Fig 1.
Schematic representation of experimental design used in this study to compare the differences between genetic sex determination and temperature dependent sex determination.
(A) Summary of experiment showing how the parental crosses were designed, and how eggs were allocated and incubated. Eggs from sex reversed females (ZZf) were initially incubated at 28°C for 10 days, then were switched to 36°C. Eggs were sampled at the same three developmental stages (6, 12, and 15) based on [19,20]. At stage 6 the gonad is bipotential, at stage 12 the gonad is in the early stages of differentiation, and it completely differentiated by stage 15. Eggs from concordant females (ZWf) were incubated at 28°C and sampled at the same three developmental stages as the ZZf eggs. (B) PCA plots showing the first and second principal components of read count per gene between ZZf (red) and ZWf (blue) at each stage of development.
Fig 2.
Schematic overview of gene-driven (blue) and temperature-driven (red) female developmental pathways in Pogona vitticeps. The pathways are initially different (from stages 6 to 12), but they ultimately converge on highly similar expression profiles when ovarian differentiation has occurred by stage 15. Both pathways are characterised by repression of a male signal, however this signal is stronger in temperature-driven females and appears to require ongoing repression when compared with the gene-driven females.
Fig 3.
(A) Expression (transcripts per million, TPM) ± SE of three genes differentially expressed at all three developmental stages between ZZf and ZWf, with KDM6B and CIRPB (outlined in red) having consistently higher expression in ZZf embryos, and GCA having higher expression in ZWf. (B) Bar graphs representing the number of differentially expressed genes in all comparisons between ZZf and ZWf, and between developmental stages. MA plots of this data are available in S1 Fig. Differentially expressed genes were determined as having P values ≤ 0.01 and log2-fold changes of 1, -1.
Fig 4.
Hypothesized pathway for the maintenance of the ovarian phenotype in stage 12 sex reversed ZZf Pogona vitticeps.
Given the upregulation of these genes, it is likely that reactive oxygen species induce the phosphorylation, and subsequent activation and nuclear translocation of STAT4, likely mediated by PDGFB. Once in the nucleus, STAT4 is able to bind to promoter regions of known target genes, NR5A1, AMHR2 and EGR2 to regulate their expression and promote ovarian development.
Fig 5.
(A) A subset of GO processes and (C) GO functions enriched in stage 6 ZZf embryos compared with ZWf. (B) A subset of GO processes and (D) GO functions enriched in stage 6 ZWf embryos compared with ZZf. Complete results of GO analysis for all developmental stages in ZZf and ZWf for enriched GO processes and functions is provided in S2 Data. Differentially expressed chromatin modifier (E) and cellular stress (F) genes in Pogona vitticeps at stage 6 comparing ZWf and ZZf females.
Fig 6.
K-means clustering analysis on normalized counts per million for ZZf (A) and ZWf (B) across all developmental stages. The colour depicts the correlation score of each gene in the cluster, where numbers approaching one (red) have the strongest correlation. All gene lists produced for each cluster are provided in S5 Data.
Fig 7.
Hypothesised cellular environment (A) of a ZZf gonad at stage 6 in Pogona vitticeps based on differential expression analysis (B) using the CaRe model as a framework [13]. We used this approach to understand the cellular context responsible for driving sex reversal in ZZf samples. This reveals that calcium signalling likely plays a very important role in mediating the temperature signal to determine sex. Influx of intracellular calcium is likely mediated primarily by TRPV2, and may also be influenced by KCNN1 and CACNB3. This influx appears to trigger significant changes in the cell to maintain calcium homeostasis. MCU, ATP2B1, CALR and TRICB all play a role in this process by sequestering calcium and pumping it back out of the cell, in which KCNN1 and CACNB3 may have a role. Calcium signalling molecules C2CD2, C2CDL2, and S100Z are likely responsible for encoding and translating the calcium signal leading to changes in gene transcription. Changes in gene expression are likely mediated primarily by the two major Polycomb Repressive Complexes, PRC1 and PRC2. Members of these two complexes (PCGF1, PCGF6, KDM6B, and JARID2) transcriptionally regulate genes by controlling methylation dynamics of their targets, the latter two of which have been previously implicated in sex reversal [14,15]. ATF5 may also play a role in gene regulation, and alternative splicing, which has been implicated in sex reversal [14] may be mediated by CLK4. High temperatures necessarily increase cellular metabolism, which in turn increases the amount of reactive oxygen species (ROS) produced by the mitochondria. ROS can cause cellular damage at high levels, so trigger an antioxidant response, which is observed here in the upregulation of MGST1, PRDX3, TXNDC11 and FOXO3. Also of note is the upregulation of CIRBP, which has numerous functions in response to diverse cellular stresses, and has been implicated in TSD.
Table 1.
All genes, full gene names, functional categories and associations with either gene (ZWf) or temperature driven (ZZf) female development mentioned in the paper.
NA denotes a gene that was mentioned, but was not differentially expressed. Genes with an asterisk are those that have previously been implicated in thermosensitive sex determination cascades, either in Pogona vitticeps, or in another reptile species.