5.1. The genetics of sex determination

The dioecy of Cannabis is genetically controlled (Figure 2). Hemp is diploid (2n = 20), with nine pairs of autosomes and one pair of sex chromosomes. Female plants are homogametic with XX chromosomes and male plants are heterogametic with an XY sex chromosome pair (Moliterni et al., 2004). Cannabis thus represents a rare case among the flowering plants in which sex chromosomes have been identified (Charlesworth, 2016).
The diploid genome size of female Cannabis plants is estimated to be 1636 Mbp, that of a male plant 1683 Mbp by flow cytometry (Sakamoto et al., 1998). The sex chromosomes of Cannabis are the largest in the chromosomal complement, they are estimated to comprise 6.5 % (Y chromosome) and 6.1 % (X chromosome) of the total length of the genome (Divashuk et al., 2014). Assuming that those estimates roughly correspond to length in base pairs (which is most certainly an oversimplification), the X chromosome would be 102.7 Mbp in size, and the Y chromosome 109.4 Mbp. This is close to the size of the X and Y chromosomes as determined by genome sequencing (Figure 5, Supplementary Table 1). However, the flow cytometry results mentioned above (Sakamoto et al., 1998) indicate that the Y chromosome is 47 Mb larger than the X chromosome (1683 Mb - 1636 Mb = 47 Mb). Flow cytometry analyses of otherCannabis varieties yielded very similar results (Faux et al., 2014). It is not clear where the discrepancy between genome sequencing and flow cytometry measurements is coming from. Structural genomic variations between different Cannabis plants as well as problems in assembling the sex chromosomes may both play a role.
Detailed analyses of the sex chromosomes revealed that the pseudo-autosomal region on the X chromosome (i.e. the region still recombining with the Y chromosome) is ca. 30 Mb in size, whereas the X-specific region (which is not recombining with the Y chromosome) is ca. 75 Mbp in size (Prentout et al., 2020). Prentout et al. also identified ca. 500 sex-linked genes, i.e. alleles that are inherited in a sex-linked fashion (e.g. only from father to daughter, not from father to son for X-hemizygous alleles). It will be especially interesting to analyse the X chromosome alleles that have no Y chromosome counterpart in detail in the future, as they may contribute to sex determination.
Whereas gene density on the X chromosome appears to be similar to autosomes, about 70 % of the genes on the Y chromosomes have been estimated to be lost (Prentout et al., 2020). One explanation for the relatively large size of the Y chromosome despite substantial gene losses seems to be the accumulation of transposons and other repetitive elements (Sakamoto et al., 2000). Prevalence of transposable elements on the Y chromosome might also help to explain difficulties in chromosome assembly and discrepancies in size estimates.
Despite progress in identification and sequencing the sex chromosomes of Cannabis , not much is known about the molecular circuits involved in sex determination. Some confusion exists as to what the genetic ‘mode’ of sex determination is (Kovalchuk et al., 2020; Vergara et al., 2016). Some reports suggest the mode is similar to humans and other mammals, where the presence or absence of the Y chromosome determines the sex: humans carrying a Y chromosome are almost always phenotypically male, those without a Y chromosome are female, with autosomes or extra X chromosomes bearing little consequence on the sex determination (Gamble and Zarkower, 2012; Sakamoto et al., 1998). An alternative view is that the X chromosome to autosome ratio determines the sex (Westergaard, 1958). This would be somehow similar to Drosophila, where the number of X chromosomes determines the sex, with the presence or absence of the Y chromosome having limited relevance (Gamble and Zarkower, 2012). In this model, the Y chromosome essentially becomes a ‘placeholder’, lowering the number of X chromosomes.
Experimental evidence supporting the one or the other mode of sex determination is surprisingly scarce. Warmke and Davidson used colchicine to produce tetraploid Cannabis plants (Warmke and Davidson, 1944). Strikingly, a cross between a tetraploid female and a diploid male plant yielded female and female-hermaphrodites, but no male plants, though half of the progeny should have an XXY chromosome constitution (Warmke and Davidson, 1944). This is evidence that the Y chromosome does not play a role as prominent as in humans in sex determination in Cannabis, but that an X to autosome ratio model for sex determination might apply.
It is important to note that, even if an X to autosome ratio model applies in Cannabis, this does not mean that the Y chromosome is dispensable for the development of male plants. In Drosophila, the Y chromosome is not involved in sex determination but encodes genes required for male fertility, such that XO flies are phenotypically male but sterile (Gamble and Zarkower, 2012). Likewise, Y chromosome specific genes in Cannabis may well play a critical role in male plant development, even if not involved in bona fide sex determination (McKernan et al., 2020).