Single-cell RNA sequencing identifies regulators of differentiation and nutritional cues in Drosophila female germ cells

Abstract

Drosophila ovarian germline stem cells (GSCs) are powerful model for stem cell research. However, due to the scarcity of GSCs in ovarian tissue, it is difficult to obtain the transcriptional profile of GSCs and identify novel GSC markers. In this study, we took advantage of single cell RNA sequencing (scRNA-seq) to profile the germline cells and somatic cells in wild type Drosophila ovary. We then performed an in vivo RNAi screen and network analysis to identify genes that are involved in the early stages of germ cell differentiation. We identified 33 genes with limited expression during early germ cell development and identified 19 genes that potentially regulate germ cell differentiation. Among these, an uncharacterized gene, which we named eggplant (eggpl), is highly expressed in GSC and downregulated in early daughter cells. Upon RNAi knockdown of eggpl, we observed an increase in germ cell proliferation, an accumulation of cysts in the early mitotic (2- and 4-cell) stages and an increase in overall ovary size compared to control when flies were maintained on a standard diet. In flies fed a rich yeast diet, the expression of eggpl was significantly lower and the effects of eggpl RNAi were suppressed, suggesting that downregulation of eggpl may link nutritional status to germ cell proliferation and differentiation. We also found that the matrix metalloproteases, Mmp1 and Mmp2 as well as the tissue inhibitor of metalloproteases (Timp) are additional regulators of eggpl. Collectively, this study provides new insight into the signaling networks that regulate early germ cell development and identifies eggpl as a key player in this process.

Two distinct waves of transcriptome and translatome remodelling drive germline stem cell differentiation

“…bgcn displays the opposite mRNA expression pattern, with declining mRNA levels in both the synchronised differentiating GSCs and wild type differentiation ( Figure 2Aiv ) 52 . Notably the synchronised differentiating GSC dataset recapitulates recently published mRNA expression patterns for CG32814 , which is reported to be downregulated after the GSC 41 , and blanks , which is downregulated at the 8cc stage 36 ( Figure S1A ). Collectively, validations by smFISH indicate that the RNA-seq data in synchronised differentiating GSCs mirrors what is observed in wild type ovaries both in terms of mRNA levels and changes during differentiation.…”

“…bgcn displays the opposite mRNA expression pattern, with declining mRNA levels in both the synchronised differentiating GSCs and wild type differentiation (Figure 2Aiv). Notably the synchronised differentiating GSC dataset recapitulates recently published mRNA expression patterns for genes such as CG32814, which is reported to be downregulated after the GSC 34 , and blanks, which is downregulated at the 8cc stage 42 (Figure S1A). The protocol used to drive Bam expression is also expected to induce acute changes in the expression of the heat shock responsive genes 46 .…”

“…We performed benchmarking analysis to confirm the quality and reproducibility of our sequencing datasets, and thoroughly validated them against normal wild type differentiation in comparison to the published literature and using orthogonal imaging-based methods. The RNA-seq data of synchronised differentiating GSCs reproduces the gene expression patterns identified through scRNA-seq [31][32][33][34] , and the increased depth of the data provided by RNA-seq allowed us to identify more differentially expressed genes in our dataset. More importantly, in our Ribo-seq experiment, we found that the translatome undergoes a more substantial transformation than the transcriptome during GSC differentiation, which could explain why it has been difficult to specifically distinguish signatures for each stage of differentiation from scRNA-seq data.…”

“…Developmental staging, genetic tools and FACS purification have been used to study the transcriptome of GSCs 34,35 , which have been compared to mixed differentiating cysts 36 and early follicle stages 37 . Most recently, single cell RNA sequencing has been performed on whole ovaries by several groups 38–41 , identifying mRNA expression signatures of undifferentiated germline cells, immature and mature nurse cells and oocytes, as well as many different somatic cell types of the ovary. Pseudotime analysis has been used to subcluster germline cells but, due to the low depth of existing single cell technologies, present a superficial view of the changes in the transcriptome.…”

“…Complementary low-throughput imaging-based methods can provide insight into some of the gene expression changes characterising GSC differentiation. Most recently, single cell RNA sequencing has been performed on whole ovaries by several groups [31][32][33][34] . These studies successfully identify mRNA expression signatures between undifferentiated germline cells, immature and mature nurse cells and oocytes, as well as many different somatic cell types of the ovary.…”