Detection of transgenerational spermatogenic inheritance of adult male acquired CNS gene expression characteristics using a Drosophila systems model

Abstract

Available instances of inheritance of epigenetic transgenerational phenotype are limited to environmental exposures during embryonic and adult gonadal development. Adult exposures can also affect gametogenesis and thereby potentially result in reprogramming of the germline. Although examples of epigenetic effects on gametogenesis exist, it is notable that transgenerational inheritance of environment-induced adult phenotype has not yet been reported. Epigenetic codes are considered to be critical in neural plasticity. A Drosophila systems model of pentylenetetrazole (PTZ) induced long-term brain plasticity has recently been described. In this model, chronic PTZ treatment of adult males causes alterations in CNS transcriptome. Here, we describe our search for transgenerational spermatogenic inheritance of PTZ induced gene expression phenotype acquired by adult Drosophila males. We generated CNS transcriptomic profiles of F(1) adults after treating F(0) adult males with PTZ and of F(2) adults resulting from a cross between F(1) males and normal females. Surprisingly, microarray clustering showed F(1) male profile as closest to F(1) female and F(0) male profile closest to F(2) male. Differentially expressed genes in F(1) males, F(1) females and F(2) males showed significant overlap with those caused by PTZ. Interestingly, microarray evidence also led to the identification of upregulated rRNA in F(2) males. Next, we generated microarray expression profiles of adult testis from F(0) and F(1) males. Further surprising, clustering of CNS and testis profiles and matching of differentially expressed genes in them provided evidence of a spermatogenic mechanism in the transgenerational effect observed. To our knowledge, we report for the first time detection of transgenerational spermatogenic inheritance of adult acquired somatic gene expression characteristic. The Drosophila systems model offers an excellent opportunity to understand the epigenetic mechanisms underlying the phenomenon. The finding that adult acquired transcriptomic alteration in soma is spermatogenically inherited across generations has potential implications in human health and evolution.

Figure 1. Hierarchical clustering of CNS expression profiles.

City Block similarity metric and average linkage methods were used for hierarchical clustering of arrays. The cluster was generated using Acuity 4.0 (Molecular Devices). Each time point represents mean of normalized log₂ ratio (635/534) of four biological replicates with balanced dye-swaps. Note clustering of PTZ treated males' profile with their grandsons (a). Cluster shown in (b) represents all the profiles in (a) and a freshly generated F₁ male profile. Reproducibility of expression profiling is evident from similarity between the two F₁ male profiles (b). PTZ profile shown here was derived from previously reported microarray data related to seven days of drug treatment. * indicates replication set.

Figure 2. Overlap between differentially expressed CNS genes.

PTZ genes used in this analysis was derived from previously reported microarray data related to seven days of drug treatment. PTZ genes were compared to those in F₁ and F₂ generations, whereas F₁ male genes were compared only to F₂. Hypergeometric distribution p values (−log₁₀) are plotted on y-axis. Note significant overlap (≥1.3) in all except F₂ female pair-wise comparisons. Differentially expressed genes used in the analysis are listed in Table S1.

Figure 3. Overlap between differentially expressed testis and CNS genes.

PTZ genes used in this analysis was derived from previously reported microarray data related to seven days of drug treatment. PTZ CNS genes were compared to those in PTZ testis (F₀) and F₁ CNS, whereas F₁ testis genes were compared only to F₁ CNS. Hypergeometric distribution p values (−log₁₀) are plotted on y-axis. Note significant overlap (≥1.3) between PTZ CNS and F₀ testis genes and an overlap between F₁ CNS and F₁ testis genes with borderline significance (1.24). Differentially expressed genes used in the analysis are listed in Table S1 and S3.

Figure 4. Hierarchical clustering of expression profiles of testis and CNS.

City Block similarity metric and average linkage methods were used for hierarchical clustering of arrays. The cluster was generated using Acuity 4.0 (Molecular Devices). Each time point represents mean of normalized log₂ ratio (635/534) of four biological replicates with balanced dye-swaps. PTZ profile shown here was derived from previously reported microarray data related to seven days of drug treatment.

Figure 5. Validation of microarrays using RT-PCR.

Genes were either upregulated (a) or not differentially regulated (b) in F₂ male microarrays. Equal amount of F₂ male RNA samples from all the four biological replicates used in microarray experiment represented in Figure 1 were pooled together for use in RT-PCR. Fold-change values are plotted on y-axis. Note general downregulation of gene expression in RT-PCR.

Figure 6. Electrophoretic analysis of F₂ male CNS total cellular RNA.

Equal amount of F₂ male RNA samples from all the four biological replicates used in microarray experiment represented in Figure 1 were pooled together for electrophoretic analysis. Note higher amount of rRNA in experimental flies (a) compared to control (b).