Description
Sex chromosome dosage differences between males and females are a significant form of natural genetic variation in many species. Like many species with chromosomal sex determination, Drosophila females have two X chromosomes, while males have one X and one Y. The model species D. melanogaster has five roughly equally sized chromosome arms, one of which is the X chromosome. However, fusions of sex chromosomes with autosomes have occurred along the lineage leading to D. pseudoobscura and D. miranda. The resulting neo-sex chromosomes are gradually evolving the properties of sex chromosomes, and neo-X chromosomes are becoming targets for the molecular mechanisms that compensate for differences in X chromosome dose between sexes. We have previously shown that D. melanogaster possess at least two dosage compensation mechanisms: the well- characterized MSL-mediated dosage compensation active in most somatic tissues, and another system active during early embryogenesis prior to the onset of MSL-mediated dosage compensation. To better understand the developmental constraints on sex chromosome gene expression and evolution, we sequenced mRNA from individual male and female embryos of D. pseudoobscura and D. miranda, from ~0.5 to 8 hours of development. Autosomal expression levels are highly conserved between these species. But, unlike D. melanogaster, we observe a general lack of dosage compensation in D. pseudoobscura and D. miranda prior to the onset of MSL-mediated dosage compensation. The extent of female bias on the X chromosomes decreases through developmental time with the establishment of MSL-mediated dosage compensation, but may do so more slowly in D. miranda than D. pseudoobscura. Thus either there has been a lineage-specific gain or loss in early dosage compensation mechanism(s), or increasing X chromosome dose may strain dosage compensation systems and make them less effective. These results also prompt a number of questions about whether species with more sex-linked genes have more sex-specific phenotypes, and how much transcript level variance is tolerable during critical stages of development. Overall design: We sequenced mRNA from D. pseudoobscura and D. miranda embryos, from eight timepoints, both female and male embryos, with three replicates (2 x 8 x 2 x 3). D. pseudoobscura embryos were F1s of a cross between Flagstaff-14 and PP1134 D. pseudoobscura lines, D. miranda embryos were F1s of a cross between the MSH22 and SP138 D. miranda lines.