Abstract Study question Is there an association between genetic polymorphisms in LIN28B with LIN28B expression and semen quality in spermatozoa from oligozoospermic men? Summary answer LIN28B is downregulated in spermatozoa of oligozoospermic men and explains 39% of sperm count variability. One SNP (rs314280) was associated with low LIN28B mRNA expression. What is known already Large-scale genomic studies have identified genetic variants in or near LIN28B to be robustly associated with pubertal traits such as age at menarche and age at voice break. The LIN28 family of genes are known to control cell division, growth and differentiation. The LIN28B orthologue is exclusively expressed at high levels in the testis and placenta; however, we have limited knowledge about the role of LIN28B in relationship to testicular function. Interestingly, other studies have showed that later onset of puberty is associated with poor semen quality, suggesting that pubertal development may influence adult testicular function. Study design, size, duration Twenty-nine oligozoospermic (cases) and fifty-four men with normal sperm concentration (controls) were recruited from patients seeking andrological work-up for male/couple infertility at a University Hospital. Subjects with abnormal karyotype, Yq-microdeletions, varicocele GIII-IV, hypogonadotropic hypogonadism, seminal infection, diabetes, BMI>35, excessive consumption of alcohol/drugs, testicular cancer, chemotherapy/radiotherapy were excluded. Semen samples were obtained for analysis (WHO, 2010) and sperm isolation by density gradient (PureSperm40®). Peripheral blood samples were analysed for reproductive hormones and used for DNA extraction. Participants/materials, setting, methods LIN28B and GAPDH-S mRNA abundance were measured by qRT-PCR (2-ΔCT method) in RNA isolated from purified sperm using SYBR® MasterMix (Takyon). Five SNPs in linkage disequilibrium (rs7759938, rs395962, rs314268, rs314277 and rs314280) were genotyped (TaqMan™ SNP Genotyping Assay). Statistical differences between groups were assessed by Mann-Whitney test and genetic association was analysed using “SNPassoc” package in R. Values are showed as median, Q1-Q3. Main results and the role of chance Cases and controls had similar age and BMI but as expected sperm concentration, morphology and vitality were differed between groups (9.9, 4.9-13.1 vs 82.8, 46.0-146.5 mill/ml; 1, 1-2 vs 3, 1-4% normal forms and 83%, 76-89 vs 86%, 82-90, respectively) (p < 0.05). No differences were observed in progressive motility, semen volume and days of abstinence. FSH and LH serum levels were higher in cases compared with controls (4.6, 2.6-7.8 vs 3.2, 2.5-4 mIU/ml and 4.2, 2.9-5.2 vs 3.1, 2.1-4.2 mIU/ml, respectively; p = 0.019 and p = 0.017). Relative expression of LIN28B was significantly diminished in cases compared with controls (0.0034, 0.0015-0.0103 vs 0.0446, 0.0122-0.1026; p < 0.001), while expression of the internal housekeeping control GAPDH-S was similar between groups. Linear regression model showed significant association of sperm concentration and total sperm count with LIN28B relative expression adjusted by age, abstinence time, % progressive motility and % normal morphology (beta=0.408, R2=0.241, p < 0.001 and beta=0.461, R2=0.393, p < 0.001). LIN28B rs314280 was associated with mRNA expression in the dominant model (p = 0.025) and two other SNPs (rs7759938, rs314268) showed the same trend (p = 0.09). Men with at least one minor allele of these 3 SNPs had a lower LIN28B mRNA abundance (p = 0.006; p = 0.07; p = 0.055). Limitations, reasons for caution Genotype frequency comparisons between cases and control were not conducted because of the small sample size. A larger cohort will be studied to address this assumption and association of genotypes with parameters of semen quality. Wider implications of the findings To our knowledge this study represents the first report on LIN28B expression and human testicular function. The significant direct association of LIN28B mRNA expression with sperm concentration suggests that LIN28B may be involved in spermatogonial proliferation. Trial registration number not applicable

STUDY QUESTION:Is the expression of steroidogenic enzyme 17α-Hydroxylase/17,20-Lyase (CYP17A1) down-regulated in Leydig cells (LCs) of men with spermatogenic failure and compensated impairment of LC function, i.e. a low testosterone to LH (T/LH) ratio?SUMMARY ANSWER:Although the transcriptional expression of CYP17A1 is increased, its protein expression is decreased, in isolated LCs of men with spermatogenic failure and reduced serum T/LH.WHAT IS KNOWN ALREADY:Primary spermatogenic defects have been associated with functional and morphological abnormalities of LCs, characterized by decreased serum testosterone (T) levels, decreased T/LH, increased 17β-estradiol (E2) and E2/T ratio, and larger clusters of LCs (LC hyperplasia). CYP17A1 is a key enzyme in the testosterone pathway and has been implicated in the steroidogenic lesion produced by E2 stimulation.STUDY DESIGN, SIZE, DURATION:We studied 18 azoospermic patients with Sertoli cell-only syndrome (SCOS) and signs of LC dysfunction (cases) and 10 obstructive azoospermic/oligozoospermic men with normal spermatogenesis (controls). The SCOS patients were sub-grouped into 9 cases with T/LH <2 and 9 cases with T/LH ≥2. All of the men underwent testicular biopsy for sperm retrieval at the Reproductive Unit of a University Hospital.PARTICIPANTS/MATERIALS, SETTING, METHODS:The transcriptional expression of CYP17A1 and SF-1 (steroidogenic factor 1) was quantified by SYBR®Green-based qPCR in LCs isolated by laser capture microdissection (LCM), and relative expression to the control pool was assessed. CYP17A1 protein expression was semi-quantified by indirect immunofluorescence (IFI) using Image-Pro Plus v7.0 (Media Cybernetics) in testicular tissue. FSH and LH serum concentrations, and serum and intratesticular T (ITT) and E2 (ITE2) were measured by IRMA and RIA, respectively.MAIN RESULTS AND THE ROLE OF CHANCE:Relative CYP17A1 mRNA expression was increased in cases with T/LH <2 compared to cases with T/LH ≥2, by a mean of 3.3-fold (P = 0.002). No corresponding increase in protein expression was found; in fact, CYP17A1 immunostaining intensity assessed by the Integrated Optical Density (IOD) parameter was lower in the cases with T/LH <2 compared to controls (P = 0.008). Relative SF-1 mRNA expression was similar in both case subgroups. CYP17A1 mRNA expression correlated with ITE2 and intratesticular E2/T (r = 0.536; P = 0.026 and r = 0.542; P = 0.016, respectively), while an inverse association was observed for ITE2 and protein level expression (r = -0.421; P = 0.05).LARGE SCALE DATA:Not applicable.LIMITATIONS REASONS FOR CAUTION:We should interpret the results of the semi-quantification of immunofluorescent staining by Image-Pro Plus software with caution, because it is a semi-quantitative method that may have certain difficulties regarding the disposition of protein in the cells. However, it is not influenced by variations in the number of cells that express the protein, as could be the case of western blot analysis in testicular tissue.WIDER IMPLICATIONS OF THE FINDINGS:Dysfunctional LCs of men with SCOS show post-transcriptional deregulation of CYP17A1, with increased mRNA and decreased protein expression, which may be modulated by increased ITE2 levels. In addition, transcriptional expression of CYP17A1 was not associated with changes in SF-1 mRNA expression.STUDY FUNDING/COMPETING INTEREST(S):This work was supported by the National Fund for Scientific and Technological Development (FONDECYT) of Chile to A.C. [grant number 1120176]. The authors declare no conflict of interest.

Several observational studies have showed a combination of lower testosterone (T) to LH ratio and higher estradiol (E2 ) to T ratio in secretory infertile men compared to men with normal spermatogenesis, suggesting a steroidogenic dysfunction of Leydig cells (Lc) that may involve increased aromatase activity. Low T/LH ratio is associated with Lc hyperplasia, which together with LH hyperstimulation may represent compensation for impaired T production. Aromatase expression and oestrogen production are mainly detected in Lc of the testis, although Sertoli and germ cells also contribute to testicular aromatase activity. The aim of this study was to assess the transcriptional expression of CYP19A1 (aromatase) in isolated Lc of subjects with Sertoli cell-only syndrome (SCOS) and signs of Lc impairment. Nineteen patients with SCOS and 10 controls with normal spermatogenesis who had medical indication of testicular biopsy for sperm retrieval were studied. Leydig cells were isolated by Laser Capture Microdissection (LCM) and CYP19A1 mRNA expression was quantified by SYBR® Green-based qPCR. In addition, testicular T and E2 and serum hormonal levels were measured. Relative to control group, CYP19A1 was overexpressed more than twofold in 10/19 cases (2.3-12.2-fold increase), showing a significant increment in cases with low T/LH ratio (T/LH < 2) compared to cases with T/LH ≥ 2 (p = 0.038, REST® ). Moreover, Rq data for CYP19A1 had a direct correlation with testicular levels of E2 and the E2 /T ratio (r = 0.869; p < 0.001 and r = 0.633; p = 0.005). In summary, Lc from infertile patients with signs of Lc dysfunction overexpressed aromatase and showed an increment of testicular E2 . Our results suggest that increased expression of aromatase in Lc leads to higher E2 production and may account for the functional impairment of the Lc in patients with SCOS.

STUDY QUESTION: Are copy number variations (CNVs) in the pseudoautosomal regions (PARs) frequent in subjects with Y-chromosome microdeletions and can they lead to abnormal stature and/or neuropsychiatric disorders?SUMMARY ANSWER: Only subjects diagnosed with azoospermia factor (AZF)b+c deletions spanning to the end of the Y chromosome (i.e. terminal deletions) harbor Y isochromosomes and/or cells 45, X that lead to pseudoautosomal gene CNVs, which were associated with abnormal stature and/or neuropsychiatric disorders.WHAT IS KNOWN ALREADY: The microdeletions in the long arm of the Y chromosome (Yq) that include the loss of one to three AZF regions, referred to as Yq microdeletions, constitute the most important known etiological factor for primary spermatogenic failure. Recently, controversy has arisen about whether Yq microdeletions are associated with gain or loss of PAR genes, which are implicated in skeletal development and neuropsychiatric function.STUDY DESIGN, SIZE, DURATION: We studied a cohort of 42 Chilean patients with complete AZF deletions (4 AZFa, 4 AZFb, 23 AZFc, 11 AZFb+c) from a university medical center, diagnosed over a period of 15 years. The subjects underwent complete medical examinations with special attention to their stature and neuropsychiatric function.PARTICIPANTS/MATERIALS, SETTING, METHODS: All subjects were characterized for Yq breakpoints by PCR, and for CNVs in PARs by multiplex ligation-dependent probe amplification (MLPA), followed by qPCR analysis for genes in PAR1 (SHOX and ZBED1), PAR2 (IL9R) and two single copy genes (SRY and DDX3Y, respectively located in Yp11.3 and AZFa). In addition, karyotypes revision and fluorescence in situ hybridization (FISH) for SRY and centromeric probes for X (DXZ1) and Y (DYZ3) chromosomes were performed in males affected with CNVs.MAIN RESULTS AND THE ROLE OF CHANCE: We did not detect CNVs in any of the 35 AZF-deleted men with interstitial deletions (AZFa, AZFb, AZFc or AZFb+c). However, six of the seven patients with terminal AZFb+c deletions showed CNVs: two patients showed a loss and four patients showed a gain of PAR1 genes, with the expected loss of VAMP-7 in PAR2. In these patients, the Yq breakpoints localized to the palindromes P8, P5 or P4. In the four cases with gain of PAR1, qPCR analysis showed duplicated signals for SRY and DDX3Y and one copy of IL9R, indicating isodicentric Yp chromosomes [idic(Y)] with breakpoint in Yq11.22. The two patients who had loss of PAR1, as shown by MLPA, had an additional reduction for SRY and DDX3Y, as shown by qPCR, associated with a high proportion of 45, X cells, as determined by FISH and karyotype. In agreement with the karyotype analysis, we detected DYZ3++ and DYZ3+ cells by FISH in the six patients, confirming idic(Y) and revealing additional monocentric Y chromosome [i(Y)]. Five patients had a history of major depressive disorders or bipolar disorder, and three had language impairment, whereas two patients showed severe short stature (Z score: -2.75 and -2.62), while a man with bipolar disorder was very tall (Z score: +2.56).LARGE SCALE DATA: N/A.LIMITATIONS, REASONS FOR CAUTION: The number of males studied with Y-chromosome microdeletions and normozoospermic controls with normal karyotypes may not be enough to rule out an association between AZF deletions and PAR abnormalities. The prevalence of Y isochromosomes and/or 45, X cells detected in peripheral blood does not necessarily reflect the variations of PAR genes in target tissues.WIDER IMPLICATIONS OF THE FINDINGS: This study shows that CNVs in PARs were present exclusively in patients with terminal AZFb+c deletions associated with the presence of Y isochromosomes and 45, X cells, and may lead to neuropsychiatric and growth disorders. In contrast, we show that men with interstitial Yq microdeletions with normal karyotypes do not have an increased risk of PAR abnormalities and of phenotypical consequences. Moreover, our results highlight the importance of performing molecular studies, which are not considered in the usual screening for patients with Yq microdeletions.

We characterised and correlated the histological and hormonal aspects of a cohort of 261 azo/oligozoospermic men, applying a quantitative/qualitative evaluation of testicular tissue and serum and intratesticular hormonal measurements. One hundred and 93 azo⁄oligozoospermic patients were diagnosed as: complete sertoli cell only syndrome (cSCOS), n = 76; focal SCOS, n = 31; maturation arrest, n = 34; hypospermatogenesis, n = 17; mixed atrophy, n = 25; and severe atrophy, n = 10. Normal spermatogenesis was observed in 68 infertile men (controls). Patients with cSCOS, focal SCOS, mixed and severe atrophy had larger LC/clusters (11.5; 11.0; 10.7; 18.9 LC/cluster) than controls (6 LC/cluster; P < 0.001). cSCOS, focal SCOS, mixed and severe atrophy patients had higher FSH, LH and lower T/LH ratio serum levels than the other groups. Intratesticular testosterone concentrations were higher in tissues with complete or focal SCOS (45.6 ng mg(-1) protein) and mixed atrophy (79.0 ng mg(-1) protein) than normal tissues (20.3 ng mg(-1) protein; P = 0.03 and P = 0.007). Considering all subjects, significant correlations were found between T/LH ratio and Leydig cells/cluster (r = 0.510, P < 0.001), FSH levels (r = -0.692, P < 0.001) and with intratesticular testosterone (r = -0.354, P = 0.001); these correlations follow the pattern of severity of spermatogenic damage. By a thorough histological evaluation, we validate the concept that the severity of spermatogenic impairment is associated with major morphological and functional disturbance of the Leydig cell compartment.

DAX-1 [dosage-sensitive sex reversal-adrenal hypoplasia congenital (AHC) critical region on the X chromosome gene 1; NR0B1] is an orphan nuclear receptor that acts as a transcriptional repressor in adrenal/gonadal development, steroidogenesis and probably spermatogenesis. An alternatively spliced form called DAX-1A (NR0B1A) has been described in several tissues including the testis, and in vitro studies have shown an inhibitory effect on DAX-1 transcriptional function. We aimed to study the mRNA and protein expression of DAX-1 in testicular tissues of 65 men with primary spermatogenic failure [complete Sertoli cell only syndrome (SCOS), focal SCOS, maturation arrest and mixed atrophy] compared with 33 controls with normal spermatogenesis. As a novel finding, we observed intense immunostaining, not only in the nucleus of Sertoli cells, but also in pachytene spermatocytes and round spermatids. The quantitative mRNA expression of DAX-1 and DAX-1A was similar between cases and controls and was not associated with the levels of gonadotrophins and steroids. Moreover, DAX-I transcript expression level was ∼750-fold higher than DAX-1A, and there was a strong positive correlation between them (r = 0.52; P< 0.001). We conclude that, in addition to Sertoli cells, DAX-1/DAX-1A is expressed in germ cells from spermatogonia to round spermatids. Besides, the similar mRNA expression of DAX-I and DAX-IA in testicular tissues from cases and controls does not support the involvement of DAX-1 in the etiology of primary spermatogenic failure. Finally, the low level of expression of the alternative transcriptional variant DAX-1A would not support its putative inhibitory function in vivo.

There is evidence that impaired spermatogenesis is associated with an imbalance in the oestradiol/testosterone ratio and with Leydig cell (LC) dysfunction. In testis, P450-aromatase, encoded by CYP19, is responsible for the conversion of testosterone to oestradiol. The aims of this study were to quantify CYP19 mRNA expression, aromatase activity and protein localization, and to measure the oestradiol to testosterone ratio in testicular tissues of men with spermatogenic impairment. Twenty-four men with complete Sertoli cell-only syndrome (SCOS), 14 with focal SCOS, 14 with maturation arrest (MA), 8 with mixed atrophy and 30 controls with normal spermatogenesis were subjected to testicular biopsy. All subjects underwent a physical examination, cytogenetic and serum hormonal studies. Testicular CYP19 mRNA was quantified using real time RT-PCR. Testicular aromatase activity was measured using the (3)H(2)0 assay and protein expression was evaluated using immunohistochemistry. In cases, serum testosterone and oestradiol were normal, but the testosterone/LH ratio was lower compared with controls (p < 0.05). Aromatase was localized in the Leydig, Sertoli and germ cells of all tissues, although stronger intensity was observed in LC. Aromatase mRNA and activity were not altered in cases and correlated positively with LC number (r = 0.516 and r = 0.369; p < 0.008). The intratesticular oestradiol/testosterone ratio was elevated (p = 0.005) in complete SCOS patients compared with controls. In conclusion, testicular aromatase seems to be normal in most subjects with impaired spermatogenesis. However, an altered intratesticular oestradiol/testosterone ratio in some patients with complete SCOS suggests that aromatase is increased, which might contribute to Leydig cell dysfunction.

Y chromosome microdeletion is the most important genetic cause of impairment of spermatogenesis. Nevertheless, a significant proportion of patients with spermatogenic failure do not have this condition. This study investigated the expression level of AZF genes, DDX3Y (DBY), RBMY1, DAZ and TSPY in testicular tissues of 42 subjects with impaired spermatogenesis compared with 33 with normal spermatogenesis. Histopathological evaluation was performed in all subjects and tissues were classified according to Johnsen Score. Transcript amounts were determined by quantitative-competitive RT-PCR. Patients with complete Sertoli cell-only syndrome (SCOS) did not exhibit RBMY1, DAZ and TSPY gene expression, however, we detected very low expression of DDX3Y transcript. Tissue samples with focal SCOS showed significantly decreased expression of all genes (P < 0.001). Maturation arrest and hypospermatogenesis tissues expressed significantly low levels of DDX3Y testicular transcript (P < 0.001), while the mRNA levels of the other genes were similar to that in tissues from the normal spermatogenesis group. Negative or diminished gene expression of DDX3Y, RBMY1, DAZ and TSPY in tissues samples with SCOS or focal SCOS reflects the absence or the lower number of germ cells, respectively. The finding that the testicular transcript of DDX3Y is significantly decreased in patients with severe spermatogenenic failure, especially in those presenting maturation arrest, suggests an important role of DDX3Y during spermatogenesis.