Cytological characterization of premeiotic versus postmeiotic defects producing hybrid male sterility among sibling species of the Drosophila melanogaster complex.
While a great deal of effort has gone into determining the genetic basis of hybrid sterility, the actual "phenotype" of hybrid male sterility has typically been ignored. From Dobzhansky's (1937a) method of measuring testis size to the prevailing method of detecting the presence of motile sperm (Zouros 1981; Coyne 1985), existing fertility assays do not allow for the demonstration of specific causes of sterility. These studies have constrained the trait of sterility to be classified as an all-or-none phenomenon. However, different hybrid genotypes may possess characteristically unique spermatogenic aberrations. Such patterns as the differential degree of sterility between reciprocal interspecific crosses as well as the observed pattern of sterility classes among different [F.sub.1] hybrids may provide insight into the evolutionary progression of hybrid male sterility. Cytological studies of hybrid male sterility can readily be performed because the developmental system of spermatogenesis has been well characterized in such species as D. melanogaster and D. hydei (Lindsley and Tokuyasu 1980; Fuller 1993). Also, the germ-cell autonomous nature of hybrid male sterility (Dobzhansky and Beadle 1936) allows for a truly independent appraisal of this developmental process.
Interspecific hybridizations between members of the melanogaster complex, with exception of one cross, follow the script of Haldane's rule. This complex includes four sibling species (D. melanogaster, D. simulans, D. sechellia, and D. mauritiana) that are morphologically identical except for conspicuous differences in the posterior process of their genital arches (Sturtevant 1919; Tsacas and David 1974; Tsacas and Bachli 1981). When D. simulans, D. sechellia, or D. mauritiana females are crossed to D. melanogaster males, sterile males result with no adult female progeny (Sturtevant 1920; David et al. 1974). This breach of Haldane's rule does not arise in the reciprocal cross, as only sterile adult female progeny are produced. Strains have been discovered from both D. simulans and D. melanogaster that can rescue the viability of the "absent" hybrid sex (Watanabe 1979; Hutter and Ashburner 1987; Hutter et al. 1990; Sawamura et al. 1993a,b). Recently, particular strains of these two species have been shown to rescue the fertility of females when hybridized in one direction of the cross (Davis et al. 1996). However, between all three sibling species of the simulans clade (D. simulans, D. sechellia, and D. mauritiana) interspecific hybridizations produce fertile female and sterile male progeny (Lachaise et al. 1988).
In this study, six sterile [F.sub.1] hybrid male genotypes produced from three sibling species of the D. simulans clade [ILLUSTRATION FOR FIGURE 1 OMITTED] were cytologically characterized for specific spermatogenic defects. Each hybrid genotype exhibited a characteristic sterility phenotype reflecting the independent evolution of sterility factors within each parental species. Furthermore, the sterility of a particular interspecific hybrid could be classified as the termination of spermatogenesis either before meiosis (premeiotic) or during the spermiogenic stages (postmeiotic). We also present evidence of a segregating allele, in D. simulans, that may cause the reversion of hybrid male sterility types when crossed to D. sechellia.
MATERIALS AND METHODS
Origin of Drosophila Stocks
Two wild type strains from each of D. simulans (denoted sim), D. sechellia (denoted sec) and D. mauritiana (denoted mau) were used in this study. These strains were: D. simulans Colombia, South Africa; D. sechellia 21, 24; D. mauritiana David 105, LG24. All strains were obtained from the Bloomington Drosophila Species Stock Centre, with the exception of D. mauritiana LG24, which was procured from Jean David. In addition, a Peruvian strain of D. melanogaster (denoted mel) was obtained from the National Drosophila Species Resource Centre at Bowling Green. To detect any apparent cytoplasmic effect stemming from D. simulans, an attached-X line, stock 2119 ([yvmgf.sup.2]/C(1)RMyw), was also obtained from Bloomington. All stocks were maintained in 250-mL glass jars with foam plugs on approximately 25 mL of standard banana medium. These stocks were reared at 22-23 [degrees] C in an incubator under a 12:12 L:D photoperiod.
All hybridizations (within and between strain as well as between species) took place in 35-mL glass vials with foam plugs and approximately 5 mL of standard banana medium. Both virgin male and female flies were collected from stock cultures within a five-day period prior to mating. Flies were anaesthetized using low levels of C[O.sub.2] to separate the sexes. In all matings, 10 females were crossed to 10 males with the exception of one cross (in the interspecific cross between D. sechellia females and D. simulans males, 15-20 males were mated to 10 females). Progeny from this cross was difficult to produce, however repeated transfers of the parents eventually yielded offspring. In all crosses, parents were transferred to fresh vials every five to seven days for a period of four weeks. Crossed flies and their progeny were subjected to the same timed-light and temperature conditions as flies from the original stock cultures.
Whole Mount Testes Analysis
Within 16 hours after their eclosion, the reproductive tract was retrieved from live males submerged under a drop of Hennig's testes buffer using a Zeiss dissecting microscope, (Ashburner 1989). Testes and seminal vesicles were dissected from the rest of the reproductive tract tissue. A Corning 22-[mm.sup.2] coverslip was gently placed on top of the testes preparation.
The testes were observed for the presence of various spermatogenic landmarks under brightfield, phase-contrast, and Nomarski optics using a Zeiss Axioplan microscope. The apical end (or apex) of the testes was defined as the area in the closed-ended tip of the testis, which is one-tenth the total testis' length. The midtestes region was approximated to be the area (one-tenth the total length of the testis) surrounding the exact midpoint of the testis. The testis' basal region was defined as the area traversing the final tenth of the testis' length. If too many sperm bundles were present at a certain location, the maximum number of sperm bundles that could clearly be resolved was listed in conjunction with a greater-than symbol (>). Emphasis was placed on midtestes sperm bundle counts, as they give a reliable quantification of fertility. (The maximum number of sperm bundles is found in the midtestes region because this site represents the origin of sperm bundle elongation; Tokuyasu et al. 1972.) Also, the midtestes region in these species does not contain coiled or "looped" sperm bundles, which may affect the accuracy of the total sperm bundle count.
Sectioning of Testes
Testes were extracted from three-day-old virgin adult males that were submerged in Schneider's liquid medium. These flies were collected within eight hours of eclosion. They were aged in standard vial conditions at low density from their time of collection until 66 h later. Hence, each adult testis was between 66 h to 74 h old posteclosion. Dissected testes were bathed in a primary fixative consisting of 0.1 M cacodylate buffer, 1% paraformaldehyde, and 2.5% gluteraldehyde at room temperature. Testes were then washed in 0.1 M cacodylate buffer (6 x 10 min) and postfixed on chilled 1.0% osmium tetroxide in 0.1 M cacodylate buffer for 1 h. After an ethanol dehydration series, testes pairs were separated and transferred to propylene oxide. Only one testis was randomly selected from each individual for analysis. Propylene oxide was gradually replaced by Spurr's resin and placed on a shaker for 24 h. Infiltrated testes were baked at 65 [degrees] C for 8 h.
An RMC MT-7 microtome was used to cut the embedded testes into 2-[[micro]meter] sections. These sections were collected on a slide, briefly stained with 1% toluidine blue in borax at 65 [degrees] C, sealed under a coverslip, and observed using brightfield microscopy. Sections from the midtestes region were quantified for the number of mature sperm bundles (cysts containing 64 spermatids), premeiotic cysts (cysts containing 16 or fewer cells) or pre-elongation cysts (cysts that contain 32 or 64 immature spermatids of large cross-sectional area).
Thin sections (60-90 nm thickness) were produced from the embedded testes (see above) and placed onto a formvar-coated copper slot grid. Dried grids were first stained with a 1:1 mixture of uranyl acetate saturated in [H.sub.2]O and ethanol and second with lead citrate. Stained sections were viewed using a JEOL 1200 E II transmission electron microscope operated at 80 kV. Images were photographed on Kodak 4489 film.
Using electron microscopy, the furthest normally developed spermatids (i.e., the most developmentally mature spermatids), were characterized. Testes from each hybrid genotype are observed to arrest developmentally at common stages of spermiogenesis. The markers of developmental stage, used in this assessment, included axonemal-mitochondrial derivative relationships as well as various characteristics of microtubule maturation.
Attached-X (2119) females from D. simulans were crossed to both D. sechellia and D, mauritiana males to test the influence of D. simulans-derived cytoplasm on the sterility of the [F.sub.1] hybrids [ILLUSTRATION FOR FIGURE 1 OMITTED]. However, the presence of sterility differences between genotypically identical [F.sub.1] hybrids does not necessarily prove the existence of a cytoplasmic influence on hybrid male sterility. The presence of D. simulans 2119-specific alleles on the autosomes and Y-chromosome may produce a confounding effect on the hybrid. Factors that may be unique to the D. simulans 2119 strain were recombined with a South African wild-type strain of D. simulans by back-crossing female hybrids for two generations. This novel line contained free X and Y-chromosomes that were completely South African in origin, autosomes and maternal effects that were largely South African in origin, and cytoplasms that fully retained their original 2119 genetic identity [ILLUSTRATION FOR FIGURE 2 OMITTED]. Therefore, the influence of D. simulans cytoplasm (from two diverse lines of D. simulans) on [F.sub.1] hybrid male sterility could be assessed.
Although there may appear to be differences in spermatogenic characters between species (Table 1), only differences among strains within species were statistically significant using nested ANOVA tests. The number of sperm bundles in midtestes cross-sections had a significant within-species between-strain effect ([F.sub.4,32] = 4.05, P = 0.009), but not a significant between-species effect ([F.sub.3,4] = 2.11, P = 0.242). However, neither the number of premeiotic cysts nor the number of pre-elongation spermatid cysts observed in midtestes cross-sections showed significant species or strain effects (nested ANOVA; P [greater than] 0.1). The number of sperm bundles projecting into the apical region of the whole mount testes was another variable trait (Table 1). The two D. sechellia strains display the lowest strain averages, but because of large between-strain variation, it is difficult to assess the significance of species differences. Kolmogorov-Smirnov two-sample tests were used to compare the sperm bundle distribution of D. sechellia 21 (mean = 0.7 [+ or -] 1.1) to both D. simulans 2119-XX (mean = 1.5 [+ or -] 2.0) and D. mauritiana David 105 (mean = 2.0 [+ or -] 1.8). In neither case were the differences significant (KS [[Chi].sup.2] = 4.15, P = 0.308 and KS [[Chi].sup.2] = 9.45, P = 0.124, respectively), which indicates that D. sechellia variation overlaps that of D. simulans and D. mauritiana.
D. simulans/D. sechellia Hybrids
The two types of [F.sub.1] hybrid genotypes produced from this species pair yielded hybrids that progressed normally through both the premeiotic and meiotic stages of spermatogenic development. The presence of sperm bundles in the testes of both reciprocal hybrids and the corresponding absence of individualized spermatids indicates a failure in the postmeiotic or spermiogenic stage of sperm development [ILLUSTRATION FOR FIGURE 3 OMITTED]. Whole mount testes surveys (within 16 h posteclosion) and cross-sections (66-72 h posteclosion) confirm the presence of elongated sperm bundles containing approximately 64 spermatids (data not shown).
Although the quantities of cross-sectional midtestes sperm bundles from the hybrids of the two reciprocal crosses of wild-type D. simulans and D. sechellia were not significantly different from each other (Mann-Whitney test; [n.sub.1] = [n.sub.2] = 5, [U.sub.s] = 12, P = 0.5; Table 2), other differences in testes structure were observed. In the whole mount analysis (testes less than 16 h old posteclosion), there was a complete absence of sperm bundles found at the apex of the testes in hybrids of D. sechellia (female) x D. simulans (male; Table 2). Also, the sperm bundles in the midtestes regions contained less densely packed spermatids (especially on the perimeter of the testis lumen) compared to the tightly demarcated sperm bundles of its reciprocal hybrid. The number of sperm bundles found at the midtestis region at three days posteclosion was similar in both reciprocal hybrids, but a difference in midtestis sperm bundle content became apparent between sim/sec [F.sub.1] hybrids in males older than five days posteclosion. In D. simulans (female) X D. sechellia (male) hybrids, an increase in sperm bundle number through time (day 0-10) took place, indicating a continuous progression of sperm bundle development. However, in the reciprocal hybrid, adult males manifested either an abrupt decrease in spermatogenic production or drastic degeneration in sperm bundles after five days (data not shown).
Ultrastructural analysis of both sim/sec [F.sub.1] hybrids revealed similarities in axonemal development; however, differences in microtubule development were also observed between the two hybrids, with D. simulans (female) x D. sechellia (male) hybrids possessing further developed microtubules (Table 3). In addition, these hybrids contained a relatively smaller amount of spermatid cytoplasmic debris than all other male hybrids of the clade [ILLUSTRATION FOR FIGURE 3 OMITTED]. It must be mentioned that only one strain per species was used in the ultrastructural survey, therefore between-strain differences were not studied.
D. simulans/D. mauritiana Hybrids
In the genotypes produced from the hybridization of D. simulans and D. mauritiana, two classes of sterility (premeiotic and postmeiotic) were observed. Drosophila mauritiana (female) x D. simulans (male) [F.sub.1] hybrids were completely devoid of sperm bundles (Table 2). The midtestes region was instead full of premeiotic cysts of varying stages, ranging from two to eight cells each. Sixteen-cell cysts were rarely seen.
In striking contrast to the above aspermic condition, D. simulans (female) x D. mauritiana (male) hybrids had numerous sperm bundles at the apical, midtestes, and basal regions of the testes. Sperm bundles at the midtestes regions [TABULAR DATA FOR TABLE 1 OMITTED] contained loosely packed spermatids and were found in relatively smaller numbers compared to the sim/sec hybrids (Table 2). Electron microscopic analysis revealed the underdevelopment of both axonemal complexes and microtubule placement. The presence of large amounts of cytoplasm also indicates that spermatids were in a preindividualized state (Table 3, [ILLUSTRATION FOR FIGURE 3 OMITTED]).
D. sechellia/D. mauritiana Hybrids
The two sets of sec/mau hybrid genotypes originating from each of the two reciprocal crosses also possessed completely different degrees of sterility. The hybrids of the cross D. sechellia (female) x D. mauritiana (male) progressed through some of the postmeiotic stages of spermatogenesis. However, hybrids of this cross manifested a very low sperm bundle count at the midtestes region in both 16-h-old and 66-72-h-old adult males relative to the other postmeiotic sterile hybrids (Table 2). This relatively low mean sperm bundle count was parallelled in ultrastructural analysis by the presence of sperm bundles that had very large areas with large amounts of cytoplasmic debris [ILLUSTRATION FOR FIGURE 3 OMITTED]. Also, immature axonemal development was observed in conjunction with a large amount of microtubule developmental abnormalities (Table 3).
The reciprocal cross, D. mauritiana (female) x D. sechellia (male), produced aspermic males with severe sterility defects that were premeiotic in nature. Four-celled cysts were the most common spermatogenic cyst type at the midtestis region.
Cytoplasmic versus Nuclear Influence
There was a large difference in [F.sub.1] sterility phenotype between D. sechellia (female) x D. simulans (male) and D. simulans-XX (female) x D. sechellia (male) hybrids. The [F.sub.1] hybrid males from these two crosses are chromosomally identical but differ in both the specific origin of their cytoplasms and the parental origin of their X- and Y-chromosomes. The former hybrid contained large amounts of midtestes sperm bundles, whereas the attached-X hybrids possessed an apparent developmental arrest prior to spermiogenesis (Table 2, [ILLUSTRATION FOR FIGURE 4 OMITTED]). This result could be attributed to cytoplasmic factors or a parental origin effect of the X-chromosome. However, a D. simulans line-to-line difference between 2119 (attached-X) and the Colombian strain may also account for this difference. This latter hypothesis was indeed validated by backcrossing female progeny from a D. simulans 2119 (female) x D. simulans 914 (South Africa) (male) to D. simulans South Africa males, then crossing the resulting female to a D. sechellia male and observing the presence of postmeiotic defects in cross-sections of the hybrid testes [ILLUSTRATION FOR FIGURE 2 OMITTED]. Of 12 independently sampled testes cross-sections (2-4 days posteclosion), at least 10 contained postmeiotic arrests in spermatogenesis (Table 2), corresponding to the same class of postmeiotic sterility as its genotypic equivalent, D. sechellia (female) x D. simulans (male) (2 of the 12 cross-sections revealed ambiguous results - either they were poorly stained or premeiotic). This difference in the hybrid sterility phenotypes is therefore due to a strain dependent effect in D. simulans. In addition, the average number of midtestes sperm bundles in the 10 "reverted" postmeiotic steriles was significantly smaller than either of the two hybrids of the reciprocal crosses ([ILLUSTRATION FOR FIGURE 3 OMITTED]; Kruskal-Wallis; [Mathematical Expression Omitted], P = 0.0430), which indicates either an incomplete reversion of sterility or a strain effect.
Hybrids produced from the cross, D. mauritiana (female) x D. simulans (male), seemed to be identical in sterility to genotypically analogous hybrids produced by an attached-X cross (Table 2, [ILLUSTRATION FOR FIGURE 4 OMITTED]). These two hybrids did not contain any sperm bundles in the postmeiotic stages of spermatogenic development. The midtestes region consisted of premeiotic cysts of different developmental stages ranging from two to eight cells. In both cases, 16-cell cysts were rarely seen.
Generally, no significant differences in midtestes spermatogenic profile were observed between the four members of the melanogaster complex (Table 1). If one strain of D. mauritiana is excluded, however, the average amount of midtestes sperm bundles is always lowest in D. sechellia. Lachaise et al. (1986) also showed that the number of ovarioles in adult females was much smaller in D. sechellia than other species of the complex. These parallel findings, in developmentally analogous meiotic products from both males and females may stem from a common genetic pathway for both sperm and egg production or may represent two independently diverged traits found solely in D. sechellia. We must stress, however, that the limited use of different strains and the high strain variation in spermatogenic characters may make any conclusion superfluous.
Drosophila sechellia may also have differences in other aspects of testes structures. In the apical end of the testes, sperm bundles were almost absent in D. sechellia, in contrast to other species of the complex (Table 1). Longer testis length in D. sechellia relative to its sibling species (Civetta and Singh 1998) may account for the lack of sperm bundles at the testis apex. However, Joly (1987) observed a significantly longer cyst length in D. sechellia compared to that of D. simulans and D. mauritiana. The existence of other factors, such as sperm bundle coiling frequency and elongation timing, may cause this deficiency in apical sperm bundle number in D. sechellia. Alternatively, there may be an absence of species specificity because, as stated in the Results, there appeared to be a large variation in the number of apical sperm bundles between strains.
Although the overall process of spermatogenesis is almost indistinguishable between the sibling species of the simulans clade, large differences were revealed between the sterility phenotypes of most of the hybrid genotypes in this study. Such differences are expected if the genetic interactions that produce these sterility incompatibilities evolve independent of each other. Different sets of interacting loci are then assumed to affect each interspecific hybrid genotype. Two distinct classes of spermatogenic defects, premeiotic and postmeiotic, were observed among the six [F.sub.1] hybrid genotypes.
Four out of the six interspecific hybrid genotypes exhibited defects that were spermiogenic, or postmeiotic, in nature. (Cross-sections of sperm bundles revealed nearly 64 spermatids per bundle in such hybrids.) Our assignment of a particular sterility as being either premeiotic or postmeiotic is based on the criterion of the amount of cells per cyst at the time of spermatogenic termination. Therefore, although these hybrids reached the postmeiotic stages of spermatogenesis, the existence of premeiotic lesions remains a possibility. All of these hybrid genotypes displayed a wide range of abnormalities from sperm bundle dynamics to ultrastructural defects. For example, drastic reductions in sperm bundle number as well as an increase in sperm bundle size were seen in both hybrids of D. mauritiana that manifested postmeiotic defects. This may indicate that the program of spermatogenesis in these hybrids had also been perturbed at an early stage.
As stated above, the majority of spermatogenic blockages found in the hybrids of this study were postmeiotic. This parallels the pattern observed in within-species sterility whereby the majority of single gene effects causing sterility in D. melanogaster take place in the spermiogenic stages of spermatogenesis (Lindsley and Lifschytz 1972; Lifschytz 1987). In addition, the wide array of postmeiotic defects that were found within a particular hybrid male sterile also parallels those that occur in within-species spermiogenic mutants. It has been suggested that such general breakdowns in a diverse range of spermatogenic traits is the consequence of the many events found in spermatid differentiation occurring through independent pathways (Lifschytz 1987). In the hybrid stemming from the cross D. simulans 2119 (female) x D. sechellia (male), for example, spermiogenic elongation was still occurred even though the completion of normal meiosis had not [ILLUSTRATION FOR FIGURE 3H OMITTED]; [ILLUSTRATION FOR FIGURE 4 OMITTED]. This interspecific hybrid [TABULAR DATA FOR TABLE 2 OMITTED] [TABULAR DATA FOR TABLE 3 OMITTED] example of spermatogenic differentiation occurring through independent pathways has an analogous phenotype in the single gene mutant, twine. In twine mutants, chromosomal segregation and cytokinesis in male meiosis does not take place but spermatids still form and elongate (Alphey et al. 1992; Courtot et al. 1992).
However, most within-species spermiogenic mutants manifest similar sterility phenotypes. At the substructural level, early defects are difficult to resolve and numerous single gene mutants share "a common classic male sterile phenotype when examined by light microscopy" (Fuller 1993). These phenotypes include sperm bundles that elongate but are disorganized and spermatids that degenerate prior to individualization (Hackstein 1991; Castrillon et al. 1993). At the ultrastructural level, such male sterile mutants exhibit spermatids with disorganized axonemal components and oddly shaped mitochondrial derivatives (Fuller 1993). The sterility phenotypes found in postmeiotically defective hybrids in the present study may also fall under the definition of common classic male sterile phenotype. However, this description essentially corresponds to our lack of understanding of the numerous pathways involved in normal spermiogenesis.
The complexity in hybrid spermiogenic sterility may be influenced by the apparent lack of transcription in the later stages of spermatogenesis (Oliveri and Oliveri 1965; Gould-Somero and Holland 1974). It has also been shown that the presence of chromosomes are unnecessary during spermiogenesis (Lindsley and Tokuyasu 1980). The lack of compensatory mechanisms in the form of transcriptional regulation that are important in other developmental processes may then cause spermiogenesis to be less buffered against any quantitative change in gene product (Wu and Davis 1993).
It is conceivable that the genetic factors that cause a preponderance of "complex" spermiogenic phenotypes in within-species spermatogenesis may be the same factors that cause a preponderance of [F.sub.1] hybrids of the simulans clade to be postmeiotically sterile. However, Perez et al. (1993) stated that the complexity of [F.sub.1] hybrid male sterility suggests that it may be due to interactions of many genes. In contrast, it can be argued that the complexity of hybrid male sterility does not necessarily establish the presence of a large number of incompatible interactions caused by a large number of genes (polygenic nature) or strong epistatic interactions (complex epistasis), but may simply be the byproduct of a few or even a single genetic incompatibility on numerous independent pathways.
Drosophila mauritiana (female) x D. simulans (male) and D. mauritiana (female) x D. sechellia (male) manifested blockages in spermatogenic development that took place before meiosis commenced. Because female hybrids of this clade are known to be fertile (Lachaise et al. 1988), these observations support the presence of inherent genetic differences between the premeiotic processes found in male spermatogenesis and female oogenesis. Furthermore, they suggest that Haldane's rule and its genetic basis may not be solely established in spermiogenesis but also in the premeiotic stages of spermatogenesis.
The results presented in this study differ from others that have attempted to characterize hybrid male sterility in the simulans clade. Lachaise et al. (1986) described the presence of fully developed (yet aspermic) testes in D. simulans/D. sechellia and D. simulans/D. mauritiana [F.sub.1] hybrids and reported a wide range of testicular phenotypes ranging from atrophied testes to fully developed (yet aspermic) testes in [F.sub.1] hybrids of the D. sechellia/D. mauritiana species pair. Observed at a lower level of resolution, such characterizations argue that particular hybrid sterilities encompass a large range of abnormalities. In contrast, the present study relates concrete spermatogenic anomalies to particular hybrid genotypes. A number of unaccounted variables in the Lachaise et al. study, including environmental differences such as variations in temperature, humidity, and rearing medium as well as strain-dependent differences, may account for the disparity in observations. For example, an important variable not mentioned by Lachaise et al. was the age of the dissected testes. Thus, the observed variation in testes structure found in their data may be explained by the sampling of a number of testes at different ages. Alternatively, the use of different strains may have elevated the amount of variation in the traits observed. In the Lachaise et al. (1986) study, numerous non-isofemale derived lines from D. simulans and D. melanogaster were employed, which may have resulted in the variable expression of certain polymorphisms.
Other studies have also briefly characterized certain hybrids of the simulans clade. Observing the two [F.sub.1] hybrids stemming from the crosses, D. simulans (female) x D. sechellia (male) and D. simulans attached-X (female) x D. sechellia (male), Coyne and Kreitman (1986) noted that the former hybrid produced 4.2% motile sperm whereas the latter attached-X cross did not produce any sperm. Perez et al. (1993) observed that hybrids of D. simulans (females) x D. mauritiana (males) progressed through spermatogenesis until the elongation stage of spermiogenesis whereas hybrids produced from D. simulans (females) x D. sechellia (males) contained many abnormal onion cells (pre-elongation cysts) and appeared to have enlarged mitochondrial derivatives. Again, these results are not in agreement with the present study.
The relative constancy of hybrid male sterility in each particular hybrid genotype characterized in this study also differs from current opinions on the plasticity of hybrid female sterility. In their recovery of female fertility between D. melanogaster and D. simulans, Davis et al. (1996) noted large line-to-line variations in the average "fertility" of the hybrids. They attribute these results to a "strain-dependent continuum of hybrid female fertility." In our study, such large "strain-dependent continuums" were generally not observed in the sterilities of the hybrid male genotype, although differences between parental strains were recorded. The variations in spermatogenic markers found between strains seemed to have become buffered and not accentuated in hybrid male sterility. However, a clear conclusion is difficult to make because only two strains per species (i.e., four crosses) were employed in the hybridizations of this study. The only strain that created a noticeably different sterility phenotype from other strains of its species was the D. simulans 2119 attached-X line. It was observed that D. simulans 2119 when crossed to D. sechellia 3590 produced a much more severe sterility phenotype than D. simulans South Africa. Such a large change in the hybrid's sterility phenotype cannot be considered continuous, but is instead discrete and may be based on the influence of loci with large effect. The question must be asked whether this mode of change is common during the postzygotic reproductive isolation of related species.
Intraspecific hybrids of D. simulans 2119 were fertile (Table 1) although the extent of their fertilities (i.e., quantity of sperm bundles) was not assayed. This observed fertility may be the result of species-compatible modifier loci that have evolved in D. simulans 2119 to counteract the effect of this novel factor(s). In fact, the number of midtestes sperm bundles of the cross D. simulans 2119/914 (female) x D. sechellia (male) was shown to be significantly lower than genotypically equivalent hybrids of the cross, D. sechellia (female) x D. simulans (male; see Results, Table 2). It is possible that the presence of nonrecombined modifier loci in the line was the cause of this decrease in fertility.
In addition to the nuclear genic interactions that are known to cause incompatibilities in the hybrid, other factors that are inherited through the egg's cytoplasm have been thought to influence hybrid male sterility (Dobzhansky and Sturtevant 1935; Dobzhansky 1937b, 1974). There have been a number of studies on the incompatibility of the hybrid's genome with maternal factors. Orr (1989) has claimed that the difference between [F.sub.1] sterile male hybrids derived from the species D. pseudoobscura pseudoobscura and D. pseudoobscura bogotana, and [F.sub.2] males of the "same" genotypic constitution can be explained by a maternal effect. In most studies, however, the disentanglement of factors originating from the X-chromosome and factors stemming from the cytoplasm are impossible. This problem can be overcome with the use of a cross employing an attached-X line. Such crosses with attached-X females produce patriclinous sons that inherit the X-chromosome from their father and both the Y-chromosome and cytoplasm from their mother (Lindsley and Zimm 1992). In this study, an attached-X line tested the presence of cytoplasmic effects on the sterility of hybrid males of the simulans clade. Using this method, changes in the degree of sterility between [F.sub.1] hybrid males were assayed. However, because the attached-X line is solely available (at least in this clade) in D. simulans, only a unidirectional evaluation of the D. simulans cytoplasm on [F.sub.1] hybrids could be made.
The presence of maternal effects or cytoplasmic factors of a D. simulans origin that affect the sterilities of sim/mau [F.sub.1] males was not observed. The sterility phenotype of the two hybrids, D. simulans-XX (female) x D. mauritiana (male) and D. mauritiana (female) x D. simulans (male) was nearly identical in their aspermic nature (Table 2). This indicates the absence of factors transmitted through the cytoplasm (maternal effect or cytoplasmic factors) of D. simulans 2119 that can interact with an autosome or X-chromosome of D. mauritiana to effect a change in the hybrid's sterility (at least at a detectable level). Davis et al. (1994) also utilized an attached-X line with the same two species. Using recombinant [F.sub.2] males of D. simulans and D. mauritiana, they showed that neither maternal effects nor cytoplasmic determinants are factors in the sterility of hybrid males of this species pair.
The effect of D. simulans cytoplasm on hybrid male sterility in hybrids of the sim/sec species pair was also investigated in this study. A large difference in sterility phenotype was observed between the two hybrids of the crosses, D. simulans-XX (female) x D. sechellia (male) and its genotypic equivalent, D. sechellia (female) x D. simulans Colombia (male; [ILLUSTRATION FOR FIGURE 1 OMITTED], Table 2). The former hybrid manifested arrests in spermatogenesis that were solely premeiotic whereas the latter hybrid displayed postmeiotic abnormalities. This large difference in sterilities may at first seem to indicate a cytoplasmic influence. However, it was shown that a D. simulans line-to-line difference between 2119 (attached-X) and Colombia (and South Africa) accounted for this difference (for cross, [ILLUSTRATION FOR FIGURE 2 OMITTED]; for sperm bundle quantification, [ILLUSTRATION FOR FIGURE 4 OMITTED]).
Using a recurrent backcross technique, Zeng and Singh (1993) were also able to show that cytoplasmic factors do not play a role in hybrid male sterility in the simulans clade. However, this method does not allow for the evaluation of maternal effects, which are an important component of cytoplasmic influence. The use of attached-X lines allows such maternal effects to be disentangled from its natural cotransmission with the X-chromosome. In the present study, components of the cytoplasm such as maternal effects that in contrast to cytoplasmic factors, are important to the classical model of speciation (Dobzhansky 1970; Goulielmos and Zouros 1995) were able to be separated from other factors involved in hybrid male sterility. It must be noted, however, that the complete isolation of maternal effects cannot be achieved because factors found on the attached-X chromosome will always remain a part of the lineage in such crosses [ILLUSTRATION FOR FIGURE 2 OMITTED].
Differences in sterility phenotype between reciprocal crosses indicate the independence of incompatibilities in each reciprocal genotype (Orr 1995). In their two-locus model of reproductive isolation, Dobzhansky (1937b) and Muller (1942) showed that the seemingly maladaptive trait of hybrid sterility or inviability can be simply achieved as a byproduct of the independent fixation of two loci between two isolated populations. If a sex chromosome harbors one of the fixed loci, the incompatibility produced will manifest itself in only one of the reciprocal males. A number of studies (Wu and Beckenbach 1983; Vigneault and Zouros 1986) have demonstrated the presence of asymmetrical genic incompatibilities in hybrid male sterility where one reciprocal male hybrid is sterile and the other is fertile. Our study reveals the presence of distinct asymmetries in the severity of hybrid male sterility between reciprocal crosses amongst the three species studies. These asymmetries ranged from the premeiotic versus postmeiotic anomalies observed in reciprocal [F.sub.1] hybrids of the D. simulans/D. mauritiana and D. sechellia/D. mauritiana species pairs to the differential degradation of postmeiotic cysts observed in D. simulans/D. sechellia [F.sub.1] hybrids. In addition, the large jump between spermiogenic or postmeiotic incompatibilities and premeiotic ones in reciprocal hybrids (such as those involving D. mauritiana) indicate that only a small number of genic incompatibilities are involved. Alternatively, the presence of many minor genes or "complex epistasis" (Perez et al. 1993) may be present solely in the spermiogenic stages.
By observing trends of premeiotic versus postmeiotic abnormalities in [F.sub.1] hybrids, inferences about the phylogenetic history of the parental species may be entertained. Assuming the probability that substitutions affecting spermiogenesis take place more often than those affecting the earlier stages of spermatogenesis (Lindsley and Tokuyasu 1980), it could be argued that D. simulans and D. sechellia have diverged most recently. Both hybrid genotypes produced from this species pair display postmeiotic defects. Drosophila mauritiana, however, produces reciprocal hybrids conferring the two classes of sterilities, postmeiotic and premeiotic, when crossed to either D. simulans or D. sechellia. In support of the premise that spermiogenic blockages precede blockages in earlier spermatogenic stages, Coyne (1985) observed that developmental disharmonies are more severe in sim/mel [F.sub.1] hybrids than sim/mau [F.sub.1] hybrids. Also, Lachaise et al. (1986) showed that crosses of D. melanogaster with any of the three sibling species of the simulans clade produced agametic testes (i.e., premeiotic steriles). This is in direct contrast to the prevalence of postmeiotic sterility observed in hybrids of the simulans clade.
Another perspective may be more informative. If we take into account the large effect of the X-chromosome on hybrid male sterility (Coyne and Orr 1989) as well as the theoretical and empirical evidence on the interaction of the X-chromosome with either the Y-chromosome or the autosomes (Coyne and Orr 1989; Zouros 1989) the relative effects of sterility incurred by the X-chromosome of different species may be evaluated. Under this scenario, the X-chromosome stemming from D. mauritiana seems to hold the largest effect on hybrid male sterility when hybridized to other species of this clade [ILLUSTRATION FOR FIGURE 1 OMITTED] because premeiotic lesions are produced in the hybrids. Drosophila sechellia and D. simulans have equal effects to each other, but have much smaller effects than D. mauritiana. These results support either the higher rate of divergence of spermatogenic loci on the X-chromosome of D. mauritiana or longer divergence time. Coyne and Kreitman (1986) have similarly shown that there has been less divergence between D. sechellia and D. simulans than between D. mauritiana and D. simulans in reproductive characters by comparing both morphological traits (sex-comb teeth number, testes color, and genital arch shape) and a hybrid incompatibility (the severity of sterility in backcross [F.sub.2] males).
However, the simple genetic basis of both premeiotic and postmeiotic defects may also explain the presence of these defects in [F.sub.1] hybrids of this clade. Such a supposition assumes that the progression of hybrid male sterility does not necessarily originate in spermiogenesis, but may commence at any stage of spermatogenesis. The polymorphism in D. simulans (as discussed above) supports this hypothesis.
We are grateful to L.-W. Zeng and A. Civetta for valuable discussions. Special thanks are due to D. Morton for his exceptional support and interest in this project. We also thank E. Zouros and an anonymous reviewer for insightful comments on the original manuscript. This work was supported by a research grant to RSS from the Natural Sciences and Engineering Research Council of Canada.
ALPHEY, A., J. JIMENEZ, H. WHITE-COOPER, I. DAWSON, P. NURSE, AND D. M. GLOVER. 1992. twine, a cdc25 homologue that functions in the male and female germline of Drosophila. Cell 69: 977-988.
ASHBURNER, M. 1989. Drosophila: a laboratory manual. Cold Spring Harbor Laboratory Press, New York.
BOCK, I. R. 1984. Interspecific hybridization in the genus Drosophila. Evol. Biol. 18:41-70.
CASTRILLON, D. H., R. GONCZY, S. ALEXANDER, R. RAWSON, C. G. EBERHART, S. VISWANATHAN, S. DINARDO, AND S. A. WASSERMAN. 1993. Toward a molecular genetic analysis of spermatogenesis in Drosophila melanogaster: characterization of male-sterile mutants generated by single P element mutagenesis. Genetics 135:489-505.
CIVETTA, A., AND R. S. SINGH. 1998. Sex and speciation: genetic architecture and evolutionary potential of sexual versus nonsexual traits in the sibling species of the Drosophila melanogaster complex. Evolution 52:1080-1092.
COURTOT, C., C. FRANKHAUSER, V. SIMANIS, AND C. LEHNER. 1992. The Drosophila cdc25 homolog twine is required for meiosis. Development 116:395-402.
COYNE, J. A. 1984. Genetic basis of male sterility in hybrids between two closely related species of Drosophila. Proc. Nat. Acad. Sci. USA 81:4444-4447.
-----. 1985. The genetic basis of Haldane's rule. Nature 314: 736-738.
COYNE, J. A., AND B. CHARLESWORTH. 1989. Genetic analysis of X-linked sterility in hybrids between three sibling species of Drosophila. Heredity 62:97-106.
COYNE, J. A., AND M. KREITMAN. 1986. Evolutionary genetics of two sibling species, Drosophila simulans and Drosophila sechellia. Evolution 40:673-691.
COYNE, J. A., AND H. A. ORR. 1989. Two rules of speciation. Pp. 180-207 in D. Otte and J. A. Endler, eds. Speciation and its consequences. Sinauer, Sunderland, MA.
DARWIN, C. R. 1859. The origin of species by means of natural selection. John Murray, London.
DAVID, J. R., F. LEMEUNIER, K. TSACAS, AND C. BOCQUET. 1974. Hybridation d'une nouvelle espece Drosophila mauritiana avec D. melanogaster et D. simulans. Ann. Genet. 17:235-241.
DAVIS, A. W., E. G. NOONBURG, AND C.-I. WU. 1994. Evidence for complex genic interactions between conspecific chromosomes underlying hybrid female sterility in the Drosophila simulans clade. Genetics 137:191-199.
DAVIS, A. W., J. ROOTE, T. MORLEY, K. SAWAMURA, S. HERRMANN, AND M. ASHBURNER. 1996. Rescue of hybrid sterility in crosses between D. melanogaster and D. simulans. Nature 380:157-159.
DOBZHANSKY, TH. 1936. Studies on hybrid sterility, II. Localization of sterility factors in Drosophila pseudoobscura hybrids. Genetics 21:113-135.
-----. 1937a. Further data on Drosophila miranda and its hybrids with Drosophila pseudoobscura. J. Genet. 34:135-151.
-----. 1937b. Genetics and the origin of species. Columbia Univ. Press, New York.
-----. 1970. Genetics and the evolutionary process Columbia Univ. Press, New York.
-----. 1974. Genetic analysis of hybrid sterility within the species Drosophila pseudoobscura. Hereditas 77:81-88.
DOBZHANSKY, TH., AND G. W. BEADLE. 1936. Studies on hybrid sterility. IV. Transplanted testes in Drosophila pseudoobscura. Genetics 21:832-840.
DOBZHANSKY, TH., AND A. H. STURTEVANT. 1935. Further data on maternal effects in Drosophila pseudoobscura hybrids. Genetics 21:566-570.
FULLER, M. T. 1993. Spermatogenesis. Pp. 71-148 in M. Bate and A. Martinez-Arias, eds. The Development of Drosophila melanogaster. Cold Spring Harbor Laboratory Press, New York.
GOULD-SOMERO, M., AND L. HOLLAND. 1974. The timing of RNA synthesis for spermiogenesis in organ cultures of Drosophila melanogaster testes Wilhelm Roux's Arch. Dev. Biol. 174:133-148.
GOULIELMOS, G., AND E. ZOUROS. 1995. Incompatibility analysis of male hybrid sterility in two Drosophila species: lack of evidence for maternal, cytoplasmic, or transposable element effects. Am. Nat. 145:1006-1014.
HACKSTEIN, J. H. P. 1991. Spermatogenesis in Drosophila. A genetic approach to cellular and subcellular differentiation. European J. Cell Biol. 56:151-169.
HALDANE, J. B. S. 1922. Sex ratio and unisexual sterility in hybrid animals. J. Genet. 12:101-109.
-----. 1932. The causes of evolution. Longmans, Green and Company, London.
HUTTER, P., AND M. ASHBURNER. 1987. Genetic rescue of inviable hybrids between Drosophila melanogaster and its sibling species. Nature 327:331-333.
HUTTER, P., J. ROOTE, AND M. ASHBURNER. 1990. A genetic basis for the inviability of hybrids between sibling species of Drosophila. Genetics 124:909-920.
JOLY, D. 1987. Between species divergence of cyst length distributions in the Drosophila melanogaster species complex. Jpn. J. Genet. 62:257-263.
KHADEM, M., AND C. B. KRIMBAS. 1991. Studies of the species barrier between Drosophila subobscura and D. madeirensis. I. The genetics of male hybrid sterility. Heredity 67:157-165.
LACHAISE, D., J. R. DAVID, F. LEMEUNIER, L. TSACAS, AND M. ASHBURNER. 1986. The reproductive relationships of Drosophila sechellia with D. mauritiana, D. simulans, and D. melanogaster from the Afrotropical region. Evolution 40:262-271.
LACHAISE, D., M.-L. CARIOU, J. R. DAVID, F. LEMEUNIER, L. TSACAS, AND M. ASHBURNER. 1988. Historical biogeography of the Drosophila melanogaster species subgroup. Evol. Biol. 22:159-225.
LIFSCHYTZ, E. 1987. The developmental program of spermiogenesis in Drosophila: a genetic analysis. Int. Rev. Cytol. 109:211-257.
LINDSLEY, D. L., AND E. LIFSCHYTZ. 1972. The genetic control of spermatogenesis in Drosophila. Pp. 203-222 in R. A. Beatty and S. Gluecksohn-Waelsch. Proceedings of the international symposium on the genetics of the spermatozoon. Bogtrykkeriet Forum, Copenhagen, Denmark.
LINDSLEY, D. L., AND K. T. TOKUYASU. 1980. Spermatogenesis. Pp. 225-294 in M. Ashburner and T. R. F. Wright, eds. The genetics and biology of Drosophila. Vol 2D. Academic Press, London.
LINDSLEY, D. L., AND G. G. ZIMM. 1992. The genome of Drosophila melanogaster. Academic Press, San Diego, CA.
MULLER, H. J. 1942. Isolating mechanisms, evolution, and temperature. Biol Symp. 6:71-125.
OLIVIERI, G., AND A. OLIVIERI. 1965. Autoradiographic study of nucleic acid synthesis during spermatogenesis in Drosophila melanogaster. Mutat. Res. 2:366-380.
ORR, H. A. 1987. Genetics of male and female sterility in hybrids of Drosophila pseudoobscura and D. persimilis. Genetics 116: 555-563.
-----. 1989. Genetics of sterility in hybrids between two subspecies of Drosophila. Evolution 43:180-189.
-----. 1995. The population genetics of speciation: the evolution of hybrid incompatibilities. Genetics 139:1805-1813.
PEREZ, D. E., C.-I. Wu, N. A. JOHNSON, AND M.-L. Wu. 1993. Genetics of reproductive isolation in the Drosophila simulans clade: DNA marker-assisted mapping and characterization of a hybrid-male sterility gene, Odysseus (Ods). Genetics 133:261-275.
SAWAMURA, K., T. TAIRA, AND T. K. WATANABE. 1993a. Hybrid lethal systems in the Drosophila melanogaster species complex. I. The maternal hybrid rescue (mhr) gene of Drosophila simulans. Genetics 133:299-305.
SAWAMURA, K., M.-T. YAMAMOTO, AND T. K. WATANABE. 1993b. Hybrid lethal systems in the Drosophila melanogaster species complex. II. The zygotic hybrid rescue (Zhr) gene of Drosophila melanogaster. Genetics 133:307-313.
STURTEVANT, A. H. 1919. A new species resembling Drosophila melanogaster. Psyche 26:153-155.
-----. 1920. Genetic studies on Drosophila simulans. I. Introduction. Hybrids with Drosophila melanogaster. Genetics 5:488-500.
TOKUYASU, K. T., W. J. PEACOCK, AND R. W. HARDY. 1972. Dynamics of spermiogenesis in Drosophila melanogaster. I. Individualization process. Z. Zelfforsch. 84:141-175.
TSACAS, L., AND G. BACHLI. 1981. Drosophila sechellia, n. sp., huitieme espece du sous-groupe melanogaster des Iles Sechelles. Revue Fr. Entomol. 3:146-150.
TSACAS, L., AND J. DAVID. 1974. Drosophila mauritiana n. sp. du groupe melanogaster de l'ile Maurice. Bull. Soc. Entomol. Fr. 79:42-46.
TURELLI, M., AND H. A. ORR. 1995. The dominance theory of Haldane's rule. Genetics 140:389-402.
VIGNEAULT, G., AND E. ZOUROS. 1986. The genetics of asymmetrical male sterility in Drosophila mojavensis and D. arizonensis hybrids: interactions between the Y-chromosome and autosomes. Evolution 40:1160-1170.
WATANABE, T. K. 1979. A gene that rescues the lethal hybrid between Drosophila melanogaster and Drosophila simulans. Jpn. J. Genet. 54:325-331.
WU, C.-I. 1992. A note on Haldane's rule: hybrid inviability versus hybrid sterility. Evolution 46:1584-1587.
WU, C.-I., AND A. T. BECKENBACH. 1983. Evidence for extensive genetic differentiation between the sex-ratio and the standard arrangement of Drosophila pseudoobscura and D. persimilis and identification of hybrid sterility factors. Genetics 105:71-86.
WU, C.-I., AND A. W. DAVIS. 1993. Evolution of postmating reproductive isolation: the composite nature of Haldane's rule and its genetic bases. Am. Nat. 142:187-212.
ZENG, L.-W., AND R. S. SINGH. 1993. The genetic basis of Haldane's rule and the nature of asymmetric hybrid male sterility between Drosophila simulans, Drosophila mauritiana, and Drosophila sechellia. Genetics 134:251-260.
ZOUROS, E. 1981. An autosome-Y chromosome combination that causes sterility in D. mojavensis x D. arizonensis hybrids. Dros. Inf. Serv. 56:167-168.
-----. 1989. Advances in the genetics of reproductive isolation in Drosophila. Genome 31:211-220.
|Printer friendly Cite/link Email Feedback|
|Author:||Kulathinal, Rob; Singh, Rama S.; Zouros, E.|
|Date:||Aug 1, 1998|
|Previous Article:||Host-parasite coevolution: evidence for rare advantage and time-lagged selection in a natural population.|
|Next Article:||Sex and speciation: genetic architecture and evolutionary potential of sexual versus nonsexual traits in the sibling species of the Drosophila...|