Aberrant gene expression by Sertoli cells in infertile men with Sertoli cell-only syndrome

Abstract

Sertoli cell-only (SCO) syndrome is a severe form of human male infertility seemingly characterized by the lack all spermatogenic cells. However, tubules of some SCO testes contain small patches of active spermatogenesis and thus spermatogonial stem cells. We hypothesized that these stem cells cannot replicate and seed spermatogenesis in barren areas of tubule because as-of-yet unrecognized deficits in Sertoli cell gene expression disable most stem cell niches. Performing the first thorough comparison of the transcriptomes of human testes exhibiting complete spermatogenesis with the transcriptomes of testes with SCO syndrome, we defined transcripts that are both predominantly expressed by Sertoli cells and expressed at aberrant levels in SCO testes. Some of these transcripts encode proteins required for the proper assembly of adherent and gap junctions at sites of contact with other cells, including spermatogonial stem cells (SSCs). Other transcripts encode GDNF, FGF8 and BMP4, known regulators of mouse SSCs. Thus, most SCO Sertoli cells can neither organize junctions at normal sites of cell-cell contact nor stimulate SSCs with adequate levels of growth factors. We propose that the critical deficits in Sertoli cell gene expression we have identified contribute to the inability of spermatogonial stem cells within small patches of spermatogenesis in some SCO testes to seed spermatogenesis to adjacent areas of tubule that are barren of spermatogenesis. Furthermore, we predict that one or more of these deficits in gene expression are primary causes of human SCO syndrome.

Conflict of interest statement

DAP is affiliated with 'Consulting Research Services, Inc.' This service was founded and funded by DAP. Funds came from his lecture fees. DAP does not receive any financial benefit from this service and the service has no employees. Thus, this service does not present a conflict of interest with the research presented in this manuscript. The authors declare that there are no other conflicts of interest. This does not alter our adherence to PLOS ONE policies on sharing data and materials.

Figures

Fig 1. Comparison of the transcriptomes of testes with complete spermatogenesis with the transcriptomes of testes with SCO syndrome.

(A) K means cluster analysis comparing the transcriptomes of testes with complete spermatogenesis with testes with SCO syndrome. Genes were clustered under 5 Gene Ontology terms; one or two representative terms per cluster are listed here. (B) The abundance in of a few well-characterized transcripts considered markers of three testicular somatic cells: Sertoli cells (VIM), peritubular myoid cells (ACTA2), and Leydig cells (INHA). Also shown are transcripts encoding receptors for three hormones required for normal testis function: FSHR, AR, and LHCGR. (C) Abundance of transcripts expressed by human SSCs (GFRA1, RET, ZBTB16, UCHL1, ENO2), by spermatocytes (SYCP2, SYCP3) and by round spermatids (PRM1, PRM2). Data in panels B and C are presented as CPM values in the total testis transcriptome divided by the CPM value for beta actin in the same sample. Asterisks over a pair of bars identify significant differences between testes with complete spermatogenesis and SCO testes (p

Fig 2. The expression of Sertoli cell signature transcripts in testes with complete spermatogenesis and in testes with SCO syndrome.

(A) PCA analysis of the expression of Sertoli cell signature transcripts in individual testes. Data were obtained from 4 testes with complete spermatogenesis (C1-4, green symbols) and seven SCO testes (SCO5, SCO7-12, red symbols) (B). A comparison of the mean level of expression of each Sertoli cell signature transcript in testes with complete spermatogenesis and in testes with SCO syndrome. Transcripts that were expressed in SCO testes at significantly higher levels (FDR<0.05), the same level or significantly lower levels are identified by blue, white and black dots, respectively.

Fig 3. Expression of Sertoli cell signature transcripts encoding proteins that form adhesive or gap junctions or that organize and polarize the Sertoli cell plasma membrane.

(A) CLDN11 and JAM3 are tight junction adhesive proteins. Nectin2 and CDH2 are adhesive proteins of ectoplasmic specializations. GJB2 encodes the gap junction protein, Connexin 26. (B) Transcripts that encode proteins that organize or polarize the plasma membrane. CRB2 is a cell polarity component protein that defines the site of separation of the apical and basal-lateral plasma membrane domains. TJP3 is an adaptor protein that links the adhesive proteins of a tight junction to the actin cytoskeleton. SPTB, SPTPN4 and SPTPN5 are scaffolding proteins that concentrate beneath the plasma membrane, seed formation of membrane microdomains, and link those domains to the actin cytoskeleton. All data in this figure are expressed as mean + SEM of CPMs. The presence of an asterisk over a pair of bars indicates a significant difference between testes with complete spermatogenesis and testes with SCO syndrome (FDR≤0.05).

Fig 4. Expression of Sertoli cell signature transcripts encoding regulators of vesicular trafficking.

(A) Expression of RRAS2 and HRAS and the catalytic subunit of phosphoinositide 3-kinase (PI3K), PIK3CA. Binding of RRAS2 and HRAS to PI3KCA stimulates its catalytic activity. (B) Expression of regulatory subunits of PI3K. (C) Small GTPase family members that regulate vesicular trafficking. RAB20 regulates endocytic trafficking; RAB40B and RAB3D regulate exocytic trafficking. Activities of these small GTPases are stimulated by PI3K. Data are expressed as mean + SEM of CPMs for each transcript. The presence of an asterisk over a pair of bars indicates a significant difference between testes with complete spermatogenesis and testes with SCO syndrome (FDR0.05).

Fig 5. Expression of Sertoli cell signature transcripts that encode growth factors that are known regulators of SSCs and/or progenitor spermatogonia.

Data are expressed as mean + SEM of CPMs. An asterisk over a pair of bars indicates a significant difference (FDR≤0.05).

Fig 6. Confirmation that FGF8 is a product of human Sertoli cells and verification that expressions of both FGF8 protein and mRNA are reduced in SCO testes.

Testes with complete spermatogenesis (A) and with SCO syndrome (B) were dispersed into single cells and then incubated with antibodies for the Sertoli cell marker SOX9 and for FGF8. Negative control cells (C) were not incubated with antibodies. Cells were then analyzed FACs. These data are representative of three independent analyses. (D) Quantification by real-time PCR of levels of FGF8 mRNA in biopsies of human testes with complete spermatogenesis and in testes with SCO syndrome. Data (mean + SEM; n = 3) were normalized for expression of beta actin in the same sample (p

Fig 7. Summary: Deficits in Sertoli cell gene expression that are associated with specific aspects of human SCO syndrome.

(A) Human testis with complete spermatogenesis. Polarity complex proteins (red) and tight junctions (gray) polarize the plasma membrane thereby generating apical and basal-lateral domains (identified by up and down arrows). Additional tight junctions, ectoplasmic specialization (yellow) and gap junctions (blue and green) are concentrated at sites of cell-cell contact. Vesicular trafficking (VT; tan circles) of adhesive and gap junction proteins facilitates the assembly and disassembly of junctions, controlled by a pathway that includes Ras GTPases, PI3K and RAB small GTPases (purple shading). Sertoli cells express normal amounts of GDNF, FGF8 and BMP4, growth factors that regulate replication and differentiation of SSCs. (B) The infertile SCO testis. Inadequate expression of polarity and adaptor proteins renders Sertoli cells incapable of properly polarizing their plasma membranes. There is reduced, inefficient vesicular trafficking of proteins that form tight junctions, ectoplasmic specializations and gap junctions, stranding these proteins in vesicles within the cytoplasm. Finally, there is reduced expression of GDNF, FGF8 and BMP4. (C) The effects of GDNF, FGF8 and BMP4 on SSCs and progenitor spermatogonia. GDNF and FGF8 promote self-renewing replication of SSCs, while BMP4 stimulates differentiation of SSCs into progenitor spermatogonia and progenitors into fully differentiated B spermatogonia.