Reinforcement of premating isolating mechanisms near a hybrid zone
Justin W. Merry
Department of Biological Sciences, Ohio University, Athens
email:jm703496@oak.cats.ohiou.edu
If
two populations become geographically isolated from one another and diverge sufficiently,
it is possible that upon secondary contact, hybrid offspring between the two populations
may have reduced fitness. In this situation,
prezygotic isolating barriers can evolve in the parental populations in response to
selection against hybridization through a process called reinforcement. While reinforcement is intuitively appealing,
until recently it was considered unlikely to ever occur in a natural setting. However, recent theoretical work, reviews, and
empirical studies provide evidence in support of the reinforcement hypothesis. With the new body of evidence that has surfaced
in the past decade, it no longer seems presumptuous to conclude that reinforcement is not
only a viable hypothesis, but has occurred, and is occurring, in several natural
populations.
Hybrid zones offer interesting opportunities to study species and speciation because it is at hybrid zones that the line blurs between species and races. The debate over the definition and classification of species has lasted for well over 100 years (e.g. Darwin, 1859; Dobzhansky, 1940; Mayr, 1942; Mayr, 1963; Paterson 1978; Zink and McKitrick, 1995) and no apparent end to the argument is in sight. Most workers would agree, however, that "good" species have reproductive isolating barriers that prevent or minimize gene flow with other species (Howard, 1993). These barriers can be loosely categorized into two classes: prezygotic and postzygotic. Prezygotic barriers include mechanisms such as mate choice, genital incompatibility, spatial and temporal habitat differences, and gamete incompatibility. Postzygotic barriers, on the other hand, include mechanisms such as embryo inviability and hybrid sterility (Mayr, 1963).
Postzygotic barriers are generally related
to the genetic dissimilarity between the two parents.
The more two populations diverge, the more dissimilar they become, and hence the
more likely it is that hybridization will disrupt adaptive gene complexes. Thus, postzygotic reproductive isolating
mechanisms are typically not under direct selection themselves, but instead are
by-products of the divergent evolution of adaptive gene complexes between two populations
(Mayr, 1963; But see Coyne, 1974 for a possible exception). Prezygotic barriers, however, can prevent
individuals of two populations from "wasting" their reproductive energy on
hybrid offspring that are unfit or inviable. As
such, natural selection can strengthen the effectiveness of these mechanisms if they
improve the fitness of a parent by preventing hybridization -- a process called
reinforcement (Howard, 1993).
The Reinforcement Hypothesis
Consider a case in which a population is
divided into two populations by a geographical barrier.
While isolated, the two populations change independently from one another, evolving
adaptive gene complexes that permit individuals to survive under conditions specific to
each population. If this barrier is later
removed and the populations have not diverged sufficiently, they might merge together once
again as a continuous population. On the
other hand, if the populations have diverged to such an extent that hybrid offspring have
reduced fitness or are inviable, they may persist as distinct populations. In this situation, Dobzhansky (1940) suggested
that selection might favor mechanisms that strengthen prezygotic isolating mechanisms that
prevent hybridization and thus the production of inferior or inviable offspring.
Where hybridization jeopardizes the
integrity of two or more adaptive complexes, genetic factors which would decrease the
frequency or prevent the interbreeding would thereby acquire a positive selective value,
even though these factors by themselves might be neutral. [Dobzhansky, 1940, p. 316]
Reinforcement, as defined by Howard (1993), is the process whereby prezygotic isolating barriers evolve "in zones of overlap or hybridization (or both) as a response to selection against hybridization." This process typically results in reproductive character displacement, a pattern of "greater divergence of an isolating trait in areas of sympatry between closely related taxa than in areas of allopatry," (Howard, 1993). It is important to note that this definition allows other processes, such as selection for specialization to different ecological conditions, to result in this pattern of character displacement (Rundle and Schluter, 1998).
Alternative definitions for these terms do exist. Butlin (1987, 1989) argued that reinforcement should be applied only to those situations in which fertile, but inferior hybrid offspring are produced. Reproductive character displacement, he argued, should refer to a conceptually similar process that occurs among two populations between which gene flow is prevented by pre-existing postzygotic barriers that preclude any gene flow. Reinforcement is therefore, in Butler's opinion, a mechanism of speciation, whereas reproductive character displacement involves pre-existing species. For the purposes of this paper, I will use the definitions of Howard, as they do not assume a particular species concept and adhere more readily to the original concepts behind the terms as written by Dobzhansky (1940), Blair (1955), and Brown and Wilson (1956). Nevertheless, a number of current studies use Butlin's definitions, and the reader should be aware of these alternatives when investigating this literature.
While initially widely accepted, Dobzhansky’s hypothesis soon came under criticism due to theoretical objections (e.g. Moore, 1957; Bigelow, 1965; Paterson, 1978; Barton and Hewitt, 1981) and a perceived lack of experimental evidence (Howard, 1993; Littlejohn, 1981; Phelan and Baker, 1987). One of the most influential arguments against reinforcement involves what are perceived as alternative, and frequently more likely, outcomes of secondary contact between two populations. If two populations that differ in size encounter one another, individuals of the more rare population will hybridize more frequently than individuals of the more common population. If the smaller population is significantly smaller, parents will encounter, and thus mate with, individuals of the opposite species more frequently than they do with individuals of their own species. This situation would result in extremely high selection intensities for the evolution of prezygotic isolation barriers. However, if the population in question is sufficiently small or has a low growth rate, then a high degree of "wasteful" reproductive effort on hybrid offspring could cause the extinction of that species before reinforcement could raise isolation barriers (Liou and Price, 1994). Other objections, along with counter-arguments by supporters of reinforcement, are reviewed in detail by Howard (1993).
To investigate the validity of the reinforcement hypothesis in a complex, multilocus genetic model, Liou and Price (1994) conducted an intensive computer simulation analysis. Their model assumed completely sympatric populations, polygynous mating, zero hybrid fitness (complete postzygotic reproductive isolation), no ecological competition between species, and no direct cost to female mate preferences. They allowed for a correlation between male indicator traits and female mate preferences, which were both determined by five-locus genes. They found that reinforcement can occur, and indeed is very likely to occur, if (1) hybrid fitness is sufficiently low and (2) there is sufficient divergence in mate recognition systems between the populations prior to secondary contact. In addition, they found that there are a number of population characters that increase the likelihood of reinforcement occurring, including high ecological carrying capacity, strong female preferences, high potential population growth rates, and high heritability of mate recognition systems.
No debate over a potentially important evolutionary process such as reinforcement should hinge on theoretical arguments alone. Indeed, as Howard (1993) stated after his review of the theoretical arguments over this hypothesis, "it has become clear that the issue cannot be settled by theoretical arguments but by evidence from natural populations." It is on this empirical evidence that I will focus in the next portion of this review.
The Evidence
There is a perception in the literature that very little empirical evidence exists in support of reinforcement. As Howard (1993) pointed out, a number of studies from the 1980's (e.g. Littlejohn, 1981; Phelan and Baker, 1987) that echoed this opinion referred to a single, highly influential survey of acoustic insects by Walker (1974). Walker examined published data on the songs of 135 species of crickets and 119 species of katydids for evidence of reproductive character displacement, but found only 5 relatively weak cases in support of reproductive character displacement. However, in all, he only found 22 studies in which there was "sufficient" data for revealing reproductive character displacement. In addition, only one experiment noted in Walker's survey actually focused on collecting data for reproductive character displacement: Hill et al. (1972), who failed to find evidence of reinforcement between two species of Australian field crickets, Teleogryllus oceanicus and T. commodus (Howard, 1993).
Howard (1993) conducted a detailed survey of the literature on reinforcement and reproductive character displacement, in which he identified 48 studies that had performed experiments in sufficient detail to assess the presence or absence of reproductive character displacement. Of these, 33 cases demonstrated reproductive character displacement, while 15 found no evidence of such a pattern. Furthermore, six of these 33 studies were able to provide convincing evidence for reinforcement by demonstrating (1) reproductive character displacement, (2) heterospecific matings in nature, and (3) selection against hybridization.
Since Howard's survey, a number of studies have been conducted that provide strong evidence for the reinforcement hypothesis. These include systems in which the cost of hybridization ranges from moderate (viable hybrid offspring of reduced fertility or viability) to extremely high (no viable offspring) and involve taxa ranging from insects to birds. Some of these studies are summarized below, all of which demonstrate strong character displacement, selection against hybridization, and reduced hybrid fitness, which indicates selection against hybridization. Several other good examples have been published (e. g. Hauffe and Searle, 1993; Grant, 1994; Noor, 1997; Rundle and Schluter, 1998), however the studies below have been selected for their experimental clarity and the unique properties of the individual systems.
Gerhardt (1994) studied a very interesting hybrid system between two species of the grey treefrog (Hyla) complex. H. chrysoscelis is a diploid species and is the apparent parental species of H. versicolor, which is tetraploid. Attempts to hybridize these two species in the laboratory and create triploid offspring have encountered extremely low viability rates in early larval stages, as well as sterility in hybrid adults (Johnson, 1963). While adult hybrids are rare in nature (Gerhardt, 1982), significant interspecific mating activity has been detected in Missouri populations (Gerhardt et al., 1994).
Gerhardt (1994) began his study by conducting oscillographic analysis of calls among the different populations. This indicated that H. versicolor males call at a consistently lower pulse-per-minute rate than H. chrysoscelis, but that call duration does not consistently vary between the two species. He then used synthetically produced calls to conduct mate choice tests on H. chrysoscelis females from populations that were (1) allopatric with, (2) allopatric, but within 60 km of, and (3) sympatric with, H. versicolor populations. He found that all types of females preferred calls with long durations, which is a possible indication of vigor. In addition, females from all the populations also preferred calls with high pulse rates, which simulates a conspecific signal. In a third experiment, however, females had to choose between a call of long duration but low pulse rate (mimicking a H. versicolor male’s call) and a call of short duration but high pulse rate (mimicking a conspecific, H. chrysoscelis male’s call). Interestingly, females from the two allopatric populations chose the conspecific stimulus 42.9% and 56.3% of the time, respectively. This was in stark contrast to the females from the three populations living in sympatry (or in nearby, allopatric populations), which preferred the conspecific male 87.5-100% of the time. Thus, because of the high costs of mating with a male from another species, females living in sympatry may have evolved stronger preferences for conspecific characters.
Ryan et al. (1996) studied another system involving high costs to hybridization. In this system, however, male mate choice, and not female mate preference, was under reinforcement selection. Amazon mollies, Poecilia formosa, are gynogenetic species, meaning that they are all female and reproduce via clones, however they require insemination by a related species to initiate embryogenesis. This male does not contribute to the genome of any offspring reproduced by such a copulation, and therefore would be wasting both gametes and time by mating with a gynogenetic female. Thus, one would expect that there would be a fitness benefit to males that avoid these females, and therefore consequential character displacement in male mating preference.
To test this hypothesis, Ryan et al. (1996) conducted choice tests on P. latipinna and P. mexicana males from populations either sympatric or allopatric with P. formosa. In each trial, a male was placed in the center portion of a tank that was divided into three sections, while a female was placed in each of the two side compartments. The dividers were lifted and male preferences were estimated based on the number of gonopodial thrusts that were made by the male towards each female over 15 minutes. They found that males from all populations showed significant preferences for females from conspecific populations when tested with conspecific and P. formosa females. However, when comparing the proportion of gonopodial thrusts directed toward P. formosa females by males from the different populations, males from sympatric populations attempted to copulate significantly less often than males from allopatric populations. Roughly 7% and 23% of thrusts were directed towards P. formosa females by sympatric and allopatric males, respectively. The males, therefore, may have evolved stronger preferences for conspecific females in areas in which there is a cost to mating indiscriminately.
Species of the genus Drosophila also provides evidence for reinforcement. Coyne and Orr (1989) conducted a large survey of 119 hybridization interactions between different species of Drosophila, and later expanded this sample in a follow-up paper to 171 hybridizations (Coyne and Orr, 1997). For each of these interactions, they computed prezygotic and postzygotic isolation indices, which range from zero to one and indicate the degree of isolation for each type of mechanism. They found that as genetic distance increased, prezygotic and postzygotic isolation increased at roughly the same rate in populations that typically exist in allopatry. However, among sympatric populations, prezygotic isolation evolves at a much faster rate than postzygotic isolation, indicating that prezygotic isolation may be reinforced by selection.
Noor (1995) conducted a more detailed study on one such hybridization system. D. pseudoobscura ranges across much of the western continental United States. The range of another fruit fly, D. persimilis, lies completely within that of D. pseudoobscura. Small numbers of hybrids from heterospecific matings are captured in the field. However, laboratory crosses show that, while viable, hybrid male offspring are sterile. Female hybrid offspring, on the other hand, are fertile (Dobzhansky, 1973; Powell, 1983), so the cost to hybridization is lower in this system than in some of the examples above. When Noor (1995) paired D. persimilis males with D. pseudoobscura females from either allopatric or sympatric populations, he found that in five out of six comparisons, the female allopatric populations mated significantly more often with D. persimilis males than did females from sympatric populations. No difference was detected in the courtship behaviors of D. persimilis males directed towards either allopatric or sympatric D. pseudoobscura females, which indicates that the difference above is due to stronger female mate choice in sympatric D. pseudoobscura females.
One tenet of the reinforcement hypothesis is that if, at secondary contact, one population is more rare than the other in the hybrid zone, that population will evolve stronger prezygotic isolating mechanisms than it would in a situation in which it was the more common species. Noor (1995) tested this hypothesis in his populations by comparing the strengths of preferences of D. persimilis females from populations in which D. persimilis was more common than D. pseudoobscura with the preferences of females from populations in which D. persimilis was less common than D. pseudoobscura. His results confirmed the hypothesis, as D. persimilis females from a rare population mated significantly less frequently with male D. pseudoobscura than did females from populations in which D. persimilisis was common.
Evidence for reinforcement has also been found in European flycatchers of the genus Ficedula. In an area of sympatry between the pied flycatcher, F. hypoleuca, and the black-and-white collared flycatcher, F. albicollis, S³tre et al. (1997) found that hybridization occurs in the field in low frequency (94.8% of all observed pairs were species assortative). Those individuals that did interbreed with heterospecifics produced viable offspring, however these usually had reduced fecundity – only 25.9% of the eggs in pairs involving at least one hybrid adult successfully hatched, compared to 95.1% in pure and mixed-species pairs (S³tre et al., 1997).
Sympatric F. hypoleuca males are typically lighter in color than their allopatric counterparts, and have no white collar around their necks. F. albicollis males, on the other hand, are typically extremely dark, but possesses a white collar around their necks. While no allopatric population of F. albicollis exists, a closely related sister taxa, the semicollared flycatcher (F. semitorquata), does live in allopatry with F. hypoleuca. Males of this species are typically darkly colored but possess a partial, less well-defined white collar around their neck than do F. albicollis males.
Mate choice tests conducted on females from both sympatric populations showed that females paired exclusively with conspecific males when presented with males featuring sympatric coloration patterns. On the other hand, when presented with males from allopatric populations, a large number of sympatric females chose a heterospecific mate, though conspecifics were still preferred overall. Therefore, the divergent male plumage patterns seen in the sympatric populations may assist females in recognizing conspecifics.
Perhaps more interesting was a second choice test, in which F. hypoleuca females from both allopatric and sympatric populations chose between allopatric and sympatric conspecific males. Females from the sympatric population strongly preferred lighter (sympatric) males. On the other hand, females from the allopatric population showed a complete reversal in preference, strongly preferring the darker, allopatric males. Previous studies on pied flycatchers and other birds have shown that darker, more intense coloration indicates strong potential direct benefits to a female in the form of better male parental care (Saetre et al., 1995; Hill, 1991). Therefore, reinforcement selection on female choice may have overridden a preference for vigorous males in favor of more accurately choosing conspecifics due to the high cost of hybridization between species.
The hypothesis that prezygotic isolating mechanisms can be selectively strengthened, or reinforced, along the edges of a hybrid zone can be traced to Dobzhansky’s (1940) writings on speciation. Though initially accepted, the concept came under scrutiny and was generally doubted by many biologists. However, within the past decade, theoretical work (Liou and Price 1994), taxonomic surveys (Coyne and Orr, 1989; Coyne and Orr, 1997; Howard, 1993), and numerous empirical studies (e.g. Gerhardt ,1994; Noor, 1995; Ryan et al., 1996; S³tre et al., 1997; Rundle and Schluter, 1998) have provided strong evidence supporting the reinforcement hypothesis.
With this recent evidence, it no longer seems presumptuous to conclude that reinforcement does occur in nature, and thus can be an important mechanism in the generation of isolation barriers between diverged populations. Therefore, future studies should focus not on simply demonstrating reinforcement in populations (stamp collecting), but on investigating how instances of reinforcement can change or alter our understanding of the evolutionary history of different groups of taxa. Some types of organisms in which reinforcement may have been particularly important include those animals that live in small populations that are frequently isolated and reintroduced by fluctuating geographic barriers, such as fish living in small streams, terrestrial animals living in mountain valleys, or island-dwelling species. If the isolation period is long enough for significant genetic divergence to occur, prezygotic isolation may evolve upon secondary contact between two populations. Reinforcement could help explain some of the impressive diversity in reproductive characters (Rauchenberger, 1990; Nunn and Cracraft, 1996; Galis and Metz, 1998) found among some of these populations.
References
Barton, N. H., and Hewitt, G. M.
(1981) Hybrid zones and speciation, in Evolution and speciation: Essays in honor of M. J. D.
White (Atchley, W. R. and Woodruff, D. S., eds.), pp. 109-145, Cambridge: Cambridge
University Press.
Bigelow, R. S. (1965) Hybrid zones and reproductive isolation, Evolution, 19, 449-458.
Blair, F. W. (1955) Mating call and stage of speciation in the Microhyla olivacea - M. Carolinensis complex, Evolution, 9, 469-480.
Brown, W. L. Jr., and Wilson, E.
O. (1956) Character Displacement,. Syst. Zool., 5, 49-64.
Butlin, R. (1989) Reinforcement of premating isolation, In Speciation and its consequences, (Otte, D. and
Endler, J. A., eds.), pp. 158-179, Sunderland, MA: Sinauer.
Butlin, R. (1987) Speciation by reinforcement, Tr. Ecol. Evol., 2, 8-13.
Coyne, J. A. (1974) The evolutionary origin of hybrid inviability, Evolution, 28, 505-506.
Coyne, J. A., and Orr, H. A.
(1989) Patterns of speciation in Drosophila, Evolution,
43(2), 362-381.
Coyne, J. A., and Orr, H. A.
(1997) "Patterns of speciation in Drosophila" revisited, Evolution, 51(1), 295-303.
Darwin, C. (1859) On the origin of species by means of natural selection,
London: John Murray.
Dobzhansky, T. (1973) Am. Nat., 107, 312-314.
Dobzhansky, T. (1940) Speciation as a stage in evolutionary divergence, Am. Nat., 74, 312-321.
Galis, F., and Metz, J. A. J.
(1998) Why are there so many cichlid species? Tr. Ecol. Evol., 13(1), 1-2.
Gerhardt, H. C. (1982) Sound pattern recognition in some North American
treefrogs (Anura: Hylidae): implications for mate choice, Am. Zool.,
22, 581-595.
Gerhardt, H. C. (1994) Reproductive character displacement of female mate
choice in the grey treefrog, Hyla chrysoscelis,
Anim. Behav., 47, 959-969.
Gerhardt, H. C., Ptacek, M. B.,
Barnett, L., & Torke, K. G. (1994) Hybridization
in the diploid-tetraploid treefrogs: Hyla
chrysoscelis and Hyla versicolor, Copeia, 1994(1), 51-59.
Grant, V. (1994) Modes and origins of mechanical and ethological
isolation in angiosperms, Proc. Nat. Acad. of
Sci.U.S.A., 91, 3-10.
Hauffe, H. C., and Searle, J. B.
(1993) Extreme karyotypic variation in a Mus musculus domesticus hybrid zone: the tobacco
mouse story revisited, Evolution, 47(5),
1374-1395.
Hill, G. E. (1991) Plumage coloration is a sexually selected indicator of
male quality, Nature, 350, 337-339.
Hill, K. G., Loftus-Hills, J. J.,
and Gartside, D. F. (1972) Pre-mating isolation
between the australian field crickets Teleogryllus
commodus and T. oceanicus (Orthoptera:
Gryllidae), Aust. J. Zool., 20, 153-163.
Howard, D. J. (1993) Reinforcement: origin, dynamics, and fate of an
evolutionary hypothesis, in Hybrid zones and the
evolutionary process (Harrison, R. G., ed.), pp.46-69, New York: Oxford University
Press.
Liou, L. W., and Price, T. D.
(1994) Speciation by reinforcement of premating
isolation, Evolution, 48(5), 1451-1459.
Littlejohn, M. J. (1981) Reproductive isolation: a critical review, in Evolution and speciation: essays in honor of M. J. D.
White (Atchley, W. R. and Woodruff, D. S., eds), pp. 298-334, Cambridge: Cambridge
University Press.
Mayr, E. (1963) Animal species and evolution, Cambridge, MA:
Belknap Press.
Mayr, E. (1942) Systematics and the origin of species, New
York: Columbia University Press.
Moore, J. A. (1957) An embryologist's view of the species concept, in The species problem (Mayr, E., ed.), pp. 325-338,
Washington, DC: American Association for the Advancement of Science.
Noor, M. A. (1995) Speciation driven by natural selection in Drosophila, Nature, 375, 674-675.
Noor, M. A. F. (1997) How often does sympatry affect sexual isolation in Drosophila, Am. Nat., 149(6), 1156-1163.
Nunn , and Cracraft. (1996) Phylogenetic relationships among the major lineages of
the birds-of-paradise (Paradisaeidae) using mitochondrial DNA gene sequences, Molec. Phylogenetics Evol., 5(3), 445-459.
Paterson, H. E. H. (1978) More evidence against speciation by reinforcement,
S. African J. Sci., 74, 369-371.
Phelan, P. L., and Baker, T. C.
(1987) Evolution of male pheromones in moths:
reproductive isolation through sexual isolation? Science,
235, 205-207.
Powell, J. R. (1983) Proc. Nat. Acad. Sci U.S.A., 80, 492-495.
Rauchenberger, M., Kallman, K.
D., and Morizot, D. C. (1990) Monophyly and
geography of the Río Pánuco basin swordtails (genus Xiphophorus) with descriptions of four new
species, Am. Mus. Novitates, 2975, 1-41.
Rundle, H. D., and Schluter, D.
(1998) Reinforcement of stickleback mate
preferences: sympatry breeds contempt, Evolution,
52(1), 200-208.
Ryan, M. J., Dries, L. A., Batra,
P., and Hillis, D. M. (1996) Male mate preferences
in a gynogenetic species complex of Amazon mollies, Anim. Behav., 52, 1225-1263.
Saetre, G. P., Fossnes, T., and
Slagsvold, T. (1995) Food provisioning in the pied
flycatcher: do females gain direct benefits from choosing bright-coloured males? J. Anim. Ecol., 64, 21-30.
Saetre, G.-P., Moum, T., Bures,
S., Kral, M., Adamjan, M., and Moreno, J. (1997) A
sexually selected character displacement in flycatchers reinforces premating isolation,
Nature, 387, 589-592.
Walker, T. J. (1974) Character displacement and acoustic insects, Am. Zool., 14, 1137-1150.
Zink, R. M., and McKitrick, M. C. (1995) The debate over species concepts and its implications for ornithology, Auk, 112 (3), 701-719.
Copyright © 1999-2000 Justin W. Merry All Rights Reserved