Article Novel mutations in testis-specific ubiquitin protease 26 gene may cause male infertility and hypogonadism

Transcription

1 RBMOnline - Vol 10. No Reproductive BioMedicine Online; on web 19 April 2005 Article Novel mutations in testis-specific ubiquitin protease 26 gene may cause male infertility and hypogonadism Darius A Paduch, MD, PhD, is currently finishing a fellowship in Male Reproductive Medicine and Microsurgery at the Cornell Institute for Reproductive Medicine, and Department of Urology, Weill Medical College of Cornell University, in New York, USA. In addition he is engaged in a post-doctoral research fellowship in the Population Council, Centre for Biomedical Research also in New York. He is a translational scientist with a major interest in the genetics of male infertility, and the role of gene dose compensation mechanisms in pathophysiology of Klinefelter syndrome. He hopes to continue his research as an assistant professor in Weill Medical College of Cornell University starting summer of Dr Darius A Paduch Darius A Paduch 1,2,3, Anna Mielnik 2, Peter N Schlegel 1,2 1 Department of Urology, Weill Medical College of Cornell University, 525 East 68th Street, Box 580, New York, NY 10021; 2 Population Council, Centre for Biomedical Research, 1230 York Avenue, New York, NY 10021, USA 3 Correspondence: Tel: ; Fax: ; dpaduch@msn.com Abstract Patients (n = 188) with non-obstructive azoospermia (NOA), and 17 fertile controls were screened for sequence changes in the ubiquitin specific protease (USP) 26 gene. Semen analysis, hormonal evaluation, and testicular biopsies were performed. DNA was extracted from whole blood. Denaturing high-performance liquid chromatography was used to screen for single nucleotide polymorphisms. Amino acid sequences were determined in samples with mutations. Twenty out of 188 (10.6%) infertile men had amino acid changes in USP26. No changes were found in fertile controls. 1090C T substitution and (363insACA; 494T C; 1423C T) change were found in 3.3 and 1.9% of patients respectively. Serum testosterone concentrations and testicular volume were lower in the mutation positive group compared with the non-mutation group (272 versus 366 ng/dl; P = 0.01) (volume: 7.88 versus 10 ml, P = 0.03). Six out of 28 (21%) patients with Sertoli cell-only syndrome, and two out of 18 (11%) patients with maturation arrest had mutations in the USP26 gene. There were no live deliveries in couples with the USP26+ mutation, and three live deliveries in the group without mutations. The USP26 gene may be of importance in male reproduction. Mutations in this gene may be associated with male infertility, and may negatively affect testicular function. Keywords: infertility, ubiquitin, USP26 Introduction Infertility affects one in 10 couples, with 30 50% of couples suffering from male factor infertility. Although varicocoele, history of undescended testis, Klinefelter syndrome, infections and drugs can account for 70 80% of male infertility, the aetiology of idiopathic male infertility is often unknown. Growing evidence suggests that genetic defects affecting spermatogenesis may be responsible for many cases of idiopathic infertility (Huynh et al., 2002). Although it is known that chromosomal aberrations such as Klinefelter syndrome (KS), and Y chromosome microdeletion are associated with infertility, the molecular mechanisms responsible are not known (Lanfranco et al., 2004). Recently reported, testis specific-genes located on the X chromosome, revealed an array of candidate genes for male infertility (Wang et al., 2001). One of those genes is ubiquitin specific protease 26 (USP26), located on the X chromosome, at Xq26.2. Ubiquitination and deubiquitination of macromolecules regulates the cell cycle, chromosomal structure, vacuolization, and gene silencing (Glickman et al., 2002). Deubiquitination of macromolecules by deubiquitinating enzymes (DUB), including ubiquitin proteases, can rescue macromolecules from degradation through substrate-specific, N-terminaldependent, enzymatic reaction (Wilkinson, 1997; Wing, 2003). Because of the importance of DUB in cell cycle regulation, as well as testis-specific expression of this gene, it was decided to choose USP26 as a novel candidate gene for the study of male 747

2 infertility. It has previously been reported that preliminary data indicates increased number of mutations in the USP26 gene in men with severe male factor infertility (Paduch et al., 2004). The initial observations have been further strengthened by a recent report describing a 363insACA in USP26, which was identified in 9.5% of patients with Sertoli cell-only syndrome (SCO; Stouffs et al., 2004). The present study provides the first extended phenotypic description of patients with mutations in USP26 and evidence linking mutations in USP26 to male infertility. Materials and methods Patient selection This study is part of a larger, IRB approved, study of novel genes in male infertility. A DNA repository with over 1500 DNA samples obtained from patients referred for Y microdeletion screening has been established. Two hundred and thirteen randomly chosen patients with azoospermia or severe oligozoospermia who were referred for genetic Y microdeletion screening were included in the current project. Seventeen fertile men served as a control group. Patients with known chromosomal aberrations, or Y chromosome microdeletions were excluded from this analysis. The results of semen analysis, serum hormones (T, FSH, LH, E), karyotype and physical examinations were obtained from patients records. Some, but not all, patients underwent miscrosurgical testicular sperm extraction (TESE), and for those patients intraoperative findings, results of intracytoplasmic sperm injection (ISCI), and final pathology of testicular biopsies were available. Testicular volume was measured by an attending physician using an orchidometer. Mutation analysis DNA from all subjects in this project was extracted using the Stratagene DNA extraction kit (200600) (Stratagene, La Jolla, CA, USA) and stored at 20 C. USP26 sequence (AF285593) was obtained from the NCBI web site (date of accession: 7/10/2003). The USP26 gene has 2794 bp and no introns, and it encodes a 913 amino acid long globular protein. The entire gene was divided into six overlapping fragments (spans: 94 to 2869) ranging from 409 to 600 bp in length. Each fragment was amplified in a single polymerase chain reaction (PCR) using a GeneAmp PCR System 9700 thermocycler (Applied Biosystems, Foster City, CA, USA). The primers used are listed in Table 1. Genomic DNA 150 ng was added to the PCR mixture containing 2.5 IU of Transgenomic Optimase polymerase (Transgenomic, Omaha, NE, USA), 1.5 mmol/l of MgSO 4, 0.5 µmol/l of each forward and reverse primer, 200 µmol/l of each dntp (deoxynucleoside triphosphate), 5 µl of reaction buffer (10 stock solution), and water up to 50 µl. Annealing temperatures were 2 C higher then calculated primer melting temperatures, as recommended by the Transgenomic Optimase product insert. Amplification was carried for 35 cycles. The presence of PCR product was verified by a high performance liquid chromatography system (HPLC; Wave Nucleic Acid Fragment Analysis System Model 3500A; Transgenomic). PCR product 5 µl was injected into the HPLC column. Concentration of the acetonitrile buffer and elution temperature was calculated for each amplicon, and controlled with Wavemaker software (Transgenomic) (Table 1). The chromatograms were stored on a hard drive. The presence of product was defined as a chromatogram peak above 2.5 mv. If the peak was lower than 2.5 mv, the PCR reaction was repeated or 15 µl of product was injected. Wild-type amplicon from fertile men was hybridized with each PCR product at 95 C, and slowly cooled down in a thermocycler. Hybridized samples were again injected into the HPLC column, and chromatograms were stored on a hard drive. The presence of sequence change in the examined sample was evident by the presence of two to four peaks on its chromatogram, whereas samples with no sequence changes had only one peak. Amplicons from samples with sequence changes were Table 1. Primers used to amplify the overlapping fragments of ubiquitin specific protease (USP) 26 gene, length of amplicon product and temperatures of high performance liquid chromatography screening used for each amplicon. bp = base pair. USP26 Fragment Primers Temperatures fragment size bp of screening no. ( C) For: 5 -ACCAATACTAGAAATAGGATCATTCTG-3 ; 55.1, 55.4, 57.5 Rev: 5 -TCCCACTTCCTTTTGCTATCTC For: 5 -GCACAACACAGAAGGAAATCAA-3 ; 56.0, 56.8 Rev: 5 -CCGTGGCATATTTTCTCTGG For: 5 -CGGTTACACAAAGTGGGATAAA-3 ; 52.0, 55.0, 56.4 Rev: 5 -TTCTTTGGGGAAGGTTGATG For: 5 -TGTTGCACTCCATTGCTTGT-3 ; 56.0, 56.8 Rev: 5 -TTGCTGCTGTCTTCTGCTTG For: 5 -TTAAAGGGGCAAGCAGAAGA-3 ; 55.6, 57.2, 58.9 Rev: 5 -TGAGGGGCTTGTTCACAGAT For: 5 -ATCTGTGAACAAGCCCCTCA-3 ; 55.8, 57.9 Rev: 5 -CCATGGAGGAAGTGGTATCG-3

3 collected using the Transgenomic fragment collector, sequenced with forward and reverse primers, and results of sequencing were compared with wild-type sequence. Nucleotides were counted using A of ATG start codon as +1 nucleotide. Reverse transcriptase (RT) PCR In order to examine USP26 mrna expression patterns in human testis, total RNA was extracted from nine testicular samples: three patients with normal histology, three patients with maturation arrest, and three patients with SCO, using RNA STAT-60 (Tel-Test, Freindswood, TX, USA). The quality of RNA extraction was verified by denaturing gel electrophoresis, using standard protocol with formaldehyde and MOPS. Two distinct ribosomal RNA bands were identified in each sample examined. Using the Titan One Tube RT-PCR system (Roche Diagnostic, Indianapolis, IN, USA), a cdna library was generated from 1 µg of total RNA, at 50 C for 30 min, followed by PCR according to the manufacturer s instructions. To amplify USP26 from cdna, the following set of primers was used: 5 -GCACAACACAGAAGGAAATCAA-3 and 5 -CCGTGGCATATTTTCTCTGG-3 (526-bp long product). Oestrogen receptor alpha, which is expressed in testis, was used to verify the quality of cdna synthesis and PCR reaction. To exclude genomic amplification, PCR was performed with the same total RNA samples without reverse transcriptase. Products were analysed on 4% agarose gel. Reactions were performed in triplicate to ensure consistent results. Statistical analysis Student s t-test was used to compare means between groups, except for comparison of testicular volumes, where Mann Whitney U-test was used. Fisher exact test was used to calculate statistical significance of assisted reproduction and sperm retrieval rates between groups with and without mutation. Results Phenotypic changes identified in patients with USP26 mutations Sequence changes with amino acid changes or insertion were found in 20 (10.64%) out of 188 patients screened (Tables 2 and 3). For detailed discussion of identified mutations, please see below. Results of testis biopsies were available in 49 patients screened for mutation. Twenty-one per cent (six out of 28) of patients with SCO had mutations in the USP26 gene, and 11% (two out of 18) patients with maturation arrest had mutations in the USP26 gene. No mutations were found in three patients with hypospermatogenesis (Tables 3 and 4). To answer the question whether mutations in the USP26 gene are functionally relevant, results of semen analysis, testicular size and hormonal evaluation were compared in 20 patients with mutations to 168 patients without mutations. Patients with KS and deleted in azoospermia factor (AZF) a, b or c deletions were excluded from the analysis. Only patients seen at Weill Medical College had data available for verification. Review of available semen analyses showed that all patients with USP26 mutation were azoospermic, whereas only 66% of patients without mutations were azoospermic. Mean testosterone concentration in patients with mutation in USP26 was lower by 104 ng/dl. This difference was statistically significant (P = 0.01) (Table 5). There were no statistically significant differences in concentrations of FSH, LH, or oestradiol between groups with and without mutation (Table 5). Patients with the USP26 gene mutation had a statistically significant difference in testicular size, as compared with patients without the mutation (7.88 ml, SD = 4.5 ml versus 10 ml, SD = 4.41 ml; P = 0.03) (Figure 1). Forty-two patients out of 188 screened underwent TESE with ICSI for male factor infertility. In this group, seven patients had the USP26 mutation. Although not statistically significant, patients with this mutation had lower sperm retrieval rate two out of seven patients (29%) versus 15 out of 35 patients (43%). There was one pregnancy and no delivery in the group with mutation, and seven pregnancies and three live deliveries in patients without the mutation. It remains to be seen if those differences will become statistically significant with an increased number of patients screened. Detailed description of identified mutations Seven patients had 1090C T substitution, with leucine to phenylalanine amino acid change in position 364 (). Another patient had 1090C T and 1274C T substitutions in the same allele, causing leucine to phenylalanine amino acid change in position 364 (), and proline to leucine change in position 425 (P425L). 1090C T mutation seems functionally significant, since two patients in this group had SCO (Table 3). Four patients had characteristic insertion of 363insACA in nucleotide position 363, causing a threonine insertion in amino acid position 121, followed by tandem of 494T C and 1423C T substitution, with amino acid sequence changes: leucine into serine, and histidine into tyrosine respectively (363insACA; 494T C; 1423C T). One patient had insertion of ACA in nucleotide position 363, causing a threonine insertion in position 121, followed by tandem of 494T C without 1423C T substitution (363insACA; 494T C). One patient had 1423C T substitution without any changes. It is interesting to note that the ACA insertion followed by 494T C and 1423C T substitution occurs in the same allele, properly described as (363insACA; 494T C; 1423C T). It seems that 363insACA together with 494T C by itself can negatively affect function of the USP26, as evident by low testosterone in patient no. 15 bearing this mutation (Table 3). At the same time, biopsy of the patient with the sole 1423C T mutation revealed SCO. This patient also had very low testosterone. There was no difference in testosterone concentration among four patients with the (363insACA; 494T C; 1423C T) sequence variant and the nine other mutations identified. 749

4 Table 2. Frequency and type of detected mutations among 188 patients with azoospermia and severe oligozoospermia. Nucleotide change, amino acid position No. Percentage patients insACA, T121ins and 494T C, L165S and 1423 H475Y 1737G A, M579I G A, M579I and 2202A C, K734N insACA, T121ins and 494T C, L165S and 1274 P425L T A, L346H H475Y G A, E500K T659M Totals Table 3. Details of identified mutations in ubiquitin specific protease 26 gene. SCO = Sertoli cell-only syndrome, MA = maturation arrest, Te = testosterone, E = oestradiol, T-L = left testicular volume (ml), T-R = right testicular volume (ml). Amino acids: C = cysteine, L = leucine, E = glutamate, M = methionine, F = phenylalanine, N = asparagine, H = histidine, P = proline, I = isoleucine, S = serine, K = lysine, T = threonine, Y = tyrosine. Yq: deletion of distal arm of Y chromosome was verified by microdeletion screening and not karyotype. Patient Histology FRG1 FRG2 FRG3 FRG4 FRG5 FRG6 Sperm Te a FSH a E a LH a T-L T-R Karyotype no. count T A, L346H 2 SCO XY G A, A C, M5791 K734N SCO XY and 1274 P425L SCO XY H475Y 7 SCO 363ins 494, XY ACA T C, Ti2lins L165S H475Y XY G A, M ins ACA, T C, C T Ti2iins L165S H475Y Yq continued on page 751

5 Table 3. continued Patient Histology FRG1 FRG2 FRG3 FRG4 FRG5 FRG6 Sperm Te a FSH a E a LH a T-L T-R Karyotype no. count XY T659M XY ins XY ACA, T C, T121ins L165S 14 MA 363ins XY ACA, T C, C T T121ins L165S H475Y 15 MA XY G A, M SCO XY G A E500K 20 SCO 363ins ACA, T C, C T Tl2lins L165S H475Y a Units as shown in Table 5. Table 4. Results of mutation screening in patients with (USP26+) and without (USP26 ) identified mutations in ubiquitin specific protease (USP) 26 gene who underwent testicular sperm extraction. Histology USP26+ (%) USP26 (%) Total SCO 6 (21) 22 (79) 28 Maturation arrest 2 (11) 16 (89) 18 Hypospermatogenesis 0 3 (100) 3 Total 8 (16) 41 (84)

6 Table 5. Hormonal characteristics of patients with (USP26+) and without (USP26 ) identified mutations in ubiquitin specific protease 26 gene. Category No. Testosterone FSH (IU/l) LH (IU/l) Oestradiol (ng/dl) (ng/l) Mean SD Mean SD Mean SD Mean SD USP USP P-value 0.01 NS NS NS NS = not significant. Figure 1. Patients with mutations in ubiquitin specific protease (USP) 26 (USP26+) have lower testicular volume than patients without mutation (USP26 ). Figure 2. Results of reverse transcriptase-polymerase chain reaction (RT-PCR) with total RNA extracted from testicular samples. N = normal histology, MA = maturation arrest, SC = Sertoli cell-only syndrome, W = water. 752 Figure 3. Schematic diagram of ubiquitination and deubiquitination pathways. Ubiquitin specific protease (USP) 26 is able to rescue proteins condemned to deactivation by proteasome system. Adequate balance between ubiquitination and deubiquitination of cell cycle and apoptotic proteins may ultimately be responsible for the fate of the cell. Ub = ubiquitin. E1 = ubiquitin-activating enzyme E1; E2 = ubiquitin-conjugating enzyme E2; E3 = ubiquitin-protein ligase E3.

7 Three patients had nucleotide substitution 1737G A, with amino acid change methionine to isoleucine (M579I); however, one of the patients in this group had the additional amino acid change lysine to asparagine (K734N), as a result of 2202A C substitution (1737G A; 1737G A). The following four substitutions occurred only once in the studied population: 1037T A, resulting in leucine to histidine change (L346H), 1423C T with amino acid change histidine to tyrosine (H475Y), 1497G A substitution causing glutamate to lysine change (E500K), and 1976C T substitution with threonine to methionine change (T659M). RT PCR It is thought that USP26 mrna should be absent in SCO syndrome; however, initial data reveal that USP26 is expressed in testis with normal histology, as well in testis samples with SCO and MA (Figure 2). This experiment was repeated three times and each time the same pattern of expression was achieved. More experiments using real time RT-PCR, northern blot, and in-situ hybridization to better assess spatial expression of USP26 are being performed in the laboratory. Since it was shown that mutations in USP26 may cause hypogonadism, it will be important to understand how and if those mutations affect function of Leydig cells. Discussion This study presents the first evidence that USP26 may be important in male infertility and testicular dysfunction. USP26 belongs to a large family of DUB. Ubiquitination and deubiquitination of macromolecules regulates the cell cycle, chromosomal structure, vacuolization, and gene silencing (Wilkinson, 1997; Wing, 2003; Ciechanover et al., 2004) During the cell cycle, molecules important for apoptosis or cell proliferation can be turned off by tagging a particular molecule with ubiquitin (Ub) (Zhang et al., 2004) (Figure 3). This molecule will then undergo degradation by the proteosome system. Deubiqutination of macromolecules by DUB, including ubiquitin proteases, can rescue macromolecules from degradation through substrate-specific, N-terminaldependent, enzymatic reaction. DUB identified so far are tissue specific with high specificity for substrate. DUB have cysteine and two histidines consensus patterns. Cysteine consensus pattern: G-(LIVMFY)-x(1,3)-(AGC)-(NASM)-x- C-(FYW)-(LIVMFC)-(NST)-(SACV)-x-(LIVMS)-Q (C is the putative active site residue) and histidine consensus pattern: Y- x-l-x-(sag)-(livmft)-x(2)-h-x-g-x(4,5)-g-h-y (The two H s are putative active site residues) (Wilkinson, 1997). Structure analysis revealed that none of the identified mutations are located within active sites; however, the majority of mutations are found within the DUB substrate recognition site. Hence, it is postulated that the activity of deubiquitination in the poliub assay test may not be affected in mutation-specific functional analysis of USP26, but identified mutations may affect the interaction between the substrate and the USP26. USP26 was first identified by Wang et al., and it is believed to be expressed only in testis. Recently, Stouffs et al. reported data on mutation analysis in the USP26 gene in patients with infertility (Stouffs et al., 2005). Their manuscript reports one of several mutations also reported by the present authors (Paduch et al., 2004) during the AUA meeting in Both this paper and Stouffs et al. reported characteristic insertion of 363insACA in nucleotide position 363 (363insACA; 494T C; 1423C T) causing a threonine insertion in position 121, followed by tandem of 494T C and 1423C T substitution, with amino acid sequence changes: leucine into serine, and histidine into tyrosine respectively. Although all eight patients in the report of Stouffs et al. had combined haploallelic change following 363ACAins, in the present study, one patient was found with only 363insACA and 1423C T (363insACA; 494T C), and one patient with 1423C T as a sole mutation. It will have to be determined from further studies which of those three mutations in one allele affect function of USP26 protein. In the current group of patients, nine new mutations were identified other than (363insACA; 494T C; 1423C T) substitution (Table 3). Although all patients with Klinefelter syndrome and AZF a, b or c deletion were excluded, one patient with (363insACA; 494T C; 1423C T) change had Y chromosome distal arm microdeletion. It is interesting that Stouffs et al. noticed the same association in one of their patients. It is unknown if (363insACA; 494T C; 1423C T) substitution plays any role in pathogenesis of Yq microdeletion or if the above finding is a pure coincidence. The differences in the total number of mutations identified in the screened population between the present report and the one by Stouffs et al. (2005) can be explained by the fact that Stouffs et al. screened the entire gene in only 42 patients with SCO, whereas the present study screened 214 patients and then excluded patients with KS and AZF a, b or c deletion. Other than 1090C T substitution and (363insACA; 494T C; 1423C T) change which occurred in 3.3 and 1.9% all the other new mutations identified by us occurred in less than 1% of patients ( %); therefore, to detect those changes, one needs to screen the entire gene in over 100 patients. It is likely that the identified mutations have functional significance of sincethe same type of mutation was found by two groups in diverse populations in Europe (Stouffs et al.) and in the USA (present study). It is important to notice that more fertile controls were so that statistical analysis of the frequency of mutations could be performed. It is astonishing that the frequency of (363insACA; 494T C; 1423C T) substitution in patients with known histology is almost exactly the same in the present report 4/44 (9.1%) and that published previously (9.5%). This study is the first to report that the presence of mutations in USP26 is associated with significantly lower testicular volume, and lower testosterone concentration. Patients with USP26 mutations have mean testosterone concentration below the low normal range (300 ng/dl) in this study. No study so far has evaluated spatial localization of the USP26 mrna or protein, hence it is possible that either the expression of USP26 is limited to germ cells and testicular dysfunction is secondary to aplasia of germinal epithelium, or alternatively USP26 may be expressed in Leydig or Sertoli cells as well. However, it has been established that patients with idiopathic infertility often suffer from hypogonadism. (Andersson et al., 2004). Preliminary data showed that USP26 is expressed abundantly in patients with normal histology, SCO, and 753

8 maturation arrest. Limited expression of USP26 was anticipated in patients with SCO, but experiments so far have proved us wrong. It is likely that in humans expression of USP26 is not limited to germ cells. This issue is currently being evaluated. Regardless of the mechanism, the data provide early evidence that USP26 mutations may cause testicular dysfunction. In the present series, 21% of patients with SCO and 11% of patients with MA had mutation in USP26. Stoufts et al. reported 7.2% rate of mutation in USP26 among patients with SCO, and no mutation in MA; however, their paper only reported one type of mutation (363insACA; 494T C; 1423C T). Since four patients with the same mutation were identified and two had SCO, the actual frequency of 363ACAins mutation among men with SCO is almost the same: 7.7% (2/44). deubiquitinating enzymes. Federation of American Societies for Experimental Biology Journal 11, Wing SS 2003 Deubiquitinating enzymes the importance of driving in reverse along the ubiquitin proteasome pathway. International Journal of Biochemistry and Cell Biology 35, Zhang HG, Wang J, Yang X et al Regulation of apoptosis proteins in cancer cells by ubiquitin. Oncogene 23, Received 3 February 2005; refereed 17 February 2005; accepted 2 March Forty-two patients underwent TESE, and the results of sperm retrieval as well as pregnancy rates are lower in the group with USP26 mutation; however, the differences are not statistically significant at this point. More assisted reproduction data are being collected to further analyse the role of USP26 mutations in outcomes of assisted reproduction. In conclusion, these data indicate that mutations in USP26 may cause significant testicular dysfunction and male infertility, and may potentially affect the outcomes of assisted reproduction. It is hoped that this report will stimulate further research into deubiquitination and its effect on male fertility. Acknowledgements This work was supported by the Frederick J and Theresa Dow Wallace Fund of the New York Community Trust. Internal Review Board of the Weill Medical College of Cornell University approved this study. References 754 Andersson AM, Jorgensen N, Frydelund-Larsen L et al Impaired Leydig cell function in infertile men: a study of 357 idiopathic infertile men and 318 proven fertile controls. Journal of Clinical Endocrinology and Metabolism 89, Ciechanover A Iwai K 2004 The ubiquitin system: from basic mechanisms to the patient bed. International Union of Biochemistry and Molecular Biology Life 56, Glickman MH Ciechanover A 2002 The ubiquitin proteasome proteolytic pathway: destruction for the sake of construction. Physiological Reviews 82, Huynh T, Mollard R, Trounson A 2002 Selected genetic factors associated with male infertility. Human Reproduction Update 8, Lanfranco F, Kamischke A, Zitzmann M, Nieschlag E 2004 Klinefelter s syndrome. Lancet 364, Paduch DA, Mielnik AN, Schlegel PN 2004 Novel mutations in testis specific ubiquitin protease 26 (USP26) in infertile males. Presented at the 2004 annual meeting of the American Urological Association, San Francisco, USA. Publishing ID: Stouffs K, Lissens W, Tournaye H et al Possible role of USP26 in patients with severely impaired spermatogenesis. European Journal of Human Genetics 13, Wang PJ, McCarrey JR, Yang F, Page DC 2001 An abundance of X- linked genes expressed in spermatogonia. Nature Genetics 27, Wilkinson KD 1997 Regulation of ubiquitin-dependent processes by