Complete Dosage Compensation and Sex-Biased Gene Expression in the Moth Manduca sexta

Sex chromosome dosage compensation balances homogametic sex chromosome expression with autosomal expression in the heterogametic sex, leading to sex chromosome expression parity between the sexes. If compensation is incomplete, this can lead to expression imbalance and sex-biased gene expression. Recent work has uncovered an intriguing and variable pattern of dosage compensation across species that includes a lack of complete dosage compensation in ZW species compared with XY species. This has led to the hypothesis that ZW species do not require complete compensation or that complete compensation would negatively affect their fitness. To date, only one study, a study of the moth Bombyx mori, has discovered evidence for complete dosage compensation in a ZW species. We examined another moth species, Manduca sexta, using high-throughput sequencing to survey gene expression in the head tissue of males and females. We found dosage compensation to be complete in M. sexta with average expression between the Z chromosome in males and females being equal. When genes expressed at very low levels are removed by filtering, we found that average autosome expression was highly similar to average Z expression, suggesting that the majority of genes in M. sexta are completely dosage compensated. Further, this compensation was accompanied by sex-specific gene expression associated with important sexually dimorphic traits. We suggest that complete dosage compensation in ZW species might be more common than previously appreciated and linked to additional selective processes, such as sexual selection. More ZW and lepidopteran species should now be examined in a phylogenetic framework, to understand the evolution of dosage compensation.


Figure S2
Average Z-linked and autosomal contig expression levels (FPKM) across four replicate males and females, filtering out contigs with <2 FPKM (for contig n see Table 1). Black lines are the median of the FPKM distribution across contigs, boxes show the interquartile range, whiskers extend to 1.5 × the interquartile range and notches approximate the 95% confidence intervals of the medians. Overlapping notches are evidence for the similarity of median values. Autosomes Z

Figure S3
Average Z-linked and autosomal contig expression levels (FPKM) across four replicate males and females, filtering out contigs with <3 FPKM (for contig n see Table 1). Black lines are the median of the FPKM distribution across contigs, boxes show the interquartile range, whiskers extend to 1.5 × the interquartile range and notches approximate the 95% confidence intervals of the medians. Overlapping notches are evidence for the similarity of median values. Autosomes Z

Figure S5
Average Z-linked and autosomal contig expression levels (FPKM) across four replicate males and females for sex-biased genes with putative physical locations (autosomal n = 570, Z n = 34). Physical locations were determined through 1:1 orthologs in B. mori. Black lines are the median of the FPKM distribution across contigs, boxes show the interquartile range, whiskers extend to 1.5 × the interquartile range and notches approximate the 95% confidence intervals of the medians. Overlapping notches are evidence for the similarity of median values.

Table S1
Average expression level of genes, and average expression difference (%), according to chromosomal location (Z or autosome), sex and filtering method. See Table 1 Spreadsheet table headings   Table S3 Mapping statistics for sub-sampled libraries. Each library was down-sampled to 13 million reads and re-mapped to the contig assembly for differential expression analysis.

Table S4
Full GOrilla functional enrichment test results. GO term IDs and descriptions are provided, along with each GO term p-value and FDR, the total number of genes (N), the total number of genes associated with a specific GO term (B), the number of genes in test set (n) and the number of genes in the intersection (b). Enrichment = (b/n) / (B/N).

Table S5
Full DAVID functional enrichment clustering test results. Each cluster includes the enrichment score (the geometric mean of single annotation term p-values), the category of annotation term, the term itself with description, the number of genes with that annotation in the test set (Count), gene IDs (Genes), fold enrichment and multiple-test correction statistics (Bonferroni, Benjamini and Hochberg, and FDR).

Table S6
List of significant sex-biased contigs, including D. melanogaster orthologs and gene ontology annotations identified through a BLAST search (see Methods).

Table S7
Database of all BLAST hits for M. sexta contigs to the D. melanogaster genome (see Methods), including M. sexta contigs IDs, D. melanogaster transcript and gene IDs, and related gene ontology terms.

Table S8
Full results of the differential expression analysis between males and females. Results for each contig (Contig ID) are shown, along with the log fold change (logFC), log counts per million (logCPM), likelihood ratio (LR) and contig-specific p-value and FDR.