Sperm-dependent parthenogens delay the spatial expansion of their sexual hosts

Abstract

It has been known for long time that asexual organisms may affect the distribution of sexual taxa. In fact, such phenomenon is inherent in the concept of geographical parthenogenesis. On the other hand, it was generally hypothesized that sperm-dependent asexuals may not exercise the same effect on related sexual population, due to their dependence upon them as sperm-donors. Recently, however, it became clear that sperm-dependent asexuals may directly or indirectly affect the distribution of their sperm-hosts, but rather in a small scale. No study addressed the large-scale biogeographic effect of the coexistence of such asexuals with the sexual species. In our study we were interested in the effect of sexual–asexual coexistence on the speed of spatial expansion of the whole complex. We expand previously published Lotka–Volterra model of the coexistence of sexual and gynogenetic forms of spined loach (Cobitis; Teleostei) hybrid complex by diffusion. We show that presence of sperm-dependent parthenogens is likely to negatively affect the spatial expansion of sexuals, and hence the whole complex, compared to pure sexual population. Given that most of the known sperm-dependent asexual complexes are distributed in areas prone to climate-induced colonization/extinction events, we conclude that such mechanism may be an important agent in determining the biogeography of sexual taxa and therefore requires further attention including empirical tests.

Introduction

More than a century after the recognition of the ‘paradox of sex’ (Weismann, 1889), there is little disagreement that strict asexuality among higher organisms generally represents an evolutionary dead-end due to the inability of clones to escape either mutation or parasite loads, or to cope with a changing environment (e.g. Dybdahl and Lively, 1998, Paland and Lynch, 2006, West et al., 1999). It should, however, be noted that some findings of exceptional longevity of clones, their adaptive potential or even the reevolution of sex contrast traditional assumptions in some model groups (Vrijenhoek, 1993, Judson and Normark, 1996, Domes et al., 2007, Normark et al., 2003, Martens et al., 2003).

Whether or not asexuals are looked at as short-lived offshoots of sex, there is one aspect of their existence that is not fully appreciated in the discussion as to whether asexual organisms should be considered no-hopers or hopeful monsters (Vrijenhoek, 1989); their potential to affect the distributions of sexual species. Although cases where asexuality has completely replaced sex (i.e. where there is no sexual species closely related to clonal species) are rare among Metazoans, asexuals do interfere with their sexual competitors even without driving sexual species to extinction. Geographical parthenogenesis may at least in some cases be viewed as the occupation by asexuals of habitats where sex would have normally existed (e.g. Bierzychudek, 1985, Peck et al., 1998, Kearney, 2003). Although the evolutionary potential of asexual organisms is traditionally studied in terms of their abilities to persist some number of generation without sex or the creation of some evolutionary novelties, we should perhaps focus on questions of what their presence means to competing sexual species. Even if a given clone eventually goes extinct, the changes that its presence induced in the sexual species—such as affected distribution ranges, ecological or behavioral adaptations, or mate recognition mechanisms (e.g. Lynch, 1984, Peck et al., 1998, Ryan et al., 1996, Gabor and Ryan, 2001, Schlupp, 2005)—may represent an evolutionary impact not generally considered.

Sperm-dependent (pseudogamous) asexuals are parthenogenetic organisms that require sperm to activate the oocyte development, and are thus dependent on a suitable sperm donor in what could be considered a kind of parasitism (see e.g. Schlupp, 2005, Choleva et al., 2008, and citations therein). It has been speculated for decades (see. e.g. Moore and McKay, 1971) that all else being equal, the sperm-dependent asexuals should out-compete their hosts and the whole complex collapse. Therefore, various mechanisms have been proposed to ensure the coexistence of both populations. For example, sexual males are likely to evolve mate-recognition since the sperm used for the activation of clonal ova is effectively wasted. This may be manifested either by the premating choice of partner (Ryan et al., 1996, Gabor and Ryan, 2001) or by differential allocation of sperm (Schlupp and Plath, 2005). In asexuals, selection among clones operates to overcome the mating preference for conspecifics in the sexual species by more aggressive mating behavior or by the evolution of sex-mimicry (Beukeboom and Vrijenhoek, 1998). Competition for resources with the parental species may select the asexual population to either minimize the niche overlap by specializing to alternative habitats (Hellriegel and Reyer, 2000, Plenet et al., 2005) or to form narrowly specialized clonal lineages able to out-compete the sexual host in given sub-niches and therefore to coexist in the same habitat (Vrijenhoek, 1979, Pound et al., 2002). Parasites may also preferentially affect asexuals, because clonal reproduction does not generate enough variability to escape them (e.g. Dybdahl and Lively, 1998, Hakoyama and Iwasa, 2004).

Aspects of spatial segregation of sexual and sperm-dependent asexual components have rarely been taken into account, but if so, the studies suggested importance of such aspects for the long-term persistence of such mixed complexes. These works either assumed the two-deme model (Hellriegel and Reyer, 2000), or searched for dispersal-extinction equilibria in a metapopulation framework (e.g. Kokko et al., 2008). Interestingly, Kokko et al. (2008) showed that metapopulation dynamics alone may promote stable coexistence. This is expected either if the sperm-parasite causes little damage to its host (thus resulting in majority of demes being occupied by both species), or if the damage is rather high, locally driving the host population quickly to extinction (thus resulting in a relatively low invasion rate of the parasite).

Both abovementioned studies assumed population dynamics in a constantly inhabited environment, because the sperm-dependent parthenogens are usually expected to have only a small effect on the large-scale biogeography of the sexual taxa on which they critically depend. Sperm dependent asexuals are generally assumed to arise in areas where their sexual progenitors are present and subsequently invade their ranges without driving sexual hosts to extinction. However, several sperm-dependent asexual ‘species’ are known to have either abandoned their parental host species or out-competed it with a subsequent switch to another sperm-host (rev. in Choleva et al., 2008), resulting in a type of geographical parthenogenesis even in the case of the sperm-dependent asexuality. While such effects of sperm-dependent asexuals are rare, they may possibly influence the large-scale biogeography of sexual species yet by a different, possibly quite frequent, process; they may attenuate the rate of spatial spread of their sexual hosts.

We noticed such a possibility when reconstructing the phylogeography of the Cobitis taenia hybrid complex. Here, six parapatric species, i.e. C. elongatoides, C. taenia, C. tanaitica, C. taurica, C. strumicae and C. melanoleuca, are known to hybridize producing several virtually all-female hybrid forms that coexist with the parental species and reproduce by gynogenesis (a type of sperm-dependent asexuality, where the sperm triggers the oocyte development, but does not contribute to genetic build-up of progeny; reviewed in Janko et al., 2007a). Such hybridizations likely take place during interglacial periods, when sexual species intermittently come into reproductive contact and the resulting gynogenetic hybrid all-female lineages subsequently invade the areas of parental species (Janko et al., 2005). Two of the parental species, C. taenia and C. elongatoides, are widespread in Europe, but whereas the former species inhabits most of the eastern, western and northern Europe, the latter is restricted to the Danubian watershed and to upper stretches of the Odra and Elbe rives. Culling et al. (2006) showed that the current distribution of C. taenia including western Europe, is a result of recent, probably post-glacial, colonization from the east.

It is not clear why in the case of Cobitis the Danube did not play an important role in the colonization of Europe, in contrast to other cases of freshwater fish (e.g. Durand et al., 1999, Nesbø et al., 1999, Kotlík and Berrebi, 2001, Bohlen et al., 2006). This is especially striking, as C. elongatoides was apparently able to expand into Central Europe, but not any further. This problem has many more or less plausible explanations, but our previous analysis (Janko et al., 2005) suggests that C. taenia expanded as a ‘pure’ sexual population, which was ‘infected’ by gynogenetic asexuals only after the postglacial reproductive contact with C. elongatoides expanding from the Danubian refuge. On the other hand, C. elongatoides expansion was about twice as slow as C. taenia and was probably coupled with the co-expansion of an ancient elongatoides-tanaitica gynogenetic hybrid lineage that survived glacial events in the Danubian refuge. Since asexuals compete with host species for resources and sperm, thereby decreasing population densities, we speculated that they may also reduce the dispersal potential of the sexual population. We proposed it as the reason why the Cobitis colonization of Western Europe came from the east rather than from the Danube.

In this paper we study this hypothesis by expanding published Lotka–Volterra model describing the local dynamics of Cobitis hybrid complex by adding a linear diffusion. Since this model assumed no differentiation between sexual and asexual forms, we also expanded model of Schley et al. (2004) describing the coexistence of sexual and sperm-dependent asexual hermaphroditic populations that may differ either in competitive ability or intrinsic capacity to exploit resources. We stress on the choice of parameters under which both population coexist (and not to perform a full bifurcation analysis of our models which could be rather complex) and show that presence of sperm-dependent parthenogens is likely to negatively affect the spatial expansion of the whole complex under some circumstances. Given that ranges of many of the known sperm-dependent asexual complexes likely underwent repeated extinction/expansion cycles, we suggest that such a mechanism may be an important agent in determining the biogeography of sexual taxa.