CDY1 and BOULE transcripts assessed in the same biopsy as predictive markers for successful testicular sperm retrieval

Transcription

1 CDY1 and BOULE transcripts assessed in the same biopsy as predictive markers for successful testicular sperm retrieval Sandra E. Kleiman, Ph.D., Ofer Lehavi, M.D., Ron Hauser, M.D., Amnon Botchan, M.D., Gedalia Paz, Ph.D., Haim Yavetz, M.D., and Leah Yogev, Ph.D. Institute for the Study of Fertility, Lis Maternity Hospital, Tel Aviv Sourasky Medical Center, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel Objective: To evaluate the molecular markers CDY1 and BOULE in the same testicular biopsy for predicting success of sperm retrieval in azoospermic men. Design: Prospective study. Setting: University-affiliated medical center. Patient(s): Azoospermic men (n ¼ 92) who underwent testicular sperm extraction (TESE) and who were classified as normal spermatogenesis, mixed atrophy, spermatocyte maturation arrest, or Sertoli cell only according to their combined histological and cytological testicular findings. Intervention(s): Quantitative and qualitative evaluation of testicular biopsies by histological and qualitative reverse transcriptase-polymerase chain reaction (RT-PCR) expression methodologies. Main Outcome Measure(s): CDY1 and BOULE expression and the presence of sperm cells in testicular tissue. Result(s): Both transcripts significantly predicted the presence of sperm cells by qualitative and quantitative methodologies. Although CDY1 had the best sensitivity by qualitative RT-PCR (98.3%), assessing both transcripts simultaneously had an additive efficacy compared with assessing CDY1 alone, improving the specificity from 84.4% to 96.3%. Conclusion(s): Assessing the expression of both CDY1 and BOULE by qualitative RT-PCR is a sensitive and feasible test for predicting the presence of sperm cells in testicular tissue and may serve as a predictive tool if repeated TESE is required. (Fertil Steril Ò 2011;95: Ó2011 by American Society for Reproductive Medicine.) Key Words: Azoospermia, spermatogenesis, BOULE, CDY1, gene expression in testis Genes that are expressed during spermatogenesis encode proteins necessary for specific stages of sperm cell development as well as for maintaining the general functions of the cells involved. CDY1 (chromodomain Y1) genes are of special interest because they are included in the AZFc (azoospermia factor-c) region, the most frequently AZF-deleted region in the Y-chromosome that has been identified in men with azoospermia or severe oligozoospermia (1 5). Their expression in testicular tissue correlates with complete spermatogenesis (6, 7). CDY1 genes belong to the CDY family of genes, which includes two identical CDY1 gene copies at the AZFc region that code for three alternative spliced transcripts called major, minor, and short (6, 8, 9); two additional genes located proximal to the AZFb region (CDY2); and two autosomal genes (CDYL and CDYL2) (8, 10, 11). The proteins encoded by the CDY family consist of a unique combination of a domain implicated in chromatin binding (the chromodomain) and a catalytic domain (resembling enoyl-coa- isomerase). Members Received January 11, 2011; revised March 2, 2011; accepted March 8, 2011; published online April 6, S.E.K. has nothing to disclose. O.L. has nothing to disclose. R.H. has nothing to disclose. A.B. has nothing to disclose. G.P. has nothing to disclose. H.Y. has nothing to disclose. L.Y. has nothing to disclose. The research was carried out under the auspices of the Alan and ADA Selwyn Chair in Clinical Infertility Research and Molecular Medicine (Melbourne, Australia), granted to G.P. Reprint requests: Sandra E. Kleiman, Ph.D., Institute for the Study of Fertility, Lis Maternity Hospital, Tel Aviv Sourasky Medical Center, 6 Weizman Street, Tel Aviv 64239, Israel ( ser@tasmc.health.gov.il). of the CDY family were postulated to be HAT (histone acetyltransferase) proteins involved in the spermiogenetic process and transcriptional corepressor (12, 13). The level of CDY1 expression in testis had not been reported earlier. An additional gene of interest that participates in the spermatogenesis process is the highly conserved BOULE gene. It belongs to the DAZ (deleted in azoospermia) gene family, and its quantitative expression in the human testis seems to be a good predictor of finding sperm cells in azoospermic men (14, 15). The DAZ gene family includes the autosomal genes BOULE and DAZL and the four DAZ genes in the AZFc region. BOULE protein expression in men with complete spermatogenesis was found to be restricted to different stages of spermatocytes (16) or round spermatids (17). BOULE protein expression was completely lacking in spermatocytes of testicular biopsies with meiotic arrest (16). It was suggested that BOULE may encode a key switch that regulates the progression of germ cells through meiosis and spermiogenesis in mouse and Drosophila and probably in humans as well (14, 18, 19). Understanding the expression of genes involved in spermatogenesis in fertile and infertile men may assist in understanding spermatogenic failure. Qualitative reverse transcriptase polymerase chain reaction (RT-PCR) and quantitative RT-PCR (Q-RT-PCR) can be applied to analyze the expression of a modest number of genes, making it potentially useful for testing gene expression for diagnostic purposes. In the era of intracytoplasmic sperm injection (ICSI), finding even a few sperm cells in the testicular tissue of azoospermic men can make them potentially fertile. Detection of testicular spermatozoa /$36.00 Fertility and Sterility â Vol. 95, No. 7, June doi: /j.fertnstert Copyright ª2011 American Society for Reproductive Medicine, Published by Elsevier Inc.

2 in azoospermic men is, however, hampered by the nonhomogeneous nature of testicular tissue (20). Since diagnosis is made on a minute section of testicular tissue, it is of the utmost importance to develop an easy and reliable test that will help to predict the presence of testicular spermatozoa. Until serum markers become available for this purpose, the search for potential molecular markers currently targets the testicular tissue (6, 7, 15, 21, 22). We performed an assessment of the expression of CDY1 and BOULE in the same testicular specimen in a large number of testicular samples using quantitative and qualitative mrna expression in an effort to establish the best predictive markers and methodology for detecting complete spermatogenesis. MATERIALS AND METHODS Patients Ninety-two testicular biopsy specimens from 70 nonobstructive and 22 obstructive azoospermic men were analyzed in this study. All testicular samples were obtained from testicular sperm extraction (TESE) procedures performed in an attempt to obtain sperm for ICSI. All the men had signed a written informed consent to undergo these procedures and for the samples to be used for further studies. The study was approved by the local Institutional Review Board committee in accordance with the Helsinki Declaration of None of these 92 men had ever been included in previous studies that assessed CDY expression (6, 7). Testicular Samples After the TESE procedure had been carried out, one biopsy (R50 mg) was divided into two pieces: one small piece (<10 mg) was taken for histopathological analysis and RNA isolation, and the remainder was minced for spermatozoa extraction. In all cases, two additional biopsies from two different locations of the testis were taken for spermatozoa extraction to be used in the ICSI process (20). The testicular biopsies were classified according to the most advanced spermatogenetic cell identified in the combined cytological and histological findings in the testes into testes with normal spermatogenesis (samples with an average of at least 17 elongated spermatids per tubule; n ¼ 25), mixed atrophy (elongated spermatids detected in some tubules; n ¼ 39), complete maturation arrest at spermatocyte stage (n ¼ 7), and Sertoli cell only (n ¼ 21). Histological quantification was performed as previously described elsewhere (23). Synthesis of cdna by RT RT was performed in 30 ml aliquots containing mg of total RNA and oligo(dt)15 using the RevertAid First strand cdna Synthesis Kit (Fermentas, Burlington, Ontario, Canada). RT-PCR Qualitative expression of RBM, DAZ, CDY1, BOULE, and b-actin was assessed in all the testicular samples by multiplex RT-PCR. The expression of b-actin was evaluated as an internal control for the RNA isolation and efficiency of the RT-PCR. DAZ and RBM expressions were taken as being indicative of the presence of germ cells, and CDY1 (minor and short) expression was indicative of the presence of spermatids in the analyzed biopsy (6) (see Supplementary Material). Two mixes were established: the previously reported mix A included b-actin, DAZ, and RBM (24), and the heretofore unreported mix B included b-actin, CDY1minor and short, and BOULE. Q-RT-PCR The relative concentration of BOULE and CDY1 gene expression was assessed in triplicate with the ABI Prism 7700 Sequence Detector (Applied Biosystems, Darmstadt, Germany) using TaqMan Universal PCR Master Mix and Assay On Demand (AOD) Gene Expression Products (Applied Biosystems). Relative gene expression analysis was performed according to the Comparative C T Method: 2 DCT (25) and Supplementary Material. The 2 DCT parameter indicates the expression fold of one gene (CDY1 or BOULE) with respect to the calibrator PPIA (cyclophilin A) gene (Supplementary Material). Statistics Fisher s exact test was used to assess the association between categorical parameters (RT-PCR gene expression and the presence of sperm cells). The significance of the relationship between the level of expression (Q-RT-PCR) of the housekeeping PPIA gene (CT), the CDY1, and the BOULE transcripts (2 DCT ) in the study groups was assessed by one-way analysis of variance (ANOVA), which was performed after log transformation to achieve an improved normal distribution. Pairwise comparisons were performed by Tukey s method. The relationship of the levels of gene expression in the study groups was assessed by Pearson s correlation test. Spearman s rho test determined the nonparametric correlations between Q-RT-PCR and the average number of germ cells (spermatocytes, round spermatids, or elongated spermatids) counted per tubule. Logistic regression was performed to analyze the prognostic value for forecasting the presence of sperm cells among the gene expressions and among the RT-PCR and Q-RT-PCR tests. All statistics were performed at the Statistical Laboratory of Tel Aviv University. RESULTS Qualitative expression of BOULE and CDY1 was assessed by a multiplex RT-PCR. The absence of amplification was confirmed by single-gene RT-PCR. For quantitative assessment, PPIA I was chosen as an endogenous control to normalize for variations in cdna amounts because no significant change in PPIA I expression was found among two biopsies with normal spermatogenesis and two with Sertoli cell only (Supplementary Material). We then assessed all 92 study specimens with PPIA I, and the results verified no significant change in its expression among the four histological/ cytological groups (ANOVA, P¼.208). BOULE Expression The RT-PCR assessment showed no expression of BOULE in all the biopsies with Sertoli cells only, with the exception of one that exclusively expressed BOULE but no other germ cell markers (DAZ and RBM). BOULE was detected in all the biopsies that had any germ cells, with the exceptions of two biopsies of men with severe mixed atrophy (Fig. 1). One of the latter two showed no expression of additional germ cells markers (DAZ and RBM), suggesting that the sample for RNA isolation contained only Sertoli cells. BOULE predicted the presence of sperm cells by the RT-PCR evaluation at a high sensitivity and a high specificity (Table 1). The AOD assay amplified sequence stretches on the boundary of exons 6-7 of BOULE, allowing the quantification of all three alternative transcripts reported (26). There was a significant difference in the Q-RT-PCR expression of BOULE among the study groups (ANOVA, P<.001). Unfortunately, Q-RT-PCR assessment of BOULE could not distinguish between the mixed atrophy and spermatocyte maturation arrest groups (comparison by Tukey s method, P>.05; Fig. 2). Specifically, the mean expression of BOULE among specimens with spermatocyte maturation arrest was very close to the mean expression among those with mixed atrophy (mean 2 DCT ¼ SE and SE, respectively; Fig. 2). The specimen with Sertoli cell only and expression of BOULE by the RT-PCR method had a very low level of expression when assessed by Q-RT-PCR ( ) compared with specimens from the other groups. BOULE predicted sperm cells with a sensitivity of 81.3% and specificity of 75% by the Q-RT-PCR assessment (logistic regression, P¼.002; odds ratio, ; 95% confidence interval [CI] for odds, ; Table 1) Kleiman et al. CDY1 and BOULE in sperm retrieval Vol. 95, No. 7, June 2011

3 FIGURE 1 Expression of CDY1 and BOULE in specimens with different testicular findings. n ¼ number of specimens. Q-RT-PCR expression level of BOULE among specimens with normal spermatogenesis (mean 2 DCT ¼ and 0.45, respectively). There was a significant difference in the expression of CDY1 among the four study groups (ANOVA, P<.001). Moreover, CDY1 could distinguish between the mixed atrophy and spermatocyte maturation arrest groups (Fig. 2). CDY1 predicted the presence of sperm cells with a higher sensitivity than BOULE (Table 1; logistic regression, P¼.006; odds ratio, 1.488; 95% CI for odds, ). CDYI Expression RT-PCR found no expression of CDY1 in all the biopsies with Sertoli cells, and it was only detected in biopsies with complete spermatogenesis, with the exception of one biopsy with spermatocyte maturation arrest (Fig. 1). This specimen was obtained from the subject s left testis, and mature sperm cells were detected in the contralateral testis, suggesting that the left testis might have had a few spermatids that were undetected by histological and cytological analyses. CDY1 was not detected in four specimens with mixed atrophy. One of these specimens had severe hypospermatogenesis in isolated tubules, most of the tubules had hyalinized and contained Sertoli cell only, and there was no expression of the markers DAZ, RBM, and BOULE. The other three specimens with severe hypospermatogenesis expressed germ cells markers, suggesting the presence of germ cells but not of spermatids in that specific biopsy sample. CDY1 predicted the presence of sperm cells at a higher sensitivity than BOULE by RT-PCR assessment (Table 1). CDY1 has only one intron that is spliced in CDY1minor and short transcripts but not in CDY1major (6, 8). Moreover, the 5 0 border of the unique intron differs between CDY1minor and short transcripts. To prevent possible contamination of the DNA, the AOD assay was designed to exclusively detect CDY1minor transcripts. The mean Q-RT-PCR expression level of CDY1minor in specimens with normal spermatogenesis was 362 times lower than the mean Comparison Between CDY1 and BOULE Both markers (CDY1 and BOULE) could significantly predict the presence of sperm cells by the RT-PCR assessment (Fisher s exact test, P<.001). The combination of both markers enhanced the predictive power by having an additive efficacy (calculated by logistic regression). The CDY1 odds ratio was (95% CI for odds, ,185), and BOULE improved it by (95% CI for odds, ). CDY1 predicted the presence of sperm cells with a better sensitivity than BOULE by the Q-RT-PCR assessment (Table 1; logistic regression, P¼.006; odds ratio, 1.488; 95% CI for odds, ). Significant correlations (P<.001) were observed between the mean number of spermatocytes and spermatids per tubule and the levels of expression of BOULE and CDY1 by Q-RT-PCR (Table 2). Logistic regression confirmed that there was no advantage in quantitatively assessing both markers in combination to predict the presence of sperm cells and that the use of CDY1 alone was sufficient. DISCUSSION Testing gene expression may be potentially useful for predicting the presence of complete spermatogenesis. In the present study, we confirmed the usefulness of analyzing the transcripts CDY1 and BOULE in combination as effective candidates for predicting the presence of complete spermatogenesis. Both transcripts significantly predicted the presence of sperm cells. Although CDY1 had better sensitivity and specificity, assessing both transcripts by RT-PCR methodology alone yielded an additive efficacy. There was no improvement in the sensitivity and specificity of predicting the presence of sperm cells by using Q-RT-PCR assessment. As far as we know, this is the first time that the expression of BOULE and CDY1 has been meticulously assessed and compared in the same testicular biopsy using two methods of expression. CDY1 had better sensitivity to distinguish between complete maturation arrest and mixed atrophy. The different predictive values for BOULE and CDY1 (Table 1) could be explained by their distinct expression pattern observed in the testis. BOULE was transcribed in TABLE 1 Comparison between CDY1 and BOULE predictive values in the present study. Transcript tested Method Specimens tested, n Sensitivity, % Specificity, % CDY1 Q-RT-PCR RT-PCR BOULE Q-RT-PCR RT-PCR CDY1þ BOULE Q-RT-PCR RT-PCR Fertility and Sterility â 2299

4 FIGURE 2 Quantitative expression of CDY1 and BOULE (95% CI) on specimens with different testicular findings. Empty circles indicate mean value. Significant differences (P<.05) were observed among the groups except among those labeled with an asterisk. n ¼ number of specimens; CT ¼ threshold cycle. spermatocytes, and its protein was observed in leptotene spermatocytes as well as in later stages (16). On the other hand, CDY1 was transcribed in late spermatogenic stages, and its protein was observed in round spermatids (6, 12) Kleiman et al. CDY1 and BOULE in sperm retrieval Vol. 95, No. 7, June 2011

5 TABLE 2 Correlation between the average number of cells and the expression level (Q-RT-PCR) of CDY1 and BOULE genes. a Germ cells Correlation coefficient CDY1 BOULE Spermatocytes Spermatids Round spermatids Elongated spermatids a All correlations were significant (P<.001), Spearman s rho test. The present study correlated the BOULE and CDY1 expression observed in a single tiny testicular sample to the presence of sperm cells as evaluated by histological/cytological approaches in three different locations of the testis. A 100% sensitivity and specificity had been reported when a cutoff level of 0.5 in the receiver operating characteristic curve of BOULE semiquantitative expression was used (15).This high predictive value may had been influenced by the size of the group tested (42 specimens), the definition of sperm retrieval (the presence of sperm cells on wet preparation), and the use of an extracted sample from the pooling of all testicular biopsies taken for the purpose of BOULE detection. The observed additional statistical benefit of testing BOULE compared with testing CDY1 alone may be due to the four severe mixed atrophy samples in which CDY1 was not detected but BOULE was. The use of a sample from the pool of the three testicular biopsies taken from each testis may improve the predictive value of the test, an approach that had been applied by other investigators (15). Neither transcript was expressed in specimens with Sertoli cell only when assessed by RT-PCR and Q-RT-PCR methodology, in agreement with the findings of most previous reports (6, 7, 12, 26, 27), with the exception of that of Lin et al., who reported observing the expression of BOULE in Sertoli cell only (15). The only specimen with Sertoli cell only in which expression of BOULE was observed by RT-PCR in our current study had a very low level of expression by Q-RT-PCR. Although this level was 10 times higher than the mean value of the Sertoli cell only group, it was still very low compared with specimens from the other histological/cytological groups. A similar level of expression was observed in another specimen with mixed atrophy in which BOULE was not detected by the RT-PCR method, implying that these low levels might be in the lower limit of sensitivity of the RT-PCR method and might reflect the upper range of the basal level. BOULE transcript was repeatedly evaluated by both methods and not detected in one specimen with mixed atrophy, although other germ cells (RBM and DAZ) and sperm cells (CDY1) markers were detected. Similarly, BOULE protein was completely absent in another three biopsies with mixed atrophy (16). These exceptional cases might suggest that BOULE is not absolutely indispensable in humans and that complete spermatogenesis can be accomplished even when there is no evidence of its expression. These observations are in contrast to the crucial role of Boule in mice and Drosophila. The absence of Boule in mice still allowed the accomplishment of meiosis, but round spermatids did not progress beyond step six (19), and its inactivation in Drosophila blocked meiosis (14, 28). It was remarkable that in spite of the high sensitivity of Q-RT- PCR, it did not improve the predictive power of the test, while assessing both transcripts by RT-PCR had better predictive power and lower cost compared with Q-RT-PCR. Mature sperm cells cannot be found in about 40% of the azoospermic men undergoing TESE (20). Consequently, keeping a small testicular sample for a further molecular assessment in case of a failure to detect sperm cells in the first TESE procedure should be considered. The present study included 92 testicular specimens. Enlarging the sample size, especially that of the normal obstructive controls, would further strengthen our findings, given that the expression levels of CDY1 and BOULE in normal fertile men have not been reported before. Assessing the protein level of both CDY1 and BOULE when specific commercial antibodies become available should also contribute to a better understanding of impaired spermatogenesis. The findings of this study imply that a sensitive and accessible test (RT-PCR) can be useful for complementing the histological analysis in predicting successful testicular sperm retrieval. The information it yields will be valuable for counseling couples about the chance of finding sperm cells if an additional TESE procedure is considered. The availability of additional criteria will be of further benefit in the andrology clinic and will spare some patients from undergoing unnecessary repeated surgical procedures. Acknowledgments: The authors thank Dikla Ben-Moshe for her excellent technical work, Orli Sharon and Rona Limor (Institute of Endocrinology, Metabolism and Hypertension, Tel Aviv Sourasky Medical Center) for their kind assistance with real-time PCR, Esther Eshkol for editorial assistance, and Ilana Gelernter (Statistical Department, Tel Aviv University) for expert statistical analysis. REFERENCES 1. Reijo R, Lee TY, Salo P, Alagappan R, Brown LG, Rosenberg M, et al. Diverse spermatogenic defects in humans caused by Y chromosome deletions encompassing a novel RNA-binding protein gene. Nat Genet 1995;10: Reijo R, Alagappan RK, Patrizio P, Page DC. Severe oligozoospermia resulting from deletions of azoospermia factor gene on Y chromosome. Lancet 1996;347: Pryor JL, Kent-First M, Muallem A, Van Bergen A, Nolten WE, Meisner L, et al. Microdeletions in the Y-chromosome of infertile men. N Engl J Med 1997;336: Kleiman SE, Yogev L, Gamzu R, Hauser R, Botchan A, Lessing JB, et al. Genetic evaluation of infertile men. Hum Reprod 1999;14: Lahn BT, Page DC. Functional coherence of the human Y chromosome. Science 1997;278: Kleiman SE, Lagziel A, Yogev L, Botchan A, Paz G, Yavetz H. Expression of CDY1 may identify complete spermatogenesis. Fertil Steril 2001;75: Kleiman SE, Yogev L, Hauser R, Botchan A, Bar- Shira Maymon B, Schreiber L, et al. Members of the CDY family have different expression patterns: CDY1 transcripts have the best correlation with complete spermatogenesis. Hum Genet 2003;113: Lahn BT, Page DC. Retroposition of autosomal mrna yielded testis-specific gene family on human Y chromosome. Nat Genet 1999;21: Erratum in Nat Genet 1999;22: Kuroda-Kawaguchi T, Skaletsky H, Brown LG, Minx PJ, Cordum HS, Waterston RH, et al. The AZFc region of the Y chromosome features massive palindromes and uniform recurrent deletions in infertile men. Nat Genet 2001;29: Repping S, Skaletsky H, Lange J, Silber S, Van Der Veen F, Oates RD, et al. Recombination between palindromes P5 and P1 on the human Y chromosome causes massive deletions and spermatogenic failure. Am J Hum Genet 2002;71: Fertility and Sterility â 2301

6 11. Dorus S, Gilbert SL, Forster ML, Barndt RJ, Lahn BT. The CDY-related gene family: coordinated evolution in copy number, expression profile and protein sequence. Hum Mol Genet 2003;12: Lahn BT, Tang ZL, Zhou J, Barndt RJ, Parvinen M, Allis CD, et al. Previously uncharacterized histone acetyltransferases implicated in mammalian spermatogenesis. Proc Natl Acad Sci USA 2002;99: Caron C, Pivot-Pajot C, van Grunsven LA, Col E, Lestrat C, Rousseaux S, et al. Cdyl: a new transcriptional co-repressor. EMBO Rep 2003;4: Maines JZ, Wasserman SA. Post-transcriptional regulation of the meiotic Cdc25 protein Twine by the Dazl orthologue Boule. Nat Cell Biol 1999;1: Lin YM, Kuo PL, Lin YH, Teng YN, Nan Lin JS. Messenger RNA transcripts of the meiotic regulator BOULE in the testis of azoospermic men and their application in predicting the success of sperm retrieval. Hum Reprod 2005;20: Luetjens CM, Xu EY, Rejo Pera RA, Kamischke A, Nieschlag E, Gromoll J. Association of meiotic arrest with lack of BOULE protein expression in infertile men. J Clin Endocrinol Metab 2004;89: Lin YM, Chung CL, Cheng YS. Posttranscriptional regulation of CDC25A by BOLL is a conserved fertility mechanism essential for human spermatogenesis. J Clin Endocrinol Metab 2009;94: Mikhaylova LM, Boutanaev AM, Nurminsky DI. Transcriptional regulation by Modulo integrates meiosis and spermatid differentiation in male germ line. Proc Natl Acad Sci USA 2006;103: VanGompel MJ, Xu EY. A novel requirement in mammalian spermatid differentiation for the DAZ-family protein Boule. Hum Mol Genet 2010;19: Hauser R, Botchan A, Amit A, Ben Yosef D, Gamzu R, Paz G, et al. Multiple testicular sampling in non-obstructive azoospermia is it necessary? Hum Reprod 1998;13: Schrader M, Muller M, Schulze W, Heicappell R, Krause H, Straub B, et al. Quantification of the expression level of the gene encoding the catalytic subunit of telomerase in testicular tissue specimens predicts successful sperm recovery. Hum Reprod 2002;17: Schrader M, Muller-Tidow C, Ravnik S, Muller M, Schulze W, Diederichs S, et al. Cyclin A1 and gametogenesis in fertile and infertile patients: a potential new molecular diagnostic marker. Hum Reprod 2002;17: Yogev L, Segal S, Zeharia E, Gamzu R, Maymon BB, Paz G, et al. Sex chromosome alignment at meiosis of azoospermic men with azoospermia factor microdeletion. J Androl 2004;25: Kleiman SE, Yogev L, Hauser R, Botchan A, Maymon BB, Paz G, et al. Expression profile of AZF genes in testicular biopsies of azoospermic men. Hum Reprod 2007;22: Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc 2008;3: Kostova E, Yeung CH, Luetjens CM, Brune M, Nieschlag E, Gromoll J. Association of three isoforms of the meiotic BOULE gene with spermatogenic failure in infertile men. Mol Hum Reprod 2007;13: Kuo PL, Wang ST, Lin YM, Lin YH, Teng YN, Hsu CC. Expression profiles of the DAZ gene family in human testis with and without spermatogenic failure. Fertil Steril 2004;81: Eberhart CG, Maines JZ, Wasserman SA. Meiotic cell cycle requirement for a fly homologue of human Deleted in Azoospermia. Nature 1996;381: Kleiman et al. CDY1 and BOULE in sperm retrieval Vol. 95, No. 7, June 2011

7 SUPPLEMENTARY MATERIALS AND METHODS Patients The absence of microdeletions in one or more of the three AZF regions was verified by Y-chromosome microdeletion analysis (4) in 58 men who agreed to undergo the genetic test. The presence of the AZFc and -b regions in the remaining 34 men without Y-chromosome assessment was implied by the detection of reverse transcriptase polymerase chain reaction (RT-PCR) expression of DAZ and RBM (RNA-binding motif) genes localized at these two regions (6). RT-PCR The oligonucleotide primers sets that were used for DAZ, RBM, CDY1, and b-actin expression were described elsewhere (6). The oligonucleotide primer set used for BOULE detection was 5 0 -TCCCAGTATGGGTCTGT- GAA-3 0 and 5 0 -GCTGGACCAATGTTCAGCTT-3 0. The amplicon size was 164 bp. All transcripts except b-actin gave differential-sized PCR products for cdna and for genomic DNA (gdna). Since the same PCR product size was obtained for cdna and gdna with b-actin primers, they were tested in each case with and without the RT step to detect any gdna contamination. Q-RT-PCR The Q-RT-PCR assays were designed to amplify sequence stretches on the boundary of exons 1-2 and 6-7 of CDY1 and BOULE cdnas, respectively. The housekeeping PPIA (cyclophilin A) gene expression (commercial assay, Applied Biosystems) was used as an endogenous control to normalize for variations in cdna amounts. Expression of five housekeeping genes, PPIA, b-actin, RPLPO (ribosomal phosphoprotein), TBP (TATA box binding protein), and GAPDH (glyceraldehyde 3-phosphate dehydrogenase), was measured in biopsies with normal spermatogenesis (n ¼ 2) and Sertoli cell only (n ¼ 2). PPIA expression was found to be the most constant one between the two types of histologies, and it was used in further experiments. Relative gene expression analysis was performed according to the Comparative C T Method, 2 DCT (25). The threshold cycle (C T ) parameter is the PCR cycle at which the fluorescent signal of the reporter dye crossed an arbitrarily placed threshold. The C T is a logarithmic parameter and is inversely related to the number of copies of the gene studied. Relative gene expression levels (DC T ) of each sample were calculated by the formula: DC T ¼ mean C T target gene (CDY1 or BOULE) mean C T housekeeping gene (PPIA), allowing the comparison of samples independently of the amount of total input cdna. Fertility and Sterility â 2302.e1