Are plant and animal sex chromosomes really all that different?

Abstract

Sex chromosomes in plants have often been contrasted with those in animals with the goal of identifying key differences that can be used to elucidate fundamental evolutionary properties. For example, the often homomorphic sex chromosomes in plants have been compared to the highly divergent systems in some animal model systems, such as birds, Drosophila and therian mammals, with many hypotheses offered to explain the apparent dissimilarities, including the younger age of plant sex chromosomes, the lesser prevalence of sexual dimorphism, or the greater extent of haploid selection. Furthermore, many plant sex chromosomes lack complete sex chromosome dosage compensation observed in some animals, including therian mammals, Drosophila, some poeciliids, and Anolis, and plant dosage compensation, where it exists, appears to be incomplete. Even the canonical theoretical models of sex chromosome formation differ somewhat between plants and animals. However, the highly divergent sex chromosomes observed in some animal groups are actually the exception, not the norm, and many animal clades are far more similar to plants in their sex chromosome patterns. This begs the question of how different are plant and animal sex chromosomes, and which of the many unique properties of plants would be expected to affect sex chromosome evolution differently than animals? In fact, plant and animal sex chromosomes exhibit more similarities than differences, and it is not at all clear that they differ in terms of sexual conflict, dosage compensation, or even degree of divergence. Overall, the largest difference between these two groups is the greater potential for haploid selection in plants compared to animals. This may act to accelerate the expansion of the non-recombining region at the same time that it maintains gene function within it.

This article is part of the theme issue ‘Sex determination and sex chromosome evolution in land plants’.

1. Introduction

Sex chromosomes have formed many times independently in a vast array of organisms [1] and their repeated origin has revealed many convergent evolutionary and genetic patterns [2]. Generally, sex chromosomes form when recombination is halted between a homologous pair of chromosomes, causing a variety of both adaptive and non-adaptive processes that produce distinct differences between the X and Y chromosomes (or Z and W—for simplicity, I refer mainly to male heterogamety in the remainder of this review unless discussing a specific female heterogametic system). The sex chromosomes of land plants (e.g. embryophytes) have often been contrasted with those in animals [3], and the many unique features of land plants suggested as possible explanations for perceived differences (please see box 1 for the operational delineation of plants and animals used here). For example, land plants tend to have homomorphic sex chromosomes with little divergence between the sex chromosomes in terms of gene content and sequence difference. This has been contrasted to the highly diverged, heteromorphic, sex chromosomes in some animal model systems, such as birds, Drosophila and therian mammals (see box 1 for definitions).

Box 1. A note on terminology.

For the purposes of this review, I contrast land plants, defined as the embryophytes, with animals, the Holozoa. Operationally, most of the work I discuss here is focused on the angiosperms in plants and the Deuterostomes and Protostomes in animals.

Although I argue here that plant and animal sex chromosomes may be more similar than different, the mating system terminology differs substantially between plant and animal biologists. Plant mating systems are arguably much more complicated than animals and include monoecy, where individual plants have male and female flowers; hermaphroditism, where flowers have both male and female reproductive parts; and androdioecy, where individuals can be male or hermaphrodites; with other rarer forms as well. Here the discussion is largely confined to dioecy—where all the flowers on individual plants are either male or female. The corresponding animal term is gonochorism, though because gonochorism is so common in animals, the term is somewhat obscure. I have used dioecy to refer to both plants and animals for simplicity.

Additionally, plant and animal scientists use the terms sex and gender in different, and often confusing ways. Although once used interchangeably, most animal biologists now shy away from using the term gender, which is increasingly associated with a social, rather than biological, distinction between males and females in humans. Instead, zoologists tend to focus on an individual's sex, whether it is male or female, as defined by either its sex chromosomes or overall phenotype. Gender is the equivalent term used by most plant biologists (sensu Lloyd [4]), and typically is taken to mean the relative genetic contribution to the next generation through pollen or ovules. This works for plant species with individuals that produce both types of gametes, but is not terribly helpful for dioecious plants, where individuals produce either one or the other type of gamete. The use of gender in plant biology has a long history and a rich literature, however, this term has become increasingly politically fraught in many parts of the world with the acknowledgement of human sexual diversity and gender identity. For simplicity and in acknowledgement that gender is recognized in many places as a social, rather than biological, descriptor [5], I will use the animal term, sex. I apologize to my plant biologist colleagues who will be uncomfortable about this.

Unlike terms related to sex and gender, sex chromosome descriptions are the same in plants and animals. Sex chromosome heteromorphy used to mean when the sex chromosomes differed visibly from each other in karyotypes, but has since broadened to mean where there are significant genomic distinctions indicating X-Y sequence divergence. Homomorphic sex chromosomes either show no karyotypic differences or very little evidence of divergence at the sequence level.

Divergence of X and Y gene content can also bring with it differences in gene dose between males (with one X chromosome) and females (with two X chromosomes). Differences in gene dose can lead to differences in gene expression, and compensation for this can be limited to dosage-sensitive genes, referred to as incomplete dosage compensation, or can affect all genes on the X chromosome regardless of their dosage sensitivity, referred to as complete dosage compensation [6]. Many plant sex chromosomes lack complete sex chromosome dosage compensation observed in some animals, including therian mammals, Drosophila and Anolis, and plant dosage compensation, where it exists, appears to be incomplete [7,8].

Even the canonical theoretical models of sex chromosome formation differ somewhat between plants and animals [9–12], and it may be that land plant and animal sex chromosomes exhibit fundamental differences from each other. Indeed, plants and animals differ in many essential ways, from the presence of a cell wall and chloroplast to a much more diverse array of mating systems; but how many of these differences are expected to affect sex chromosome evolution? Essentially, how different are the underlying processes driving the evolution of plant and animal sex chromosomes? For example, the highly divergent sex chromosomes observed in some animal groups are actually the exception, not the norm, and sex chromosomes in many animal clades are largely homomorphic and similar to plants in their degree of divergence [1]. Furthermore, although some animals do indeed exhibit complete sex chromosome dosage compensation, these are also the exception, and most animals are similar to plants with incomplete dosage compensation [6].

Which of the many unique properties of land plants would be expected to affect sex chromosomes differently than animals, and in what ways? How can we use these properties to elucidate fundamental evolutionary processes?

2. Dioecy is derived in most land plants and is ancestral in animals

One of the most important factors distinguishing plants and animals in terms of sex chromosome evolution is the difference in the prevalence of hermaphroditism and monoecy, where a single organism expresses both sexes, versus dioecy, where an individual is either male or female, in the two groups. Dioecy is derived, and relatively rare in most land plants [3,13,14], whereas the vast majority of animals, and the presumed ancestor, are outcrossing gonochorists [1], referred to hereafter as dioecious for simplicity (see Box 1). There are several ways this might lead to differences between plant and animal sex chromosome systems.

Land plant sex chromosomes are generally homomorphic [1,3], although with notable exceptions in Cannabis [15], Rumex [16] Silene [17] and Coccinia [18,19]. Several have hypothesized that the low level of divergence between plant sex chromosomes, in general, is a result of the fact that dioecy is derived and, therefore, phylogenetically relatively young in land plants [20,21]. The prevalence of homomorphic plant sex chromosomes is in sharp contrast to the highly conserved and diverged sex chromosomes in some animal groups, such as therian mammals, birds and Lepidoptera [22,23]; but is heteromorphy really the rule in animals? One could argue that these clades represent the exception in metazoans, as the vast majority of animals either determine sex without sex chromosomes [1] or with homomorphic sex chromosomes [24,25].

Just as Silene, Cannabis, Coccinia and Rumex represent the heteromorphic exceptions in plants [15–19], heteromorphic animal sex chromosomes are an exception as well and homomorphy is a general, rather than plant-specific, rule. Furthermore, we now know that the rate of sex chromosome divergence is neither unidirectional nor correlated with time [1,26–29], and degeneration of the Y can occur very rapidly in both animals and plants [17,30]. It, therefore, remains unclear whether the preponderance of homomorphy is a function of the relatively recent origins of dioecy in the clade, or homomorphic sex chromosomes are simply more common in general, and heteromorphy is an exception.

One way to test the link between recent dioecy and sex chromosome homomorphy is to look to animal clades with frequent transitions among sexual systems. Most illustrative are perhaps the ray-finned (Actinopterygiian) fishes, a large clade comprising most fish species, originating 400 Ma [31]. The majority of Actinopterygiian species are dioecious, as are their outgroups, and it is worth noting that although there are multiple instances of hermaphroditism in fishes, they are nearly all recently evolved and confined to the tips of the phylogeny [32]. Dioecy is, therefore, the presumed ancestral state for the Actinopterygii, exactly opposite to land plants, where monoecy/hermaphroditism is the presumed ancestral mating system. Despite this key difference, most sex chromosomes in the Actinopterygii are either homomorphic or very slightly diverged [25]. Indeed, there is only one clade observed as yet where recombination has been suppressed between the X and Y for all but a highly restricted pseudoautosomal region, observed in Poecilia picta and Poecilia parae [26,33]. Interestingly, this phylogenetic distribution of this sex chromosome indicates that it probably originated only roughly 20 Ma [26,34], and has remained largely homomorphic in other species, including Poecilia reticulata [26,35–38], highlighting the fact that sex chromosome divergence is not necessarily a linear process correlated with time.

Furthermore, there is a clear counter point to the link between homomorphy and recent dioecy in the Salicaceae (willows and poplars). This clade has ancient dioecy [39] and yet homomorphic sex chromosomes remain the rule owing to repeated turnover of sex chromosomes [40–45]. Given the preponderance of homomorphic sex chromosomes in many animal clades, and the fact that Salicaceae retain homomorphic sex chromosomes despite ancient dioecy, perhaps heteromorphy is simply an exception rather than the rule of sex chromosome evolution, both for land plants and animals. This remains an interesting area for further quantitative and comparative study.

3. Does dioecy require the presence of sex chromosomes?

The classic model for sex chromosome evolution in plants is slightly different from that in animals. In animals, the classic model posits that sex chromosomes arise from a pair of autosomes following the acquisition of a single master sex-determining locus which determines males (if on a Y chromosome) or females (if on a W) [9,10]. Whether or not recombination is arrested in the region around this region between the emerging X and Y chromosomes owing to adaptive [9,10,12] or non-adaptive processes [46,47], a single locus is sufficient to trigger the development of one sex from the other (figure 1a).

Figure 1.

Model of sex chromosome evolution and turnover. (a) Rewiring of the sex determination pathway (dashed line) is a common cause of sex chromosome turnover in animals, and this often involves the change in regulatory position of a gene in the sex determination pathway [48]. In these cases, the chromosome with the new apical sex determining locus has the potential to become a new sex chromosome if recombination suppression develops. The old X chromosome (in red) reverts to being an autosome and the Y (blue) is likely to be lost. (b) Rewiring of the plant two-locus model is somewhat more complicated, as it requires close proximity to dominant female sterility (F^S) male fertility (M) loci on the Y, and recessive female fertility (f^S)and male sterility (m) loci on the X. A turnover of sex chromosomes through the change in regulatory precedence (dashed lines) of the sex determining cascade would require proximity of the new male and female sex determination genes, and this is unlikely, hampering turnover events through this mechanism. Figure by Jacelyn Shu of jacelyndesigns.com.

There are excellent examples of animal sex chromosomes that follow this model. The autosomal origin of sex chromosomes is evident as well in the shared gene content of X and Y chromosomes observed in therian mammals [49] and Z and W chromosomes in birds [50,51] and snakes [52], among others. The single sex-determining locus is exemplified by the tiger pufferfish, where maleness is associated with a single missense single nuceotide polymorphism (SNP) in anti-Müllerian hormone receptor type II (Amhr2) [53]. Individuals with two functional copies of Amhr2 develop as females, those with one copy of the missense SNP develop as males and this SNP is inherited through the patriline as a proto-Y chromosome.

However, there is an ever-increasing list of exceptions to this model, both in terms of the autosomal origin of the sex chromosomes and the single sex-determining locus. For example, there is evidence that Y chromosomes in Rhinocola aceris and Cacopsylla peregrina [54,55], as well as a W chromosome in some Lake Malawi cichlids [56] and possibly some Lepidoptera [23] arose from B chromosomes, parasitic supernumary chromosomes, and do not share homology with their corresponding X or Z chromosome. In the case of the pillbug (Armadillium vulgare), the W chromosome arose from a Wolbachia feminizer [57]. Moreover, some fishes do not have a single sex-determining locus, and instead, sex is a polygenic threshold trait [58]. These examples do not so much challenge the canonical model of animal sex chromosome evolution, but rather indicate that there are a plethora of viable models, and no single one explains all origins of sex chromosomes.

The plant model of sex chromosome evolution is also based on an autosomal origin, but instead of a single locus initiating the development of one sex, the plant model requires two linked sterility loci, one each for female and male reproductive function [11,59] (figure 1b). This key difference results from the assumption that dioecy is derived in land plants from ancestral monoecy or hermaphroditism, in contrast to the assumption of ancestral dioecy in animals. The two-locus model has some clear support, for example in grapes [60], kiwifruit [61] and asparagus [62]. However, just as there are many exceptions to the canonical animal model of sex chromosome formation [27], the explosion of sex chromosome studies in plants has also offered up many exceptions. For example, in Mercurialis annua, fertility for at least one sex is quantitative rather than controlled by a single locus [63], and this may allow for the rapid transition away from dioecy in response to ecological shifts [64]. Furthermore, evidence from persimmon [65], the Salicaceae [40,42] and spinach [66] among others suggest sex determination is single-locus in these species, rather than controlled by two linked reciprocal sterility loci.

This increasing evidence for a diversity of sex chromosome models in land plants leads to the question of whether the origin of dioecy is dependent upon, and concomitant with, the origin of sex chromosomes. Sex chromosomes and dioecy are sometimes used interchangeably in the botanical literature, and the two-locus model of sex chromosome formation implies that the evolution of true dioecy is dependent upon linkage and recombination suppression between the male- and female- sterility loci on the sex chromosomes. Said another way, the emergence of sex chromosomes is required for the origin of complete dioecy in this model. However, the increasing number of exceptions to this rule opens the door to the possibility that this need not be the case. Instead, there may be dioecious lineages that lack sex chromosomes entirely, where sex is a threshold trait based on a polygenic architecture, as observed in some species of fishes [58,67]. Genetic mapping of sex in a greater array of dioecious and sub-dioecious species will be useful to determine whether this is the case, as will selection experiments to determine the genetic architecture of the transition between dioecy and subdioecy (e.g. [63]).

Ancient dioecy in the Salicaceae [39] coupled with the frequent turnover of sex chromosomes throughout the clade [41,42,44] with single-locus sex determination [40,42] also suggests that the origin of dioecy may be somewhat decoupled from the origin of sex chromosomes in this group of plants. However, at this point, it is unclear whether all dioecious species in the clade have sex chromosomes, or there are intermediate states of dioecy without sex chromosomes.

Incidentally, turnover of sex chromosomes, like that observed in the Salicaceae, may be more prevalent in systems that do not follow the two-locus model. Although turnover of the two-locus model is possible through transposition of the entire linkage group, such as that seen in wild strawberry [68], the two-locus model may make turnover by changes in the master sex-determining loci somewhat unlikely. This is because it would require rewiring of both male and female sex determination pathways and physical proximity of the new reciprocal sterility loci. By contrast, a single-locus sex determination model only requires regulatory rewiring of an existing sex determination pathway [69], allowing a new locus to initiate the sex determination cascade and precipitating the formation of a new sex chromosome pair (figure 1).

There is as yet little evidence that some plant sex chromosomes do not derive from a pair of homologous autosomes, in contrast to the B chromosome and Wolbachia origins in some animals [23,54–57]. This is perhaps somewhat surprising given the prevalence of cytoplasmic male sterility factors in plants. One could imagine a system similar to the acquisition of a Wolbachia feminizer as a W chromosome in pillbugs [57], where the male sterility factor could become a W chromosome, or the male sterility resistance locus could conceivably become a Y chromosome (J. Willis 2019, personal communication).

Given the remarkable diversity of plant genomes, it would be surprising if there were not as many exceptions to the canonical model of sex chromosome evolution as we observe in animals. It may well be that time and detailed studies of more independent plant sex chromosome systems will reveal the same wealth of mechanisms.

4. Plants may have less sexual conflict…

Intralocus sexual conflict, where an allele benefits the fitness of one sex at some cost to another, is part of the classic adaptive model of sex chromosome divergence. In essence, this model posits that there is selection against recombination between the X and Y chromosome to sequester male benefit/female harm alleles on the Y chromosome, so they are always transmitted with the region of the genome that is associated with male development [9,12,70]. In theory, if sexual conflict is important in the formation and subsequent expansion of the non-recombining region, we might expect more origins and more rapid enlargement of the non-recombining region, and therefore sex chromosome heteromorphy, in species with greater sexual conflict.

The sex chromosome model of sexual conflict has been difficult to assess directly owing to difficulties in quantifying sex conflict and identifying the underlying genetic architecture. Although there is evidence of sexual conflict in some land plants, including Silene [71–73] and Collinsia [74,75], it may be that key differences in plant mating systems may make them less prone to sexual conflict than animals. However, sexual conflict is actually quite difficult to estimate directly, as this requires determining male and female reproductive fitness of the same genotypes [76,77], or long-term pedigrees coupled with fitness data [78], and we have precious few examples in animals. The best examples of studies of sexual conflict have been done under laboratory conditions (e.g. [79]) and the simple environment in the laboratory may exacerbate sexual conflict compared to more complex natural settings [80,81]. This means that quantification of the degree and genetic architecture of sexual conflict in natural populations remains unclear in both plants and animals. More direct estimates of sexual conflict in both taxonomic groups, particularly in the wild or under semi-natural conditions in the laboratory, would be extremely helpful in establishing the prevalence, magnitude and locus of sexual conflict.

Although differences between plants and animals in the manifestation of sexual conflict are not clear, many have noted that overall sexual dimorphism in plants may be less than that we frequently observe in animals [82–84]. Phenotypic sex differences are an indirect measure of resolved sexual conflict, as dimorphism evolves to allow each sex to reach separate fitness optima. Although only fitness experiments can reveal the degree of potential sexual conflict remaining unresolved within the genome, phenotypic dimorphisms can be useful given the difficulties of estimating conflict directly. In theory, if sexual dimorphism is less prevalent in plants, this might suggest the potential for sexual conflict is also somewhat less than in animals, and therefore we might expect less selection against recombination suppression around the sex-determining region.

However, it is also possible that sexual dimorphism in plants is simply less visible than in animals, but still substantial. For example, sex-biased genes, loci with differences in transcription between males and females, have been suggested to underlie at least some phenotypic sexual dimorphism and represent the resolution of sexual conflict over gene regulation [85]. Proportions of the transcriptome with sex-biased gene expression vary extensively depending on developmental stage and among tissues, particularly between reproductive and vegetative tissue, as well as the statistical thresholds. Although the proportion of expressed genes with sex-biased expression in reproductive tissues appears to be slightly less in land plants than animals (table 1), they are still substantial, particularly for all species other than Silene. Moreover, the proportion of sex-biased genes in vegetative tissues is within the range seen in animal somatic tissues (table 1).

Table 1.

Proportion of transcriptomes with sex-biased gene expression in land plants and animals.

	tissue type	statistical thresholds	% sex-biased	ref.	notes
land plants
reproductive tissues
Populus balsamifera	flower	1% FDRa, 2 FCb	35%	[86]	flower
Silene latifolia	flower	5% FDRa	19%	[87]	includes 903 contigs with sex-limited expression
Salix viminalis	catkin	5% FDRa, 2 FCb	43%	[88]
vegetative tissues
Asparagus officinalis	spear tip	5% FDRa, 2 FCb	2%	[89]	excludes 15 contigs with male-limited expression
10 Leucadendron species	leaf	5% FDRa, 2 FCb	0.1–2.5%	[90]
Mercurialis annua	leaf	5% FDRa, 2 FC	0.2%	[91]	from stage III, includes sex-specific genes
Populus balsamifera	leaf	1% FDRa, 2 FCb	0.0003%	[86]
Silene latifolia	leaf	5% FDRa	2%	[87]
Salix viminalis	leaf	5% FDRa, 2 FCb	0.09%	[88]
animals
reproductive tissue
Drosophila melanogaster	gonad	5% FDRa, 2 FCb	64%	[92]
Meleagris gallopavo	gonad	5% FDRa, 2 FCb	51%	[93]	excludes all Z-linked genes
two Ruditapes species	gonad	5% FDRa, 2 FCb	41%	[94]	adult
Gryllus bimaculatus	gonad	5% FDRa, 2 FCb	31%	[95]
somatic tissue
Drosophila melanogaster	brain	5% FDRa, 2 FCb	2%	[96]
Meleagris gallopavo	brain, spleen	5% FDRa, 2 FCb	0–0.02%	[93]	excludes all Z-linked genes
					adult
Gryllus bimaculatus	ventral nerve cord, brain	5% FDRa, 2 FCb	1–4%	[95]

^aFalse discovery rate correction. Lower numbers are more stringent, with 5% a common threshold.

^bFold-change in difference in expression between males and females. Higher numbers are more stringent, and twofold differences (2 FC), expressed twice as much in one sex compared to the other, is a common threshold.

This may suggest that sexual dimorphism in plants exists in traits not readily perceived by human senses, such as cell type abundance or immunity for example. Recent measures of more cryptic forms of dimorphism reveal substantial differences in Salix purpurea [97], suggesting the exact nature of possible cryptic plant sexual dimorphism remains an interesting area for further study. Systematic measurements of the phenotypic effects of knockout mutations in males and females in dioecious land plants (e.g. [98]) may be needed to determine cryptic sex differences, as well as single-cell RNA-Seq studies to determine sex differences in cell type abundance. Similarly, an expansion of species-specific studies quantifying sexual conflict over various fitness traits will also be a valuable addition, particularly across different environments (e.g. [77]).

If plants do experience less sexual conflict than animals, it implies smaller non-recombining regions if sexual conflict is important in recombination suppression [3], and might suggest that plant sex chromosomes will largely remain homomorphic over time when compared with animals. However, there are a few reasons to question whether lower overall sexual conflict in plants, if this is indeed the case, might predispose them to greater homomorphy than animals. First, there is increasing discussion that the sexual conflict model may not explain recombination suppression in some systems [46,47], and it may be that the non-recombining region of a Y chromosome may simply represent a male recombination cold spot around the sex determining gene [99]. Other adaptive models that do not invoke sexual conflict have also been proposed [100,101]. It is worth noting that one of the key animal sexual selection model systems, the Trinidadian guppy (Poecilia reticulata) [36] has only a small ancestral conserved non-recombining region [26,35,37,38] while the same sex chromosomes in close relatives are highly diverged [26,33]. In this case at least, sexual conflict, sexual dimorphism and sexual selection are by no means a good predictor of X-Y divergence across species. If sexual conflict does not predict sex chromosome divergence, it cannot affect land plant and animal sex chromosomes differently.

Interestingly, some neutral models of sex chromosome divergence imply that heterochiasmy, differences in the male and female recombination landscapes, may foster sex chromosome divergence [46,47]. Although sex differences in the recombination landscape are well documented in animals and often quite striking [102–104], they may be less pronounced in land plants [105], possibly reducing the relative potential for sex chromosome heteromorphy.

5. … but haploid selection is more prevalent in plants

Although it is unclear whether sexual conflict differs between plants and animals, or how it important it is in sex chromosome divergence, plants definitely do experience a much longer haploid phase and stronger haploid selection [106,107] than animals as a result of fundamental differences in life history between the two kingdoms (figure 2). At first thought, this might suggest that sex chromosome divergence would be retarded in plants more than animals. The loss of recombination suppression reduces the power of selection on the Y chromosome, often leading to a rapid loss of gene activity for Y-linked genes [2]. Selection to preserve gene dose balance between the sex chromosomes and autosomes can act to maintain gene activity for dosage sensitive Y-linked loci [108], although expression for most dosage sensitive Y loci is lost as buffering and dosage compensation on the X chromosome reduces selection for dosage balance on the Y [109–111] for diploid expressed genes.

Figure 2.

Haploid selection and the maintenance of gene activity on the Y chromosome. Proportion of life cycle and proportion of genes expressed in the haploid phase in animals (a) and seed plants (b). Figure by Jacelyn Shu of jacelyndesigns.com.

For Y-genes expressed in the haploid phase, buffering and compensation by the X will not counter loss of Y expression (figure 2), as only either the X or Y chromosome is present in each haploid cell. We might, therefore, expect greater selection to maintain expression from essential Y-linked genes in species with greater haploid selection, e.g. plants. Indeed, haploid-expressed Y-linked loci are more likely to be maintained in Rumex [112] and Silene [113]. For bryophytes and brown algae, where the majority of the life cycle is spent in the haploid phase, gene loss from the sex chromosomes in these groups is even more retarded than in land plants [114–116].

This difference in the length of the haploid phase and the strength of haploid selection could lead to differences between plants and animals in dosage compensation. As gene activity from the Y chromosome is lost, males (with one X) experience a reduction in gene dose compared to females (with two X chromosomes). Differences in gene dose can lead to differences in gene expression, and compensation for this can be limited to dosage-sensitive genes, referred to as incomplete, or can encompass the entire X chromosome, referred to as complete dosage compensation [6]. Plant dosage compensation, where it exists, appears to be incomplete [7,8], and the maintenance of expression for haploid-expressed dosage-sensitive and essential Y-linked genes might lessen the need for complete X chromosome dosage compensation [7,8,117]. That said, dosage compensation is more often incomplete in animals as well, with only a few well-known examples of complete dosage compensation [6]. Therefore, it is unclear whether plants and animals really differ in the type of sex chromosome dosage compensation.

At the same time that it might prevent degeneration of some Y-expressed genes, the extended haploid phase in plants might actually accelerate recombination suppression between the X and Y. Haploid selection has been proposed to drive suppressed recombination and even sex chromosome turnover [100,101,107], and it is reasonable that this might lead to greater expansions on the non-recombining region in plants than animals, with their extended haploid phase. The longer haploid phase in plants also carries with it the potential for substantial ploidally antagonistic selection [118], where alleles are under opposing selection pressures during the diploid and haploid phases. Ploidally antagonistic selection can also lead to sex chromosome divergence in theoretical models [21]. Overall, the extended haploid phase in plants has the potential to increase the region of recombination suppression, and therefore the number of Y-linked genes, at the same time that it may preserve gene activity from these loci.

6. Concluding remarks

How different are sex chromosomes in land plants and animals? Sex chromosomes are definitely rarer in land plants [3], as is dioecy overall [14], and there are few instances of sex chromosome conservation over vast evolutionary distances, as we observe in birds and therian mammals. However, most animals display rapid sex chromosome origin, turnover and loss, similar to land plants in general, and interesting questions remain about whether the rate of origin and divergence between the sex chromosomes really differs between these two groups. Furthermore, the role of sexual conflict in sex chromosome divergence, and the degree to which it differs between animals and land plants, also remains an interesting area for further work. Finally, new models about the role of haploid selection [100,101,107] and ploidally antagonistic selection [21] in sex chromosome divergence have been proposed, and these suggest that plant sex chromosomes might actually experience greater rates of divergence than sex chromosomes in animals. Broader comparative and detailed species-specific data are still very much needed to understand the key differences between sex chromosome evolution in these two groups.

Data accessibility

This article has no additional data.

Authors' contributions

J.E.M. wrote the manuscript.

Competing interests

I declare I have no competing interests.

Funding

I am extremely grateful for funding from the European Research Council (grant agreement 680951), a Canada 150 Research Chair and the National Sciences and Engineering Research Council of Canada.