The Open Collections website will be unavailable July 27 from 2100-2200 PST ahead of planned usability and performance enhancements on July 28. More information here.

UBC Undergraduate Research

Do Caenorhabditis elegans exhibit spatial learning? Using a t-maze to explore association of a spatial… Law, Jackie WY 2009-05-08

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


42446-Law_Jackie_Caenorhabditis_elegans_spatial_learning_2009.pdf [ 221.81kB ]
JSON: 42446-1.0103632.json
JSON-LD: 42446-1.0103632-ld.json
RDF/XML (Pretty): 42446-1.0103632-rdf.xml
RDF/JSON: 42446-1.0103632-rdf.json
Turtle: 42446-1.0103632-turtle.txt
N-Triples: 42446-1.0103632-rdf-ntriples.txt
Original Record: 42446-1.0103632-source.json
Full Text

Full Text

T-maze and C. elegans 1 Running Head: SPATIAL LEARNING IN THE C. ELEGANS         Do Caenorhabditis elegans exhibit spatial learning? Using a t-maze to explore association of a spatial environment with an attractant Jackie Law University of British Columbia T-maze and C. elegans 2 Abstract This study investigated spatial learning in Caenorabditis elegans; the ability to associate reinforcing cues with a location. Naive, wildtype C. elegans were trained in a microfluidic t-maze in the presence of diacetyl (a volatile attractant associated with food) and subsequently tested to see if they could associate diacetyl with one arm of the t-maze. 70-80% of the subjects chemotaxed towards diacetyl during training phase, but they randomly chose left or right when diacetyl was absent (number of subjects that chose diacetyl being under 65%). From our experiments, it is unlikely that the worms are associating diacetyl with one arm of the t-maze, but appears to be using some component of the atmosphere as a cue. T-maze and C. elegans 3           Do Caenorhabditis elegans exhibit spatial learning? Using a t-maze to explore                            association of a spatial environment with an attractant  Caenorhabditis elegans, a soil-dwelling nematode, has been a widely used invertebrate research model for studying learning and memory (Wicks and Rankin, 1995; Rankin, Beck & Chiba, 1990). The nematode has 302 neurons, of which the cell lineage and connectivity have been mapped (White, Southgate, Thomson & Brenner, 1986). Despite its small number of neurons, researchers have demonstrated that C. elegans can exhibit both non-associative learning (such as habituation and sensitization) and associate learning (such as context conditioning; Rankin, 2004). In nature, these forms of learning and memory are thought to play a substantial role in survival. For three decades, mazes have been used as a tool to test learning and memory in vertebrate models, most commonly in the rat (Sutherland and Hamilton, 2004). Experiments using Morris’ water maze, Barnes maze and T-maze investigated spatial memory in rats in response to danger (such as with the Morris’ and Barnes mazes) (Faes, Aerts, Geys & De Schaepdrijver, 2009) or in response to reward expectancy (Feeney & Roberts, 2008), thus establishing the knowledge that rats are capable of spatial learning and can associate locations with a reward (or salvation from danger) within simple mazes. Using a reward-dependent learning paradigm, Dudchenko and Davidson (2001) trained rats to reach an arm of the maze (where they were rewarded with cereal), then tested them without rewards in a T-maze. At 12 trials per rat, 10 out of 14 rats were observed to choose the correct side. However, the vast number of neurons in vertebrate models, such as the rat, makes learning and memory difficult to study at a molecular level, especially when trying to identify where and how neural plasticity occurs in T-maze and C. elegans 4 specific neurons. The rat has over 2.79 million neurons in its right basal ganglia alone (Oorschot, 1998), a number that is significantly greater than the total number of neurons in C. elegans, making the C. elegans a comparitively more suitable organism to study when trying to pinpoint changes in neural connectivity underlying behaviour. There is a lack of current research that extends the t-maze model for spatial learning to invertebrates, so our study will investigate whether an invertebrate model such as the C. elegans can demonstrate learning by association within a t-maze.  Recently, Qin and Wheeler (2006) demonstrated that T-mazes can be used to evaluate learning in the Caenorhabditis elegans. They suggested that C. elegans may be able to show spatial learning in a maze by associating spatial location with the presence of volatile food odour. Qin and Wheeler produced microfluidic PDMS (silicone rubber) T-mazes and trained 20 worms to turn to the right chamber of the maze for five trials, where a hanging droplet of Escherichia coli OP50 was present as a food reward. The worms were subsequently tested in the absence of E. coli for five more trials. They found that 70-80% of total tested worms chemotaxed towards the right chamber and concluded that the worms can be conditioned to associate a food reward with spatial location. However, there were possible confounds in this study: for example, it is unclear whether Qin and Wheeler covered both of the maze’s open chambers. In addition, their lab environment lacked consistency in temperature and humidity over the course of their experiment (Lee Lau, by personal communication). These factors may produce detectable changes in the microenvironment within the maze that could affect C. elegans’ behaviour. T-maze and C. elegans 5  In this experiment, naïve, wildtype C. elegans were trained in microfluidic T-mazes under the exposure of an attractant, diacetyl. Our purpose is to re-establish and extend Qin and Wheeler’s observations by replicating their study under a more controlled environment. In addition, we manipulated environmental variables to identify how and what the worms might be using as cues to choose one chamber of the maze over the other. We hypothesized that worms will associate diacetyl’s presence in the right chamber and so, choose the right chamber during testing.   Method  Subjects   Wildtype N2 hermaphroditic Caenorhabditis elegans was obtained from the Caenorhabditis Genetics Centre. They were reared on Escherichia coli OP50 on nematode growth medium (NGM) at 20 degrees C (plus or minus 0.5C). The worms were synchronized (laid as eggs within a difference of plus or minus 2 hours) and tested at four days of age. Four-day old worms were used because in comparison to other ages, they exhibit the greatest attraction to diacetyl (Matsuura, 2007). Materials The NGM used during experimentation was made of 3% agar to prevent worms from burrowing out of the T-mazes. T-shaped, microfluidic silicone mazes were rinsed with EtOH and distilled water for cleaning, then dried before they were used. Tweezers were used to handle the mazes during washing and during its placement onto agar. Coverslips were broken into pieces that fitted and completely covered a maze chamber. The attractant used was 0.01% diacetyl, prepared by diluting it with EtOH. To set up the attractant for use within the maze, diacetyl was pipetted and gently tapped onto a T-maze and C. elegans 6 coverslip piece so that there was one drop of diacetyl on the center of the coverslip piece. Diacetyl is a volatile substance released by E. coli (bacteria that the subjects feed on in their lifetime); thus, it can act as a reinforcing cue to the presence of food. Matsuura   Fig. 1. Maze schematic. The t-maze was made with silicone rubber and leaves an air-filled avenue between the maze walls and the agar for passage. A coverslip was hung over a maze chamber. Reprinted from Qin and Wheeler (2006).  (2007) has also shown that diacetyl attracts C. elegans. A VCR (Panasonic AG1960) and monitor (NEC PM-1271A) connected to a dissecting microscope provided convenient viewing of a worm’s progress through a maze.  Procedure All experiments were performed in a temperature (20 degrees C plus or minus 0.5C) and humidity (34-41%) controlled room. Before the worms were released into the maze, they were washed free of E. coli in a culture tube filled with S-buffer for 60 minutes. During this period, any food remaining in the worms’ digestive systems should T-maze and C. elegans 7 have passed through, effectively rendering the worm without food for the duration of the experiment (adapted from benzaldehyde-starvation assay of Anne Hart (2006)). C. elegans is known to defecate once every 45 seconds (SD of 3 seconds) during feeding (Liu and Thomas, 1994). Worms were then transferred to NGM plates, and one worm was picked (using a tool made of a 1.5cm long platinum rod held in place by a pipet melted at the tip) onto a 100mm-wide Petri dish with 3% agar NGM. The lid of a Petri dish was stacked under the agar-filled dish to eliminate condensation build up as the worms were observed under a microscope. A worm was given a period of 5 minutes to acclimatize on the plate before it proceeded to the training phase. The worm started at the bottom of the T-maze and timed for completion of a trial; its choice of either the left or right chamber was indicated when the worm’s entire body has entered a chamber. When the worm has completed a trial, the maze was peeled off, then, it was repositioned on fresh, un-travelled agar on the same plate. The worm started at the bottom of the T-maze at the beginning of every trial. There was no time limit to complete the maze.  A total of 4 experimental assays were done. In 3 of the 4 assays, 0.01% diacetyl, a known attractant for C. elegans (Matsuura, 2007), was applied by placing one droplet onto a coverslip, which was hung with the droplet facing down over a maze chamber. The droplet did not come into contact with any part of the maze. The results were calculated by the proportion of worms that selected the right chamber in a pool of 20 worms in each assay. Averages, variances and statistical significances were evaluated by students’ t-test.  Experimental Assay 1 T-maze and C. elegans 8 To investigate attraction-associated learning, a naive N2 worm proceeded through two phases of an assay: the training and testing phase. The training phase represents the unconditioned response of a worm seeking an attractant (diacetyl). The testing phase proceeded immediately after the training phase, and represents the conditioned response of a worm’s choice for either the left or right chamber based on its previous experience in the training phase. Each phase consisted of 5 consecutive trials, where a worm’s choice was recorded, along with the time it took to complete a maze. During training, a hanging droplet of 0.01% was administered on a coverslip, which was flipped over the maze’s right chamber. Diacetyl was carefully reapplied every 20 minutes for the duration of the training phase due to evaporation. To test for memory retention of the maze, the testing phase lacked diacetyl. Both left and right chambers were covered in training and testing. Experimental assay 2   To investigate a possible confounding factor for differences in the microenvironment and its effect on the worm’s choice, the left chamber was uncovered during training phase while the right chamber donned a coverslip with diacetyl. During the testing phase (no diacetyl), only the left chamber was covered.  Experimental assay 3 To investigate a worm’s preference between open and covered chambers, the right chamber was covered during training, while the left chamber was covered during testing. No diacetyl was applied in either phase so that the only manipulated change in variable was the placement of the coverslip. Experimental assay 4 T-maze and C. elegans 9  To investigate whether a change in the microatmosphere acted as a cue to enhance memory retention of the spatial environment in the maze given the presence of diacetyl, the left chamber was uncovered for both training and testing. Diacetyl was applied on the right chamber in training only. The right chamber was covered during testing. Results  In trials 3-4 of the training phase, C. elegans demonstrated exploratory behaviour, characterized by frequent turns and reversals, as well as head waggles that did not accompany much movement in the direction the head was facing at any one instance. The worms stayed close to the maze walls and displayed very little movement during those times. They often explored both arms of the maze before choosing a chamber. Worms often lingered just beyond the border of a chamber with diacetyl before entering that chamber. If one chamber was uncovered (i.e. assays 2, 3, and 4), worms sometimes reversed away from the open chamber-this behaviour was also sometimes observed as worms partially entered a chamber with diacetyl, although the latter occurred much less frequently. In trials 4-10, worms no longer lingered in one place for very long. The worms also did not tend to travel along the maze walls as they did in earlier trials, but travelled with increased speed along the centre of the air-filled avenues. Unlike earlier trials, the worms tended to swim directly into a chamber (whether it be covered or uncovered) with little or no pause at a chamber’s border.    Worms in the training phase were more likely to choose the right arm containing diacetyl compared to the testing phase; that proportion ranged from an average of 70-80%. This indicates that they were attracted to diacetyl. The highest proportion of worms that chose the right arm was found in the training phases of assays that had diacetyl in the T-maze and C. elegans 10 right chambers (fig. 2, fig. 3 and fig. 5). The number of worms choosing the right arm during testing phases for all assays was at close to chance proportions. Although that number for assay 4 (fig. 5) in testing phase was higher than chance in comparison to other assays, it was not statistically significant. Significant differences were p<0.001, p<0.0001, p<0.0001, p<0.2 for assays 1, 2, 3, 4, respectively.  The time it took for the worms to complete a trial in the t-maze decreased with time for all assays. For instance, the average times elapsed for assay 1 were 597 seconds and 249 seconds for trials 1 and 10, respectively (p=0.130). There were noticeable drops in elapsed time as soon as trial 2 (fig. 4) or trial 3 (fig. 2, 3, 5). With the exception of assay 3, the times of completion reached plateaus in the testing phase. Discussion   The results were not consistent with the hypothesis because C. elegans chose randomly between left and right during testing (when diacetyl was removed). C. elegans were evaluated for their ability to associate a spatial location within a t-maze with the presence of an environmental cue. Diacetyl was used as an US to train wildtype worms to select the right chamber of the t-maze. However, as soon as diacetyl was taken away in the testing phase, there was a noticeable drop in the proportion of worms that chose the right-sided chamber. This result is inconsistent with Qin and Wheeler (2006), who found that the number of worms choosing the right chamber was 20-30% above chance in the testing phase.  The worms did not appear to be associating diacetyl with an arm of the t-maze. In assays 2-4, we tested the possibility of an additional environmental cue that could act as a US. Thus, we manipulated an aspect of the atmospheric in one arm by covering the right T-maze and C. elegans 11 chamber during training, then covering the left chamber during testing (assays 2 and 3). Results show that worms tended to select the covered chamber 60-65% of the time during testing (fig. 3, 4), suggesting that they were not conditioned to diacetyl but were selecting for a different environmental cue. Finally, assay 4 tested for the possibility that both diacetyl and the presence of a coverslip acted in conjunction to stimulate a worm’s memory for turning right in testing. The right chamber was covered in both training and testing, while diacetyl acted as the US during training. Testing results for assay 4 showed that more worms (65-70%) chose the right chamber (fig. 5), but the differences between testing and training were insignificant. These results suggest that the worms most likely do not show association with its spatial environment with an attractant but were following an environmental gradient within a maze (by olfaction during training or presence of coverslip).   In this study, worms have shown preference to a covered chamber, however, we do not know exactly what component of the atmospheric gradient to which the worms were responding. We speculate that oxygen levels, carbon dioxide levels or air flow may play a role that affects C. elegans’ choice in a t-maze, which can be investigated by using strains that lack the sensory mechanism to detect those aspects. For example, the C. elegans strain, gcy-35, do not aerotax and do not sense changes to oxygen levels (Cheung, Cohen, Rogers, Albayram, de Bono, 2005), therefore we can use these worms to see if O2 concentrations affect their performance in the t-maze. Understanding the cue to which worms respond can help gauge C. elegans’ sensitivity to changes within a microenvironment. T-maze and C. elegans 12 Based on the change in speed and number of turns and reversals during the worms’ time in the maze, the worms may be learning some unknown aspect of the maze. In all assays, a worm’s travelling speed increased as it spent more time in the maze. Increased speed was coupled with the tendency to travel near the centre of the maze avenues, and a distinct decrease in the number of pirouettes and reversals. This change in behaviour was independent of the presence of diacetyl, because the same behaviour was observed in assay 3 (where no diacetyl was used). One possible explanation is that the initial behaviour is a response to a worm’s introduction to a novel environment, and the worm switched to long-range travelling as it became acclimatized with the environment (Gray, Hill & Bargmann, 2004). However, without testing the effects of diacetyl quantitatively, we cannot rule out its effect on worms in the t-maze. A future direction is to test diacetyl concentrations at 0.0001M, 0.001M, 0.01M (Matsuura, 2007) in a t-maze while recording the time of completion for each trial. In addition, a more complicated maze, such as a multidirectional U-maze (more than two arms) can be used to further explore worm navigation in a maze-which may lead to interesting speculations on the extent of the C. elegans’ ability to learn to navigate its environment.  This will address the question of how maze-learning correlates to the speed with which worms swim through a maze, thus, relating to the time of completion of a trial. Based on the findings of this study, C. elegans is unable to associate a reinforcing cue (diacetyl, a volatile attractant) with a specific spatial location within a t-maze. Furthermore, they are making a choice between the left or right arm by following an environmental aspect that may be a difference in oxygen or carbon dioxide levels when diacetyl is absent. Although our study does not echo past findings on spatial learning in T-maze and C. elegans 13 C. elegans, we found indication that the worms may be learning some unknown aspect of the t-maze, demonstrated by the decreasing amount of time they require to complete the maze.T-maze and C. elegans 14    Fig. 2. Assay 1 where both chambers were covered during training and testing (n=20). (a) Proportion of worms that chose the right arm grouped by phase (p=0.0003). Worms chose the right arm during training but choice between left and right was random during testing. (b) Proportion of worms that chose the right arm grouped by trial. (c) Time taken (seconds) for worms to choose a chamber. As number of trials increased, time taken to complete the maze decreased. T-maze and C. elegans 15  Fig. 3. Assay 2 where the right chamber had diacetyl and was covered during training while the left chamber was covered (no diacetyl present) during testing (n=20). Worms chose the right arm during training but choice between left and right was random during testing. (a) Proportion of worms that chose the right arm grouped by phase (p=0.0003). (b) Proportion of worms that chose the right arm grouped by trial. (c) Time taken (seconds) for worms to choose a chamber. As number of trials increased, time taken to complete the maze decreased.    T-maze and C. elegans 16   Fig. 4. Assay 3 where only the right chamber was covered during training and only the left chamber was covered during testing (n=20); no diacetyl was present during both training and testing. Choice between left or right was near random during training while about 35% of the worms chose the right arm during testing. (a) Proportion of worms that chose the right arm grouped by phase (p<0.0001). (b) Proportion of worms that chose the right arm grouped by trial. (c) Time taken (seconds) for worms to choose a chamber. As number of trials increased, time taken to complete the maze decreased. T-maze and C. elegans 17  Fig. 5. Assay 4 where only the right chamber was covered in both phases, and diacetyl was present in training (n=20). Worms chose the right arm during training but choice between left and right was random during testing. (a) Proportion of worms that chose the arm grouped by phase (p=0.1724). (b) Proportion of worms that chose the right arm grouped by trial. (c) Time taken (seconds) for worms to choose a chamber. As number of trials increased, time taken to complete the maze decreased. T-maze and C. elegans 18                                                                                   References  Cheung, B. H. H., Cohen, M., Rogers, C., Albayram, O., & de Bono, M. (2005). Experience-dependent modulation of C. elegans behavior by ambient oxygen. Current Biology, 15(10), 905-917.  Dudchenko, P. A. (2001). How do animals actually solve the T maze? Behavioral Neuroscience, 115(4), 850-860.  Faes, C., Aerts, M., Geys, H., & De Schapdrijver, L. (2009). Modeling spatial learning in rats based on morris water maze experiments. Pharmaceutical Statistics,  Feeney, M. C., & Roberts, W. A. (2008). Rats show preference for delayed rewards on the radial maze. Learning Behaviour, 36(1), 42-54.  Gray, J. M., Hill, J. J., & Bargmann, C. I. (2004). A circuit for navigation in caenorhabditis elegans. Proceedings of the National Academy of Sciences of the United States of America, 102(9), 3184-3191.  Hart, A. C. (2006). Benzaldehyde starvation learning. Behaviour (1.87.1 ed., pp. 27) Retrieved from  Liu, D. W., & Thomas, J. H. (1994). Regulation of a periodic motor program in C. elegans . J.Neurosci., 14(4), 1953-1962.  T-maze and C. elegans 19 Matsuura, T., Endo, S., Iwamoto, R., Takahashi, H., & Ichinose, M. (2007). Developmental changes in chemotactic response and choice of two attractants, sodium acetate and diacetyl, in the nematode caenorhabditis elegans . Comparative Biochemistry and Physiology - Part A: Molecular & Integrative Physiology, 147(4), 920-927.  Oorschot, D. E. (1998). Total number of neurons in the neostriatal, pallidal, subthalamic, and substantia nigral nuclei of the rat basal ganglia: A stereological study using the cavalieri and optical disector methods. Journal of Comparative Neurology, 366(4), 580-599.  Qin, J., & Wheeler, A. (2006). Maze exploration and learning in C. elegans. Lab on a Chip, doi:10.1039/b613414a  Rankin, C. H., Beck, C. D. O., & Chiba, C. M. (1990). Caenorhabditis elegans: A new model system for the study of learning and memory. Behavioural Brain Research, 37, 89-92.  Rankin, C. H. (2004). Invertebrate learning: What can't a worm learn? Current Biology, 14(15), R617-R618.  Sutherland, R. J., & Hamilton, D. A. (2004). Rodent spatial navigation: At the crossroads of cognition and movement. Neuroscience & Biobehavioral Reviews, 28(7), 687-697.  White, J. G., Southgate, E., Thomson, J. N., & Brenner, S. (1986). The structure of the nervous system of nematode caenorhabditis elegans. Philosophical Transactions of the Royal Society of London, 314, 1-340.  T-maze and C. elegans 20 Wicks, S. R., & Rankin, C. H. (1995). Integration of mechanosensory stimuli in caenorhabditis elegans. J.Neurosci., 15, 2434-2444.   


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items