BEHAVIORAL AND NEURAL BIOLOGY 58, 211--221 (1992)
An Analysis of Behavioral Plasticity in Male Caenorhabditis elegans K I M BILL M A H AND CATHARINE H . R A N K I N 1
Department of Psychology, University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z4
ulation and observation make C. elegans an excellent model system in which to study the relationships between neurons and behavior. C. elegans also has a rich repertoire of behavioral plasticity (Rankin & Chiba, 1988; Rankin, Beck, & Chiba, 1990; Kumar, Williams, Culotti, & van der Kooy, 1989). Rankin et al. (1990) have shown that the tap withdrawal reflex (in response to a tap to the dish holding the worm) in C. elegans exhibits the major forms of nonassociative learning: habituation, dishabituation, and sensitization. The neural circuit for tail-touch-induced forward swimming and head-touch-induced backward swimming in the hermaphrodite, described by Chalfie, Sulston, White, Southgate, Thomson, and Brenner (1985), has been shown to underlie the tap withdrawal reflex as well (Rankin & Chalfie, 1989; Wicks & Rankin, 1991). Thus, a great deal is known about the C. elegans hermaphrodite; however, C. elegans also occurs infrequently as a male as a result of X-chromosome nondisjunction (Hodgkin, Horvitz, & Brenner, 1979). Males do not produce eggs and thus must mate with hermaphrodites in order to reproduce. To that end, males have specialized neuroanatomy, sensory organs, and behavior. The male has a larger complement of somatic nuclei than the hermaphrodite: 1031 in total, 72 more than the hermaphrodite (Hodgkin, 1983, 1988). Most of these additional cells are accounted for by the existence of the copulatory apparatus of the tail region, including the vas deferens, the cloaca, and the sclerotic hook. In addition to having the two-tail sensilla possessed by the hermaphrodite (Sulston, Albertson, & Thomson, 1980), the male C. elegans can receive additional sensory input from the tail via 32 additional sensory neurons associated with the various structures that make up the specialized male tail (Wood, 1988). In order for males to reproduce they must mate; the additional structures in the tails of male C. elegans appear to be associated with specialized
Caenorhabditis elegans is a simple soil-dwelling nematode which has two sexes, hermaphrodite and male. The male C. elegans is differentiated from the hermaphrodite by the presence of 14 sensory structures in the tail. In this study, we compared the behavioral responses of males and hermaphrodites to head-touch and to tap. We hypothesized that the anatomical difference in sensory structures might result in behavioral differences in the reversal response to vibratory stimulation (a tap to the side of the holding dish). In the response to increasing intensities of tap, both sexes showed an increase in response magnitude, with the males showing larger responses than hermaphrodites. In addition, the male was shown to be capable of simple nonassociative learning: it demonstrated habituation and recovery from habituation in a similar manner as the hermaphrodite. Tail-touchinduced inhibition of the reversal response appeared to be similar in males and hermaphrodites. The evidence suggests that the touch withdrawal circuit in hermaphrodites is also present in the male C. elegans, and that the subtle differences in response to tap seen in males may result from the additional sensory receptors of the copulatory bursa of the tail. It seems clear from these studies that these structures do not play a key role in the male worm's response to tap. ©1992AcademicPress, Inc. The nematode Caenorhabditis elegans provides a powerful model system for the study of the relationship between behavior and its neural underpinnings. C. elegans possesses a simple, well-mapped nervous system of 302 neurons, a small genome which is currently the object of intensive mapping studies, and a short reproductive cycle, developing from egg to adult in approximately 31/2 to 4 days at 20°C. These characteristics, coupled with a behavioral repertoire amenable to experimental manip1 The authors acknowledge Catherine Chiba and two anonymous reviewers for editing assistance. The research was funded, in part, by a n NSERC operating g r a n t to C.H.R. Address correspondence and reprint requests to C a t h a r i n e H. Rankin, Dept. of Psychology, University of British Columbia, Vancouver, BC, Canada V6T 124. 211
0163-1047/92 $5.00 Copyright © 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
212
MAH AND RANKIN
mating behaviors. Mating behavior begins when a male contacts a hermaphrodite with its tail; the male then rubs the copulatory bursa on the side of the hermaphrodite, moving backward until it contacts the vulva. The male attaches to the hermaphrodite using the spicules, opens the vulva, and ejaculates sperm into it, and then it removes the spicules and swims away (Chalfie & White, 1988). The neuroanatomy and function of the touch circuit in males has not been examined; however, males respond to head-touch with backward movement and tail-touch with forward movement as do hermaphrodites. Although this suggests that the underlying neural circuit is similar in both sexes, the additional sensory structures in the male tail may interact with elements of the touch circuit. Thus, it is possible that the differences in sensory structures between males and hermaphrodites may result in subtle differences in response to vibrational stimulus such as a tap and perhaps differences in nonassociative learning. Using hermaphrodites Chiba and Rankin (1990) showed that although there were no changes in behavior in the touch response during development, there were changes in the response to a tap. Thus, response to tap may be a more sensitive measure of mechanosensory organization than response to touch. The following experiments were designed (1) to examine whether males responded to touch and tap the same as hermaphrodites, (2) to compare the plasticity of the tap response behaviors mediated by the touch circuit in male and hermaphroditic C. elegans, and (3) to provide clues as to the nature of the touch circuit in the male. METHODS
Subjects In all experiments, C. elegans hermaphrodites were controls while male C. elegans were experimental subjects. A total of 200 hermaphrodites and 200 males (strain N2 Bristol) were used. Age-synchronized colonies were set up; all worms were tested at 4 days of age (adults; Byerly, Cassada, & Russell, 1976).
Materials All animals were maintained at 20°C on 5-cm petri plates filled with Nematode Growth Medium (Brenner, 1974) and streaked with Escherichia coli (strain OP50) as a source of food for C. elegans. For testing, individual animals were transferred to unstreaked (i.e., no food present) agar-filled plates. In
all cases, worms were transferred to the test plate a minimum of 1 to 2 min prior to experimental manipulation.
Apparatus The apparatus used was the same as that described in Chiba and Rankin (1990). Test plates were viewed using a Wild M3Z stereomicroscope with an attached Panasonic D5000 high-resolution video camera and Panasonic video cassette recorder. Taps were delivered to the side of the test plate using an electromagnetically controlled tapping arm. Shocks were administered with a spanning electrode consisting of two wires placed in the agar, one on either side of the animal. Both the tap and shock parameters were controlled by a Grass S88 stimulator.
Scoring and Statistical Analysis Using stop-frame video analysis, the reversal responses of each animal were traced onto acetate sheets. Measurements of the magnitude (length) of the tracing (and hence of the reversal) were rendered on an Apple Macintosh SE computer using a Bit Pad Plus digitizing tablet and MacMeasure software. Data involving numbers of animals within groups showing reversals were analyzed using an overall ANOVA with dichotomous data (Winer, Brown, & Michels, 1991). Data involving magnitude information were analyzed using overall ANOVAs with post hoc analyses as warranted. Trend analysis was used to analyze magnitude data obtained in response to stimuli of graded intensity. Regression analysis was employed to test rate measures (i.e., rates of habituation or recovery from habituation). Repeated measures data were analyzed with a repeated measures ANOVA. EXPERIMENT I: SPONTANEOUS REVERSALS Although the reversal response (changing from forward to backward swimming) is seen in response to stimuli such as taps or tail touch, both male and hermaphroditic C. elegans also reverse in the absence of such applied stimulation. We have defined such reversals as spontaneous reversals. Chiba and Rankin (1990) demonstrated that adult C. elegans hermaphrodites show two to three spontaneous reversals/minute. Since the male finds and mates with the hermaphrodite by touching its tail to the hermaphrodite's body, it was predicted that, to increase the probability of mating, males would show more spontaneous reversals than hermaphrodites.
213
BEHAVIORAL PLASTICITY IN MALE C. elegans
This experiment was designed to test whether males show larger and/or more frequent spontaneous reversals than to hermaphrodites.
Subjects and Procedure Twenty male and 20 hermaphroditic 4-day-old C.
elegans were used in this experiment. Worms were transferred to test plates at least 1 min prior to beginning the experiment. A 10-min period of spontaneous activity was then recorded.
A oo ,,,.I
30
O0 n" ILl
20
U.I n'LI. 0
lo
[]
TRUESPONTANEOUS
[]
SELF-CONTACT
Z I.g
=E
o MALE
HERM
Scoring and Statistical Analyses Videotapes were scored for total number and magnitude of reversals during the 10-min observation period; data were analyzed using overall ANOVAs. On average, males are smaller than hermaphrodites of the same age. To ensure that differences in reversal magnitude were not due to an effect of worm length, the magnitude of each reversal was standardized by dividing the reversal length by the length of that particular worm (Chiba & Rankin, 1990; Beck, 1991).
B W a I,m Z
1.0
Number of reversals. An examination of the videotaped behavior revealed that males showed two types of "spontaneous" reversals: (1) self-contact reversals and (2) true spontaneous reversals resembling those shown by hermaphrodites. A self-contact reversal was defined as a reversal in which the tail of the animal made contact with its own body (described in Chalfie & White, 1988). Self-contact reversals accounted for approximately 19% of the total number of reversals observed in males while accounting for less than 3% of reversals shown by hermaphrodites. Overall, males showed a significantly greater number of both true spontaneous reversals [Fig. 1A, F(1, 38) = 8.35, p < .01] and selfcontact reversals [Fig. 1B, F(1, 38) = 13.69, p < .001] than did hermaphrodites. This effect was not due to differences in activity level, since there were no significant differences between males and hermaphrodites in time active as measured by the percentage of total time (over 10 min) swimming forward and backward, [(male mean = 99.875, SEM = 0.091), (hermaphrodite mean = 99.058, SEM = 0.663); t(19) = 1.209, p = .1207]. Reversal magnitude. There were no significant differences between males and hermaphrodites in the magnitudes of true spontaneous reversals, [F(1, 38) = 0.34, p = .5639]. However, the magnitude of self-contact reversals was significantly larger in
TRUE SPONTANEOUS
[]
SELF-CONTACT
0.8
0.8
=E ,,.I
0.4
O~ n,. U.I
0,2
U.I
n-
0.0
MALE
Results
[]
HERM
FIG. 1. (A) Mean number of true and self-contact spontaneous reversals + SEM shown by males (n = 20) and hermaphrodites (n = 20) over a 10-min observation period. Males showed significantly more true and self-contact spontaneous reversals than hermaphrodites. (B) Magnitude (distance traveled/body length) of true and self-contact reversals + SEM shown by males (n = 20) and hermaphrodites (n = 20) over a 10-min observation period. The magnitude of true spontaneous reversals did not differ; however, the magnitude of self-contact reversals was significantly higher in males.
males than in hermaphrodites [F(1, 38) = 8.27, p < .01].
Discussion Males show significantly more true spontaneous reversals and significantly more self-contact reversals than hermaphrodites. In self-contact reversals males reverse in response to their tails touching their own bodies; they then press the entire fan-like tail structure firmly against their cuticles and rub it along the body vigorously for several seconds (this is similar to the mating behavior when a hermaphrodite is contacted). In contrast, when the tail of the hermaphrodite contacts its own body, it only briefly brushes its cuticle with its tail. Thus, it appears that the self-contact reversal of the male m a y be an active attempt to mate while that of the hermaphrodite is more perfunctory in nature. In contrast to self-contact reversals, true spon-
214
MAH AND RANKIN
taneous reversals occurred with the same magnitude in both sexes, although with greater frequency in the males. By reversing more frequently males may increase the probability of bringing their tail copulatory structures into contact with hermaphrodites, and thus increase the likelihood of successful mating.
A ,-I
20
,< UJ W n" u.
15
10
0 W
5
w
EXPERIMENT II: RESPONSE TO GRADED TAP INTENSITY
Z
0
MALE
Adult hermaphrodites reverse direction and swim backward following a tap to the side of the plate and show a graded response to vibrational stimuli of different intensities, showing larger reversals to taps of larger intensities Beck and Rankin (in press). This experiment was designed to uncover differences in responses to vibrational stimuli between males and hermaphrodites by comparing their responses to stimuli of varying intensity.
• []
WEAK INTERMEDIATE
[]
STRON
1
HERM
B u.I a .-j IwZ
1.5-
•
WEAK
[]
INTERMEDIATE
1.0
.J ¢j)
O.5'
W
Subjects and Procedure Single worms were placed on individual unstreaked test plates approximately 2 h prior to testing. The force of the tap was varied by altering the voltage to the electromagnetic relay controlling the tapper. To produce a weak tap, a 34-V intensity was used, to produce an intermediate tap a 40-V intensity was used, while to produce a strong tap a 60V intensity was used. Twenty males and 20 hermaphrodites 4 days old were tested at each tap intensity.
Scoring and Statistical Analyses The number of animals in each group showing reversals to the differing intensities of tap and the magnitude of those reversals that did occur were scored. The number of worms reversing was analyzed using a two-factor ANOVA. Between-group and within-group differences were analyzed with the Least-Square Means method. Within group magnitude data were analyzed using an overall twofactor ANOVA.
Results Number of reversals. As tap intensity increased, the number of reversals shown by both males and hermaphrodites increased [Fig. 2A, F(2, 114) = 7.171, p = .0012]. Worms showed significantly fewer reversals to weak taps than to intermediate or strong taps (p < .05). Furthermore, there were no
KI
n"
0.0
MALE
HERM
FIG. 2. (A) The number of males (n = 20) and hermaphrodites (n = 20) showing reversals to weak, intermediate, and strong taps. Among both males and hermaphrodites significantly fewer worms reversed to a weak tap, whereas most worms reversed to intermediate and strong taps. There were no sex differences in the number of worms reversing at any tap intensity. (B) Mean reversal magnitude (distance traveled/body length) + SEM of males (n = 20) and hermaphrodites (n = 20) in response to taps of weak, intermediate, and strong intensity. Reversal magnitude for each animal was standardized by dividing the reversal observed by the length of the worm showing that reversal. Both males and hermaphrodites showed larger reversals as tap intensity increased. Response magnitudes were significantly higher in males than those in hermaphrodites at each tap intensity.
differences between males and hermaphrodites in number of reversals at different tap intensities [F(2, 114) = 0.222, p = .8014].
Reversal magnitude. Animals of both sexes responded to increases in tap intensity with increasing reversal magnitudes [Fig. 2B, F(2, 114) = 4.957, p = .0086]. At all tap intensities, response magnitudes were greater in males than in hermaphrodites, [F(1, 114) = 11.983, p = .0008]. As tap intensity increased, the increase in reversal magnitude shown by males paralleled that shown by hermaphrodites, and no significant interactions between sex and tap intensity were observed [F(1, 114) = 0.201, p = .8180].
BEHAVIORAL PLASTICITYIN MALEC. elegans
215
Discussion
Subjects and Procedure
Both males and hermaphrodites showed graded responses to tap and made larger reversals in response to taps of larger intensities. Males, however, showed significantly larger reversals than hermaphrodites at all stimulus intensities tested, suggesting that they were more responsive to tap than were hermaphrodites. The differences between the sexes does not appear to be a simple case of the male tail sensory structures making the male more sensitive to all tactile stimulation of the tail. In contrast to hermaphroditic ~ail-touch receptors which produce forward movement the additional sensory structures in the tail appear to enhance the magnitude of male worm's reversals to tap.
Twenty males and 20 hermaphrodites 4 days old were used. Animals were placed individually on test plates 24 h prior to testing (to replicate Rankin, 1991). A small amount ofE. coli was placed on the test plate to ensure that the subject was not food deprived. Each animal received seven different stimulus conditions, separated by intervals of approximately 15-20 min (adult hermaphrodites show no habituation at a 10-min ISI). The conditions were (1) single tap, (2) tail-touch followed 1 s later by tap, (3) tail-touch followed 5 s later by tap, (4) tail-touch followed 10 s later by tap, (5) tail-touch followed 20 s later by tap, (6) tail-touch followed 30 s later by tap, and (7) single tap alone. Condition i established a baseline level of response to tap alone. Conditions 2-6 tested for the existence of the inhibition effect and explored the temporal parameters of the effect itself. Condition 7 tested whether the response to tap alone showed habituation due to stimuli from conditions 2-6. Since the intent was to study the effect of tail-touch on reversals to tap, worms that did not reverse to the initial tap were not tested (fewer than 10% of worms). Tail-touch was administered by hand by drawing an eyelash mounted to the tip of a 30-gauge needle across the tail of a worm.
EXPERIMENT III: INHIBITION Given that males and hermaphrodites respond to tap with reversals it is likely that the touch circuit responds to tap in roughly the same way in both sexes, with perhaps, as suggested in the previous experiment, enhanced excitation of the reversal circuit in males. Touch is a very different sort of tactile stimulus from tap, although it does activate the same circuit and produce the same types of responses. In contrast to tap, direct touch to the tail of hermaphrodites stimulates only the tail-touch receptors, without stimulating the receptors that would produce the competing response of backward movement. In both males and hermaphrodites tail-touch produces forward swimming. In hermaphrodites the activation of forward swimming by tail-touch inhibits the competing response of reversal to tap (Rankin, 1991). Tail-touch administered 1 s prior to tap decreases both the likelihood and magnitude of reversal to tap. This inhibition of reversal to tap by tail-touch lasted about 10 s and was followed by facilitation of the reversal magnitude to tap above baseline levels. Since the graded response experiment suggested that males might have a bias toward larger reversals, males might show a different pattern of behavior than hermaphrodites when tail-touch and tap are put into competition. The following experiment examines response competition in males and in hermaphrodites in order to assess the relative contributions of head- and tail-touch receptors to the response to tap.
Scoring and Statistical Analyses The number of reversals to tap was examined across sexes to determine between-sex differences. Within each sex, an ANOVA on the number of reversals was compared across increasing tail-touch tap intervals. The magnitude of the tap response was analyzed in each sex with a one-factor repeated measures ANOVA and multiple comparisons with Fisher's PLSD.
Results Number of reversals. All animals reversed to the initial tap. Both males and hermaphrodites showed a large decrease in the number of reversal responses to tap at the 1-s tail-touch/tap interval (p < .01 in both cases). At all other intervals, there were no significant differences in the number of animals reversing when compared to the baseline tap. Additionally, males and hermaphrodites did not differ from each other in the number of reversals shown at any of the tap intervals (Fig. 3A).
216
MAH AND RANKIN A
• []
•
20
n-
lO
MALE HERM
Discussion
P, s ~ z
o TAP 1S 5S 1 0 S 2 0 S 3 0 S TAP TAIL-TOUCH/TAP INTERVAL
200
•
MALE
T
50
100
.~ o~
In both males and hermaphrodites, tail-touch administered 1 s prior to tap resulted in a decrease or inhibition of both the frequency of reversals and the magnitude of reversals at short touch/tap intervals. EXPERIMENT IV: H A B I T U A T I O N AND DISHABITUATION
B
/~
but nonsignificant trend of facilitation 20 and 30 s following tail-touch. The sexes did not differ significantly from one another at any of the tailtouch/tap intervals in the magnitude of reversals to tap [F(1, 38) = 0.929, p = .3413].
so o TAP 1S 5S lOS TAIL-TOUCH/TAP
2OS 3OS TAP INTERVAL
FIG. 3. (A) Number of males (n = 20) and hermaphrodites (n = 20) showing reversals to tap and to tap following tail-touch at intervals of 1, 5, 10, 20, and 30 s. Both males and hermaphrodites show a significant decrease in the number of reversals to tap at the 1-s tail-touch/tap interval. (B) Mean magnitude of reversals + SEM of males (n = 20) and hermaphrodites (n = 20) to tap and tail-touch/tap intervals of 1, 5, 10, 20, and 30 s. Reversal magnitude is expressed as a percentage of each animals reversal to the initial tap (which was set at 100%). Males and hermaphrodites showed a significant decrease in reversal magnitude to tap at the 1- and 5-s tail-touch/tap interval. Hermaphrodites showed significant facilitation of reversal magnitude to tap at the 20- and 30-s tail-touch/tap interval and at the final tap. Males did not show significant facilitation.
Reversal magnitude.
To directly compare inhibition between males and hermaphrodites, reversal magnitude was standardized by expressing each worm's response as a percentage of its initial response to tap, which was set at 100%. Both males and hermaphrodites made significantly smaller reversals to taps 1 and 5 s following tail-touch than they did to taps alone [males: F(6, 114) = 13.814, p < .001, Fisher's PLSD: p < .01 and p < .05, respectively; hermaphrodites: F(6, 114) = 15.844, p < .001, Fisher's PLSD: p < .01 in both cases). Hermaphrodites also showed facilitation of reversal magnitude to tap 20 and 30 s following tail-touch when compared to reversal magnitude to tap alone (p < .01, Fig. 3B). Males showed a similar
Habituation is defined as a decreased response to repeated stimulation that is not due to sensory or motor fatigue. Dishabituation is the facilitation of the habituated response due to the presentation of a novel or noxious stimulus (Groves & Thompson, 1970). Both habituation and dishabituation of the reversal response have been demonstrated in C. elegans hermaphrodites (Rankin & Chiba, 1988; Rankin et al., 1990; Rankin & Broster, 1992). The current experiment was designed to test whether male C. elegans were capable of habituation and dishabituation and to compare their response patterns to those of hermaphrodites.
Subjects and Procedure Twenty male and 20 hermaphroditic 4-day-old worms were used. Sixty single taps with an interstimulus interval (ISI) of 10 s were delivered followed by the administration of a 60-V shock within 10 s of the last tap. Within 10-20 s of the shock, an additional series of 10 taps with an ISI of 10 s was administered.
Scoring and Statistical Analysis Only animals which showed a reversal to the initial tap were scored (greater than 90% of all anireals tested). A response in which an animal responded to the tap by moving forward more rapidly than before was replaced with the group mean; this was because the relationship between reversals and such infrequent accelerations is unclear. This replacement did not have a large impact on statistical analyses since fewer than 1% of data points were replaced. The responses to the first seven stimuli were analyzed with a repeated measures ANOVA (number of reversals) or a regression analysis (magnitude data) to determine whether habituation was present and, if so, to determine its rate. The final
217
BEHAVIORAL PLASTICITY IN MALE C. elegans
10 stimuli prior to the administration of the dishabituating stimulus (shock) were analyzed with a repeated measures ANOVA (number of reversals) to compare the final levels of habituation reached between the sexes. Finally, dishabituation was tested by averaging the magnitude of the last 10 responses and comparing this mean to the response immediately following the shock using a paired t test. The last 10 responses were used to provide a better estimate of preshock responding than would be obtained by a single point.
Results Number of reversals. The number of males and hermaphrodites reversing to tap decreased significantly over the course of the first seven taps [F(6, 228) = 5.565, p = .0001]. Although males and hermaphrodites showed a similar pattern of decrement in the number of reversals, over the first seven taps males responded with a greater number of reversals than did hermaphrodites [Fig. 4A, F(1, 38) = 4.163, p = .O483. During the last 10 stimulus presentations preceding the shock, the number of reversals in both sexes had stabilized, with males showing a significantly higher number of reversals than hermaphrodites [F(1, 38) = 22.537, p = .0001]. To determine whether the shock increased the number of reversals shown over habituated levels, the mean number of reversals to the last 10 stimuli prior to shock was compared to the number of reversals immediately following the shock. Following shock there was an increase in the number of males reversing; however, this increase just missed significance [#19) = 1.612, p = .06]. The hermaphrodites did show a significant increase in the number of worms reversing, and thus expressed significant dishabituation [t(19) = 2.398, p = .0134].
Reversal magnitudes.
To compare habituation and dishabituation in males and hermaphrodites, reversal magnitude was standardized by expressing
each worm's response as a percentage of its initial response to tap, which itself was set at 100%. The rate of response decrement was determined by computing the mean of the slopes of regression lines obtained for individual animals for the first seven stimulus presentations. Both males and hermaphrodites showed habituation, as indicated by A 03 .J 69 rr uJ
20-
MALE HERM
15-
I
LU n," I,,.L.
10'
o n-' ILl m
5-
Z
0 10
20
40
50
60
7
B ILl 69 Z
15°t
----B----
MALE
o
0. 69 LU n-
100
...I m I--
5O
i
10
20
30
40
50
7O
STIMULUS
B 150
HERM
uJ 03 Z
0a. 69 uJ 0c
100
_J I.-
0
•
10
FIG. 4. (A) Number of reversals by males (n = 20) and hermaphrodites (n = 20) to each stimulus during habituation training and following the administration of dishabituating stimulus (60-V shock, represented by the vertical line appearing at immediately after stimulus 60). Both males and hermaphrodites showed habituation, although the number of males reversing to tap was significantly higher than hermaphrodites at almost every tap. Hermaphrodites showed significant dishabituation following shock, whereas males did not. (B) Habituation curve for reversal magnitude of males (n = 20) showing the mean magnitude of reversals 4- SEM at each stimulus presentation. Reversal magnitude is expressed as a percentage of each animal's initial re-
30
STIMULUS
20
.
30
.
40
-T'---.---
50
.
60
70
STIMULUS
versal• Shock administration is represented by a vertical line after stimulus 60. Males showed rapid habituation of response magnitude and did not show dishabituation in response to a 60V shock. (C) Habituation curve for reversal magnitude of hermaphrodites (n = 20) showing the mean magnitude of reversals 4- SEM at each stimulus presentation• Reversal magnitude was standardized as a percentage of the initial reversal of individual animals. Shock administration was represented by a vertical line after stimulus 60. Hermaphrodites showed a rapid decline in response magnitude, reaching asymptote near zero. Significant dishabituation was seen following shock administration.
218
MAH AND RANKIN
negative slopes which were significantly different from zero [Figs. 4B and 4C, males: mean slope = -13.29, SEM = 2.841, t(19) = 5.574, p < .0001; hermaphrodites: mean slope = - 9 . 4 8 9 , SEM = 1.702, t(19) = 5.574, p < .0001]. There were no significant differences in the rate of habituation between sexes as measured by comparing the slopes of the males and the hermaphrodites [t(19) = - 3 . 8 0 3 , p = .1457]. The response to tap 7 was significantly lower than the response to the initial tap [males: t(19) = 7.763, p < .0001; hermaphrodites: t(91) = 3.583, p = .0011], indicating that both sexes showed habituation after 7 taps. Furthermore, the response to tap 7 did not differ between males and hermaphrodites [t(38) = - 0.406, p = .3435]. After 60 stimuli the response magnitude in hermaphrodites was significantly lower than in males. This was determined by comparing the mean response for both males and hermaphrodites for the final 10 stimuli prior to shock administration [males: mean = 23.32, SEM = 5.382; hermaphrodites: mean = 0.692, SEM = 0.165, t(18) = 4.185, p = .0003]. Furthermore, males showed a significantly higher degree of variability than did hermaphrodites in their response magnitudes over the last 10 stimuli [F(20, 20) -- 892.064, p < .0001]. Males did not show significant dishabituation of response magnitude following the shock [t(19) = 0.522, p = .304] whereas hermaphrodites did show dishabituation [t(19) -- 1.188, p = .0376].
Discussion Both males and hermaphrodites showed habituation to repeated stimulation. In an earlier experiment investigating factors affecting habituation and recovery from habituation in hermaphrodite C. elegans Rankin and Broster (1992) suggested that there were two components to habituation, the initial rapid response decrement and the long steady asymptotic response level of the later portion of the curve. In this experiment there were differences between males and hermaphrodites in both components of the habituation curve. Males showed significantly more reversals to the first 7 and to the last 10 stimuli than did hermaphrodites. In addition although there were no sex differences in response magnitude to the first 7 stimuli, males showed significantly larger responses and greater variability to the last 10 stimuli than did the hermaphrodites. Thus, although habituation is present in both sexes, some aspects of the mechanisms governing habituation of this reflex m a y differ between males and hermaphrodites. The finding of no significant dishabituation in
males should be interpreted with caution. Close inspection of the final 10 responses in the habituation series prior to shock administration revealed a large amount of variability in the responsiveness of males (Fig. 4B). It is possible that this high variability in response magnitude overshadowed dishabituation; however, it is also possible that males might lack dishabituation. Research using a larger number of animals might decrease the response variability during the final taps, thereby increasing the probability of determining whether dishabituation can occur in males. EXPERIMENT V: S P O N T A N E O U S RECOVERY FROM H A B I T U A T I O N Following the habituation of a response, the magnitude of the habituated response eventually recovers back to baseline levels in the absence of further habituating stimuli. This process is referred to as spontaneous recovery from habituation (Groves & Thompson, 1970). Rankin and Broster (1992) have shown that hermaphrodite C. elegans recover from habituation to trains of taps at a 10-s ISI within 20 min of the last stimulus. The current experiment explored the temporal parameters that govern spontaneous recovery from habituation resulting from repeated presentation of single taps in males and hermaphrodites.
Subjects and Procedure Twenty male and 20 hermaphroditic 4 days old were used. A series of 60 single taps with an ISI of 10 s was delivered to produce habituation. To test for spontaneous recovery from habituation, a single tap was administered 30 s and 10, 20, and 30 min following the last tap of the habituation series.
Scoring and Statistical Analysis Only animals which showed a reversal to the initial tap were scored and used in the analyses. A response in which an animal accelerated forward was replaced with the group mean. Only responses obtained to the initial tap, the last tap of the habituation series, and the 30-s, 10-, 20-, and 30-min posthabituation taps were analyzed. Proportions of animals reversing were analyzed with one-factor repeated measures ANOVA for dichotomous data. Magnitude data were analyzed with trend and regression analyses.
Results Number of reversals. The number of males responding changed significantly across the treatment [Fig. 5A, F(5, 95) = 8.018, p < .0001, repeated
BEHAVIORAL PLASTICITY IN MALE C. elegans
A ,.,.I
[] []
20
TAP 1 TAP 60
[] []
10MIN 20MIN
OS
N
,>,, n,-
10
P, 6 o HERM
MALE
B
•
TAP 1
[] []
10 MIN 20 MIN IN
1°° t ~
60
,~ V_~
40 20
o
1
MALE
HERM
FIG. 5. (A) Number of males and hermaphrodites (n = 20 for each sex) showing reversals in response to the initial tap, tap 60 (the last tap of the habituation series), 30 s, and 10, 20, and 30 min following the last habituation tap. Significantly fewer animals reversed in response to tap 60 compared to the initial tap in both sexes, indicating significant habituation. Males and hermaphrodites showed significantly more reversals at 30 s and 10, 20, and 30 min posthabituation than to the 60th tap. There were no significant differences between males and hermaphrodites in the number of reversals shown. (B) Magnitude of reversals shown by males and hermaphrodites (n = 20 for each sex) + SEM for reversals in response to the initial tap, tap 60 (the last tap of the habituation series), 30 s, and 10, 20, and 30 rain following the last habituation tap. Both males and hermaphrodites showed recovery from habituation in a linear fashion. There were no sex differences in the rate of spontaneous recovery from habituation.
measures ANOVA on initial tap, final habituation tap, and tap at 30 s, 10, 20, and 30 min]. All males reversed to the initial tap; by the final tap, significantly fewer were showing reversals (p < 0.01). Recovery was evident by 10-min posthabituation: the number of males reversing 10, 20, and 30 min was significantly larger than those reversing at tap 60 (p < .01). The number of hermaphrodites reversing also changed across the treatments [Fig. 5A, F(5, 95) = 15.299, p < .0001]. Significantly fewer worms were reversing at tap 60 than to the initial tap (p < .01). Recovery occurred by 30 s and continued over time, with the number of animals reversing at 30 s, 10
219
min (p < .05), and at 20 and 30 min (p < .01) larger than the number reversing at tap 60. There were no differences between males and hermaphrodites in the number of worms reversing at any of the times tested [F(1, 38) = 1.185, p = .2832].
Reversal magnitude. To compare magnitude data across recovery for males and hermaphrodites, the reversal magnitudes were standardized by expressing each worm's response as a percentage of its initial response to tap, which was set at 100%. Male C. elegans showed habituation to repeated taps, as the response to the final tap was significantly smaller than the response to the initial tap [t(19) = 83.918, p < .0001]. Spontaneous recovery from habituation in males occurred in a positive linear fashion over the 30-min posthabituation period [F(1, 57) = 10.338, p < .01; for trend analysis including tap 60, 10-, 20-, and 30-min taps]. Also, there was significant recovery above habituated levels, since the mean slope for the regression line in the linear trend was significantly different than zero [Fig. 5B, t(15) -- - 2 . 7 4 5 , p = .0075]. In hermaphrodites, habituation and significant recovery from habituation were also evident [Fig. 5B, initial tap vs final tap response: t(16) = - 2 . 5 1 1 , p = .0116; positive linear trend, F(1, 57) = 6.727, p < .025; slope of regression line, t(16) = - 2 . 5 1 1 , p = .0116]. There were no significant differences between sexes in the rate of recovery from habituation, shown by comparing the mean slope of the regression line for tap 60, and the 10-, 20-, and 30-min tap [males: mean slope = 25.963, SEM = 9.459; hermaphrodite: mean slope -- 18.2877, SEM = 7.283; t(13) = 0.526, p = .3038]. Discussion Since there were no significant differences between the males and hermaphrodites in either the response frequency or the rate of recovery as measured by response magnitude, we hypothesize that the processes which govern recovery from habituation are likely to be similar in the two sexes of
C. elegans. GENERAL DISCUSSION The purpose of these experiments was to test whether the presence of additional sensory receptors in the C. elegans males might lead to differences between the males and hermaphrodites in spontaneous behavior, reflexive reversal behavior in response to tap, in competition between responses to tail-touch and tap, and in nonassociative learning.
220
MAH AND RANKIN
Differences between the male and hermaphrodite do exist in spontaneous activity; males showed a larger number of spontaneous reversals of all types. This is consistent with the hypothesis that reversal activity in males may serve to increase contacts with other worms to initiate mating. Both sexes showed graded responses to increasing tap intensity; however, males demonstrated larger reversal magnitudes than hermaphrodites to all tap intensities. This suggested a possible bias of the touch circuit toward the head-touch-backwardmovement portion of the system. However, when the tail-touch and head-touch circuits were put into opposition in tests of inhibition, males and hermaphrodites showed the same amount of inhibition of reversal frequency and magnitude at the 1- and 5-s tail-touch/tap intervals. In addition, there were no significant sex differences in the amount of inhibition produced by tail-touch at any of the tailtouch/tap intervals. Taken together these data suggest that the additional sensory receptors in the tails of males do not greatly change the organisms response to tap, and we thus hypothesize that the circuit underlying the touch and tap withdrawal responses in hermaphrodites (Chalfie et al., 1985) exists in the same or similar form in the males, and that tail-touch activates the same neural circuits leading to inhibition of reversal to tap in both sexes. Both males and hermaphrodites showed habituation to repeated single taps of 10 s ISI. The rate of habituation did not differ between the two sexes in either the habituation or recovery experiments; however, the final level of habituation and the variance associated with that level were significantly higher in males than in hermaphrodites. Thus, males appeared to habituate, but seemed to be more likely to respond at each of the habituation trials and to respond with larger reversals than hermaphrodites. This suggests that although habituation occurs in all worms there are subtle differences in both the descending portion of the habituation curve and the asymptotic portion of the habituation curve of males. It is conceivable that the larger number and magnitude of reversals displayed by males at asymptote are related to the importance of tactile stimulation in the mating behavior of male C. elegans. If the animal were completely habituated to tactile stimulation, whether from a tap or from contact with another worm, very few or no reversals would occur. The absence of reversal activity might prevent, or at the very least seriously hamper, successful mating in males. Protecting the touch circuit from complete habituation would allow the male to continue to reverse, thereby affording the male the opportunity to investigate stimulation of the pos-
terior sensory receptors that might be indicative of a potential mating partner. We have suggested that the lack of dishabituation in males may have been due to the high variability of the 10 responses preceding the administration of the shock. Dishabituation is often considered evidence that habituation is not the result of sensory adaptation or motor fatigue (Groves & Thompson, 1970). Since dishabituation was not shown by males, it is possible that the response decrement shown by males might simply be the result of sensory adaptation or muscular fatigue. If fatigue or sensory adaptation is the cause of response decrement in males while learning is the cause of response decrement in hermaphrodites, then the two sexes would be unlikely to show the same rates of recovery following habituation training. However, Experiment V demonstrated the same rate of recovery from habituation in males and hermaphrodites; therefore, it is unlikely that entirely different processes underlie the decrement in responding seen in the male and the hermaphrodite. In conclusion, our behavioral observations support the hypothesis that the neural circuit underlying touch-withdrawal in hermaphrodites is also present in male C. elegans. The results also suggest that although the male responds similarly to hermaphrodites, showing habituation, inhibition, etc., the subtle differences in behavior are consistent with male mating behavior and the functional role of the additional tail structures in male mating.
REFERENCES Beck, C. D. O. (1991). The effects of aging on non-associative learning in the nematode Caenorhabditis elegans. Unpublished master's thesis, University of British Columbia, Vancouver. Beck, C. D. O., and Rankin, C. H. (in press). Effects on aging on non-associative learning in the nematode C. elegans. Behavioural Processes. Brenner, S.T. (1974). The genetics of Caenorhabditis elegans. Genetics, 77, 71-94. Byerly, L., Cassada, R. C., & Russell, R. L. (1976). The life cycle of the nematode Caenorhabditis elegans. Developmental Biology, 51, 23-33. Chalfie, M., Sulston, J. E., White, J. G., Southgate, E., Thomson, J.N., & Brenner, S. (1985). The neural circuit for touch sensitivity in Caenorhabditis elegans. The Journal of Neuroscience, 5, 956-964. Chalfie, M., and White, J. G. (1988). The nervous system. In W. B. Wood (Ed.), The nematode Caenorhabditis elegans (pp. 337391). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Chiba, C. M., & Rankin, C. H. (1990). A developmental analysis of spontaneous and reflexive reversals in the nematode Caenorhabditis elegans. Journal of Neurobiology, 21, 543-554.
BEHAVIORAL PLASTICITY IN MALE C. elegans Groves, P. M., & Thompson, R. F. (1970). Habituation: A dualprocess theory. Psychological Review, 77, 419-450. Hodgkin, J. (1983). Male phenotypes and mating efficiency in Caenorhabditis elegans. Genetics, 103, 43-64. Hodgkin, J. (1988). Sexual dimorphism and sex determination. In W. B. Wood (Ed.), The nematode Caenorhabdit~s elegans (pp. 243-279). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Hodgkin, J., Horvitz, H. R., & Brenner, S. (1979). Nondisjunction mutants of the nematode Caenorhabd~tis elegans. Genetics, 91, 67-94. Kumar, N., Williams, M., Culotti, J., and van der Kooy, D. (1989). Evidence for associative learning in the nematode C. elegans. Society for Neuroscience Abstracts, 15, 1141. Rankin, C.H. (1991). Interaction between two antagonistic reflexes in Caenorhabditis elegans. Journal of Comparative Physiology A, 169, 59-67. Rankin, C. H., & Chiba, C. M. (1988). Short and long term learning in the nematode C. elegans. Society for Neuroscience Abstracts, 14, 607. Rankm, C. H., and Chalfie, M. (1989). Analysis of non-associative
221
learning in C. eIegans: I. Neural circmt mutations. Society for Neuroscience Abstracts, 15, 1118. Rankin, C.H., Beck, C. D. O., & Chiba, C.M. (1990). Caenorhabd~tis elegans: A new model system for the study of learning and memory. Behavioral Brain Research, 37, 89-92. Rankin, C. H., & Broster, B. S. (1992). Factors affecting habituation and recovery from habituation in the nematode Caenorhabditis elegans. Behavioral Neuroscience, 106, 239-249. Sulston, J. E., Albertson, D. G., & Thomson, J. N. (1980). Postembryonic development of nongonadal structures. Developmental Biology, 78, 542-576. Wicks, S., and Rankin, C.H. [1991). Circuit analysis of interactions between two antagonistic reflexes in C. elegans. Society for Neuroscience Abstracts, 17, 1055. Winer, B. J., Brown, D. R., & Michels, K. M. (1991). Statistical Principles ~n Experimental Design (Third ed.). New York: McGraw-Hill. Wood, W. B. (1988). Introduction to C. elegans biology. In W. B. Wood (Ed.), The nematode Caenorhabditis elegans (pp. 116). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
An Analysis of Behavioral Plasticity in Male Caenorhabditis elegans K I M BILL M A H AND CATHARINE H . R A N K I N 1
Department of Psychology, University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z4
ulation and observation make C. elegans an excellent model system in which to study the relationships between neurons and behavior. C. elegans also has a rich repertoire of behavioral plasticity (Rankin & Chiba, 1988; Rankin, Beck, & Chiba, 1990; Kumar, Williams, Culotti, & van der Kooy, 1989). Rankin et al. (1990) have shown that the tap withdrawal reflex (in response to a tap to the dish holding the worm) in C. elegans exhibits the major forms of nonassociative learning: habituation, dishabituation, and sensitization. The neural circuit for tail-touch-induced forward swimming and head-touch-induced backward swimming in the hermaphrodite, described by Chalfie, Sulston, White, Southgate, Thomson, and Brenner (1985), has been shown to underlie the tap withdrawal reflex as well (Rankin & Chalfie, 1989; Wicks & Rankin, 1991). Thus, a great deal is known about the C. elegans hermaphrodite; however, C. elegans also occurs infrequently as a male as a result of X-chromosome nondisjunction (Hodgkin, Horvitz, & Brenner, 1979). Males do not produce eggs and thus must mate with hermaphrodites in order to reproduce. To that end, males have specialized neuroanatomy, sensory organs, and behavior. The male has a larger complement of somatic nuclei than the hermaphrodite: 1031 in total, 72 more than the hermaphrodite (Hodgkin, 1983, 1988). Most of these additional cells are accounted for by the existence of the copulatory apparatus of the tail region, including the vas deferens, the cloaca, and the sclerotic hook. In addition to having the two-tail sensilla possessed by the hermaphrodite (Sulston, Albertson, & Thomson, 1980), the male C. elegans can receive additional sensory input from the tail via 32 additional sensory neurons associated with the various structures that make up the specialized male tail (Wood, 1988). In order for males to reproduce they must mate; the additional structures in the tails of male C. elegans appear to be associated with specialized
Caenorhabditis elegans is a simple soil-dwelling nematode which has two sexes, hermaphrodite and male. The male C. elegans is differentiated from the hermaphrodite by the presence of 14 sensory structures in the tail. In this study, we compared the behavioral responses of males and hermaphrodites to head-touch and to tap. We hypothesized that the anatomical difference in sensory structures might result in behavioral differences in the reversal response to vibratory stimulation (a tap to the side of the holding dish). In the response to increasing intensities of tap, both sexes showed an increase in response magnitude, with the males showing larger responses than hermaphrodites. In addition, the male was shown to be capable of simple nonassociative learning: it demonstrated habituation and recovery from habituation in a similar manner as the hermaphrodite. Tail-touchinduced inhibition of the reversal response appeared to be similar in males and hermaphrodites. The evidence suggests that the touch withdrawal circuit in hermaphrodites is also present in the male C. elegans, and that the subtle differences in response to tap seen in males may result from the additional sensory receptors of the copulatory bursa of the tail. It seems clear from these studies that these structures do not play a key role in the male worm's response to tap. ©1992AcademicPress, Inc. The nematode Caenorhabditis elegans provides a powerful model system for the study of the relationship between behavior and its neural underpinnings. C. elegans possesses a simple, well-mapped nervous system of 302 neurons, a small genome which is currently the object of intensive mapping studies, and a short reproductive cycle, developing from egg to adult in approximately 31/2 to 4 days at 20°C. These characteristics, coupled with a behavioral repertoire amenable to experimental manip1 The authors acknowledge Catherine Chiba and two anonymous reviewers for editing assistance. The research was funded, in part, by a n NSERC operating g r a n t to C.H.R. Address correspondence and reprint requests to C a t h a r i n e H. Rankin, Dept. of Psychology, University of British Columbia, Vancouver, BC, Canada V6T 124. 211
0163-1047/92 $5.00 Copyright © 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
212
MAH AND RANKIN
mating behaviors. Mating behavior begins when a male contacts a hermaphrodite with its tail; the male then rubs the copulatory bursa on the side of the hermaphrodite, moving backward until it contacts the vulva. The male attaches to the hermaphrodite using the spicules, opens the vulva, and ejaculates sperm into it, and then it removes the spicules and swims away (Chalfie & White, 1988). The neuroanatomy and function of the touch circuit in males has not been examined; however, males respond to head-touch with backward movement and tail-touch with forward movement as do hermaphrodites. Although this suggests that the underlying neural circuit is similar in both sexes, the additional sensory structures in the male tail may interact with elements of the touch circuit. Thus, it is possible that the differences in sensory structures between males and hermaphrodites may result in subtle differences in response to vibrational stimulus such as a tap and perhaps differences in nonassociative learning. Using hermaphrodites Chiba and Rankin (1990) showed that although there were no changes in behavior in the touch response during development, there were changes in the response to a tap. Thus, response to tap may be a more sensitive measure of mechanosensory organization than response to touch. The following experiments were designed (1) to examine whether males responded to touch and tap the same as hermaphrodites, (2) to compare the plasticity of the tap response behaviors mediated by the touch circuit in male and hermaphroditic C. elegans, and (3) to provide clues as to the nature of the touch circuit in the male. METHODS
Subjects In all experiments, C. elegans hermaphrodites were controls while male C. elegans were experimental subjects. A total of 200 hermaphrodites and 200 males (strain N2 Bristol) were used. Age-synchronized colonies were set up; all worms were tested at 4 days of age (adults; Byerly, Cassada, & Russell, 1976).
Materials All animals were maintained at 20°C on 5-cm petri plates filled with Nematode Growth Medium (Brenner, 1974) and streaked with Escherichia coli (strain OP50) as a source of food for C. elegans. For testing, individual animals were transferred to unstreaked (i.e., no food present) agar-filled plates. In
all cases, worms were transferred to the test plate a minimum of 1 to 2 min prior to experimental manipulation.
Apparatus The apparatus used was the same as that described in Chiba and Rankin (1990). Test plates were viewed using a Wild M3Z stereomicroscope with an attached Panasonic D5000 high-resolution video camera and Panasonic video cassette recorder. Taps were delivered to the side of the test plate using an electromagnetically controlled tapping arm. Shocks were administered with a spanning electrode consisting of two wires placed in the agar, one on either side of the animal. Both the tap and shock parameters were controlled by a Grass S88 stimulator.
Scoring and Statistical Analysis Using stop-frame video analysis, the reversal responses of each animal were traced onto acetate sheets. Measurements of the magnitude (length) of the tracing (and hence of the reversal) were rendered on an Apple Macintosh SE computer using a Bit Pad Plus digitizing tablet and MacMeasure software. Data involving numbers of animals within groups showing reversals were analyzed using an overall ANOVA with dichotomous data (Winer, Brown, & Michels, 1991). Data involving magnitude information were analyzed using overall ANOVAs with post hoc analyses as warranted. Trend analysis was used to analyze magnitude data obtained in response to stimuli of graded intensity. Regression analysis was employed to test rate measures (i.e., rates of habituation or recovery from habituation). Repeated measures data were analyzed with a repeated measures ANOVA. EXPERIMENT I: SPONTANEOUS REVERSALS Although the reversal response (changing from forward to backward swimming) is seen in response to stimuli such as taps or tail touch, both male and hermaphroditic C. elegans also reverse in the absence of such applied stimulation. We have defined such reversals as spontaneous reversals. Chiba and Rankin (1990) demonstrated that adult C. elegans hermaphrodites show two to three spontaneous reversals/minute. Since the male finds and mates with the hermaphrodite by touching its tail to the hermaphrodite's body, it was predicted that, to increase the probability of mating, males would show more spontaneous reversals than hermaphrodites.
213
BEHAVIORAL PLASTICITY IN MALE C. elegans
This experiment was designed to test whether males show larger and/or more frequent spontaneous reversals than to hermaphrodites.
Subjects and Procedure Twenty male and 20 hermaphroditic 4-day-old C.
elegans were used in this experiment. Worms were transferred to test plates at least 1 min prior to beginning the experiment. A 10-min period of spontaneous activity was then recorded.
A oo ,,,.I
30
O0 n" ILl
20
U.I n'LI. 0
lo
[]
TRUESPONTANEOUS
[]
SELF-CONTACT
Z I.g
=E
o MALE
HERM
Scoring and Statistical Analyses Videotapes were scored for total number and magnitude of reversals during the 10-min observation period; data were analyzed using overall ANOVAs. On average, males are smaller than hermaphrodites of the same age. To ensure that differences in reversal magnitude were not due to an effect of worm length, the magnitude of each reversal was standardized by dividing the reversal length by the length of that particular worm (Chiba & Rankin, 1990; Beck, 1991).
B W a I,m Z
1.0
Number of reversals. An examination of the videotaped behavior revealed that males showed two types of "spontaneous" reversals: (1) self-contact reversals and (2) true spontaneous reversals resembling those shown by hermaphrodites. A self-contact reversal was defined as a reversal in which the tail of the animal made contact with its own body (described in Chalfie & White, 1988). Self-contact reversals accounted for approximately 19% of the total number of reversals observed in males while accounting for less than 3% of reversals shown by hermaphrodites. Overall, males showed a significantly greater number of both true spontaneous reversals [Fig. 1A, F(1, 38) = 8.35, p < .01] and selfcontact reversals [Fig. 1B, F(1, 38) = 13.69, p < .001] than did hermaphrodites. This effect was not due to differences in activity level, since there were no significant differences between males and hermaphrodites in time active as measured by the percentage of total time (over 10 min) swimming forward and backward, [(male mean = 99.875, SEM = 0.091), (hermaphrodite mean = 99.058, SEM = 0.663); t(19) = 1.209, p = .1207]. Reversal magnitude. There were no significant differences between males and hermaphrodites in the magnitudes of true spontaneous reversals, [F(1, 38) = 0.34, p = .5639]. However, the magnitude of self-contact reversals was significantly larger in
TRUE SPONTANEOUS
[]
SELF-CONTACT
0.8
0.8
=E ,,.I
0.4
O~ n,. U.I
0,2
U.I
n-
0.0
MALE
Results
[]
HERM
FIG. 1. (A) Mean number of true and self-contact spontaneous reversals + SEM shown by males (n = 20) and hermaphrodites (n = 20) over a 10-min observation period. Males showed significantly more true and self-contact spontaneous reversals than hermaphrodites. (B) Magnitude (distance traveled/body length) of true and self-contact reversals + SEM shown by males (n = 20) and hermaphrodites (n = 20) over a 10-min observation period. The magnitude of true spontaneous reversals did not differ; however, the magnitude of self-contact reversals was significantly higher in males.
males than in hermaphrodites [F(1, 38) = 8.27, p < .01].
Discussion Males show significantly more true spontaneous reversals and significantly more self-contact reversals than hermaphrodites. In self-contact reversals males reverse in response to their tails touching their own bodies; they then press the entire fan-like tail structure firmly against their cuticles and rub it along the body vigorously for several seconds (this is similar to the mating behavior when a hermaphrodite is contacted). In contrast, when the tail of the hermaphrodite contacts its own body, it only briefly brushes its cuticle with its tail. Thus, it appears that the self-contact reversal of the male m a y be an active attempt to mate while that of the hermaphrodite is more perfunctory in nature. In contrast to self-contact reversals, true spon-
214
MAH AND RANKIN
taneous reversals occurred with the same magnitude in both sexes, although with greater frequency in the males. By reversing more frequently males may increase the probability of bringing their tail copulatory structures into contact with hermaphrodites, and thus increase the likelihood of successful mating.
A ,-I
20
,< UJ W n" u.
15
10
0 W
5
w
EXPERIMENT II: RESPONSE TO GRADED TAP INTENSITY
Z
0
MALE
Adult hermaphrodites reverse direction and swim backward following a tap to the side of the plate and show a graded response to vibrational stimuli of different intensities, showing larger reversals to taps of larger intensities Beck and Rankin (in press). This experiment was designed to uncover differences in responses to vibrational stimuli between males and hermaphrodites by comparing their responses to stimuli of varying intensity.
• []
WEAK INTERMEDIATE
[]
STRON
1
HERM
B u.I a .-j IwZ
1.5-
•
WEAK
[]
INTERMEDIATE
1.0
.J ¢j)
O.5'
W
Subjects and Procedure Single worms were placed on individual unstreaked test plates approximately 2 h prior to testing. The force of the tap was varied by altering the voltage to the electromagnetic relay controlling the tapper. To produce a weak tap, a 34-V intensity was used, to produce an intermediate tap a 40-V intensity was used, while to produce a strong tap a 60V intensity was used. Twenty males and 20 hermaphrodites 4 days old were tested at each tap intensity.
Scoring and Statistical Analyses The number of animals in each group showing reversals to the differing intensities of tap and the magnitude of those reversals that did occur were scored. The number of worms reversing was analyzed using a two-factor ANOVA. Between-group and within-group differences were analyzed with the Least-Square Means method. Within group magnitude data were analyzed using an overall twofactor ANOVA.
Results Number of reversals. As tap intensity increased, the number of reversals shown by both males and hermaphrodites increased [Fig. 2A, F(2, 114) = 7.171, p = .0012]. Worms showed significantly fewer reversals to weak taps than to intermediate or strong taps (p < .05). Furthermore, there were no
KI
n"
0.0
MALE
HERM
FIG. 2. (A) The number of males (n = 20) and hermaphrodites (n = 20) showing reversals to weak, intermediate, and strong taps. Among both males and hermaphrodites significantly fewer worms reversed to a weak tap, whereas most worms reversed to intermediate and strong taps. There were no sex differences in the number of worms reversing at any tap intensity. (B) Mean reversal magnitude (distance traveled/body length) + SEM of males (n = 20) and hermaphrodites (n = 20) in response to taps of weak, intermediate, and strong intensity. Reversal magnitude for each animal was standardized by dividing the reversal observed by the length of the worm showing that reversal. Both males and hermaphrodites showed larger reversals as tap intensity increased. Response magnitudes were significantly higher in males than those in hermaphrodites at each tap intensity.
differences between males and hermaphrodites in number of reversals at different tap intensities [F(2, 114) = 0.222, p = .8014].
Reversal magnitude. Animals of both sexes responded to increases in tap intensity with increasing reversal magnitudes [Fig. 2B, F(2, 114) = 4.957, p = .0086]. At all tap intensities, response magnitudes were greater in males than in hermaphrodites, [F(1, 114) = 11.983, p = .0008]. As tap intensity increased, the increase in reversal magnitude shown by males paralleled that shown by hermaphrodites, and no significant interactions between sex and tap intensity were observed [F(1, 114) = 0.201, p = .8180].
BEHAVIORAL PLASTICITYIN MALEC. elegans
215
Discussion
Subjects and Procedure
Both males and hermaphrodites showed graded responses to tap and made larger reversals in response to taps of larger intensities. Males, however, showed significantly larger reversals than hermaphrodites at all stimulus intensities tested, suggesting that they were more responsive to tap than were hermaphrodites. The differences between the sexes does not appear to be a simple case of the male tail sensory structures making the male more sensitive to all tactile stimulation of the tail. In contrast to hermaphroditic ~ail-touch receptors which produce forward movement the additional sensory structures in the tail appear to enhance the magnitude of male worm's reversals to tap.
Twenty males and 20 hermaphrodites 4 days old were used. Animals were placed individually on test plates 24 h prior to testing (to replicate Rankin, 1991). A small amount ofE. coli was placed on the test plate to ensure that the subject was not food deprived. Each animal received seven different stimulus conditions, separated by intervals of approximately 15-20 min (adult hermaphrodites show no habituation at a 10-min ISI). The conditions were (1) single tap, (2) tail-touch followed 1 s later by tap, (3) tail-touch followed 5 s later by tap, (4) tail-touch followed 10 s later by tap, (5) tail-touch followed 20 s later by tap, (6) tail-touch followed 30 s later by tap, and (7) single tap alone. Condition i established a baseline level of response to tap alone. Conditions 2-6 tested for the existence of the inhibition effect and explored the temporal parameters of the effect itself. Condition 7 tested whether the response to tap alone showed habituation due to stimuli from conditions 2-6. Since the intent was to study the effect of tail-touch on reversals to tap, worms that did not reverse to the initial tap were not tested (fewer than 10% of worms). Tail-touch was administered by hand by drawing an eyelash mounted to the tip of a 30-gauge needle across the tail of a worm.
EXPERIMENT III: INHIBITION Given that males and hermaphrodites respond to tap with reversals it is likely that the touch circuit responds to tap in roughly the same way in both sexes, with perhaps, as suggested in the previous experiment, enhanced excitation of the reversal circuit in males. Touch is a very different sort of tactile stimulus from tap, although it does activate the same circuit and produce the same types of responses. In contrast to tap, direct touch to the tail of hermaphrodites stimulates only the tail-touch receptors, without stimulating the receptors that would produce the competing response of backward movement. In both males and hermaphrodites tail-touch produces forward swimming. In hermaphrodites the activation of forward swimming by tail-touch inhibits the competing response of reversal to tap (Rankin, 1991). Tail-touch administered 1 s prior to tap decreases both the likelihood and magnitude of reversal to tap. This inhibition of reversal to tap by tail-touch lasted about 10 s and was followed by facilitation of the reversal magnitude to tap above baseline levels. Since the graded response experiment suggested that males might have a bias toward larger reversals, males might show a different pattern of behavior than hermaphrodites when tail-touch and tap are put into competition. The following experiment examines response competition in males and in hermaphrodites in order to assess the relative contributions of head- and tail-touch receptors to the response to tap.
Scoring and Statistical Analyses The number of reversals to tap was examined across sexes to determine between-sex differences. Within each sex, an ANOVA on the number of reversals was compared across increasing tail-touch tap intervals. The magnitude of the tap response was analyzed in each sex with a one-factor repeated measures ANOVA and multiple comparisons with Fisher's PLSD.
Results Number of reversals. All animals reversed to the initial tap. Both males and hermaphrodites showed a large decrease in the number of reversal responses to tap at the 1-s tail-touch/tap interval (p < .01 in both cases). At all other intervals, there were no significant differences in the number of animals reversing when compared to the baseline tap. Additionally, males and hermaphrodites did not differ from each other in the number of reversals shown at any of the tap intervals (Fig. 3A).
216
MAH AND RANKIN A
• []
•
20
n-
lO
MALE HERM
Discussion
P, s ~ z
o TAP 1S 5S 1 0 S 2 0 S 3 0 S TAP TAIL-TOUCH/TAP INTERVAL
200
•
MALE
T
50
100
.~ o~
In both males and hermaphrodites, tail-touch administered 1 s prior to tap resulted in a decrease or inhibition of both the frequency of reversals and the magnitude of reversals at short touch/tap intervals. EXPERIMENT IV: H A B I T U A T I O N AND DISHABITUATION
B
/~
but nonsignificant trend of facilitation 20 and 30 s following tail-touch. The sexes did not differ significantly from one another at any of the tailtouch/tap intervals in the magnitude of reversals to tap [F(1, 38) = 0.929, p = .3413].
so o TAP 1S 5S lOS TAIL-TOUCH/TAP
2OS 3OS TAP INTERVAL
FIG. 3. (A) Number of males (n = 20) and hermaphrodites (n = 20) showing reversals to tap and to tap following tail-touch at intervals of 1, 5, 10, 20, and 30 s. Both males and hermaphrodites show a significant decrease in the number of reversals to tap at the 1-s tail-touch/tap interval. (B) Mean magnitude of reversals + SEM of males (n = 20) and hermaphrodites (n = 20) to tap and tail-touch/tap intervals of 1, 5, 10, 20, and 30 s. Reversal magnitude is expressed as a percentage of each animals reversal to the initial tap (which was set at 100%). Males and hermaphrodites showed a significant decrease in reversal magnitude to tap at the 1- and 5-s tail-touch/tap interval. Hermaphrodites showed significant facilitation of reversal magnitude to tap at the 20- and 30-s tail-touch/tap interval and at the final tap. Males did not show significant facilitation.
Reversal magnitude.
To directly compare inhibition between males and hermaphrodites, reversal magnitude was standardized by expressing each worm's response as a percentage of its initial response to tap, which was set at 100%. Both males and hermaphrodites made significantly smaller reversals to taps 1 and 5 s following tail-touch than they did to taps alone [males: F(6, 114) = 13.814, p < .001, Fisher's PLSD: p < .01 and p < .05, respectively; hermaphrodites: F(6, 114) = 15.844, p < .001, Fisher's PLSD: p < .01 in both cases). Hermaphrodites also showed facilitation of reversal magnitude to tap 20 and 30 s following tail-touch when compared to reversal magnitude to tap alone (p < .01, Fig. 3B). Males showed a similar
Habituation is defined as a decreased response to repeated stimulation that is not due to sensory or motor fatigue. Dishabituation is the facilitation of the habituated response due to the presentation of a novel or noxious stimulus (Groves & Thompson, 1970). Both habituation and dishabituation of the reversal response have been demonstrated in C. elegans hermaphrodites (Rankin & Chiba, 1988; Rankin et al., 1990; Rankin & Broster, 1992). The current experiment was designed to test whether male C. elegans were capable of habituation and dishabituation and to compare their response patterns to those of hermaphrodites.
Subjects and Procedure Twenty male and 20 hermaphroditic 4-day-old worms were used. Sixty single taps with an interstimulus interval (ISI) of 10 s were delivered followed by the administration of a 60-V shock within 10 s of the last tap. Within 10-20 s of the shock, an additional series of 10 taps with an ISI of 10 s was administered.
Scoring and Statistical Analysis Only animals which showed a reversal to the initial tap were scored (greater than 90% of all anireals tested). A response in which an animal responded to the tap by moving forward more rapidly than before was replaced with the group mean; this was because the relationship between reversals and such infrequent accelerations is unclear. This replacement did not have a large impact on statistical analyses since fewer than 1% of data points were replaced. The responses to the first seven stimuli were analyzed with a repeated measures ANOVA (number of reversals) or a regression analysis (magnitude data) to determine whether habituation was present and, if so, to determine its rate. The final
217
BEHAVIORAL PLASTICITY IN MALE C. elegans
10 stimuli prior to the administration of the dishabituating stimulus (shock) were analyzed with a repeated measures ANOVA (number of reversals) to compare the final levels of habituation reached between the sexes. Finally, dishabituation was tested by averaging the magnitude of the last 10 responses and comparing this mean to the response immediately following the shock using a paired t test. The last 10 responses were used to provide a better estimate of preshock responding than would be obtained by a single point.
Results Number of reversals. The number of males and hermaphrodites reversing to tap decreased significantly over the course of the first seven taps [F(6, 228) = 5.565, p = .0001]. Although males and hermaphrodites showed a similar pattern of decrement in the number of reversals, over the first seven taps males responded with a greater number of reversals than did hermaphrodites [Fig. 4A, F(1, 38) = 4.163, p = .O483. During the last 10 stimulus presentations preceding the shock, the number of reversals in both sexes had stabilized, with males showing a significantly higher number of reversals than hermaphrodites [F(1, 38) = 22.537, p = .0001]. To determine whether the shock increased the number of reversals shown over habituated levels, the mean number of reversals to the last 10 stimuli prior to shock was compared to the number of reversals immediately following the shock. Following shock there was an increase in the number of males reversing; however, this increase just missed significance [#19) = 1.612, p = .06]. The hermaphrodites did show a significant increase in the number of worms reversing, and thus expressed significant dishabituation [t(19) = 2.398, p = .0134].
Reversal magnitudes.
To compare habituation and dishabituation in males and hermaphrodites, reversal magnitude was standardized by expressing
each worm's response as a percentage of its initial response to tap, which itself was set at 100%. The rate of response decrement was determined by computing the mean of the slopes of regression lines obtained for individual animals for the first seven stimulus presentations. Both males and hermaphrodites showed habituation, as indicated by A 03 .J 69 rr uJ
20-
MALE HERM
15-
I
LU n," I,,.L.
10'
o n-' ILl m
5-
Z
0 10
20
40
50
60
7
B ILl 69 Z
15°t
----B----
MALE
o
0. 69 LU n-
100
...I m I--
5O
i
10
20
30
40
50
7O
STIMULUS
B 150
HERM
uJ 03 Z
0a. 69 uJ 0c
100
_J I.-
0
•
10
FIG. 4. (A) Number of reversals by males (n = 20) and hermaphrodites (n = 20) to each stimulus during habituation training and following the administration of dishabituating stimulus (60-V shock, represented by the vertical line appearing at immediately after stimulus 60). Both males and hermaphrodites showed habituation, although the number of males reversing to tap was significantly higher than hermaphrodites at almost every tap. Hermaphrodites showed significant dishabituation following shock, whereas males did not. (B) Habituation curve for reversal magnitude of males (n = 20) showing the mean magnitude of reversals 4- SEM at each stimulus presentation. Reversal magnitude is expressed as a percentage of each animal's initial re-
30
STIMULUS
20
.
30
.
40
-T'---.---
50
.
60
70
STIMULUS
versal• Shock administration is represented by a vertical line after stimulus 60. Males showed rapid habituation of response magnitude and did not show dishabituation in response to a 60V shock. (C) Habituation curve for reversal magnitude of hermaphrodites (n = 20) showing the mean magnitude of reversals 4- SEM at each stimulus presentation• Reversal magnitude was standardized as a percentage of the initial reversal of individual animals. Shock administration was represented by a vertical line after stimulus 60. Hermaphrodites showed a rapid decline in response magnitude, reaching asymptote near zero. Significant dishabituation was seen following shock administration.
218
MAH AND RANKIN
negative slopes which were significantly different from zero [Figs. 4B and 4C, males: mean slope = -13.29, SEM = 2.841, t(19) = 5.574, p < .0001; hermaphrodites: mean slope = - 9 . 4 8 9 , SEM = 1.702, t(19) = 5.574, p < .0001]. There were no significant differences in the rate of habituation between sexes as measured by comparing the slopes of the males and the hermaphrodites [t(19) = - 3 . 8 0 3 , p = .1457]. The response to tap 7 was significantly lower than the response to the initial tap [males: t(19) = 7.763, p < .0001; hermaphrodites: t(91) = 3.583, p = .0011], indicating that both sexes showed habituation after 7 taps. Furthermore, the response to tap 7 did not differ between males and hermaphrodites [t(38) = - 0.406, p = .3435]. After 60 stimuli the response magnitude in hermaphrodites was significantly lower than in males. This was determined by comparing the mean response for both males and hermaphrodites for the final 10 stimuli prior to shock administration [males: mean = 23.32, SEM = 5.382; hermaphrodites: mean = 0.692, SEM = 0.165, t(18) = 4.185, p = .0003]. Furthermore, males showed a significantly higher degree of variability than did hermaphrodites in their response magnitudes over the last 10 stimuli [F(20, 20) -- 892.064, p < .0001]. Males did not show significant dishabituation of response magnitude following the shock [t(19) = 0.522, p = .304] whereas hermaphrodites did show dishabituation [t(19) -- 1.188, p = .0376].
Discussion Both males and hermaphrodites showed habituation to repeated stimulation. In an earlier experiment investigating factors affecting habituation and recovery from habituation in hermaphrodite C. elegans Rankin and Broster (1992) suggested that there were two components to habituation, the initial rapid response decrement and the long steady asymptotic response level of the later portion of the curve. In this experiment there were differences between males and hermaphrodites in both components of the habituation curve. Males showed significantly more reversals to the first 7 and to the last 10 stimuli than did hermaphrodites. In addition although there were no sex differences in response magnitude to the first 7 stimuli, males showed significantly larger responses and greater variability to the last 10 stimuli than did the hermaphrodites. Thus, although habituation is present in both sexes, some aspects of the mechanisms governing habituation of this reflex m a y differ between males and hermaphrodites. The finding of no significant dishabituation in
males should be interpreted with caution. Close inspection of the final 10 responses in the habituation series prior to shock administration revealed a large amount of variability in the responsiveness of males (Fig. 4B). It is possible that this high variability in response magnitude overshadowed dishabituation; however, it is also possible that males might lack dishabituation. Research using a larger number of animals might decrease the response variability during the final taps, thereby increasing the probability of determining whether dishabituation can occur in males. EXPERIMENT V: S P O N T A N E O U S RECOVERY FROM H A B I T U A T I O N Following the habituation of a response, the magnitude of the habituated response eventually recovers back to baseline levels in the absence of further habituating stimuli. This process is referred to as spontaneous recovery from habituation (Groves & Thompson, 1970). Rankin and Broster (1992) have shown that hermaphrodite C. elegans recover from habituation to trains of taps at a 10-s ISI within 20 min of the last stimulus. The current experiment explored the temporal parameters that govern spontaneous recovery from habituation resulting from repeated presentation of single taps in males and hermaphrodites.
Subjects and Procedure Twenty male and 20 hermaphroditic 4 days old were used. A series of 60 single taps with an ISI of 10 s was delivered to produce habituation. To test for spontaneous recovery from habituation, a single tap was administered 30 s and 10, 20, and 30 min following the last tap of the habituation series.
Scoring and Statistical Analysis Only animals which showed a reversal to the initial tap were scored and used in the analyses. A response in which an animal accelerated forward was replaced with the group mean. Only responses obtained to the initial tap, the last tap of the habituation series, and the 30-s, 10-, 20-, and 30-min posthabituation taps were analyzed. Proportions of animals reversing were analyzed with one-factor repeated measures ANOVA for dichotomous data. Magnitude data were analyzed with trend and regression analyses.
Results Number of reversals. The number of males responding changed significantly across the treatment [Fig. 5A, F(5, 95) = 8.018, p < .0001, repeated
BEHAVIORAL PLASTICITY IN MALE C. elegans
A ,.,.I
[] []
20
TAP 1 TAP 60
[] []
10MIN 20MIN
OS
N
,>,, n,-
10
P, 6 o HERM
MALE
B
•
TAP 1
[] []
10 MIN 20 MIN IN
1°° t ~
60
,~ V_~
40 20
o
1
MALE
HERM
FIG. 5. (A) Number of males and hermaphrodites (n = 20 for each sex) showing reversals in response to the initial tap, tap 60 (the last tap of the habituation series), 30 s, and 10, 20, and 30 min following the last habituation tap. Significantly fewer animals reversed in response to tap 60 compared to the initial tap in both sexes, indicating significant habituation. Males and hermaphrodites showed significantly more reversals at 30 s and 10, 20, and 30 min posthabituation than to the 60th tap. There were no significant differences between males and hermaphrodites in the number of reversals shown. (B) Magnitude of reversals shown by males and hermaphrodites (n = 20 for each sex) + SEM for reversals in response to the initial tap, tap 60 (the last tap of the habituation series), 30 s, and 10, 20, and 30 rain following the last habituation tap. Both males and hermaphrodites showed recovery from habituation in a linear fashion. There were no sex differences in the rate of spontaneous recovery from habituation.
measures ANOVA on initial tap, final habituation tap, and tap at 30 s, 10, 20, and 30 min]. All males reversed to the initial tap; by the final tap, significantly fewer were showing reversals (p < 0.01). Recovery was evident by 10-min posthabituation: the number of males reversing 10, 20, and 30 min was significantly larger than those reversing at tap 60 (p < .01). The number of hermaphrodites reversing also changed across the treatments [Fig. 5A, F(5, 95) = 15.299, p < .0001]. Significantly fewer worms were reversing at tap 60 than to the initial tap (p < .01). Recovery occurred by 30 s and continued over time, with the number of animals reversing at 30 s, 10
219
min (p < .05), and at 20 and 30 min (p < .01) larger than the number reversing at tap 60. There were no differences between males and hermaphrodites in the number of worms reversing at any of the times tested [F(1, 38) = 1.185, p = .2832].
Reversal magnitude. To compare magnitude data across recovery for males and hermaphrodites, the reversal magnitudes were standardized by expressing each worm's response as a percentage of its initial response to tap, which was set at 100%. Male C. elegans showed habituation to repeated taps, as the response to the final tap was significantly smaller than the response to the initial tap [t(19) = 83.918, p < .0001]. Spontaneous recovery from habituation in males occurred in a positive linear fashion over the 30-min posthabituation period [F(1, 57) = 10.338, p < .01; for trend analysis including tap 60, 10-, 20-, and 30-min taps]. Also, there was significant recovery above habituated levels, since the mean slope for the regression line in the linear trend was significantly different than zero [Fig. 5B, t(15) -- - 2 . 7 4 5 , p = .0075]. In hermaphrodites, habituation and significant recovery from habituation were also evident [Fig. 5B, initial tap vs final tap response: t(16) = - 2 . 5 1 1 , p = .0116; positive linear trend, F(1, 57) = 6.727, p < .025; slope of regression line, t(16) = - 2 . 5 1 1 , p = .0116]. There were no significant differences between sexes in the rate of recovery from habituation, shown by comparing the mean slope of the regression line for tap 60, and the 10-, 20-, and 30-min tap [males: mean slope = 25.963, SEM = 9.459; hermaphrodite: mean slope -- 18.2877, SEM = 7.283; t(13) = 0.526, p = .3038]. Discussion Since there were no significant differences between the males and hermaphrodites in either the response frequency or the rate of recovery as measured by response magnitude, we hypothesize that the processes which govern recovery from habituation are likely to be similar in the two sexes of
C. elegans. GENERAL DISCUSSION The purpose of these experiments was to test whether the presence of additional sensory receptors in the C. elegans males might lead to differences between the males and hermaphrodites in spontaneous behavior, reflexive reversal behavior in response to tap, in competition between responses to tail-touch and tap, and in nonassociative learning.
220
MAH AND RANKIN
Differences between the male and hermaphrodite do exist in spontaneous activity; males showed a larger number of spontaneous reversals of all types. This is consistent with the hypothesis that reversal activity in males may serve to increase contacts with other worms to initiate mating. Both sexes showed graded responses to increasing tap intensity; however, males demonstrated larger reversal magnitudes than hermaphrodites to all tap intensities. This suggested a possible bias of the touch circuit toward the head-touch-backwardmovement portion of the system. However, when the tail-touch and head-touch circuits were put into opposition in tests of inhibition, males and hermaphrodites showed the same amount of inhibition of reversal frequency and magnitude at the 1- and 5-s tail-touch/tap intervals. In addition, there were no significant sex differences in the amount of inhibition produced by tail-touch at any of the tailtouch/tap intervals. Taken together these data suggest that the additional sensory receptors in the tails of males do not greatly change the organisms response to tap, and we thus hypothesize that the circuit underlying the touch and tap withdrawal responses in hermaphrodites (Chalfie et al., 1985) exists in the same or similar form in the males, and that tail-touch activates the same neural circuits leading to inhibition of reversal to tap in both sexes. Both males and hermaphrodites showed habituation to repeated single taps of 10 s ISI. The rate of habituation did not differ between the two sexes in either the habituation or recovery experiments; however, the final level of habituation and the variance associated with that level were significantly higher in males than in hermaphrodites. Thus, males appeared to habituate, but seemed to be more likely to respond at each of the habituation trials and to respond with larger reversals than hermaphrodites. This suggests that although habituation occurs in all worms there are subtle differences in both the descending portion of the habituation curve and the asymptotic portion of the habituation curve of males. It is conceivable that the larger number and magnitude of reversals displayed by males at asymptote are related to the importance of tactile stimulation in the mating behavior of male C. elegans. If the animal were completely habituated to tactile stimulation, whether from a tap or from contact with another worm, very few or no reversals would occur. The absence of reversal activity might prevent, or at the very least seriously hamper, successful mating in males. Protecting the touch circuit from complete habituation would allow the male to continue to reverse, thereby affording the male the opportunity to investigate stimulation of the pos-
terior sensory receptors that might be indicative of a potential mating partner. We have suggested that the lack of dishabituation in males may have been due to the high variability of the 10 responses preceding the administration of the shock. Dishabituation is often considered evidence that habituation is not the result of sensory adaptation or motor fatigue (Groves & Thompson, 1970). Since dishabituation was not shown by males, it is possible that the response decrement shown by males might simply be the result of sensory adaptation or muscular fatigue. If fatigue or sensory adaptation is the cause of response decrement in males while learning is the cause of response decrement in hermaphrodites, then the two sexes would be unlikely to show the same rates of recovery following habituation training. However, Experiment V demonstrated the same rate of recovery from habituation in males and hermaphrodites; therefore, it is unlikely that entirely different processes underlie the decrement in responding seen in the male and the hermaphrodite. In conclusion, our behavioral observations support the hypothesis that the neural circuit underlying touch-withdrawal in hermaphrodites is also present in male C. elegans. The results also suggest that although the male responds similarly to hermaphrodites, showing habituation, inhibition, etc., the subtle differences in behavior are consistent with male mating behavior and the functional role of the additional tail structures in male mating.
REFERENCES Beck, C. D. O. (1991). The effects of aging on non-associative learning in the nematode Caenorhabditis elegans. Unpublished master's thesis, University of British Columbia, Vancouver. Beck, C. D. O., and Rankin, C. H. (in press). Effects on aging on non-associative learning in the nematode C. elegans. Behavioural Processes. Brenner, S.T. (1974). The genetics of Caenorhabditis elegans. Genetics, 77, 71-94. Byerly, L., Cassada, R. C., & Russell, R. L. (1976). The life cycle of the nematode Caenorhabditis elegans. Developmental Biology, 51, 23-33. Chalfie, M., Sulston, J. E., White, J. G., Southgate, E., Thomson, J.N., & Brenner, S. (1985). The neural circuit for touch sensitivity in Caenorhabditis elegans. The Journal of Neuroscience, 5, 956-964. Chalfie, M., and White, J. G. (1988). The nervous system. In W. B. Wood (Ed.), The nematode Caenorhabditis elegans (pp. 337391). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Chiba, C. M., & Rankin, C. H. (1990). A developmental analysis of spontaneous and reflexive reversals in the nematode Caenorhabditis elegans. Journal of Neurobiology, 21, 543-554.
BEHAVIORAL PLASTICITY IN MALE C. elegans Groves, P. M., & Thompson, R. F. (1970). Habituation: A dualprocess theory. Psychological Review, 77, 419-450. Hodgkin, J. (1983). Male phenotypes and mating efficiency in Caenorhabditis elegans. Genetics, 103, 43-64. Hodgkin, J. (1988). Sexual dimorphism and sex determination. In W. B. Wood (Ed.), The nematode Caenorhabdit~s elegans (pp. 243-279). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Hodgkin, J., Horvitz, H. R., & Brenner, S. (1979). Nondisjunction mutants of the nematode Caenorhabd~tis elegans. Genetics, 91, 67-94. Kumar, N., Williams, M., Culotti, J., and van der Kooy, D. (1989). Evidence for associative learning in the nematode C. elegans. Society for Neuroscience Abstracts, 15, 1141. Rankin, C.H. (1991). Interaction between two antagonistic reflexes in Caenorhabditis elegans. Journal of Comparative Physiology A, 169, 59-67. Rankin, C. H., & Chiba, C. M. (1988). Short and long term learning in the nematode C. elegans. Society for Neuroscience Abstracts, 14, 607. Rankm, C. H., and Chalfie, M. (1989). Analysis of non-associative
221
learning in C. eIegans: I. Neural circmt mutations. Society for Neuroscience Abstracts, 15, 1118. Rankin, C.H., Beck, C. D. O., & Chiba, C.M. (1990). Caenorhabd~tis elegans: A new model system for the study of learning and memory. Behavioral Brain Research, 37, 89-92. Rankin, C. H., & Broster, B. S. (1992). Factors affecting habituation and recovery from habituation in the nematode Caenorhabditis elegans. Behavioral Neuroscience, 106, 239-249. Sulston, J. E., Albertson, D. G., & Thomson, J. N. (1980). Postembryonic development of nongonadal structures. Developmental Biology, 78, 542-576. Wicks, S., and Rankin, C.H. [1991). Circuit analysis of interactions between two antagonistic reflexes in C. elegans. Society for Neuroscience Abstracts, 17, 1055. Winer, B. J., Brown, D. R., & Michels, K. M. (1991). Statistical Principles ~n Experimental Design (Third ed.). New York: McGraw-Hill. Wood, W. B. (1988). Introduction to C. elegans biology. In W. B. Wood (Ed.), The nematode Caenorhabditis elegans (pp. 116). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
