SEL1L regulates ER homeostasis in Sertoli cells but is dispensable for their function

Abstract

Endoplasmic reticulum (ER)–associated protein degradation (ERAD) plays a vital role in maintaining ER homeostasis by degrading misfolded ER proteins. The SEL1L–HRD1 complex, the most evolutionarily conserved branch of ERAD, has been implicated in various physiological processes in both mice and humans, including cellular stress responses, immune function, and development. However, its role in Sertoli cells, which are critical for supporting spermatogenesis, remains unexplored. Here, we show that Sertoli cell SEL1L is not essential for their function or spermatogenesis. SEL1L and HRD1 proteins are expressed in Sertoli cells, and the deletion of SEL1L in Sertoli cells reduces HRD1 protein levels and impairs ERAD function. This leads to elevated ER stress responses and increased expression of ER chaperones, suggesting a potential compensatory adaptation to maintain ER homeostasis. Despite these changes, Sertoli cell-specific Sel1L deletion does not disrupt testicular histology, sperm count, or male fertility. These findings reveal the adaptation of Sertoli cells to SEL1L and ERAD dysfunction and highlight their ability to sustain spermatogenesis under ER stress.

• ERAD is crucial for ER quality control, yet its role in Sertoli cells and its impact on spermatogenesis remain unclear.

• Using a Sertoli cell-specific Sel1L knockout mouse model, we demonstrate that SEL1L and HRD1 are expressed in Sertoli cells, and that SEL1L deletion disrupts ERAD, causing ER stress. However, spermatogenesis remains unaffected due to compensatory ER chaperone upregulation.

• These findings reveal the adaptation of Sertoli cells to ERAD dysfunction, highlight their ability to sustain spermatogenesis under ER stress, and contribute to a broader understanding of ER homeostasis in physiology.

INTRODUCTION

Endoplasmic reticulum (ER)–associated protein degradation (ERAD) is a crucial quality-control mechanism that maintains protein homeostasis by identifying misfolded proteins within the ER, retrotranslocating them to the cytosol, and targeting them for proteasomal degradation (Guerriero and Brodsky, 2012; Qi et al., 2017). Suppressor/enhancer of Lin-12-like (SEL1L) is an adaptor protein for the E3 ligase hydroxymethylglutaryl reductase degradation protein 1 (HRD1), and together they form a complex that is evolutionarily conserved from yeast to humans and serves as a core component of the ERAD machinery (Sundaram and Greenwald, 1993; Hampton et al., 1996; Plemper et al., 1999; Gardner et al., 2000; Carvalho et al., 2006; Mueller et al., 2006; Mueller et al., 2008; Burr et al., 2011; Christianson et al., 2012). It has been reported to play crucial roles in regulating metabolism, immunity, and hematopoietic cell differentiation in various tissues and cell types (Sun et al., 2014; Wu et al., 2014; Sun et al., 2015b; Shi et al., 2017; Bhattacharya et al., 2018; Kim et al., 2018; Wei et al., 2018; Bhattacharya and Qi, 2019; Liu et al., 2020; Shrestha et al., 2020; Xu et al., 2020; Zhou et al., 2020; Kim et al., 2021; Wei et al., 2021; Yoshida et al., 2021; Sim et al., 2022; Thepsuwan et al., 2023; Song et al., 2024). However, the role of SEL1L–HRD1 ERAD in male germ cell differentiation during spermatogenesis remains unexplored.

Sertoli cells, a somatic cell type within the seminiferous tubules of the testes, support male germ cell differentiation by providing structural scaffolding, creating a nurturing environment, and regulating hormonal signals critical for spermatogenesis and testicular function (Franca et al., 2016; O'Donnell et al., 2022). Various secretory and membrane proteins are produced in the ER of Sertoli cells to ensure proper spermatogenesis (Griswold, 1988; Mruk and Cheng, 2004; Kopera et al., 2010). Modulation of ER homeostasis in Sertoli cells appears crucial for their function, as disruptions in this process have been linked to impaired germ cell differentiation and infertility under pathological conditions (Tabuchi et al., 2006; Gong et al., 2009; Zheng et al., 2018; Ge et al., 2022). Despite the importance of ER homeostasis, the role of the SEL1L–HRD1 EARD complex in Sertoli cells during spermatogenesis remains unknown.

In this study, we investigate the function of the SEL1L–HRD1 ERAD pathway in Sertoli cells by generating the cell type-specific Sel1L knockout mouse model. Our findings show that, while the absence of SEL1L in Sertoli cells disrupts HRD1-mediated ERAD, this dysfunction surprisingly has no impact on male fertility, sperm count, or testicular histology. Our data suggest that compensatory mechanisms may support normal spermatogenesis despite impaired ERAD function.

RESULTS

Expression of SEL1L and HRD1 proteins in Sertoli cells in mouse testes

Analysis of previously published single-cell RNA-sequencing data revealed that the transcripts of Sel1L and Hrd1, as well as their ER lectin cofactors Os9 and Xtp3b, are expressed at relatively higher levels in Sertoli cells compared with spermatogenic cells (Figure 1A) (Green et al., 2018). To assess the expression of SEL1L and HRD1 proteins, we performed immunofluorescent staining using vimentin, a Sertoli cell-specific intermediate filament protein, as a marker. In adult mice, SEL1L and HRD1 proteins were observed in vimentin⁺ Sertoli cells, as well as other spermatogenic cells in the seminiferous tubules at different cyclic stages (Figure 1, B and C). Similarly, the ER chaperone glucose-regulated protein 94 (GRP94) was expressed in both Sertoli and spermatogenic cells (Figure 1D). Together, these results demonstrate that SEL1L–HRD1 ERAD is expressed in both Sertoli and spermatogenic cells, indicating potential physiological functions during spermatogenesis.

FIGURE 1:

SEL1L and HRD1 are expressed in Sertoli cells. (A) Heatmap showing the re-analysis of single-cell RNA-sequencing data from mouse testes (GSE112393), illustrating differentially expressed ERAD genes in various cell types. (B–D) Immunofluorescence staining of testis sections from 7-wk-old WT mice with insets of higher magnification shown below. Blue arrowheads point to vimentin⁺ Sertoli cells. Representative data from n = 4 mice per cohort.

Generation of Sertoli cell-specific Sel1L knockout mice

To investigate the role of SEL1L–HRD1 ERAD in spermatogenesis, we generated a Sertoli cell-specific Sel1L knockout mouse model (Sel1L^flox/flox;Amh-Cre, KO^Amh) by crossing Sel1L^flox/flox mice with transgenic mice expressing Cre recombinase under the control of Sertoli cell-specific promoter Amh (Amh-Cre) (Lecureuil et al., 2002) (Figure 2, A and B). Wild-type (Sel1L^flox/flox or Sel1L^flox/+, combined as WT^Amh) and Sel1L heterozygous (Sel1L^flox/+;Amh-Cre, HET^Amh) littermates were included as controls. Western blot analysis of whole testes, which contain a heterogeneous mix of cells, showed that SEL1L protein levels were reduced by ∼40% in KO^Amh testes compared with WT^Amh controls, with HET^Amh testes showing intermediate SEL1L protein levels (Figure 2C). Immunofluorescence staining revealed a specific loss of SEL1L protein in vimentin⁺ Sertoli cells, but not other spermatogenic cells, of KO^Amh mice (Figure 2D; Supplemental Figure S1). Thus, we generated a Sertoli cell-specific Sel1L knockout mouse model.

FIGURE 2:

Generation of Sertoli cell-specific Sel1L knockout mice. (A) Diagram illustrating the generation of Sertoli cell-specific Sel1L knockout mice (KO^Amh), wild-type (WT^Amh), and heterozygous (HET^Amh) littermates. Representative genotyping results are shown in (B). (C) Western blot analysis of testes from 7-wk-old WT^Amh, HET^Amh, and KO^Amh littermates, with quantitation of SEL1L normalized to HSP90 shown below. N = 3 per cohort. (D) Immunofluorescence staining of testis sections from 7-wk-old WT^Amh and KO^Amh littermates, showing SEL1L expression in vimentin⁺ Sertoli cells of WT^Amh mice and its specific ablation in KO^Amh mice. Insets of higher magnification are shown on the right. White dotted lines indicate individual Sertoli cells. The average immunofluorescent signal intensity of SEL1L in individual Sertoli cells is quantified on the right. Negative control staining (−ve CON) of SEL1L using normal rabbit serum in vimentin⁺ Sertoli cells is quantified as well. Representative images from n = 4 mice per cohort. For quantitation, a total of 40 Sertoli cells from three individual mice were quantitated for each cohort. More representative images are shown in Supplemental Figure S1. Values, mean ± SD. **, p < 0.01; ***, p < 0.001; ns, not significant by one-way ANOVA with Tukey multiple comparison tests.

Sel1L ablation leads to reduced HRD1 protein and ERAD defects in Sertoli cells

We next examined the effect of SEL1L ablation on HRD1 expression and ERAD function. Although Western blot analysis of whole testes showed comparable HRD1 protein levels across all three cohorts (Figure 3A), immunofluorescent staining revealed a specific reduction of HRD1 protein in Sertoli cells, but not other spermatogenic cells, in KO^Amh testes (Figure 3B; Supplemental Figure S2). To assess HRD1 activity and ERAD function, we measured the protein levels of two known ERAD substrates, inositol requiring enzyme 1 (IRE1α) and ER lectin OS9 (Tyler et al., 2012; Sun et al., 2014; Sun et al., 2015b). Consistent with a deficiency of SEL1L and HRD1, we observed an accumulation of both IRE1α and OS9 proteins in the testes of young and old KO^Amh mice, independent of transcriptional induction (Figure 3, C–E). HET^Amh testes did not show significant deficiency in HRD1 ERAD activity (Figure 3, C and D). These data demonstrate that SEL1L deficiency in Sertoli cells leads to impaired HRD1 ERAD function, consistent with our previous reports in other tissues (Sun et al., 2014; Sun et al., 2015a; 2015b; Shi et al., 2017; Bhattacharya et al., 2018; Yoshida et al., 2021).

FIGURE 3:

Sertoli cell-specific Sel1L ablation leads to HRD1 ERAD dysfunction. (A) Western blot analysis of testes from 7-wk-old WT^Amh, HET^Amh, and KO^Amh littermates, with quantitation of HRD1 normalized to HSP90 shown below. N = 3 per cohort. (B) Immunofluorescence staining of testis sections from 7-wk-old WT^Amh and KO^Amh littermates, showing the specific reduction of HRD1 in vimentin⁺ Sertoli cells of KO^Amh mice. Insets of higher magnification are shown on the right. White dotted lines indicate individual Sertoli cells. The average immunofluorescent signal intensity of HRD1 in individual Sertoli cells is quantified on the right. Negative control staining (−ve CON) of HRD1 using normal rabbit serum in vimentin⁺ Sertoli cells is quantified as well. Representative images from n = 4 mice per cohort. For quantitation, a total of 40 Sertoli cells from three individual mice were quantitated for each cohort. More representative images are shown in Supplemental Figure S2. (C and D) Western blot analysis of ERAD substrates in testes from 7-wk-old (C) and 8-mo-old (D) WT^Amh, HET^Amh, and KO^Amh littermates. Quantitation normalized to HSP90 shown below. N = 6 for 7-wk-old cohorts, n = 3 for 8-mo-old cohorts. (E) qPCR analysis of testes from 7-wk- and 8-mo-old WT^Amh and KO^Amh littermates. N = 6 for 7-wk-old cohorts, n = 4 for 8-mo-old cohorts. Values, mean ± SD. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ns, not significant by one-way ANOVA with Tukey multiple comparison tests in A to D, and by two-tailed Student's t test in E.

Sertoli cell-specific Sel1L ablation does not affect male fertility or spermatogenesis

All KO^Amh mice survived to adulthood without any visible abnormalities or growth defects, similar to their WT^Amh and HET^Amh littermates (Figure 4A). To assess fertility, WT^Amh, HET^Amh, and KO^Amh males of varying ages were bred with 2- to 4-mo-old wild-type females. KO^Amh males demonstrated normal fertility, producing litters of typical size at the ages of 2 to 8 mo (Figure 4B). Testes from KO^Amh mice exhibited comparable weights and histological features across various stages of the seminiferous epithelial cycle (Figure 4, C–E). Furthermore, sperm count in the cauda epididymis and epididymal histology were unchanged in KO^Amh mice compared with the other two control cohorts (Figure 4, F and G).

FIGURE 4:

Sel1L deficiency in Sertoli cells does not affect fertility. (A) Growth curve of WT^Amh, HET^Amh, and KO^Amh littermates. N = 9 for WT^Amh and HET^Amh, n = 10 for KO^Amh. (B) Breeding test mating WT^Amh, HET^Amh, and KO^Amh males of varying ages with 2- to 4-mo-old wild-type females. The number of pups born per litter is shown. N = 5 for WT^Amh and HET^Amh, n = 7 for KO^Amh. (C and D) Weight (C) and normalized weight (D) of testes from 7-wk- and 8-mo-old cohorts. N = 4 to 9. (E) Representative images of H&E-stained testis sections from 7-wk-old WT^Amh and KO^Amh littermates, showing various cell types at different stages of the seminiferous epithelium cycle. Representative images from n = 4 mice per cohort. (F and G) Mature sperm count (F) and H&E staining (G) of epididymis from WT^Amh, HET^Amh, and KO^Amh littermates. N = 4 to 6. Values, mean ± SD. *, p < 0.05; ns, not significant by one-way ANOVA with Tukey multiple comparison tests.

To determine whether SEL1L deficiency in Sertoli cells affects spermatogenesis, we next analyzed the distribution and organization of spermatogenic cells in seminiferous tubules of 7-wk-old mice. immunofluorescence staining was performed using germ cell-specific differentiation markers: germ cell nuclear acidic peptidase (GCNA) for spermatogonia (Figure 5A) (Enders and May, 1994), DEAD box helicase 4 (DDX4) for spermatocytes and round spermatids (Figure 5B) (Hickford et al., 2011; Azizi et al., 2021), and lectin-PNA for round and elongating spermatids (Figure 5C) (Mortimer et al., 1990). The number and distribution of each population across different stages of the seminiferous epithelial cycle were comparable between WT^Amh and KO^Amh littermates. Taken together, these data suggest that SEL1L in Sertoli cells is dispensable for male fertility and spermatogenesis.

FIGURE 5:

Sel1L deficiency in Sertoli cells does not affect spermatogenesis. Immunofluorescence staining of testes from 7-wk-old WT^Amh and KO^Amh littermates, showing (A) GCNA, a germ cell-specific marker; (B) DDX4, a spermatocyte and round spermatid-specific marker; and (C) Lectin PNA, a sperm acrosome marker at different stages of the seminiferous epithelium cycle. Insets of higher magnification shown on the right. Representative images from n = 4 mice per cohorts.

Elevated UPR and expression of ER chaperones in Sel1L-deficient Sertoli cells

The aforementioned findings suggest that, while SEL1L deficiency leads to ERAD defects, it does not impact Sertoli cell function. This prompted us to further investigate mechanistically whether and how SEL1L deficiency affects ER homeostasis in Sertoli cells. We first examined the activation of two unfolded protein response (UPR) pathways, initiated by UPR sensors IRE1α and PKR-like ER kinase (PERK). Both RT-PCR and quantitative PCR (qPCR) analyses of the IRE1α RNase substrate X box-binding protein 1 (Xbp1) mRNA splicing revealed a moderate but significant 1.5- to 3-fold induction of Xbp1 splicing in both young and old KO^Amh testes compared with control cohorts, indicating an activation of the IRE1α pathway (Figure 6, A–D). Phosphorylation of PERK, along with its mRNA levels, were significantly increased in KO^Amh testes (Figure 6, B, D, E, and F). However, we did not detect the PERK-induced expression of activating transcription factor 4 (ATF4) or induction of its downstream effector C/EBP homologous protein (Chop) mRNA or protein in KO^Amh testes, suggesting a relatively low level of UPR (Figure 6, B, D, and E).

FIGURE 6:

Sertoli cell-specific Sel1L ablation leads to altered ER homeostasis. (A–D) RT-PCR analysis of Xbp1 splicing (A and C) and qPCR analysis (B and D) in testes from 7-wk- and 8-mo-old WT^Amh and KO^Amh littermates. In A and C, u and s denote unspliced and spliced Xbp1 mRNA, respectively. The percentage of Xbp1 splicing is quantitated below the images. In C, male germ GC-4 cells treated with 2 µg/ml tunicamycin (Tuni) for 4 h serve as a positive control. N = 6 in A and B, n = 4 in C and D. (E and F) Western blot analysis of testes from 7-wk- and 8-mo-old cohorts, with quantitation shown below. In E, GC-4 cells treated with tunicamycin for 24 h serve as a positive control. N = 3. (G) TUNEL staining (green) with nuclei stained with DAPI in testes from 7-wk-old WT^Amh and KO^Amh littermates. White arrowheads, TUNEL-positive cells. The cell death index represents the percentage of seminiferous tubules containing at least one apoptotic cell, quantified from low-magnification whole-testis sections of n = 8 mice per cohort. Values, mean ± SD. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ns, not significant by two-tailed Student's t test in A to D, and by one-way ANOVA with Tukey multiple comparison tests in E to G. (H and I) Immunofluorescence staining of testis sections, showing the induction of GRP94 and BiP in vimentin⁺ Sertoli cells (blue arrowheads) of KO^Amh mice. Insets of higher magnification are shown on right. Representative data from n = 4 mice per cohort.

Despite this moderate UPR activation, there was no evidence of elevated cell death in the testes of KO^Amh mice, as indicated by an absence of cleaved caspase-3, a marker of apoptosis, and negative TUNEL staining (Figure 6, E and G). This is in contrast with the pharmacologically induced excessive ER stress in male germ GC-4 cells treated with tunicamycin, which led to extensive Xbp1 splicing, PERK phosphorylation, and apoptosis marked by elevated caspase-3 cleavage (Figure 6, C and E).

These seemingly paradoxical findings led us to investigate whether Sertoli cells adapt to SEL1L deficiency by upregulating the expression of ER chaperones. The total transcript levels of key ER chaperones, including binding immunoglobulin protein (BiP) and Grp94, were mildly elevated in KO^Amh testes compared with the control cohorts, respectively (Figure 6, B and D). Immunofluorescence staining revealed a striking upregulation of both BiP and GRP94 proteins specifically in Sertoli cells following SEL1L deficiency (blue arrowheads, Figure 6, H and I). Notably, BiP expression was markedly induced throughout the cytoplasm of KO^Amh Sertoli cells, extending from the basal lamina to the lumen of the seminiferous tubules (blue arrowheads, Figure 6I). Thus, SEL1L deficiency in Sertoli cells induces the expression of ER chaperones without triggering cell death.

Ultrastructural analysis of various cell types in testes

Finally, we performed transmission electron microscopy (TEM) to assess ultrastructural changes. In both WT^Amh and KO^Amh testes, the ER in Sertoli cells appeared as sheets (yellow arrowheads, Figure 7A). In line with a moderate UPR response, no noticeable differences were observed in the morphology or abundance of ER, nor in other organelles such as mitochondria and lipid droplets, comparing Sertoli cells in WT^Amh and KO^Amh mice (Figure 7A). Given the importance of Sertoli cells in nursing spermatogenic cells, we also examined the ultrastructure of other cell types in seminiferous tubules. We failed to observe any significant changes in KO^Amh mice, including but not limited to, ER sheets and mitochondria in spermatocytes (Figure 7B), acrosome caps and adjacent Golgi in round spermatids (Figure 7C), and condensed nuclei, overlaying acrosomes, and axonemes encased by mitochondria sheaths in elongating spermatids (Figure 7D). Taken together, these data demonstrate that SEL1L–HRD1 ERAD deficiency is well-tolerated by Sertoli cells, possibly compensated by a modest activation of the UPR, upregulation of ER chaperones, and potentially other pathways.

FIGURE 7:

Sel1L deficiency does not affect the ultrastructure of Sertoli and spermatogenic cells. TEM images of testes from 7-wk-old WT^Amh and KO^Amh littermates, showing Sertoli cells with prominent nuclei on basement membrane (A), primary spermatocytes with large rounded nuclei (B), round spermatids with the cap-shaped acrosome (C), and elongating spermatids with condensed chromatin and elongated acrosome (D). Green dotted lines illustrate individual cells, and insets of higher magnification are numbered and shown below. Yellow arrowheads, ER; blue arrows, acrosome; red arrows, axoneme. N = 2 per cohort.

DISCUSSION

In this study, we investigated the role of SEL1L–HRD1 ERAD in Sertoli cells and uncovered its dispensable function in supporting spermatogenesis. SEL1L and HRD1 proteins were both expressed in Sertoli cells, and the loss of SEL1L resulted in reduced HRD1 protein levels and impaired ERAD function. However, in the absence of SEL1L–HRD1 ERAD, Sertoli cells were able to restore ER homeostasis and maintain spermatogenesis, potentially through compensatory upregulation of ER chaperones and other adaptive mechanisms. This possible adaptive response needs to be experimentally tested and may underscore the remarkable plasticity of Sertoli cells in managing ER stress and preserving their critical functions in nurturing germ cells, facilitating germ cell movement, and maintaining immune privilege.

Our study reveals a surprisingly dispensable role of SEL1L–HRD1 ERAD in Sertoli cells despite its well-established importance in maintaining ER homeostasis and protein quality control in many other cell types (Sun et al., 2014; Wu et al., 2014; Sun et al., 2015b; Shi et al., 2017; Bhattacharya et al., 2018; Kim et al., 2018; Wei et al., 2018; Bhattacharya and Qi, 2019; Liu et al., 2020; Shrestha et al., 2020; Xu et al., 2020; Zhou et al., 2020; Kim et al., 2021; Wei et al., 2021; Yoshida et al., 2021; Sim et al., 2022; Thepsuwan et al., 2023; Song et al., 2024). This unexpected result provides new insights into the cellular mechanisms that support Sertoli cell function and point to several important avenues for further investigation. Our results are particularly surprising considering a previous study on a key ERAD cofactor ubiquitin-conjugating enzyme E2 J1 (UBE2J1), whose deficiency led to male infertility (Koenig et al., 2014). UBE2J1 is a critical component of the ERAD machinery, functioning as an ER-anchored E2 enzyme for the ubiquitination of misfolded proteins (Mueller et al., 2008). The infertility observed in Ube2j1 deficient male mice highlights an essential role of ERAD in certain reproductive contexts. In contrast, the normal fertility in our Sertoli cell-specific Sel1L-knockout model suggests that Sertoli cells may compensate for the loss of SEL1L–HRD1 ERAD by utilizing alternative ERAD components or different protein quality-control pathways to preserve their function. The precise underlying mechanisms remain unclear but may involve compensatory upregulation of parallel pathways, such as ER chaperones, other E3 ubiquitin ligases, or autophagy, to maintain ER homeostasis. Understanding the functional redundancy and interplay among these pathways in Sertoli cells, which requires in vivo experimentation through the generation of compound knockout mouse models, will be crucial to resolving this apparent paradox and uncovering the full spectrum of mechanisms that support spermatogenesis.

Our findings align with observations in certain other cell types, such as gut epithelial cells and macrophages, where SEL1L–HRD1 ERAD deficiency does not lead to overt phenotypes under basal conditions (Sun et al., 2015b; Ji et al., 2023). In these cells, enhanced expression of ER chaperones and/or adoption of alternative protein quality-control mechanisms help maintain ER homeostasis and cellular function. In contrast, in adipocytes, SEL1L deficiency results in intracellular accumulation of lipoprotein lipase and systemic postprandial hyperlipidemia, demonstrating a clear disruption of cell function without causing cell death (Sha et al., 2014; Wu et al., 2023). These observations suggest that the cellular response to SEL1L deficiency is highly context-dependent, driven by cell type-specific functions and substrate specificity. In SEL1L-deficient Sertoli cells, the elevated ER stress is likely driven by an accumulation of misfolded substrate proteins in the ER that are normally targeted for degradation by the SEL1L–HRD1 ERAD pathway. Nevertheless, the absence of SEL1L–HRD1 ERAD may spare the essential functions of Sertoli cells, possibly due to a relatively low burden of misfolded proteins or the presence of alternative quality-control pathways. The Sertoli cell-specific SEL1L–HRD1 ERAD substrates remain unknown and warrant further investigation. This cell type- and substrate-specific effect of SEL1L deficiency underscores the complexity of ERAD function across different physiological contexts.

Recent studies have highlighted the interplay between SEL1L–HRD1 ERAD and autophagy in maintaining cellular homeostasis, particularly in pancreatic β cells and white adipocytes (Shrestha et al., 2023; Wu et al., 2023). In these cells, ERAD and autophagy cooperate to regulate the degradation of misfolded proteins and protein aggregates, ensuring metabolic and cellular function (Shrestha et al., 2023; Wu et al., 2023). Notably, autophagy has been identified as a critical pathway in Sertoli cells, with its disruption leading to cytoskeletal disorganization and impaired germ–cell interactions (Liu et al., 2016). Our findings suggest that SEL1L–HRD1 ERAD is dispensable for Sertoli cell function under basal conditions, potentially due to compensatory upregulation of autophagic activity. However, the extent and dynamics of this cross-talk in Sertoli cells remain unknown, particularly under pathological or stress conditions associated with elevated ER stress, such as exposure to environmental toxins (Tabuchi et al., 2006; Gong et al., 2009; Zheng et al., 2018; Ge et al., 2022). Future studies are needed to investigate the coordinated roles of ERAD and autophagy in Sertoli cells, particularly in maintaining ER homeostasis, cytoskeletal integrity, and the support of germ cells, which represents an important direction requiring in vivo studies conducted in a Sertoli cell-specific manner.

MATERIALS AND METHODS

Generation of Sertoli cell-specific Sel1L knockout mice

Sertoli cell-specific Sel1L knockout mice were generated by breeding Sel1L^flox/flox females (Sun et al., 2014) with male mice expressing Sertoli cell-specific Amh promoter driven Cre (Amh-Cre, a gift from Dr. Florian Guillou) (Lecureuil et al., 2002). Genotyping primers for Amh-Cre are, forward: 5′-GAACCTGATGGACAT GTTCA GG-3′; reverse: 5′-AGTGC GTTCGAACGCTAGAGCCTGT-3′. Mice were fed a standard rodent chow diet and housed in a temperature-controlled environment with 12 h light/dark cycles with free access to food and water. Mice were killed by cervical dislocation under isoflurane-induced anesthesia. Tissues were collected and immediately either fixed for histology analysis or frozen in liquid nitrogen.

Assessment of fertility and spermatozoa counting

For the breeding test, each 2-mo-old male was housed with one wild-type female continuously for 6 mo. Cages are checked and recorded daily for pregnancy and pups per delivery. For spermatozoa counting, the cauda epididymis was collected, incised, and incubated in 1 ml warm PBS for 10 min to release the sperm. The sperm numbers were counted using a hemocytometer chamber.

Western blot and quantitation

Preparation of cell and tissue lysates and Western blotting was performed as described previously (Sun et al., 2015b). The following antibodies were used in this study: SEL1L (Abclonal customized) (Wei et al., 2024), HRD1 (Abclonal customized, E15102) (Mao et al., 2024), OS9 (Abcam, ab109510), IRE1α (Cell Signaling Technology [CST], 14C10), PERK (CST, 3192), ATF4 (CST, 11815), CHOP (CST, L63F7), Caspase-3 (CST, 9662), and HSP90 (Santa Cruz Biotechnology, sc7947). Secondary antibodies, goat anti-rabbit IgG HRP (Bio-Rad), were used at 1:6000 dilution. Western blot membranes were developed using the Clarity Western ECL Substrate (Bio-Rad), and the signal was detected with a ChemiDOC imager (Bio-Rad). Image LAB software was used to quantify the band density.

TEM

Mice were anesthetized and perfused with perfusion fixation buffer containing 2.5% glutaraldehyde, 2% formaldehyde in 0.1 M sodium cacodylate buffer (electron microscopy sciences). Testis tissues were collected, cut into small pieces (about 1–2-mm cubes), and incubated in primary fixation buffer containing 3% glutaraldehyde, 3% formaldehyde in 0.1 M sodium cacodylate overnight at 4 °C. The tissues were then submitted to the University of Michigan Microscopy Core for washing, embedding, and sectioning at 70 nm. The grid containing tissue sections was imaged by a Talos F200X transmission electron microscope at Wayne State University Lumigen Instrument Center.

RNA extraction, RT-PCR, and quantitative RT-PCR

Testis RNA was isolated as described previously (Coate et al., 2017). Briefly, total RNA was extracted from the testis using RNA-Stat 60 (IsoTex Diagnostics), chloroform, and precipitated by isopropanol. RNA quality was determined by measuring the OD260/280 and visualizing on an agarose gel. cDNA was generated using the Superscript III reverse transcriptase (Thermo Fisher Scientific). RT-PCR for Xbp1 splicing was performed as described previously (Sun et al., 2019) using GoTaq PCR Master Mixes (Promega). qPCR was performed using 2X Universal SYBR Green Fast qPCR Mix (Abclonal). U36B4 was used as the reference. RT-PCR and qPCR primer sequences are listed below:

• Following are for RT-PCR:

mouse Xbp1 (ACACGCTTGGGAATGGACAC, CCATGGGAAGATGTTCTGGG) and

mouse L32 (GAGCAACAAGAAAACCAAGCA, TGCACACAAGCCATCTACTCA).

• Following are for qPCR:

mouse OS9 (AGGAGGAGGAGGAGGAAGAG, TGGGTCCAGTTCCTTCTCTG),

mouse Ire1α (TGAAACACCCCTTCTTCTGG, CAGGGGGACAGTGATGTTCT),

mouse Xbp1s (GAGTCCGCAGCAGGTG, GTGTCAGAGTCCATGGGA),

mouse Perk (TCAAGTTTCCTCTACTGTTCACTCA, CGGGAAACTCCAAGTTCTCA),

mouse BiP (CAAGGATTGAAATTGAGTCCTTCTT, GGTCCATGTTCAGCTCTTCAAA),

mouse Chop (CCAGAAGGAAGTGCATCTTCA, ACTGCACGTGGACCAGGTT),

mouse GRP94 (CTACCAGACACCAAGGCGTA, TGCGTCTTCTGAGGTGTCTT), and

mouse U36B4 (CGTCCTCGTTGGAGTGACA, CGGTGCGTCAGGGATTG).

Histology, immunofluorescence, and TUNEL staining

Testes were fixed in a fixation buffer containing 11.1% formalin, 5% glacial acetic acid, and 15% ethanol overnight at 4°C and processed by the University of Michigan In Vivo Animal Core (Ann Arbor, MI) on a fee-for-service basis. The H&E slides were imaged using an Aperio Scanscope (Leica). For immunofluorescent staining, paraffin sections were rehydrated and boiled for antigen retrieval in sodium citrate buffer (18 mM citric acid and 9.35 mM sodium citrate, pH 6.0) for 20 min, followed by blocking (5% normal donkey serum, 0.5% Tween-20 in PBS) at room temperature for 1 h. The slides were incubated with primary antibodies at 4°C overnight, followed by secondary antibodies at room temperature for 1 h using Alexa fluor 488 or 594 affinipure donkey anti-mouse or rabbit IgG (H + L) (Jackson ImmunoResearch). For negative controls, normal serum of the host animals of the respective primary antibodies was used in place of the primary antibody, followed by the same secondary antibody incubation. Both primary and secondary antibodies were diluted in PBS containing 5% normal donkey serum and 0.5% Tween-20. The nuclei and slides were stained and mounted by prolong diamond antifade reagent with DAPI (Invitrogen P36971). The following primary antibodies were used for staining: SEL1L (Abcam, ab78298), HRD1 (Abclonal customized, E15102), BiP (CST, C50B12), GRP94 (Novus, NBP2-42379), DDX4 (Abcam, ab13840), lectin PNA (Invitrogen, L21409), GCNA (DSHB, 10D9G11), and vimentin (BioLegend, 699301). Fluorescent images were captured using a ZeissLSM880 confocal microscope. Immunofluorescence signal intensity was quantitated using ImageJ software with the freehand tool. TUNEL staining was performed using an in-situ cell death detection kit (Roche, 11684795910) according to the manufacturer's protocol. TUNEL images collected from low-magnification whole-testis sections were quantified by calculating the percentage of seminiferous tubules containing at least one TUNEL-positive cell relative to the total number of seminiferous tubules in the testis, that is, cell death index.

Cell culture

GC-4 cells (a gift from Z.Z.) were cultured in DMEM (Life Technologies) containing 10% FBS and 1% penicillin–streptomycin. For tunicamycin (Tuni, Tocris 3516) treatment, cells were treated with 2 µg/ml Tuni for 4 or 24 h, then collected for RNA or protein extraction. Mycoplasma contamination was tested by PCR monthly.

Re-analysis of single-cell RNA-sequencing data

The published dataset of mouse testes was retrieved from the GEP repository (GSE112393). Cells pooled from 25 sample sets were first filtered based on the number of features at a cutoff threshold of 800 features per cell and subsequently clustered using the Seurat package (version 4.0.1) in the R environment. Cell types were assigned based on the expression of the marker genes defined in the original study (Green et al., 2018). The mean expression values of genes in each cell type were calculated using the AverageExpression function in the Seurat package. The mean expression value for each gene was scaled and centered before being plotted as a heatmap using the ComplexHeatmap package (version 2.8.0) in R.

Statistical analysis

The unpaired two-tailed Student's t test was used for two-group analyses. One-way ANOVA with the Tukey multiple comparison test was used for multigroup analyses (GraphPad Prism). Data were presented as the mean ± SD; a p value of < 0.05 was considered significant.

Study approval

All animal procedures were approved by the Institutional Animal Care and Use Committee of University of Virginia and Wayne State University.

Abbreviations used:

ATF4

Activating transcription factor 4

BIP

Binding immunoglobulin protein

CHOP

C/EBP homologous protein

ERAD

Endoplasmic reticulum-associated protein degradation

GCNA

Germ cell nuclear acidic peptidase

GRP94

Glucose-regulated protein 94

DDX4

DEAD box helicase 4

HRD1

Hydroxymethylglutaryl reductase degradation protein 1

IRE1α

Inositol-requiring enzyme 1

OS9

Osteosarcoma Amplified 9 PERK=PKR-like ER kinase

SEL1L

Suppressor/enhancer of Lin-12-like

TEM

Transmission electron microscopy

UPR

Unfolded protein response

UBE2J1

ubiquitin-conjugating enzyme E2 J1

XBP1

X box-binding protein 1