Impaired Spermatogenesis and gr/gr Deletions Related to Y Chromosome Haplogroups in Korean Men

Microdeletion of the Azoospermia Factor (AZF) regions in Y chromosome is a well-known genetic cause of male infertility resulting from spermatogenetic impairment. However, the partial deletions of AZFc region related to spermatogenetic impairment are controversial. In this study, we characterized partial deletion of AZFc region in Korean patients with spermatogenetic impairment and assessed whether the DAZ and CDY1 contributes to the phenotype in patients with gr/gr deletions. Total of 377 patients with azoo-/oligozoospermia and 217controls were analyzed using multiplex polymerase chain reaction (PCR), analysis of DAZ-CDY1 sequence family variants (SFVs), and quantitative fluorescent (QF)-PCR. Of the 377 men with impaired spermatogenesis, 59 cases (15.6%) had partial AZFc deletions, including 32 gr/gr (8.5%), 22 b2/b3 (5.8%), four b1/b3 (1.1%) and one b3/b4 (0.3%) deletion. In comparison, 14 of 217 normozoospermic controls (6.5%) had partial AZFc deletions, including five gr/gr (2.3%) and nine b2/b3 (4.1%) deletions. The frequency of gr/gr deletions was significantly higher in the azoo-/oligozoospermic group than in the normozoospermic control group (p = 0.003; OR = 3.933; 95% CI = 1.509–10.250). Concerning Y haplogroup, we observed no significant differences in the frequency of gr/gr deletions between the case and the control groups in the YAP+ lineages, while gr/gr deletion were significantly higher in azoo-/oligozoospermia than normozoospermia in the YAP− lineage (p = 0.004; OR = 6.341; 95% CI = 1.472–27.312). Our data suggested that gr/gr deletion is associated with impaired spermatogenesis in Koreans with YAP− lineage, regardless of the gr/gr subtypes.


Ethics Statement
This study was approved by the Institutional Review Board of CHA Gangnam Medical Center (IRB number: 09-06), and written informed consent was obtained from all participants.

Study Population
A total of 619 men who were born to ethnic Korean parents, were analyzed for classical AZF deletions and partial AZFc region deletions in the Fertility Center of CHA Gangnam Medical Center between January 2009 and December 2010. Of these, 210 patients were excluded because they had either numerical or structural chromosome abnormalities, known causes of spermatogenic failure (such as obstruction of the vas deferens, history of orchitis and active orchitis, or history of unilateral, bilateral cryptochidism and varicocele) or insufficient clinical data. In addition, 32 patients with classical AZF deletions were also excluded. Thus, the subjects were composed of 377 men with azoospermia or oligozoospermia (sperm concentration of ,20610 6 /ml, in all three semen analyses). The normozoospermic control group comprised 217 men who consulted the same hospital for a routine fertility workup. All of the control subjects were clinically healthy and possessed sperm concentrations of .20610 6 /ml, normal sperm motility and morphology, and hormonal parameters. Semen analysis was performed according to the World Health Organization criteria [23].

Characterization of the Partial AZFc Deletions
Genomic DNAs were extracted from peripheral blood samples using the QIAampH DNA Blood Mini Kit (QIAGEN, Hilden, Germany). Multiplex PCR reactions were performed using three STSs for the AZFc region (sY1191, sY1291, sY1206) and organized into two multiplex PCRs including a PCR control marker (SOHLH2). The amplification conditions were 94uC for 5 min, and then 30 cycles of 95uC for 30 sec, 61uC for 30 sec, and 72uC for 30 sec and a final elongation at 72uC for 10 min. The PCR reaction was always performed with a male control sample, a female sample, and a blank sample. The reaction products were analyzed by electrophoresis on 2% agarose gel. We identified partial AZFc deletions by the following STS results: the absence of sY1191, sY1291, sY1191/sY1291, and sY1206 represents the b2/ b3, gr/gr, b1/b3, and b3/b4 deletions, respectively [7,19].

gr/gr Subtypes Analysis
Deletion copy types of CDY1 and DAZ gene were analyzed by the previous described method [14]. For DAZ, a sequence family variant (SFV), STS-sY587 placed in intron 10, which discriminates DAZ1/2 from DAZ3/4, was used. And for CDY1, we used a C/A SFV located in 7750 bp upstream from the CDY1 translation start codon (CDY1-7750), which distinguishes CDY1a from CDY1b. SFVs were analyzed by PCR followed by restriction enzyme digestion: DAZ-sY587/DraI (DAZ1/2 cut); CDY1-7750/PvuII (CDY1b cut). For quantitative analysis of CDY1 and DAZ, quantitative fluorescent-PCR (QF-PCR) was performed according to the previously described method [14,25]. Briefly, CDY1 and DAZ were co-amplified with CDY2 and DAZL, respectively as a control with known gene copies. One primer (forward primer) in each set was labeled with 6-FAM (fluorescein amidite) fluorescent dye. The amplified products were loaded on the ABI 3130xl genetic analyzer and analyzed with the GeneMapper IDHversion 3.2 (Applied Biosystems, Foster City, CA). The copy numbers of the genes were estimated based on the relative DAZ/DAZL and CDY1/CDY2 ratios.

Statistical Analysis
Statistical analysis was performed using the statistical package SPSS for Windows (version 20, Chicago, IL, USA) software. The frequency of azoo-/oligozoospermia compared with normozoospermia was analyzed using both the chi-square test and Fisher's exact test (two-tailed). Continuous variables were analyzed by the ttest for independent samples. Probability (p) values ,0.05 were considered statistically significant.

Discussion
In this study, we investigated the types of partial AZFc deletions and their clinical implications in Korean population. Firstly, we  screened Yq microdeletions using STS markers for AZFa, AZFb and AZFc region. And then, gr/gr deletions were classified according to the Y-haplogroup, deletion copy types and gene copy number of CDY1 and DAZ genes.
Regardless of deletion type, as we expected, the overall frequency of partial AZFc deletions in azoo-/oligozoospermic men was higher than in normozoospermic men (15.6% vs. 6.5%, p = 0.001). This result suggested that such mutations could be a risk factor for impaired spermatogenesis in the Korean population. Four types of partial AZFc deletions were identified in our population and only gr/gr deletions were statistically associated with impaired spermatogenesis (Table 1). Our result is consistent with two recent meta-analyses [20,26] and with the largest study on Caucasians [21]. However, several other studies showed no association between gr/gr deletions and azoo-/oligozoospermia in Caucasian [10,27], Asian [4,[28][29][30][31] and admixed ethnic populations [32]. The phenotype of gr/gr deletion carriers is reported in Table S1. For the other three types, the frequency of b2/ b3deletion was not significant difference in between azoo-/ oligozoospermic males and normozoospermic controls, which suggested that b2/b3 deletion is not associated with spermatogenetic impairment in our study population. Similar observations have been reported in various studies [6,19,27,33]. Data from very large study populations from China and, very recently, a North African population suggest that b2/b3 deletion is a risk factor for impaired sperm production [28,29,34]. However, the limited number of subjects with this deletion in our study does not allow to define its role in the Korean population. The b1/b3 and b3/b4 deletions were identified only in patients with azoo-/oligozoospermia indicating that deletions might affect on spermatogenesis with mechanism yet to be revealed. The results, however, were limited due to the small number of cases. We compared the clinical features including total sperm count, combined testicular volume, FSH and testosterone levels between subjects with gr/gr deletion and others (Table S2). No statistically significant differences in sperm count were found when comparing sperm count of gr/gr deletion carriers versus non carriers. This may derive from the peculiar composition of our study population which shows a high prevalence of azoospermic men. Gr/gr deletion carriers are mainly oligozoospermic and in fact this deletion is more likely to be associated with oligozoospermia than with azoospermia [21,26].
The overall frequency of gr/gr deletions in Korean patient with spermatogenetic failure (8.5%) was higher than Europeans (,4.5%) and similar to Han Chinese populations (10.0 and 10.6%) [11,26]. This might be resulting from different origins of the study populations. It has been reported that some Asian populations, including Korean, Japanese, and Tibetan, showed the higher frequency of YAP+ haplogroup compared to other populations [35,36]. And the haplogroup D2b derived from YAP+ lineage always possessed gr/gr deletion and showed normal phenotype [31,37]. So, we reclassified gr/gr deletions based on YAP haplogroups, YAP+ (hgr.DE) and YAP2 [hgr.Y*(xDE)] lineages. Our data showed that the frequency of gr/gr deletion with YAP+ lineage was not significantly different between azoo-/ oligozoospermic males and normozoospermic controls. However, the frequency of gr/gr deletion with YAP2 lineage was much higher in azoo-/oligozoospermic males than in controls (Fig. 2). So, we concluded that only gr/gr deletion with YAP2 lineage was associated with spermatogenetic impairment in Korean population. Hereby, we demonstrated that the gr/gr deletion might effect on spermatogenetic impairment inY haplogroup-dependent manner.
We also investigated gr/gr deletion subtypes according to deletion patterns of DAZ and CDY1 gene copies. Normally, four copies of DAZ gene and two copies of CDY1 gene are assigned in the AZFc region. Several studies related to gr/gr deletion subtypes and spermatogenetic impairments have presented different conclusions. Some studies showed DAZ1/2 deletions were associated with spermatogenetic impairment [17,19,21,38,39], and Machev et al. [14] presented DAZ3/4-CDY1a deletions were linked to the phenotype. More recently, no association between subtypes of gr/ gr deletion and phenotypic abnormalities was also reported [6]. In our study, subjects with YAP+ lineage showed a homogeneous pattern of gr/gr deletion, DAZ1/2 2 , CDY1a 2 and no significant difference between azoo-/oligozoospermic and normozoospermic controls. This result was the same as a previous report [37]. In YAP2 lineage, deletion of the CDY1a was found only in spermatogenetic impairment group (Table 2). Although our result might not be sufficient to verify the association between these deletion subtypes and clinical consequences, this phenomenon is similar to previous studies in a correlation between the absence of CDY1a and male infertility [14,39]. So, further studies will be required.
Finally, we carried out CDY1 and DAZ gene copy number analysis to identify gr/gr rearrangement types (Table 3). Krausz et al. [6] reported that gr/gr deletions could be classified into five rearrangement types based on the copy number of CDY1 and DAZ gene. In our study, three out of five rearrangement types were identified and the majority of gr/gr deletions (83.8%, 31/37) were simple gr/gr deletion type, one copy of CDY1 and two copies of DAZ. This result was similar to European population (80%, 128/ 160) [6]. The other two types, gr/gr deletion-b2/b4 duplications (four copies of DAZ and two copies of CDY1) and gr/gr deletion-CDY1 amplification (two copies of DAZ and two copies of CDY1), were found in patients with spermatogenetic impairment with frequencies of 13.5% (5/37) and 2.7% (1/37), respectively, but none in normozoospermic controls. These results suggested that regardless of having an additional duplication of AZFc region, the simple gr/gr deletion might be associated with spermatogenetic impairment in Korean population. Recently, Lu et al. [22] reported that b2/b3 deletion with secondary duplication was also a risk factor for spermatogenetic impairment in Han Chinese population. Meanwhile, two possible mechanisms to generate the gr/gr deletion-CDY1 amplification are proposed and the recombinant products resulting from both ways were not distinguishable as shown in Figure S1. So, we could not verify what happened first with current technology.
In conclusion, we analyzed 377patients with spermatogenetic impairments and 217normozoospermic controls. As far as we know, this is the first report that only gr/gr deletion with YAP2 lineage, among several types of partial AZFc deletions, was associated with spermatogenetic impairment in Korean population. Although the role of CDY1 and DAZ on spermatogenesis is still not clear, further studies on other genes related to spermatogenesis and larger scale population study will be assured to understand the spermatogenesis pathology. Figure S1 Two possible mechanisms of gr/gr del-CDY1 amplification; one is that the g1/g2 recombination resulting in gr/gr deletion arises first and then the CDY1 amplification occurs and the other is vice versa. The recombinant products from both ways are not distinguishable. (TIF)