Causes and Clinical Features of Infertile Men With Nonobstructive Azoospermia and Histopathologic Diagnosis of Hypospermatogenesis

Accepted Manuscript Title: Causes and Clinical Features of Infertile Men with Non-Obstructive Azoospermia and Histopathological Diagnosis of Hypospermatogenesis Author: Yu-Sheng Cheng, Chun-Wun Lu, Tsung-Yen Lin, Pei-Yu Lin, YungMing Lin PII: DOI: Reference:

S0090-4295(17)30283-2 http://dx.doi.org/doi: 10.1016/j.urology.2017.03.026 URL 20358

To appear in:

Urology

Received date: Accepted date:

29-11-2016 15-3-2017

Please cite this article as: Yu-Sheng Cheng, Chun-Wun Lu, Tsung-Yen Lin, Pei-Yu Lin, YungMing Lin, Causes and Clinical Features of Infertile Men with Non-Obstructive Azoospermia and Histopathological Diagnosis of Hypospermatogenesis, Urology (2017), http://dx.doi.org/doi: 10.1016/j.urology.2017.03.026. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Causes and Clinical Features of Infertile Men with Non-Obstructive Azoospermia and Histopathological Diagnosis of Hypospermatogenesis

Yu-Sheng Chenga,b, Chun-Wun Lua, Tsung-Yen Lina, Pei-Yu, Lina, Yung-Ming Lina

a

Department of Urology, bGraduate Institute of Clinical Medicine, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan

Correspondence Yung Ming Lin, MD, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, 138, Sheng Li Road, Tainan, Taiwan Tel: 886-6-276-6179; Fax: 886-6-238-3678; E-mail: [email protected]

Acknowledgments: This work was supported by the Ministry of Science and Technology of Taiwan (100-2314-B-006-017). Computational analyses and data mining was provided by the Bioinformatics Core, National Cheng Kung University, Tainan, Taiwan.

Word counts Abstract: 242 Manuscript text: 2993

1

Page 1 of 22

ABSTRACT Objective: To analyze the causes and the clinical features of infertile men with non-obstructive azoospermia (NOA) and hypospermatogenesis (HS). Methods: This retrospective cohort study included 100 patients with NOA and HS and eight patients with obstructive azoospermia and normal spermatogenesis. The severity of HS was subdivided into three groups (mild, moderate, and severe) based on spermatogenic score. Data of history, physical findings, serum hormone profiles, genetic studies and sperm retrieval rate (SRR) were collected. Whole genome DNA methylation analysis and microarray mRNA expression analysis were used to identify the candidate genes of methylation dysregulation in HS. Results: Thirty-two (32%) patients had at least one prior/current testicular insults and 13 (13%) patients had genetic anomalies. Fifty-five (55%) patients were categorized as idiopathic HS. Patients with mild HS had a higher frequency of testicular insults, and patients with severe HS had a significantly higher frequency of genetic anomalies. SRR was 100%, 100% and 88.4% for patients with mild, moderate and severe HS, respectively. Four sterility-related genes, including BOLL, DDX4, HORMAD1and MAEL, were found to have increased methylation at CpGs of the promoter regions and decreased mRNA expressions in HS testis. Conclusions: The causes of HS are complex and multifactorial. The main causes of HS were prior or current testicular insults and chromosomal or genetic anomalies. More than half of the patients were categorized as idiopathic HS. With high throughput analysis, methylation dysregulations of BOLL, DDX4, HORMAD1and MAEL are believed to be associated with HS.

Key words: azoospermia; hypospermatogenesis; genetic anomaly; methylation 2

Page 2 of 22

INTRODUCTION Non-obstructive azoospermia (NOA) is a failure of spermatogenesis intrinsic to the testis and accounts for approximately 60% of infertile men with azoospermia.1 The histopathologies of testicular biopsy in NOA can be subcategorized into hypospermatogenesis (HS), maturation arrest (MA), Sertoli cell-only syndrome (SCOS), and tubular sclerosis. HS is the most common histopathological pattern of NOA in different studies, with prevalence of up to 55.8%.2,3 HS is histopathologically characterized by the presence of all stages of spermatogenesis but a decline in the number of germ cells. Some researchers have hypothesized that the imbalance of cell proliferation and apoptosis of spermatogonia is responsible for the pathogenesis of HS.4,5

Clinically, the causes of HS seem to be heterogeneous, with chromosomal and genetic anomalies, endocrinopathy, developmental disorders, metabolic disorders, heat exposure, varicocele, infection, and environmental and cytotoxic factors interacting to cause the disease. 6

To date, testicular sperm retrieval is currently used to treat patients with NOA and HS, and

the sperm retrieval rates (SRR) reported in the literature vary from 81% to 100%.7,8 For patients with NOA and HS, there remain some issues to be explored. First, the distribution of causes of HS has yet to be studied. Secondly, it is unclear whether the causes are related to the severity of HS. Third, novel aberrant genetic or epigenetic events involved in HS await further investigation.

3

Page 3 of 22

To date, there is accumulating evidence disclosing aberrant expressions of certain reproduction-related genes in male infertility; however, specific mutation/polymorphism of DNA sequence could not fully explain the underlying mechanisms. It is reasonable to believe that epigenetic dysregulation of reproduction-related genes may contribute to the causes of HS. Because HS is highly prevalent in NOA, we retrospectively reviewed the causes, clinical characteristics and sperm retrieval outcomes in 100 patients with NOA and HS. We stratified these patients into three groups based on the severity of spermatogenic defects and analyzed the differences between the groups. Moreover, we studied the genome-wide DNA methylation pattern in human testis, and tried to uncover the possible role of hypermethylation of reproduction-related genes in HS.

MATERIAL AND METHODS Selecting NOA patients with HS This study was approved by the Institutional Review Board of National Cheng Kung University Hospital, Tainan, Taiwan (BR-99-140 and A-ER-103-208). Two hundred thirty-two infertile men with NOA who had complete medical records and had undergone sperm retrieval were retrospectively reviewed and analyzed. Patients with biopsy-documented HS but without a concurrent seminal tract anomaly or obstruction were enrolled in this study. The term “hypospermatogenesis” is defined as all stages of spermatogenesis are present, but there is a decline in the number of germ cells. This definition includes varying patterns: uniform reduction in the number of late spermatids in all 4

Page 4 of 22

tubules; reduction in the number of late spermatids in tubules showing no progression past early (round) spermatids, spermatocytes, or spermatogonia; reduction in the number of late spermatids in tubules showing no germ cells; reduction in the number of late spermatids in hyalinized tubules; and a discordant pattern in the bilateral testes with a reduction in the number of late spermatids on one side.9,10

History taking, physical examination, and laboratory workup All the enrollees had undergone a detailed history taking, physical examination, bilateral testicular volume measurement, serum reproductive hormone testing, chromosome analysis, and Y-chromosome microdeletion screening. History of testicular insults included varicocele, heat exposure, genitourinary tract infection/inflammation, undescended testis and testicular trauma. The diagnosis of varicocele is made by physical examination, and is clinically classified into three grades.11 Heat exposure is defined as current occupational exposure for a duration of at least 20 hours per week and at least 3 months, such as welder, cook, baker, driver and fireman.12 History of genitourinary tract infection/inflammation includes past and current infection/inflammation of the testis, epididymis, prostate, and urethra. Testicular volume was measured using an orchidometer (Seager, Switzerland).

Testicular biopsy and spermatogenic score All the enrollees had undergone testicular sperm retrieval and testicular biopsy. The biopsy was done before or simultaneously with sperm retrieval. For patients with varicocele and 5

Page 5 of 22

NOA, they were advised to receive varicocele repair simultaneously with testicular biopsy. Then, the patients were followed up for at least 12 months, and testicular sperm retrieval was performed in those cases of persistent azoospermia. We harvested 3  3  3-mm testicular tissue from each testis. The Bouin’s-fixed, paraffin-embedded specimen was sent for 4-µm-thick sectioning and hematoxylin and eosin (H&E) staining. This method allowed us to examine more than 100 cross-sections of seminiferous tubules for each patient. Two specialists (one pathologist and one andrologist) reviewed all testicular slices. The severity of spermatogenic failure was determined by a spermatogenic scoring system with slight modification.13 Briefly, the percentage of all tubules with elongated spermatid were calculated. The spermatogenic score represents the [number of tubules containing elongated spermatids]  [total number of tubules evaluated]. In this study, we sub-classified our patients into three groups based on the spermatogenic score: ≥ 0.8 = mild HS;  0.4 to < 0.8 = moderate HS; < 0.4 = severe HS.

Genome-wide methylation profiling and microarray mRNA expression analysis of human testes To further investigate the role of methylation dysregulation in HS, testicular specimens from patients without testicular insults or genetic anomaly were collected for subsequent DNA methylation profiling, mRNA expression profiling, quantitative real-time RT-PCR (qRT-PCR) and pyrosequencing analysis. A total of 17 patients were enrolled in this study. In addition, nine patients with obstructive azoospermia (OA) and histologically proved normal 6

Page 6 of 22

spermatogenesis (NR) were enrolled as controls. NR is defined as full spermatogenesis in the entire biopsy, thin basement membrane and tunica propria, and the presence of a normal interstitial tissue.10 Informed consent was obtained from each patient. One pair of testicular samples, one patient with NR with spermatogenic score of 1 and one patient with moderate HS with spermatogenic score of 0.58, were used for whole genome DNA methylation analysis and microarray mRNA expression analysis. To diminish the effect of different cell compositions in a testicular sample, isolation of germ cell-enriched fraction from testis was performed prior to high throughput study. The isolation procedure which is based on enzymatic digestion and countercurrent centrifugal elutriation was performed according to previous report with slight modification.14 DNA methylation profiling of NR and HS germ cell-enriched cells was performed using Agilent custom 1M Promoter-CpG island Microarray (Agilent Technologies, Santa Clara, CA), which covers CpGs in all promoter areas 4500 bp upstream and 1500 bp downstream from transcription start site. Messenger RNA expression profiling of NR and HS germ cells was performed using Agilent SurePrint G3 Human GE 8 x 60 K Microarray (Agilent).

Identification of candidate gene by bioinformatics analysis Genes with increased DNA methylation and decreased mRNA expression in HS versus NR testes were exported for bioinformatics analysis. We used the MetaCoreTM software (GeneGo, MI, USA) for gene ontology enrichment analysis, which provides a ranked representation of ontologies that are highly-saturated among the given gene list. A literature review on the top 7

Page 7 of 22

ten comprehensive processes was done. Those genes, which were known as male reproduction-related in mammalians, were selected as candidate genes for further validation.

Quantitative real-time RT-PCR The other eight patients with NR due to obstruction and 16 patients with HS were enrolled for qRT-PCR and subsequent pyrosequencing analysis. The qRT-PCR was performed, by using germ cell-enriched cells, to determine the transcript level of each candidate gene.15 The primer sequences are available upon request. The standard curve method was used for quantification, and the steady-state concentration of mRNA for gene in each testicular sample was normalized to the amount of r18S mRNA (endogenous control).

DNA methylation analysis of selected promoter regions Pyrosequencing was used to validate the Agilent custom 1M Promoter-CpG island Microarray data. DNA was purified from germ cell-enriched samples by using QIAamp DNA Mini kit (Qiagen, Valencia, CA). EpiTect Bisulfite Kit (Qiagen) was used for bisulphite conversion. Primers were designed by using the online program Methprimer (http://www.urogene.org/methprimer). All primer sequences contain at least one CpG at 3′ end to facilitate maximal discrimination between methylated and unmethylated DNA sequences. The primer sequences are available upon request. Pyrogram was obtained with the Pyromark Q24 2.0.6 software program (Qiagen). The degree of methylation at each CpG was determined by the ratio of C to T (percentage of methylation). The mean percent methylation of each CpG was used to compare the difference between two groups. 8

Page 8 of 22

Statistical analysis Data were analyzed with Prism 5 (GraphPad Software, San Diego, CA). Continuous variables are presented as the mean  standard error (SE). Differences between variables within three groups were examined using One-way ANOVA or 2 tests.

RESULTS Clinical features of patients with NOA and HS One hundred (43.1%) of the 232 patients with NOA had histopathologically proved HS, and the clinical characteristics of these100 patients are shown in Table 1. Thirty-two patients had at least one prior or current testicular insult, and the most common testicular insult was varicocele, followed by heat exposure, genitourinary tract infection/inflammation, undescended testis and testicular trauma. Seven patients had abnormal chromosome karyotypes (47, XXY), and another six patients had Y-chromosome microdeletion (AZFc deletion). The incidence of overall genetic anomalies in HS patients was 13%. Spermatozoa were retrieved in 95 of 100 patients. These patients were further subdivided into three groups based on spermatogenic score, and the representative histological images and the clinical characteristics are shown in Supplementary Figure S1 and Table 2, respectively. There was no significant difference in the age distribution among the three groups. The frequency of prior or current testicular insults was significantly higher in patients with mild HS (P = .042), and varicocele was most significantly related to the difference. However, no significant 9

Page 9 of 22

difference in the varicocele grade distribution was noted (P = .373 for grade I; P = .316 for grade II; P = . 208 for grade III). A significant difference in testicular volumes was noted among the three groups (P < .0001). Patients with severe HS presented with significantly elevated serum FSH (P < .0001) and LH (P = .006) levels; however, no significant difference in serum T or PRL levels was noted. One patient with moderate HS and six patients with severe HS had abnormal chromosome karyotypes (47, XXY), but none of the patients with mild HS did (P = .054). Similarly, two patients with moderate HS and four patients with severe HS had AZFc microdeletions, but none of the patients with mild HS did (P = .276). The frequency of overall genetic anomalies was significantly lower in patients with mild HS (P = .016). The mean operation time was significantly longer in patients with severe HS (P < .0001), and higher frequency of bilateral surgeries was also noted in severe HS (P = .004). Spermatozoa were retrieved in 27 of 27 patients with mild HS, 30 of 30 patients with moderate HS, and 38 of 43 patients with severe HS (P = .031). All five patients with failed sperm retrieval presented with a spermatogenic score of less than 0.1.

Differential DNA methylation and mRNA expression in testes from HS versus NR To search the candidate genes regulated by DNA methylation in HS, we compared the DNA methylation and mRNA expression profiling from the testes of HS and NR. Figure 1a shows the stepwise approach for the identification of candidate genes. In this study, genes with both absolute methylation difference ( ≥ 5%) and significant mRNA expression difference (log 2 fold change ≤ -1) were filtered. A total of 255 genes were characterized as hypermethylated in 10

Page 10 of 22

the promoter DNA and low expressed in HS testis. Bioinformatics analysis suggested that these 255 genes were significantly enriched for several comprehensive processes. The top ten gene ontologies are shown in Supplementary Table S1. Then, we performed an intensive literature review, and 14 genes (ANKRD36, BOLL, CRISP2, DAZL, DDX3Y, DDX4, DMRTC2, HIST1H1T, HORMAD1, MAEL, RAN, RPS21, SOHLH2, TCP11) with known biological functions in male reproduction in mammalians were identified.

Validation of mRNA expression and DNA methylation We proceeded to use qRT-PCR to measure the transcript levels and pyrosequencing to determine the methylation levels of these 14 genes in 16 patients with HS and 8 patients with NR. We focused on looking at the genes with differences in methylation among those displaying difference in mRNA expression, implying that the promoter hypermethylation occurring in testis may affect their expression. Of the 14 genes, four genes, including BOLL (boule homolog, RNA binding protein), DDX4 (DEAD-box helicase 4), HORMAD1(HORMA domain containing 1) and MAEL (maelstrom spermatogenic transposon silencer) reached our criterion selection (Figure 1b and 1c). A total of eight hypermethylated regions among the four genes were detected in our HS testes. MAEL had up to five adjacent segments and BOLL, DDX4, HORMAD1 had only one segment representing hypermethylation in their promoters. The details of chromosome location, differentially methylated CpG site and known male reproduction-related functions of these four genes are shown in Table 3. 11

Page 11 of 22

COMMENT We reported the clinical features and sperm retrieval outcomes of 100 patients with NOA and HS, which is the largest series published to date. HS accounts for 43.1% of patients with NOA in our infertility clinic. In the literature, the reported prevalence of HS varied widely from 13 to 55.8%.2,3 In this study, the definition of HS is clearly stipulated as consisting of five histopathological patterns. All testicular specimens were reviewed by 2 independent specialists blinded to the patients’ medical histories. We believe that our result is closer to the real situation, and further meta-analyses should provide reasonable estimates of HS prevalence in NOA. The frequency of prior or current exposure to testicular insults, especially in patients with varicocele, was higher in patients with mild HS. In contrast, the frequency of overall genetic anomalies was significantly higher in patients with severe HS. These findings suggest that severe HS is largely caused by primary genetic anomalies, and mild HS, by acquired testicular insults.

The overall SRR in the present study was 95%. This is similar to the findings of other studies reporting SRRs of HS that ranged from 81% to 100%.7,8,16 Thus, biopsy-proved HS does not ensure a successful microTESE. The inconsistency between histological findings and the results of sperm retrieval may be due to the removal of vital foci of spermatogenesis for diagnostic biopsy in an already severely defective testis or due to a limited surgeon and embryologist experience. In this study, all five patients with failed sperm retrieval presented 12

Page 12 of 22

with severe HS (spermatogenic score < 0.1) that had mixed histological patterns of minute focal spermatogenesis and SCOS or MA. Our findings suggest that the probability of sperm recovery depends upon the severity of HS, and that the status of the tubules retrieved was much more critical than were the histological pattern.

The underlying causes of NOA and HS are complex and multifactorial. Thirty-two of our 100 patients had prior or current testicular insults and 13 had chromosomal or genetic anomalies. Therefore, we identified possible causes of HS in 45 patients but thought that the other 55 (55%) patients had idiopathic HS. Currently, the genetic tests for patients with NOA are limited to chromosome karyotype and Y-chromosome deletion test. In the literature, a number of mutations and single nucleotide polymorphisms of autosomal genes or X-linked genes and Y-chromosome structural variants have been reported to be associated with spermatogenic failure.17 In the post-Human Genome Project era, epigenetic regulation deciphers the phenomenon of gene inactivation without sequence change, and methylation of CpG island in promoter region is known to repress gene expression. Recent studies have revealed that aberrant DNA methylation of not only imprinted genes but also non-imprinted developmental genes are related to spermatogenic failure. For example, promoter methylations of methylenetetrahydrofolate reductase (MTHFR) and discoidin domain receptor 1 (DDR1) have been shown to be associated with NOA.18,19 In this study, we propose that methylation dysregulations of BOLL, DDX4, HORMAD1and MAEL genes may be involved in the pathophysiology of HS. When comparing to normal spermatogenesis, 13

Page 13 of 22

increased methylation at CpGs of the promoter regions and decreased mRNA expression of these four genes were found in HS testes. All of them are exclusively expressed in the germ cell lineage and are recognized as reproduction-related genes.20-26

Aberrant DNA methylation provides insight to the mechanism of spermatogenic failure. Our and previous studies confirm the existence of different testicular methylation pattern between NOA and OA groups,19,27 suggesting that specific DNA hyper- or hypo-methylation may be useful in the prediction of testicular phenotype as well as successful sperm retrieval. Recently, cell-free DNA from seminal fluid has been found to be an alternative to testicular tissue for the evaluation of methylation aberration of testis-expressed genes.28 Significant correlations were observed between methylation percentage in testicular DNA and paired cell-free seminal DNA. Thus, it is reasonable to expect that non-invasive DNA methylation biomarker may serve to clinical application in the future.

Our study revealed that DNA methylation of four testis-expressed genes seems to be a valuable indicator in HS; however, several issues remain to be studied. First, it is of great interest to investigate the methylation variation of these four genes in NOA patients with different testicular phenotype, such as MA and SCOS. Composition of the cell types, cell-specific gene expression, alternation in genomic imprinting and dynamic methylation change might be responsible for the different methylation status among the phenotypes. Secondly, given the requirement of reference values for DNA methylation in clinical practice, 14

Page 14 of 22

the temporal stability and inter-individual variability of DNA methylation should be determined. Third, if DNA methylation is a key factor in spermatogenesis, methylation dysregulation of germ cells may influence the outcomes of subsequent assisted reproductive technology. Thus, further studies in the transgenerational inheritance of epigenetic genes, fertilization rate, embryo development, congenital abnormalities and pregnancy outcome are needed.

CONCLUSIONS We reviewed a subset of patients with NOA who presented with varied severity of HS. Severe HS was caused mostly by primary genetic anomalies, and mild HS was more likely caused by acquired testicular insults. Biopsy-proved HS might not always yield a successful sperm retrieval. With high throughput analysis, we suggest that epigenetic dysregulation of BOLL, DDX4, HORMAD1and MAEL may be associated with HS.

15

Page 15 of 22

Figure 1. Identification of epigenetic dysregulated genes and validations of their methylation and mRNA expression levels. (A) Schematic diagram shows stepwise approach to identify the potential epigenetic dysregulated genes in hypospermatogenesis. (B) The mRNA expression levels of identified genes (BOLL, DDX4, HORMAD1 and MAEL) were determined by qRT-PCR. r18S is used as internal control. (C) The methylation levels of each CpG was measured by pyrosequencing analysis. Mean percent methylation was used to compare the difference between HS and NR groups. The asterisk indicates significant difference (P < 0.05). HS, hypospermatogenesis; NR, normal spermatogenesis.

Supplementary Figure S1. Representative histopathologies of testicular specimens with various levels of severity of HS. The scoring system we used for quantitatively evaluating spermatogenesis is based on the percentage of seminiferous tubules showing elongated spermatids. (A and B) The spermatogenic score of this specimen was evaluated as 0.94. B is a magnification of A. (C and D) Spermatogenic score is 0.77. D is a magnification of C. (E and F) Spermatogenic score is 0.51. F is a magnification of E. (G and H) Spermatogenic score is 0.25. H is a magnification of G. Scale bar = 100 m

16

Page 16 of 22

References 1. 2.

3.

4.

5.

6. 7. 8.

9. 10.

11.

12.

13. 14.

15.

Matsumiya K, Namiki M, Takahara S, Kondoh N, Takada S, Kiyohara H and Okuyama A: Clinical study of azoospermia. Int J Androl. 17: 140-2, 1994. Abdullah L and Bondagji N: Histopathological patterns of testicular biopsy in male infertility: A retrospective study from a tertiary care center in the western part of Saudi Arabia. Urol Ann. 3: 19-23, 2011. Haddad FH, Omari AA, Malkawi OM, Ajour WK, Izat A, Khasrof H and Khawaldeh AK: Patterns of testicular cytology in men with primary infertility: any change since the Gulf War? Acta Cytol. 48: 807-12, 2004. Takagi S, Itoh N, Kimura M, Sasao T and Tsukamoto T: Spermatogonial proliferation and apoptosis in hypospermatogenesis associated with nonobstructive azoospermia. Fertil Steril. 76: 901-7, 2001. Hentrich A, Wolter M, Szardening-Kirchner C, Luers GH, Bergmann M, Kliesch S and Konrad L: Reduced numbers of Sertoli, germ, and spermatogonial stem cells in impaired spermatogenesis. Mod Pathol. 24: 1380-9, 2011. Cerilli LA, Kuang W and Rogers D: A practical approach to testicular biopsy interpretation for male infertility. Arch Pathol Lab Med. 134: 1197-204, 2010. Ramasamy R and Schlegel PN: Microdissection testicular sperm extraction: effect of prior biopsy on success of sperm retrieval. J Urol. 177: 1447-9, 2007. Tsujimura A, Matsumiya K, Miyagawa Y, Tohda A, Miura H, Nishimura K, Koga M, Takeyama M, Fujioka H and Okuyama A: Conventional multiple or microdissection testicular sperm extraction: a comparative study. Hum Reprod. 17: 2924-9, 2002. Bergmann M and Kliesch S: Testicular Biopsy and Histology. 155-167, 2010. McLachlan RI, Rajpert-De Meyts E, Hoei-Hansen CE, de Kretser DM and Skakkebaek NE: Histological evaluation of the human testis--approaches to optimizing the clinical value of the assessment: mini review. Hum Reprod. 22: 2-16, 2007. Jungwirth A, Giwercman A, Tournaye H, Diemer T, Kopa Z, Dohle G, Krausz C and European Association of Urology Working Group on Male I: European Association of Urology guidelines on Male Infertility: the 2012 update. Eur Urol. 62: 324-32, 2012. De Fleurian G, Perrin J, Ecochard R, Dantony E, Lanteaume A, Achard V, Grillo JM, Guichaoua MR, Botta A and Sari-Minodier I: Occupational exposures obtained by questionnaire in clinical practice and their association with semen quality. J Androl. 30: 566-79, 2009. Bergmann M, Kliesch S (1998) Hodenbiopsie In: Krause W, Weidner W (eds) Andrologie. Enke Verlag, Stuttgart, pp 66-71 Chang YF, Lee-Chang JS, Panneerdoss S, MacLean JA, 2nd and Rao MK: Isolation of Sertoli, Leydig, and spermatogenic cells from the mouse testis. Biotechniques. 51: 341-2, 344, 2011. Lin YM, Chung CL and Cheng YS: Posttranscriptional regulation of CDC25A by BOLL is a conserved fertility mechanism essential for human spermatogenesis. J Clin 17

Page 17 of 22

16.

17. 18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

Endocrinol Metab. 94: 2650-7, 2009. Okada H, Dobashi M, Yamazaki T, Hara I, Fujisawa M, Arakawa S and Kamidono S: Conventional versus microdissection testicular sperm extraction for nonobstructive azoospermia. J Urol. 168: 1063-7, 2002. Massart A, Lissens W, Tournaye H and Stouffs K: Genetic causes of spermatogenic failure. Asian J Androl. 14: 40-8, 2012. Khazamipour N, Noruzinia M, Fatehmanesh P, Keyhanee M and Pujol P: MTHFR promoter hypermethylation in testicular biopsies of patients with non-obstructive azoospermia: the role of epigenetics in male infertility. Hum Reprod. 24: 2361-4, 2009. Ramasamy R, Ridgeway A, Lipshultz LI and Lamb DJ: Integrative DNA methylation and gene expression analysis identifies discoidin domain receptor 1 association with idiopathic nonobstructive azoospermia. Fertil Steril. 102: 968-973 e3, 2014. Kee K, Angeles VT, Flores M, Nguyen HN and Reijo Pera RA: Human DAZL, DAZ and BOULE genes modulate primordial germ-cell and haploid gamete formation. Nature. 462: 222-5, 2009. Castrillon DH, Quade BJ, Wang TY, Quigley C and Crum CP: The human VASA gene is specifically expressed in the germ cell lineage. Proc Natl Acad Sci U S A. 97: 9585-90, 2000. Sugimoto K, Koh E, Sin HS, Maeda Y, Narimoto K, Izumi K, Kobori Y, Kitamura E, Nagase H, Yoshida A et al.: Tissue-specific differentially methylated regions of the human VASA gene are potentially associated with maturation arrest phenotype in the testis. J Hum Genet. 54: 450-6, 2009. Shin YH, Choi Y, Erdin SU, Yatsenko SA, Kloc M, Yang F, Wang PJ, Meistrich ML and Rajkovic A: Hormad1 mutation disrupts synaptonemal complex formation, recombination, and chromosome segregation in mammalian meiosis. PLoS Genet. 6: e1001190, 2010. Daniel K, Lange J, Hached K, Fu J, Anastassiadis K, Roig I, Cooke HJ, Stewart AF, Wassmann K, Jasin M et al.: Meiotic homologue alignment and its quality surveillance are controlled by mouse HORMAD1. Nat Cell Biol. 13: 599-610, 2011. Costa Y, Speed RM, Gautier P, Semple CA, Maratou K, Turner JM and Cooke HJ: Mouse MAELSTROM: the link between meiotic silencing of unsynapsed chromatin and microRNA pathway? Hum Mol Genet. 15: 2324-34, 2006. Soper SF, van der Heijden GW, Hardiman TC, Goodheart M, Martin SL, de Boer P and Bortvin A: Mouse maelstrom, a component of nuage, is essential for spermatogenesis and transposon repression in meiosis. Dev Cell. 15: 285-97, 2008. Ferfouri F, Boitrelle F, Ghout I, Albert M, Molina Gomes D, Wainer R, Bailly M, Selva J and Vialard F: A genome-wide DNA methylation study in azoospermia. Andrology. 1: 815-21, 2013. Wu C, Ding X, Tan H, Li H and Xiong C: Alterations of testis-specific promoter methylation in cell-free seminal deoxyribonucleic acid of idiopathic nonobstructive azoospermic men with different testicular phenotypes. Fertil Steril. 106: 1331-1337, 18

Page 18 of 22

2016.

19

Page 19 of 22

Table 1. Clinical characteristics of 100 patients with NOA and HS Number Characteristic of patients Age (years) History of at least one testicular insults 32 Varicocele 15 Grade I 6 Grade II 6 Grade III 3 Heat exposure 14 Genitourinary tract infection/inflammation 4 Undescended testis 3 Testicular trauma 3 Testicular volume (mL) Right Left Hormonal profilea FSH (mIU/mL) LH (mIU/mL) T (ng/mL) PRL (ng/mL) Overall genetic anomaly 13 Karyotype abnormality 47,XXY 7 Y-chromosome microdeletion

Mean ± SE 33.2 ± 0.5

10.4 ± 0.8 10.4 ± 0.7 19.4 6.8 4.4 12.2

± ± ± ±

1.9 0.6 0.4 1.2

AZFc deletion 6 Successful sperm retrieval 95 a: Reference values: FSH: 1.1-13.5 mIU/mL; LH: 0.4-5.7 mIU/mL; T: 2.6-15.9 ng/mL; PRL: 3.1-16.5 ng/mL

20

Page 20 of 22

Table 2. Comparison of clinical characteristics of three subgroups based on different histological findings Moderate Severe Mild HS HS HS (n = 27) (n = 30) (n = 43) P Clinical Characteristic Age (years) 33.2 ± 0.8 33.1 ± 0.7 33.3 ± 0.8 .972 History of at least one testicular insult 12 (44.4) 12 (40.0) 8 (18.65) .042 (n (%)) Varicocele (n (%)) 7 (25.9) 6 (20.0) 2 (4.7) .035 Grade I 2 (7.4) 3 (10.0) 1 (2.3) .373 Grade II 3 (11.1) 2 (6.7) 1 (2.3) .316 Grade III 2 (7.4) 1 (3.3) 0 (0) .208 Heat exposure (n (%)) 5 (18.5) 5 (16.7) 4 (9.3) .491 Genitourinary tract 1 (0) 2 (6.7) 1 (2.3) .876 infection/inflammation (n (%)) Undescended testis (n (%)) 0 (0) 1 (3.3) 2 (4.7) .535 Testicular trauma (n (%)) 1 (3.7) 1 (3.3) 1 (2.3) .940 Testicular volume (mL) Right side 16.7 ± 0.8 11.0 ± 1.4 6.5 ± 0.8 < .0001 Left side 16.5 ± 0.7 10.6 ± 1.4 6.6 ± 0.9 < .0001 Hormonal profile FSH (mIU/mL) 8.1 ± 1.5 17.3 ± 2.9 27.8 ± 3.1 < .0001 LH (mIU/mL) 3.9 ± 0.7 6.5 ± 1.0 8.9 ± 1.1 .006 T (ng/mL) 4.0 ± 0.4 4.6 ± 0.5 4.6 ± 0.9 .869 PRL (ng/mL) 13.4 ± 2.9 12.6 ± 2.1 10.9 ± 1.5 .707 Overall genetic anomaly (n (%)) 0 (0) 3 (10) 10 (23.3) .016 Karyotype abnormality 47,XXY (n (%)) 0 (0) 1 (3.3) 6 (14.0) .054 Y-chromosome microdeletion AZFc deletion (n (%)) 0 (0) 2 (6.7) 4 (9.3) .276 Testicular sperm retrieval Operation time (mins) 67.2 ± 2.9 84.7 ± 3.7 97.5 ± 4.1 < .0001 Unilateral and bilateral procedure .004 (n (%)) Unilateral 27 (100) 26 (86.7) 30 (69.8) Bilateral 0 (0) 4 (13.3) 13 (30.2) Successful sperm retrieval (n (%)) 27 (100) 30 (100) 38 (88.4) .031 The significance threshold was set at P < .05 (boldface numbers).

21

Page 21 of 22

Table 3.

Hypermethylated probe sequences in promoter region of four genes and their known reproduction-related functions Known reproduction-related function Function in meiotic G2/M transition; DAZ gene family

Gene

Chromosome location

Sequences and CpG position

BOLL

Chr 2:198651614-198651673

CAGCCTTGGGATCGCCTTTGGAAGAAGATTGCGCT TTTTTTCTTAAATTAATGAATTAAT

DDX4

Chr 5:55033796-55033840

TAGTCACCAGCCAATCGTCAACAGACGCCATTTGTT GTTGGAGCC

Function in germ cell development; potential sperm motility marker

HORMAD1

Chr 1:150693387-150693431

ACAGTGGCTTTTTGACCCCTGGTCGAGGTCACCTTT CCCGCCAAG

Function in formation of synaptonemal complex

MAEL

Chr 1:166958215-166958261

GTAAAACGCGGAAACACCCGCAGCTCCTCTTTTTCT CTACCTACTTT AGCTCCTCTTTTTCTCTACCTACTTTGTCGCACATTA CGCAATAATGTACAGAAATTGCT ACGCACCCAGCCAATCAGAGCACTTGGCACCTGCG ACGGCGCTCT CAAGGCGGCACGAGCCGGAATCTTCCAGTCTCAGG CTGTTTGTTC CGAGAGTTGCGGGCTGCGTGCGCAGGCGCCTACC TCTGTTACTTAG

Function in piRNA pathway to repress transposable elements during meiosis

Chr 1:166958236-166958295 Chr 1:166958330-166958374 Chr 1:166958391-166958435 Chr 1:166958478-166958522

22

Page 22 of 22