Range-wide study in a sexually polymorphic wild strawberry reveals climatic and soil associations of sex ratio, sexual dimorphism and sex chromosomes

Abstract

1 INTRODUCTION

Separate sexes (male and female) have evolved from hermaphroditism hundreds of times in flowering plant evolution (Renner, 2014), indicating that under some circumstances the benefits of sex specialization can outweigh the costs of reproductive uncertainty (reviewed in Pannell & Jordan, 2022; Spigler & Ashman, 2011). Indeed, there is a wide range of polymorphic sexual systems: gynodioecious (hermaphrodite and female), subdioecious (male, hermaphrodite and female) or dioecious (male and female), and these can form a continuum with intraspecific variation among populations within species (e.g. Costich & Meagher, 2001; Dorken & Barrett, 2003). Because the male and female reproductive pathways have different resource/mating demands, environmental variation can drive geographic patterns of sexual systems (reviewed in Varga & Soulsbury, 2020). Recent concerns over anthropogenic change in climate factors and soil fertility (IPCC, 2022; Penuelas et al., 2013; Singh et al., 2020) have led ecologists to call for a broader understanding of the abiotic drivers of the key features of sexually polymorphic populations, such as sex ratio, sexual dimorphism and sex determination (Hangartner et al., 2022; Hultine et al., 2016; Varga & Soulsbury, 2020). Yet, for most sexually polymorphic species the geographic relationship between-sex ratio and environment remains unexamined (Varga & Soulsbury, 2020) and evidence of clinal variation in sexual dimorphism or sex-determining factors is all but absent in plants (but see Bürli et al., 2022; Puixeu et al., 2019)—especially at the range-wide scale.

In sexually polymorphic plant populations, the sex ratio is determined by the reproductive fertility of each sex morph, the genetic mechanism of sex determination, and the degree of sex environmental lability (reviewed in Käfer et al., 2022; Schenkel et al., 2023; Spigler & Ashman, 2011). Given the lower energetic costs (Ashman, 1994; Obeso, 2002) of reproducing solely as a male (e.g. pollen production) than a female (e.g. ovule and seed production) in insect-pollinated plants, sex ratios can vary across gradients of environmental stressors (reviewed in Spigler & Ashman, 2011; Varga & Soulsbury, 2020). For instance, in gynodioecious species where hermaphrodites produce both pollen and as many seeds as females, they bear a higher reproductive cost than females. Female frequency is thus predicted to increase with increasing environmental stresses in gynodioecious species. For example, a survey of Illinois populations of Lobelia spicata demonstrated female frequency increased with increasing temperature stress (Ruffatto et al., 2015). Comparatively, in subdioecious and dioecious species, where males exist or hermaphrodites produce few seeds, the females bear the highest physiological demands of reproduction. In such systems, female frequency is predicted to decrease with increasing environmental stresses (Spigler & Ashman, 2011; Varga & Soulsbury, 2020). Yet across several subdioecious species, support for this hypothesis was mixed: female frequency increased with higher temperatures but lower water (Varga & Soulsbury, 2020). And while no environmental association was found in dioecious species (Varga & Soulsbury, 2020), higher female expenditure (and higher mortality [Marais & Lemaître, 2022]) can lead to a male-biased sex ratio, especially in long-lived iteroparous, clonal and fleshy-fruited dioecious species (Field et al., 2013). The difficulty of characterizing sexually polymorphic species as purely gynodioecious or subdioecious could also lead to murky environment-sex ratio associations. For these species, it is possible that a more complete landscape-wide view is needed to sufficiently capture variation that can reveal the underlying associations.

Sexual dimorphism of traits may also vary across environmental gradients because of sex-specific resource requirements. Sexual dimorphism arises from trait divergence in response to sex-specific adaptation or phenotypic plasticity, and these can also be context dependent (Ashman, 2005; Case & Ashman, 2007; Delph, 2019; Hangartner et al., 2022; Morgan & Ashman, 2003; Obeso, 2002). Specifically, traits may be under divergent selection through male and female fertility because of differential costs for these modes of reproduction (reviewed in Singh & Punzalan, 2018). For instance, females may be selected to invest more in leaves than males because more carbon is required to successfully mature fruit than to produce pollen (Ashman, 2005). In contrast, selection to ensure adequate mate access and pollen transfer may limit sexual dimorphism in traits like flowering time and petal sizes (Case & Ashman, 2007). When the local environment modifies the cost or the benefit of a given allocation pattern, selection and sexual dimorphism may change. For example, experimental modification of water availability impacts the direction of sex-specific selection on leaf size in dioecious Silene latifolia (Delph, 2019) and the degree of pollen limitation affects the strength of sex-specific selection on petals in experimental populations of Fragaria virginiana (Case & Ashman, 2007; Morgan & Ashman, 2003). Differences in sexual dimorphism can also arise when one sex is more responsive to environmental variation, and recent theory shows that sex-specific plasticity can promote population persistence (Hangartner et al., 2022). In Vallisneria spinulosa, females displayed higher plasticity than males in vegetative growth in response to water depth in aquatic mesocosms (Li et al., 2019), but F. virginiana hermaphrodite fruit production was more plastic than females' in response to experimentally manipulated resource availability (Spigler & Ashman, 2011). Either mechanism can lead to clinal variation in sexual dimorphism. For instance, in a common garden study, Puixeu et al. (2019) found genetic differentiation in sexual dimorphism in height and inflorescence size in Rumex hastatulus related to mean annual temperature at the source location. Likewise, Bürli et al. (2022) found that the degree of sexual dimorphism varied along climatic and elevation gradients in three wind-pollinated dioecious species due at times to greater environmental sensitivity of females.

Finally, environmental variation in sex ratio and sexual dimorphism could reflect geographic variation in genetic sex determination. Sex chromosomes are dynamic, and rapid changes in the sex-determining region (‘SDR’) can lead to polymorphism (types, haplotypes or races) within species (Palmer et al., 2019; Renner & Müller, 2021). Environmental stresses, such as those known to induce ‘leaky’ sex expression (e.g. temperature, drought, pollen limitation; Cossard & Pannell, 2021; Delph & Wolf, 2005) can also trigger evolutionary divergence of sex-determining mechanisms leading to variation along environmental gradients (Schenkel et al., 2023). Moreover, because sex chromosomes can be rich in variation for sex-specific adaptations (reviewed in Dean & Mank, 2014), sex chromosome variation may contribute to variation in sexual dimorphism. It is worth noting, however, that genes for sexually dimorphic traits can also be autosomal (Ashman, 2005; Lande, 1980; Spigler et al., 2011), and thus sexual dimorphism may not vary with sex chromosome type. Interestingly, Puixeu et al. (2019) found pronounced east–west geographic separation of the XY and XY1Y2 sex chromosome races of dioecious Rumex hastatulus as well as some, though not systemic, differences in sexual dimorphism between them. The majority of geographically widespread studies of intraspecific sex chromosome variation, however, have been performed in animals (e.g. Sniegula et al., 2022), leaving the generality of the Puixeu et al. (2019) findings an open question.

Most studies of environmental determinants of sex ratio or sexual dimorphism use only a small sample of geographically restricted sets of populations. A recent meta-analysis reported an average of six populations studied per plant species, and the highest sampled species (121 populations) did not cover the entire range (Varga & Soulsbury, 2020). These limited views restrict our ability to make inferences regarding sex-specific factors and their environmental drivers. Landscape-scale studies, however, can address how sex ratio and sexual dimorphism respond to climate and soil variation across broad and relevant sets of environmental gradients and thus offer a powerful means to address this limitation. Recent digitization of herbarium specimens and the exponential growth of iNaturalist observations have opened vast troves of data fit for novel range-wide exploration of these issues (Heberling, 2022; Heberling et al., 2021). Accordingly, we conducted the first ever continent-wide study of Fragaria virginiana, a widespread sexually polymorphic wild strawberry. We characterized ~15,000 herbarium and iNaturalist records to determine whether sex ratio and sexual dimorphism varied spatially or with climatic or soil gradients. We then used genotyping of SDR haplotypes to identify geographic and abiotic associations with known sex chromosome types for 172 herbarium samples, 47 germplasm accessions and 21 previously sequenced female plants. We explicitly tested hypotheses that (1) sex ratio (female frequency) correlates with environmental stress one of two ways: (a) increases (as predicted for gynodioecious species) or (b) decreases (as predicted for subdioecious species); (2) sexual dimorphism varies with environment in ways that reflect contrasting (resource acquisition) or similar (mate access) needs by the sexes; (3) SDR haplotypes are geospatially structured and contribute to observed variation in sex dimorphism and sex ratio.