Abstract
Both somatic cells and germ cells must establish their correct sexual identity for proper gametogenesis. In Drosophila, sex determination in somatic cells is controlled by the switch gene Sex lethal (Sxl), which is activated in females by the presence of two X chromosomes. Though germline sex determination is much less well understood, Sxl is also essential for the female identity in germ cells. Loss of Sxl function in the germline results in ovarian germline tumors, a characteristic of male germ cells developing in a female soma. Further, Sxl expression is sufficient for XY (male) germ cells to produce eggs when transplanted into XX (female) somatic gonads. As in the soma, the presence of two X chromosomes activates Sxl in the germline, but the mechanism for “counting” X chromosomes in the germline is thought to be different from the soma. Here we have explored this mechanism at both cis- and trans- levels. Our data support the model that the Sxl “establishment” promoter (SxlPE) is activated in a female-specific manner in the germline, as in the soma, but that the timing of SxlPE activation, and the DNA elements that regulate SxlPE, are different in the germline. Nevertheless, we find that the X chromosome gene sisterless A (sisA), which helps activate Sxl in the soma, is also essential for Sxl activation in the germline. Loss of sisA leads causes of Sxl expression in the germline, and to ovarian tumors and germline loss. These defects can be rescued by Sxl expression, demonstrating that sisA lies upstream of Sxl in germline sex determination. We conclude that sisA acts as an X chromosome counting element in both the soma and the germline, but that additional factors regulating female-specific expression of Sxl in the germline remain to be discovered.
Author summary
The production of sperm and eggs requires proper sexual identity to be established in both somatic cells and the germ cells, which ultimately produce the gametes. While somatic sex determination has been well studied in a number of organisms, how germ cells establish their sexual identity is much less well understood. In Drosophila, the RNA binding protein Sex lethal (Sxl) is essential for female sexual identity in both the soma and the germline, but its regulation in the germline is thought to be different than in the soma. Here we explore how Sxl is activated in the germline. We find that the germline uses a different set of DNA elements to control activation of the key sex-specific Sxl promoter. Nonetheless, one of the activators of Sxl in the soma, the transcription factor Sisterless A (SisA), also acts to activate Sxl in the germline. Our data indicate that, while SisA acts as a common activator in both the soma and germline, additional, germline-specific Sxl activators remain to be discovered.
Introduction
Sex determination influences the development of many different tissues, but is particularly important in the germline to control the production of either sperm or eggs. While in some species somatic sex is sufficient to determine germline sex via inductive signaling, in flies and humans, the sex chromosome karyotype also plays a role intrinsically in the germline. Since a failure to match germline and somatic sex leads to defects in gametogenesis, understanding how intrinsic sex determination is regulated in the germline is important for our understanding of gonad development, reproductive biology, and human health.
Sex determination in Drosophila is under the control of the switch gene Sex lethal (Sxl) which is activated in females by the presence of two X chromosomes [1–11]. Sxl is both necessary and sufficient for the female sexual identity of somatic cells [3,6,12–19]. Expression of Sxl in the soma is regulated by two different promoters [20,21]. The Sxl “establishment promoter” (SxlPE) is sex-specific and is activated by the presence of two X chromosomes [2]. Default splicing of the SxlPE transcript produces a pulse of ‘Early’ Sxl protein starting at embryonic nuclear cycle 12. Later, at nuclear cycle 14, dosage compensation equalizes X chromosome expression in males and females and so X chromosomes can no longer be counted. Thus, there is a transition to the “maintenance promoter,” SxlPM, which is activated in both sexes [20,21]. The transcript from SxlPM requires alternative splicing regulated by the early Sxl protein (only made in females) in order to produce more ‘Late’ Sxl protein, creating an auto-regulatory loop that maintains Sxl expression in females [3]. The initial, female-specific activation of SxlPE in the soma is in direct response to a diploid dose of X-linked Signaling Elements (XSEs) – genes on the X chromosome that activate SxlPE when present in two copies, but not one. The somatic XSEs include the transcription factors Sisterless A (SisA), Sisterless B (SisB) and Runt, along with the JAK/STAT ligand Unpaired (Upd) [9,22–30].
Unlike primary sex determination in the soma, which is autonomously dependent on a cell’s sex chromosome genotype, sex determination in the Drosophila germline is regulated by non-autonomous signals from the soma in addition to the sex chromosome make-up of the germline (reviewed in [31]). In order for proper gametogenesis to occur, the “sex” of the germ cells has to match the sex of the soma. For example, XX germ cells developing in a male soma result in a testis with a severely atrophic germline [32–45], while XY germ cells developing in a female soma produce an ovary with a tumorous germline (“ovarian tumor”) and a complete failure to produce eggs [32,46–48]. Similar to the soma, Sxl is both necessary and sufficient for determining the autonomous component of germline sexual identity. Sxl loss-of-function in the germline results in ovarian germline tumors, similar to male germ cells developing in a female soma [32,33,37,39,42,46,48,49]. Further, XY (male) germ cells expressing Sxl are able to produce eggs when transplanted into XX (female) somatic gonads, demonstrating that Sxl is sufficient for female sexual identity in the germline as long as the surrounding soma is female [50]. Female-specific Sxl expression in the germline is dependent on the X chromosome dose: XX germ cells express Sxl while XY germ cells do not [32,39]. Evidence also indicates that SxlPE is important for female-specific expression of Sxl in the germline. Transcript data from early germ cells show a Sxl RNA that matches the SxlPE transcript in the soma [50], and data indicate that the positive auto-regulation of Sxl expression that occurs in the soma also acts in the germline [49]. However, the mechanism for activating Sxl expression in the germline appears to be different from the soma [33,46,50–54]. When germ cells simultaneously heterozygous for Sxl and the somatic XSEs sisA, sisB, and runt, were transplanted into wildtype female embryos, they were still able to undergo oogenesis indicating that the germ cells were not masculinized, even though this same genotype is sufficient to masculinize somatic cells [52]. Further, germ cells homozygous for mutations in sisB, or maternally deficient for the sisB co-factor daughterless, can also make normal eggs [33,46]. Thus, different XSEs, or at least a different combination of XSEs, is required to activate Sxl expression in the germline.
Here we utilize a combination of approaches to investigate the female-specific activation of Sxl in the germline. Using RNA FISH and CRISPR-tagging of specific Sxl isoforms, we show that the timing of SxlPE activation relative to SxlPM is different in the germline than in the soma. Using promoter/enhancer reporter constructs, we demonstrate that the regulatory sequences required for female-specific activation of SxlPE in the germline are different from those required in the soma. Lastly, we show that the somatic XSE, sisA, is also required for female-specific expression of Sxl in the germline. Together our data support a model in which sisA acts as a germline XSE, but that additional factors, different from the somatic XSEs, are also important for Sxl activation in the female germline.
Results
Analysis of Sex lethal transcription in PGCs
Since the activation of SxlPE appears to be the key sex-specific decision in germline sex determination, similar to what is observed in the soma, we first decided to examine when and how SxlPE is activated in the germline. We hypothesized that, as in the soma, SxlPE would be activated prior to SxlPM. To study this, we performed RNA fluorescent in situ hybridization (FISH) using oligopaints to examine nascent RNA (nRNA) being transcribed from SxlPE and SxlPM at the Sxl locus. As SxlPM is upstream of SxlPE, we generated one set of probes that exclusively targets SxlPM-derived transcripts (PM-probe, Cy5, Green), and another that targets the common transcript from both SxlPE and SxlPM (Com-probe, Cy3, Red) (Fig. S1A). Thus, if SxlPE alone is active, we should observe signal from the Com-probe (Red) but not the PM-probe (Green), while if SxlPM is active we should observe signal from both probes (and cannot make a conclusion about SxlPE).
In the soma, female-specific activation of SxlPE occurs prior to expression of SxlPM. Consistent with this, in the female soma we initially observed RNA FISH signals as fluorescent nuclear foci with the Com-probe but not the PM-probe, indicating that only SxlPE was activated (Fig. S1C-S1C’’). Subsequently, we observed signals from both probes, indicating that SxlPM had been activated (Data not shown). Conversely, in the male soma, we always observed a signal from both the PM- and Com-probes simultaneously (Fig. S1B-S1B’’), indicating that that SxlPE is not expressed (or does not precede SxlPM activation). We observed two Sxl foci per nucleus in females and only one focus in males, which was expected as Sxl is an X chromosome locus. These data are consistent with the previous understanding of Sxl activation in the soma - that SxlPE is on initially only in females followed by SxlPM activation in both sexes.
In contrast, in both female and male germ cells, we only observed RNA FISH signals from both the PM-probes and the Com-probes simultaneously, and never from the Com-probe alone (Fig. 1A-1B’’). This indicates that SxlPE is not activated before SxlPM in the germline. (Note: we only observed a single focus in both XX and XY germ cells suggesting that the X chromosomes may be paired in the germline.) The simultaneous expression of PM-probes and Com-probes occurred in a subset of female germ cells at stage 5 (11.1%), and increased to 100 % of germ cells by stage 11 (Fig. 1C). Similarly, both probe sets were expressed simultaneously in a subset of male germ cells, starting at stage 5 (7%) and increased to 100% of germ cells by stage 11 (Fig. 1D). The gradual increase in the number of cells expressing SxlPM transcripts is similar to what is observed in the soma [21]. We conclude that, unlike in female somatic cells, SxlPE does not precede SxlPM activity in the germline, but instead is likely activated either simultaneously with, or later than, SxlPM. This is still consistent with a model where ‘Early Sxl’ protein produced from default splicing of the SxlPE-derived transcript is required for productive RNA splicing of the SxlPM-derived transcript, and thus for the subsequent production of ‘Late Sxl’ protein; even if SxlPM is active prior to SxlPE activation, these transcripts would not be able to produce Sxl protein until SxlPE is activated.
SxlPE activity in the germline requires different cis-regulatory elements than in the soma
We next wanted to identify if and when SxlPE is activated in the germline, and also to compare the cis-regulatory logic of SxlPE activation in the germline to that in the soma. To do this we used transcriptional reporter constructs specific for SxlPE. We first tested a SxlPE-EGFP transcriptional reporter that contains the 1.5kb somatic enhancer immediately upstream of SxlPE (SxlPE-1.5kb), which had previously been reported to recapitulate both somatic and germline SxlPE activity [50,74]. Contrary to previous reports, we did not observe EGFP expression from SxlPE-1.5kb in developing PGCs, even though SxlPE-1.5kb showed sex-specific EGFP expression in somatic cells (Data not shown). Thus, we conclude that the cis-regulatory region sufficient for sex-specific expression of SxlPE in the soma is insufficient for SxlPE expression in the germline. We next extended the SxlPE transcriptional reporter to contain the entire 5.2kb genomic sequence upstream of SxlPE (but excluding SxlPM) (SxlPE-5.2kb, Fig. S2A). Once again, even though we observed female-specific nuclear EGFP expression from the SxlPE-5.2kb reporter in somatic cells, we found no evidence of EGFP expression in the germline of either male or female embryos up to stage 15 (Fig. S2B, S2F). To accommodate the possibility of a delay between the activation of the SxlPE-5.2kb transgene and our ability to detect EGFP expression (as is observed in the somatic cells), we also characterized its expression at later stages of development. However, we failed to observe any germline EGFP expression at the first, second, or third larval instar (L1, L2, or L3 respectively) stages (Fig. S2B-S2I). Thus, SxlPE-5.2kb also does not contain the cis-regulatory elements sufficient for SxlPE activity in the germline. Interestingly, SxlPE-5.2kb exhibited sex-specific EGFP expression generally in the soma until L1, and this expression persisted in the somatic gonad until L2 (Fig. S2F-S2H). This was unexpected as SxlPE has been reported to shut off in the soma following activation of dosage compensation in the early embryo. Our observations could be due to the stability of the EGFP reporter used or may point at a possible mechanism that keeps SxlPE active even after X chromosomes can no longer be counted.
We next generated a SxlPE-EGFP reporter that includes the 5.2kb sequence discussed above, as well as an additional 5kb downstream of SxlPE up to the start of Exon 4 (Fig. S2A) (SxlPE-10.2kb). Flies carrying this transgene showed sex-specific nuclear EGFP expression in the soma by stage 15 of embryogenesis (Fig. S2J-S2K). Excitingly, we also observed sex-specific nuclear EGFP expression in the female germline during the first larval instar (L1) stage (GFP immunostaining Fig. 2A-2B’, endogenous GFP and quantification, Fig. 6F,G,I). Together, these data suggest that SxlPE activity in the germline is sex-specific as it is in the soma, and that it requires additional cis-regulatory elements that lie downstream of SxlPE. Further, germline SxlPE activation appears to occur later in development than previously thought (L1), although it is possible that there is a delay between when SxlPE is actually activated and when we are first able to detect EGFP expression.
Late Sex lethal protein is expressed only after SxlPE is activated in the germline
Since SxlPE is sex-specifically activated in the germline, we hypothesized that, like in the soma, Early Sxl protein derived from SxlPE would be required to produce Late Sxl protein derived from SxlPM. To study expression of the Early and Late Sxl proteins in the germline, we used CRISPR-Cas9 genome editing to separately tag Sxl’s Early and Late protein products. We inserted a 2x human influenza hemagglutinin (2xHA) epitope at the N-terminus of the coding sequence of exon E1, which is specific to the mRNA produced by SxlPE, to generate a HA:SxlE1 ‘Early (E) Sxl’ tag (Fig. S3A). Separately, we inserted a 3xFLAG epitope at the N-terminus of the coding sequence of exon L2, which is specific to the mRNA produced by SxlPM, to generate a FLAG:SxlL2 ‘Late (L) Sxl’ tag (Fig. S3A) (Note: these tags are in separate stocks). Both of these tagged alleles of Sxl appear wildtype for Sxl function as homozygous females are viable and fertile.
Using anti-HA antibody, we were able to observe female-specific Sxl Early protein in somatic cells beginning at stage 9, and this expression was primarily nuclear, consistent with Sxl’s role in alternative splicing. Early Sxl expression in the soma persisted until the L1 stage (Fig. S3B-S3G), and no expression was observed after this time. In addition to the somatic gonadal tissue, we confirmed Early Sxl expression in the L1 stage in another somatic tissue, the developing gut (Fig. S3H-S3I). The continued presence of Early Sxl protein at the L1 stage is consistent with what we observed with the SxlPE-10.2kb reporter and suggests that Early Sxl protein is expressed, or at the very least maintained, much later in development than previously thought, and well after dosage compensation has been initiated. To our surprise, we were unable to detect anti-HA immunolabeling in the developing germline at any stage examined (embryo, and L1-L2). A low level of signal was observed in the cytoplasm of the germline (arrows in Fig. 2C-D’), but this was not clearly different between males and females, and may represent background staining. Detection of Early Sxl in the soma required tyramide amplification of the antibody signal, and so Early Sxl protein may be expressed at very low levels, which was undetectable in the germline even after tyramide amplification. Regardless, this prevents us from making a conclusion about Early Sxl protein expression in the germline.
Using the FLAG-tagged Late Sxl allele (FLAG:SxlL2), we observed anti-FLAG immunolabeling in female germ cells starting at the second larval instar (L2) stage (Fig. 2E-2F’). Female germ cells remained positive for anti-FLAG labeling through to adulthood and expression patterns were consistent with anti-Sxl antibody labeling (Fig. S3J-S3M’), which can be used to visualize Sxl in the germline starting at the third larval instar stage and onwards (Data not shown). No anti-FLAG immunoreactivity was observed in male germ cells. Though our RNA FISH analysis indicates that expression of SxlPM begins in the germline at stage 5, and is active in most germ cells by stage 10 (Fig. 1C), the SxlPE reporter (SxlPE-10.2kb) is only detectable in L1 germ cells (Fig. 2B), indicating that SxlPE expression begins much later in the germ cells than the soma. This is consistent with our ability to detect Late Sxl protein only in L2 germ cells; if Early Sxl protein is required to splice SxlPM transcripts to produce Late Sxl protein, then we would expect to detect Late Sxl protein only after SxlPE is active in the germline.
sisterless A loss-of-function in the female germline results in ovarian tumors and germline loss
We next wanted to investigate the trans-acting factors that regulate Sxl in the germline. As discussed above, previous work indicates that germ cells use a different combination of XSEs than the soma. However, it remains possible that some of the individual somatic XSEs contribute to the X chromosome counting mechanism in the germline, perhaps along with unknown germline-specific XSEs. To investigate the potential role of the individual somatic XSEs in germline Sxl activation, we knocked down their expression specifically in the germline using RNAi (nanos-GAL4; UAS-sisA RNAi, UAS-sisB RNAi, UAS-sisC RNAi, or UAS-runt RNAi). We then immunolabeled adult gonads to look for phenotypes resembling Sxl loss of function, such as the formation of ovarian tumors. We found that RNAi knockdown of sisB, sisC, or runt in the germline did not result in any aberrant ovarian phenotypes (Fig. S4A-S4D’). Germ cells in these ovaries appear to differentiate properly as indicated by the progressive increase in the size of their nuclei, characteristic of polyploid nurse cells during oogenesis.
In contrast, sisA germline RNAi (sisA RNAi 1) resulted in ovaries that exhibited ovarian germ cell tumors (Fig. 3C-3C’, S5C, S5D) similar to those observed in Sxl LOF (Fig. 3B-3B’, S5C) or when XY germ cells develop in an XX soma [32,46–48]. While a high percentage of ovaries exhibited germline tumors (Fig. S5C), not all ovarioles in each ovary were tumorous. To confirm that this phenotype was not due to off-target effects sometimes seen with RNAi, we generated an additional UAS-sisA RNAi transgene (sisA RNAi 2) (Fig. S5D). Expression of this RNAi line in the germline also produced ovarian germ cell tumors (Fig. S5B-S5B’, S5D), but these tumors were less severe and occurred with lower frequency (Fig. S5C). An additional UAS-sisA RNAi transgene did not result in any observable phenotypes (Fig. S5D, Data not shown). When 2 copies of nanos-GAL4 were used to drive increased expression of UAS-sisA RNAi 1, we observed severe germline loss resulting in germ cell-less ovaries (Fig. 3D-3D’, S5C) and subsequent infertility. This is similar to what is observed in strong alleles of other sex determination genes such as ovo and ovarian tumor (otu), as well as when XY germ cells develop in an XX soma [40,75–79].
Next we wanted to investigate the null phenotype of sisA in the germline. Since homozygous sisA mutant females and hemizygous sisA mutant males are embryonic lethal [57], we generated sisA germline-specific loss of function mutations using tissue-specific CRISPR-Cas9 genome editing (“G0 CRISPR”, [69–71]). Male flies that ubiquitously express either 2 or 4 different guide RNAs (gRNAs) targeting the sisA gene region (Fig. S5D) were mated to females expressing Cas9 in the germline under the nanos promoter. F1 females from this cross also exhibited ovarian tumors as well as germ cell-less ovaries (Fig. 3E-3G’, S5C), similar to what was observed using RNAi (Fig. 3C-D’). Taken together, these data indicate that sisA is required for female germline differentiation and maintenance. Additionally, germline sisA loss of function is similar to Sxl loss of function, suggesting sisA is a candidate XSE for SxlPE activation in the germline.
sisterless A expression in the embryonic germline precedes Sex lethal activation
In order for sisA to act as an XSE for activating Sxl in the germline, it should be expressed zygotically in germ cells prior to sex-specific activation of SxlPE in the germline. It has been reported that sisA mRNA is excluded from the nuclei that bud off to form pole cells [28]. However, reported microarray expression data demonstrates that sisA transcripts are enriched in PGCs at the 1-to-3 hour time point [80]. To observe zygotic transcription of sisA in the germline we utilized RNA FISH against nRNA being transcribed from the sisA locus. As a validation of our approach, we observed signals from sisA RNA FISH probes as fluorescent nuclear foci in somatic nuclei starting at nuclear cycle 8 (Fig. S6A-S6A’) and also at later stages in yolk cell nuclei, consistent with previous reports (Fig. S6B-S6B’) [28,57]. Interestingly, we were able to detect sisA RNA FISH signals in PGCs at stages 3-6 (3-5 shown in Fig. 4A-4C’). At later stages, the strong expression levels of sisA RNA in yolk cell nuclei made it difficult to distinguish fluorescent nuclear foci in neighboring PGCs. This suggests that sisA is indeed zygotically transcribed in the early embryonic germline. We quantified this expression in sexed embryos and found that sisA is expressed in the PGCs of both sexes (Fig. 4D).
We next wanted to observe SisA protein in the germline. Since no antibodies against SisA have been generated to date, we used CRISPR-Cas9 genome editing to insert a super-folder Green Fluorescent Protein (sfGFP) epitope tag at the N-terminus of the SisA protein (sfGFP:SisA) (Fig. S6C). Importantly, the epitope tag does not appear to affect SisA function as homozygous females and hemizygous males were viable and had no observable gonad phenotypes (Fig. S6D-S6E’). Furthermore, anti-GFP immunolabeling is restricted to the nucleus, which is consistent with the predicted function of SisA as a bZIP transcription factor. In somatic cells, we observed expression of sfGFP:SisA in pre-blastoderm syncytial nuclei (Fig. S6F-S6F’) and in yolk cell nuclei (Fig. S6G-S6G’), consistent with somatic sisA RNA expression.
Excitingly, similar to what we observed using RNA FISH, sfGFP:SisA expression was also observed in PGCs at stage 3, 4, and 5 (Fig. 4E-4G’). Additionally, this expression persisted in PGCs until stage 10 (Fig. S6H-S6H’). This late expression of SisA is consistent with its potential role in acting as an XSE for SxlPE activation in the germline. Like with sisA RNA, we were only able to detect sfGFP:SisA in a subset of PGCs in any one embryo. sfGFP:SisA expression was not observed later in first, second, or third larval instar stages, in either the germline or the soma (Data not shown). We conclude that SisA is expressed in the embryonic germline and this expression precedes the activation of SxlPE, which is consistent with a role as a germline XSE for the activation of Sxl.
sisterless A is required for expression of Sxl in the female germline
If sisA acts as an activator of Sxl in the female germline, we would expect a loss of Sxl expression upon loss of sisA function. Strikingly, using anti-Sxl immunolabeling, we observed that the germ cell tumors in sisA RNAi ovaries had a strong reduction in Sxl expression when compared with controls (Fig. 5A-5C’, S7A-S7B’). Tumorous germ cells that were mutant for sisA using G0 CRISPR also had lower levels of Sxl expression (Fig. S7C-S7D’). Somatic Sxl expression remained unaffected in both cases. To examine Sxl expression specifically in those germ cells that express highest levels of Sxl (i.e. the early, undifferentiated germline), we performed sisA RNAi in the germline of bag of marbles (bam) mutants, which are enriched for germ cells robustly expressing Sxl (Fig 5D-5D’). We found that knocking down sisA using RNAi led to a dramatic reduction of Sxl antibody labeling in the germ cells of bam mutant ovaries (Fig. 5E-5E’), demonstrating that knocking down sisA leads to a loss of Sxl expression in the germline. To test whether sisA acts as a regulator of the Sxl PE promoter, we examined the effects of knocking down sisA function in the germline (nos>sisARNAi) on the expression of the Sxl PE-GFP reporter (SxlPE-10.2kb). Indeed, loss of sisA caused a significant reduction of germline GFP expression from the Sxl PE-GFP reporter, but did not reduce expression down to levels observed in males (Fig. 6F-I). This is consistent with SisA being one of, but not the only, regulator of Sxl PE expression in the germline.
These data show that sisA is necessary for Sxl expression in the germline, and we next wanted to test whether ectopic expression of sisA would be sufficient to promote Sxl expression. Expression of UAS-Sxl in the male germline leads to inhibition of the JAK/STAT pathway and loss of germline stem cells ([86] and Fig. 6B-6B’). Expression of UAS-sisA did not lead to detectible expression of Sxl in the testis (Fig. 6C-6C’). This indicates that the overexpression of sisA alone is not sufficient to activate Sxl in the germline and that, like in the soma, additional XSEs are required to activate SxlPE in the germline.
If sisA is required for Sxl expression, and this is its primary role in the germline, then expression of Sxl should bypass the need for sisA in the germline and rescue the germline defects observed with sisA LOF. To test this we expressed Sxl under the control of a female germline-specific promoter (otu) in a sisA LOF background. We used the strongest sisA loss of function condition (2x nos-GAL4 > sisA RNAi 1) which results in severe germline loss, tumor formation, and infertility (Fig. 6D-6D’, S7E). Excitingly, 50% of sisA RNAi + otu > Sxl ovaries exhibited a wildtype germline morphology (Fig. 6E-6E’, S7E) and were fertile, suggesting Sxl was able to provide a robust rescue. Another 20% of these ovaries showed a partial rescue of the germline, but contained pervasive germline tumors (Fig. S7E). Taken together, these data indicate that sisA is necessary, but not sufficient, for Sxl expression in the female germline and that activation of Sxl is the primary germline role for sisA.
Discussion
The sex-specific activation of Sxl in the germline
Previous evidence indicated that sex-specific activation of SxlPE is important for Sxl expression in germline, as it is in the soma. Transcripts from the SxlPE promoter were detected in early germ cells [50] and positive autoregulation of Sxl was observed in the germline [49], suggesting that the same mechanism of Early Sxl protein from SxlPE being required for splicing of SxlPM transcripts occurs in germ cells and soma. Our data also support this model. Using a SxlPE transcriptional reporter, we found that SxlPE is active in female germ cells by L1 (Fig. 2B-2B’). Further, while we observed activation the SxlPM promoter in embryonic germ cells (Fig. 1B-1B’’, 1C), we only observed expression of Late Sxl protein in the germ cells at L2 (Fig. 2F-2F’), after we observed activation of SxlPE. These data are consistent with SxlPM transcripts being unable to produce Sxl protein until after sex-specific activation of SxlPE. One difference between the germline and the soma appears to be the timing of SxlPE activation relative to that of SxlPM. In the soma, SxlPE is activated prior to SxlPM, while in the germline, there is no time at which we detect SxlPE alone by RNA FISH (Fig. 1C), and SxlPE is activated later in development, as judged by the SxlPE transcriptional reporter. This is still consistent with the autoregulatory model for Sxl activation, but suggests that the control of SxlPE activation is different in the germline than the soma.
We also found that the cis-regulatory elements that regulate female-specific SxlPE expression in the germ cells are different from those in the soma. Previously, it was reported that a 1.5kb genomic fragment just upstream of SxlPE was sufficient to regulate female-specific expression in both the soma [20,74] and the germline [50]. However, we were unable to observe germline expression from this SxlPE reporter or even a larger SxlPE reporter containing more of the upstream DNA (Fig. S2F-I). We only observed activation of SxlPE when we included an even larger genomic region, including sequences both upstream and downstream of SxlPE (Fig. S2A, 2B-2B’). These data do not exclude sequences within the 1.5 kb region upstream of SxlPE being important for its activation in the germline, but indicate that these sequences are not sufficient for activation, as they are in the soma. Similarly, sequences outside this 1.5 kb region may also augment expression in the soma. The fact that the cis-regulatory logic of SxlPE is different in the germline than the soma is consistent with previous work demonstrating that the combination of trans-acting factors, the XSEs, that activate SxlPE in the germline are different from the soma [33,52,81]. Future work will involve identifying the specific enhancers that regulate SxlPE expression in the germline, which could in turn help identify additional germline trans-regulators of Sxl. In summary, we have shown that the cis-regulatory logic of Sxl activation is different between the germline and the soma, however, the key step in determining the sex of both these cell types is the female-specific activation of SxlPE.
Finally, a surprising observation from our study was the presence of sex-specific SxlPE transcriptional reporter activity, as well as Early Sxl protein, in somatic cells as late as the L1 larval stage (Fig. S2F-S2G, S3E-S3F, S3I). In order for XSEs to act, they need to be expressed at higher levels in XX cells than XY cells. Once X chromosome dosage compensation is initiated at the late blastoderm stage, there should no longer be a difference in XSE expression between XX and XY cells. This is the entire logic behind the autoregulatory model for Sxl expression; SxlPE is activated by XSEs only in females, and is then turned off prior to the time when dosage compensation equalizes XSE expression. Why then is Early Sxl protein and SxlPE activity detected at the L1 stage? One possibility is that both the GFP reporter and the Early Sxl protein are stable and perdure to the L1 stage. An alternative possibility, however, is that some other mechanism, such as feedback regulation within the sex determination system, maintains SxlPE activity much longer than previously thought. Interestingly, Sxl contains a highly-conserved binding site for the Doublesex transcription factor which could facilitate such an XSE-independent feedback regulation of SxlPE [82].
sisterless A as an activator of Sxl
While the combination of XSEs at work in the germline may be different than in the soma, our data also indicate that at least one XSE, SisA, is shared between them. Loss of sisA in the germline leads to the formation of germline tumors (and germline loss) (Fig. 3C-3D’, 3F-3G’, S5B-S5C) as well as a loss of Sxl expression in the female germline (Fig. 5C-5E’, S7B-S7D’). Additionally, we report that sisA is expressed in the germline prior to SxlPE (Fig. 4) and that the expression of Sxl is able to rescue the sisA loss-of-function phenotype in the germline (Fig. 6D-6E’, S7E), indicating that sisA lies upstream of Sxl in the germline sex determination cascade. Taken together, these data indicate that SisA is a germline activator of Sxl and may act as a germline XSE.
It remains to be seen whether SisA acts directly on Sxl in the germline or acts on other Sxl regulators, such as ovo or otu. Current thinking is that SisA is a transcriptional activator for SxlPE in the soma [9,28], and the simplest model would be that it acts similarly in the germline. However, the gap in timing we observe between sisA expression in the germline (Stage 3-10, Fig. 4E-4G’, S6H-S6H’) and our ability to detect SxlPE activation (L1 larvae, Fig. 2B-2B’) leaves open the possibility that there are intermediate steps between sisA and SxlPE. SisA is a somewhat unusual leucine zipper protein and its DNA binding activity and specificity have never been determined. Thus, there is no evidence that it regulates SxlPE directly in the soma or germline. The transient nature of SisA expression during embryogenesis makes it difficult to conduct chromatin immunoprecipitation (ChIP) experiments under native conditions. Instead, we expressed GFP:SisA in Drosophila S2 tissue culture cells for ChIP analysis, but this experiment did not identify SisA genomic binding sites. Lastly, yeast two-hybrid analysis has indicated that SisA might physically interact with the basic-helix-loop-helix protein Daughterless (Da) [83], but co-expressing GFP:SisA and Da in S2 cells also did not result in an identifiable ChIP signal. Thus, while our data clearly indicate that sisA is required for activation of Sxl in the germline, and that this is its primary role in the germline, evidence that SisA is a direct, transcriptional activator of SxlPE will require identification and study of a SisA-binding enhancer within the Sxl locus.
It is puzzling that Sxl is a critical regulator of both germline and somatic sex determination, and is regulated by a two-X dose in both cell types, yet its mechanism of activation and its role in these cell types is so different. It is somewhat simplifying that at least one factor, sisA, is important for activation of Sxl in both cell types. Yet evidence still suggests that other aspects of Sxl activation in these cells will be different. Further, the most significant Sxl targets in the soma, transformer for sexual identity and msl-2 for dosage compensation, do not play a role in the germline [45,47]. Instead, distinct factors such as Tdrd5l and the Jak/Stat pathway, which controls Phf7 expression, are regulated by Sxl in the germline [67,84-86]. It will be interesting to examine how these independent roles for Sxl came to be, and how conserved each role is in different species.
Materials and methods
Fly stocks
The following fly stocks were used: SxlPE-EGFP (BDSC# 24105), SxlPE-EGFP (BDSC# 32565), vas-Cas9 (BDSC# 51324), vas-Cas9 (BDSC# 56552), alphatub-piggyBac (BDSC# 32070), nos-Gal4 ([55]), nos-Cas9 (BDSC# 54591), nos-Cas9 (BDSC# 78782), UAS-sisA-RNAi 1 (TRiP.HMC03864, BDSC# 55181), UAS-Sxl-RNAi (TRiP.HMS00609, BDSC# 34393), UAS-sisB-RNAi (TRiP.GL01130, BDSC# 41594), UAS-sisC-RNAi (TRiP.HMS00545, BDSC# 33680), UAS-run-RNAi (TRiP.HMS01186, BDSC# 34707), UAS-mCherry-RNAi control (BDSC# 35785), bam[1] (Gift from A. Spradling, [56]), bam[delta86] (BDSC# 5427), sisA[5] (Gift from J. Erickson, [57]), UAS-Sxl (BDSC# 58484), otu-GFP.K10 (BDSC# 29727), otu-GFP.SV40 (BDSC# 29729), otu-Sxl (BDSC# 58491). Oregon R flies were used as wildtype flies, w[1118];Sco/CyO;MKRS/TM6B,Tb,Hu double balancers (Gift from X. Chen) were used for 2nd and 3rd chromosome balancing. FM7c was used for X chromosome balancing.
Antibody staining and Tyramide Signal Amplification (TSA)
Adult testes were dissected in 1X PBS and fixed at room temperature for 20 minutes in 4.625% formaldehyde in PBS containing 0.1% Triton X-100 (PBTx). Adult ovaries, and larval gonads (first, second, and third instar) were dissected in 1X PBS and fixed at room temperature for 20 minutes in 5.25% formaldehyde in PBTx. Blocking and immunostaining was performed as previously described [58], and samples were mounted in 2.5% DABCO. Embryo collection, fixing, blocking and staining was performed as described previously [59]. All images were taken with a Zeiss LSM 700 confocal microscope. The following primary antibodies (sources) and concentrations were used: chicken anti-Vas 1:10,000 (K. Howard); rabbit anti-Vas 1:10,000 (R. Lehmann); mouse anti-Sxl 3:100 (M18, DSHB); mouse anti-Fas3 1:50 (7G10, DSHB); rat anti-NCad 3:100 (DN-Ex #8, DSHB); mouse anti-Lamin B 1:100 (ADL67.10, DSHB); mouse anti-Pros 1:10 (MR1A, DSHB); guinea pig anti-TJ 1:1,000 (J. Jemc); rabbit anti-GFP 1:1,000 (ab290, Abcam); mouse anti-FLAG 1:50 (F3165, Millipore Sigma); rat anti-HA 1:100 (ROAHAHA Clone 3F10, Roche); rabbit anti-HA 1:800 (C29F4 #3724, CST). DSHB: Developmental Studies Hybridoma Bank; CST: Cell Signaling Technologies. Secondary antibodies were used at 1:500 (Alexa Fluor, Host:Goat, Invitrogen).
TSA was performed using the manufacturer’s protocol (Tyramide SuperBoost Kits with Alexa Fluor Tyramides, Invitrogen). Larval gonads and embryos were fixed as previously described (See above). Samples were blocked in 10% Goat Serum for 1 hour at room temperature followed by incubation with a single primary antibody (target of signal amplification) as previously described [58,59] overnight at 4°C with nutation. Primary antibody was washed off using 1X PBS and samples were incubated with poly-HRP-conjugated secondary antibody for 1 hour at room temperature. Secondary antibody was washed off using 1X PBS and samples were incubated in tyramide working solution for 10 minutes at room temperature before stopping the reaction. Samples were washed with 1X PBS and immunolabeling with TSA was multiplexed with other antibodies following the standard immunolabeling protocol.
Developmental Staging and Sexing
To obtain stage-specific embryos, embryos were pre-collected overnight on apple juice plates with fresh yeast paste and discarded to synchronize egg-laying. Embryos were then collected for 2 hours at 25°C after which the flies were removed and the embryos were aged as required. To obtain first (L1) and second (L2) instar larvae, embryos were aged 20 hours. The plates were cleared of any larvae that had hatched early. L1 larvae were collected after 4-8 hours. For L2 larvae, the L1 larvae were transferred to fresh plates with yeast paste and aged for an additional 24 hours. Third instar (L3) larvae were collected directly from vials. Sex of embryos was determined by counting the number of Sex lethal or sisterless A RNA FISH signals corresponding to number of X chromosomes or by somatic EGFP expression using the transcriptional SxlPE-EGFP reporters. Stage of the embryos was approximately determined by the position of the primordial germ cells (PGCs). Sex of the first and second larval instar gonads was determined by looking for presence or absence of a hub using anti-FasIII or anti-NCad immunolabeling. Sex of the third larval instar gonads was determined by morphology.
Oligopaints Probe Design and Synthesis
Oligopaint probes for RNA FISH were designed using Oligopaints [60–62] by Dr. Kayla Viets from Dr. Robert Johnston’s laboratory, Department of Biology at Johns Hopkins University. Gene target sequences (Sex lethal and sisterless A) were run through an open-source bioinformatics pipeline made available by the Wu lab at Harvard Medical School (http://genetics.med.harvard.edu/oligopaints/) to identify sets of 50-bp (Sex lethal) or 30-bp (sisterless A) optimized probe sequences (libraries) to tile the nascent RNA transcript [63]. Library of Cy3-conjugated oligos against sisterless A RNA was ordered individually. Library of oligos against Sex lethal and its sub-libraries (PM vs. PE+PM probes) were ordered as part of a 90k oligopool including oligos for other gene targets (Gift from R. Johnston, Johns Hopkins University). To isolate target-specific probes (Sex lethal), five 19-bp barcording primers, target F and R; universal (univ) F and R; and sublibrary (sub) F were appended to the 5’ and 3’ ends of each probe.
PCR using target F and R primers allowed amplification of libraries of probes against target RNA from oligopool. PCR using sub F and target R primers allowed amplification of sub-libraries from whole target libraries. PCR using univ F and R primers allowed the conjugation of fluorophores (Cy3, Cy5), generation of single-stranded DNA (ssDNA) probes as well as addition of secondary sequences to allow amplification of RNA FISH signal using secondary probes (conjugated with fluorophores) that bind to primary probes. RNA FISH probes were generated as previously described [60–62]). Total number of probes per transcript are as follows: sisA RNA – 12; SxlPM RNA – 19; SxlPE+PM RNA – 36.
RNA fluorescent in situ hybridization (FISH) and immunofluorescence
RNA FISH was performed using modified versions of the protocols described previously [60–63]. Oregon R embryos were collected (after a pre-collection was discarded) on apple juice plates with fresh yeast paste for 2 hours at 25°C and aged as needed. Embryos were collected, rinsed with 1X PBTx, and dechorionated using 50% bleach for 90 seconds and washed with 1X PBTx. Fixing was performed in scintillation vials in 50μL 10X PBS, 100μL 0.25M EGTA, 125μL fresh 16% formaldehyde, 225μL Milli-Q H2O, 1μL NP-40 (Tergitol solution), and 500μL heptane. Embryos were shaken vigorously by hand for 1 minute, then fixed for 20 minutes at room temperature with gentle agitation. The aqueous phase was removed and the embryos were devitellinized in by adding 500μL methanol and vigorous shaking for 2 minutes. Devitellinized embryos were collected washed three times in methanol. Embryos were rehydrated via 5-minute serial single washes in 75%, 50%, and 25% methanol in 1X PBTx. This was followed by three 5-minute washes in 1X PBTx and two 15-minute washes in 1X PBTx with 0.2U/μL RNase inhibitor. Embryos were blocked for 1 hour at room temperature in blocking buffer (1X PBTx + Western Blocking Reagent, 1:1) with nutation. They were then incubated in primary antibody diluted in 1X PBTx with 3% normal goat serum and 0.2U/μL RNase inhibitor overnight at 4°C with nutation. Primary antibody was rinsed off followed by three 20-minute washes with 1X PBTx. Embryos were then incubated with secondary antibody diluted in 1X PBTx with 3% normal goat serum and 0.2U/μL RNase inhibitor for two hours at room temperature. Secondary antibody was rinsed off followed by two 20-minute washes in 1X PBTx and one 20-minute wash in 1X PBS. Embryos were then washed as follows: four 5-minute washes in 2X SSCT, one 10-minute wash in 20% formamide in 2X SSCT, one 10-minute wash in 50% formamide in 2X SSCT, one 4-hour wash in 50% formamide in 2X SSCT at 37°C with shaking. Embryos were incubated with primary probe at a concentration of ≥5 pmol fluorophore/μL in hybridization buffer (50% formamide in 2X SSCT with 10% dextran sulfate (w/v) + 0.2U/μL RNase inhibitor) for 16-20 hours at 37°C with shaking. Primary probe was rinsed and followed by a 1-hour wash in 50% formamide in 2X SSCT at 37°C with shaking. Secondary probes were hybridized for 1 hour at 37°C with shaking at a concentration of ≥5 pmol fluorophore/μL in 50% formamide in 2X SSCT and 0.2U/μL RNase inhibitor. This was followed by two 30-minute washes in 50% formamide in 2X SSCT at 37°C, one 10-minute wash in 20% formamide in 2X SSCT, two 10-minute washes in 2X SSCT at room temperature, and one 10-minute wash in 2X SSC at room temperature. Embryos were washed for 10 minutes in 1X PBTx containing DAPI and incubated at room temperature for 10 minutes in SlowFade Diamond with DAPI, followed by mounting. All images were taken with a Zeiss LSM 700 confocal microscope.
SxlPE reporter transgenes and constructs
SxlPE-10.2kb, a 10.2kb genomic sequence from a Sxl genomic clone BAC# CH321-74P19 (BACPAC Resources Center) cloned into the pJR16 vector in two steps (Gift from R. Johnston, Johns Hopkins University) using HiFi DNA Assembly (New England Biolabs). pJR16 uses an EGFP reporter with a nuclear localization sequence (nls). Fragment 1 extended from 116 bp downstream of SxlPE TSS to 5,116 bp upstream of SxlPE TSS and was assembled into pJR16 digested with AgeI and AscI to generate SxlPE-5.2kb. Fragment 2 extended from 117 bp downstream of SxlPE TSS to 5,108 bp downstream of SxlPE TSS and was assembled into Fragment 1 + pJR16 digested with AgeI.
Fragment 1 SxlPE-10.2kb Fw (5’-3’) – CCACCCCGGTGAACAGCTCCTCGCCCTTGCTCACCATGGTGGCGACCGGTAA TGGGATAATCACAAAGTT
Fragment 1 SxlPE-10.2kb Rv (5’-3’) – CATGCTGCAGCAGATCTGGTCTAGAGCCCGGGCGAATTCGCCGGCGCGCCGT AATTTTTCTTTGCTCCTCCTG
Fragment 2 SxlPE-10.2kb Fw (5’-3’) – CAGCTCCTCGCCCTTGCTCACCATGCTGTACGATGAATCGA
Fragment 2 SxlPE-10.2kb Rv (5’-3’) – GAAAAACGTAACTTTGTGATTATCCCATTATGGATTTCAATTTTGATAC
The primers ensured that EGFP remained in frame with Exon E1’s coding sequence (CDS). Constructs were injected into embryos and integrated via PhiC31 integrase-mediated transgenesis (done at BestGene Inc.) into the same genomic location at P{CaryP}attP40 on Chromosome II.
Generation of sisA RNAi 2 and 3
Transcript-specific knockdown was achieved using the VALIUM vector system (PMID: 21460824). A 21-nucleotide (nt) sequence without any off-targets greater than 16nt was selected based on the algorithm of [64]. The following sequences were selected:
sisA-RNAi 2 Sense – CGCCGACGAGGAGCAACGCUA
sisA-RNAi 2 Antisense – UAGCGUUGCUCCUCGUCGGCG
sisA-RNAi 3 Sense – CCGGUUCUGGUUCGGAUGUCA
sisA-RNAi 3 Antisense – UGACAUCCGAACCAGAACCGG
The top and bottom strands for the short hairpin were as follows:
sisA-RNAi 2 Top Strand – CTAGCAGTCGCCGACGAGGAGCAACGCTATAGTTATATTCAAGCATATAGCG TTGCTCCTCGTCGGCGGCG
sisA-RNAi 2 Bottom Strand – AATTCGCCGCCGACGAGGAGCAACGCTATATGCTTGAATATAACTATAGCGT TGCTCCTCGTCGGCGACTG
sisA-RNAi 3 Top Strand – CTAGCAGTCCGGTTCTGGTTCGGATGTCATAGTTATATTCAAGCATA-TGACATCCGAACCAGAACCGGGCG
sisA-RNAi 3 Bottom Strand – AATTCGCCCGGTTCTGGTTCGGATGTCATATGCTTGAATATAACTATGACATC CGAACCAGAACCGGACTG
The top and bottom strands were annealed and cloned into the VALIUM20 vector #1467 (Drosophila Genomics Resource Center, DGRC). Constructs were injected into embryos and integrated via PhiC31 integrase-mediated transgenesis (done at BestGene Inc.) into the same genomic location at P{CaryP}attP40 on Chromosome II. For RNAi-mediated germline-specific knockdown of sisA, sisB, sisC, runt, Sxl, Ovo, and mCherry, male flies carrying shRNA transgenes were mated with nanos-GAL4:VP16 virgin females [55] and the crosses were maintained at 29°C. The progeny were reared at 29°C until 3-5 days post-eclosion (unless otherwise specified).
CRISPR-tagging
sfGFP:sisA, 2xHA:SxlE1, and 3xFLAG:SxlL2 were generated using scarless genome editing described in [65] to generate N-terminal tags. Guide RNA (gRNA) sequences were selected using the FlyCRISPR algorithm (http://flycrispr.molbio.wisc.edu) [66], contain 20 nucleotides each and have no predicted off-targets. The following gRNAs were selected:
sfGFP:sisA gRNA Target (5’-3’) – GTCCAATGGCAAGCTACCTG
2xHA:SxlE1 gRNA Target (5’-3’) – GCCTCCTTCGATCTTCTACC
3xFLAG:SxlL2 gRNA Target (5’-3’) – GACTTGTTGTTGTAGCCATA
The guides were cloned into the pU6-2-BbsI-gRNA vector #1363 (DGRC) using the listed primers that were phosphorylated using T4 polynucleotide kinase (PNK).
sfGFP:sisA gRNA Target sense (5’-3’) – CTTCGTCGTTGGCCAATCCGGATGC
sfGFP:sisA gRNA Target antisense (5’-3’) – AAACGCATCCGGATTGGCCAACGAC
2xHA:SxlE1 gRNA Target sense (5’-3’) – CTTCGCCTCCTTCGATCTTCTACC
2xHA:SxlE1 gRNA Target antisense (5’-3’) – AAACGGTAGAAGATCGAAGGAGGC
3xFLAG:SxlL2 gRNA Target sense (5’-3’) – CTTCGACTTGTTGTTGTAGCCATA
3xFLAG:SxlL2 gRNA Target antisense (5’-3’) – AAACTATGGCTACAACAACAAGTC
Donor plasmids to facilitate homology dependent repair were generated. The coding sequences of the different tags are cloned from genomic DNA of vas-Cas9 expressing flies (BDSC# 51324, 56552) adjacent to a piggyBac transposon that contained a DsRed expression construct resulting in a selectable-tagging-cassette in vectors pHD-sfGFP-ScarlessDsRed #1365, pHD-2xHA-ScarlessDsRed #1366, pHD-3xFLAG-ScarlessDsRed #1367 (DGRC). Approximately 1kb 5’ and 3’ Homology arms were assembled using HiFi DNA Assembly (New England Biolabs) upstream and downstream of this cassette with the following primers (5’-3’):
sfGFP:sisA 5’ Arm Fw (5’-3’) – GAATTCGCCAAAGGGATTTC
sfGFP:sisA 5’ Arm Rv (5’-3’) – CCGGAACCTCCAGATCCACCGGTGATTTTTTTCGATGTGTG
sfGFP:sisA cassette Fw (5’-3’) – ACACATCGAAAAAAATCACCGGTGGATCTGGAGGTTCC
sfGFP:sisA cassette Rv (5’-3’) – AAGTAAAGATGACTCCGTTCGGAACCTCCTGAACCACC
sfGFP:sisA 3’ Arm Fw (5’-3’) – CTGGTGGTTCAGGAGGTTCCGAACGGAGTCATCTTTACTTGCC
sfGFP:sisA 3’ Arm Rv (5’-3’) – GGTACCGCATTGGCCCAATTC
2xHA:SxlE1 5’ Arm Fw (5’-3’) – GCAATCTGTGTTCTTGGTATTTTG
2xHA:SxlE1 5’ Arm Rv (5’-3’) – GGAACATCGTATGGGTACATAATGGGATAATCACAAAGTTAC
2xHA:SxlE1 cassette Fw (5’-3’) – AACTTTGTGATTATCCCATTATGTACCCATACGATGTTCC
2xHA:SxlE1 cassette Rv (5’-3’) – ACAGTATCAAAATTGAAATCGGAACCTCCTGAACCACC
2xHA:SxlE1 3’ Arm Fw (5’-3’) – CTGGTGGTTCAGGAGGTTCCGATTTCAATTTTGATACTGTGAC
2xHA:SxlE1 3’ Arm Rv (5’-3’) – CGATCGAAGGTGAGTTTC
3xFLAG:SxlL2 5’ Arm Fw (5’-3’) – CGACCATGTCGTCCTACTATAAC
3xFLAG:SxlL2 5’ Arm Rv (5’-3’) – TCATGGTCTTTGTAGTCCATATCCTGAGAGTTGGGAGTG
3xFLAG:SxlL2 cassette Fw (5’-3’) – ACACTCCCAACTCTCAGGATATGGACTACAAAGACCATGAC
3xFLAG:SxlL2 cassette Rv (5’-3’) – CCCGGATTATTGTTGCCGTAGGAACCTCCTGAACCACC
3xFLAG:SxlL2 3’ Arm Fw (5’-3’) – CTGGTGGTTCAGGAGGTTCCTACGGCAACAATAATCCG
3xFLAG:SxlL2 3’ Arm Rv (5’-3’) – GCTAATGAGGGGATTCCTATG
Assembled donor constructs were cloned into pCR2.1-TOPO (TOPO TA Cloning Kit, ThermoFisher). Site-specific mutagenesis was also used to mutate the PAM sites in each donor plasmid. For sfGFP:sisA, the included linker sequence between 3’ UTR and the start of sfGFP was removed using site-specific mutagenesis (QuikChange, Agilent). The gRNA expression plasmid and donor plasmid were injected by BestGene into embryos of Vas-Cas9 flies (#56552, #51324, BDSC). DsRed positive flies were then crossed to a piggyBac transposase expressing line (#32070, #32073, BDSC) to excise DsRed resulting in an in-frame fusion of the tags’ and the respective genes’ coding sequences. Successful generation of CRISPR-tags was confirmed by sequencing.
Generation of overexpression constructs
UAS-sfGFP, UAS-sisA, and UAS-da were generated by cloning the ORFs of sfGFP, sisA, and da-PD into pUASpB [67] using the listed primers. pUASpB is a modified version of pUASP [68] including an attB site for phiC31-mediated integration. UAS-sfGFP:sisA was generated by cloning a flexible linker sequence followed by the ORF of sisA (lacking a start codon) using the listed primers into UAS-sfGFP that had been linearized with SpeI. Constructs were assembled using HiFi DNA Assembly (New England Biolabs).
UAS-sfGFP Fw (5’-3’) – TACCCGCCCGGGGATCAGATCCGCGGCCGCATGGTGTCCAAGGGCGAG
UAS-sfGFP Rv (5’-3’) – GACTCTAGAGGATCCAGATCCACTAGTTCACTTGTACAGCTCATCCATGC
UAS-sfGFP:sisA Fw (5’-3’) –GCCGGCATCACCCTGGGCATGGATGAGCTGTACAAGATTAAGGCCGGCGGGT CG
UAS-sfGFP:sisA Rv (5’-3’) – ACGTTAACGTTCGAGGTCGACTCTAGAGGATCCAGATCCATCACTGCTCCATT TCCAGGC
UAS-sisA Fw (5’-3’) – CTGTTCATTGGTACCCGCCCGGGGATCAGATCCGCATGGAACGGAGTCATCTT TACTTGC
UAS-sisA Rv (5’-3’) – ACGTTAACGTTCGAGGTCGACTCTAGAGGATCCAGATCCATCACTGCTCCATT TCCAGGC
UAS-da-PD Fw (5’-3’) – AGGTCCTGTTCATTGGTACCCGCCCGGGGATCAGATCCGCATGGCGACCAGT GACGATG
UAS-da-PD Rv (5’-3’) – ACGTTAACGTTCGAGGTCGACTCTAGAGGATCCAGATCCATTAAAAGTGTTG TACATTTTGTAGGGG
Constructs flies were injected into embryos and integrated via PhiC31 integrase-mediated transgenesis (done at BestGene Inc.) into the same genomic location at P{CaryP}attP40 on Chromosome II. For germline-specific overexpression, male flies carrying UAS transgenes were mated with nanos-GAL4:VP16 virgin females [55] and the crosses were maintained at 29°C. The progeny were reared at 29°C until 3-5 days post-eclosion (unless otherwise specified).
G0 CRISPR (Tissue-specific CRISPR)
Guide RNA (gRNA) sequences were selected using the FlyCRISPR algorithm (http://flycrispr.molbio.wisc.edu) [66], contain 20 nucleotides each and have no predicted off-targets. For sisA, four different gRNA sequences were selected, one within the coding region (Target 4) and one within the 3’ UTR (Target 3), and two upstream of the gene locus (Targets 1 and 2). As a control, four different gRNA sequences were selected against GFP. The following gRNAs were selected:
sisA-G0 gRNA Target 1 (5’-3’) – TCGTTGGCCAATCCGGATGCAGG
sisA-G0 gRNA Target 2 (5’-3’) – CACTGAGTCTACCTGATAATTGG
sisA-G0 gRNA Target 3 (5’-3’) – ATAGTGTAGCTATGTGTCGCAGG
sisA-G0 gRNA Target 4 (5’-3’) – GCTGAAAACGGAGCTTGCTATGG
gfp-G0 gRNA Target 1 (5’-3’) – CAGGGTCAGCTTGCCGTAGG
gfp-G0 gRNA Target 2 (5’-3’) – AGCACTGCACGCCGTAGGTC
gfp-G0 gRNA Target 3 (5’-3’) – CGGCCATGATATAGACGTTG
gfp-G0 gRNA Target 4 (5’-3’) – CATGCCGAGAGTGATCCCGG
Four guides (or two) were cloned into the pCFD5 vector #73914 (Addgene) as previously described [69] using the listed primers.
sisA-G0-4-PCR1-Fw (5’-3’) – GCGGCCCGGGTTCGATTCCCGGCCGATGCATCGTTGGCCAATCCGGATGCGT TTTAGAGCTAGAAATAGCAAG
sisA-G0-4-PCR1-Rv (5’-3’) – ATTATCAGGTAGACTCAGTGTGCACCAGCCGGGAATCGAACCC
sisA-G0-4-PCR2-Fw (5’-3’) – CACTGAGTCTACCTGATAATGTTTTAGAGCTAGAAATAGCAAG
sisA-G0-4-PCR2-Rv (5’-3’) – GCGACACATAGCTACACTATTGCACCAGCCGGGAATCGAACCC
sisA-G0-4-PCR3-Fw (5’-3’) – ATAGTGTAGCTATGTGTCGCGTTTTAGAGCTAGAAATAGCAAG
sisA-G0-4-PCR3-Rv (5’-3’) –ATTTTAACTTGCTATTTCTAGCTCTAAAACTAGCAAGCTCCGTTTTCAGCTGC ACCAGCCGGGAATCGAACCC
GFP-G0-4-PCR1-Fw (5’-3’) – GCGGCCCGGGTTCGATTCCCGGCCGATGCACAGGGTCAGCTTGCCGTAGGGT TTTAGAGCTAGAAATAGCAAG
GFP-G0-4-PCR1-Rv (5’-3’) – GACCTACGGCGTGCAGTGCTTGCACCAGCCGGGAATCGAACCC
GFP-G0-4-PCR2-Fw (5’-3’) – AGCACTGCACGCCGTAGGTCGTTTTAGAGCTAGAAATAGCAAG
GFP-G0-4-PCR2-Rv (5’-3’) – CAACGTCTATATCATGGCCGTGCACCAGCCGGGAATCGAACCC
GFP-G0-4-PCR3-Fw (5’-3’) – CGGCCATGATATAGACGTTGGTTTTAGAGCTAGAAATAGCAAG
GFP-G0-4-PCR3-Rv (5’-3’) –ATTTTAACTTGCTATTTCTAGCTCTAAAACCCGGGATCACTCTCGGCATGTGC ACCAGCCGGGAATCGAACCC
Constructs were injected into embryos and integrated via PhiC31 integrase-mediated transgenesis (done at BestGene Inc.) into the same genomic location at P{CaryP}attP40 on Chromosome II. For germline-specific CRISPR-mediated knockout of sisA and gfp, male flies carrying gRNA expressing transgenes were mated with nanos-Cas9 virgin females ([70], BDSC# 54591 or [71], BDSC# 78782) and the crosses were maintained at 29°C. The progeny were reared at 29°C until 3-5 days post-eclosion (unless otherwise specified).
Generating sisA mutant
CRISPR-Cas9 was used as previously described to generate a sisA deletion mutant [72,73]. 2 guide RNA (gRNA) sequences were selected flanking the sisA gene using the FlyCRISPR algorithm (http://flycrispr.molbio.wisc.edu) [66], contain 20 nucleotides each, and have no predicted off-targets. The following gRNAs were selected:
sisA-del gRNA Target 1 (5’-3’) – GTCCAATGGCAAGCTACCTG
sisA-del gRNA Target 2 (5’-3’) – GATTACCTTCGGCCAGGCGA
The guides were cloned into the pU6-2-BbsI-gRNA vector #1363 (DGRC) using the listed primers that were phosphorylated using T4 PNK.
sisA-del gRNA Target 1 sense (5’-3’) – CTTCGTCGTTGGCCAATCCGGATGC
sisA-del gRNA Target 1 antisense (5’-3’) – AAACGCATCCGGATTGGCCAACGAC
sisA-del gRNA Target 2 sense (5’-3’) – CTTCGCACTGAGTCTACCTGATAAT
sisA-del gRNA Target 2 antisense (5’-3’) – AAACATTATCAGGTAGACTCAGTGC
A donor plasmid to facilitate homology dependent repair was generated. A 5’ homology arm and a 3’ homology arm were PCR-amplified from genomic DNA of vas-Cas9 expressing flies (BDSC# 51324, 56552) and cloned into multiple cloning sites found upstream and downstream of a removable 3xP3-DsRed marker in the pHD-DsRed-attP vector #1361 (DGRC).
sisA-del 5’ Arm Fw (5’-3’) – AGCACACCTGCACGACCGATGAAAATGGAGCAAGTGGAAAGCACACCTGCA CGACCGATGAAAATGGAGCAAGTGGAA
sisA-del 5’ Arm Rv (5’-3’) – AGCACACCTGCACGATTAACCTTAGGCAATATGTCAGCCAGCACACCTGCAC GA TTAACCTTAGGCAATATGTCAGCC
sisA-del 3’ Arm Fw (5’-3’) – GGCGGCTCTTCCTAAAAAACGAATGCTTTCTTATTGGCGGCTCTTCCTAA AAAACGAATGCTTTCTTATT
sisA-del 3’ Arm Rv (5’-3’) – GGCGGCTCTTCCCGGTTTAATATACATATATTTGTGGCGGCTCTTCCCGG TTTAATATACATATATTTGT
The gRNA expression plasmid and donor plasmid were injected by BestGene into embryos of Vas-Cas9 flies (#56552, #51324, BDSC). DsRed-positive flies were balanced without DsRed removal to maintain a selectable marker. Successful deletion of the sisA gene and replacement with the DsRed marker was confirmed by sequencing.
Acknowledgements
We thank VDRC, Vienna, and the Bloomington Drosophila Stock Center, Indiana, for flies; the Developmental Studies Hybridoma Bank for antibodies; and Flybase (www.flybase.org) for essential information. We thank Integrated Imaging Center at The Johns Hopkins University, Baltimore, for imaging. This work was supported by NSF Fellowship DGE-123285 (to R.G.) NIH grant R01GM113001 (to M.V.D).
Figure legends
Cartoons in Figures S1A, S2A, S3A, S5D, and S6C were created with Biorender.com
References
- 1.Direct proof through non-disjunction that the sex-linked genes of drosophila are borne by the x-chromosomeScience 40:107–109https://doi.org/10.1126/science.40.1020.107
- 2.Indirect effects of ploidy suggest X chromosome dose, not the X:A ratio, signals sex in DrosophilaPLoS Biol 5https://doi.org/10.1371/journal.pbio.0050332
- 3.Autoregulatory functioning of a Drosophila gene product that establish es and maintains the sexually determined stateGenetics 107:231–277
- 4.Primary events in the determination of sex in Drosophila melanogasterThe origin and evolution of sex New York, USA: A.R. Liss
- 5.The sex combs in males and intersexes of Drosophila melanogasterJ Exp Zool 134:533–556https://doi.org/10.1002/jez.1401340307
- 6.Positive autoregulation of sex-lethal by alternative splicing maintains the female determined state in DrosophilaCell 65:229–239https://doi.org/10.1016/0092-8674(91)90157-t
- 7.Genetic and molecular analysis of the autosomal component of the primary sex determination signal of Drosophila melanogasterGenetics 141:1451–1471
- 8.Triploid intersexes in drosophila melanogasterScience 54:252–254https://doi.org/10.1126/science.54.1394.252
- 9.Evidence that sisterless-a and sisterless-b are two of several discrete “numerator elements” of the X/A sex determination signal in Drosophila that switch Sxl between two alternative stable expression statesGenetics 119:829–862
- 10.Sex-lethal, a link between sex determination and sexual differentiation in Drosophila melanogasterCold Spring Harb Symp Quant Biol 50:595–604https://doi.org/10.1101/sqb.1985.050.01.072
- 11.Developmental distribution of female-specific Sex-lethal proteins in Drosophila melanogasterGenes Dev 5:403–415https://doi.org/10.1101/gad.5.3.403
- 12.PRODUCT OF THE MATERNALLY-INFLUENCED SEX-LETHAL GENE DETERMINES SEX IN DROSOPHILA-MELANOGASTERGenetics
- 13.A male-specific lethal mutation in Drosophila melanogaster that transforms sexDev Biol 72:266–275https://doi.org/10.1016/0012-1606(79)90117-9
- 14.Sex determination and dosage compensation in Drosophila melanogaster: production of male clones in XX femalesEMBO J 2:485–491https://doi.org/10.1002/j.1460-2075.1983.tb01451.x
- 15.Clonal analysis ofsex-lethal, a gene needed for female sexual development inDrosophila melanogaster. Wilhelm Roux’Archiv 191:211–214https://doi.org/10.1007/BF00848339
- 16.The interaction between daughterless and sex-lethal in triploids: a lethal sex-transforming maternal effect linking sex determination and dosage compensation in Drosophila melanogasterDev Biol 95:260–274https://doi.org/10.1016/0012-1606(83)90027-1
- 17.The link between dosage compensation and sex differentiation in Drosophila melanogasterChromosoma. 82:217–227
- 18.Intersexuality resulting from the interaction of sex-specific lethal mutations in Drosophila melanogasterDev Biol 94:153–162https://doi.org/10.1016/0012-1606(82)90078-1
- 19.Studies on the sex-specific lethals of Drosophila melanogaster. V. Sex transformation caused by interactions between a female-specific lethal, Sxlf 1, and the male-specific lethals mle(3)132, msl-2(27), and mleGenetics 102:233–243
- 20.The primary sex determination signal of Drosophila acts at the level of transcriptionCell 68:933–943https://doi.org/10.1016/0092-8674(92)90036-c
- 21.A shared enhancer controls a temporal switch between promoters during Drosophila primary sex determinationProc Natl Acad Sci USA 105:18436–18441https://doi.org/10.1073/pnas.0805993105
- 22.The Drosophila segmentation gene runt acts as a position-specific numerator element necessary for the uniform expression of the sex-determining gene Sex-lethalGenes Dev 5:2176–2187https://doi.org/10.1101/gad.5.12a.2176
- 23.The JAK/STAT signaling pathway is required for the initial choice of sexual identity in Drosophila melanogasterMol Cell 5:581–587https://doi.org/10.1016/s1097-2765(00)80451-7
- 24.Sex determination in Drosophila melanogaster: X-linked genes involved in the initial step of sex-lethal activationDev Genet 15:251–264https://doi.org/10.1002/dvg.1020150307
- 25.An extracellular activator of the Drosophila JAK/STAT pathway is a sex-determination signal elementNature 405:970–973https://doi.org/10.1038/35016119
- 26.Interpretation of X chromosome dose at Sex-lethal requires non-E-box sites for the basic helix-loop-helix proteins SISB and daughterlessMol Cell Biol 21:1581–1592https://doi.org/10.1128/MCB.21.5.1581-1592.2001
- 27.The segmentation gene runt is needed to activate Sex-lethal, a gene that controls sex determination and dosage compensation in DrosophilaGenet Res 59:189–198https://doi.org/10.1017/s0016672300030470
- 28.A bZIP protein, sisterless-a, collaborates with bHLH transcription factors early in Drosophila development to determine sexGenes Dev 7:1688–1702https://doi.org/10.1101/gad.7.9.1688
- 29.Direct activation of Sex-lethal transcription by the Drosophila runt proteinDevelopment 126:191–200
- 30.Drosophila JAK/STAT pathway reveals distinct initiation and reinforcement steps in early transcription of SxlCurr Biol 17:643–648https://doi.org/10.1016/j.cub.2007.02.038
- 31.The regulation of germline sex determination in Drosophila by Sex lethalSex Dev 16:323–328https://doi.org/10.1159/000521235
- 32.Cell-autonomous and inductive signals can determine the sex of the germ line of drosophila by regulating the gene SxlCell 57:157–166https://doi.org/10.1016/0092-8674(89)90181-5
- 33.Sex determination in Drosophila: sis-b, a major numerator element of the X:A ratio in the soma, does not contribute to the X:A ratio in the germ lineDevelopment 117:763–767
- 34.Sex determination signals control ovo-B transcription in Drosophila melanogaster germ cellsGenetics 160:537–545
- 35.Studies on the expression of the transformer gene of drosophila melanogasterGenetics 46:143–156
- 36.Regulatory and functional interactions between the somatic sex regulatory gene transformer and the germline genes ovo and ovarian tumorDevelopment 126:861–871
- 37.Selection and maintenance of sexual identity in the Drosophila germlineGenetics 141:1521–1535
- 38.The somatic sex determines the requirement for ovarian tumor gene activity in the proliferation of the Drosophila germlineDevelopment 121:579–587
- 39.Sex determination in the germ line of Drosophila depends on genetic signals and inductive somatic factorsDevelopment 107:505–518
- 40.Sex-lethal, master and slave: a hierarchy of germ-line sex determination in DrosophilaDevelopment 119:897–908
- 41.A Gene in Drosophila Melanogaster That Transforms Females into MalesGenetics 30:297–299
- 42.Somatic sex-determining signals act on XX germ cells in Drosophila embryosDevelopment 122:4065–4071
- 43.[experimental studies on the principles of sterility of transformer (tra) males in drosophila melanogaster]Z Vererbungsl 94:217–241
- 44.Sex determination in germ line chimeras of Drosophila melanogasterJ Embryol Exp Morphol 37:173–185
- 45.Is sex determination in germ line and soma controlled by separate genetic mechanisms?Nature 272:249–251https://doi.org/10.1038/272249a0
- 46.Normal female germ cell differentiation requires the female X chromosome to autosome ratio and expression of sex-lethal in Drosophila melanogasterGenetics 109:529–548
- 47.Autosomal mutations that interfere with sex determination in somatic cells of Drosophila have no direct effect on the germlineDev Biol 89:117–127https://doi.org/10.1016/0012-1606(82)90300-1
- 48.Functional changes associated with structural alterations induced by mobilization of a P element inserted in the Sex-lethal gene of DrosophilaGenetics 117:221–231
- 49.Induction of female Sex-lethal RNA splicing in male germ cells: implications for Drosophila germline sex determinationDevelopment 124:5033–5048
- 50.Drosophila Sex lethal gene initiates female development in germline progenitorsScience 333:885–888https://doi.org/10.1126/science.1208146
- 51.The Drosophila sex determination gene daughterless has different functions in the germ line versus the somaCell 48:479–487https://doi.org/10.1016/0092-8674(87)90198-x
- 52.Sex determination in the germ line of Drosophila melanogaster: activation of the gene Sex-lethalDevelopment 118:813–816
- 53.Sxl in the germline of Drosophila: a target for somatic late inductionDev Genet 15:265–274https://doi.org/10.1002/dvg.1020150308
- 54.Sex determination of the Drosophila germ line: tra and dsx control somatic inductive signalsDevelopment 120:707–716
- 55.Regulation of zygotic gene expression in Drosophila primordial germ cellsCurr Biol 8:243–246https://doi.org/10.1016/s0960-9822(98)70091-0
- 56.Spradling AC. bag-of-marbles: a Drosophila gene required to initiate both male and female gametogenesisGenes Dev 4:2242–2251https://doi.org/10.1101/gad.4.12b.2242
- 57.The Drosophila melanogaster sex determination gene sisA is required in yolk nuclei for midgut formationGenetics 155:191–202
- 58.DiNardo S. bag-of-marbles and benign gonial cell neoplasm act in the germline to restrict proliferation during Drosophila spermatogenesisDevelopment 124:4361–4371
- 59.The establishment of sexual identity in the Drosophila germlineDevelopment 136:3821–3830https://doi.org/10.1242/dev.042374
- 60.Single-molecule super-resolution imaging of chromosomes and in situ haplotype visualization using Oligopaint FISH probesNat Commun 6https://doi.org/10.1038/ncomms8147
- 61.Versatile design and synthesis platform for visualizing genomes with Oligopaint FISH probesProc Natl Acad Sci USA 109:21301–21306https://doi.org/10.1073/pnas.1213818110
- 62.Visualizing genomes with Oligopaint FISH probesCurr Protoc Mol Biol 105https://doi.org/10.1002/0471142727.mb1423s105
- 63.Characterization of Button Loci that Promote Homologous Chromosome Pairing and Cell-Type-Specific Interchromosomal Gene RegulationDev Cell 51:341–356https://doi.org/10.1016/j.devcel.2019.09.007
- 64.An accurate and interpretable model for siRNA efficacy predictionBMC Bioinformatics 7https://doi.org/10.1186/1471-2105-7-520
- 65.Advances in Engineering the Fly Genome with the CRISPR-Cas SystemGenetics 208:1–18https://doi.org/10.1534/genetics.117.1113
- 66.Highly specific and efficient CRISPR/Cas9-catalyzed homology-directed repair in DrosophilaGenetics 196:961–971https://doi.org/10.1534/genetics.113.160713
- 67.Phf7 controls male sex determination in the Drosophila germlineDev Cell 22:1041–1051https://doi.org/10.1016/j.devcel.2012.04.013
- 68.Gal4 in the Drosophila female germlineMech Dev 78:113–118
- 69.Augmenting CRISPR applications in Drosophila with tRNA-flanked sgRNAsNat Methods 13:852–854https://doi.org/10.1038/nmeth.3972
- 70.Optimized CRISPR/Cas tools for efficient germline and somatic genome engineering in DrosophilaProc Natl Acad Sci USA 111:E2967–76https://doi.org/10.1073/pnas.1405500111
- 71.Optimized gene editing technology for Drosophila melanogaster using germ line-specific Cas9Proc Natl Acad Sci USA 110:19012–19017https://doi.org/10.1073/pnas.1318481110
- 72.CRISPR-Cas9 Genome Editing in DrosophilaCurr Protoc Mol Biol 111https://doi.org/10.1002/0471142727.mb3102s111
- 73.Precise Genome Editing of Drosophila with CRISPR RNA-Guided Cas9Methods Mol Biol 1311:335–348https://doi.org/10.1007/978-1-4939-2687-9_22
- 74.Sex-specific GFP-expression in Drosophila embryos and sorting by Copas flow cytometry technique45th Annual Drosophila Research Conference
- 75.Suppression of distinct ovo phenotypes in the Drosophila female germline bymaleless− andSex-lethalMDev Genet 23:335–346https://doi.org/10.1002/(SICI)1520-6408(1998)23:4<335::AID-DVG8>3.0.CO;2-M
- 76.Genetic evidence that the ovo locus is involved in Drosophila germ line sex determinationGenetics 125:535–550
- 77.Evidence for sex transformation of germline cells in ovarian tumor mutants of DrosophilaDev Biol 161:318–320https://doi.org/10.1006/dbio.1994.1032
- 78.The ovo locus is required for sex-specific germ line maintenance in DrosophilaGenes Dev 1:913–923https://doi.org/10.1101/gad.1.9.913
- 79.Function of Drosophila ovo+ in germ-line sex determination depends on X-chromosome numberDevelopment 120:3185–3195
- 80.Genome-wide analysis of the maternal-to-zygotic transition in Drosophila primordial germ cellsGenome Biol 13https://doi.org/10.1186/gb-2012-13-2-r11
- 81.A female-specific lethal lesion in an X-linked positive regulator of the Drosophila sex determination gene, Sex-lethalGenetics 113:641–663
- 82.Sex- and tissue-specific functions of Drosophila doublesex transcription factor target genesDev Cell 31:761–773https://doi.org/10.1016/j.devcel.2014.11.021
- 83.Protein-protein interactions among components of the Drosophila primary sex determination signalMol Gen Genet 248:182–189https://doi.org/10.1007/bf02190799
- 84.Maintenance of Drosophila germline stem cell sexual identity in oogenesis and tumorigenesisDevelopment 142:1073–1082https://doi.org/10.1242/dev.116590
- 85.Tudor-domain containing protein 5-like promotes male sexual identity in the Drosophila germline and is repressed in females by Sex lethalPLoS Genet 15https://doi.org/10.1371/journal.pgen.1007617
- 86.Germline sex determination regulates sex-specific signaling between germline stem cells and their nicheCell Rep 39https://doi.org/10.1016/j.celrep.2022.110620
Article and author information
Author information
Version history
- Sent for peer review:
- Preprint posted:
- Reviewed Preprint version 1:
Copyright
© 2024, Goyal et al.
This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.
Metrics
- views
- 167
- downloads
- 4
- citations
- 0
Views, downloads and citations are aggregated across all versions of this paper published by eLife.