An essential role for insulin and IGF1 receptors in regulating sertoli cell proliferation, testis size, and FSH action in mice

Abstract

Testis size and sperm production are directly correlated to the total number of adult Sertoli cells (SCs). Although the establishment of an adequate number of SCs is crucial for future male fertility, the identification and characterization of the factors regulating SC survival, proliferation, and maturation remain incomplete. To investigate whether the IGF system is required for germ cell (GC) and SC development and function, we inactivated the insulin receptor (Insr), the IGF1 receptor (Igf1r), or both receptors specifically in the GC lineage or in SCs. Whereas ablation of insulin/IGF signaling appears dispensable for GCs and spermatogenesis, adult testes of mice lacking both Insr and Igf1r in SCs (SC-Insr;Igf1r) displayed a 75% reduction in testis size and daily sperm production as a result of a reduced proliferation rate of immature SCs during the late fetal and early neonatal testicular period. In addition, in vivo analyses revealed that FSH requires the insulin/IGF signaling pathway to mediate its proliferative effects on immature SCs. Collectively, these results emphasize the essential role played by growth factors of the insulin family in regulating the final number of SCs, testis size, and daily sperm output. They also indicate that the insulin/IGF signaling pathway is required for FSH-mediated SC proliferation.

Figure 1.

Expression profiles of Insr, Igf1r_, and their ligands in purified gCs and SCs at P5. A, qRT-PCRs using as template total RNAs from purified populations of immature SCs (GFP⁺ cells) and a GC-enriched fraction (GFP⁻ cells) originating from P5 testis expressing the Sox9-eGFP allele. Note the high expression of Igf1 and Igf2 in male GCs, whereas transcripts coding for Insr, Igf1r are present both in the SCs and the enriched GC fraction. B, qRT-PCR using as template total RNAs from purified populations of immature SCs (GFP⁺ cells) of both control Sox9-eGFP and mutant SC-Insr;Igf1r;Sox9-eGFP mice at P5. Note the massive reduction in transcripts coding for the Insr and Igf1r genes, whereas the expression of control genes such as Amh and Igf2r was not affected. Results are means ± SEM (n = 3/group). C, IF staining for INSR (green) or IGF1R (green) in both control and SC-Insr;Igf1r testes at P5. Note the absence of staining for the 2 receptors specifically in SCs. Arrowheads, SCs; arrows, GCs. Scale bar corresponds to 50 μm. DAPI, 4′,6′-diamidino-2-phenylindole.

Figure 2.

Reduced testis weight and sperm count in SC-Insr, SC-Igf1r, and SC-Insr;Igf1r testes at P180. Photomicrographs (A–D) and histological H&E sections (E–H) of wt (A and E), SC-Insr (B and F), SC-Igf1r (C and G), and SC-Insr;Igf1r (D and H) testes. Testis weights of SC-Insr, SC-Igf1r, and SC-Insr;Igf1r display 13.6%, 34.6%, and 72.4% reductions, respectively, compared with those for wt (I). Similarly, epididymal sperm concentrations were decreased by more than 38% and 58%, respectively, in SC-Igf1r and SC-Insr;Igf1r mutant animals (J). Testicular sperm concentration, a reflect of daily sperm production, was reduced by 72.4% in SC-Insr;Igf1r animals (K). Similarly, total seminiferous tubule's lengths (L) and GFP⁺ SCs (M) were reduced by 75% in SC-Insr;Igf1r animals. In contrast, sperm viability assessed by ethidium monomer staining of the spermatozoa was not affected (N). A fertility assay performed with WT or SC-Insr, SC-Igf1r, and SC-Insr;Igf1r males mated with 2 C57BL/6 females over a period of 6 months (n = 3/genotype) did not reveal any reproductive defects because the number of litters (O) and the average pups per litter (P) were not affected. Values are expressed as means ± SEM. ns, not statistically significant; *P < .05 vs control; **P < .01 vs control; ***P < .001 vs control. Scale bar in H corresponds to 50 μm.

Figure 3.

Reduced proliferation of immature SCs during the late fetal and early neonatal stage in SC-Insr;Igf1r mutants. A, Graph showing testis weight according to developmental stages (postnatal days). B, Graph depicting the number of GFP⁺ SCs isolated from control (square) and SC-Insr;Igf1r (triangle) testes from E15 up to P60. In parentheses are the percentages of reduction at each relevant stages. C, Mean daily rate of increase in GFP⁺ SCs during selected periods. D, Percentage of proliferating SCs at selected developmental stages. Note the significant reduction in the SC proliferative index during the E15.5 to E17.5 and P0 to P5 periods. Double IF using the SC marker GATA4 (red) and the proliferative marker Ki-67staining (green) is shown in both control (E and F) and SC-Insr;Igf1r mice (G and H) at E17.5 (E and G) and P5 (F and H). Values are expressed as means ± SEM. ns, not statistically significant; *P < .05 vs control; **P < .01 vs control; ***P < .001 vs control. Scale bar in H corresponds to 50 μm.

Figure 4.

Seminiferous epithelium displays a transient delay in maturation between P0 and P42 in SC-Insr;Igf1r mutants. H&E staining from wt (A–D, I, and J) and SC-Insr;Igf1r (E–H, M, and N) mutant testes at P0 (A and E), P5 (B and F), P10 (C and G), P15 (D and H), P21 (I and M), and P42 (A, J, and N) reveals that mutant SCs display a delay in their maturation. Periodic acid-Schiff staining (K, L, O, and P) from wt (K) and SC-Insr;Igf1r (O and enlargement in L and P) mutant testes at P42 (A and E) showing a desynchronized progression of GCs as reflected by the simultaneous presence of spermatogenic stages VIII (L) and XII (M) within the same seminiferous tubule. IF against p27 and ZO1 confirm a partial delay of SC maturation (Q–Z). At P15, mutant testes display an absence of P27 staining (compare R with W). This delay is partial because the expression of ZO1 appears unaffected at P15 and P21 (T, U, Y, and Z). Scale bars correspond to 50 μm. DAPI, 4′,6′-diamidino-2-phenylindole.

Figure 5.

Both the morphology and function of LCs and PMCs were not affected in SC-Insr;Igf1r mutant testes. IF for SMA (A and B) and 3β-HSD (C and D) in both control (A and C) and SC-Insr;Igf1r (B and D) testes at P60 revealed that peritubular myoid cells (PMC) were not affected in terms of morphology or number/tubule (E), whereas the density of LCs in the interstitial compartment shows a slight increase in mutant testes (68% vs 61% in wt) (F). Plasma levels of FSH, LH, activin, and testosterone were evaluated in both control and mutant individuals (G–J). Basal testosterone (G), activin (H), and LH (I) levels were not affected, whereas testosterone levels under hCG stimulation were increased by 7.7- and 3-fold in control and SC-Insr;Igf1r individuals (G). FSH levels are 2.5-fold higher in the serum of in SC-Insr;Igf1r mutant animals (J). Values are expressed as means ± SEM. ns, not statistically significant; *P < .05 vs control; **P < .01 vs control; ***P < .001 vs control. Scale bars in B and D correspond to 50 μm. DAPI, 4′,6′-diamidino-2-phenylindole.

Figure 6.

FSH action but not thyroid hormones requires insulin/IGF signaling to regulate SC proliferation and testis growth. Relative testis weight (A, D, and G), testicular reserve (B), epididymal reserve (E and H), and seminiferous tubule length (C and F) from control (wt) and SC-Insr;Igf1r mutant animals at P60 exposed postnatally to 0.1% PTU (A–C) or neonatally hemicastrated (D–F) or FSH treated between P0 and P10 (G and H). qRT-PCRs performed on cDNAs of Sox9:GFP⁺ sorted SCs isolated by fluorescence-activated cell sorting at E17.5, P0, P5, P15, and P60 show a significant decrease in transcripts coding for both Fshr (P) and inhibin β (I) at P5 in SC-Insr;Igf1r mutant animals. IF staining against pan-AKT (J, K, M, and N) and phospho-AKT (L and O) at P5 in both control (J–L) and mutant animals (M–O). Note the drastic reduction of AKT and phospho-AKT signals in mutant compared with wt individuals. Values are expressed as means ± SEM. ns, not statistically significant; *P < .05 vs control; **P < .01 vs control; ***P < .001 vs control. Scale bars in M and O correspond to 50 μm.