Loss of WNT4 in the gubernaculum causes unilateral cryptorchidism and fertility defects

Abstract

Undescended testis (UDT) affects 6% of male births. Despite surgical correction, some men with unilateral UDT may experience infertility with the contralateral descended testis (CDT) showing no A-dark spermatogonia. To improve our understanding of the etiology of infertility in UDT, we generated a novel murine model of left unilateral UDT. Gubernaculum-specific Wnt4 knockout (KO) mice (Wnt4-cKO) were generated using retinoic acid receptor β2-cre mice and were found to have a smaller left-unilateral UDT. Wnt4-cKO mice with abdominal UDT had an increase in serum follicle-stimulating hormone and luteinizing hormone and an absence of germ cells in the undescended testicle. Wnt4-cKO mice with inguinal UDT had normal hormonal profiles, and 50% of these mice had no sperm in the left epididymis. Wnt4-cKO mice had fertility defects and produced 52% fewer litters and 78% fewer pups than control mice. Wnt4-cKO testes demonstrated increased expression of estrogen receptor α and SOX9, upregulation of female gonadal genes, and a decrease in male gonadal genes in both CDT and UDT. Several WNT4 variants were identified in boys with UDT. The presence of UDT and fertility defects in Wnt4-cKO mice highlights the crucial role of WNT4 in testicular development.

Keywords: Cryptorchidism; Gubernaculum; Infertility; SOX9; Undescended testis; WNT4.

Fig. 1.

WNT4 and retinoic acid receptor β2 (Rarb) expression in the testis and gubernaculum. Rarb-cre mice were crossed with the reporter line Gt(ROSA)26Sor^{tm14(CAG-tdTomato)Hze} to determine Rarb-cre expression in adult tissues. Gubernaculum and testis of four hemizygous mice were sectioned (5 μm) and stained with WNT4 and mCherry antibodies. (A) WNT4 is highly expressed in the adult gubernaculum. (B) WNT4 has low adult testicular expression level with a low expression in the interstitial cells (black arrow). (C,D) mCherry antibody was used to detect the tdTomato reporter that indicated Rarb expression. mCherry was detected in the adult gubernaculum (C) and detected at high levels in the interstitial cells of the testis (black arrow). (E) Gubernaculum negative control. (F) Testis negative control (omission of primary antibody).

Fig. 2.

Wnt4-cKO mice have left UDT. (A,B) Images indicating the testicular position in the abdominal cavity of WT mice with normal testicular descent into the scrotum (A) and Wnt4-cKO mice with right testis in the scrotum and left testis high in the abdominal cavity (B). (C) Left abdominal UDT in Wnt4-cKO is smaller with a longer gubernaculum compared with WT testis. (D) Comparison of testicular weight in WT (blue), Wnt4-cKO with inguinal UDT (red) and Wnt4-cKO with abdominal UDT (green). n=15; *P<0.01 (one-way ANOVA). Box whiskers indicate minimum to maximum, middle line indicates mean, and the error bars indicate s.e.m. Abd, abdominal; G, gubernaculum; Ing, inguinal; L, left; R, right; T, testis.

Fig. 3.

Wnt4-cKO have gubernacular, testicular and epididymal abnormalities. (A-E) Masson's trichrome staining of mouse gubernaculum from different mice indicating muscle (red) and collagen (blue) content. (F-J) Testicular comparison using PAS staining of WT and Wnt4-cKO mice with testes of different positions. (F) WT mice displaying normal ST morphology and distribution. (G) Right-CDT displaying some abnormalities, including the presence of round cells in the lumen of STs. (H) Inguinal-UDT testes of Wnt4-cKO mice displaying some abnormalities, including the presence of round cells in the lumen of STs. (I) Inguinal-UDT testes of Wnt4-cKO mice displaying an MA phenotype. (J) Abdominal-UDT testes of Wnt4-cKO mice displaying an SCO phenotype. (K-O) Epididymal comparison using PAS staining of WT and Wnt4-cKO mice. (K) WT mice displaying abundant sperm in the epididymis. (L,M) Wnt4-cKO R-CDT mice (L) and Wnt4-cKO inguinal L-UDT mice (M) displaying abundant sperm in the epididymis with some isolated round cells. (N,O) Wnt4-cKO inguinal L-UDT mice (N) and Wnt4-cKO abdominal L-UDT mice (O) displaying a few round cells only.

Fig. 4.

Wnt4-cKO mice with abdominal UDT have elevated levels of FSH and LH. (A-C) Comparison of levels of serum testosterone (A), LH (B) and FSH (C) among five mice from each of the following groups: WT, Wnt4-cKO with inguinal UDT (cKO-Ing) and Wnt4-cKO with abdominal UDT (cKO-Abd). *P<0.01 (one-way ANOVA). Box whiskers indicate minimum to maximum, middle line indicates mean, and the error bars indicate s.e.m.

Fig. 5.

Wnt4-cKO mice show fertility defects. (A) Graph showing fertility of male mice after 6 months of mating with WT females. All eight WT males produced one litter per month and were considered fertile (blue bar). Of the eight Wnt4-cKO males, three had left inguinal UDT (red) and five had abdominal UDT (green). All the Wnt4-cKO with inguinal UDT were subfertile. Of the Wnt4-cKO with abdominal UDT, one was infertile and the other four were subfertile. (B) Total number of litters produced per mouse group over a period of 6 months. Each of the eight WT male mice produced six litters for a total of 48 litters in a 6-month period (blue line). The three Wnt4-cKO with inguinal UDT produced a total of 11 litters in a 6-month period (red line), similar to the five Wnt4-cKO with abdominal UDT that produced a total of 13 litters in a 6-month period (green line). All Wnt4-cKO (yellow line) produced a total of 24 litters, which was significantly different (P<0.01; one-way ANOVA) from WT mice. (C) Total number of pups produced per mouse group over a period of 6 months. The eight WT male mice produced a total of 395 pups in a 6-month period (blue line). The three Wnt4-cKO with inguinal UDT produced a total of 47 pups in a 6-month period (red line), similar to the five Wnt4-cKO with abdominal UDT that produced a total of 46 pups in a 6-month period (green line). All Wnt4-cKO (yellow line) produced a total of 93 pups, which was significantly different (P<0.01; one-way ANOVA) from WT mice.

Fig. 6.

Characterization of Wnt4-cKO CDT and UDT in different testicular cell populations. Testes of WT, Wnt4-cKO left inguinal UDT, Wnt4-cKO abdominal right CDT and Wnt4-cKO left abdominal UDT were stained with different testicular cell markers. (A-D) Leydig cells showed 3β-HSD expression in all testes analyzed. (E-L) The spermatogonial markers PLZF (E-H) and CCND1 (I-L) were expressed in all testes analyzed including the abdominal UDT, in which most of the STs exhibited an SCO phenotype. (M-P) The spermatogonial marker KI67 was highly expressed in WT spermatogonia, but not in Wnt4-cKO spermatogonia independent of the testis position.

Fig. 7.

Wnt4-cKO mice have increased ERα and SOX9. Testes of WT, Wnt4-cKO abdominal right CDT, left inguinal UDT (showing normal-appearing ST or MA) and Wnt4-cKO left abdominal UDT were stained for ERα and SOX9. (A-E) Leydig cells showed ERα expression (green arrow) in all testes analyzed, but Wnt4-cKO testes had more ERα-expressing cells than did WT testes. (F-J) Expression of the Sertoli cell marker SOX9 at low and high magnification indicates an increase in the number of SOX9-positive cells in Wnt4-cKO testes independent of the position. (K) Quantification of the number of SOX9-positive cells in WT testes (10.4±4.5) and in Wnt4-cKO inguinal CDT (15.4±7.5) and UDT (21.2±7.5) indicates a significant increase of SOX9-positive cells in Wnt4-cKO mice compared with WT mice (P<0.001; one-way ANOVA).

Fig. 8.

Altered gene expression profile in Wnt4-cKO mouse testes. (A) qPCR analysis of 27 selected genes. Colors indicate gene categories: cell cycle (orange), WNT pathway (yellow), male gonad sex determination and function (blue) and female gonad sex determination and function (pink). Relative gene quantification compared with WT testes (normalized to 1) is indicated for both testes of mice with inguinal UDT and abdominal UDT as well as the corresponding CDT. The numerical quantification is indicated. Genes that were downregulated are lower than 1 and indicated by variants of red color depending on the degree of downregulation. Genes that were upregulated are higher than 1 and indicated by variants of green color depending on the degree of upregulation. Statistical analyses were conducted using the REST Software from QIAGEN to determine significance (*P<0.005). (B) BMP4 western blots normalized to β-actin and quantified using ImageJ of WT left testis and both left and right testis of Wnt4-cKO with inguinal UDT. Graphs show western blot band intensities. Data are average±s.d.

Fig. 9.

Structural mapping of WNT4 variants in boys with GU defects. (A) Sequence alignment of human WNT4 and WNT3 using Jalview. The sequences of the signal peptide, the WNT N-terminal domain (NTD), the linker region, and the elongated C-terminal domain (CTD) are highlighted in gray, purple, green and pink, respectively. The residues that are altered in boys with GU defects are labeled in red. The UniProtKB accession numbers are shown to the left of the sequences. (B) The homology model of WNT4 was superimposed onto the crystal structure of the WNT signaling complex (PDB ID: 6AHY), which consists of human WNT3 and mouse frizzled 8 Cys-rich domain (CRD). For clarity, the structure of WNT3 is not shown, and the Frizzled-8 CRD is shown in blue ribbon representation. Each domain of WNT4 is color coded as in A. The mutated residues of WNT4 are shown as spheres. The inset on the right shows the close-up view of residues affected by variants shown in ball-and-stick models. The atoms are colored with oxygen in red, nitrogen in blue and sulfur in yellow. (C-F) Close-up views of mutations (Y80H, Q167H, R258C, P264Q and Q282R) in WNT4. The WT residues are colored as in A, and the mutant models are shown in gray.

Fig. 10.

Model for testicular dysfunction in Wnt4-cKO mice. (A) In normal conditions, gubernacular WNT4 is required for the testis to descend in the scrotum. In the descended testis, there is limited expression of estrogen (E) and activated estrogen receptors (ERs). Nuclear translocation of SOX9 in Sertoli cells occurs freely, causing the increased expression of its own gene and its downstream targets, such as Dmrt1 and Dhh. (B) In mice lacking WNT4 in the gubernaculum, the left testis does not descend into the scrotum and could be located in the inguinal canal or in the abdominal cavity. The testis of Wnt4-cKO seems to have increased ERα and female gonadal genes, such as WNT4 and FOXL2. Recently, it was shown that estrogen action, through the binding of ERα, regulates SOX9 in Sertoli cells by blocking SOX9 nuclear translocation. A consequent increase in cytoplasmic SOX9 prevents the activation of downstream targets of SOX9 and promotes the expression of WNT4 and FOXL2 (Stewart et al., 2020). We propose that ER activation in Wnt4-cKO testis could inhibit SOX9 nuclear translocation in Sertoli cells and cause a decrease in male genes, such as Dmrt1 and Dhh, and allow for the expression of female genes, such as Wnt4, Rspo1 and Foxl2. P, postnatal day.