Logo of geneticsGeneticsCurrent IssueInformation for AuthorsEditorial BoardSubscribeSubmit a Manuscript
Genetics. 2007 Apr; 175(4): 1761–1771.
PMCID: PMC1855125

Diversity in Mating Behavior of Hermaphroditic and Male–Female Caenorhabditis Nematodes


In this study, we addressed why Caenorhabditis elegans males are inefficient at fertilizing their hermaphrodites. During copulation, hermaphrodites generally move away from males before they become impregnated. C. elegans hermaphrodites reproduce by internal self-fertilization, so that copulation with males is not required for species propagation. The hermaphroditic mode of reproduction could potentially relax selection for genes that optimize male mating behavior. We examined males from hermaphroditic and gonochoristic (male–female copulation) Caenorhabditis species to determine if they use different sensory and motor mechanisms to control their mating behavior. Instead, we found through laser ablation analysis and behavioral observations that hermaphroditic C. briggsae and gonochoristic C. remanei and Caenorhabditis species 4, PB2801 males produce a factor that immobilizes females during copulation. This factor also stimulates the vulval slit to widen, so that the male copulatory spicules can easily insert. C. elegans and C. briggsae hermaphrodites are not affected by this factor. We suggest that sensory and motor execution of mating behavior have not significantly changed among males of different Caenorhabditis species; however, during the evolution of internal self-fertilization, hermaphrodites have lost the ability to respond to the male soporific-inducing factor.

EVOLUTIONARY changes in reproductive modes of metazoans require modifications not only in the development of reproductive tissues, but also in neuromuscular-based behaviors that relate to species propagation. In nematodes of the genus Caenorhabditis, hermaphroditism has been proposed to have evolved twice (Cho et al. 2004; Kiontke et al. 2004). Sex-determining molecules that regulate gametogenesis between different Caenorhabditis species have been studied as targets that direct evolutionary changes in reproductive mechanisms (Haag and Kimble 2000; Baird 2002; Haag et al. 2002; Stothard et al. 2002; Hill et al. 2006; Stothard and Pilgrim 2006). However, the repercussions of reproductive modifications of the evolution of behaviors have not been intensively investigated.

The nematode Caenorhabditis elegans contains two genders, self-fertilizing hermaphrodites and males. The predominant sex generated from internal self-fertilization is the hermaphrodite; however, at a frequency of 0.2%, males can be generated via spontaneous nondisjunction events during meiosis. In contrast, when hermaphrodites are cross-fertilized, equal numbers of progeny males and hermaphrodites are produced. The ability of C. elegans males to mate with hermaphrodites has been documented to be inefficient in comparison to mating behaviors exhibited by Caenorhabditis species that display a gonochorism mode of reproduction (male–female copulation) (Chasnov and Chow 2002). This has been shown in population dynamics studies, which demonstrate that the frequency of males in androdioecious (male–hermaphrodite) C. elegans populations declines within a few generations (Chasnov and Chow 2002; Stewart and Phillips 2002). A likely reason for the reduction of cross-progeny males in androdioecious populations is that most hermaphrodites are not cross-fertilized; alternative postcopulatory, prezygotic explanations are unlikely since male sperm will out-compete the hermaphrodite sperm for fertilization (Ward and Carrel 1979). Thus, some aspect of C. elegans male mating behavior might not be optimized for successful cross-fertilization into hermaphrodites.

The first cellular dissection of C. elegans male mating behavior was conducted using detailed behavioral observations coupled with laser ablation analysis (Liu and Sternberg 1995). From observations using uncoordinated hermaphrodites as mating partners, C. elegans male mating behavior was dissected into several sensory and motor sub-behaviors. Mating behavior initiates when the ray and ventral sensilla of the male tail contact the hermaphrodite cuticle. These sensory neurons signal the male to press his tail against the hermaphrodite cuticle and begin moving backward along his mate, scanning for the vulva. If the male reaches the end of the hermaphrodite without encountering the vulva, then the posterior ray sensilla signal the male to make a ventral turn and continue to search for the vulva on the hermaphrodite's opposite side. When the HOA and HOB sensory neurons, located in the hook, and the PCA, PCB, and PCC sensory neurons, located in the postcloacal sensilla, sense the vulva, the male stops its backward locomotion and attempts to insert his copulatory spicules into the hermaphrodite. The spicules will repeatedly thrust at the tightly closed vulva until the tips partially penetrate the vulval slit. The putative proprioceptive SPC spicule neurons then trigger the spicules to penetrate completely through the vulva. After the spicules completely insert, the male gametes are then transferred from the seminal vesicle, through the vas deferens, out the cloaca and into the hermaphrodite.

This stereotyped male mating behavior can be consistently observed in couplings between males and genetically paralyzed hermaphrodites. However, not all of the mating substeps are performed with equal efficiency. Spicule insertion is the most difficult step for males to accomplish (Liu and Sternberg 1995). During matings with young paralyzed adult hermaphrodites, some males spend >10 min attempting to insert their spicules before the vulva is breached. During matings with constantly moving hermaphrodites, anecdotal observations from our laboratory, as well as from others, suggest that the ability of males to complete any defined behavioral substep, including spicule insertion, is variable. Thus many studies of C. elegans mating behavior use nonmoving hermaphrodites as the male's partner (Loer and Kenyon 1993; Liu and Sternberg 1995; Barr and Sternberg 1999; Yoda et al. 2000; Garcia et al. 2001; Lints and Emmons 2002; Carnell et al. 2005; Schindelman et al. 2006).

For species that display gonochorism, copulation is strictly essential for propagation. In contrast, the C. elegans self-fertilization mode of reproduction might relax selection on genes that are used in optimizing male mating behavior. This could contribute to the reduced ability of males to mate with normally behaving hermaphrodites. In this study, we asked why males of gonochoristic Caenorhabditis species are more proficient than their hermaphroditic relatives in performing the behavioral steps of copulation. We focused on spicule insertion behavior, since this step is the most difficult for C. elegans males to perform, and probably the major reason they fail to impregnate their hermaphrodite mates.

We found that males from the gonochoristic Caenorhabditis species 4, PB2801 and C. remanei species executed the various steps of mating similarly to the hermaphroditic species C. elegans and C. briggsae. However, the major difference between hermaphrodite–male and female–male couplings was that C. remanei and PB2801 males induced rapid paralysis in their virgin female mates. They also triggered the female vulval slit to widen, allowing instant spicule insertion. C. elegans and C. briggsae hermaphrodites were not affected this way during matings with their conspecific males or with gonochoristic males. Interestingly, C. briggsae males, but not C. elegans males, induced inactivity in virgin females of both gonochoristic species. We suggest that the nervous system of males from hermaphroditic and gonochoristic species are designed to copulate with nonmoving partners and that the nonoptimal mating efficiency that occurs in hermaphroditic species is due to the loss of male or hermaphrodite gene functions that induce paralysis in the hermaphrodites.



Strains were propagated at 20° on NGM agar plates seeded with Escherichia coli OP50 (Brenner 1974). We used the laboratory strain C. elegans Bristol N2 containing the mutation him-5(e1490). The him-5(e1490) allele (Hodgkin et al. 1979), located on LG V, increases the rate of X nondisjunction during meiosis and consequently increases the frequency of spontaneous males from 0.2 to 30%. The male–female strain of C. elegans Bristol (N2) contains the fog-2(q71) allele, located on LG V (Schedl and Kimble 1988). We also used the C. elegans wild strains, Palo Alto CB4855 and Hawaii CB4856. For C. briggsae, we used the strains AF16 VT847 and PB826. For C. remanei and Caenorhabditis species 4, PB2801, we used the strains PB4641 and PB2801, respectively. PB4641 and PB2801 are 20 times inbred derivatives of C. remanei and Caenorhabditis species 4. To generate males in C. briggsae and non-him-5 hermaphroditic C. elegans, late L4 hermaphrodites were heat-shocked at 30° for 5 hr and then shifted to 20°. Spontaneously generated males were then serially propagated with hermaphrodites to maintain androdioecious populations.

Laser ablations:

Males were laser operated on agarose pads containing NaN3 using standard protocols (Bargmann and Avery 1995). For gonad and M-cell ablations, newly hatched L1 animals of mixed sexes were anesthetized using pads that contain 1 mm NaN3. Ablated animals were then separated by sex when they reached the L4 stage. For Pn.p ablations, L1/L2 animals were anesthetized using pads that contain 5 mm NaN3. Males were then discarded at the L4 stage. For P9.p, P10.p, cloacal ganglia, and linker cell ablations, males were sexed using a dissecting microscope before being anesthetized on pads containing 10 mm NaN3. P9.p, P10.p cells were ablated in early L3 males. Cloacal ganglia and linker cell ablations were conducted at mid-L4 stage. At this stage, the tail spike is partially retracted, and the linker cell is ~10–20 μm from the cloacal cavity.

Mating assays:

From mixed-staged stocks of nematodes, 10–20 L4 males at the developmental stage of tail spike retraction and 10–20 L4 hermaphrodites or females at the stage where a defined clearing was formed in the vulval region (“the Christmas tree stage” of vulval development) were separated from their opposite sex and placed on separate NGM agar plates. Females and hermaphrodites were assayed for mating behavior 24, 48, and 72 hr later. To make the mating lawns, E. coli OP50 was grown overnight in LB media at 37° without aeration. A total of 1 ml of culture was concentrated to 10 μl and then spotted onto an NGM agar plate. The diameter of the bacterial lawn was ~5 mm. Ten 24-, 48-, or 72-hr adult females or hermaphrodites were placed onto the lawn using a platinum wire pick. Then the 24-hr male of interest (nonoperated or laser ablated) was placed onto the lawn.

We did not use older mating lawns since hermaphrodites and females behaved differently on 24 hr or older preformed mating lawns, as compared to freshly made ones. On older lawns, some hermaphrodites and females crawled to the border of the bacterial lawn and sometimes stopped moving once they reached the edge. In contrast, on freshly made bacterial lawns, hermaphrodites and females continually moved around the lawn.

Great care was taken in manually transferring animals from one plate to another. Harsh manipulations sometime caused a male to immediately crawl off the mating lawn. In those cases, the wandering male was removed and another male was added. Mating observations lasted for 10 min or until the male inserted his spicules into a partner; timing started when the male was placed on the lawn. The number of partners that a male attempted to mate with during the observation window was recorded. If the male tail lost contact with a mate during the mating sequence, the concentration of moving hermaphrodites or females was high enough to allow the male to quickly contact a different mate, but if the male recontacted the same hermaphrodite or female, we still recorded that as a new mating encounter. Males and mated hermaphrodites/females were removed from the lawn after the observation was completed. A new hermaphrodite/female was then added to the lawn.

In the course of this study, we noted some interesting observations regarding C. briggsae and C. remanei mating behavior. C. briggsae AF16 and VT847 males appeared to have no ability to discriminate themselves from their mates. Between 12 to 24 hr after L4 molt, 95–100% of C. briggsae adult males contained a copulatory plug deposited over their excretory pore. This phenomenon is similar to what has been reported for the wild C. elegans strain AB2 (Gems and Riddle 2000). We observed that when the C. briggsae male's tail contacted any part of his own body, he immediately initiated mating behavior on himself. This would ultimately result in the insertion of his spicules, transferring of sperm, and depositing of a copulatory plug over his own excretory pore. During mating observations with C. briggsae males, if a male started to mate with himself, we gently prodded the male with a pick or an eyelash hair to disrupt the self-mating behavior. In contrast, males from the C. briggsae strain PB826 did not show this self-plugging behavior. Also, PB826 males did not show much behavioral interest in mating, either with hermaphrodites from their own strain or with conspecific hermaphrodites. Stocks of PB826 that contained males were serially maintained by setting up mating crosses consisting of five to six PB826 males and one 48- to 72-hr PB826 hermaphrodite. To maximize the probability that the males would contact the hermaphrodite, matings were conducted on plates that contained a 1-cm bacterial lawn.

We also noted that virgin C. remanei females were strongly attracted to copulating C. remanei males. When a C. remanei male induced behavioral inactivity in a mate, the other females would quickly crawl toward the pair and clump around the couple. During mating observations with C. remanei, we gently prodded females that moved toward the mating pair with a pick to keep them from disrupting copulation of the mating couple.


Caenorhabditis gonochoristic females behave differently from hermaphrodites during mating:

We initiated this study to understand under what conditions the neuromuscular circuit that controls C. elegans spicule insertion is designed to function. Previous studies have shown that N2 C. elegans males insert their spicules into older, rather than younger, genetically paralyzed hermaphrodites (Garcia et al. 2001; Garcia and Sternberg 2003; Gruninger et al. 2006). However, the mutations that produce paralysis in hermaphrodites could additionally cause defective vulval musculature that disrupts the males' ability to insert their spicules. To address if males can insert their spicules into young wild-type hermaphrodites efficiently, we observed mating behavior of 10 24-hr virgin N2 males coupled to 24- and 72-hr virgin N2 hermaphrodites for 10 min or until the males inserted their spicules (Figure 1A). We found that the constant hermaphrodite movement caused males to lose contact easily with their mates before the mating sequence was completed. The inability of males to maintain contact led to a greater number of mating attempts with different hermaphrodites. On average, males attempted to mate with 7 ± 3 (mean ± SD) 24-hr hermaphrodites and 3 ± 2 (mean ± SD) 72-hr hermaphrodites during the 10 min of observation. Males contacted fewer 72-hr than 24-hr hermaphrodites (P = 0.001, Mann–Whitney test) because 90% of males inserted their spicules into the older virgins, whereas 10% of males inserted into the younger hermaphrodites (P = 0.001, Fischer's exact test).

Figure 1.
Distribution of spicule insertion ability in different virgin Caenorhabditis males. Each circle represents the behavior of a single male. Males that inserted their spicules into their mates are represented as an open circle. Those that did not insert ...

One possible explanation for better spicule insertion into older hermaphrodites is that older hermaphrodites have depleted much of their self-sperm, which might trigger some mechanism to allow the vulva to be more accessible. To test this possibility, we observed matings between males and fog-2(lf) females. The fog-2 mutation does not affect males, but inhibits sperm production in hermaphrodites, causing them to become reproductively female (Schedl and Kimble 1988). Results of matings between males and fog-2(lf) females were similar to males and hermaphrodites; males penetrated older females more easily than younger ones (Figure 1B). Therefore, the presence of sperm does not inhibit male spicule insertion into young hermaphrodites. Some other age-related factor, whether it is physiological or behavioral, affects how efficiently the vulva can be breached. Although not a central focus of this study, we additionally observed that young fog-2(lf) females did not appear to move away from males at the same frequency as young hermaphrodites. On average, males attempted to mate with 2 ± 1 (mean ± SD) 24-hr females (P < 0.001 vs. 24-hr hermaphrodites, Mann–Whitney test) during the observation window. This is consistent with reported observations that the fog-2(lf) mutation alters behaviors in females that promote copulation (G. A. Kleemann, personal communication; Lipton et al. 2004) and that sperm-depleted wild-type hermaphrodites are more behaviorally receptive to males than their self-sperm-containing cohorts (Kleemann and Basolo 2007).

Inefficient spicule insertion could be due to relaxed selection for genes that regulate motor functions used for mating behavior in hermaphroditic species. To determine if C. elegans males lost some ability for mating, we asked how spicule insertion behavior of C. elegans compares with males of the related Caenorhabditis species 4, PB2801, C. remanei and C. briggsae. Phylogenetic relationships between these nematodes have been determined through analyzing par-6, pkc-3, RNA polymerase II, and large and small subunit ribosomal RNA sequences. Like C. elegans, C. briggsae is a hermaphroditic nematode. However, it is more closely related to C. remanei and to PB2801 than to its hermaphrodite relative (Figure 1C), indicating that hermaphroditism had evolved independently in the two species (Cho et al. 2004; Kiontke et al. 2004). When spicule insertion behavior was observed in C. briggsae AF16 males coupled to 24- and 72-hr AF16 C. briggsae hermaphrodites, males had difficulty inserting into both ages of hermaphrodites (Figure 1D). The efficiency of inserting into 72-hr C. briggsae hermaphrodites was different when compared to C. elegans males inserting into their 72-hr hermaphrodites (4/10 C. briggsae AF16 hermaphrodites were penetrated, whereas 9/10 C. elegans N2 hermaphrodites were penetrated; P < 0.001, Fischer's exact test), suggesting that behavioral changes that occur in older C. elegans might be specific to their physiology. In contrast, C. remanei and PB2801 males instantaneously inserted their spicules into their own 24-hr virgin females (Figure 1D). This demonstrated that, unlike C. elegans and C. briggsae, males of the two gonochoristic species are more efficient at penetrating and impregnating their mates.

C. remanei, C. briggsae, and C. elegans males have been shown to transfer sperm into different heterospecific partners after extended periods of mating (Hill and L'Hernault 2001). This suggests that Caenorhabditis females and hermaphrodites display common physical and chemical cues that can be recognized by male tail sensilla. Since C. remanei and PB2801 males were efficient at penetrating their own 24-hr females, we asked if they could penetrate young heterospecific mates with similar proficiency. We found that, similar to C. elegans and C. briggsae males, males of the two gonochoristic species do not insert their spicules very efficiently into young hermaphrodites, although they can insert their spicules into females of their heterospecific relatives (Table 1). Surprisingly, although C. briggsae AF16 males do not insert their spicules into their own hermaphrodites easily, they instantly penetrated both PB2801 and C. remanei females, indicating that the inability to insert their spicules into their conspecific mate is partly due to the C. briggsae hermaphrodite (Table 1). C. briggsae AF16 was originally isolated from Ahmedabad, India (Fodor et al. 1983). To determine how males from different C. briggsae strains behave, we analyzed the spicule insertion behavior of males from the C. briggsae strain VT847, isolated from Hawaii, and from the C. briggsae strain PB826, isolated from Ohio (Kiontke and Sudhaus 2006). Similar to AF16 males, we found that VT847 males inserted their spicules into both PB2801and C. remanei females better than into C. briggsae AF16 or C. elegans N2 hermaphrodites (Table 1). In contrast, C. briggsae PB826 males did not insert their spicules very efficiently into any of the hermaphrodites or females tested. This was because none of the observed 40 PB826 males showed interest in executing mating behavior when their tails contacted their mates. However, two PB2801 and C. remanei females and one C. elegans hermaphrodite were penetrated during the observation period (Table 1).

Efficiency of spicule insertion during heterospecific matings

Locomotor activity of young hermaphrodites is one potential factor that reduces the efficiency of C. elegans and C. briggsae male spicule insertion behavior. During matings, young C. elegans and C. briggsae hermaphrodites continue to move around the bacterial lawn as males try to breach the vulva with their spicules. The hermaphrodite movements generally cause the male tail to move off the vulva, and in many instances the male will also lose contact with the hermaphrodite (Kleemann and Basolo 2007). In contrast, we noted that when the cloacal region of C. remanei PB2801 and C. briggsae AF16 VT847 and also PB826 males contacted the virgin female vulva, the males immediately insert their spicules, and the females simultaneously stop locomotion, defecation, and reduce pharyngeal pumping behavior (supplemental video S1 at http://www.genetics.org/supplemental/). In C. remanei, this striking behavioral inactivity continues for ~64 ± 38 sec (mean ± SD n = 15 females), which is sufficient time for the male to transfer sperm and deposit a copulatory plug over the vulva. C. remanei and PB2801 virgin females do not display this soporific behavior when C. elegans N2 males attempt to insert; instead, the virgin females behave like C. elegans and C. briggsae hermaphrodites by moving off the copulating male (supplemental video S2 at http://www.genetics.org/supplemental/). Similarly, C. elegans and C. briggsae hermaphrodites do not display this behavioral inactivity during matings with C. briggsae, C. remanei, and PB2801 males. This demonstrates that virgin hermaphrodites and females behaviorally respond to males differently.

Since the 1950s, C. elegans Bristol N2 have been grown in the laboratory under various culture methods (propagation in liquid culture containing axenic media, liquid monoxenic media containing E. coli, and surface propagation on E. coli-containing NGM plates) (Hodgkin and Doniach 1997). The lack of selection for mating due to these laboratory culture conditions could contribute to the loss of certain N2 male behaviors. We tested C. elegans males from the Palo Alto CB4855 isolate and the Hawaii CB4856 isolate to determine if males of other C. elegans strains can induce the soporific behavior in virgin gonochoristic females. Unlike N2 males, males from both of these C. elegans strains have the ability to deposit a copulatory plug over the vulva after sperm transfer. In addition, Palo Alto CB4855 males have been reported to sire progeny up to 12 days of adulthood, whereas N2 males are impotent after 4 days (Hodgkin and Doniach 1997). However, despite the additional mating attributes that these males display, we found that CB4855 and CB4856 males do not induce behavioral inactivity in C. remanei and PB2801 virgin females, nor do they insert their spicules into 24-hr adult C. briggsae AF16 or N2 hermaphrodites better than N2 males (Table 1). Thus C. elegans males do not display the ability to modify their mate's behavior to promote spicule insertion.

Contact between the C.remanei female vulva and the male tail induces female behavioral inactivity:

To determine how gonochoristic females respond to males during mating, we narrowed our analysis to determining what aspects of C. remanei male mating behavior trigger C. remanei virgin females to attenuate locomotion, defecation, and pharyngeal pumping. C. remanei virgin females display the soporific behavior when the male contacts the vulva, inserts his spicules, and ejaculates. To determine if the female vulva was required for the soporific behavior, we ablated the Pn.p cells: P4.p, P5.p, P6.p, P7.p, P8.p, and P9.p in 10 late L1 to mid-L2 females. In C. elegans, this ablation removes not only the putative vulval precursor cells, but also other Pn.p cells that might adopt a vulval precursor fate (Kimble 1981). We found that 0/10 vulvaless females displayed behavioral inactivity as the males continually moved backward along the female.

To rule out the possibility that ablating nonvulval Pn.p cells might indirectly affect the female soporific behavior, we tested the requirement for the vulva in a different way; we ablated the somatic gonad, which is necessary for the Pn.p cells to adopt a vulval fate. Z1, Z2, Z3, and Z4 are gonadal precursor cells in newly hatched C. elegans larvae. Descendants of Z1 and Z4 give rise to the somatic gonad, and descendants of Z2 and Z3 give rise to the germline (Kimble and Hirsh 1979). Ablating the equivalent cells of Z1, Z2, Z3, and Z4 in five L1 C. remanei females results in adults that do not contain a somatic gonad, a germline, or vulval tissue, similar to C. elegans. In addition, 0/5 operated animals displayed behavioral inactivity as C. remanei males attempted to mate with them. When Z2 and Z3 were ablated in two C. remanei females, operated adult animals contained a somatic gonad and vulval tissue, but no germline. In contrast to Pn.p-and gonad-ablated females, both germline-ablated females displayed behavioral inactivity when males contacted their vulva. These experiments suggest that the C. remanei vulva can act as a sensor to modify female behavior. However, since the C. remanei female does not display the soporific effect when the cloaca of C. elegans males touches the vulva, the C. remanei vulva must sense something more specific than mechanical contact.

Neurons associated with the C.remanei male cloaca are required to induce the soporific behavior in females:

The soporific behavior that virgin C. remanei females display occurs when males contact the vulva, insert their spicules, and transfer sperm. Since these male mating substeps occur very quickly, we laser ablated various cells in C. remanei males that, on the basis of analogy to C. elegans male mating behavior, should uncouple vulva location, spicule insertion, and ejaculation behavior (Figures 2 and and3).3). We then asked if the operated males could attenuate locomotion, pharyngeal pumping, and defecation for at least 20 sec in their female mates.

Figure 2.
M-cell-ablated male. (A) DIC photomicrograph of the left lateral region of an adult C. remanei male tail. Anterior to the left, dorsal to the top. Bar, 10 μm. (B) DIC photomicrograph of a ventral view of the tail region of an M-cell-ablated C. ...
Figure 3.
Positions of cloacal ganglia cells and linker cell in Caenorhabditis males. Anterior to the left, dorsal to the top. Bars, 10 μm. DIC photomicrographs of the left lateral cloacal ganglia region of (A) a mid-L4 C. elegans male tail and (B) a mid-L4 ...

We first removed the spicule protractor muscles to ask if complete spicule insertion triggers female inactivity during normal mating. To do this, the M cell was laser ablated in L1 males. In C. elegans, M-cell-ablated males lack the M-lineage-derived body wall muscles, coelomocytes, and all sex muscles, including the spicule muscles. During normal development, the spicule muscles are required for the copulatory spicules to elongate into their characteristic blade-like appearance (Figure 2A). Without these muscles, the spicules do not elongate and consequently form stunted structures (Sulston et al. 1980). Ablation of the M cell in C. remanei resulted in males that lack their sex muscles and also develop crumpled spicules (Figure 2B). During mating behavior, the M-cell-ablated males show variable reduction in the efficiency of backing and turning behavior, but all of the operated males located and stopped at the vulva on the first contact. Because the males lack the muscles that control spicule movements, none of the observed males inserted their crumpled spicules; however, 90% of operated males induced the C. remanei females to display the soporific behavior (Table 2). Interestingly, although locomotion, pharyngeal pumping, and defecation were reduced, contact between the male cloaca and the vulva caused the vulva muscles to contract periodically in all paralyzed females (Figure 2, C and D) (supplemental video S3 at http://www.genetics.org/supplemental/).

Ablation of C. remanei male structures that affect inactivity in females

Since the M-cell ablation result demonstrated that spicule movements were not necessary to induce the soporific behavior, we asked if sustained vulval contact between the male cloaca and the female vulva was required. In C. elegans, the male senses the vulva using the hook sensillum neurons HOA and HOB and the postcloacal sensilla (p.c.s.) neurons PCA, PCB, and PCC. We ablated the blast cells P9.p and P10.p in L3 C. remanei males, which in C. elegans results in the loss of the HOA and HOB neurons in adults. Similar to C. elegans, all P9.p- and P10.p-ablated C. remanei males fail to locate the vulva efficiently; however, 90% of the males induced the soporific behavior either when their cloaca transiently passed over the vulva or when they inserted their spicules (Table 2). When the cells analogous to the C. elegans PCA, PCB, and PCC were ablated (Figure 3, A and B), all operated males hesitated in the general region of the vulva; however, none of the males positioned their cloaca over the vulval slit for longer than 2–3 sec, and, consequently, very few of the males inserted their spicules (Table 2). This behavioral defect is similar to that displayed by PCA-, PCB-, and PCC-ablated C. elegans males (Liu and Sternberg 1995), indicating that these C. remanei cells have functions in sensing the vulva similar to those of the C. elegans p.c.s. neurons. Similarly, like the P9.p- and P10.p-ablated males, 7/10 PCA-, PCB-, and PCC-ablated C. remanei males also induced inactivity in the females when their cloaca transiently contacted the vulva.

In C. elegans, the p.c.s. and hook neurons not only function redundantly to sense the vulva (Liu and Sternberg 1995), but also act to initiate spicule insertion behavior (Garcia et al. 2001). When the spicules partially breach the vulval slit, the SPC motor neurons then promote complete spicule protraction and ejaculation. In C. remanei, ablating the p.c.s. and hook neurons did not strongly reduce the males' ability to induce the soporific affect. However, ablating the p.c.s. neurons did reduce spicule insertion and ejaculation, suggesting that the C. remanei SPC and the p.c.s. neurons might share some redundant functions. To test if these neurons might have some functional overlap during mating, we ablated the SPC cells alone and with the p.c.s. neurons. In contrast to C. elegans, where the SPC motor neurons are essential for complete spicule penetration, 10/11 SPC-ablated C. remanei males induced the soporific behavior and inserted their spicules completely, but of the 10 males that inserted, only 1 transferred sperm (Table 2). Thus, the SPC cells in both C. remanei and C. elegans are required for sperm transfer; however, their functions in spicule insertion behavior have diverged between the two species. When the C. remanei SPC cells were ablated with the PCA, PCB, and PCC cells, 0/10 males induced the soporific effect or inserted their spicules (Table 2). Also, the females would crawl back and forth trying to shift the males off their cuticle, reminiscent of how 24-hr C. briggsae and C. elegans adult hermaphrodites behaved with males. Since 70 and 91% of p.c.s. and SPC-ablated males, respectively, induced the soporific effect in comparison to 0% of the combination-ablated males, the SPC and the p.c.s. neurons in intact C. remanei males must act redundantly to trigger some process that ultimately results in the attenuation of female behavior.

A soporific-inducing factor from C. remanei, PB2801, and C.briggsae males requires a connection between the somatic gonad and the cloacal opening:

Partial EM reconstructions of the C. elegans male indicate that the p.c.s. neurons PCB and PCC and the spicule neuron SPC also synapse the proximal somatic gonad (S. W. Emmons, D. H. Hall and M. Xu, personal communication; http://www.wormatlas.org). The relevance of these connections has not been determined; however, they might be used to coordinate vulva location and spicule insertion behavior with ejaculation (Gower et al. 2005). Although C. elegans males do not induce the soporific effect, the connections between these neurons and the somatic gonad made us ask if, in C. remanei, the gonad might act with the p.c.s. and SPC neurons to induce the soporific behavior in females.

When Z1, Z2, Z3, and Z4 of C. remanei males were ablated, none of the adults contained a somatic gonad or a germline, but development of other male structures appeared to be superficially normal. Ablation of the gonad eliminated the ability of all operated males to induce the soporific effect (Table 2). In addition, none of the males were able to hold their position on the vulva for >5 sec or to insert their spicules, a defect that is similar to that caused by ablating the p.c.s. and the SPC neurons.

Since the gonad was required, we then asked if the germline provided any contribution to the soporific effect. Z2- and Z3-ablated males lacked a germline, but they still contained the somatic gonad, which is connected to the cloaca. A total of 9/10 operated males induced female inactivity, inserted their spicules, transferred spermless seminal fluid, and deposited a copulatory plug. This indicated that the germline is not necessary for the males to sedate the females.

We then asked if a connection between the proximal somatic gonad and cloaca was required for the soporific effect. The proximal end of the male somatic gonad consists of the vas deferens, which is the conduit between the seminal vesicle and the cloacal cavity; the cloaca opens to the environment. In C. elegans, these somatic gonadal cells are guided to the cloacal cavity via the linker cell (Figure 3C). When the linker cell reaches the cloaca, cells associated with the cloacal cavity kill the linker cell, thus facilitating the lumen of vas deferens in connecting with the cloacal opening (Kimble and Hirsh 1979; Sulston et al. 1980). If the linker cell is laser ablated before it reaches the cloaca, then no connection is made between the vas deferens and the outside. In mid-L4 males, when the tail spike begins retraction, we ablated the cell that shares similar morphology to the C. elegans linker cell (Figure 3, C and D). During mating, all operated males, despite containing a somatic gonad and germline, behaved similarly to Z1-, Z2-, Z3-, and Z4-ablated males; all linker-cell-ablated males failed to sedate their mates, maintain their position over the vulva for >5 sec, or insert their spicules (Table 2) (supplemental video S4 at http://www.genetics.org/supplemental/). Taken together, these results suggest that, prior to spicule insertion, the somatic gonad or some substance that it might release requires access to the cloacal opening when the male cloaca contacts the female vulval lips.

Since PB2801 and C. briggsae males also induce the soporific effect in gonochoristic females, we ablated the linker cell in PB2801 and C. briggsae L4 males (Figure 3, E and F) and then asked if the operated adults could still induce behavioral inactivity in virgin PB2801 females. We found that 0/10 PB2801 and 0/10 C. briggsae males can induce the soporific effect. This suggests that the cellular mechanism used to sedate virgin females might be common to C. remanei, PB2801, and C. briggsae males.

Components of the male germline suppress the soporific behavior in nonvirgin females:

We casually noted that C. remanei and PB2801 females that contain a mating plug and eggs in their uterus behaved like C. elegans and C. briggsae hermaphrodites during male mating attempts. To test this more carefully, we mated 10 virgin C. remanei females with males. We allowed the males to transfer sperm and then deposit a mating plug. One hour later, we retested the nonvirgin females with virgin males. We found that only 1/10 nonvirgin female displayed the soporific effect. In parallel, we mated 10 virgin C. remanei females with germline-ablated males. Each of the germline-ablated C. remanei males (n = 10) induced behavioral inactivity, transferred spermless fluid from their somatic gonad, and then deposited a copulatory plug. We then retested the females with virgin C. remanei nonoperated males. Interestingly, although all 10 females had a copulatory plug covering their vulva, they all displayed behavioral inactivity when males passed over or around the plug (supplemental video S5 at http://www.genetics.org/supplemental/). Eventually, the males displaced the copulatory plug, presumably via their hook structure, inserted their spicules, and transferred sperm. Therefore, some component of the germline, perhaps sperm, can change the physiology of females so that they either become refractive to or inhibit the production of the soporific factor produced by the male somatic gonad.


In the species C. remanei and in the Caenorhabditis species 4, PB2801, male–female copulation is essential for propagating the species. This necessity predicts that various requirements must be fulfilled to ensure that mating behavior is successful. One requirement is to bring individuals of opposite genders in close proximity to initiate copulation. In C. remanei and C. elegans, this is accomplished by secreted substance(s) that attract and retain males close to females and hermaphrodites (Chasnov and Chow 2002; Simon and Sternberg 2002; Lipton et al. 2004). Another requirement is for the mating couple to remain in contact until sperm transfer is complete. Our study provides a mechanism that explains a major behavioral difference between hermaphrodites and gonochoristic females during this stage of mating (Figure 4).

Figure 4.
One possible model of how gonochoristic males immobilize their mates. (A) The male moves backward along the female cuticle scanning for the vulva. The female is also moving during this period. (B) The hook and postcloacal sensilla neurons sense the vulva ...

The behavioral steps of copulation leading up to spicule insertion are similar among the males of the Caenorhabditis genus used in this study. However, upon vulval contact, copulation behaviors between the species differ. For C. elegans and C. briggsae, males repetitively prod the vulval slit of virgin hermaphrodites with their spicules before they completely insert. Depending on the age and the locomotor activity of the hermaphrodite, duration of spicule prodding can range from seconds to minutes. In contrast, when C. remanei and PB2801 males contact the vulva, their female mates stop moving, and spicule insertion occurs instantly.

We found that, upon contact with the vulva, a factor associated with gonochoristic males rapidly immobilizes the virgin female, while simultaneously inducing the vulval opening to widen. Once females are impregnated, they become refractive to the soporific-inducing factor. The soporific-inducing effect is remarkable, considering that topical exposure to the vulva causes multiple muscles used in locomotion, pharyngeal pumping, and defecation behaviors to become simultaneously relaxed, whereas genital muscles that are connected to the vulva contract. This behavior is superficially similar to the behavior that C. elegans and other Caenorhabditis males display when they transfer sperm. During sperm transfer, the spicule muscles remain contracted, whereas muscles involved in pharyngeal pumping, locomotion, and defecation are relaxed (Liu and Sternberg 1995; Gruninger et al. 2006). Possibly, mating-induced inactivity in females and ejaculation behavior in males might be controlled by the same underlying mechanism.

The male soporific-inducing factor might be a structural component of the male cloaca or a substance that is released from the cloacal opening upon vulval contact. We favor the latter since the effectiveness of the soporific-inducing factor requires the somatic gonad, which produces seminal fluid; the postcloacal sensilla neurons, which facilitate sensing the vulva and inducing spicule insertion; and the SPC neurons, which facilitate the transfer of seminal fluid and sperm from the gonad. Our experiments do not differentiate between whether the soporific-inducing factor is directly secreted by the SPC and the p.c.s. neurons or by the somatic gonad. Nothing is known about the development of the male SPC and the p.c.s. neurons in the gonochoristic Caenorhabditis species. In C. elegans, the somatic gonad is not required for SPC and the p.c.s. functions during mating behavior. However, in C. remanei, laser ablation of the somatic gonad mimics laser damage to the p.c.s. and the SPC neurons. It is possible that the SPC and the p.c.s neurons directly secrete the soporific-inducing factor through the cloacal cuticle and out the cloacal opening, but this function requires interactions with the somatic gonad. An alternative possibility is that, upon contact with the vulva, the p.c.s. and SPC neurons trigger the somatic gonad directly to release the soporific-inducing factor through the gonadal–cloacal junction and out the cloacal opening (Figure 4). In Drosophila melanogaster, peptides found in male seminal fluid have been shown to affect behavior, physiology, and gene expression of impregnated females (Lawniczak and Begun 2004; McGraw et al. 2004). Therefore, substances in nematode pre-ejaculation fluid might similarly cause behavioral changes in virgin female nematodes.

The hermaphroditic C. elegans and C. briggsae species are believed to have evolved independently from gonochoristic ancestors (Braendle and Felix 2006). For C. briggsae, this is consistent with our observation that C. briggsae males can induce behavioral inactivity in both C. remanei and PB2801 females. The putative soporific-inducing factor must be vestigial in C. briggsae males, since C. briggsae hermaphrodites do not respond to their males in the same way as C. remanei or PB2801 females do. Presumably, over evolutionary time, the hermaphroditic mode of reproduction either relaxed the maintenance of genes used in transducing the effects of the male factor or selected for activated alleles of genes that normally inhibit the effects of the male factor. In C. elegans, neither hermaphrodite nor male displays any obvious aspect of this behavior. By analogy with C. briggsae, either one or both genders lost this behavioral trait. Because the genome of C. elegans is more divergent relative to C. briggsae, C. remanei, and PB2801, it is possible that the gonochoristic ancestor of C. elegans might have used a different mechanism to promote copulation. However, since C. elegans males can execute all steps of mating behavior with heterospecific partners (Hill and L'Hernault 2001), it is likely that its ancestors probably also shared the ability to induce and respond to the same type of soporific factor.

Mating behavior of C. elegans males is not efficient with young moving hermaphrodites. The probable reason is that the C. elegans male nervous system, like that of males of the other Caenorhabditis species, was not designed to mate with mobile partners. Additionally, spicule insertion is difficult for C. elegans males since the young hermaphrodites do not actively facilitate spicule penetration. C. elegans males do not induce behavioral inactivity in young hermaphrodites, but are more efficient in inserting their spicules and transferring sperm into older partners. The 72-hr hermaphrodites used in this study are probably an extreme case of what is the most efficient copulation partner for C. elegans males, since cross-fertilized hermaphrodites at that age are not as fertile as younger cross-fertilized hermaphrodites (L. R. Garcia, unpublished observation). The age of the hermaphrodite mate, after L4 molt, that is optimized for copulation and fecundity should be between 24 and 72 hr. The reasons for the correlation between age and copulation efficiency is not obvious. The quantity of self-sperm is likely involved (Kleemann and Basolo 2007), but perhaps in addition, the mechanism that induces mating receptivity in gonochoristic virgin females partially functions constitutively in older C. elegans hermaphrodites.


We thank T. Gruninger, R. Lints, Y. Liu, and D. Gualberto for reading the manuscript. We also thank the journal reviewers for helpful suggestions in the revision of this manuscript. Nematode strains were provided by the Caenorhabditis Genetics Center, which is funded by the National Institutes of Health (NIH) Center for Research Resources. B.L. and L.R.G. were supported by NIH grant GM70431 and by the Searle Scholars Program.


  • Baird, S. E., 2002. Haldane's rule by sexual transformation in Caenorhabditis. Genetics 161: 1349–1353. [PMC free article] [PubMed]
  • Bargmann, C. I., and L. Avery, 1995. Laser killing of cells in Caenorhabditis elegans. Methods Cell Biol. 48: 225–250. [PubMed]
  • Barr, M. M., and P. W. Sternberg, 1999. A polycystic kidney-disease gene homologue required for male mating behavior in C. elegans. Nature 401: 386–389. [PubMed]
  • Braendle, C., and M.-A. Felix, 2006. Sex determination: ways to evolve a hermaphrodite. Curr. Biol. 16: 468–471. [PubMed]
  • Brenner, S., 1974. The genetics of Caenorhabditis elegans. Genetics 77: 71–91. [PMC free article] [PubMed]
  • Carnell, L., J. Illi, S. W. Hong and S. L. McIntire, 2005. The G-protein-coupled serotonin receptor SER-1 regulates egg-laying and male mating behaviors in Caenorhabditis elegans. J. Neurosci. 25: 10671–10681. [PubMed]
  • Chasnov, J. R., and K. L. Chow, 2002. Why are there males in the hermaphroditic species Caenorhabditis elegans? Genetics 160: 983–994. [PMC free article] [PubMed]
  • Cho, S., S. W. Jin, A. Cohen and R. E. Ellis, 2004. A phylogeny of Caenorhabditis reveals frequent loss of introns during nematode evolution. Genome Res. 14: 1207–1220. [PMC free article] [PubMed]
  • Fodor, A., D. L. Riddle, F. K. Nelson and J. W. Golden, 1983. Comparison of a new wild-type Caenorhabditis briggsae with laboratory strains of C. briggsae and C. elegans. Nematologica 29: 203–217.
  • Garcia, L. R., and P. W. Sternberg, 2003. Caenorhabditis elegans UNC-103 ERG-like potassium channel regulates contractile behaviors of sex muscles in males before and during mating. J. Neurosci. 23: 2696–2705. [PubMed]
  • Garcia, L. R., P. Mehta and P. W. Sternberg, 2001. Regulation of distinct muscle behaviors controls the C. elegans male's copulatory spicules during mating. Cell 107: 777–788. [PubMed]
  • Gems, D., and D. L. Riddle, 2000. Genetic, behavioral and environmental determination of male longevity in Caenorhabditis elegans. Genetics 154: 1597–1610. [PMC free article] [PubMed]
  • Gower, N. J., D. S. Walker and H. A. Baylis, 2005. Inositol 1,4,5-trisphosphate signaling regulates mating behavior in Caenorhabditis elegans males. Mol. Biol. Cell 16: 3978–3986. [PMC free article] [PubMed]
  • Gruninger, T. R., D. G. Gualberto, B. LeBoeuf and L. R. Garcia, 2006. Integration of male mating and feeding behaviors in Caenorhabditis elegans. J. Neurosci. 26: 169–179. [PubMed]
  • Haag, E. S., and J. Kimble, 2000. Regulatory elements required for development of Caenorhabditis elegans hermaphrodites are conserved in the tra-2 homologue of C. remanei, a male/female sister species. Genetics 155: 105–116. [PMC free article] [PubMed]
  • Haag, E. S., S. Wang and J. Kimble, 2002. Rapid coevolution of the nematode sex-determining genes fem-3 and tra-2. Curr. Biol. 12: 2035–2041. [PubMed]
  • Hill, K. L., and S. W. L'Hernault, 2001. Analyses of reproductive interactions that occur after heterospecific matings within the genus Caenorhabditis. Dev. Biol. 232: 105–114. [PubMed]
  • Hill, R. C., C. E. de Carvalho, J. Salogiannis, B. Schlager, D. Pilgrim et al., 2006. Genetic flexibility in the convergent evolution of hermaphroditism in Caenorhabditis nematodes. Dev. Cell 10: 531–538. [PubMed]
  • Hodgkin, J., and T. Doniach, 1997. Natural variation and copulatory plug formation in Caenorhabditis elegans. Genetics 146: 149–164. [PMC free article] [PubMed]
  • Hodgkin, J. A., H. R. Horvitz and S. Brenner, 1979. Nondisjunction mutants of the nematode Caenorhabditis elegans. Genetics 91: 67–94. [PMC free article] [PubMed]
  • Kimble, J., 1981. Alterations in cell lineage following laser ablation of cells in the somatic gonad of Caenorhabditis elegans. Dev. Biol. 87: 286–300. [PubMed]
  • Kimble, J., and D. Hirsh, 1979. The postembryonic cell lineages of the hermaphrodite and male gonads in Caenorhabditis elegans. Dev. Biol. 70: 396–417. [PubMed]
  • Kiontke, K., and W. Sudhaus, 2006. Ecology of Caenorhabditis species, pp. 1–14 in Wormbook, edited by The C. elegans Research Community (http://www.wormbook.org). [PubMed]
  • Kiontke, K., N. P. Gavin, Y. Raynes, C. Roehrig, F. Piano et al., 2004. Caenorhabditis phylogeny predicts convergence of hermaphroditism and extensive intron loss. Proc. Natl. Acad. Sci. USA 101: 9003–9008. [PMC free article] [PubMed]
  • Kleemann, G., and A. L. Basolo, 2007. Facultative decrease in mating resistance in hermaphoditic Caenorhabditis elegans with self-sperm depletion. Anim. Behav. (in press).
  • Lawniczak, M. K. N., and D. J. Begun, 2004. A genome-wide analysis of courting and mating responses in Drosophila melanogaster females. Genome 47: 900–910. [PubMed]
  • Lints, R., and S. W. Emmons, 2002. Regulation of sex-specific differentiation and mating behavior in C. elegans by a new member of the DM domain transcription factor family. Genes Dev. 16: 2390–2402. [PMC free article] [PubMed]
  • Lipton, J., G. Kleemann, R. Ghosh, R. Lints and S. Emmons, 2004. Mate searching in Caenorhabditis elegans: a genetic model for sex drive in a simple invertebrate. J. Neurosci. 24: 7427–7434. [PubMed]
  • Liu, K. S., and P. W. Sternberg, 1995. Sensory regulation of male mating behavior in Caenorhabditis elegans. Neuron 14: 79–89. [PubMed]
  • Loer, C. M., and C. J. Kenyon, 1993. Serotonin-deficient mutants and male mating behavior in the nematode Caenorhabditis elegans. J. Neurosci. 13: 5407–5417. [PubMed]
  • McGraw, L. A., G. Gibson, A. G. Clark and M. F. Wolfner, 2004. Genes regulated by mating, sperm, or seminal proteins in mated female Drosophila melanogaster. Curr. Biol. 14: 1509–1514. [PubMed]
  • Schedl, T., and J. Kimble, 1988. fog-2, a germ-line-specific sex determination gene required for hermaphrodite spermatogenesis in Caenorhabditis elegans. Genetics 119: 43–61. [PMC free article] [PubMed]
  • Schindelman, G., A. J. Whittaker, J. Y. Thum, S. Gharib and P. W. Sternberg, 2006. Initiation of male sperm-transfer behavior in Caenorhabditis elegans requires input from the ventral nerve cord. BMC Biol. 4: 26. [PMC free article] [PubMed]
  • Simon, J. M., and P. W. Sternberg, 2002. Evidence of a mate-finding cue in the hermaphrodite nematode Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA 99: 1598–1603. [PMC free article] [PubMed]
  • Stewart, A. D., and P. C. Phillips, 2002. Selection and maintenance of androdioecy in Caenorhabditis elegans. Genetics 160: 975–982. [PMC free article] [PubMed]
  • Stothard, P., and D. Pilgrim, 2006. Conspecific and interspecific interactions between the FEM-2 and the FEM-3 sex-determining proteins despite rapid sequence divergence. J. Mol. Evol. 62: 281–291. [PubMed]
  • Stothard, P., D. Hansen and D. Pilgrim, 2002. Evolution of the PP2C family in Caenorhabditis: rapid divergence of the sex-determining protein FEM-2. J. Mol. Evol. 54: 267–282. [PubMed]
  • Sulston, J. E., D. G. Albertson and J. N. Thomson, 1980. The Caenorhabditis elegans male: postembryonic development of nongonadal structures. Dev. Biol. 78: 542–576. [PubMed]
  • Ward, S., and J. S. Carrel, 1979. Fertilization and sperm competition in the nematode Caenorhabditis elegans. Dev. Biol. 73: 304–321. [PubMed]
  • Yoda, A., H. Sawa and H. Okano, 2000. MSI-1, a neural RNA-binding protein, is involved in male mating behavior in Caenorhabditis elegans. Genes Cells 5: 885–895. [PubMed]

Articles from Genetics are provided here courtesy of Genetics Society of America
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...