Alterations in the steroid hormone receptor co-chaperone FKBPL are associated with male infertility: a case-control study

Abstract

Background: Male infertility is a common cause of reproductive failure in humans. In mice, targeted deletions of the genes coding for FKBP6 or FKBP52, members of the FK506 binding protein family, can result in male infertility. In the case of FKBP52, this reflects an important role in potentiating Androgen Receptor (AR) signalling in the prostate and accessory glands, but not the testis. In infertile men, no mutations of FKBP52 or FKBP6 have been found so far, but the gene for FKBP-like (FKBPL) maps to chromosome 6p21.3, an area linked to azoospermia in a group of Japanese patients.

Methods: To determine whether mutations in FKBPL could contribute to the azoospermic phenotype, we examined expression in mouse and human tissues by RNA array blot, RT-PCR and immunohistochemistry and sequenced the complete gene from two azoospermic patient cohorts and matching control groups. FKBPL-AR interaction was assayed using reporter constructs in vitro.

Results: FKBPL is strongly expressed in mouse testis, with expression upregulated at puberty. The protein is expressed in human testis in a pattern similar to FKBP52 and also enhanced AR transcriptional activity in reporter assays. We examined sixty patients from the Japanese patient group and found one inactivating mutation and one coding change, as well as a number of non-coding changes, all absent in fifty-six controls. A second, Irish patient cohort of thirty showed another two coding changes not present in thirty proven fertile controls.

Conclusions: Our results describe the first alterations in the gene for FKBPL in azoospermic patients and indicate a potential role in AR-mediated signalling in the testis.

Figure 1

FKBPL structure and conservation. A. Structure of the human and mouse genes. The gene consists of two exons, joined by a short intron. The open reading frame is in grey, lengths in base pairs are indicated. The positions of the gene alterations seen in patients and listed in Table 1 are indicated above the human sequence. B. Protein structure. Peptidyl-prolyl cis-trans isomerase (PPI) domains are shown in yellow, tetratricopeptide repeats (TPR) in green. FKBP12 has a PPI domain but contains no TPR. FKBP52 and FKBP51 contain a duplication of the PPI domain, but the C-terminal copy is inactive (X). FKBP6 and FKBPL have N-terminal regions with some homology to the PPI.C. Alignment of the PPI domains from FKBP6, FKBP12 and FKBPL by CLUSTALW: residues conserved in PPI with good enzymatic activity are indicated above the alignment but can be seen to be poorly conserved in FKBPL.

Figure 2

Fkbpl transcription in mouse. A. A Multiple Tissue Array normalised polyA+ RNA blot was hybridised to a radiolabelled Fkbpl cDNA. The locations of testis, epididymis and submaxillary gland RNA are indicated. Top row: brain (A1), eye (A2), liver (A3), lung (A4), kidney (A5); Second row: heart (B1), skeletal muscle (B2), smooth muscle (B3), pancreas; Third row: (C1), thyroid (C2), thymus (C3), submaxillary gland (C4), spleen (C5); Fourth row: testis (D1), ovary (D2), prostate (D3), epididymus (D4), uterus (D5); Fifth row: embryo 7 days (E1), embryo 11 days (E2), embryo 15 days (E3) and embryo 17 days (E4); Last row: negative controls (F1-4). Positions not listed (e.g. B4) are blank. B. Northern blotting. Total RNA (20 ug) was extracted from the tissues indicated and fractionated on a 1% gel containing formamide before transferring to nitrocellulose and hybridising to the cDNA probe used in A. The positions of relevant size markers are indicated at right. The experiment was repeated twice. C. RT-PCR of total RNA isolated from testis at different days postnatally. The primers span the intron, allowing any contaminating genomic product to be easily distinguished (right); β-actin is used as an internal positive control: RT- are negative controls. The experiment was repeated twice.

Figure 3

FKBPL expression in human testis. A. Immunohistochemical staining of human testis sections with the indicated antibodies. AR shows nuclear staining in Sertoli (Si), Leydig (L) and peritubular myoid (P) cells, but is negative elsewhere, including spermatids (Sd). FKBP52 is cytoplasmic, staining Sertoli and spermatogenic cells, but shows weak or absent staining in Leydig and myoid cells. FKBPL is also cytoplasmic, matching FKBP52 in most respects except for strongly staining the Leydig cells. Preimmune serum for FKBPL (shown) and other antibodies was used as a negative control (neg.). Experiments were repeated three times. Staining for FKBPL and FKBP52 was almost identical in rat testis (not shown). B. Western blot of total protein from untransfected HT29 cells (untransfect.) or transfected with a GFP-FKBPL fusion protein (transfect.). The size of the endogenous protein (endog.) is indicated. The experiment was repeated three times. C. LNCaP cells were transfected with a construct (pPA) containing luciferase driven by the prostate specific antigen transcriptional regulatory elements either with or without FKBPL as shown. The indicated cultures were additionally exposed to 1 nM of AR ligand (R1881). Luciferase activity was measured using a fluorimeter and normalised to an internal control. The experiment was carried out more than three times with similar results. D. LnCaP cells were transfected with pPA and FKBPL from normal controls (WT) or from the mutated allele seen in patient 83 (p83- listed as #6 in Table 1) and exposed to R1881. The results shown represent the median of three experiments.