reversals. Finally asymptotic level of decelerations were more tightly associated with reversals than with accelerations. While more information is needed about the competition within tap withdrawal circuit, correlation analyses can lead to hypotheses about how a worm shifts its behaviour over the course of habituation depending on the relative strength of each sub-circuit. Wicks (1995) found that if ALM and AVM (neurons in the forward sub-circuit) were ablated, these worms would only respond to tap with accelerations. Conversely, if PLM & PVD were ablated, animals would  31  only respond to tap with reversals. This demonstrated that, when there is no competition from the forward sub-circuit, C.elegans will respond to tap with a reversal. Conversely, when there is no competition from the reverse sub-circuit, C.elegans will respond to tap with accelerations (Wicks, 1995). Presumably, as the relative strength of the forward sub-circuit increases over habituation training, worms shift behaviours from reversals to accelerations. If acceleration and reversals represent two behavioural extremes with response to tap, then correlation analysis between these two responses with other non-reversal behaviours can give insight into how the degree of competition within the tap-withdrawal circuit alters a worms’ response to tap. Decelerations are more closely associated with reversals (asymptotic level) (p<0.01) whereas pauses are more associated with accelerations (p<0.01). Both of these correlation coefficients are significant after Holm-Bonferroni correction. Together this suggests decelerations may occur when there is a relative small level of competition from the forward sub-circuit. When competition increases, pausing behaviour may increase. This conclusion might be an oversimplification since behaviours could be modulated in overlapping areas of the forward and reversal sub-circuits (i.e. the command interneurons) or possibly outside of our defined circuit. Additionally, the relatively low sample size for this type of analysis (N=12 plates) and the reliance of a line of best-fit to find rate and asymptotic level (which has error associated with it) may skew the calculated correlation coefficients. Nevertheless, this analysis does give a starting point in which to speculate about the relationship between tap-induced behaviours. In order to test the relationships between the tap-induced behaviours and confirm their association with either reversals or accelerations, one could attempt to manipulate the forward or reverse circuit independently and determine whether, for example, decreasing the strength of the  32  anterior circuit decreased the rate of pauses and decelerations. .  One way to decrease the strength of the anterior circuit might be to manipulate dopamine  signaling. In Kindt et al. (2007) a food/dopamine dependent pathway that modulated reversal probability after repeated taps was characterized. Wild-type C.elegans habituate more rapidly in the absence of food compared to in its presence, while worms deficient in dopamine signalling show the same slower rate of habituation on or off of food (Kindt 2007). Kindt et al reported a dopamine dependent more rapid decrease in Ca+ signalling to repeated taps in the ALM sensory neurons, but not in the PLM sensory neurons. This fast rate of habituation in the ALM neurons would weaken the reversal sub-circuit. My hypothesis is that if ALM excitability is reduced off of food then the PLM circuit should be stronger and I predict that greater proportional input from the tail cells will lead to a higher probability of decelerations compared to when worms are on food. I tested this hypothesis by analyzing the behaviour of worms on and off food (Figure 2.4.2). Consistent with the correlation analysis, the decrement in rate and asymptotic level for reversal probability is coupled with a striking increase in the frequency of decelerations (and relatively little change in accelerations or pauses). If down-regulating ALM (and the reversal sub-circuit) results in increased decelerations, perhaps the combination of down-regulating ALM and up-regulating PLM or neurons within the forward sub-circuit is needed to produce pauses and accelerations.  33  Figure 2.4.2 Tap-induced responses of N2 worms on and off food. (Left) worms on food show a near equal increment of pauses, accelerations and decelerations corresponding to a decrement in the proportion of reversals. (Right) Worm undergoing training in the absence of food exhibit an increase in decelerations coupled with an increase rate of habituation for reversal  2.4.3 Nature of habituation for reversal probability and the C.elegans startle response Habituation is defined as the reduction of behaviour in response to repeated stimuli. Previous studies in C.elegans habituation have focused on the reversal response and its decrement after repeated taps, however, C.elegans produce multiple behaviours in response to tap. By studying all of the behaviours produced after a tap, one can interrogate the mechanics of the worm’s startle response (which encompasses all behavioural responses to tap opposed to just the reversal responses). Since the tap-reversal response is mediated by at least two opposing subcircuits, decreasing the output of one sub-circuit might result in an inevitable increase in output of the opposing sub-circuit. In this case, the apparent strengthening of the opposing circuit would be (relatively) passive since, mechanistically, what occurred within the opposing circuit to cause 34  the increased behavioural output happened in reaction to changes in its counter-part. Another possibility is that tap-stimuli induces changes in both circuits which results in greater competition and a bias away from reversals. Since the relative strength of each sub-circuit is unknown, neither model can be ruled out. Interestingly, worms undergoing habituation off of food show an increase in the proportion of decelerations without significant changes to the proportion of accelerations or pauses. Since the absence of food should result reduced cell excitability of ALM specifically, perhaps more competition within the circuit is required to bias the non-reversal tap-induced behaviours from decelerations to pauses or accelerations. Based on this the sequence of gradual change in the relationship between head and tail input would be reversals dominant, followed by increased decelerations, followed by increased pauses, followed by increased accelerations. There would be no increase in “no responses” until both the head and tail circuits have habituated. I would predict that there would be an increase in no responses after 40 or 50 tap stimuli.  2.4.4 The magnitude of the reversals and accelerations decrements over repeated taps Consistent with Wicks (1996a), the magnitude of accelerations showed a slight facilitation which then decremented with repeated taps. This is consistent with the hypothesis that the final habituation curve is the result of integration of the forward and reversal subcircuits.  2.4.5 Mechanisms of changes in tap-induced behaviour Unlike the results for reversal and acceleration probability, the reversal distance for all of these behavioural types decreased during habituation training. One riddle to decipher is how and where behavioural types are regulated (for magnitude and probability) to produce such diverse 35  changes in behaviour simultaneously. There are three possible ways in which decrements in magnitude for one behavioural type can occur at the same time as the probability of the same behavioural type increases. First, it is possible that magnitude of response is regulated in a part of the tap-withdrawal circuit that effects both forward and reverse movement equally (i.e. localized to a neuron or set of neurons that form chemical and/or electrical connections to both the forward and reverse sub-circuits). In this case, it is possible that a neuron (or set of neurons) act downstream or in parallel of the locus of habituation for probability to decrease the output from the forward and reverse sub-circuit equally. The second possibility is that the mechanisms that govern habituation of response magnitude are localized in the same neuron(s) that reversal probability is regulated, however, the molecular mechanisms that regulate probability and magnitude are independent of each other; activation of one pathway has no effect on the other pathway. In this scenario, either differences in the level of activation of the forward or reversal sub-circuit, dissimilarities in gene expression and/or non-autonomous signaling between either sub-circuit could account for why the magnitude of locomotion (be it forward or reverse) decreases whereas the probability of a type of movement occurring simply shifts. The third possibility is that reversal magnitude and reversal probability are regulated differently both in terms of molecular mechanisms and locus in the tap-withdrawal circuit. The mode of regulation governing the magnitude and probability of behaviour is beyond the scope of this study. In order to gain insight into this topic, there are several approaches that can be taken. We know that in larval worms the AVM neuron has not yet developed, and the probability of a reversal or an acceleration are equal (Chiba and Rankin 1992). Analyzing data from habituation of larval worms in which the two subcircuits are equally weighted might shed light on these issues. Further studies using genetic manipulations can also be used. In broad  36  terms, more information of the molecular mechanisms behind habituation for reversal magnitude and probability can be teased out from Giles et al.’s large-scale screen. Predictions based on cluster analysis can be confirmed and cell-specific rescue experiments can be used to find the locus of habituation for magnitude and probability. The major caveat in this approach is that the mechanisms that regulate habituation kinetics are complex and involve at least two possible categories of defects (rate and asymptotic level). Rescue experiments typically result in overexpression of the gene being rescued, which may affect the habituation curve.  2.5 Conclusion Through detailed behavioural analysis, I showed that worms exhibit several different behavioural responses to tap that change in probability over the course of habituation; worms shift their behaviour from reversals to pauses, decelerations or accelerations. Worm that crawled through a tap without changing their speed only represented a small portion of the overall behaviour. Correlation analysis demonstrated that both rate and asymptotic level for accelerations is not correlated with the corresponding value for reversals. Additionally, the asymptotic level of decelerations was correlated with reversals while the asymptotic level of pauses was correlated with accelerations. Unlike the results for probability, the magnitude of reversals and accelerations both decremented after repeated taps. This suggests that one of three types of regulation is occurring. Either the locus of plasticity for the magnitude of response occurs parallel to or downstream of the site of plasticity for probability; that both response types are regulated by different molecular mechanisms occurring in the same neurons; or a combination of diverse sets of neurons and independent molecular mechanisms regulate behaviour for magnitude and probability. Testing the mechanisms proposed in Giles et al.(in preparation) for probability and magnitude will give further insights into the locus and nature of 37  how habituation is regulated. This analysis offers a new tool with which to examine the genes that play a role in habituation. Now that accelerations, pauses and decelerations can be analysed, genes that preferentially alter one sub-circuit in the tap-withdrawal circuit can be identified and allow us to gain insight the circuit-based regulation of habituation and behavioural shifts. This approach might also help in the identification of genes that distinguish the head and tail mechanosensory neurons, as to date no genes have been identified that are expressed in just ALMs or just PLMs.  3 Experiment 2a: Role of G protein pathways in habituation 3.1 Introduction Heterotrimeric G-proteins have important and diverse roles in development and neuronal function. There are three families of heterotrimeric G-protein signaling pathways whose members are highly expressed in neurons. These are Gαi, Gαq and Gαs. Through a series of experiments involving rate of egg-laying, locomotion and aldicarb sensitivity studies, the general role of each of these families of heterotrimeric G-protein signaling pathway in C.elegans have been elucidated (Miller 1999, Nurrish 1999, Yvonne M. Hajdu-Cronin 1999, Schade 2005, Reynolds 2005). Generally speaking, Gαi inhibits neurotransmitter release whereas Gαq and Gαs promote it. The Gαq pathway has been described as a critical component of vesicle priming pathway and acts by catalyzing the synthesis of diacylglycerol (DAG) and IP3. DAG then in activates downstream responses and binds directly to unc-13, a protein that directly regulates neurotransmitter release (Lackner et al., 1999). Gαi inhibits neurotransmitter release by inhibiting the Gαq pathway (Lackner 1999, Miller 1999, Nurrish 1999). Once the Gαi signaling pathway has been activated, goa-1 (the α subunit in the Gαi pathway) positively regulates eat-16, an RGS protein that inhibits egl-30 (the α subunit in the Gαq pathway) (Hajdu-Cronin et al, 38  1999). Finally, the Gαs pathway modulates the activity of the Gαq signaling pathway (Reynolds et al, 2005). Through epistatic analysis, it has been shown that Gαs pathway activity relies heavily on the function of the Gαq pathway, however, it also has Gαq-independent roles. Taken together these data led to the hypothesis that Gαs modulates Gαq signaling at specific synapses and is important for coordinated movement (Reynolds et al, 2005). The heterotrimeric G-protein signaling pathway in C.elegans was first shown to be important in habituation in a study by Kindt et al.(2007) who found that C.elegans decrease their reversal responses to repeated tap stimuli more rapidly when food is present compared to when food is absent. Through a candidate gene approach, Kindt et al. found that the food-dependent modulation of habituation acts through the G-protein coupled (dopamine) receptor, dop-1. By testing mutations of α subunits in each family of heterotrimeric G-protein signaling pathways, they found that egl-30/Gαq phenocopied dop-1 and likely acted downstream of the receptor. Since dop-1 is expressed in the mechanosensory neurons ALM and PLM, calcium imaging was performed on these cells in order to determine whether the observed behavioural changes corresponded to changes in calcium influx. The results showed that ALM but not PLM exhibited a more rapid decrement in calcium influx in dop-1 mutants compared to controls. This suggested that the fast habituation phenotype corresponded to reduced cell excitability of the ALM neurons. Loss-of-function of dop-1 attenuated the decrement in cell-excitability in the presence of food and caused the animals to remain responsive to repeated tap stimuli when food was present. This process was mediated by the Gαq/egl-30 signaling pathway (Kindt et al, 2007). Further evidence that heterotrimeric G-protein signaling has a role in governing habituation came after Giles et al. completed a high-throughput screen for mutants with habituation defects. Giles et al. found that goa-1 and eat-16 (members of the Gαi signaling  39  pathway) exhibited large differences in the rate and asymptotic level of the habituation curve for both reversal distance and probability. These results lead to the hypothesis that the Gαi signaling pathway has a major role in regulating habituation kinetics (Giles et al, in preparation). Even though over two dozen different behavioural phenotypes were characterized during the high-throughput habituation screen, the only habituation measures analyzed were reversal probability, reversal distance and reversal duration. However, when a worm does not reverse to tap, it sometimes does other behaviours in response to tap. These behaviours are accelerations, decelerations, pauses and continuing normal movement (i.e. ignoring the tap). Using the approach developed for Experiment 1, members of the Gαi Gαq and Gαs were tested (or in some cases, retested) and the proportion of all tap-induced behaviours was analyzed.  3.2 Methods 3.2.1 Animals Nematodes were stored at 20ºC and cultured according to standard C.elegans techniques (Brenner 1974) Strains were ordered from the Caenorhabditis Genetics Center and subsequently tested for habituation defects: egl-30(ep271), egl-30(n715), egl-30(js126), egl-30(ad806), egl-30(n686), egl-8)32917), egl-8(n488), egl-8(ok934, egl-10(nu62), egl-10(md176), goa-1(sa734)/, , goa1(n1134), goa-1(n3055), eat-16(ce71), dgk-1(nu62), gsa-1(ce81), gsa-1(ce94), acy-1(nu329), acy-1(ce2), acy-1(md1756), acy-4(ok1806), pde-4(ce268), pde-4(ok1290), kin-2(ce179), cnb1(jh103), mod-5(n822), mod-5(n3314), gpr-1(ok2126), gpr-2(ok1179), grk-1(ok1239), twk18(cn110), unc-103(n500n1211)  40  3.2.2 Behavioural assay MWT plates were made by pouring 12 ml of NGM into 5cm petri-dishes. Plates were allowed to dry for 3 days. The day before colonies were made, 50 µl of OP50 was pipetted to the MWT plates, spread to form a uniform layer of bacteria and allowed to dry for up to 24h. Worms were synchronized by allowing 5 N2 worms to lay eggs on E. coli seeded plates for 4-5 hours. This resulted in approximately 60-80 worms per plate. When worms were 4 days old 4-6 plates (containing 60-80 worms) of each strain were tested on the multi-worm tracker. For each experiment on the MWT, worms were tracked for 10 minutes before the first stimulus was delivered to the plate (unless otherwise specified). Taps were delivered to the side of the petri dish via a push-type solenoid that fired every 10 seconds for 30 stimuli. In age-experiments, strains were tested at 96h and 120h. First, both control and mutant strains were tested at 96h. 60-80 worms were picked off their original plate and transferred to a new plate that had been seeded up to 24h beforehand. These were then stored in a 20ºC incubator for 24h and retested.  3.2.3 Behavioural data analysis For each habituation experiment .blob files generated by the multi-worm tracker were analyzed with choreography version 1.3.0. To collect locomotion data, the “Trigger” argument was used to find the average speed and bias of individual worms 1 second before the tap stimulus and 1 second afterwards. Data was outputted as .trig files with 4 columns: speed, standard deviation of speed, bias and instance of tap. A Python script was used to concatenate all .dat files for each plate. Worm bias (the direction the worm is moving) was multiplied by its speed. Since the average bias was calculated as an average, when bias was not 1 (representing forward), 0 (representing no movement) or -1 (representing reversal movement), the bias was changed from 41  its value to -1. This was to remove instances where worms moved in more than one direction during the time-frame their speed was measured. The Python script generated .err files that were used to calculate the standard-deviation between the speeds of the worms on the plate. It also generated .dat files which took the values of reversal probability and reversal distance calculated by the MeasureReversal plugin for choreography and .txt files which contained the speed data for all worms tracked during an experiment. The speed data was used to determine the proportion of accelerations, decelerations, pauses and “no responses”. Each behavioural type was calculated using the following formulas:  num_swmthr = (nansum(acc<0.5 & acc./A>-0.5 & acc~=0 & B>=2*sem & A>=2*sem)) + num_accel = nansum(acc>2 & B>2*sem & A>=0); num_pause = nansum(B<=0.5 & B>=0.5*-sem & A>=2*sem); num_decel = nansum(acc2>2 & A>2*sem & B>2*sem);  See below for a description of the variables: num_swmthr=number of worms that do not respond to tap num_accel=number of worms that accelerate to tap num_pause=number of worms that pause to tap num_decel=number of worms that decelerate to tap A=average speed 1-0 seconds before a tap stimuli B=average speed 0-1 seconds after a tap stimuli acc=B-A acc2=A-B s.e.m.=standard error of the mean for a worm’s speed  The proportion of worms engaged in each behavioural type was calculated by finding the sum of the worms meeting one of the four conditions and dividing that by the total population of worms. The sum of the behavioural types was added to insure that the value was equal to 100% + 10%. Reversal curves using the speed data were also generated to ensure that they approximated that calculated by the MeasureReversal plugin. 42  3.2.4 Statistics For each strain, 4-6 plates with 60-80 worms were analyzed. The mean for each plate was calculated and the s.e.m. was found using n=number of plates tested/strain. For each reversal curve, a line of best fit was found using the “power1” function in Matlab (based on the model y=axb). Using the best-fit line, the reversal rate, initial response and asymptotic level were found. The rate was calculated by finding the half-life of the line of best fit. The initial response and asymptotic level were found by finding the y-value at the first and last tap. Z-scores were calculated by comparing the mean and standard error of 12 control groups against all the strains tested. Z-scores were converted to p-values using the Normsdist function in excel. The Pearson correlation coefficients were found between all behaviours for rate and asymptotic level. The Holm-Bonferroni method was use to correct for multiple comparisons in table 2.4.2.1. Specifically,p-values that were less than 0.05/n-k (where n is the number of comparison and k is the rank of the p-value in order of smallest to largest) were considered significant.  3.3 Results Giles et al (in preparation) used correlation analyses of the different characteristics of habituation to demonstrate that habituation is regulated by at least 4 genetically independent pathways that control rate of habituation for reversal distance, final or asymptotic level for reversal distance, rate of habituation for reversal probability and final or asymptotic level for reversal probability. In this study habituation of multiple alleles of Gαs, Gαq, and Gαi orthologues were tested to identify which aspect of habituation (if any) they regulated. In addition, 11 strains associated with at least one G-protein pathway were tested in order to find  43  potential downstream targets or regulators. In these analyses the habituation curves were characterized in terms of rate and asymptotic level for both distance and probability.  3.4 Habitation kinetics of Gαq signaling pathway mutants Gαq is regulated by Gαi and is the core pathway responsible for acetylcholine release (Nurrish 1999, Reynolds 2005, Lackner 1999). Four strains encoding reduction of function alleles of egl-30 (the alpha subunit of the Gαq signaling pathway) (Harris et al., 2003) and one strain encoding an egl-30 gain-of-function were tested for deficits in habituation. These are egl30(js126), which results in a missense mutation, egl-30(ad806), egl-30(n686) and egl-30(n715), which have mutations in either a splice donor or splice accepter site (Brundage et al., 1996 ). Finally, egl-30(ep271) represents a gain-of-function mutation that converts a methionine to isoleucine at residue 244. Since the Gαq pathway is thought to be critical for synaptic priming, gain-of-function mutations in egl-30 should lead to increased synaptic priming and presumably increased neurotransmitter release whereas reduction of function alleles should lead to reduced neurotransmission.  Table 3.4.1 Five alleles of egl-30 that were tested for habituation defects  Name (allele)  Strain  Mutation type  Function  egl-30 (ep271)  CE1047  α subunit, Gαq  egl-30 (n715) egl-30 (js126) egl-30 (ad806) egl-30 (n686)  MT1520 NM1380 DA1084 MT1434  Missense M244I (gf) Other Missense V180M Other Other  Human Orthologue GNAQ  α subunit, Gαq α subunit, Gαq α subunit, Gαq α subunit, Gαq  GNAQ GNAQ GNAQ GNAQ  44  3.4.1 Habituation kinetics of egl-30/GNAQ for reversal distance The habituation curve for reversal distance not only gives information on the worm’s ability to learn; it also gives insight into the worm’s ability to move. egl-30 is widely expressed in the C.elegans nervous system and strong reduction-of-function mutations have been reported to be paralyzed. In order to confirm that the strains being tested for habituation defect had relatively normal locomotion, initial reversal distance was compared against controls. Initial distances for egl-30(n686) and egl-30(js126) were not significantly different compared to controls while egl-30(n715) and egl-30(ad806) showed relatively low initial reversal response distances (p=<0.001 for both alleles). In habituation of reversal distance, both egl-30(n686) and egl-30(js128) showed a slow rate of habituation (p=0.003 and p=0.005, respectively) coupled with a high asymptotic level (p=0.0023 and p=0.007, respectively) (Figure 3.4.1).  egl-30 (distance) Reversal distance (mm)  1.2 1 0.8  N2 egl-30(js126)  0.6  egl-30(ad806) 0.4  egl-30(n715) egl-30(ep271)  0.2 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29  Tap Figure 3.4.1 Reversal distance habituation of four alleles of egl-30. All alleles are reduction of function mutations except egl-30 (ep271), a gain-of-function strain.  45  3.4.2 Habituation kinetics of egl-30/GNAQ for reversal probability For each strain, the rate and asymptotic level of the habituation curve was analyzed and compared to control worms and other mutants in the same pathway. egl-30(ep271), the gain-offunction allele, exhibited a slow rate of reversal probability habituation and high asymptotic level (p=0.007 and p=0.004, respectively). This suggests that increasing neurotransmission results in a slower and shallower level of habituation of reversal probability (Figure 3.4.2.1).  Revrsal probability  1.2  egl-30 (probability)  1  N2  0.8  egl-30(ep271) 0.6  egl-30(js126)  0.4  egl-30(n686)  0.2  egl-30(n715) egl-30(ad806)  0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29  Tap Figure 3.4.2.1 Reversal probability curve of egl-30 mutants tested between 80-96 hours. Two reduction-of-function alleles of egl-30 have a fast rate of habituation compared to controls whereas two others showed slow habituation and high asymptotic level. egl-30(ep271)(gf) showed a slow rate of habituation and high asymptotic level.  The reversal probability results of the gain-of-function allele, egl-30(ep271) were compared to the reduction of function alleles (Figure 3.4.2.1). Since the gain-of-function mutation exhibited slow habituation of reversal probability, one would expect that the reduction of function to show rapid habituation for distance and probability. egl-30(js126) is the only reduction of function allele that has been mapped. egl-30(js126) encodes a missense mutation converting valine to a methionine in a coding region of the protein. Like the gain-of-function, 46  this allele exhibited a slow rate of habituation and significantly high asymptotic level (p=0.003 and p=0.007, respectively) along with egl-30(n715) (p <0.001 and p<0.001, respectively). Conversely, two other alleles of egl-30 (egl-30(n686) and egl-30(ad806)) exhibited a fast rate of habituation of reversal probability and a low asymptotic level. Timbers et al. (2012) previously found that habituation for reversal probability is shallow for younger animals. Since egl-30 mutants are known to develop more slowly than wild-type, to rule out the possibility that the shallow reversal probability phenotype was due to delayed development in the mutants, these two alleles along with the gain-of-function alleles were retested at 96 and 120 hours. If the phenotype was due to delayed development, the 120h animals should have a less severe phenotype compared to 96hs.  Reversal probability  1.2  egl-30 (120h) (probability)  1 0.8 0.6  N2  0.4  egl-30(ep271) 96h egl-30(ep271) 120h  0.2 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29  Tap Figure 3.4.2.2 egl-30(ep271) gain of function allele tested at 96h and 120h. egl-30 strains were tested at 96h and to determine whether these mutants were developmentally delayed. Older animals showed an increased rate of habituation and lower asymptotic level suggesting egl-30 mutants are slow growing .  47  egl-30 (120h) (probability)  Reversal probability  1 0.8 0.6  N2 0.4  egl-30(js126) 96h  0.2  egl-30(js126) 120h  0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29  Tap Figure 3.4.2.3 egl-30(js126) gain of function allele tested at 96h and 120h. egl-30 strains were tested at 96h and to determine whether these mutants were developmentally delayed. Older animals showed an increased rate of habituation and lower asymptotic level suggesting egl-30 mutants are slow growing.  Reversal probability  1  egl-30 (120h) (probability)  0.8 0.6  N2 0.4  egl-30(n686) 96h  0.2  egl-30(n686) 120h  0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29  Tap Figure 3.4.2.4 egl-30(n686) gain of function allele tested at 96h and 120h. egl-30 strains were tested at 96h and to determine whether these mutants were developmentally delayed. Older animals showed an increased rate of habituation and lower asymptotic level suggesting egl-30 mutants are slow growing.  The reversal probability habituation curve of 120h old mutants of egl-30(js126) and egl-30(n686) was compared with 96h old control animals (Figures 3.4.2.2-3.4.2.4). These strains either habituated more quickly or at the same rate as N2 suggesting the possibility that the slow habituation phenotype was due to age-effects, and not due to a role in habituation (p<0.001 for 48  egl-30(js126) at 120h and p<0.001 for egl-30(n686) at 120h). In contrast, egl-30(n715) did not show a robust change in habituation rate or asymptotic level (reversal probability) at the older age. The gain-of-function allele, egl-30(ep271), showed a slight increase in rate of habituation, however, the rate was still much slower compared to controls (p<0.001). Taken together these results suggest that disrupting egl-30 function resulted in slower development in the worm and this delay in development altered the rate and asymptotic level of the reversal probability habituation curve. When egl-30 function was constitutively active, habituation was slow and less deep. The effect of reducing egl-30 function on habituation was less clear. Of the four egl-30 reduction-of-function mutant strains tested at 80-96 hours, two alleles showed fast habituation with low asymptotic level while two others showed slow habitation with high asymptotic level. When retesting egl-30 mutants at an older age, egl-30(js126) and egl-30(n686) but not egl30(n715) showed an increase in habituation rate suggesting that reducing function of egl-30 results in fast habituation, however, this effect can be masked by the slower rate of development compared to controls.  3.4.3 Habituation kinetics of genes downstream of egl-30. Because egl-30 appeared to have a strong effect on habituation, genes acting downstream of egl-30 were tested for habituation defects. In both vertebrate and C.elegans, the egl-30 homologue, GNAQ, activates PLCβ and subsequently hydrolyzes phosphatidylinositol bisphosphate (PIP2) to inositol 1,4,5-trisphosphate (IP3) and diacylglycerol (DAG). IP3 mediates the release of calcium from the endoplasmic reticulum and/or the sarcoplasmic reticulum. DAG mediates vesicle fusion at the synapse, and consequently, synaptic transmission (reviewed in Sternweis and Smrcka, 1993; Jiang and Simon, 1994). Likewise, egl-8, the orthologue of phospholipase Cβ (PLCβ), acts downstream of egl-30 to produce DAG and IP3 49  and ultimately regulates acetylcholine release (reviewed in Bastiani and Mendel, 2006). The activity of the Gαq signaling pathway is modulated by egl-10 and dgk-1. egl-10 is an RGS protein that negatively regulates goa-1 and the Gαi signaling pathway while dgk-1 is a diacyclglycerol kinase that is activated indirectly by goa-1/ Gαi and negatively regulates Gαq (Koelle and Horvitz, 1996; Hajdu-Cronin et al., 1999). Loss-of-function alleles of egl-10 should result in reduced Gαq signaling whereas loss-of-function mutations in dgk-1 should alleviate some repression of the Gαq pathway and increase Gαq signaling. To test this, the habituation kinetics for three alleles of egl-8, 3 alleles of egl-10 and 1 allele of dgk-1 were examined.  Table 3.4.3 Members of the Gαq signaling pathway that were tested for habituation defects  Name (allele)  Strain  Mutation type  Function  egl-8 (e2917) egl-8 (n488) egl-8 (ok934) egl-10 (n692) egl-10 (md176) dgk-1 (nu62)  CB6614 MT1083 RB1012 MT1443 MT8504 KP1097  Insertion Deletion Other Nonsense Other Nonsense  Lipase Lipase Lipase RGS RGS Diacylglycerol kinase  Human Orthologue PLCβ PLCβ PLCβ RGS7 RGS7 -  3.4.4 Reversal distance of genes downstream of egl-30/GNAQ Six strains representing different alleles of egl-8, egl-10 and dgk-1 were investigated for defects in habituation for reversal distance. elg-8/PLCβ acts downstream of egl-30 to catalyze the formation of DAG and facilitate acetylcholine release (reviewed in Bastiani and Mendel, 2006). Three alleles of egl-8 were tested for defects in habituation of reversal distance (Figure 3.4.4). All three alleles of egl-8 showed significantly higher asymptotic level than wildtype for reversal distance (p<0.001, p=0.016, and p<0.001, for egl-8(e2917), egl-8(n488) and egl-8(ok934), respectively) however, egl-8(e2917) also showed a slightly lower initial distance compared to 50  wild-type suggesting that it may have some locomotion defects (p=0.004). Next, two alleles of egl-10 were tested. egl-10 is a negative regulator of the Gαi signaling pathway and loss-offunction mutations should result in an increased Gαq activity. Both alleles showed a striking increase in rate of habituation (p=0.032 for egl-10(md176) and p<0.001 for egl-10 (n692)) and low asymptotic level for reversal distance. Finally, dgk-1 (a diacylglycerol kinase) was tested. dgk-1 degrades DAG generated by the Gαq signaling pathway. Loss-of-function mutations in dgk-1 should phenocopy gain-of-function mutations in egl-30. Indeed, dgk-1(nu62) showed slow rate of habituation and high asymptotic level for reversal distance (p<0.001 for both measures) (Figure 3.4.4).  Gαq associated genes (distance) Reversal distance (mm)  1.4 1.2  N2  1  egl-8(ok934)  0.8  egl-8(e2917)  0.6  egl-8(n488)  0.4  egl-10(n692)  0.2  egl-10(md176) dgk-1(nu62)  0 1  3  5  7  9 11 13 15 17 19 21 23 25 27 29  Tap Figure 3.4.4 Reversal distance habituation of members of the Gαq signaling pathway. Error bars represent s.e.m.  3.4.5 Reversal probability of genes downstream of egl-30/GNAQ Giles et al. (in preparation) showed that reversal distance and reversal probability are not correlated, thus for each of the genes tested I also analyzed reversal probability. As with reversal distance, egl-8(e2917)/PLCβ showed the strongest phenotype for reversal probability which 51  consisted of a mild increase in the rate of habituation and a slightly lower asymptotic level compared to controls. Neither egl-8(n488) nor egl-8(ok934) were different from wild-type for reversal probability. Next, habituation of reverse probability for two alleles of egl-10 (a negative regulator of the Gαi pathway) was examined. egl-10(n962) has a premature stop site and is likely null. egl-10(md176) contains an uncharacterized mutation. Again, neither allele showed any deficits in habituation of reversal probability. Lastly, dgk-1(nu62) was tested. If the slow habituation phenotype exhibited by the egl-30 gain-of-function mutation is the result of increased DAG production, loss-of-function mutations in dgk-1 should phenocopy gain-of-function mutations in egl-30. Indeed, dgk-1 exhibits a slow rate of habituation (p<0.001) and high asymptotic level for reversal probability (p<0.001 ). Unlike egl-30, dgk-1 is not reported to have a slow growth rate. This suggests that egl-30 and dgk-1 regulate the rate of habituation with respect to reversal probability by modulating DAG levels (Figure 3.4.5).  Gαq associated genes (probability) 1.2  Reversal probability  1  N2  0.8  egl-8(ok934) egl-8(e2917)  0.6  egl-8(n488) 0.4  egl-10(n692) egl-10(md176)  0.2  dgk-1(nu62) 0 1  3  5  7  9  11 13 15 17 19 21 23 25 27 29  Tap Figure 3.4.5. Reversal probability habituation of members of the Gαq signaling pathway. Error bars represent s.e.m.  52  3.5 Habitation kinetics of Gαi signaling pathway mutants According to previous studies a core function of the Gαi signaling pathway is to negatively regulate Gαq and, in general, reduce neurotransmitter release (Willson et al., 2004). Mutations in members of the Gαi pathway would presumably result in increased neurotransmitter release and phenocopy gain-of function mutations in the Gαq signaling pathway. In order to determine whether this relationship holds for habituation, mutants from the Gαi signaling pathway were tested habituation defects. Three different alleles of goa-1/GNAO (the α subunit in the Gαi signaling pathway) were used in this study: 3.5 Alleles of goa-1 (from the Gαi signaling pathway) tested for defects in habituation  Name (allele)  Strain  Mutation type  Function  goa-1 (sa734) goa-1 (n1134) goa-1 (n3055)  DG1856 MT2426 MT8628  Nonsense Missense M1I Other  α subunit, Gαi α subunit, Gαi α subunit, Gαi  Human Orthologue GNAO GNAO GNAO  Of the three alleles tested, goa-1(n1134), goa-1(sa734) are putative null mutations. goa1(n1134) has a point mutation which converts the initial start codon to isoleucine whereas goa1(sa734) contains a nonsense mutation which results in the conversion of Q to stop codon in the first exon. The last allele tested goa-1(n3055) has an uncharacterized mutation and was isolated in a forward genetic screen. This allele had a relatively weak locomotion phenotype compared to goa-1(n363) (another uncharacterized mutant) suggesting that it is a weak loss-of-function allele (Sawin et al., 2000).  3.5.1 Habituation kinetics of goa-1/GNAO for reversal distance Of these mutations in the GNAO pathway goa-1(n1134) and goa-1(sa734) had the strongest habituation phenotypes and exhibited defects in all categories of habituation. Similar to  53  the egl-30 gain-of-function mutant, goa-1(n1134) and goa-1(sa734) showed a slow rate of reversal distance habituation compared to controls (p<0.001 and p<0.001, respectively). The last allele of goa-1 to be tested was goa-1(n3055) a mutation that has not yet been characterized and has the weakest phenotype of all goa-1 strains. goa-1(n3055) had a normal rate of habituation (reversal distance) and higher asymptotic level (p<0.001 and p<0.001, respectively).  Gαi signaling pathway (distance) 1.2  Reversal distance  1 0.8 N2  0.6  goa-1(sa734) goa-1(n1134)  0.4  goa-1(n3055) 0.2 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 Tap  Figure 3.5.1 Reversal distance habituation curves of goa-1 (of the Gαi signaling pathway). Error bars represent s.e.m.  3.5.2 Habituation kinetics of goa-1/GNAQ for reversal probability Similar to the result from the reversal distance, goa-1(n1134) and goa-1(sa734) showed a slow rate of habituation (reversal probability) (p<0.001 and p<0.001, respectively) and higher asymptotic level compared to wild-type worms (p<0.001 for goa-1(n1134) and p<0.001 for goa1(sa734)). goa-1(n3055), the weakest allele of the four showed a slightly higher asymptotic level 54  for reversal probability habituation (p=0.037). Although not all alleles photocopied each other in terms of habituation of reversal probability, the phenotypes exhibited by goa-1(sa734) and goa1(n1134) probably represent the best picture of the ways habituation is disrupted in goa-1 null mutants. This is because goa-1(n3055) has previously been described as a weak allele and is likely not null.  goa-1 (probability) 1  Reversal probability  0.9  0.8 0.7 0.6  N2  0.5  goa-1(sa734)  0.4  goa-1(n1134)  0.3  goa-1(n3055)  0.2 0.1 0 1  3  5  7  9  11 13 15 17 19 21 23 25 27 29  Tap Figure 3.5.2 Reversal probability habituation of goa-1. Error bars represent s.e.m.  3.5.3 Habituation kinetics of genes downstream of goa-1/GNAQ for reversal probability Next, known regulators of the Gαi signaling pathway were tested for their specific habituation kinetics. After goa-1/GNAO is activated, a downstream effector of goa-1 called eat16 suppresses activity of egl-30/GNAQ and the Gαq signaling pathway. goa-1 activity also indirectly activates dgk-1, an enzyme that decreases DAG levels produced from the Gαq 55  signaling pathway. Loss of eat-16 and dgk-1 should result in decreased Gαi signaling. Two alleles of eat-16 and one allele of dgk-1 were compared with controls and goa-1 mutants for differences in habituation kinetics.  3.5.3 Members of the Gαi signaling pathway that were tested for habituation defects  Name (allele)  Strain  Mutation type  Function  eat-16 (sa609) eat-16 (ce71) dgk-1 (nu62)  JT609 KG571 KP1097  Missense R396C Other Nonsense  regulator (rgs) regulator (rgs) diacylglycerol kinase  Human Orthologue -  3.5.4 Reversal distance of genes downstream of goa-1/GNAO Like goa-1(sa734), eat-16(sa609) exhibited deficits both reversal distance and reversal probability. eat-16(sa609) showed a slow rate of habituation of reversal distance (p<0.001) coupled with a high asymptotic level (p<0.001). The initial reversal distance was not different from wild-type. eat-16(ce71) also showed a slow rate of reversal distance habituation and high asymptotic level (p=0.006 and p<0.001). Finally, as described before, dgk-1(nu62) shares a similar phenotype to that of eat-16(sa609) and the gain-of-function allele of egl-30. Specifically, dgk-1(nu62) exhibited slow habituation rate (p<0.001) and high asymptotic level for reversal distance (p<0.001) (Figure 3.5.4).  56  Gαi signaling pathway (distance) Reversal distance  1.2 1 0.8  N2  0.6  eat-16(sa609)  0.4  eat-16(ce71)  0.2  dgk-1(nu62)  0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 Tap Figure 3.5.4 Reversal Distance habituation of members of the Gαi signaling pathway. Error bars represent s.e.m.  3.5.5 Reversal probability of genes downstream of goa-1/GNAO Like goa-1, both alleles of eat-16 showed significantly high asymptotic levels and slow habituation rates for reversal probability, For reversal probability, dgk-1(nu62) showed a slow habituation rate and high asymptotic level (p<0.001 and p<0.001, respectively). In general, the eat-16 and dgk-1 mutants strains showed strong similarities to the goa-1 loss-of-function mutants. These results strongly favour the proposition that eat-16 and dgk-1 act downstream of goa-1 to regulate habituation rate and asymptotic level of reversal probability. Overall, lesions in goa-1 resulted in slow habituation and high asymptotic level for both distance and probability. Since the egl-30 gain-of-function mutation phenocopies goa-1 loss-offunction and lesions in egl-30 and goa-1 result in widespread habituation defects, the Gαi and Gαq pathways may interact act as general regulators of habituation (Figure 3.5.5).  57  Gαi associated genes (probability) Reversal probability  1.2 1 0.8  N2  0.6  eat-16(sa609)  0.4  eat-16(ce71)  0.2  dgk-1(nu62)  0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29  Tap Figure 3.5.5 Reversal probability habituation of members of the Gαi signaling pathway. Error bars represent s.e.m.  3.6 Habitation kinetics of Gαs signaling pathway mutants The main function of the Gαs pathway is less well understood. Like Gαq, Gαs is thought to positively regulate neurotransmitter release. Counter-intuitively, Gαs is not required for vesicle priming; Gαs null synapses still exhibit normal or even elevated neurotransmitter release and mutant strains defective in Gαs signaling still show normal steady-state neurotransmitter release in motorneurons (Hou 2003 et al.; Renden et al. 2003; Reynolds et al. 2005; Charlie et al. 2006). Epistatic analysis between the Gαq and Gαs pathway demonstrated that the function of Gαs is largely dependent on Gαq and that the Gαs pathway likely converges on Gαq downstream of DAG (Reynolds et al.2005). Reynolds et al. postulate that, at least for locomotion, the Gαq pathway regulates neurotransmitter release and the Gαs pathway modulates the release at specific synapses to produce coordinated movement. To understand the role of the Gαs signaling pathway in habituation, two gain-of-function alleles of gsa-1/GNAS (the α subunit of the Gαs signaling pathway) were tested for their effect on habituation. Loss-of-function alleles of gsa-1 could not be tested as null alleles result in lethality.  58  Table 3.6 Alleles of gsa-1 (from the Gαs signaling pathway) that were tested for habituation defects  Name (allele)  Strain  Mutation type  Function  Human Orthologue  gsa-1 (ce81)  KG421  α subunit, Gαs  gsa-1 (ce94)  KG524  Missense R185C (gf) Other (gf)  α subunit, Gαs  3.6.1 Habituation kinetics of gsa-1/GNAS for reversal distance Gain-of-function alleles of gsa-1 were tested for their effect on habituation. The Gαs signaling pathway is thought to both function in parallel and downstream of the Gαq signaling pathway. Of the two alleles of gsa-1 tested, both have point mutations that render the protein constitutively active. gsa-1 gain-of-function mutants showed relatively normal habituation kinetics for reversal distance. gsa-1(ce81) did not show any change in reversal distance habituation. resulted in a fast rate of habituation and a normal asymptotic level. gsa-1(ce94) had a significantly low asymptotic level (p=0.038), however, this could be the result of a type II error.  gsa-1 (distance)  Reversal distance (mm)  1.4 1.2  N2  1 0.8  gsa-1(ce81)  0.6  gsa-1(ce94)  0.4 0.2 0 1  3  5  7  9  11 13 15 17 19 21 23 25 27 29  Tap Figure 3.6.1 Reversal distance habituation of gsa-1. Error bars represent s.e.m.  59  3.6.2 Habituation kinetics of gsa-1/GNAS for reversal probability Both alleles of gsa-1 exhibited a rapid rate of habituation and no effect on asymptotic level of reverse probability. The initial response was not significantly different from controls (Figure 3.6.2).  1.2  gsa-1 (probability)  Reversal probability  1 0.8  N2  0.6  gsa-1(ce81) 0.4  gsa-1(ce94)  0.2 0 1  3  5  7  9 11 13 15 17 19 21 23 25 27 29  Tap Figure 3.6.2 Reversal distance habituation of gsa-1. Error bars represent s.e.m.  3.6.3 Habituation kinetics of genes downstream of gsa-1/GNAS acy-1, which encodes an adenylyl cyclase that converts ATP to cAMP, is a major downstream target of gsa-1. cAMP then activates cAMP-dependent kinases including protein kinase A (PKA). cAMP levels are regulated by pde-4, a phosphodiesterase that converts cAMP to AMP (reviewed in Sunahara et al., 1996; Walsh and Van Patten, 1994; Hobson et al. 2003). Three alleles of acy-1 and one allele of kin-2 (a subunit of PKA) and two alleles of pde-4 were tested for their role in habituation.  60  Table 3.6.3 Members of the Gαs signaling pathway that were tested for habituation defects  Name (allele)  Strain  Mutation type  Function  acy-1 (nu329) acy-1 (ce2) acy-1 (md1756) kin-2(ce172)  KP1182 KG518 KG522 KG532  Other Other (gf) Other (gf) Other  pde-4 (ce268) pde-4 (ok1290)  KG744 RB1231  Missense D323N Other  adenylyl cyclase adenylyl cyclase adenylyl cyclase negative regulator PKA phosphodiesterase phosphodiesterase  Human Orthologue ADCY9 ADCY9 ADCY9 PRKAR1B -  3.6.4 Reversal distance of genes downstream of gsa-1/GNAS Of the three alleles of acy-1 that were tested, two strains represented gain-of-function mutations and one encoded a reduction-of-function allele of acy-1. acy-1 gain-of-function mutations showed a rapid (although not significantly different) rate of habituation and normal asymptotic level (Figure 3.6.5). acy-1(nu329), the reduction of function allele exhibited a normal rate and asymptotic level of habituation. The Gαs signaling pathway results in the production cAMP (Hobson et al. 2003). Therefore, mutant strains of pde-4, a gene responsible for degrading cAMP, was tested for habituation defects. Loss-of-function mutations in pde-4 should result in increased levels of cAMP (as gsa-1 and acy-1 gain-of-function mutations do; Figure 3.6.4). pde-4(ce268) contains a missense mutation that effects five of the six isoforms whereas pde-4(ok1290) contains a deletion whose location is unknown. For reversal distance pde-4(ok1290) exhibited a high asymptotic level and fast rate of habituation. pde-4(ce268) mutant worms that underwent habituation training did not show a significant difference in either habituation rate or asymptotic level of reversal distance. Activation of gsa-1 also results in the activation of both protein kinase A (PKA)dependent and independent pathways (reviewed in Sunahara et al., 1996; Walsh and Van Patten, 61  1994). The role of PKA in habituation kinetics was examined. Loss-of-function mutations in kin1/PKA result in lethality; however, kin-1 forms a complex with and is negatively regulated by kin-2 (Schade et al. 2005). Loss-of-function mutations in kin-2 result in a constitutively active kin-1/PKA and should provide insights into its role in habituation. kin-2(ce179) mutants showed normal reversal distance habitation rate and asymptotic level, however, the initial reversal distance was also lower than wild-type (Figure 3.6.4).  Gαs-associated genes (distance) 1.6  Reversal distance (mm)  1.4 1.2  N2 1  acy-1(ce2) acy-1(md1756)  0.8  acy-1(nu392)  0.6  pde-4(ce268) 0.4  pde-4(ok1290)  0.2  kin-2(ce179)  0 1  3  5  7  9 11 13 15 17 19 21 23 25 27 29  Tap Figure 3.6.4 Reversal distance habituation of members of the Gαs signaling pathway. Error bars represent s.e.m.  3.6.5 Reversal probability of genes downstream of gsa-1/GNAS The effect of mutations in genes downstream of gsa-1 on habituation of reversal probability was also determined. Two gain-of-function alleles were tested along with one reduction-of-function. The acy-1 gain-of-function alleles showed a relatively weak effect on habituation of reversal probability compared to gsa-1 (Figure 3.6.5). acy-1(ce2) showed an 62  increase in rate for reversal probability and no difference in asymptotic level compared to wildtype. acy-1(md1756) did not show significant differences in habituation of reversal probability compared to wild-type. The reduction of function allele acy-1(nu329), showed a slow rate of reversal probability habituation (p=0.002) and a higher asymptotic level (p=0.003)compared to wild-type for reversal probability (Figure 3.6.5). Next, the reversal probability habituation kinetics of the two alleles of pde-4 were examined. The initial reversal probability was normal for pde-4(ce268), however, this strained showed an increased rate of habituation (p<0.001) along with a low asymptotic level (p<0.001) (reversal probability; Figure 3.6.5). The low initial response and decreased reversal probability may be indicative of locomotion defects rather than habituation defects. Unlike pde-4(ce268), pde-4(ok1290) did not show any habituation phenotype for reversal probability. Finally, kin-2 heterozygous worms were examined for reversal probability habituation defects. kin-2(ce172) mutants showed a fast rate of reversal probability habituation (p<0.001) with a normal asymptotic level (Figure 3.6.5).  63  1.2  Gαs associated genes (probability)  Reversal probability  1 0.8  N2 acy-1(ce2)  0.6  acy-1(md1756) pde-4(ce268)  0.4  pde-4(ok1290) kin-2(ce179)  0.2 0 1  3  5  7  9  11 13 15 17 19 21 23 25 27 29  Tap Figure 3.6.5 Reversal probability habituation of members of the Gαs signaling pathway. Error bars represent s,e,m,  3.7 Discussion 3.7.1 DAG levels and Gαq signaling pathway regulate habituation Gαq signaling pathway is considered important for synaptic priming and neurotransmitter release in motoreurons (Reynolds et al. 2005, Lackner et al. 1999). The role of Gαq in habituation was assessed by testing mutants that either increased or decreased its function. Both egl-30(ep271) and dgk-1(nu62), mutations that increased function of egl-30 and the Gαq signaling pathway resulted in slow habituation rate and high asymptotic level (for reversal distance and probability). Taken together, these results suggest that slow rate and high asymptotic level for reversal distance are linked to increased DAG levels. Interestingly, mutations in egl-10(n692), a gene which should increase Gαq activity, resulted in an increased rate of habituation and low asymptotic level for reversal probability. Unlike egl-30, egl-10 is predominately expressed in neuronal process at chemical synapses (as 64  opposed to the cell body) (Koelle et al. 1996). It is possible that increases in DAG levels at specific compartments in the neuron account for the fast rate of habituation in the mutant; perhaps general increases in DAG in a neuron (as would be seen in egl-30 gain-of-function mutations) effects protein function at both the synapse and neuronal cell body to change the habituation kinetics whereas egl-10 only exerts its effect in neuronal processes, particularly at chemical synapses. Since chemical synapses between the mechanosensory neuron and interneurons are hypothesized to be inhibitory (Chalfie et al., 1985), an intriguing possibility is that increases in inhibitory synapses within the tap-withdrawal circuit result in a fast habituation rate. Since egl-10 is expressed in almost all neurons, to test this theory, one would have to identify where egl-10 is acting to modulate habituation through cell specific rescue. Next, the myristoylation consensus sequence (myr) can be added to a dominant negative (dn) version of dgk-1 which will result in a targeted increases in DAG at the synapse. For example, one could inject a pmec4::myr::dgk-1construct into worms with a wild-type background. The addition of the myristoylation consensus sequence would result in a lipid to be covalently attached to dgk-1 and would anchor dgk-1(dn.) protein. The mec-4 promoter would cause this construct to be expressed in mechanosensory cells. I hypothesize that these transgenic worms should show an increased rate in habituation for reversal probability similar to that seen in egl-10 mutants. Although it is tempting to conclude that decreased DAG levels would result in an increased rate of reversal probability habituation and a decreased asymptotic level, the results for the habituation kinetics of the egl-30 reduction-of-function mutants tested here were confounded by locomotion defects.  65  3.7.2 Gαi interacts with Gαq to regulate habituation rate an asymptotic level In vertebrate systems, Gαi primarily inhibits the Gαs signaling pathway (opposed to Gαq) (Oldman, 2008). In C.elegans, Gαi has been shown to interact with Gαq in terms of acetylcholine release and, as found with in this study, habituation. Gain-of-function alleles of egl-30 phenocopy loss-of-function mutations of goa-1, eat-16 and dgk-1 for both reversal distance and reversal probability. The severity of these mutations (which effect at least four different characteristics of habituation) suggests that these pathways are major regulators of tapinduced responses, however, given the widespread expression of members of goa-1 and egl-30 in neurons, one cannot rule out the possibility that the observed defects in habituation in these mutants are influenced by pleiotropic effects.  3.7.3 Gαs specifically regulates habituation rate The Gαs signaling pathway results in the production of cAMP, a second messenger which in turn activates cAMP-dependent proteins such as transcription factors and kinases (Hobson, 2003). Unlike Gαi and Gαq, mutants in the Gαs signaling pathway only affect reversal rate rather than asymptotic level. In terms of reversal distance, many gain-of-function mutants in the Gαs signaling pathway resulted a fast rate of habituation coupled with a normal asymptotic level and initial response. These data suggest that gsa-1 acts specifically to alter rate of habituation of reversal distance. In Reynolds (2005) locomotion studies, epistatic analysis between the Gαq and Gαs signaling pathway indicated that Gαs acts predominately through the Gαq signaling pathway meaning Gαs largely depends on Gαq to exert its effect. This led to the hypothesis that Gαq acts as the main synaptic priming pathway with respect to acetylcholine release, whereas Gαs acts synergistically with Gαs to strengthen certain synapses and allow for coordinated movement 66  (Reynolds 2005). Given its similarity in phenotype to egl-10 (a negative regulator of Gαi) with respect to reversal probability, it is possible that the primary role of Gαs in habituation is to enhance Gαq activity at defined synapses to control habituation rate for reversal probability. Interestingly, acy-1 mutants did not completely phenocopy gsa-1 mutants; acy-1 exhibited a significantly high initial response whereas gsa-1 did not. The initial response is equivalent to the naïve response to tap and gives insight into the homeostatic maintenance of neuronal function and/or sensitivity to tap stimuli. A 2012 paper by Timbers et al. investigated the role of tap intensity to reversal magnitude and probability (Timbers 2012). While the asymptotic level of reversal distance was not affected by increased tap intensity, the initial response magnitude in 72h old animals was larger at the most intense stimuli compared to less intense stimuli. The high initial response in acy-1 mutants could be due to a hyper-sensitivity to tap. However, if this were the case, one would not expect acy-1 mutant to also show a fast rate of habituation for reversal probability. Alternatively, the high initial response and fast habituation rate could be due to homeostatic changes at the neuron. Perhaps acy-1 animals maintain a higher cell-excitability in neurons compared to wild-type worms. To test this hypothesis, cameleon imaging of calcium currents can be performed in the mechanosenory neurons to determine whether cell excitability differs from controls.  3.7.4 PKA regulates some aspects of habituation As described, Gαs activation results in the production of cAMP. One major target of the Gαs signaling pathway is PKA (reviewed in Sunahara et al., 1996; Walsh and Van Patten, 1994). Surprisingly, mutations in kin-2 (a negative regulator of kin-1/PKA) did not result in habituation defects for reversal distance. This, however, may be due to the fact that only heterozygotes could be tested. kin-2(ce179) did result in a fast habituation phenotype for reversal probability. 67  Together, this suggests that habituation rate for reversal probability is at least partially regulated by a PKA-dependent mechanism. The results for reversal distance are more difficult to interpret. While the lack of phenotype in kin-2 suggests that reversal distance is regulated by a PKAindependent pathway, it is also possible that pleiotropic effects are masking its habituation phenotype. In order to be sure, one would have to confirm the kin-2 habituation phenotype using RNAi or by generating additional kin-2 mutants and testing them. Additionally, PKA antangonists and agonists could be used to alter PKA function using pharmaceuticals.  3.8 Experiment 2b: Non-reversal responses elicited by tap in Heterotrimeric G-protein mutants In experiment 1, analysis of N2 behaviour after taps demonstrated that worms shift their behaviour in response to repeated tap stimuli, showing a decrease in reversals and an increase in decelerations and pauses over the course of habituation. Since each family of heterotrimeric Gprotein signaling pathways had different profiles for reversal behaviour, the other response-types for these strains were analyzed to determine how their behaviour shifted over habituation training. In this analysis, the data from the habituation tests described in Experiment 2a were analyzed with the methods described in Experiment 1. Genes from each heterotrimeric Gprotein signaling pathways were examined for potential shifts in non-reversal tap responses.  3.8.1 Non-reversal tap responses in Gαq In order to develop a more complete picture of the worms’ tap-induced behaviour, the proportion of tap-induced accelerations, decelerations, pauses and “no responses” were measured for each allele of egl-30. First, the responses of 80-96 hour old animals were compared with wild-type for differences in behavioural shifts. The most striking difference between egl-30 mutants compared to controls was the number of pauses and decelerations performed in response 68  to tap. All alleles of egl-30 except for egl-30(ad806) executed a much lower proportion of pauses compared to controls (p-values reported in Table B.2 ) (Figure 3.8.1.1). egl-30(ad806) exhibited a particularly fast rate of habituation of reversal probability, low asymptotic level for reversal probability and normal rate and asymptotic level for pause frequency. This allele, however, also showed a high-proportion for reversal distance, it is possible that locomotion or sensory defects are affecting its habituation profile. Recall that egl-30 mutants are known to develop more slowly than N2 worms and that age can affect habituation rate. Therefore, the same analysis was performed on worms 120h old. These results showed that all alleles of the 120h egl-30 mutants tested exhibited a lower proportion of pauses than wild-type. At 120h, egl-30(js126) and egl-30(n686) showed an increase in habituation rate compared to 96h and this corresponded with an increase in decelerations and “no responses”, respectively. egl-30(n715) did not show a difference in habituation rate between 96h and 120h but did show a slightly higher proportion of accelerations at 120h (Figure 3.8.1.2). Together, this demonstrates that regardless of age, egl-30 mutants shift their behaviour from pauses to reversals, which may correspond to a weaker reverse sub-circuit. This would be consistent with Kindt et al.’s finding that egl-30 mutants exhibited decreased cell excitability in ALM Deeper habituation found in older mutants of egl-30 corresponded with an increase in a different type of behaviour, however, the behaviour that increased was variable between alleles (either decelerations or no responses).  69  Figure 3.8.1.1 Behavioural profile of members of the Gαq signaling pathway at 96 hours. Accelerations (top left); pauses (top right); decelerations (bottom left) and no responses (bottom right). Dashed lines represent reversal responses. Solid lines represent the non-reversal behaviour for that graph.  70  Figure 3.8.1.2 Behavioural profile of members of the Gαq signaling pathway at 120h. Accelerations (top left); pauses (top right); decelerations (bottom left) and no responses (bottom right). Dashed lines represent reversal responses. Solid lines represent the non-reversal behaviour for that graph.  3.8.2 Non-reversal tap responses in Gαi The Gαi and Gαq signaling pathway seem to be closely linked with respect to habituation phenotype; gain-of-function alleles of egl-30 phenocopy loss-of-function alleles of goa-1 for habituation of reversal probability. To determine whether this pattern held for all tap responses,  71  the proportion of accelerations, decelerations, pauses and “no responses” were measured to determine the overall patterns of behavioural responses for each of the mutant strains. The proportion of accelerations, decelerations and pauses for four alleles of goa-1 and two alleles of eat-16 and 1 allele of dgk-1 were measured. All alleles tested showed a smaller proportion of pauses and decelerations (p-values reported in Table B.2). The results for accelerations were variable and appear to depend on the severity of the reversal habituation phenotype. Gαi signaling pathway mutants show a much lower tendency to decelerate or pause, suggesting that Gαi signaling pathway mutants shift their behaviour from pauses and decelerations to reversals (p-values reported in table B.2 ). Any decrement in the proportion of reversal during habituation training corresponds to an increase in accelerations (Figure 3.8.2). There was one strain, goa-1(n3055) that did not completely phenocopy the defects found in the other Gαi signaling mutants. goa-1(n3055), which didn’t show a deficit in the rate of habituation for reversal probability, has previously been characterized as a weak allele of goa-1 which helps to explain why its habituation profile is different from the other alleles of goa-1. Interestingly, the deep habituation phenotype of goa-1(n3055) is coupled by a rapid increase in accelerations (p<0.001 for acceleration asymptotic level) (rather than pauses or decelerations) suggesting that, while the reversal behaviour differs, its shift in behaviour (from reversals to accelerations) is consistent with the other members of the Gαi signaling pathway. Overall, Gαi mutants show a consistent shift from pauses and decelerations to reversals. During habituation training, any decrement in reversals corresponded to an increment in accelerations. This demonstrates that mutations in the Gαi pathway biases the tap-withdrawal circuit towards reversals and suggests that there is little competition from the forward sub-circuit.  72  Figure 3.8.2 Behavioural profile of members of the Gαi signaling pathway. Accelerations (top left); pauses (top right); decelerations (bottom left) and no responses (bottom right). Dashed lines represent reversal responses. Solid lines represent the non-reveral behaviour for that graph.  73  3.8.3 Non-reversal tap responses in Gαs To gain better insight into changes Gαs mutants exhibit in response to tap, accelerations, decelerations, pauses, and “no responses” were measured. Two gain-of-function alleles of gsa-1 along with two gain-of-function alleles of acy-1 and one reduction of function allele of acy-1 were analyzed. All mutations that resulted in an increase in Gαs signaling showed a higher proportion of accelerations coupled with a lower proportion of pauses and decelerations (pvalues reported in table B.2 )(Figure 3.8.3). Conversely, acy-1(nu329) (the reduction of function allele of acy-1) showed a higher proportions of reversals coupled with fewer accelerations and decelerations (p<0.001 for acceleration rate, p<0.001 for acceleration asymptote; p<0.001 for deceleration rate and p=0.004 for deceleration asymptotic level). Together, this demonstrates that the shift in behaviour from reversals to accelerations and vice versa is linked to Gαs signaling and cAMP levels. Additionally, the shift in behaviour from reversals to accelerations implies that there is a greater level of competition from the forward sub-circuit compared to controls in mutants that increased cAMP levels.  74  Figure 3.8.3 Behavioural profile of members of the Gαs signaling pathway. Accelerations (top left); pauses (top right); decelerations (bottom left) and no responses (bottom right). Dashed lines represent reversal responses. Solid lines represent the non-reveral behaviour for that graph.  3.8.4 Strains predicted to interact in heterotrimeric G-protein signaling Eleven strains representing ten genes were tested to determine whether their phenotype recapitulated the general phenotypes from the Gαi, Gαq and Gαs analyses (Table B.2 in appendix). These genes were selected because they had putative genetic or predicted interactions 75  with either goa-1, egl-30 or gsa-1. Two genes that showed interesting defects in habituation will be described. These were egl-3 and unc-103. C.elegans unc-31 is the orthlogue of human CAPS (calcim-dependent activator of protein secretion) . A 2006 study determined that unc-31 (required for neuropeptide release among other things) acts upstream of the Gαs signaling pathway and is required for Gαs pathway activation (Charlie et al. 2006). In order to identify whether neuropeptides contributed to the regulation of habituation, egl-3 was tested. egl-3 is a proprotein convertase that cleaves inactive propeptides into neuroactive peptides that are released from dense core vesicles. If neuropeptides released from dense-core vesicles activate Gαs signaling during habituation training, one would expect that mutation of egl-3 would have the opposite phenotype to gsa-1 gain-of-function alleles. One allele of egl-3 was tested: egl-3(n150), a missense mutation in exon 2. Surprisingly, egl-3(n150) exhibited only mild defects in habituation (for distance), however, egl-3 mutants did show a severe reduction in habituation rate (probability) and a higher asymptotic level for probability. The proportion of non-reversal behaviours were tested to see if they recapitulate that of the acy-1 reduction-of-function allele. Mutations in egl-3 did not show an increase in accelerations. Instead, all behaviours were shifted to reversal suggesting neuropeptides have a larger role in habituation (either directly or indirectly) than the Gαs signaling pathway (Figure 3.8.4, left). Another potential downstream effector of heterotrimeric G-protein signalling, unc-103, was identified; unc-103 encodes an outwardly rectifying potassium channel that is expressed in muscle and neuronal cells including the command interneurons AVA, AVD and AVE (Garcia et al. 2003, Gruninger et al. 2006). unc-103(n500n1211) is a revertant mutant that is phenotypically  76  wild-type with respect to locomotion, however, it has severe defects in habituation. unc103(n500n1211) has a normal initial response (for reversal probability) but the rate of habituation is extremely slow and shallow. Reversal distance habituation has a more mild phenotype, however, the rate of habituation is slow compared to controls and the asymptotic level is relatively high. Non-reversal responses to tap were analyzed to see whether unc-103 exhibited the same shift in behaviour as goa-1. This analysis showed that all non-reversal behaviours were shifted in favour of reversals suggesting that there was little (if any) competition from the forward subcircuit (Figure 3.8.4, right).  Figure 3.8.4 Behavioural profile of egl-3 (left) and unc-103 (right).  2.8.5 Caveats of non-reversal behavioural measurements The sum of accelerations, decelerations, pauses, reversals and “no responses” should account for approximately 100% of the behaviour analyzed in the worms, however, in some mutants the sum of all behaviour measured was much lower than expected. This was particularly apparent in worms that had severe locomotion phenotypes. Reversal probability is found by finding the number of worms that reversed (showed a bias value of -1) within a 1 second period 77  after tap. In N2 and heterotrimeric G-protein mutants, this measures agreed the proportion of worms showing an average negative speed between 0-1 second after tap. In mutants with extreme defects in locomotion rate, the reversal probability measured based on average speed differs from that based on whether there was a negative bias at any point within a 1 second window after tap. In this scenario, measurements of non-reversal behaviours become unreliable. Therefore, only mutants whose sum of behaviours was 100 + 10% were considered. One strain out of approximately 30 was rejected in this study.  3.9 Discussion 3.9.1 Gαq and Gαi shift their behaviour from pauses to reversals Analyzing the non-reversal behaviours demonstrated that mutations disrupting Gαi, Gαq or Gαs result in a characteristic shifts in behaviour. Analysis of habituation of reversal probability demonstrated that Gαq gain-of-function mutations phenocopied Gαi loss-of-function mutations suggesting that Gαq and Gαi share a close relationship in the regulation of reversal rate and asymptotic level. Corroborating this association, both goa-1(loss-of-function) and egl-30 (gain-of-function) mutants tend to shift their behaviour from pauses to reversals. The pattern of shifting behaviour from pauses to reversals was consistent between 96h and 120h old egl-30 mutants. All alleles of egl-30 showed a significantly low level of pauses. Unlike the egl-30 gain-of-function mutant, however, Gαi signaling mutants also showed a much lower proportion of decelerations. A possible explanation is that the Gαi signaling pathway suppresses Gαs or another unidentified signaling cascade in addition to Gαq, which results in a more severe phenotype. Epistatic analysis has shown that, with respect to locomotion, the Gαs signaling pathway feeds into the Gαq pathway downstream of DAG production and hyperactivation of the Gαq signaling pathway is capable of restoring some locomotion defects 78  found in Gαs mutants (Reynolds et al. 2005). Since hyperactivating the Gαq signaling pathway via the use of an egl-30 gain-of-function mutation was not sufficient to cause a phenotype in the proportion of decelerations, this suggests that Gαi acts on another signaling cascade in addition to Gαq. However, since only one egl-30 gain-of-function mutant was tested, one cannot rule out the possibility that a stronger gain-of-function mutant (if it exists) would phenocopy goa-1 lossof-function strains. Regardless, both the Gαi and Gαq bias behaviour towards reversals opposed to other stimuli-induced behaviours suggesting that the forward sub-circuit is relatively weak compared to the reverse sub-circuit.  3.9.2 Gαs mutants shift behaviour from reversals to accelerations Gαs gain-of-function mutants exhibited a fast habituation rate for reversal probability. Coupled with this were increases in accelerations and decreases in pauses and decelerations. One reduction-of-function allele was tested and showed decreased accelerations and decelerations. These results demonstrate that the Gαs signaling pathway results in a shift from reversals and pauses to accelerations. Gαs has been proposed to facilitate locomotion by selectively strengthening neurotransmitter release at specific synapses (Reynolds et al. 2005). egl-30 gain-of-function mutants and Gαs gain-of-function mutants both show a decrement in pauses and decelerations, however, Gαs mutants also show an increase in accelerations. Assuming the Gαs signaling pathway is acting somewhere in the tap-withdrawal circuit to effect habituation, hyperactivating the Gαs signaling pathway might increase neurotransmitter release in the forward sub-circuit. Since gsa-1 and acy-1 are widely expressed in neurons, further research to determine where Gαs acts to regulate habituation is required. 79  3.9.3 Heterotrimeric G-protein signaling pathways Disruptions of Gαq, Gαi or Gαs signaling pathways result in shifts in behaviour from (or to) reversals to another behavioural type. When Gαs signaling is hyperactivated, the decrement in reversal probability directly corresponds to the increment of accelerations. Similarly, reducing Gαi signaling resulted in a shift away from pauses and reversals with a relatively small effect on accelerations. Mutations in heterotrimeric G-protein signaling may cause changes in the relative strength of each sub-circuit and bias the worm towards one non-reversal behavioural type over another.  3.9.4 Genes have a similar phenotype to members of the heterotrimeric Gprotein signaling pathway Ten mutant genes were investigated as potential upstream or downstream effectors of Gprotein signaling (see appendix, Table A.1 for description and Table B.3 for p-values of all behavioural types). Two candidate genes showed striking habituation defects. These were egl-3 and unc-103. egl-3 is a proprotein convertase that cleaves proproteins into bioactive neuropeptides. Previous studies have shown that Gαs pathway activation relies on unc-31, a gene required secretion of neuropeptides (along with ion channels and other proteins) (Charlie et al. 2006). Since bioactive neuropeptides are generated by egl-3-dependent cleavages of proproteins, loss-of-function mutations in egl-3 should exhibit similar phenotypes as gsa-1 or acy-1 reduction-of-function mutations. Indeed, egl-3(n150) showed a slow habituation rate and high asymptotic level for both reversal probability and distance (p<0.001 and p<0.001). Although defects in egl-3 suggest that neuropeptides are involved in modulating habituation, this does not give information about which neuropeptide (or neuropeptides) act as modulators of habituation or if they act directly or indirectly. Of the hundreds of g-protein coupled receptors, 50 receptors  80  are likely to bind to neuropeptides (Bargmann, 1998). Epistatic experiments with egl-3; gsa1(gain-of-function) double mutants should help to determine whether these genes are acting in the same pathway with respect to habituation. Additionally, testing mutant versions of these genes may give insight into the neuropeptide(s) responsible for modulating Gαs signaling in response to habituation training. Next, unc-103 is a potential downstream effector of goa-1 as unc-103 and goa-1 were shown to interact genetically with respect to locomotion (Robatzek 2000). unc-103 worms consistently responded to tap with a reversal rather than any other behaviour (habituation probability) (p<0.001 for habituation rate and p<0.001 for asymptotic level). Additionally, reversal distance habituation was relatively slow and had a high asymptotic level. Electrophysiology studies show that expression of the analogous mutation of allele of unc103(n500) (an unc-103 gain-of-function mutation) in Xenopus ooycytes resulted in a hyperpolarizing shift in half-activation potential (Petersen et al. 2004). This would suggest that the unc-103(n500) mutant would hyperpolarize the neurons it is expressed in and result in reduced cell excitability and is consistent with the behavioural defects of this allele (flaccid paralysis and reduced pharyngeal pumping) (Petersen et al. 2004). The unc-103(n500n1211) revertant mutant that encodes a gain-of-function and an early stop mutation (likely resulting in a truncated or null protein) exhibited normal locomotion (speed) compare to wild-type, however, showed a large increase in the proportion of reversals coupled with an increase in reversal distance compared to controls (p<0.001 for distance habituation rate and p<0.001 for asymptotic level). Since loss-of-function mutation in resulted in hyper-excitability of muscle cells, perhaps hyper-excitability of the command interneurons AVA, AVD and AVE result in increased reversal probability and distance. This would indicate that changes in excitability at the  81  command interneuron level play a role in shifting behaviours. To test this, perhaps cell excitability of AVB can be manipulated by expressing unc-103(n500) mutation in AVB (the n500 allele is semi-dominant) or by rescuing a loss-of-function mutation in AVA, AVD, and AVE. Alternatively it might affect the ALM sensory neurons and not the PLM sensory neurons, strengthening the input from the head circuit. The same type of experiments could be used to test this hypothesis. Further studies would be needed to determine whether unc-103 acts downstream of goa-1 with respect to habituation. To test this, the unc-103 gain-of-function mutation should be tested for habituation defects and, if this allele exhibits an opposite phenotype to goa-1 loss-of-function, epistatic analysis can be used to determine if they interact genetically in the regulation of habituation.  4 Conclusion An organism’s ability to understand its environment and make changes its behaviour in response to it is a fundamental requirement for survival. Habituation is a phenomenon that has been found in every organism tested for it suggesting that it is an important tool that allows animals to understand what they should ignore and what it should pay attention to. Defects in habituation can be indicative of aberrant neuronal function. Habituation defects are associated with a number of neurological disorders including Schizophrenia, Autism, Parkinson’s disease, Tourette’s Syndrome etc. The overarching aim of this study was to gain a better understanding of habituation of the C.elegans startle response at the behavioural, circuit and molecular level. This goal was tackled in three ways: first, non-reversal behaviours were characterized along with reversals in wild-type worms. This analysis showed that both the magnitude of reversals and accelerations decrease over repeated tap suggesting that both are regulated by the same set of neurons. Conversely, the 82  as reversal probability decreased, the probability of accelerations, decelerations and pauses increased suggesting that habituation for reversal probability is superimposed on facilitation of accelerations, pauses and decelerations. Additionally, worms habituated off of food showed an increase in decelerations (but not accelerations or pauses). Using this analysis, one can now start to understand how relative strengths of the two sub-circuits can regulate the type of behavioural response a worm performs after a tap. The proportion of decelerations was increased under conditions that caused decreases in ALM cell excitability (without changes in PLM cell excitability) meaning relatively small changes to the circuit caused the nematode to slow down after a tap. If reversals, decelerations, no-responses, pauses and accelerations represent a continuum of behaviour in response to tap habituation, perhaps the shift in these behaviours during habituation training are indicative of the relative competition within the tap circuit. To test this hypothesis, more information about the relative strength of the sub-circuits during habituation is needed. This can be collected either measuring the relative strength using Ca 2+ imaging techniques like G-cAMP and/or by studying the shift in worms with ablations in ALM or PLM. If decelerations are caused by the downregulation of the reversal sub-circuit, then ablations in ALM should result in a shift from reversals to decelerations. Similarily, worms with PLM ablation should shift their behaviour from accelerations to pauses. Next, the habituation kinetics of three families of heterotrimeric G-protein signaling pathways was investigated for their role in habituation. Generally speaking, mutations that resulted in the hyperactivity of Gαq (egl-30 gain-of-function mutations and goa-1 loss-offunction mutations) resulted in a slow rate in habituation and high asymptotic level for both reversal distance and probability. Mutations that increased Gαs pathway activity resulted in fast habituation. This demonstrated that the Gαi signaling pathway primarily inhibits the Gαq  83  signaling pathway and that Gαs plays a more minor role in habituation which might involve enhancing specific synapses. The Gαi and Gαq signaling pathway either direct or indirectly control the levels of the second messenger DAG. Since all mutations that result in increased DAG exhibited slow habituation with high asymptotic level, Gαq-dependent neurotransmitter release in conjunction with DAG signaling may regulate multiple aspects of the habituation curve. To investigate this further, one could test the habituation curve of nematodes treated with endogenous DAG analogues. This should directly increase the levels of DAG and have same impact on habituation kinetics as mutants that upregulate Gαq or downregulate Gαi. Mutants that increase Gαs signaling result in increased neurotransmitter release, increased cAMP and a fast habituation rate. This suggests that the Gαs has a more specific role in regulating habituation and that habituation kinetics are altered with changes in cAMP. Finally, heterotrimeric G-proteins were tested to determine how their behaviour shifts from reversal probability and whether it differs from wild-type animals. While it was difficult to interpret the results from mutants that down-regulated the Gαq signaling pathway, mutants that upreguated Gαq resulted in a shift from pauses to reversals. This might suggest an increase in input from the reverse sub-circuit or decreased input from the forward sub-circuit. Behavioural shifts in Gαi exhibited decreased decelerations and pauses in favor of reversals. This demonstrates that goa-1 and the Gαi signaling pathway play another role in regulating habituation aside from inhibiting Gαq. This role may be direct (by inhibiting Gαs, for example) or indirect (by regulating the development of neurons important for locomotion and habituation). Finally, mutants that upregulated the Gαs signaling pathway showed a shift from reversals to accelerations and may be the result of a rapid weakening in reversal sub-circuit strength in conjunction with the strengthening of the forward sub-circuit.  84  4.1 Contributions to research on non-associative learning From behavioural studies of the wild-type worm, we’ve learned how important it is to consider more than one behavioural response when studying the habituation. In characterizing the habituation pattern of the worm’s startle response, I found that worm’s rarely ignore the tapstimuli completely (as measured by the proportion of “no responses”) and instead shift their behaviour. This shows that worms remain much more responsive to tap than what’s been reported in reversal probability habituation data. In studying the heterotrimeric G-protein signaling pathway, I found that Gαi and Gαq are important for regulating all aspects of habituation whereas Gαs specifically regulates the rate of habituation. Overall, I was able to characterize how each family of heterotrimeric G-protein signaling affects habituation and gain insight into the mechanism in which it is regulated. Finally, by studying the behavioural shift that occurs during habituation training, I was able to show that alterations in heterotrimeric G-protein signaling biases how a worm will react to tap. Heterotrimeric G-protein mutants generally shift their responses away from more moderate behaviours (like pauses and decelerations) to extreme responses (acceleration and reversals). This shows that the tap-withdrawal circuit has modulated in a way to favor either the forward sub-circuit or the reverse sub-circuit and gives insight into how habituation is regulated at the circuit-level.  4.2 Potential genes acting upstream or downstream of heterotrimeric Gprotein signaling pathways Heterotrimeric G-protein signaling pathways are generally well-defined until the point where the second messenger is generated. Second messengers such as cAMP, IP3 and DAG are signaling amplifiers and bind of downstream effectors. One challenge is to determine which  85  proteins the second messengers act on to regulate habituation. Genes that are predicted to interact with at least one family of heterotrimeric G-protein signaling pathway were tested for habituation defect. In particular, genes that alter cell-excitability or modulate neurotransmitter release such as potassium and ion channels were tested. unc-103 encodes an either-a-go-go related (ERK) channel which is expressed in AVA, AVD and AVE, command interneurons that are post-synaptic to ALM and/or PLM (Garcia et al. 2003, Gruninger et al. 2006). unc-103 has been shown to interact with goa-1 with respect to locomotion (Garcia 2003). Gain-of-function mutations in unc-103 result in hyper-polarization of xenopus ooycytes and may result in hyperpolarization of the neurons its expressed in (Petersen et al. 2004). Conversely, unc103(n500n1211) allele represents a loss or reduction of function allele and should result in hyper-excitation of the neurons its expressed in. unc-103(n500n1211) showed a striking shift from non-reversal behaviour to reversals suggesting that unc-103, and possibly the command interneurons, play a part in regulating behavioural shifts.  4.3 Future directions A major goal of this study is to gain a better understanding of the mechanisms regulating habituation. Heterotrimeric G-proteins were tested for habituation defects because they’ve been shown to play a role in habituation and learning and memory (Kindt et al., 2007). One major caveat is that g-proteins are widespread and it is hard to differentiate between habituation defects and general locomotion defects and developmental effects. Analysis of worm spontaneous behaviour and morphology can suggestive of a defect; however, if the mechanisms of habituation are to be defined, information on where each gene is acting to regulate habituation is required. This can be done through cell-specific rescue experiments.  86  Next, to understand the cellular changes underlie habituation and behavioural shifts, information about the relative strength of each sub-circuit can be measured using cameleon imaging. Manipulations into the relative strength of each sub-circuit can be paired with cellexcitability data to determine what cellular changes are needed to elicit a specific behavioural response. Finally, although two candidate downstream effectors of heterotrimeric G-protein signaling were identified, further studies are needed to confirm that these genes in fact act downstream of heterotrimeric G-protein signaling and that their misregulation results in the habituation phenotype. Localization experiments coupled with cell-specific rescue and epistatic analysis should shed light in whether unc-103 and neuropeptide signaling do in fact directly regulate habituation through heterotrimeric G-protein signaling pathways.  87  Bibliography Bailey, C., Chen, M. "Morphological basis of short-term habituation in Aplysia." The Journal of neuroscience : the official journal of the Society for Neuroscience, 1988: 2452-2459. Bargmann, C. "Neurobiology of the Caenorhabditis elegans genome." Science, 1998: 2028– 2033. Bargmann, C., Avery, L. "Laser Killing of Cells in Caenorhabditis elegans." Methods in Cell Biology, 1995. Bastiani, C. and Mendel, J. "Heterotrimeric G proteins in C. elegans." Wormbook. October 13, 2006. http://www.wormbook.org (accessed 2013). Brenner, S. "The genetics of Caenorhabditis elegans." Genetics, 1974: 71-94. Brockie, P., Mellem, J., Hills, T., Madsen, D., Maricq, A. "The C. elegans glutamate receptor subunit NMR-1 is required for slow NMDA-activated currents that regulate reversal frequency during locomotion." Neuron, 2001: 617-630. Brundage, L., Avery, L., Katz, A., Kim, U.J., Mendel, J.E., Sternberg, P.W., and Simon, M.I. "Mutations in a C. elegans Gqα gene disrupt movement, egg laying, and viability." Neuron, 1996: 999-1009. Chalfie, M., Sulston, J., White ,J.,Southgate, E. "The Neural Circuit for Touch Sensitivity in Caenorhabditis elegans." The Journal of Neuroscience, 1985: 956-964. Charlie, N., Schade, M., Thomure, A., Miller, K. "Presynaptic UNC-31 (CAPS) is required to activate the G alpha(s) pathway of the Caenorhabditis elegans synaptic signaling network." Genetics, 2006: 943–961. Chase, D.L., Patikoglou, G.A., and Koelle, M.R. "Two RGS proteins that inhibit Gαo and Gαq signaling in C. elegans neurons require a Gβ5-like subunit for function." Current Biology, 2001: 222-231. Cuppen, E., van der Linden, A.M., Jansen, G., and Plasterk, R.H. "Proteins interacting with Caenorhabditis elegans Gα subunits." Comparative and Functional Genomics, 2003: 479-491. Garcia, L., & Sternberg, P. "Caenorhabditis elegans UNC-103 ERG-like potassium channel regulates contractile behaviors of sex muscles in males before and during mating." Journal of Neuroscience, 2003: 2696-705. Geng, W., Cosman P., Berry, C., Feng, Z., Schafer, W. "Automatic tracking, feature extraction and classification of C. elegans phenotypes." Biomedical Engineering, 2004: 3716-3719. Gover, T., Jiang, X., Abrams, T. " Persistent, exocytosis-independent silencing of release sites." The Journal of neuroscience: the official journal of the Society for Neuroscience, 2002: 1942-1955. Giles, A. "Candidate gene and high throughput genetic analysis of habituation in Caenorhabditis elegan." Ph.D. Diss. of University of British Columbia. Dissertation. December, 2011. Giles, A. (in preparation) Groves P., Thompson R. "Habituation: A dual-process theory." Psychological Review, 1970: 419-450. Gruninger, T., Gualberto, D., LeBoeuf, B., Garcia, L. "Integration of male mating and feeding behaviors in Caenorhabditis elegans." Journal of Neuroscience, 2006: 169-179. Hajdu-Cronin, Y., Chen, W., Patikoglou, G., Koelle, M., Sternberg, P. "Antagonism between Goα and Gqα in Caenorhabditis elegans: the RGS protein EAT-16 is necessary for Goα signaling and regulates Gqα activity." Genes Dev., 1999: 1780-1793.  88  Harris, T. Wormbase. 2003. http://www.wormbase.org/db/get?name=WBGene00001196;class=Gene (accessed 2013). Hart, A., Sims, S., Kaplan, J. "Synaptic code for sensory modalities revealed by C. elegans GLR1 glutamate receptor." Nature, 1995. Hobson R., Geng J., Gray A., Komuniecki R. "SER-7b, a constitutively active Gαs coupled 5HT7-like receptor expressed in the Caenorhabditis elegans M4 pharyngeal motorneuron." J. Neurochem, 2003: 414–419. Horvitz, H., Chalfie, M., Trent, C., Sulston, J., Evans, P. "Serotonin and octopamine in the nematode Caenorhabditis elegans." Science, 1982: 1012-1014. Hou, D., K. Suzuki, W. J. Wolfgang, C. Clay, M. "Presynaptic impairment of synaptic transmission in Drosophila embryos lacking Gsα." Journal of Neuroscience, 2003: 5897– 5905. Huang, K., Cosman, P., Schafer, W. "Automated tracking of multiple C. elegans with articulated models." Biomedical Imaging, 2007: 1240-1243. Jansen, G., Thijssen, K.L., Werner, P., van der Horst, M., Hazendonk, E., and Plasterk, R.H. " The complete family of genes encoding G proteins of Caenorhabditis elegans." Nature Genetics, 1999: 414-419. Jiang, H., Wu, D., and Simon, M.I. "Activation of phospholipase Cβ4 by heterotrimeric GTPbinding proteins." Journal of Biological Chemistry, 1994: 7593–7596. Kawano T., Po, M., Gao, S., Leung, C., Ryu, W. Zhen, M. "An Imbalancing Act: Gap Junctions Reduce the Backward Motor Circuit Activity to Bias C. elegans for Forward Locomotion." Neuron, 2011: 1010-1015. Kindt, K., Quast, K., Giles, A, De, S., Hendrey, D., Nicastro, I., Rankin, C., Schafer, W. "Dopamine mediates context-dependent modulation of sensory plasticity in C. elegans." Neuron, 2007: 662-667. Koelle, M.R., and Horvitz, H.R. "EGL-10 regulates G protein signaling in the C. elegans nervous system and shares a conserved domain with many mammalian proteins." Cell, 1996: 115125. Korswagen, H., Park, J., Ohshima, Y., Plasterk, R. " An activating mutation in a Caenorhabditis elegans Gs protein induces neural degeneration." Genes and Development, 1997: 1493– 1503. Lackner, M.R., Nurrish, S.J., and Kaplan, J.M. "Facilitation of synaptic transmission by EGL-30 Gqα and EGL-8 PLCβ: DAG binding to UNC-13 is required to stimulate acetylcholine release." Neuron, 1999: 335–346. Lee, R., Sawin, E.,Chalfie, M.,Avery, L. ". EAT-4, a homolog of a mammalian sodiumdependent inorganic phosphate cotransporter, is necessary for glutamatergic neurotransmission in Caenorhabditis elegans." The Journal of Neuroscience, 1999: 159167. Li, C., Timbers, T. Rose, J. Bozorgmehr, T., McEwan, A. Rankin, C. "The FMRFamide-related neuropeptide FLP-20 is required in the mechanosensory neurons during memory for massed training in C. elegans." Learning and Memory, 2013: 103-108. Loer, C., Kenyon, C. "Serotonin-deficient mutants and male mating behavior in the nematode Caenorhabditis elegans." The Journal of Neuroscience, 1993: 5407-5417. Maricq, P., Driscoll, M., Bargmann, C. "Bargmann Ci. Mechanosensory signalling in C. elegans mediated by the GLR-1 glutamate receptor." Nature, 1995: 78-81.  89  Mendel, J.E., Korswagen, H.C., Liu, K.S., Hajdu-Cronin, Y.M., Simon, M.I., Plasterk, R.H., and Sternberg, P.W. "Participation of the protein Go in multiple aspects of behavior in C. elegans." 1995: 1652-1655. Miller, K.G., Emerson, M.D., and Rand, J.B. "Goα and diacylglycerol kinase negatively regulate the Gqα pathway in C. elegans." Neuron, 1999: 323-333. Nurrish, S., Ségalat, L., and Kaplan, J.M. "Serotonin inhibition of synaptic transmission: Gαo decreases the abundance of UNC-13 at release sites." Neuron, 1999: 231–242. Oldham, W., Hamm, H. "Heterotrimeric G protein activation by G-protein-coupled receptors." Nature reviews, 2008: 60-71. Petersen, C.,McFarland, T., Stepanovic, S., Yang, P. "In vivo identification of genes that modify ether-a-go-go-related gene activity in Caenorhabditis." PNAS, 2004: 11773–11778. Piggott, B., Liu, J., Feng, Z., Wescott, S. and Xu, X. "The neural circuits and synaptic mechanisms underlying motor initiation in C.elegans." Cell, 2011: 922-933. Ramot, D., Johnson, B., Berry, T., Carnell, L., Goodman, M. "The Parallel Worm Tracker: a platform for measuring average speed and drug-induced paralysis in nematodes." PLoS One, 2008. Rankin, C., Chiba, C., Beck, C. "Caenorhabditis elegans: A new model system for the study of learning and memory." Learning and Memory, 1990: 89-92. Renden R., Broadie K. "Mutation and activation of Gαs similarly alters pre- and postsynaptic mechanisms modulating neurotransmission." Journal of Neurophysiology, 2003: 26202038. Reynolds, N.K., Schade, M.A., and Miller, K. "Convergent, RIC-8 Dependent Gα Signaling Pathways in the C. elegans Synaptic Signaling Network." Genetics, 2005: 651-670. Richmond, J., Davis, W., and Jorgensen, E. " UNC-13 is required for synaptic vesicle fusion in C. elegans." Nature Neuroscience, 1999: 959–964. Richmond, J.E., Weimer, R.M., and Jorgensen, E.M. "An open form of syntaxin bypasses the requirement for UNC-13 in vesicle priming." Nature, 2001: 338–341. Robatzek, M., Thomas, J. "Calcium/calmodulin-dependent protein kinase II regulates Caenorhabditis elegans locomotion in concert with a G(o)/G(q) signaling network." Genetics, 2000: 1069-1082. Sawin, E., Ranganathan, R., Horvitz, R. "C. elegans Locomotory Rate Is Modulated by the Environment through a Dopaminergic Pathway." Neuron, 2000: 619-631. Schade, M.A., Reynolds, N.K., Dollins, C.M., and Miller, K.G. " Mutations that Rescue the Paralysis of C. elegans ric-8 (Synembryn) Mutants Activate the Gαs Pathway and Define a Third Major Branch of the Synaptic Signaling Network." Genetics, 2005: 631-649. Ségalat, L., Elkes, D.A., and Kaplan, J.M. "Modulation of serotonin-controlled behaviors by Go in Caenorhabditis elegans." Science, 1995: 1648-1651. Simonetta, S., Golombek, D. "An automated tracking system for Caenorhabditis elegans locomotor behavior and circadian studies application." Journal of Neuroscience Methods, 2007: 273-280. Sternweis, P.C., and Smrcka, A.V. "G proteins in signal transduction: the regulation of phospholipase C." Ciba Foundation Symposium, 1993: 106-111. Sulston, J. Dew, M., Brenner, S. "Dopaminergic neurons in the nematode Caenorhabditis elegans." The Journal of comparative neurology, 1975: 215-226. Sulston, J., Schierenberg E., White, J., Thompson, J. "The embryonic cell lineage of the nematode Caenorhabditis elegans." Developmental Biology, 1983: 215-226. 90  Sunahara, R, Dessauer, C., and Gilman, A. " Complexity and diversity of mammalian adenylyl cyclases." Annu. Rev. Pharmacol. Toxicol, 1996: 461-480. Swierczek, N., Giles, A., Rankin, C., Kerr, R. "High-throughput behavioral analysis in C. elegans." Nature Methods, 2011: 592-598. "The integration of antago- b0215." n.d. Timbers, .T, Giles, A., Ardiel, E., Kerr, R., Rankin C. "Intensity discrimination deficits cause habituation changes in middle-aged Caenorhabditis elegans." Neurobiology of aging, 2012: 621-631. Trent, C., Tsung, N., and Horvitz, H.R. "Egg-laying defective mutants of the nematode Caenorhabditis elegans." Genetics, 1983: 619-647. Walsh, D.,Van Patten, S. " Multiple pathway signal transduction by the cAMP-dependent protein kinase." Faseb J., 1994: 1227-1236. Wickman, K., and Clapham, D. " G protein regulation of ion channels." Current opinion of Neurobiology, 1995b: 278–285. Wickman, K., Clapham, D. "Ion channel regulation by G proteins." Physiol. Rev., 1995a: 865– 885. Wicks, S., Rankin, C. "Effects of tap withdrawal response habituation on other withdrawal behaviors: The localization of habituation in the nematode Caenorhabditis elegans." Behavioural Neuroscience, 1997: 342. Wicks, S., Rankin, C. "Integration of mechanosensory stimuli in Caenorhabditis elegans." The Journal of Neuroscience, 1995: 2434-2444. Wicks, S., Rankin, C. "The integration of antagonistic reflexes revealed by laser ablation of identified neurons determines habituation kinetics of the Caenorhabditis elegans tap withdrawal response." Journal of comparative physiology, 1996a: 675-685. Wicks, S., Roehrig, C., Rankin, C. "A Dynamic Network Simulation of the Nematode Tap Withdrawal Circuit: Predictions Concerning Synaptic Function Using Behavioral Criteria." The Journal of Neuroscience, 1996b: 675-685. Willson, J., Amliwala, K., Davis, A., Cook, A., Cuttle, M.F., Kriek, N., Hopper, N.A., O'Connor, V., Harder, A., Walker, R.J., and Holden-Dye, L. "Latrotoxin Receptor Signaling Engages the UNC-13-Dependent Vesicle-Priming Pathway in C. elegans." Curr. Biol., 2004: 1374–1379. Wood, W. Introduction to C.elegans Biology . New York, NY: Cold Spring Harbor Laboratory Press, 1988. Yvonne M. Hajdu-Cronin, Wen J. Chen, Georgia Patikoglou, Michael R. Koelle, and Paul W. Sternberg. "Antagonism between Goα and Gqα in Caenorhabditis elegans: the RGS protein EAT-16 is necessary for Goα signaling and regulates Gqα activity." Genes and Development, 1999: 1780-1793.  91  Appendices Appendix A: Strains tested for habituation defects Table A.1: Stains tested  Name (allele)  Strain  Mutation type  Function  egl-30 (ep271) egl-30 (n715) egl-30 (js126) egl-30 (ad806) egl-30 (n686) egl-8 (e2917) egl-8 (n488) egl-8 (ok934) egl-10 (n692) egl-10 (md176) goa-1 (sa734) goa-1 (sa734) goa-1 (n1134) goa-1 (n3055) eat-16 (sa609) eat-16 (ce71) dgk-1 (nu62) gsa-1 (ce81)  CE1047 MT1520 NM1380 DA1084 MT1434 CB6616 MT1083 RB1012 MT1443 MT8504 DG1856 JT734 MT2426 MT8628 JT609 KG571 KP1097 KG421  α subunit, Gαq α subunit, Gαq α subunit, Gαq α subunit, Gαq α subunit, Gαq Lipase Lipase Lipase RGS RGS α subunit, Gαi α subunit, Gαi α subunit, Gαi α subunit, Gαi regulator (rgs) regulator (rgs)  gsa-1 (ce94) acy-1 (nu329) acy-1 (ce2) acy-1 (md1756) acy-4 pde-4 (ce268) pde-4 (ok1290) kin-2 twk-18(cn110) egl-3 (n150)  KG524 KP1182 KG518 KG522 VC1331 KG744 RB1231 KG532 TN110 MT150  Missense M244I Other (gf) Missense V180M Other Other Insertion Deletion Other Nonsense Other Nonsense Nonsense Missense M1I Other Missense R396C Other Nonsense Missense R185C (gf) Other (gf) Other Other (gf) Other (gf) Insertion Missense D323N Other Other Substitution Missense G498E  unc103(n500n1211) mod-5(n822)  MT2633  Other  MT8944  Nonsense  mod-5(n3314)  MT9772  Nonsense  grp-2(ok1179) grk-1(ok1239)  RB1150 RB1194  Other Other  α subunit, Gαs  Human orthologue GNAQ GNAQ GNAQ GNAQ GNAQ PLCβ PLCβ PLCβ RGS7 RGS7 GNAO GNAO GNAO GNAO GNAS  α subunit, Gαs adenylyl cyclase adenylyl cyclase adenylyl cyclase adenylyl cyclase phosphodiesterase phosphodiesterase Negative regulator K+ channel proprotein convertase ERG K+ channel  GNAS ADCY9 ADCY9 ADCY9 ADCY5 PKA KCNK18 PSCK2  NaCl dependent transporter NaCl dependent transporter G-protein regulator serine/threonin  -  KCNH6  GRK5 92  Name (allele)  Strain  Mutation type  aap-1(ok282) arr-1(ok401) Name (allele)  RB552 RB660 Strain  Other Deletion Mutation type  acy-4(ok1806) gpr-1(ok2126)  VC1331 VC1670  Deletion Deletion  Function  Human orthologue amino peptidase PI3KR1 beta-arresting ARRB1 Function Human orthologue Adenylyl cyclase ACDY5 G-protein regulator -  93  Appendix B: p-values from statistical analysis Table B.1 p-values of Pearson correlation coefficients  Rate accelerations no responses pauses decelerations reversals  accelerations 0.530 0.450 0.577 0.780  Asymptotic level reversals 0.780 0.028 0.042 0.609 -  accelerations 0.278 0.001 0.1757 0.867  reversals 0.867 0.934 0.448 0.012 -  94  Table B.2 p-values of strains tested for defects in reversal probability and non-reversal behaviours Response rate (probability) asymptotic level (probability)  acc  n.r.  pause  decel  rev  acc  egl-30(ep271) egl-30(ad806)  0.188 0.001  0.475 0.156  0.440 0.272  0.049 0.019  0.007 0.016  0.034 0.023  egl-30(n686)  5.91E05 0.002 0.371 0.028 0.009 0.406  0.427 399 0.048 0.105 0.226 0.161 0.373  0.48799 4  0.48097 8 5.29E-08 0.209 0.005 5.26E-05 8.20E-05  0.485003  0.42739 9 0.159 0.056 0.193 0.390 0.026  0.314 0.009 0.245  0.003 0.011 0.323 0.450 0.140 0.090 0.309 0.273 0.391 0.347  0.051 0.002 0.051 0.122 0.222 0.024 0.019 0.005  1.09E-01 8.08E-05 6.79E-06 0.189 9.70E-05 0.090 2.19E-05 3.15E-05 3.09E-06 2.48E-04  3.72E-06 5.31E-05 0.258 1.18E-10 0.003 7.82E-06 0.156 0.372 0.129 0.022  egl-30(n715) egl-30(js126) egl-8(e2917) egl-8(n488) egl10(md176) goa-1(sa734) goa-1(n1134) goa-1(n3055) eat-16(sa609) eat-16(ce71) dgk-1(nu62) gsa-1(ce81) gsa-1(ce94) acy-1(ce2) acy1(md1756) acy-1(nu329) pde-4(ce268) pde-4 (ok1290)  1.03E-05  0.142 0.004 0.018 0.049 0.258 0.097 1.03E-04  7.26E-13  0.210 0.055 0.143 0.173  1.30E-07  0.002  0.467  0.473 0.342  0.250 8.10E-06  4.77E-04 1.68E-07  0.135  0.463  0.478  0.016  7.06E-10  0.003 2.48E-07 0.057 0.032  n.r.  pause  0.097 1.48E08 0.46100 6 0.033  5.69E-08  7.50E-06 3.06E-09 1.10E-06 1.17E-06  decel  Initial response (probability)  rev  Reversals  0.036 0.058  0.004 0.001  0.226 0.009  0.45382 6  0.44275  0.358302  0.19683  1.82E-06 1.30E-06 2.69E-04  3.62E-06  5.96E-04 0.007  1.75E-08 0.049 0.004 0.373 0.216  0.166  0.018 0.007  0.006 7.23E-05 3.83E-04 9.66E-04  1.99E-06  0.016 0.010  9.63E-08 2.98E-07 3.19E-08 7.25E-06  1.79E-06 2.32E-06  0.007  2.57E-05  1.79E-06 2.32E-06  3.48E-05  0.240 0.058 0.104 0.273 0.108 0.075 0.016 0.443 0.005 0.033  0.003 0.038 0.094 0.001 0.112  0.037 2.62E-04 2.62E-04 9.25E-07 0.033 0.024 0.246 0.277  0.002 5.46E-05  1.34E-04 3.33E-13  0.031 0.013  0.120 2.52E-05  0.004 0.446  0.003 1.73E-05  0.094 0.252  0.067  0.002  0.270  0.005  0.094  0.048  0.359  0.009 0.017 1.98E-07  0.003 0.068 0.012 3.38E-10 1.73E-08 2.11E-06  0.002 4.23E-06 1.18E-05  0.008 0.001  0.005 3.97E-04  4.61E-05 0.009 1.98E-07 4.38E-01 0.438 0.089 6.52E-09 5.69E-07  0.100 0.001  95  Table B.3 p-values of heterotrimeric G-protein-associated genes tested for defects in reversal probability and non-reversal behaviours Response rate (probability) asymptotic level (probability) Initial response (probability) acc  n.r.  pause  decel  rev  acc  n.r.  pause  decel  8.28E-08  1.06E-06  rev  Reversals  twk-18(cn110)  7.22E-06 0.004  0.271  0.345  4.19E-08  3.47E-04  0.095  3.13E-10  2.32E-08  1.49E-09  0.006  0.009  0.099  egl-3 (n150)  0.010 0.101  unc103(n500n1211)  mod-5(n822) mod-5(n3314) grp-2(ok1179) grk-1(ok1239) aap-1(ok282) arr-1(ok401) acy-4(ok1806) gpr-1(ok2126)  3.5E-04 0.355 0.110 0.008 0.154 0.185 0.001 0.003 1.77  0.182 0.476 0.347 0.184 0.426 0.348 0.043 0.363 0.29  3.16E-12 0.061 0.041 6.10E-05 0.003 8.05E-05 0.005 0.017 1.64  8.28E-08 2.19E-05 0.074 0.008 0.115 0.096 0.008 9.83E-10 3.03  5.19E-07 0.483 0.042 0.258 0.435 0.198 0.005 0.011 -0.44  0.028 0.424 9.62E-07 5.78E-05 2.55E-10 1.38E04 0.017 2.22E-07 2.75E-06 1.19E-08 0.007 0.011 3.18E-07 1.32E-05 0.200 0.151 3.57E-04 1.17E-04 0.001 0.009 0.329 0.419 0.006 3.31E-04 0.377 0.211 0.015 3.75E-05 0.003 0.352 0.493 0.310 1.44E-04 0.002 0.323 0.001 0.004 2.12E-04 0.001 0.001 0.098 1.68E-04 0.471 3.66E-06 0.006 3.16 2.29 -3.55 -3.20 -0.64  0.086 0.028 0.103 0.165 0.354 0.130 0.341 0.247 0.366  96  Table B.4: p-values of all strains tested for reversal distance  Allele  Strain  rate initial asymtotic (distance) response level (distance) (distance)  egl-8(e2917) egl-30(ep271) egl-30(ad806) goa-1(sa734) eat-16(sa609) gsa-1(ce81) acy-1(ce2) gsa-1(ce94) kin-2(ok248) eat-16(ce71) dgk-1(nu62) acy-1(nu329) egl-10(n692) egl-30(n715) goa-1(n1134) unc103(n500n1211) egl-10(md176) goa-1(n3055) mod-5(n822) mod-5(n3314) egl-309(js126) egl-8(ok934) grp-2(ok1179) grk-1(ok1239) pde-4(ok1230) aap-1(ok282) arr-1(ok401) acy-4(ok1806) gpr-1(ok2126)  CB6614 CE1047 DA1084 DG1856 JT609 KG421 KG518 KG524 KG532 KG571 KP1097 KP1182 MT1443 MT1520 MT2426 MT2633  7.82E-05 2.35E-05 2.33E-07 1.12E-10 1.31E-05 0.293932 0.19047 0.106783 0.436459 0.005655 1.52E-05 0.081265 0.004472 5.84E-11 9.26E-13 0.003133  0.004245 0.41694 7.82E-05 0.014941 0.132008 0.397861 0.124385 0.409977 0.375627 0.467815 0.131054 0.321157 0.490959 0.012018 0.025166 0.432448  1.86E-08 1.38E-13 0.003549 5.08E-06 5.1E-07 0.191632 0.454844 0.038149 0.469636 6.48E-07 7.3E-07 0.041341 0.000791 1.33E-05 8.8E-11 3.61E-08  MT8504 MT8628 MT8944 MT9772 NM1380 RB1012 RB1150 RB1194 RB1231 RB552 RB660 VC1331 VC1670  0.008651 0.004663 0.115834 0.470502 0.008213 0.298886 0.103309 0.086412 0.209166 0.384514 0.404729 0.249639 0.024718  0.028487 0.204271 0.214557 0.12269 0.403042 0.21307 0.397861 0.487649 0.045549 0.450082 0.2686 0.003289 0.182229  8.3E-05 0.001619 0.176325 0.10983 6.03E-05 0.017233 0.028635 0.007703 0.315926 0.201957 0.307544 0.043482 0.035425  97  

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.24.1-0073828/manifest

Comment

Related Items