If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
To whom correspondence should be addressed: Dr. De-juan Wan, Department of Urology, The Sixth Affiliated Hospital of Sun Yat-Sen University, No. 26 Erheng Road, Guangzhou, Guangdong 510655, P.R.China. Tel: 86-15918683723.
To whom correspondence should be addressed: Dr. Wei Li, Department of Human Anatomy, Histology and Embryology, Air Force Medical University, No. 169 Changle West Road, Xi’an 710032, P.R.China. Tel: 86-29–84774511.
Grim-19 (gene associated with retinoid-IFN-induced mortality 19), the essential component of complex I of mitochondrial respiratory chain, functions as a non-canonical tumor suppressor by controlling apoptosis and energy metabolism. However, additional biological actions of Grim-19 have been recently suggested in male reproduction. We investigated here the expression and functional role of Grim-19 in murine testis. Testicular Grim-19 expression was detected from mouse puberty, and increased progressively thereafter, and GRIM-19 protein was observed to be expressed exclusively in interstitial Leydig cells (LCs), with a prominent mitochondrial localization. In vivo lentiviral vector–mediated knockdown of Grim-19 resulted in a significant decrease in testosterone production and triggered aberrant oxidative stress in testis, thus impairing male fertility by inducing germ cell apoptosis and oligozoospermia. The control of testicular steroidogenesis by GRIM-19 was validated using the in vivo knockdown model with isolated primary LCs and in vitro experiments with MA-10 mouse Leydig tumor cells. Mechanistically, we suggest that the negative regulation exerted by GRIM-19 deficiency-induced oxidative stress on steroidogenesis may be the result of two phenomena: A direct effect through inhibition of phosphorylation of steroidogenic acute regulatory protein (StAR) and subsequent impediment to StAR localization in mitochondria, and an indirect pathway that is to facilitate the inhibiting role exerted by the extracellular matrix (ECM) on the steroidogenic capacity of LCs via promotion of integrin activation. Altogether, our observations suggest that Grim-19 plays a potent role in testicular steroidogenesis and that its alterations may contribute to testosterone deficiency-related disorders linked to metabolic stress and male infertility.
GRIM-19 (gene associated with retinoid-IFN-induced mortality 19), coding for an approximately 16-kDa protein that induces apoptosis in a number of cancer cell lines, has been identified as a cell death activator involved in IFN-beta- and retinoic acid-induced cell death (
). Subsequent study has characterized GRIM-19 as an intrinsical subunit of mitochondrial NADH: ubiquinone oxidoreductase (complex I) in bovine heart, thus pointing out a novel link between the mitochondrion function and cell apoptosis (
). Therefore, GRIM-19 is an indispensable component of mitochondrial complex I and is essential for early embryonic development. Moreover, GRIM-19 has also been reported to be able to suppress constitutive STAT3-induced cellular transformation by down-regulating the expression of a number of cellular genes involved in cell proliferation and apoptosis (
Notably, the biological effects of GRIM-19 known to date are mostly focused on its antiapoptotic potential and pathological relevance, i.e. tumorigenesis. However, additional as yet poorly-characterized physiological actions of GRIM-19 cannot be ruled out. For example, it has been demonstrated that GRIM-19 regulates Ca2+ homeostasis and the Ca2+-dependent NFAT signaling pathway, and its expression is necessary for early heart development in Xenopus. GRIM-19−/− blastocysts display abnormal mitochondrial structure, morphology and cellular distribution as expected (
). Furthermore, somatic missense mutations in GRIM-19 were detected in three of 20 sporadic Hürthle cell carcinomas of the thyroid. The latter report not only provided novel clues to the consistent linkage of increased mitochondrial number and increased cell growth that characterizes Hürthle cell tumors, but also implicated a potential involvement of GRIM-19 in normal steroidogenesis (
). Nevertheless, the functional roles, if any, of GRIM-19 in such peripheral systems remain unexplored.
Testis is a complex endocrine organ in which different cell types interplay in the fine tuning of the reproductive function under the control of a plethora of endocrine, paracrine, and autocrine regulatory signals (
). Among different factors with key roles involved in testicular homeostasis, GnRH (gonadotropin-releasing hormone)-LH (Luteinizing hormone)-testosterone axis plays an essential role in the regulation of testicular function (
). Emerging data of recent years have confirmed that mitochondria act as a key control point for the regulation of steroid hormone biosynthesis since the first and rate-limiting step in steroidogenesis is the transfer of cholesterol across the intermembrane space from the outer mitochondrial membrane to the inner mitochondrial membrane, a process dependent on the action of steroidogenic acute regulatory protein (StAR) (
). △ψm, mitochondrial ATP synthesis, and mitochondrial pH are all required for acute steroid biosynthesis, suggesting that mitochondria must be energized, polarized, and actively respiring to support Leydig cell (LC) steroidogenesis (
). The identification of GRIM-19 as an essential component of complex I for maintenance of normal mitochondrial membrane potential prompted us to evaluate whether this signal is expressed in murine testis. Our current data are suggestive of a possible involvement of GRIM-19 signaling in the direct control of gonadal function in male mice, underscoring an unexpected reproductive facet of this conventional tumor suppressor.
Characterization of Grim-19 expression in mammalian testis
Testicular expression profile of Grim-19 at different stages of postnatal development was first explored in mice. Testes were obtained from 5-, 14-, 28-, 45-, 50-, and 70-day-old mice (n=7/group, Fig. 1A), corresponding to the initiation of spermatogenesis (5 d), appearance of pachytene spermatocytes (14 d), appearance of spermatids (21 d), beginning of puberty (28 d), appearance of adult-staged adult LCs (45 d), beginning of adulthood (50 d), and adult (70 d) stages of postnatal maturation (
). Persistent expression of Grim-19 mRNA was detected from puberty (28 d), and increased progressively thereafter, with maximum values observed in adulthood (70-day-old) samples (Fig. 1B). Immunoblotting analysis verified such an expression profile at protein level (Fig. 1C, Supplementary Table 1). Subsequently, the pattern of testicular expression of GRIM-19 peptide in developing testis was evaluated by immunofluorescence (IF), which demonstrated an exclusive distribution of GRIM-19 immunoreactivity within testicular interstitium (Fig. 1D). Specificity of GRIM-19 immunostaining was confirmed by pre-absorption of the primary antibody or using nonspecific rabbit IgG; a procedure that completely blocked labeling of testis sections (inserted panel in Fig. 1D, SF. 1). Additionally, GRIM-19 protein localized to LCs was visualized by MTA3 (a specific LC marker)-positive expression in adult testis using double-labeling IF (Fig. 1E) (
). The predominant expression of GRIM-19 in rodent LCs was further confirmed by immunoblotting in purified testicular cells (Fig. 1F), as well as by analysis of GRIM-19 expression using rat testis sections following selective elimination of mature LCs by systemic administration of the cytotoxic drug ethane dimethane sulphonate (EDS) (SF. 2) (
), so it is important to determine where our protein of interest resides. Immunofluorescence revealed a puncate distribution of GRIM-19 in the cytoplasm of primary LCs, which interestingly, was overlapped with Mitotracker labeling in these cells (Fig. 1G). We also detected a significant amount of GRIM-19 protein in the mitochondrial fraction, along with the well-known mitochondrial protein Cytochrome c (
), whereas GRIM-19 protein was weakly expressed in the cytoplasmic fraction and undetectable in the nuclear fraction of LCs (Fig. 1H). These findings, in accordance with previous reports which demonstrate that GRIM-19 contains several mitochondrial localization signals and acts as an essential component of mitochondrial complex I (
), suggest that the biological roles of GRIM-19 are likely to be based on its function in mitochondria.
Indispensible involvement of Grim-19 in testicular steroidogenesis and spermatogenesis
To elucidate the biological roles of Grim-19, we employed a previously validated lentiviral microinjection approach to inhibit Grim-19 expression in vivo. About 20 μl of pLV-Grim-19 sh-eGFP or scramble shRNA (7 ng of viral capsid proteins/μl) were microinjected into testicular interstitium using a fiber optics probe under a dissecting microscope. Mice were then allowed to recover, and were sacrificed at d 71 after shRNA injection (Fig. 2A). This timepoint was chosen because one cycle of murine spermatogenesis consists of 35 days (
), we found that lentivirus-mediated bioluminescence was specifically enriched in the testicular interstitium at d 35 and d 70 following microinjection, compared to a negative expression of bioluminescence in Mock controls (Fig. 2B). Relative to Scramble sh-treated testes, treatment with Grim-19 shRNA resulted in a ∼53.8% reduction in the expression levels of Grim-19 mRNA (P<0.05), and Grim-19 sh-mediated inhibition of Grim-19 expression could not be rescued by supplement with exogenous testosterone [testosterone propionate (TP), Fig. 2C]. The effectiveness of inhibition of Grim-19 by shRNA, as well as the inability of exogenous testosterone to rescue the expression of GRIM-19 in shRNA-treated testis, was further confirmed at protein level by immunostaining (Fig. 2D). The apparent inhibition of Grim-19 in LCs made us curious whether Grim-19 deficiency has a role in affecting male fertility, which was then investigated by assessment of a qualitative endpoint (i.e. morphological changes), as well as several quantitative endpoints (i.e. pregnancy, litter size, density and mobility of epididymal sperm, levels of plasma testosterone and testicular apoptosis). Histological examination revealed morphological defects in Grim-19 shRNA-treated testes, including thinning of seminiferous epithelium, desquamation of GCs and scarcity of mature sperm (Fig. 2E). Consequently, the mice treated with Grim-19 shRNA showed remarkable defects in male fertility, relative to their respective Scramble sh-treated counterparts (29.9±6.8 versus 87.6 ± 3.4 pregnancy rate for Grim-19 sh versus Scramble sh, P < 0.01, Fig. 2F; 4.0 ± 2.2 versus 8.8 ± 1.5 litter size for Grim-19 sh versus Scramble sh, P < 0.05, Fig. 2G; 15.3 ± 3.6 versus 29.1 ± 2.7 epididymal sperm×106 for Grim-19 sh versus Scramble sh, P < 0.01, Fig. 2H). The detects in male fertility were probably caused by testosterone deficiency-induced germ cell apoptosis, evidenced by hormonal and biochemical analysis (31.0 ± 6.5 versus 60.9 ± 4.1 plasma testosterone for Grim-19 sh versus Scramble sh, P < 0.01, Fig. 2I; 1.8 ± 0.4 versus 0.2 ± 0.1 germ cell apoptosis for Grim-19 sh versus Scramble sh, P < 0.01, Fig. 2J). The latter hypothesis become more feasible when supplement with TP effectively ameliorated reproductive defects in Grim-19 shRNA-treated mice (Fig. 2E-2J). Thus, Grim-19 inhibition in LCs yields a progressive testis degeneration resulting from a disrupted testicular steroidogenesis.
Validation of the role of Grim-19 in testicular steroidogenesis using in vitro experiments with primary LCs and MA-10 cells
To better understand the molecular events underlying the disruption in testicular steroidogenesis in Grim-19-deficient males, we isolated and purified primary LCs from Scramble sh and Grim-19 sh-treated testes (SF. 3 and Fig. 3A). Primary LCs were then challenged for 12 h with 100 ng/ml LH/hCG. Relative to LCScramble sh, the stimulated testosterone concentration in culture media was significantly reduced in LCGrim-19 sh (6.4±1.8 versus 19.5 ± 1.4 LH treatment for Grim-19 sh versus Scramble sh, P < 0.01; 5.8±1.9 versus 22.4 ±3.9 hCG treatment for Grim-19 sh versus Scramble sh, P < 0.05), whereas ablation of Grim-19 had no effects on steroidogenesis at the resting basal state in both cells (Fig. 3B). To determine whether transient Grim-19 knockdown could also affect testosterone synthesis, we tranfected primary LCs from wild-type testes with Scramble sh and Grim-19 sh (Fig. 3C). In line with the findings obtained from in vivo knockdown model with isolated primary LCs, knockdown of Grim-19 also attenuated the gonadotropin-induced testosterone production in LCs but had no effects on steroidogenesis at the resting basal state (Fig. 3D). To further define functional details of Grim-19, we generated the MA-10Grim-19-/- cells (Fig. 3E). As a relatively homogeneous clonal strain of mouse Leydig tumor cells, MA-10 cells have been proved to be a suitable model system for the study of the regulation of differentiated functions of LCs and gonadotropin actions. MA-10 cells retain many of the properties of LCs except that the production of androgens by these cells is almost undetectable due to lack of 17α-hydroxylase/17,20 lyase/17,20 desmolase expression and activity. Thus, the study on the steroidogenesis by using MA-10 is more precise to be referred as progesterone but not testosterone production (Fig. 3F) (
). Consistent with the results obtained from primary LCs, relative to MA-10Scramble sh, the db-cAMP-stimulated progesterone concentration in culture media was significantly decreased in MA-10Grim-19 sh (6.4±2.0 versus 12.2 ± 1.9 for Grim-19 sh versus Scramble sh, P < 0.05, Fig. 3G). To identify at which step the steroidogenic process was impaired by Grim-19 deficiency, MA-10Grim-19 sh cells were challenged with different stimulus including db-cAMP, 22-ROH or pregnenolone. Intriguingly, a noticeable inhibition of progesterone synthesis was still observed when db-cAMP or 22-ROH was used as a substrate. In contrast, the inhibitory effect of Grim-19 deficiency was totally reversed by the addition of pregnenolone in MA-10Grim-19 sh cells (Fig. 3H). Of note, Grim-19 deficiency had no effects on cell viability and apoptosis, as revealed by MTT and apoptotic ELISA assays (SF. 4). Thus, at least one inhibitory effect of Grim-19 deficiency on the stimulated steroidogenesis appears to occur at the steroidogenic acute regulatory protein (StAR) step.
Indirect control of phosphorylation of StAR by GRIM-19 in stimulated LCs
As further exploration of the core signaling pathways responsible for Grim-19 action in stimulated steroidogenesis, we examined expression levels of several genes known to be essential for Leydig cell biology (Ccnd1, cell cycle progress and proliferation; Hsd3b1, testosterone biosynthesis; Loxl4, metal ion binding and oxidoreductase activity; Cyp11a1, steroid synthesis; Pdk4, ATP binding and pyruvate dehydrogenase kinase activity; Serpina6, lipid and steroid binding; Star, transport of cholesterol from the outer mitochondrial membrane to the inner mitochondrial membrane; Miox, NADP binding, metal ion binding, and oxidoreductase activity; Nr4a1, transcription regulator activity; Bax, pro-apoptotic gene; Bcl-xl, anti-apoptotic gene) (
), in primary LCs isolated from Scramble sh or Grim-19 sh-treated testes. Relative to LCs from Scramble sh-treated testes, Loxl4 and Miox transcripts were significantly increased while Nr4a1 transcripts were noticeably decreased in LCs isolated from Grim-19 sh-treated testes (Fig. 4A). We speculated, therefore, that ablation of Grim-19 may impair steroidogenesis [Nr4a1 is an androgen-regulated gene (
)] and augment oxidative stress, but has no effects on cell proliferation, apoptosis or cellular metabolism of metal ion. We then evaluated the expression levels of steroidogenic enzymes known to be essential for testicular steroidogenesis by means of immunostaining. StAR was the only enzyme that was observed to be significantly down-regulated in Grim-19 sh-treated testes compared to Scramble sh-treated testes (Fig. 4B). Therefore, it is very likely that Grim-19 deficiency may affect post-translational modification of StAR protein in LCs. Upon intracellular pulses of cAMP, StAR is rapidly synthesized and then regulates fundamentally the transfer of cholesterol from outside into mitochondria, thus serving as a key factor mediating the acute steroidogenesis (
). Intriguingly, db-cAMP treatment of Scramble sh or Grim-19 sh-treated MA-10 cells induced the expression of the 3.4-kb precursor form of StAR mRNA as well as the 2.9-kb processed form in both cells (Fig. 4C). Contrarily, db-cAMP induced the expression of the 37-kDa precursor form of StAR protein as well as the 30-kDa processed form in Scramble sh-treated MA-10 cells but failed to do so in Grim-19 sh-treated MA-10 cells (Fig. 4D). To further confirm that the inhibitory effect of Grim-19 deficiency on steroidogenesis occurs at the level of StAR from a functional standpoint, we transiently overexpressed exogenous StAR in the MA-10Grim-19-/- cells (Fig. 4E). Relative to empty vector-transfected MA-10Grim-19-/- cells, overexpression of StAR significantly rescued db-cAMP stimulated progesterone production in the MA-10Grim-19-/- cells, while ablation of Grim-19 or overexpression of StAR had no effects on basal progesterone production in MA-10 cells. By contrast, pretreatment with the broad serine/threonine kinase inhibitor staurosporine (STA) substantially blocked the StAR overexpression-rescued progesterone production in MA-10Grim-19-/- cells (Fig. 4F). Because phosphorylation of StAR protein plays an essential role in the modulation of localization and steroidogenic activity of StAR (
), we sought to determine whether Grim-19 deficiency affects phosphorylation of StAR in stimulated LCs. Treatment of LCsScramble sh with db-cAMP significantly enhanced phosphorylation of StAR in WT LCs, and this stimulatory effect was totally abolished in LCsGrim-19 sh (Fig. 4G). To test whether Grim-19 is directly involved in phosphorylation of StAR, we examined the interaction between StAR and GRIM-19. The reciprocal coimmunoprecipitation experiments in NIH/3T3 cells overexpressing both StAR and GRIM-19 demonstrated that there was no direct interaction between StAR and GRIM-19 (Fig. 4H). Together, GRIM-19 may regulate the phosphorylation of StAR in an indirect manner during gonadotropin-stimulated steroidogenesis.
Indirect control of phosphorylation of StAR by GRIM-19 in stimulated LCs
Given that GRIM-19 is a core subunit of the mitochondrial complex I, we asked whether aberrant oxidative stress was involved in GRIM-19 deficiency-disrupted steroidogenesis. Evidenced by IF, ablation of GRIM-19 in primary LCs significantly enhanced intracellular ROS (DCFH-DA) and mROS (MitoSOX) productions, as well as 8-hydroxydeoxyguanosine (8-OHdG) expression, a biomarker consequent to oxidative stress (
) (Fig. 5A, 5B). Consistently, GRIM-19 deficiency noticeably reduced the NADP+/NADPH ratio but induced the GSH/GSSG ratio in stimulated primary LCs (Fig. 5C). Thus, GRIM-19 deficiency triggers aberrant oxidative stress by ROS generation in LCs. To elucidate the essential involvement of ROS in GRIM-19 deficiency-driven defects of steroidogenesis, we administrated the ROS scavenger-N-acetyl-L-cysteine (NAC) intervention in our in vivo knockdown model, to evaluate whether ROS scavenger could ameliorate GRIM-19 deficiency-disrupted steroidogenesis (Fig. 5D). Intriguingly, supplement with NAC significantly restored epithelial thickness of seminiferous tubule (Fig. 5E), and effectively but partially increased plasma testosterone levels (Fig. 5F) and reduced germ cell apoptosis (Fig. 5G), in Grim-19 sh-treated testes. Similarly in MA-10 cell cultures, co-incubation with NAC markedly but partially improved progesterone synthesis in the presence of db-cAMP stimulation (Fig. 5H). To further clarify the critical role of ROS in GRIM-19 deficiency-driven defects of steroidogenesis, we determined whether ROS inhibition could reverse GRIM-19 deficiency-driven phosporylation of StAR. As expected, NAC cotreatment markedly but partially restored p-StAR levels in db-cAMP-challenged MA-10Grim-19-/- cells (Fig. 5I). Thus, GRIM-19 deficiency, may at least partially, induce aberrant phosporylation of StAR via ROS-dependent manner.
GRIM-19 deficiency potentiates adhesion of integrin-expressing MA-10 cells to distinct extracellular matrix (ECM) components
LCs reside in the testicular interstitium where a large set of locally produced factors (e.g. ECM proteins) act in a paracrime and/or autocrine way to modulate fundamentally steroidogenesis by LCs (
). To test the possibility that GRIM-19 deficiency may alter adhesion of LCs to adjacent ECM proteins, we investigated the ability of the MA-10Grim-19-/- cells to adhere to several EMC proteins. Relative to Scramble sh-treated MA-10 cells, MA-10Grim-19-/- cells tended to unadhere to wells coated with fibronectin, vitronectin or collagen IV, but tended to adhere to wells coated with laminin under db-cAMP-stimulating conditions. By contrast, co-treatment with NAC substantially compromised GRIM-19 deficiency-elicited adhesion to Laminin in db-cAMP-challenged MA-10Grim-19-/- cells (Fig. 6A). Because the subfamily of integrins comprises a monophyletic group of closely related glycoproteins critically involved critically in cell-ECM interactions (
), one can suppose that modifications in expression of these receptors underlie the found difference between the analyzed cell lines. Indeed, immunoblotting showed that Scramble sh-treated MA-10 and MA-10Grim-19-/- cells strongly differ in the spectrum of expressed integrins. Upon db-cAMP challenge, laminin-specific integrin α6 and β1 were significantly down-regulated in Scramble sh-treated MA-10 cells but markedly induced in MA-10Grim-19-/- cells (Fig. 6B, 6C). In agreement, enhancement of collagen-specific integrin α6β1 expression was observed in cell membrane by means of immunofluorescence (Fig. 6D). The observed changes in integrin expression might be considered as a resultant feature related to ROS mechanisms, because co-treatment with NAC totally abolished H2O2-induced α6β1 expression in MA-10 cells (Fig. 6E). To determine a relative contribution of the integrins on LCs to these adhesion and steroidogenesis processes, we examined the effect of antibodies directed against anti-α6 and anti-β1 integrin subunits using MA-10Grim-19-/- cells. As shown in Fig. 6F, MA-10Grim-19-/-cells efficiently migrated toward laminin-coated well and this process depended on both α6 and β1 because anti-α6 and anti-β1 mAbs both blocked the response, while combination of anti-α6 and anti-β1 mAbs completely blocked migration of MA-10Grim-19-/- cells on laminin. Consequently, treatment with anti-α6 or anti-β1 mAbs both effectively improved stimulated steroidogenesis in MA-10Grim-19-/- cells, with the most dramatic rescuing effects been observed in combination of two mAbs (Fig. 6G). By contrast, blockage with anti-α6β1 mAbs had no effects on phosphorylation of StAR protein in db-cAMP-challenged MA-10Grim-19-/- cells (Fig. 6H). Together, inhibition of GRIM-19 induced α6β1 activation, thus controlling cell adhesion to distinct ECM component and negatively affecting steroidogenesis in stimulated LCs.
As an essential component of mitochondrial Complex I, GRIM-19 plays a pivotal role in the regulation of mitochondrial membrane potential and dynamics (
). Previous clinical observations using spermatozoa samples from patients with asthenozoospermia, as well as a recent animal study using Grim-19 knockout mice suggest that Grim-19 is functionally involved in the control of male reproduction, a function that appears to be mediated, at least partially, by its ability to modulate steroidogenesis and intracellular reactive oxygen species (ROS) (
). However, the testicular cell types expressing GRIM-19 and its precise role during spermatogenesis, and its possible interaction with autocrine/paracrine regulatory loop within seminiferous tubules, remain so far unexplored. Our results demonstrate an exclusive expression of Grim-19 in interstitial LCs of rodent testis, and unveil a peripheral regulatory hub, involving GRIM-19-mediated phosphorylation of StAR and ligand-binding properties in stimulated LCs of murine testis, as a critical mechanism regulating testicular steroidogenesis.
Studies in testicular tissues, obtained serially along postnatal development, documented a striking profile of expression of testicular Grim-19, with a progressive increase of both mRNAs and proteins from peripubertal period (postnatal d 28, Fig. 1B and 1C), a profile that appears to precede the elevation of testicular NADH-dehydrogenase (NADH-DH) activity that takes place right after 21 days of age, when rodent enters puberty and spermatids start to appear (
), our findings are compatible with an eventual stimulatory role of Grim-19 on synthesis of testosterone from (early) pubertal maturation onward. This possibility is in keeping with a recent mass spectrometry study in which Grim-19 has been observed to be involved in the formation of macromolecular protein complexes aiding in the regulation and efficiency of steroid-synthesizing MA-10 mouse LCs (
To gain anatomical resolution, immunostaining and immunoblotting studies for mapping the expression of GRIM-19 protein were performed in testicular sections and primary spermatogenic cells respectively, and depicted an exclusive expression of GRIM-19 in interstitial LCs (Fig. 1). Leydig cell-specific expression of testicular GRIM-19 was further verified by immunoblotting analyses after selective elimination of mature LCs by administration of the cytotoxic compound EDS (SF. 2). Of note, EDS treatment provokes a subsequent wave of proliferation and further differentiation of preexisting undifferentiated Leydig cell precursors, which is well known to mimics the normal developmental events of adult-type LCs during puberty (
). Thus, the lack of GRIM-19 expression in testis tissue 7 days following EDS treatment not only confirmed the Leydig cell-specific expression of GRIM-19 but also supported the notion that only LCs at advanced stages of differentiation do express GRIM-19. Intriguingly, subcellular localization of GRIM-19 appears not to be unchangeable under different scenarios. GRIM-19 has been shown to localize predominantly in the nucleus of transfected HeLa cells (
), careful analysis of GRIM-19 localization in our study indicated that this protein was mainly present in the mitochondria of LCs (Fig. 1G). Altogether, this somewhat contradictory evidence regarding GRIM-19 localization within the cell may actually reflect different roles of this protein in cell biology (
). This functional variability may affect its subcellular expression in a very cell type-specific manner.
In agreement with a unique expression pattern of Grim-19 in LCs during testicular development, a lentiviral vector–mediated inhibition of Grim-19 expression in vivo significantly disrupted spermatogenesis by inducing germ cell apoptosis (Fig. 2D and 2E) and oligozoospermia (Fig. 2F-2H), and this deleterious effect was largely ascribed to testosterone deficiency since supplement with TP effectively ameliorated reproductive defects in Grim-19 shRNA-treated mice (Fig. 2E-2J). A tempting explanation for the causal nature of Grim-19 is that it plays a prominent role in mitochondrial oxidative phosphorylation. Indeed, monoallelic loss of Grim-19 causes decline of mitochondrial electron transport machinery viz., respiratory complex (RC)-2, RC-4 and RC-5 and a modest increase in RC-3 levels, thus promoting tumorigenesis in mice (
), the available data open up the possibility that Grim-19 may regulate testosterone synthesis directly by maintaining mitochondrial dynamics. Additionally, in contrast to the general belief that mitochondrial GRIM-19 deficiency antagonizes cell apoptosis, GRIM-19 deficiency had no effects on Leydig cell viability and apoptosis (SF. 4). This is totally understandable since GRIM-19 is a core component of mitochondrial Complex I, whereas ROS production by complex I in physiological state is relatively humble presenting nonlethal oxidative damage to cells (
). Therefore, it is a very logical observation that GRIM-19 deficiency triggers oxidative stress in LCs, and GRIM-19 deficiency-influenced phosphorylation landscape of the StAR protein (Fig. 5I), as well as reconstruction of ECM-induced activation of Integrin expression, can be effectively abrogated by the pharmacological inhibition of ROS (Fig. 6A). Further mechanistic analysis revealed that GRIM-19 deficiency-induced oxidative stress compromised Leydig cell function mainly via modulating two biochemical events. On one hand, GRIM-19 appears to operate as central posttranslational modification link between stimulated steroidogenesis and StAR expression in LCs, by changing the phosphorylation state of the StAR protein. StAR predominantly mediates the rate-limiting step in steroid biosynthesis, i.e., the transport of the substrate of all steroid hormones, cholesterol, from the outer to the inner mitochondrial membrane, thus regulating fundamentally the synthesis of testosterone in LCs (
). The molecular switch from a repressive to a permissive configuration of StAR is seemingly dictated by the hormonal influence on StAR activity at posttranslational level: phosphorylation of StAR not only potentiates its steroidogenic activity (
) in db-cAMP or hCG-stimulated cells. In our study, the db-cAMP-induced phosphorylation of StAR was totally abolished in the presence of Grim-19 depletion (Fig. 4G), and reciprocal coimmunoprecipitation experiments further demonstrated that there was no direct interaction between StAR and GRIM-19 (Fig. 4H). These findings indicate that the phosphorylation of StAR is also subjected to GRIM-19-mediated posttranslational regulation during gonadotropin-stimulated steroidogenesis, probably in an indirect manner. On the other hand, it is gradually recognized association of LCs with ECM initiates an extensive cell matrix cross-talk which regulates fundamentally cell adhension, activation of various second messengers and testicular steroidogenesis (
). As yet, how the response of LCs to ECM is modulated remains ill-defined. We observed that GRIM-19 deficiency-induced oxidative stress may regulate response to ECM components (namely laminin) by controlling expression of distinct transmembrane α and β subunits in LCs. Ablation of GRIM-19 or treatment with H2O2 alone promoted α6β1 activation, whereas application of the ROS scavenger NAC totally blocked α6β1 activation in LCs (Fig. 6B, 6E). In favor of our hypothesis, cell adhesion and integrin expression have been shown to be modulated by oxidative stress in EA.hy 926 cells (
). In addition, on MA-10Grim-19-/- cells, both α6 and β1 integrins contribute to adhesion to selected ECM proteins (i.e. laminin, Fig. 6F). Because laminin has been shown to suppresses progesterone production by human luteinizing granulosa cells via interaction with α6β1integrins (
). Thus, the available data supports the general laminin-α6β1 integrins interacting mechanism by which progesterone production is disrupted in GRIM-19-deficient LCs.
Collectively, our data raise the hypothesis that the negative regulation exerted by GRIM-19 deficiency-induced oxidative stress on steroidogenesis is the result of two phenomena. A direct effect through inhibition of phosphorylation of StAR and subsequent impediment to StAR localization in mitochondria, and an indirect pathway that is to facilitate the inhibiting role exerted by the ECM on the steroidogenic capacity of LCs via promotion of integrin activation (Fig. 7). Future studies will have to be designed to determine the soundness of such a hypothesis. Moreover, because GRIM-19 also functions as a potent regulator of energy metabolism (
Animal work was complied with the institutional guidelines and the criteria outlined in the “Guide for Care and Use of Laboratory Animals” (NIH publication 86-123), and was approved by Institutional Animal Care and Use Committee (IACUC) of the First Affiliated Hospital of Sun Yat-sen University under protocol #SYSU 2018-041-2b and IACUC of Air Force Medical University (Approval #: KY20194015). Pregnant C57BL/6 mice at 14 days of gestation and adult Sprague-Dawley rats at 4 months of age were purchased from the Animal Research Center of Sun Yat-sen University, and were housed under a constant 12 h light:12 h darkness cycle (lights on at 0800 h) and controlled conditions of humidity (between 70 and 80%) and temperature (22±1°C), with free access to pellet mouse chow and tap water. Mice were allowed to acclimatize at least for 7 days before experiments (
). At 5, 14, 21, 28, 45, 50 and 70 days post-partum (dpp), mice were euthanatized by pentobarbital anesthesia [0.04-0.05 mg/g body weight, intraperitoneally (i.p.)] followed by cervical dislocation. Testes were either fixed in Bouin’s/4% paraformaldehyde solution and embedded in paraffin for histological analysis or snap-frozen in liquid nitrogen and stored at -80 °C for biochemical analysis.
In vivo knockdown of Grim-19 was accomplished by using a previously validated lentiviral vector–mediated gene transfer procedure (
). Following pentobarbital anesthesia, testes were exposed under a Leica EZ4 dissecting microscope (Leica, Beijing, China) and were injected with 20 μl of pLV-Grim-19 sh-eGFP or scramble shRNA (7 ng of viral capsid proteins/μl, Inovogen, Chongqing, China) by gentle syringe pressure using a fiber optics probe (diameter, 1.65 mm). The testes were then gently put back into scrotum, muscle and skin layers were sutured accordingly, and animals were allowed to recover (n=7/group). Two days after lentiviral injection, some mice were injected subcutaneously (s.c.) with 3 mg/kg TP (Sigma-Aldrich) every day for consecutive 68 days. Mice were euthanatized at d 71 after lentiviral injection, and testes and blood samples were collected accordingly.
Detailed information on other procedures pertinent to this paper can be found in Supporting Information.
The authors confirm that the data supporting the findings of this study are available within the article [and/or] its supplementary materials.
Conflict of interest
No potential conflicts of interest relevant to this article were reported for each author.
The authors thank Dr. Shun Zhang for technical assistance and Miss. Zhenzhen Chen for secretarial support.
Study design: LW and WD. Study conduct: QH, HK and ZZ. Data collection: QH, HK, ZZ NG, LL, YB and ZW. Data analysis: LW, QH, HK, ZZ and WD. Data interpretation: LW, QH, HK, ZZ and WD. Contribution of reagents: LW and WD. Drafting and submission of manuscript: LW and QH. Research supervision: LW and WD.
This work was supported by the National Natural Science Foundation of China (NSFC): 31971070, Key R&D Programmes of Shaanxi Province: 2022SF-241 and Science and Technology Development Fund of Air Force Medical University: 2022XB002.
CRediT author statement
Hu Qu: Methodology, Validation, Formal analysis, Investigation.
Ke He: Methodology, Validation, Formal analysis, Investigation.