The dynamic expression of YAP is essential for the development of male germ cells derived from human embryonic stem cells – Nature.com

Derivation of spermatogenic cells from human embryonic stem cells (hESCs)

To determine the spermatogenic differentiation potential of non-manipulated hESCs, MUSIe002-A were subjected to a spermatogenic differentiation procedure. At the end of culture, the hESC-derived spermatogenic cells expressed several human spermatogenic marker proteins, including the pan-germ cell marker (VASA), spermatogonia stem cell markers (PLZF, CD90 and GPR125), and a spermatocyte marker (PIWIL1) (Fig.1A) compared to non-differentiate hESCs on culture day 0 (Fig. S1). In accordance with their protein expression, hESC-derived spermatogenic cells were subjected to transcriptional analysis by qRT-PCR. Up-regulation of SOX17 and TFAP2C was observed, which indicated the development toward germ cell lineages35. Furthermore, an early PCG marker BLIMP1, the post-migratory gonocyte PIWIL4 and the germline-specific gene VASA were up-regulated on day 5 and were maintained to the end of culture. Spermatogonia markers, including PLZF, GFRA, and NANOS3, were found to be highly expressed on day 5 and to remain throughout the culture period, while expressions of GRP125, ID4, and c-Kit were found to be upregulated on day 5 after induction (Fig.1B). Furthermore, the expression levels of the meiotic marker PIWIL2, and the post-meiotic haploid marker ACR, in differentiated hESCs also increased steadily and reached their highest levels at the end of culture (Fig.1B).

Generation of hESC-derived spermatogenic cells. (A) Representative immunofluorescence micrographs show the expression of germ cell markers VASA, the hSSC markers PLZF, CD90, GPR125, and the spermatocyte marker PIWIL1 in hSSLC. Non-differentiated hESCs on culture day 0 and IgG staining served as a control. Scale bar: 100m in the first and middle panel, 20m in the last panel. The micrographs shown in the last column are the zoomed-in of the area indicated in red box (B). The graphs show the expression levels of the early germ cell markers (STELLA, TFAP2C, BLIMP1, SOX17, PIWIL4 and VASA), hSSC markers (PLZF, GFRA, GPR125, ID4, NANOS3 and c-Kit), spermatocyte (PIWIL2) and spermatid marker (ACR) during spermatogenic differentiation of hESCs relative to those on culture day 0. Data are presented as meanSD of 3 experiments. (C) Expression of CD90+ cells (top row), PLZF+ cells in the sorted CD90+ cell population determined by flow cytometry (middle row) and immunofluorescent staining (bottom row). The picture in the right panel showed the zoomed-in area indicated in red. (D) Analysis of DNA content by flow cytometry shows the percentages of hESC-derived haploid spermatid-like cells at the end of spermatogenic differentiation. Non-differentiated hESCs on culture day 0 serves as a negative control, while mature human sperm serves as a positive control. (E) Representative immunofluorescence micrographs show the expression of the haploid germ cell marker Acrosin in human sperm cells (top row), Acrosin and TNP1 in hESC-derived spermatid-like cells. (F) Representative micrographs show a single copy of chromosome 13, 21, X and Y chromosome in hESC-derived spermatid-like cells determined by fluorescent in situ hybridization (FISH). Non-differentiated hESCs, which served as a control, had two copies of these chromosomes.

Next, we determined the percentage of hESC-derived hSSLCs after spermatogenic differentiation using the cell surface marker CD90, which has previously been used to identify SSLC in human and mouse models20,36. The results show that 66% of differentiated hESCs expressed CD90 (Fig.1C; top row). We also found that more than 90% of the CD90+hSSLCs also co-expressed PLZF, another widely used SSLC marker, as determined by flow-cytometry and immunofluorescent staining (Fig.1C; middle and bottom rows, respectively). These results confirm that hESC-derived hSSLCs are present in the culture after spermatogenic differentiation and could be identified with a combination of CD90 and PLZF markers.

To determine whether the differentiated hESCs generated any haploid spermatid-like cells, a DNA content analysis was performed. The results show that 6.70.72% of the differentiated hESCs are haploid cells, while no haploid cell was detected in non-differentiated hESCs (Figs.1D and S2). The haploid spermatid-like cells derived from hESCs expressed human spermatid markers, Acrosin and TNP1 (Figs.1E and S2). Furthermore, fluorescent in situ hybridization (FISH) analysis confirmed that our sorted haploid spermatid-like cell derived from hESCs has a single copy of chromosome 13, 21 and sex chromosome compared to two copies of the same chromosomes in non-differentiated hESCs (Fig.1F). These results demonstrated that the spermatogenic differentiation procedure used in this study successfully generated haploid spermatid-like cells from hESCs. The pattern of marker expression during in vitro spermatogenic differentiation closely matched that of human testicular germ cells that underwent in vivo spermatogenesis (Supplementary Fig. S3AG)34. Collectively, these results confirm that we could derive male germ cells from hESCs in vitro using the protocol previously described by Easley et al.21. Next, we determine whether Hippo/YAP is expressed in human testicular germ cells. Analysis of scRNA sequencing data showed YAP expression during spermatogenesis (Supplementary Fig. S3H,I). Like PLZF, the expression level of YAP was higher in undifferentiated human spermatogonia compared to the later stages of human spermatogenic differentiation (Supplementary Fig. S3H,I). This result suggested that YAP might play a crucial role during the early stages of human spermatogenesis and could be involved in the expression of the undifferentiated spermatogonia marker PLZF.

To investigate the role of YAP during human spermatogenesis, the isogenic YAP-depleted hESCs (YAP-KD cells) were established by transducing hESCs (MUSIe002-A cells) with the YAP-targeted CRISPR/Cas9 plasmid33. Western blot analysis showed a 50% reduction in the YAP level in YAP-KD cells compared to their wild-type control (WT cells) (Fig.2A). The YAP-KD cells retained a typical morphology of non-differentiated wild-type hESCs even after being expanded for more than 25 passages (Figs.2B and S4). After spermatogenic differentiation, the YAP-KD cells generated several clusters of cells, which increased in size and number toward the end of the culture (indicated by black arrows in Figs. 2C and S5). These cell clusters were also observed in differentiated WT cells, but their number was less than in the differentiated YAP-KD cells (Figs. 2C and S5).

Spermatogenic differentiation of YAP knockdown hESCs (YAP-KD cells). (A) Western blot shows the level of YAP relative to -ACTIN in YAP-KD cells compared to wild-type hESCs (WT). (B) Representative micrograph showing the morphology of a non-differentiated YAP-KD cell colony. Scale bar: 200m. (C) Representative micrographs show the morphology of differentiated YAP-KD cells during their spermatogenic differentiation. Wild-type hESCs cultured under the same conditions (WT) serve as controls. Scale bar: 200m in micrographs, 50m in zoomed areas. (D) Representative immunofluorescence micrographs show the expression of the SSC marker PLZF in YAP-KD cells during their spermatogenic differentiation compared to WT cells. Scale bar: 200m. (E) Representative dot plots show the percentages of PLZF+ hSSLCs in YAP-KD cells during their spermatogenic differentiation compared to those of WT cells. (F) Graphs show the number of PLZF+ hSSLCs derived from YAP-KD cells during their spermatogenic differentiation compared to those of WT cells. Data are presented as meanSD from 3 experiments. ****P<0.0001. (G) Western blot shows the levels of the YAP and PLZF proteins relative to -ACTIN in YAP-KD cells during their spermatogenic differentiation compared to those of WT cells. (H) Graphs show the expression level of YAP protein relative to -ACTIN in YAP-KD cells during their spermatogenic differentiation compared to that of WT cells. Data are presented as meanSD of 3 experiments. *P<0.05, **P<0.01, ****P<0.0001. (I) Graphs show the expression level of PLZF protein relative to -ACTIN in YAP-KD cells during their spermatogenic differentiation compared to that of WT cells. Data are presented as meanSD of 3 experiments. **P<0.01, ***P<0.001. (J) Correlation analysis between YAP protein levels and hESC-derived PLZF+ hSSC numbers during spermatogenic differentiation of WT and YAP-KD cells. The uncropped gels associated with Fig.2A are shown in Supplementary Fig. S8.

YAP protein is highly expressed in spermatogonia residing in testicular germ cells10,11. Similar to WT cells, differentiated YAP-KD cells highly expressed PLZF, a spermatogonia marker. Immunofluorescence shows that the level of PLZF expression in differentiated YAP-KD cells increased steadily throughout the entire culture period and was much higher than that of WT cells (Fig.2D). Consistent with the level of PLZF expression, the percentages of PLZF-expressing hSSLCs (PLZF+ cells) derived from the differentiated YAP-KD cells increased steadily throughout the differentiation period and were 23 times higher than those derived from WT cells throughout the culture period (Figs. 2E and S6). Consistent with these data, when the absolute number of PLZF+ cells was determined by BD Truecount, the YAP-KD cell also generated a higher number of PLZF+ cells than the WT cells, resulting in an approximately threefold increase in the number of PLZF+ cells at the end of the culture (8.61105cells vs 2.19105 cells on culture day 12) (Fig.2F). This result indicated that the depletion of YAP in hESCs enhances their spermatogenic differentiation toward PLZF+ hSSLCs.

Next, we quantified the expression levels of YAP and PLZF proteins at different time points during the spermatogenic differentiation of YAP-KD cells compared to WT cells. The expression level of YAP in WT cells was initially downregulated on culture day 5, bounced back on culture day 10, and continued to increase towards the end of the culture (Fig.2G,H). Interestingly, YAP-KD cells, whose YAP expression was 50% lower than that of WT cells at the beginning of culture, rapidly increased their YAP expression on culture day 5 (Fig.2G,H). Similar to WT cells, the expression level of YAP in YAP-KD cells continued to increase toward the end of the culture. At the end of the culture (day 12), the expression level of YAP in YAP-KD cells is similar to that of WT cells. However, unlike WT cells, down-regulation of YAP during the early phase of spermatogenic differentiation (day 5) was not observed in differentiated YAP-KD cells (Fig.2G,H).

The up-regulation of YAP in WT and YAP-KD cells coincided with the expression of PLZF, whose level also increased steadily towards the end of the culture (Fig.2G,I). However, the differentiated YAP-KD cells expressed PLZF earlier than WT cells (day 5 vs day 10), and their levels of PLZF were significantly higher than those of WT cells throughout the culture period (Fig.2G,I).

Furthermore, the level of YAP protein in both WT and YAP-KD cells was positively correlated with the expression level of PLZF+ hSSLCs generated during their spermatogenic differentiation (R2=0.8577, P=0.001) (Fig.2J). These results suggested that depletion of YAP at the beginning of spermatogenic differentiation leads to higher and earlier expression of PLZF, resulting in a higher number of hESC-derived hSSLCs.

As shown above, early depletion of YAP in hESCs increased the percentage and absolute number of hSSLCs during spermatogenic differentiation. Next, we determine whether hESC-derived hSSLCs express c-Kit, a differentiated spermatogonia marker. We performed a flow cytometric analysis for the expression of c-Kit in hSSLCs derived from WT and YAP-KD cells at different time points during their spermatogenic differentiation. The results showed that both the percentage and the absolute number of c-Kit+differentiated spermatogonial cells were up-regulated in the WT and YAP-KD groups as differentiation progressed towards the end of the culture. In which the percentages and absolute numbers of c-Kit+cells in the YAP-KD groups are significantly higher than those of WT on days 7 and 12 of culture (Fig.3A). This pattern of c-kit expression is similar to that of PLZF expression and therefore confirms that YAP depletion increases the spermatogenic differentiation of hESCs.

Derivation of haploid spermatid-like cells from YAP-KD cells. (A) Representative dot plots show the percentages of c-Kit+ cells in differentiated WT and YAP-KD cells during their spermatogenic differentiation (left). The graph shows the absolute number of c-Kit+ differentiated spermatogonia derived from YAP-KD cells during their spermatogenic differentiation compared to those of WT cells (right). Data are presented as meanSEM from 3 replicates. *P<0.05, **P<0.01 (B) Western blot shows the levels of VASA proteins relative to -ACTIN in YAP-KD cells during their spermatogenic differentiation compared to those of WT cells. (C) Graphs show the expression level of VASA protein relative to -ACTIN in YAP-KD cells during their spermatogenic differentiation compared to that of WT cells. Data are presented as meanSD of 3 experiments. *P<0.05, ****P<0.0001. (D) Western blot shows the levels of Acrosin proteins relative to -ACTIN in YAP-KD cells during their spermatogenic differentiation compared to those of WT cells. (E) Graphs show the expression level of Acrosin protein relative to -ACTIN in YAP-KD cells during their spermatogenic differentiation compared to that of WT cells. Data are presented as meanSD of 3 experiments. **P<0.01, ****P<0.0001. (F) Analysis of DNA content by flow cytometry shows the percentages of haploid spermatid-like cells derived from YAP-KD cells during their spermatogenic differentiation compared to those of WT cells. (G) The graphs show the percentages of haploid spermatid-like cells derived from YAP-KD cells during their spermatogenic differentiation compared to those of WT cells. Data are presented as meanSD of 3 experiments. *P<0.05. The uncropped gels associated with Fig.3A,C are shown in Supplementary Fig. S9.

We then investigated the effect of YAP depletion during the early stages of spermatogenic differentiation by measuring the expression level of VASA, an early germ cell marker. We found that YAP-KD cells expressed a significantly higher level of VASA protein than WT cells throughout the entire culture period (Fig.3B,C).

Next, we investigated the effect of YAP depletion during the later stages of spermatogenic differentiation by measuring the expression level of Acrosin, a specific marker of haploid spermatids, in differentiated YAP-KD cells. Similar to PLZF, differentiated YAP-KD cells up-regulated their acrosin expression earlier than WT cells (day 5 vs day 10) and their acrosin levels were significantly higher than those of WT cells on day 5 and day 10 of culture (Fig.3D,E). Consistent with the level of acrosin expression, the percentages of haploid spermatid cells in differentiated YAP-KD cells were significantly higher than those of their wild-type counterparts on day 5 (4.40.05% vs. 1.80.68%) and day 10 of culture (6.22.09% vs. 20.10%) (Fig.3F,G). These results suggest that YAP depletion at the beginning of spermatogenic differentiation could accelerate the progression of hESC-derived hSSLCs into haploid spermatid-like cells. However, we found that acrosin+spermatid-like cells could not be maintained in culture, since the level of acrosin in differentiated YAP-KD cells on day 12 of culture was lower than on day 10 (Fig.3D).

Due to the heterozygous nature of the mutated YAP gene in our YAP-KD cells, up-regulation of YAP expression was observed during the later stages of spermatogenic differentiation (days 512) (Fig.2G,H). This pattern of YAP expression in YAP-KD cells suggests that although YAP depletion at the beginning of spermatogenic differentiation appears to promote the germline specification of hESCs, up-regulation of YAP during the later stages of spermatogenic differentiation could be crucial for the expression of germ cell markers and the derivation of both hSSLCs and haploid spermatid-like cells.

To prevent the up-regulation of YAP in YAP-KD cells during the later stages of spermatogenic differentiation, a short hairpin RNA (shRNA) targeting the YAP transcript37 was transfected into YAP-KD cells to establish a YAP double knockdown hESCs (YAP-DKD cells). YAP-DKD cells showed a further reduction in YAP (Fig.4A; Supplementary Table S2) and YAP target genes, CTGF and CYR61, compared to YAP-KD cells (Fig.4B). Like YAP-KD cells, YAP-DKD cells exhibited a typical morphology of non-differentiated wild-type hESCs (Supplementary Fig. S4).

Spermatogenic differentiation of YAP double knockdown hESCs (YAP-DKD cells) (A). Western blot shows the level of YAP relative to -ACTIN in YAP-DKD cells compared to that of YAP-KD cells and wild-type hESCs (WT). ****P<0.0001. (B) Graphs show the expression levels of YAP target genes, CTGF and CYR61, in YAP-DKD cells compared to those of YAP-KD cells and WT. Data are presented as meanSD of 3 experiments. **P<0.01, ***P<0.001. (C) Western blot shows YAP, PLZF, VASA and acrosin protein levels relative to -ACTIN in WT, YAP-KD, and YAP-DKD cells during their spermatogenic differentiation. (D) Graphs show the level of PLZF protein relative to -ACTIN in YAP-DKD cells during their spermatogenic differentiation compared to YAP-KD cells and WT. (E) Graphs show the expression levels of spermatogenic cell markers (VASA, PLZF, GFRA1, GPR125 and ACR) during spermatogenic differentiation of YAP-DKD, compared to those of YAP-KD cells and WT. Data are presented as means SD of 3 experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001. (F) Representative micrographs show the morphology of YAP-DKD cells during their spermatogenic differentiation. White dots indicate areas of dead cells. Scale bar: 200m for the upper panel and 100m for the lower panel. (G) Western blot shows the level of the apoptotic marker cleaved Caspase3 relative to -ACTIN in WT and YAP-DKD cells during their spermatogenic differentiation. The uncropped membranes associated with this figure are provided in Supplementary Fig. S10.

When subjected to spermatogenic differentiation, different expression patterns of YAP and early germ cell markers of each cell type were observed. Unlike YAP-KD cells, the level of YAP expression in YAP-DKD cells was barely detected on day 10 and day 12 of culture (Fig.4C). This result confirmed that the shRNA successfully prevented the up-regulation of YAP in YAP-DKD cells during the later stages of their spermatogenic differentiation. Further analysis found that constitutive depletion of YAP in YAP-DKD inhibited the expression of the early germ cell marker VASA (Fig.4C), the SSC marker PLZF (Fig.4C,D), and the spermatid marker acrosin (Fig.4C). These results suggest that the up-regulation of YAP level during the later stages of spermatogenesis might be critical for the expression of germ cell markers and the derivation of hSSLCs and haploid spermatid-like cells from hESCs.

Consistent with the protein expression study, differentiated YAP-DKD cells expressed lower transcript levels of germ cell marker (VASA), hSSC marker genes (PLZF, GFRA1, and GPR125), and meiotic germ cell markers (ACR) compared to WT (Fig.4E). In contrast, YAP-KD cells, whose YAP was depleted only at the beginning of their spermatogenic differentiation, expressed higher levels of hSSC markers, PLZF, GPR125, and GFRA1 compared to WT cells (Fig.4E). Taken together, these results suggest that depleting YAP at the beginning of spermatogenic differentiation promotes germline specification of hESCs, and increasing the level of YAP during the later stages of spermatogenic differentiation is critical for the derivation and survival of hESC-derived hSSLCs and haploid spermatid-like cells.

Furthermore, we also found that the differentiated YAP-DKD cells contained several large areas of dead cells (dotted areas in Figs. 4F and S5). Consistent with this, western blots also showed a high level of cleaved-caspase 3 in YAP-DKD cells at the later stages of spermatogenic differentiation (D5-12) compared to D0 (Fig.4G; YAP-DKD), while the expression of cleaved-caspase 3 in WT did not show a sharp contrast compared to its control D0. These results confirm the higher level of apoptosis in differentiated YAP-DKD cells during the later stages of spermatogenesis.

To confirm that an initial depletion followed by an up-regulation of YAP promotes spermatogenic differentiation of hESCs, we modulated the expression level of YAP at specific time points during spermatogenic differentiation of wild-type hESCs using two small molecules. Dobutamine hydrochloride (DH), which has previously been shown to promote YAP phosphorylation and suppress YAP activity38,39, was used to inhibit YAP expression, while lysophosphatidic acid (LPA) activated YAP expression in wild-type hESCs. DH-treated hESCs showed a 50% reduction in YAP level compared to untreated hESCs (Supplementary Fig. S7B,C), and also expressed lower levels of YAP target genes, CTGF and CCND1, compared to untreated hESCs (Supplementary Fig. S7D). Our results demonstrate that DH treatment successfully inhibits the expression of YAP and its target genes in wild-type hESCs. Although LPA-treated hESCs increased CTGF, CCND1, and CYR61 expression levels compared to untreated hESCs (Supplementary Fig. S7D), their YAP protein level did not increase compared to the untreated cells (Supplementary Fig. S7B,C). These results suggest that although LPA did not increase YAP level in hESCs, it might enhance YAP activity resulting in higher levels of YAP target genes.

We then supplemented DH and LPA at various time points during spermatogenic differentiation of wild-type hESCs (Fig.5A). Consistent with the results generated from YAP-KD cells, the highest number of hESC-derived hSSLCs was observed under the condition that DH was added to inhibit YAP expression during the first 5days of spermatogenic differentiation before being removed for the rest of the culture period (Fig.5B, D12: DH-ET in Group 2). On the contrary, hESCs treated with LPA to increase YAP activity during the first 5days of their spermatogenic differentiation generated a lower number of hSSLCs than both D12: DH-ET and untreated cells (Fig.5B, D12: LPA-ET in Group 2).

Modulation of YAP expression during the various stages of spermatogenic cell differentiation of wild-type hESCs by using DH and LPA. (A) Diagrams show the timeline of DH and LPA supplementation during spermatogenic differentiation of wild-type hESCs. (B) Graphs show the total number of differentiated hESCs and the number of PLZF+ hSSCs derived from various conditions described in (A) at the end of the culture. Data are presented as meanSD of 3 experiments. **P<0.01, ***P<0.001, ****P<0.0001 and ### P<0.001 indicated the statistically significant of spermatogonia number from DH treated cells between each group. (C) Representative immunofluorescence micrographs show the level of PLZF expression in DH-ET and LPA-ET cells compared to that of their untreated counterpart. Scale bar: 200m. (D) Western blot shows the expression of YAP, PLZF, Acrosin, and -ACTIN proteins in DH-ET cells during their spermatogenic differentiation compared to those of DH-AT cells, untreated control cells, and non-differentiated hESCs. Red numbers indicate the expression level of each sample relative to -ACTIN. (E) Analysis of DNA content by flow cytometry shows the percentage of haploid spermatid-like cells derived from DH-ET cells at the end of their spermatogenic differentiation compared to that of LPA-ET and untreated cells. (F) The graph shows the percentage of haploid spermatid-like cells derived from DH-ET cells at the end of their spermatogenic differentiation compared to that of LPA-ET and untreated cells. Data are presented as meanSD of 3 experiments. **P<0.01, ***P<0.001. The uncropped membranes associated with Fig.5D are provided in Supplementary Fig. S11.

Unlike YAP suppression during the early stage of spermatogenic differentiation (days 05), suppressing YAP during the middle phase (days 510) of spermatogenic differentiation reduced the number of hESC-derived hSSLCs compared to untreated hESCs (Fig.5B, D12: DH-MT in Group 3). Moreover, suppressing YAP during the middle stage (DH-MT in Group 3) or throughout the spermatogenic differentiation (DH-AT in Group 1) also reduced the total number of differentiated hESCs in culture (Fig.5B), which could result in the decrease in the number of hESC-derived hSSLCs at the end of culture. This is consistent with our previous result with YAP-DKD cells showing that continued suppression of YAP during the later stages of spermatogenic differentiation increased the apoptotic level of differentiated hESCs (Fig.4F,G).

Furthermore, our immunofluorescence study shows that the level of PLZF expression in DH-ET cells on day 12 was higher than that of the untreated control, while the level of PLZF expression in LPA-ET cells was lower than that of the untreated control (Fig.5C). Western blot analysis further confirmed that DH-ET cells expressed higher levels of PLZF than the untreated control at the end of the culture (Fig.5D), while LPA-ET cells expressed lower levels of PLZF than the untreated control at the end of the culture (Supplementary Fig. S7E,F). Unlike suppression of YAP during the early stage of spermatogenic differentiation (DH-ET), suppression of YAP throughout spermatogenic differentiation (DH-AT) decreased the expression level of PLZF compared to the untreated control (Fig.5D). Although DH-ET treatment produces a higher number of haploid germ cells than the control, the level of Acrosin expression in the DH-ET group is lower than that of the control (Fig.5D). Because the level of Acrosin is associated with spermatid maturity, it is possible that while DH-ET treatment increases the number of hESC-derived spermatids, these spermatids may be less mature than those derived from the untreated control.

We further investigated haploid germ cell production using cell cycle analysis to observe DNA content of the cells. The DH-ET cells also generate significantly higher percentages of haploid spermatid-like cells at the end of culture than their untreated counterpart (14.162.61% vs 8.302.86%) (Fig.5E,F). These results support our hypothesis that inhibition of YAP activity during the early stage of spermatogenic differentiation facilitates the derivation of hSSLCs and spermatid-like cells from hESCs. In contrast, inhibition of YAP activity during the later stages of spermatogenic differentiation reduced the number of hESC-derived hSSLCs and spermatid-like cells, possibly by increasing the apoptotic rate of hSSLCs and preventing their maturation into haploid spermatid-like cells.

View post:
The dynamic expression of YAP is essential for the development of male germ cells derived from human embryonic stem cells - Nature.com