Weak coupling among barrier loci and waves of neutral and adaptive introgression across an expanding hybrid zone

Demography-driven and adaptive introgression in a hybrid zone of the Armeria syngameon

Syngameons represent networks of otherwise distinct species connected by limited gene exchange. Although most studies have focused on how species maintain their cohesiveness despite gene flow, there are additional relevant questions regarding the evolutionary dynamics of syngameons and their drivers, as well as the success of their members and the network as a whole. Using a ddRADseq approach, we analysed the genetic structure, genomic clines and demographic history of a coastal hybrid zone involving two species of the Armeria (Plumbaginaceae) syngameon in southern Spain. We inferred that a peripheral population of the sand dune-adapted A. pungens diverged from the rest of the conspecific populations and subsequently hybridized with a locally more abundant pinewood congener, A. macrophylla. Both species display extensive plastid DNA haplotype sharing. Genomic cline analysis identified bidirectional introgression, but more outlier loci with excess A. pungens than A. macrophylla ancestry, suggesting the possibility of selection for A. pungens alleles. This is consistent with the finding that the A. pungens phenotype is selected for in open habitats, and with the strong correlation found between ancestry and phenotype. Taken together, our analyses suggest an intriguing scenario in which bidirectional introgression may, on the one hand, help to avoid reduced levels of genetic diversity due to the small size and isolated location of the A. pungens range-edge population, thereby minimizing demographic risks of stochastic extinction. On the other hand, the data also suggest that introgression into A. macrophylla may allow individuals to grow in open, highly irradiated, deep sandy, salt-exposed habitats.

Does selection favour the maintenance of porous species boundaries?

The endpoint of speciation has been viewed as complete isolation and the absence of gene flow between species. If the influx of genes from another species is maladaptive because species have different adaptations and genetic backgrounds, selection should favour the closing of species boundaries and zero gene flow, a process known as reinforcement. Recently, numerous cases of gene flow between species have been identified, many of which involved adaptive introgression of beneficial alleles. These cases could reflect transient states on the way to closed species boundaries or the result of declining strength or efficacy of selection for reinforcement as the level of gene flow approaches zero. An alternative hypothesis, however, is that selection favours porous species boundaries that allow beneficial alleles to cross, especially in changing environments. This perspective evaluates the conditions that would be needed for selection to favour porous species boundaries and the evidence for them. A contrast is made between hybridization in sexual eukaryotes and gene transfer via homologous recombination in bacteria. Current evidence is inconclusive on whether non-zero gene flow is favoured by selection. Studies are needed that quantify selection gradients on rates of gene flow and test for evolution towards intermediate values, especially experiments that manipulate conditions and track evolution for multiple generations. Estimation of gene flow networks for more clades and regional assemblages using emerging genome data will also allow the evolutionary determinants of interspecific gene flow to be better understood.

The role of introgressive hybridization in shaping the geographically isolated gene pools of wax palm populations (genus Ceroxylon)

The speciation continuum is the process by which genetic groups diverge until they reach reproductive isolation. It has become common in the literature to show that this process is gradual and flickering, with possibly many instances of secondary contact and introgression after divergence has started. The level of divergence might vary among genomic regions due to, among others, the different forces and roles of selection played by the shared regions. Through hybrid capture, we sequenced ca. 4,000 nuclear regions in populations of six species of wax palms, five of which form a monophyletic group (genus Ceroxylon, Arecaceae: Ceroxyloideae). We show that in this group, the different populations show varying degrees of introgressive hybridization, and two of them are backcrosses of the other three 'pure' species. This is particularly interesting because these three species are dioecious, have a shared main pollinator, and have slightly overlapping reproductive seasons but highly divergent morphologies. Our work supports shows wax palms diverge under positive and background selection in allopatry, and hybridize due to secondary contact and inefficient reproductive barriers, which sustain genetic diversity. Introgressed regions are generally not under positive selection. Peripheral populations are backcrosses of other species; thus, introgressive hybridization is likely modulated by demographic effects rather than selective pressures. In general, these species might function as an 'evolutionary syngameon' where expanding, peripheral, small, and isolated populations maintain diversity by crossing with available individuals of other wax palms. In the Andean context, species can benefit from gained variation from a second taxon or the enhancement of population sizes by recreating a common genetic pool.

Quantitative Analyses of Coupling in Hybrid Zones

In hybrid zones, whether barrier loci experience selection mostly independently or as a unit depends on the ratio of selection to recombination as captured by the coupling coefficient. Theory predicts a sharper transition between an uncoupled and coupled system when more loci affect hybrid fitness. However, the extent of coupling in hybrid zones has rarely been quantified. Here, we use simulations to characterize the relationship between the coupling coefficient and variance in clines across genetic loci. We then reanalyze 25 hybrid zone data sets and find that cline variances and estimated coupling coefficients form a smooth continuum from high variance and weak coupling to low variance and strong coupling. Our results are consistent with low rates of hybridization and a strong genome-wide barrier to gene flow when the coupling coefficient is much greater than 1, but also suggest that this boundary might be approached gradually and at a near constant rate over time.

The puzzle of plant hybridisation: a high propensity to hybridise but few hybrid zones reported

An interesting conundrum was recently revealed by R. Abbott when he found that the number of hybrid zones reported in the literature for plants is very low, given the propensity of plants to hybridise. In another literature survey on hybrid zones performed over the period 1970-2022, we found that the number of hybrid zones reported for vertebrates was 2.3 times greater than that reported for vascular plants, even though there are about six times more vascular plant species than vertebrates. Looking at the number of papers reporting hybrid zones, there are 4.9 times more on vertebrates than on vascular plants. These figures support the relevance of this conundrum. In this paper we aim to shed light on this question by providing a structured discussion of the causes that may underlie this conundrum. We propose six non-mutually exclusive factors, namely lack or deficit of spatial structure, lack or deficit of genetic structure, effects of hybridisation between non-closely related species, lability of plant hybrid zones over time, botanists' perception of hybridisation, and deficit of population genetic data. There does not appear to be a single factor that explains our puzzle, which applies to all cases of plants where hybridisation is detected but no hybrid zone is reported. It is argued that some plant features suggest that the puzzle is not, at least entirely, due to insufficient knowledge of the specific cases, a hypothesis that should be addressed with a wider range of empirical data across different taxonomic groups.

A complex genomic architecture underlies reproductive isolation in a North American oriole hybrid zone

Natural hybrid zones provide powerful opportunities for identifying the mechanisms that facilitate and inhibit speciation. Documenting the extent of genomic admixture allows us to discern the architecture of reproductive isolation through the identification of isolating barriers. This approach is particularly powerful for characterizing the accumulation of isolating barriers in systems exhibiting varying levels of genomic divergence. Here, we use a hybrid zone between two species-the Baltimore (Icterus galbula) and Bullock's (I. bullockii) orioles-to investigate this architecture of reproductive isolation. We combine whole genome re-sequencing with data from an additional 313 individuals amplityped at ancestry-informative markers to characterize fine-scale patterns of admixture, and to quantify links between genes and the plumage traits. On a genome-wide scale, we document several putative barriers to reproduction, including elevated peaks of divergence above a generally high genomic baseline, a large putative inversion on the Z chromosome, and complex interactions between melanogenesis-pathway candidate genes. Concordant and coincident clines for these different genomic regions further suggest the coupling of pre- and post-mating barriers. Our findings of complex and coupled interactions between pre- and post-mating barriers suggest a relatively rapid accumulation of barriers between these species, and they demonstrate the complexities of the speciation process.

Range-wide whole-genome resequencing of the brown bear reveals drivers of intraspecies divergence

Population-genomic studies can shed new light on the effect of past demographic processes on contemporary population structure. We reassessed phylogeographical patterns of a classic model species of postglacial recolonisation, the brown bear (Ursus arctos), using a range-wide resequencing dataset of 128 nuclear genomes. In sharp contrast to the erratic geographical distribution of mtDNA and Y-chromosomal haplotypes, autosomal and X-chromosomal multi-locus datasets indicate that brown bear population structure is largely explained by recent population connectivity. Multispecies coalescent based analyses reveal cases where mtDNA haplotype sharing between distant populations, such as between Iberian and southern Scandinavian bears, likely results from incomplete lineage sorting, not from ancestral population structure (i.e., postglacial recolonisation). However, we also argue, using forward-in-time simulations, that gene flow and recombination can rapidly erase genomic evidence of former population structure (such as an ancestral population in Beringia), while this signal is retained by Y-chromosomal and mtDNA, albeit likely distorted. We further suggest that if gene flow is male-mediated, the information loss proceeds faster in autosomes than in X chromosomes. Our findings emphasise that contemporary autosomal genetic structure may reflect recent population dynamics rather than postglacial recolonisation routes, which could contribute to mtDNA and Y-chromosomal discordances.

Effects of the neo-X chromosome on genomic signatures of hybridization in Rumex hastatulus

Natural hybrid zones provide opportunities for studies of the evolution of reproductive isolation in wild populations. Although recent investigations have found that the formation of neo-sex chromosomes is associated with reproductive isolation, the mechanisms remain unclear in most cases. Here, we assess the contemporary structure of gene flow in the contact zone between largely allopatric cytotypes of the dioecious plant Rumex hastatulus, a species with evidence of sex chromosome turn-over. Males to the west of the Mississippi river, USA, have an X and a single Y chromosome, whereas populations to the east of the river have undergone a chromosomal rearrangement giving rise to a larger X and two Y chromosomes. Using reduced-representation sequencing, we provide evidence that hybrids form readily and survive multiple backcross generations in the field, demonstrating the potential for ongoing gene flow between the cytotypes. Cline analysis of each chromosome separately captured no signals of difference in cline shape between chromosomes. However, principal component regression revealed a significant increase in the contribution of individual SNPs to inter-cytotype differentiation on the neo-X chromosome, but no correlation with recombination rate. Cline analysis revealed that the only SNPs with significantly steeper clines than the genome average were located on the neo-X. Our data are consistent with a role for neo-sex chromosomes in reproductive isolation between R. hastatulus cytotypes. Our investigation highlights the importance of studying plant hybrid zones for understanding the evolution of sex chromosomes.