We first sought to gain an indication of the frequency by which aneuploidy arises in sexually reproducing and parthenogenetic strains of D. mercatorum. We noted aneuploidy within neuronal tissue of a parthenogenetic strain as well as 6 sexually reproducing strains. We observed that 2.8-11.8% of cells had a loss or a gain of 1 or more chromosomes (Figure 1A). When we compared the incidence of aneuploidy to previously documented parthenogenetic ability (Figure 1B) [6], we found no clear correlation between parthenogenesis and aneuploidy in the different D. mercatorum strains. All the sexually reproducing strains examined were capable of different degrees of facultative parthenogenesis, some producing only embryos and others producing adult flies [6]. We found that the completely parthenogenetic strain displayed 8.7% of cells with aneuploidy whereas the strain with the highest level of facultative parthenogenesis showed only 2.8% aneuploidy. Therefore, the level of facultative parthenogenesis may not be an indicator of the prevalence of aneuploidy in the larval brain, but it may be a consequence of the underlying cause of parthenogenesis. It therefore seemed possible that a connection between parthenogenesis and aneuploidy might only be apparent when comparing parthenogens to sexually reproduced females from the same strain.
We proceeded to investigate the brain tissue of both sexually reproduced and parthenogenetic D. mercatorum offspring from a strain that exhibited an average of one adult progeny per ten virgin females screened (equivalent to 10.7% offspring per total number of females screened) (Figure 1B, 15082-1527.03) [6]. The karyotype of D. mercatorum comprises the X chromosome (corresponding to Muller element A); the 2L chromosome arm (Muller element B) fused to the 3R arm (Muller element E); a telocentric 2R arm (Muller element C); a telocentric 3L arm (Muller element D) which is slightly larger than the telocentric 2R arm; and the 4th chromosome (Muller element F) (Figure 2A) [6, 20]. The Muller elements terminology reflects the conservation of chromosome arm content in Drosophila species [21-24]. D. mercatorum shows chromosome polymorphisms for the 4th chromosome (Muller element F) and inversions present on the 2L chromosome arm (Muller element B) [6]. The karyotype of facultative parthenogenic D. mercatorum was the same as previously published [6]. However, 3 of the sexually reproduced female-appearing offspring of D. mercatorum were mosaic for the presence of one or more Y chromosomes in addition to two X chromosomes, representing a germline derived aneuploidy (Figure 2A). In D. mercatorum the Y chromosome is exceptionally small (Figure 2A) [20]. Due to the karyotype polymorphisms, we first verified the chromosome arms of the facultative parthenogen matched either the previously published karyotypes of the sexually reproducing and parthenogenetic D. mercatorum [6]. Using a fluorescence in situ hybridization chain reaction (HCR) protocol adapted for DNA sequence localization [6], we unambiguously identified the individual chromosome arms in the facultative strain of D. mercatorum (Figure 2B). This enabled us to identify which chromosome arms contribute to the aneuploidy phenotype. We were also able to classify the karyotype as matching that of the parthenogenetic D. mercatorum for the 4th chromosome polymorphism.
We observed that while both sexually reproduced and parthenogenetic offspring derived from the same lineage were diploid, 10.3% of the mitotic cells in 3rd instar larval brains of parthenogens were aneuploid in contrast to 3.6% in sexually reproduced offspring (Figure 2C-D and Table S1A). In the parthenogenetic offspring, losses of chromosome arms were more prevalent than gains (Figure 2E). The greater frequency of losses than gains has also been reported in neuroblasts and hepatocytes of mice [8, 9]. The chromosomes most frequently were lost in D. mercatorum were the X (24%) and 4th (39%) chromosome arms (Figure 2E). The mitotic cells were not clustered within the tissue and the incidence of aneuploidy was evenly distributed (Figure 2F). The nuclei in the tissue of the parthenogenetic offspring were less densely packed indicating the cells are larger compared to the sexually reproducing offspring. Therefore, parthenogenesis correlates with both aneuploidy and morphological changes to the tissue.
We previously identified genotypes able to confer facultative parthenogenetic ability to D. melanogaster by mimicking transcriptomic differences between parthenogenetic and sexually reproducing strains of D. mercatorum. We therefore asked whether we could detect aneuploidy in this genetically induced facultative parthenogenetic strain of D. melanogaster. D. melanogaster diverged from D. mercatorum approximately 47 million years ago [25], and thus its genome architecture is substantially different. D. melanogaster’s karyotype includes the X chromosome (Muller element A); chromosome 2, comprising a 2L arm (Muller element B) and a 2R arm (Muller element C); chromosome 3, consisting of a 3L arm (Muller element D) and a 3R arm (Muller element E); and the 4th chromosome (Muller element F). We began by examining the karyotypes in the larval brains of a wild-type strain of sexually reproducing Drosophila (CB1, wild-caught in 2019) [6]. None of the 99 cells examined in the 6 animals tested displayed cells with either germline or somatically derived aneuploidy (Figure 3A,E).
We next examined sexually reproducing Drosophila able to generate 1.4% offspring by facultative parthenogenesis. The maternal genotype is two extra copies of polo as a GFP tagged transgene, one extra copy of Myc as a genome duplication of the locus translocated on the 3rd chromosome (endogenous Myc is on the X chromosome), and a 16bp deletion in the 5’ UTR of Desat2 causing decreased expression [26] (GFP-polo+; Myc+dp Desat2- / TM6B). None of the animals had germline-derived aneuploidy. However, in the brain tissue of sexually reproduced GFP-polo+; Myc+dp Desat2- / TM6B larvae, we identified 3.1% of mitotic cells having somatically derived aneuploidy (Figure 3A,E). These sexually reproduced animals were exclusively diploid, in contrast to the facultative parthenogenetically produced offspring, which exhibited diploid, triploid, and tetraploid karyotypes (Figure 3B,E). We observed aneuploidy in up to 28.6% of larval brain cells of the parthenogen offspring (Figure 3D-E and Table S1B). The average percentage of aneuploid cells was 8.1% in the brains of diploid offspring; 10.4% in triploid offspring; and 12.6% in tetraploid offspring (Figure 3E), correlating the increase in ploidy with the proportion of aneuploid cells in the brain tissue of the engineered parthenogens. Notably, the 4th chromosome was most frequently lost, accounting for 21% of aneuploid cells. This is likely because the small number of genes present on the 4th chromosome makes its loss more tolerable. We did not observe any clustering of mitotic cells exhibiting aneuploidy, suggesting that a clonal event is not propagated (Figure 3G). However, similar to the findings in larval brains of D. mercatorum parthenogens, the nuclei in all larval brain cells of diploid parthenogenetic females appeared larger and less densely packed. Elevated Myc is known to result in enlarged cells in diploid tissues of Drosophila [27], and since Myc expression is high in both D. mercatorum and D. melanogaster parthenogens, we propose that this is the cause of the increased cell size in both species. Why this does not occur in the sexually reproduced animals of the same genetic background is not clear, although we speculate that there is an additional, and as of yet unidentified, factor that stochastically changes gene expression during parthenogenesis that contributes to this process.
These findings led us to carry out a more detailed exanimation of the larval brains of both the sexually reproducing and parthenogenetic D. mercatorum and the sexually reproducing and parthenogenetic GFP-polo+; Myc+dp Desat2- / TM6B D. melanogaster. In D. mercatorum and GFP-polo+; Myc+dp Desat2- / TM6B D. melanogaster there was no apparent sign of dysplastic or abnormal tissue in either sexually reproduced or parthenogenetic offspring (Figure 4A,B). The brains appeared healthy and exhibited overall similar morphological characteristics between larvae produced by the two reproductive modes within the same strain. Hence, despite the presence of aneuploidy within the brain tissue, it does not seem to have a negative impact on the overall morphology of the brain tissue. However, all cells within the tissue appeared more densely packed in the sexually reproduced offspring compared to the parthenogenetic offspring from both species (Figure 4C,D), confirming the results of the karyotype preparations (Figure 2F and 3G). Mitotic cells in sexually and parthenogenetically reproduced D. mercatorum and genetically engineered GFP-polo+; Myc+dp Desat2- / TM6B D. melanogaster larval brain tissue all appeared normal when in metaphase, regardless of reproductive mode. However, we observed that there were lagging chromosomes or anaphase bridges present in 27.1% of anaphase cells in D. mercatorum parthenogens compared to 4.2% in the sexually preproduced offspring and 35.9% in D. melanogaster parthenogens compared to 14.1% in the sexually reproduced offspring (Figure 4E-G). It seems therefore that chromosome nondisjunction is the likely cause of aneuploidy in both D. mercatorum and genetically engineered GFP-polo+; Myc+dp Desat2- / TM6B D. melanogaster parthenogens.