Self- and cross-fertilization in the solitary ascidian Ciona savignyi.
Hermaphroditism is a universal character of ascidians (Satoh, 1994). In solitary ascidians, gametes of both sexes are often released to the environment simultaneously. In some species, such as Halocynthia roretzi and Ciona intestinalis, a self-incompatibility mechanism is in place to avoid self-fertilization (autologous fertilization) (Morgan, 1944; Rosati and de Santis, 1978; Fuke and Numakunai, 1982). Only sperm of different individuals of the same species can bind and fertilize the mature oocytes (cross-fertilization or homologous fertilization). It is thought that this constraint is achieved by gamete surface proteins, encoded by several highly polymorphic gene loci, that distinguish self from non-self (Morgan, 1944; Marino et al., 1998; Sawada et al., 2004). Self-incompatibility is believed to have evolved to prevent inbreeding or selfing.
In contrast to H. roretzi and C. intestinalis, several solitary ascidians of the families Ascidiidae and Corellidae are self-fertile (Svane and Young, 1989). Whether they self-fertilize in nature, thereby running the risk of inbreeding depression, is not known because these species can also cross-fertilize. The likelihood that selfing occurs in the wild has been inferred from internal fertilization and from the absence of inbreeding depression over one generation in a selfing species. Corella inflata (Cohen, 1996). However, the possibility of cross-fertilization and inbreeding depression over several generations was not ruled out. The lack of experimental approaches, such as genetic markers with which to trace the offspring of self- and cross-fertilization of self-fertile individuals, has limited investigation in this area. Recently, however, ascidian genetic mutants (Jiang et al., 2005a, b) and a stable line with a genetic marker (Deschet et al., 2003) have become available.
We report here on the ability of a Santa Barbara, California, population of the solitary ascidian Ciona savignyi Herdman, 1882 to self-fertilize. Solitary individuals of C. savignyi are rarely found in the wild, and the individuals within a group will spawn in concert to environmental cue (i.e., sunrise), simultaneously releasing both sperm and eggs into the sea to be fertilized externally. Because of this mode of reproduction, we asked whether self-fertilization could occur when non-self sperm were also present. We find that although most individuals of C. savignyi are capable of self-fertilization, there is a strong preference for cross-fertilization.
Materials and Methods
Wild animals. Specimens of Ciona savignyi were collected several times between February 19 and July 11, 2003, from different spots within Santa Barbara Yacht Harbor (Santa Barbara, CA). These animals were used to analyze self-fertilization in C. savignyi and were the source of non-self sperm used in the competition assays described below.
Laboratory strains. The strains aimless (aim) and immaculate (imc) are recessive nonlethal mutants deficient in notochord morphogenesis and pigmentation of the sensory organs, respectively (Jiang et al., 2005a, b). Whereas homozygous aim embryos have short tails due to a mutation in the prickle gene (Jiang et al., 2005a), homozygous imc embryos have a defect in melanin biosynthesis and fail to develop pigmentation in the sensory vesicle (Jiang et al., 2005b). Self-fertilized offspring of homozygous aim or imc adults are 100% mutant. Bra::GFP is a stable transgenic line carrying a C. intestinalis Brachyury promoter followed by a coding sequence for green fluorescent protein (GFP) (Deschet et al., 2003). Bra::GFP animals used in this study are heterozygous for the transgene. These laboratory strains were the source of self-sperm used in the competition assays (see below).
Assessment of self-fertilization in wild C. savignyi
Wild-caught animals were kept in constant light for one week to accumulate gametes. They were then placed in separate cups containing about 200 ml of seawater and subjected to an 8-h dark period. Spawning was induced by illumination and was allowed to proceed for about 90 min. The spawned eggs (typically 200-300 in a clutch) were collected, washed of sperm, and transferred to a 6-cm petri dish. The level of self-fertilization was assessed by observing cleavage-stage embryos or hatched tadpoles, and expressed as the percentage of embryos or tadpoles in the total number of embryos or tadpoles, plus unfertilized eggs.
The sperm competition assay
Solitary ascidians were kept in constant light for 7 days to accumulate gametes. Self-sperm and self-eggs were dissected from a single homozygous aim (or imc or heterozygous Bra::GFP) animal, and non-self sperm were obtained from a single wild-type animal. Approximately 200 eggs were added into 20 ml of seawater in a 6-cm petri dish. Dissected sperm were stored on ice and then diluted with seawater to an O[D.sub.260] of 1.0. This optical density corresponds to about 2 X [10.sup.9] sperm/ml, as determined with a hemacytometer counting chamber (American Optical, Buffalo, NY). Ten microliters of O[D.sub.260] 1.0 self-sperm, or 10- and 100-fold dilutions, were added to 20 ml of seawater containing eggs, to give final sperm concentrations of 1 X [10.sup.6], 1 X [10.sup.5], or 1 X [10.sup.4] sperm/ml, respectively, with or without the presence of an equal volume of non-self sperm of various concentrations. Self-fertilization was assessed by observing the percentage of tadpoles having short-tail as the result of fertilization of aim eggs by aim sperm.
Egg chorions were removed using a protease cocktail (10 ml seawater, 1% sodium thioglycolate [Sigma T0632], 0.01% protease [Sigma P5147], at pH 10). Eggs were transferred to 10 ml of the protease cocktail in an agarose-coated petri dish and mixed gently with a Pasteur pipette until the majority of eggs were stripped of their chorions (usually 5 min). The protease cocktail was then replaced by fresh seawater, and the eggs were washed four times in fresh seawater to remove residual protease. An aliquot containing about 200 eggs was transferred to an agarose-coated petri dish containing 20 ml of seawater to be fertilized.
C. savignyi is highly self-fertile
A total of 121 animals were collected from the Santa Barbara Yacht Harbor and were induced to spawn by light. The ability of gametes from these animals to self-fertilize, as assessed at the early cleavage stage, is summarized in Figure 1. Similar results (data not shown) were obtained when self-fertilization was assessed 24 h after spawning. Within the typical clutch size of 200-300 eggs from a single animal, the percentage of eggs fertilized by self-sperm ranged from 0% to 100%. Only 5 of the 121 animals examined showed no ability to self-fertilize. In contrast, 51 animals produced gametes that had a self-fertilization rate of greater than 90%; of these, 16 were 100% self-fertilized.
Non-self sperm out-compete self sperm
We then took advantage of laboratory strains that carry phenotypic markers, which allowed us to assay the competition between self- and cross-fertilization. In our competition assay, we provide mature eggs from an aim/aim (or imc/imc, data not shown) individual with self-sperm, in the absence or the presence of sperm from an unrelated wild-type individual (Fig. 2). When non-self sperm were absent, self-sperm could readily fertilize self-eggs, producing tadpoles with the short-tailed aim phenotype. In the presence of non-self wild-type sperm, no short-tailed tadpoles were observed, even when the self-sperm was in 10-fold excess over non-self sperm. Identical results were obtained with both mutations. We repeated the competition assay using the Bra::GFP transgenic strain that carries a dominant marker--the expression of GFP in the notochord; the results with this strain were identical (data not shown). Taken together, we conclude that the non-self sperm out-compete self-sperm.
Cross-fertilization is faster than self-fertilization
Previous studies have shown that fewer self-sperm are able to bind to self-eggs, and that self-fertilization occurs at a slower rate than cross-fertilization (Kawamura et al., 1987). To further investigate the kinetics of self/non-self sperm competition, we performed a time-course study (Fig. 3). Self-aim sperm were provided to self-eggs at the beginning of the experiment, while non-self sperm were added at progressively later times. The percentages of mutant tadpoles produced by self-fertilization, and wild-type tadpoles by cross-fertilization, were scored. When self-sperm were mixed with the eggs for 1 h before the addition of non-self sperm, 37% of the tadpoles were short-tailed; but when self-sperm were added 2 h before non-self sperm, 89% of the tadpoles were short-tailed.
The kinetics of self- and cross-fertilization was also compared in an assay where the time required for fertilization by self- and non-self sperm was measured. Self-sperm appeared to require about 60-90 more minutes to fertilize than non-self sperm, which required 10-15 min (data not shown.) These data confirm the competition assay result, and together they indicate that self-fertilization is slower than cross-fertilization.
Self- and non-self sperm fertilize naked self-eggs equally well
To test whether the chorion, or the vitelline coat, is the site of discrimination, we performed the competition assay using de-chorionated mature eggs. To avoid polyspermy, we used 10-fold fewer sperm to fertilize. Their relative concentrations notwithstanding, when both self- and non-self sperm were present, about half of the tadpoles were mutant (by self-fertilization) and the other half were wild-type (by cross-fertilization) (Fig. 4). Thus, self/non-self discrimination takes place only at the chorion, and both self- and non-self sperm perform equally well on naked eggs.
Our study reveals that many of the Ciona savignyi in Santa Barbara can self-fertilize. By taking advantage of genetic markers developed in our laboratory, we demonstrated that non-self sperm overwhelmingly outcompete self-sperm in a fertilization competition assay. More time is required for self-sperm to fertilize. Furthermore, the discrimination between self- and non-self sperm requires the presence of the chorion. We conclude from these observations that self/non-self gamete recognition in C. savignyi is not absolute, but relative.
Our results with C. savignyi demonstrate that 42% of the animals are self-fertile, and less than 10% of the individuals in the Santa Barbara population exhibited complete self-sterility. In contrast, the only previous study of C. savignyi used a limited number of animals collected from Long Beach, California, and concluded that C. savignyi is self-sterile (Byrd and Lambert, 2000). This discrepancy may be understood in the light of work on C. intestinalis, a congeneric species. Unlike C. savignyi, C. intestinalis has been the subject of many fertilization studies, and although not unanimous, the prevailing conclusion is that C. intestinalis is not completely self-sterile. In defining self-sterility and self-fertility, several authors have set the cut-off point for self-fertility at those individuals that show greater than 90% of eggs fertilized with self-sperm. By this criterion, Rosati and De Santes (1978) reported that 15% of C. intestinalis from the Gulf of Naples were self-fertile, while in the rest (85%), less than 10% of the eggs were self-fertilized. In a study on a Japanese population in the Uranouchi Inlet, 21% of the animals could self-fertilize, while 71% could fertilize less than 10% of self-eggs (Kawamura et al., 1987). In a series of experiments conducted between January 1930 and May 1935 using animals collected from Newport Harbor, California, Morgan (1938) recorded that less than 10% of individuals were self-fertile, and 68% could fertilize less than 10% of self-eggs. In an extreme case, self-fertile individuals were completely absent from a population in Tokyo Bay (Murabe and Hoshi, 2002).
Environmental, experimental, and population factors may contribute to the differences among these results. Environmental conditions such as location and season might cause reproductive behaviors to differ and fluctuate (Millar, 1971; Lambert and Lambert, 2003). Our study reveals that C. savignyi from the Santa Barbara population in the spring and summer of 2003 had the potential to self-fertilize. Three experimental variables--sperm concentration (Rosati and de Santis, 1978), time allowed for fertilization (Kawamura et al., 1987), and egg condition (Marino et al., 1998)--may also affect fertilization. Vastly different experimental procedures have been used in the above-cited studies. In our experiment, we provided sufficient seawater (200 ml) to avoid a high concentration of sperm in the self-spawn experiment. The estimated upper limit for sperm concentration is about 1 X [10.sup.6] sperm/ml, which is within the range used in previously published fertilization experiments (Rosati and de Santis, 1978). Time was limited to 90 min to prevent prolonged exposure of eggs to self-sperm. Previous studies have demonstrated that self/non-self discrimination is established progressively during oogenesis in both Halocynthia roretzi (Fuke and Numakunai, 1996; Sawada et al., 2004) and C. intestinalis (Marino et al., 1998). Because self-fertile immature oocytes become self-sterile spontaneously by 3 h after germinal vesicle breakdown in vivo as well as in vitro, we allowed animals one week to accumulate mature gametes, followed by natural spawning instead of dissection to avoid contamination of immature oocytes. To further rule out the possibility of self-fertilization of immature oocytes, we incubated eggs from self-fertile individuals in vitro for 0, 3, 5, and 8 h before they were fertilized with self-sperm, and found that the fertilization rates were indistinguishable (data not shown). Taken together, we conclude that C. savignyi, or at least the Santa Barbara population, is highly self-fertile.
Other than C. savignyi (family Cionidae) described here, several self-fertile species (in the families Ascidiidae and Corellidae) of solitary ascidians have been reported (Svane and Young, 1989). Whether the absence of a block to self-fertilization indicates that eggs are routinely self-fertilized in the field is not known, although the converse (self-sterility) clearly implies that self-fertilization does not occur in nature. This issue could not be addressed before in these species because no genetic markers were available to trace the offspring from either self- or cross-fertilization. In the colonial ascidian Botryllus schlosseri, pigmentation differences were used to study self- and cross-fertilization (Sabbadin, 1971). It was concluded that self-fertilization is avoided because sperm of the same clone (self) mature 2-3 days after the eggs mature, during which time the eggs are fertilized by sperm from clones that are genetically different (non-self.) Self-fertilization can occur artificially when sperm maturation is accelerated and these sperm are used to inseminate self-eggs. Taking advantage of genetic markers developed recently in our laboratory, we have now approached this question in C. savignyi. Because the mutations and transgene involve very different processes, and because these strains have been bred in our laboratory for several years with no obvious reduced viability relative to wild type, we are confident that the failure of self-sperm to compete with non-self sperm is not related to the mutations and transgene. The results of our competition assay clearly indicate that self-fertilization is not favored when non-self sperm are present, a situation mostly likely found in the wild.
Why does C. savignyi have the ability to self-fertilize when cross-fertilization always overwhelmingly out-competes selfing? One could imagine that for a hermaphroditic species, self-fertilization would make it possible to colonize a new location single-handedly. This is the "reproductive assurance" hypothesis (Charlesworth and Wright, 2001). Whether this actually has happened is not known. However, our attempts to generate inbred strains through continuous selfing have never been successful (W. C. Smith laboratory, unpubl.), suggesting that inbreeding depression is a strong force against selfing in C. savignyi. Other investigators have made similar observations in C. intestinalis. For example, Murabe and Hoshi (2002) reported that the fitness of F1 generation offspring was dramatically reduced by selfing, when compared with that of offspring from cross-fertilization. Kano et al. (2001) were able to carry out successive selfing only to the fourth generation before animals became unviable. These results argue against the hypothesis that ability to self is reserved for the special situation when a single individual needs to propagate. However, it is possible that selfing occurs when a population, under environmental stresses, arrives at a critically low density, where remaining animals are sparse and cross-fertilization is impossible (Ghiselin, 1969). Progeny of the self-fertilization can quickly increase the population density and shorten the distances among individuals so that cross-fertilization can proceed, and with the improvement of the environment, the population recuperates. The fact that cross-fertilization occurs more rapidly than self-fertilization suggests a possible adaptive mechanism that allows self-fertilization to occur only as a "last resort." Because the eggs and sperm are shed simultaneously, the eggs will initially be exposed to a high concentration of self-sperm, and in the absence of such a mechanism would be preferentially self-fertilized were cross- and self-fertilization to proceed at the same rate.
Alternatively, the coexistence of selfing ability and self/non-self discrimination may be a "transitional state" of an evolutionary process towards complete self-incompatibility. Self-incompatibility in hermaphroditism is one of two major strategies in multicellular organisms to avoid inbreeding depression (the other being dioecy) (Bell, 1982). Self-incompatibility in ascidians can be achieved by at least two strategies: desynchronized spermatogenesis and oogenesis, as found in B. schlosseri, and the expression of polymorphic incompatibility gene products that recognizes and blocks self-gametes, as in C. intestinalis, H. roretzi, and C. savignyi. It is reasonable to theorize that the polymorphic incompatibility system evolved after the emergence of hermaphroditism, because this type of system was unnecessary before hermaphroditism. By this reasoning, the first hermaphrodites were self-compatible. Nevertheless, inbreeding depression must have presented itself, from the first moment in the history of hermaphroditism, as a strong selection pressure for some kind of incompatibility to evolve.
We do not know which, if either, of these two hypotheses, the "low-density model" or the "transitional-state model," explains the selfing capacity/self-incompatibility in C. savignyi. A synthesis of the two may best account for our observations. We speculate that the evolutionary process did indeed begin with self-compatibility. Gradually, self-incompatibility evolved to avoid inbreeding depression. In the meantime, the "low-density" crisis encountered by these solitary hermaphrodites presented itself as a strong force to maintain some selfing ability. The current state reflects an equilibrium in which both selfing capability and self/non-self discrimination are maintained.
We are thankful to Mr. Ed Lowry and Dr. Shota Chiba for their helpful discussion, and to Drs. Kathleen R. Foltz and Matthew J. Kourakis for their critical comments on the manuscript. This work was supported by a grant from the National Institutes of Health (HD38701) to W. C. S.
Bell, G. 1982. The Masterpiece of Nature. University of California Press, Berkeley.
Byrd, J., and C. C. Lambert. 2000. Mechanism of the block to hybridization and selfing between the sympatric ascidians Ciona intestinalis and Ciona savignyi. Mol. Reprod. Dev. 56: 541.
Charlesworth, D., and S. I. Wright. 2001. Breeding systems and genome evolution. Curr. Opin. Genet. Dev. 11: 685-690.
Cohen, C. S. 1996. The effects of contrasting modes of fertilization on levels of inbreeding in the marine invertebrate genus Corella. Evolution 50: 1896-1907.
Deschet, K., Y. Nakatani, and W. C. Smith. 2003. Generation of Ci-Brachyury-GFP stable transgenic lines in the ascidian Ciona savignyi. Genesis 35: 248-259.
Fuke, M. T., and T. Numakunai. 1982. Allogenic cellular reactions between intra-specific types of a solitary ascidian, Halocynthia roretzi. Dev. Comp. Immunol. 6: 253-261.
Fuke, M. T., and T. Numakunai. 1996. Establishment of self-sterility of eggs in the ovary of the solitary ascidian, Halocynthia roretzi. Roux's Arch. Dev. Biol. 205: 391-400.
Ghiselin, M. T. 1969. The evolution of hermaphroditism among animals. Q. Rev. Biol. 44: 189-208.
Jiang, D., E. M. Munro, and W. C. Smith. 2005a. Ascidian prickle regulates both mediolateral and anterior-posterior cell polarity of notochord cells. Curr. Biol. 15: 79-85.
Jiang, D., J. W. Tresser, T. Horie, M. Tsuda, and W. C. Smith. 2005b. Pigmentation in the sensory organs of the ascidian larva is essential for normal behavior. J. Exp. Biol. 208: 433-438.
Kano, S., S. Chiba, and N. Satoh. 2001. Genetic relatedness and variability in inbred and wild populations of the solitary ascidian Ciona intestinalis revealed by arbitrarily primed polymerase chain reaction. Mar. Biotechnol. (NY) 3: 58-67.
Kawamura, K., H. Fujita, and M. Nakauchi. 1987. Cytological characterization of self incompatability in gametes of the ascidian, Ciona intestinalis. Dev. Growth Differ. 29: 627-642.
Lambert, C. C., and G. Lambert. 2003. Persistence and differential distribution of nonindigenous ascidians in harbors of the Southern California Bight. Mar. Ecol. Prog. Ser. 259: 145-161.
Marino, R., M. R. Pinto, F. Cotelli, C. L. Lamia, and R. De Santis. 1998. The hsp70 protein is involved in the acquisition of gamete self-sterility in the ascidian Ciona intestinalis. Development 125: 899-907.
Millar, R. H. 1971. The biology of ascidians. Adv. Mar. Biol. 9: 1-100.
Morgan, T. H. 1938. The genetic and the physiological problems of self-sterility in Ciona. I. Data on self- and cross-fertilization. J. Exp. Zool. 78: 271-318.
Morgan, T. H. 1944. The genetic and the physiological problems of self-sterility in Ciona. VI. Theoretical discussion of genetic data. J. Exp. Zool. 95: 37-59.
Murabe, N., and M. Hoshi. 2002. Re-examination of sibling cross-sterility in the ascidian, Ciona intestinalis: genetic background of the self-sterility. Zool. Sci. 19: 527-538.
Rosati, F., and R. De Santis. 1978. Studies on fertilization in the ascidans. I. Self-sterility and specific recognition between gametes of Ciona intestinalis. Exp. Cell Res. 112: 111-119.
Sabbadin, A. 1971. Self- and cross-fertilization in the compound ascidian Botryllus schlosseri. Dev. Biol. 24: 379-391.
Satoh, N. 1994. Developmental Biology of Ascidians. Cambridge University Press, Cambridge.
Sawada, H., E. Tanaka, S. Ban, C. Yamasaki, J. Fujino, K. Ooura, Y. Abe, K. Matsumoto, and H. Yokosawa. 2004. Self/nonself recognition in ascidian fertilization: vitelline coat protein HrVC70 is a candidate allorecognition molecule. Proc. Natl. Acad. Sci. USA 101: 15615-15620.
Svane, I., and C. M. Young. 1989. The ecology and behaviour of ascidian larvae. Oceanogr. Mar. Biol. Annu. Rev. 27: 45-90.
DI JIANG* AND WILLIAM C. SMITH
Department of Molecular, Cellular, and Developmental Biology, University of California
Santa Barbara 93106
Received 3 May 2005; accepted 29 July 2005.
* To whom correspondence should be addressed. E-mail: firstname.lastname@example.org
|Printer friendly Cite/link Email Feedback|
|Author:||Jiang, Di; Smith, William C.|
|Publication:||The Biological Bulletin|
|Date:||Oct 1, 2005|
|Previous Article:||Endosymbiotic yeast maternally transmitted in a marine sponge.|
|Next Article:||Gamete compatibility and sperm competition affect paternity and hybridization between sympatric Asterias sea stars.|