Human male infertility and Y chromosome deletions: role of the AZF-candidate genes DAZ, RBM and DFFRY

A. Ferlin, E. Moro, A. Garolla and C. Foresta1

Department of Medical and Surgical Sciences, Clinica Medica 3, University of Padova, Via Ospedale 105, 35128 Padova, Italy


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Microdeletions in Yq11 overlapping three distinct `azoospermia factors' (AZFa–c) represent the aetiological factor of 10–15% of idiopathic azoospermia and severe oligozoospermia, with higher prevalence in more severe testiculopathies, such as Sertoli cell-only syndrome. Using a PCR-based screening, we analysed Yq microdeletions in 180 infertile patients affected by idiopathic Sertoli cell-only syndrome and different degrees of hypospermatogenesis, compared with 50 patients with known causes of testicular alteration, 30 with obstructive azoospermia, and 100 normal fertile men. In idiopathic severe testiculopathies (Sertoli cell-only syndrome and severe hypospermatogenesis), a high prevalence of microdeletions (34.5% and 24.7% respectively) was found, while milder forms were not associated with Yq alteration. No deletions were found in testiculopathies of known aetiology, obstructive azoospermia, normal fertile men and male relatives of patients with deletions. Deletions in the AZFc region involving the DAZ gene were the most frequent finding and they were more often observed in severe hypospermatogenesis than in Sertoli cell-only syndrome, suggesting that deletions of this region are not sufficient to cause complete loss of the spermatogenic line. Deletions in AZFb involving the RBM gene were less frequently detected and there was no correlation with testicular phenotype, with an apparent minor role for such gene in spermatogenesis. The DFFRY gene was absent in a fraction of patients, making it a candidate AZFa gene. Our data suggest that larger deletions involving more than one AZF-candidate gene are associated with a more severe testicular phenotype.

Key words: DAZ/DFFRY/male infertility/RBM/Y-chromosome


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
In humans the Y chromosome is essential for normal testicular differentiation and spermatogenesis. Cytogenetic studies on azoospermic subjects two decades a-*go led to the postulation of the existence of a gene locus in the distal euchromatic part of Yq (Yq11), defined as azoospermia factor (AZF) (Tiepolo and Zuffardi, 1976Go). Recently, evidence for this genetic function has been derived from the observation of microdeletions on Yq11 in a number of infertile men (Chandley and Cooke, 1994Go; Kobayashi et al., 1995Go; Reijo et al., 1995Go, 1996Go; Najmabadi et al., 1996aGo; Qureshi et al., 1996Go; Stuppia et al., 1996Go, 1997Go; Vogt et al., 1996Go; Foresta et al., 1997Go, 1998Go; Girardi et al., 1997Go; Pryor et al., 1997Go; Simoni et al., 1997Go; Liow et al., 1998Go). These molecular studies have suggested that interstitial microdeletions in Yq11 represent the aetiological factor of 10–15% of idiopathic azoospermia and severe oligozoospermia, and this prevalence is even higher when more severe primary testiculopathies are selected, such as idiopathic Sertoli cell-only syndrome (Reijo et al., 1995Go; Najmabadi et al., 1996aGo; Vogt et al., 1996Go; Foresta et al., 1997Go, 1998Go; Girardi et al., 1997Go; Simoni et al., 1997Go; Liow et al., 1998Go), that represent about one-third of idiopathic non-obstructive azoospermia. Furthermore, not one but multiple non-overlapping spermatogenesis loci in interval 5 and 6 on Yq11 have been identified, that map to regions defined AZFa, AZFb and AZFc (Vogt et al., 1996Go, 1997Go), indicating the involvement of several genes on Yq in spermatogenesis.

Candidate genes for these regions have been isolated in recent years: (i) the RBM (RNA-binding motif, formerly YRRM) family (Ma et al., 1993Go), consists of about 30 copies on both arms of the Y chromosome belonging to at least six subfamilies (Chai et al., 1997Go). Only RBM-I is actively transcribed, the most functional copies of this gene are located on interval 6B (Elliott et al., 1997Go), thus making it the best candidate for the AZFb region (Vogt et al., 1997Go); (ii) the DAZ (deleted in azoospermia) family (Reijo et al., 1995Go; Saxena et al., 1996Go), consisting of multiple (at least three) functional genes clustered on interval 6D (Gläser et al., 1997Go; Yen et al., 1997Go, 1998Go) and representing the AZFc candidate; (iii) DFFRY (Drosophila fat-facets related Y) (Brown et al., 1998Go), a recently characterized gene located in interval 5C and supposed to represent the AZFa candidate.

At present, AZFc represents the most frequently deleted region in infertile men. Such deletions appear to remove the entirety of the DAZ gene cluster and have been associated with a variety of spermatogenic alterations, ranging from azoospermia due to Sertoli cell-only to oligozoospermia with different testicular phenotype (Reijo et al., 1995Go, 1996Go; Najmabadi et al., 1996aGo; Stuppia et al., 1996Go; Vogt et al., 1996Go; Foresta et al., 1997Go, 1998Go; Girardi et al., 1997Go; Pryor et al., 1997Go; Simoni et al., 1997Go; Liow et al., 1998Go). Deletions in AZFb overlapping the RBM gene and deletions in AZFa occur less frequently, and in such cases they have been detected in different abnormalities of spermatogenesis (Reijo et al., 1995Go; Najmabadi et al., 1996aGo; Vogt et al., 1996Go; Foresta et al., 1997Go, 1998Go; Girardi et al., 1997Go; Pryor et al., 1997Go; Liow et al., 1998Go). DFFRY is the least studied gene, as until now it was found to be absent only in three subjects previously that were deleted in the AZFa interval, and presenting with Sertoli cell-only syndrome or severe hypospermatogenesis (Brown et al., 1998Go). Therefore, a relationship between Yq microdeletions and testicular phenotype is still lacking, and the roles of DAZ, RBM and DFFRY in determining the disruption of spermatogenesis have not yet been elucidated. Such difficulties in genotype–phenotype associations could arise from the different patient selection criteria utilized in these studies, that may be based only on clinical (infertility, history, hormonal concentrations, testicular volume), and/or seminological data (normo-, oligo-, azoospermia) and/or testicular structure (Sertoli cell-only syndrome, hypospermatogenesis, spermatogenic arrest, obstructive forms).

In this study, we performed a polymerase chain reaction (PCR)-based Yq analysis including DAZ, RBM and DFFRY genes in a large group of infertile patients affected by azoospermia or oligozoospermia and characterized by well-defined testicular alterations, in order to determine the importance of AZF regions in disrupting spermatogenesis.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Patient selection
Our study was approved by the Hospital Ethical Committee, and informed consent was obtained from each patient. In total, 180 adult men were selected who were affected by idiopathic azoospermia or oligozoospermia and in whom bilateral testicular fine needle aspiration cytology (FNAC) showed Sertoli cell-only syndrome or hypospermatogenesis respectively. In particular, 55 patients presented azoospermia with a picture of Sertoli cell-only in both testes (group 1), 85 severe oligozoospermia (sperm count <5x106/ml) and bilateral severe hypospermatogenesis (group 2), 20 moderate oligozoospermia (sperm count 5–10x106/ml) with moderate hypospermatogenesis (group 3), and 20 mild oligozoospermia (sperm count 10–20x 106/ml) with mild hypospermatogenesis (group 4), as described below.

Semen samples were obtained on two different occasions, separated by a 3-week interval, following a 3-day period of sexual abstinence, and complete semen analyses were performed according to WHO guidelines (WHO, 1992). The diagnosis of azoospermia was established by pellet analysis after semen centrifugation (1000 g, 20 min). Plasma concentrations of follicle stimulating hormone (FSH) and luteinizing hormone (LH) were measured in each subject by radioimmunoassay using 125I-labelled FSH and LH (Ares-Serono, Milan, Italy). Plasma concentrations of testosterone were determined by radioimmunoassay using 3H-labelled testosterone (Radim, Rome, Italy). Only patients with an apparently normal 46,XY karyotype were included in this study. All patients were studied with a comprehensive history and general investigation for exclusion of possible causes of testicular damage, such as cryptorchidism, varicocele, seminal tract infections, drug use, endocrinopathies, post-mumps orchitis, testicular trauma or torsion.

As a comparison, we have studied 50 azoospermic and severely oligozoospermic men (sperm count <5x106/ml) with known causes of testiculopathy (group 5), such as orchi-epididymitis, testicular trauma or chemoradiotherapy, 30 obstructive azoospermic men (congenital or acquired obstruction of the seminal tract) (group 6), and 100 healthy normozoospermic fertile men (group 7).

Testicular fine needle aspiration and cytological quantification
The testicular structure was analysed by means of bilateral FNAC, as described previously (Foresta and Varotto, 1992Go; Foresta et al., 1992Go, 1995Go). Briefly, bilateral fine needle aspiration was performed using 23-gauge (0.6 mm) butterfly needles and aspirating with a 20-ml syringe. The cellular material was stained with May-Grünwald and Giemsa, examined under a light microscope and at least 200 spermatogenic cells (spermatogonia, primary spermatocytes, secondary spermatocytes, early and late spermatids and spermatozoa) were counted per smear. Spermatogenic cells were expressed as relative percentages, while the interposed Sertoli cells were expressed as the Sertoli index (SEI, the number of Sertoli cells/100 spermatogenic cells), which has been found to be a reliable index of the tubular germ potential (Foresta and Varotto, 1992Go; Foresta et al., 1992Go, 1995Go).

As described in previous studies (Foresta and Varotto, 1992Go; Foresta et al., 1992Go, 1995Go), this cytological analysis allows us to classify azoo/oligozoospermic subjects as follows: (i) Sertoli cell-only syndrome, defined as the complete absence of germ cells; (ii) hypospermatogenesis, defined as quantitative reduction of the germ line with respect to Sertoli cells; different degrees of hypospermatogenesis were distinguished on the basis of the SEI (see Results); (iii) spermatogonia or spermatocytes arrest; (iv) spermatids arrest; and (v) obstructive forms, presenting with a normal germ line with increased percentage of mature spermatozoa. Only patients showing Sertoli cell-only syndrome, various degrees of hypospermatogenesis and, as controls, obstructive azoospermia were included in this study, therefore excluding qualitative spermatogenesis alterations (maturation arrest).

Sequence-tagged site–PCR amplification
A set of 38 Y-specific sequence-tagged sites (STS) (Vollrath et al., 1992Go; Ma et al., 1993Go; Kobayashi et al., 1995Go; Reijo et al., 1995Go, 1996Go; Brown et al., 1998Go) spanning the euchromatic region of Yq was tested in each patient. The order of STS and Yq deletion intervals as described previously are shown in Figure 1Go (Reijo et al., 1995Go, 1996Go); AZF regions are defined as previously reported (Vogt et al., 1997Go). AZF-candidate genes studied were as follows: DFFRY-5', DFFRY 4.1, DFFRY J/D and DFFRY 3.1 (Brown et al., 1998Go) amplified the regions 2015–2136, 5753–6206, 8853–9157 and 9297–9451 respectively of the coding region of the DFFRY gene: sY277, sY254, sY279, sY283 and sY255 (Reijo et al., 1995Go, 1996Go) amplified the regions 1277–1588 (from intron 1–2 to intron 2–3), 1409–1789 (from exon 2 to exon 3), 1663–2497 (from intron 2–3 to intron 4–5), 2506–3002 (from intron 4–5 to intron 5–6) and 3042–3165 (within exon 6) respectively of the genomic DNA of the DAZ gene (clone 63C9); F19/E355 (Ma et al., 1993Go; Kobayashi et al., 1995Go) amplified the region 1292–1733 of the RBM coding sequence, corresponding to the distal part of exon 11 to the most part of exon 12.



View larger version (36K):
[in this window]
[in a new window]
 
Figure 1. Sequence-tagged sites–polymerase chain reaction (STS–PCR) results of Yq deletions. A schematic representation of the Y chromosome, deletion intervals, AZF regions and Y-chromosomal STS used are listed above. Position of AZF-candidate genes with the corresponding STS are indicated. Black boxes: STS present; lines: STS absent. (A) Patients affected by idiopathic Sertoli cell-only syndrome. (B) Patients affected by idiopathic severe hypospermatogenesis. Patient numbers are listed on the left; asterisks indicate de-novo deletions confirmed by the study of fathers or brothers of the patients.

 
PCR was carried out in 50 µl of reaction volume containing 200 ng of genomic DNA extracted from peripheral blood cells, Taq polymerase (2U), dNTPs (0.2 mM each of dTTP, dCTP, dGTP, dATP), oligonucleotide primers (10 pmol each) made up in a final concentration of 1x PCR reaction buffer (10 mM Tris–HCl pH 8.3, 1.5 mM MgCl2, 50 mM KCl). All reagents were obtained from Pharmacia (Milan, Italy). Amplification was performed for 30–35 sequential cycles, each including 1 min of denaturation at 94°C, 1 min of primer annealing at 55–60°C, and 1–2 min of extension at 72°C. Before the first cycle, all samples were incubated for 10 min at 94°C. PCR reaction products were stored at 4°C and then separated on 2% agarose gel by electrophoresis in Tris–acetic acid–EDTA (TAE) buffer at room temperature using a voltage gradient of 8 V/cm for 30–60 min.

All primers were analysed in 100 normal healthy men of proven fertility (positive controls) before their application in patients, in order to ascertain that each of them produced a single amplification product of the expected size. The Y-specificity was determined in 10 normal women (negative controls). Patients were considered normal if the PCR product was of the expected size and negative only after three amplification failures. The repeated experiments were performed on new samples of DNA extracted from a different blood collection. Furthermore, only DNA extracts which gave a normal PCR amplification of the SRY locus on Yp (sY14, Vollrath et al., 1992Go) were considered.

Twenty-four fathers or brothers of 40 deleted patients were also investigated under the same experimental conditions, while no male relatives were available for the other 16 patients. None of the male relatives of patients with Yq deletion had a history of infertility.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Testicular FNAC performed on both testes allowed us to select idiopathic infertile patients affected by well-defined testiculopathies related to the seminal pattern. In particular, we have recruited azoospermic and oligozoospermic men showing the complete absence (Sertoli cell-only syndrome) or a quantitative reduction of germ cells (hypospermatogenesis). The relationship between seminal and testicular cytological patterns is shown in Table IGo.


View this table:
[in this window]
[in a new window]
 
Table I. Seminal and testicular cytological patterns with results of Yq analysis in the different groups of patients
 
Sertoli cell-only syndrome has been diagnosed when the extensive analysis of all the aspirated material from both testes demonstrated the presence of only Sertoli cells and the complete absence of any spermatogenic cell (Foresta and Varotto, 1992Go; Foresta et al., 1992Go, 1995Go). Hypospermatogenesis was characterized by a reduction in the absolute number of germ cells with respect to Sertoli cells (Foresta and Varotto, 1992Go; Foresta et al., 1992Go, 1995Go), and was distinguished as mild, moderate and severe on the basis of the SEI (SEI 50–100, 100–300 and >300 respectively, with normal values <50). In these forms, each cell type was observed in relatively normal proportions, and we have excluded qualitative alterations of spermatogenesis, as seen in maturation arrests at different levels. Patients of group 6 affected by obstructive azoospermia showed the presence of all spermatogenic cells with an increased percentage of mature spermatozoa due to intratubular stasis (Foresta et al., 1992Go, 1995Go).

Each of the 38 STS produced an amplification product of the expected size in all normal fertile men and failed to amplify in normal women. In particular, primer pairs sY277, sY254, sY279, sY283 and sY255 gave an amplification product of 311, 380, 834, 496 and 123 bp respectively, corresponding to the DAZ genomic DNA sequence; F19/E355 amplified 800 bp, according to previous reports (Ma et al., 1993Go; Kobayashi et al., 1995Go). Only the cDNA sequence is available for the DFFRY gene (Brown et al., 1998Go) and we obtained PCR products of approximately 1.5 kb for DFFRY-5', 2 kb for DFFRY 4.1, 1.2 kb for DFFRY J/D and 200 bp for DFFRY 3.1, which represent the normal bands (Nabeel A.Affara, personal communication).

Deletions of Yq were observed in 19 out of 55 patients affected by idiopathic Sertoli cell-only syndrome (group 1, 34.5%) and in 21 out of 85 patients affected by idiopathic bilateral severe hypospermatogenesis (group 2, 24.7%). No deletions were found in patients presenting with idiopathic moderate and mild oligozoospermia (groups 3 and 4), or in patients affected by severe testiculopathies of known aetiology (group 5) and in obstructive azoospermic patients (group 6) (Table IGo).

Figure 1Go summarizes the PCR results of patients with deletions, that will be discussed below. Briefly, three patients had a terminal deletion of Yq, while the others showed interstitial deletions. A specific deletion only in the AZFc region involving the DAZ gene was present in 17 out of 40 patients (42.5%), and it was more frequent in patients with severe hypospermatogenesis (13/21, 61.9%) than in patients with Sertoli cell-only syndrome (4/19, 21.1%); among these, deletions confined to the DAZ gene, not including other flanking STS, were present in five out of 17 cases and they were associated both with severe hypospermatogenesis and Sertoli cell-only syndrome. A deletion involving only the RBM gene with or without some flanking regions was present in six out of 40 subjects (15.0%) and was equally frequent in both groups, with a prevalence of 15.8% in Sertoli cell-only syndrome (3/19) and of 14.3% in severe hypospermatogenesis (3/21). More proximal deletions confined to regions overlapping the DFFRY gene were present in five out of 40 patients (12.5%), with a slightly lower percentage in Sertoli cell-only syndrome (2/19, 10.5%) than in severe hypospermatogenesis (3/21, 14.3%). Among these, a deletion involving only the DFFRY gene was present only in one patient (no. 327) and it was associated with severe hypospermatogenesis. There was a higher prevalence of deletions involving more than one AZF-candidate gene in Sertoli cell-only syndrome than in severe hypospermatogenesis: seven of 19 (36.8%) patients of the first group and only two of 21 (9.5%) of the second group had a deletion of two genes; three patients (15.8%) with Sertoli cell-only syndrome and none with severe hypospermatogenesis had a large Yq deletion involving all three genes.

The father or brothers of 24 deleted patients were also investigated and no deletions were found. No male relatives were available for the remaining patients, but on the basis of these and previous results it is possible to conclude that these represent de novo deletions and may be considered the aetiological factor of the spermatogenic defect.

Testicular volumes were significantly reduced and FSH plasma concentrations significantly increased compared with controls (10.8 ± 3.7 ml and 16.4 ± 5.3 IU/l respectively, P <0.05), without significant differences in plasma concentrations of LH and testosterone in subjects affected by a severe testiculopathy (groups 1, 2 and 5), confirming the primary testiculopathy involving only the spermatogenic system. All these parameters were not different from controls in the other groups of patients (groups 3, 4 and 6). Furthermore, testicular volume and hormone concentrations were not different in patients with and without deletion (data not shown).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Candidate genes for AZF regions appear to belong to two different super-families: the first includes DAZ and RBM that, although not related, encode for proteins with a single RNA recognition motif and an internally repeated sequence called DAZ repeat and SRGY box respectively (Ma et al., 1993Go; Reijo et al., 1995Go; Najmabadi et al., 1996bGo; Saxena et al., 1996Go; Cooke and Elliott, 1997Go; Yen et al., 1997Go; Chai et al., 1998Go); they are each homologous to autosomal genes and probably originated from transposition and amplification of autosomal ancestors (Saxena et al., 1996Go; Delbridge et al., 1997Go; Agulnik et al., 1998Go; Chai et al., 1998Go). The second consists of DFFRY that shows homologies with an X-linked gene and is likely to function as a C-terminal ubiquitin hydrolase, as it contains the conserved Cys and His domains characteristic of deubiquitinating proteins (Brown et al., 1998Go). These genes also share different expression patterns, as DAZ and RBM are testis-specific with transcription restricted to germ cells (Reijo et al., 1995Go; Saxena et al., 1996Go; Elliott et al., 1997Go, 1998Go; Menke et al., 1997Go; Habermann et al., 1998Go; Lee et al., 1998Go), while DFFRY is ubiquitously expressed in a wide range of tissues (Brown et al., 1998Go). DAZ and RBM, being testis-specific, should play a major role in the control of human spermatogenesis, and this suggestion is supported by the evidence of deletion of these genes in infertile subjects. To confirm their role further, detrimental mutations within these genes should be determined in patients, but at present this proof has not been demonstrated (Chai et al., 1997Go; Vereb et al., 1997Go; Yen et al., 1997Go), probably because of their multicopy nature. The DAZ gene represents the best candidate for AZF, since it is the most frequently deleted gene in infertile men and it shares high homology with a Drosophila male infertility gene boule (Eberhart et al., 1996Go) which, when mutated, causes spermatogenic arrest. DFFRY is the first major gene identified in AZFa, but its role in spermatogenesis remained to be confirmed. To date, only three infertile patients have been reported carrying a deletion of this gene and in these men the entire AZFa region was absent, so that the involvement of other unknown genes in this region could not be excluded (Brown et al., 1998Go). Furthermore, the exact characterization of the AZFa deletion breakpoints is under investigation and therefore the critical interval and the candidate genes are not defined (Mazeyrat et al., 1998Go). Apart from these AZF-candidate spermatogenesis genes, other regions of Yq with unknown functions may be deleted in infertile men (Najmabadi et al., 1996aGo; Stuppia et al., 1996Go; Foresta et al., 1997Go, 1998Go; Girardi et al., 1997Go; Pryor et al., 1997Go). However, it now seems that such deletions could theoretically include a number of recently identified genes, testis- and Y-specific as well as ubiquitously expressed and with X-homologues (Lahn and Page, 1997Go), whose functions are still unidentified.

In the present study we have looked for the relative importance in spermatogenesis of Yq regions with particular interest to AZF-candidate genes, screening different populations of infertile men affected by azoospermia and oligozoospermia and selected on the basis of their testicular phenotype and clinical history. We found a high prevalence of Yq deletions in idiopathic severe testiculopathies characterized by the complete absence (Sertoli cell-only syndrome) or a strong reduction (severe hypospermatogenesis) of germ cells (40 out of 140, 28.6%), while milder forms of idiopathic testiculopathy were not associated with Yq alterations. No Yq region was deleted in patients whose infertility was ascribed to known aetiologies, further supporting the hypothesis that Yq deletions are responsible for the testicular damage observed in `idiopathic' forms. The confirmation of the pathogenic role for such deletions derives also from the analysis of fertile normozoospermic men and male relatives of patients with deletions, that allowed us to exclude normal polymorphisms and to consider them as de novo deletions. The prevalence of Yq deletions found in this study is undoubtedly high but in agreement with previous papers reporting that this frequency increases with the severity of testicular tubular damage (Reijo et al., 1995Go; Najmabadi et al., 1996aGo; Vogt et al., 1996Go; Foresta et al., 1997Go, 1998Go; Girardi et al., 1997Go; Simoni et al., 1997Go; Liow et al., 1998Go). In fact, idiopathic Sertoli cell-only syndrome that represents the worst condition of male infertility shows the highest prevalence of deletion (19/55, 34.5%); furthermore, when such genetic anomalies are associated with residual spermatogenesis, the final sperm output is compromised, and no Yq deletion is found when the sperm concentration is >5x106/ml. The quite small number of patients with mild and moderate oligozoospermia may hide a very low rate of deletion, as only isolated cases have been described in these groups of patients (Pryor et al., 1997Go).

Our results confirm the predominant role of deletions in the AZFc region overlapping the DAZ gene family among other Yq regions in determining a severe tubular damage. A deletion confined to this region was the most frequent finding, as it was present overall in more than 40% of patients with deletions, and more often found in severe hypospermatogenesis rather than in Sertoli cell-only syndrome, since 13 out of 17 (76.5%) patients deleted for this gene belong to this group. However, only five out of 17 patients presented a deletion of the DAZ gene not including other flanking regions, and they were affected either by severe hypospermatogenesis or Sertoli cell-only syndrome. Therefore, a non-unique phenotype is associated with DAZ deletions, even if the absence of DAZ appears not sufficient to determine the complete loss of the spermatogenic line, but rather seems to produce a reduction in number of these cells. The multicopy nature of this gene has previously made it difficult to detect deletions of some copies or intragenic mutations. The mechanisms by which DAZ gene deletions cause this spermatogenic impairment remain unclear, since little is known about the functions of this RNA-binding protein in human spermatogenesis, except that it is expressed in different phases of the spermatogenetic cycle (Menke et al., 1997Go; Habermann et al., 1998Go) and it may regulate splicing events or translation (Elliott et al., 1997Go, 1998Go).

Deletions in AZFb involving the RBM gene are less frequently detected, both in Sertoli cell-only syndrome and in severe hypospermatogenesis. These data confirm previous reports on the role of this gene family in male infertility and do not allow us to correlate testicular phenotype with RBM deletions. Like DAZ, RBM is in multiple copies on the Y chromosome and this has complicated attempts to prove its role in human spermatogenesis, as detrimental mutations have not yet been identified. Furthermore, the exact copy number of functional RBM genes is uncertain, even if the critical region for RBM expression has been mapped to interval 6B (Elliott et al., 1997Go). RBM is a nuclear protein with dynamic modulations in its spatial location in the different spermatogenic cells, suggesting that it possesses different functions related to pre-mRNA splicing (Elliott et al., 1998Go), though how it functions during male germ cell development is not known.

The most intriguing feature of the present study regards the analysis and role of the DFFRY gene. Our results seem to suggest this gene as the AZFa candidate, since it was absent in a fraction of patients and one of them (no. 327) presented a normal PCR amplification of other STS contained in the AZFa region. Deletions of this gene seem to be more frequently associated with a depopulation of germ cells rather than with their complete absence. However, the mechanism by which a spermatogenic impairment occurs is unknown. Mutations or microdeletions in DFFRY, if found, will confirm that this gene is responsible for the AZFa phenotype. It must be noted, however, that deletions overlapping DFFRY frequently include also neighbouring regions, in which new genes have been mapped (Lahn and Page, 1997Go). In particular, interval 5C/D seems to harbour DBY and UTY genes (Lahn and Page, 1997Go; Mazeyrat et al., 1998Go) that may be absent in some patients and may contribute with DFFRY to the AZFa phenotype. However, DBY and UTY seem to lie between markers sY87 and sY88 (Mazeyrat et al., 1998Go) and therefore should be normally present in patient no. 327, whose phenotype should be due only to the absence of DFFRY.

The microdeletion pattern observed in this study differs from the AZF classification and genotype–phenotype relation proposed previously (Vogt et al., 1996Go, 1997Go), but suggests that larger deletions are associated with more severe spermatogenic phenotype. This hypothesis is supported by the evidence that in patients with Sertoli cell-only syndrome Yq deletions frequently involved large regions of interval 5 and 6 and that, in a high proportion of cases, two or even three AZF-candidate genes were absent, while these associations were extremely rare in the group of patients affected by hypospermatogenesis. Even if a correlation between Yq deletions and testicular phenotype is lacking, it could be argued that larger deletions may show the additional effects of each single gene deletion.

As previously noted by us and other groups (Najmabadi et al., 1996aGo; Stuppia et al., 1996Go; Foresta et al., 1997Go, 1998Go; Girardi et al., 1997Go; Pryor et al., 1997Go; Duell et al., 1998Go; Rossato et al., 1998Go), PCR analysis frequently showed non-contiguous deletions. The Y chromosome seems to be highly unstable and prone to deletions, probably since it is rich in repetitive elements and repeats (Girardi et al., 1997Go; Yen et al., 1998Go), and interstitial double deletions may be explained by different hypotheses: (i) the PCR observations may reflect really separated microdeletions; (ii) some STS may be from repetitive sequences (as demonstrated for example for sY146, sY153 and sY155) (Yen et al., 1998Go); and (iii) a complex rearrangement (e.g. an inversion with subsequent interstitial deletion) in the father may be the cause. We do not think that these observations reflect a population effect, as reported patients are not from a single geographical area but come from different regions of Italy.

Finally, the determination of Yq deletions in infertile men has important clinical and ethical implications, especially in severely oligozoospermic patients, as it has been clearly demonstrated that spermatozoa carrying a Yq deletion are able to fertilize and give rise to pregnancies by means of assisted reproduction techniques (Mulhall et al., 1997Go; Rossato et al., 1998Go). Therefore, such analysis provides a careful diagnosis and allows the andrologist to forego empirical and often expensive treatments; at the same time it dictates that infertile men undergoing such procedures should be informed about the possibility of passing on a Yq deletion to male offspring and a screening for microdeletions of the Y chromosome should be offered to them.


    Acknowledgments
 
We thank Dr Nabeel A.Affara for information on PCR-amplification of the DFFRY gene and Dr Pauline Yen for comments on the manuscript. The financial support of Telethon-Italy (grant no. E.C699) is gratefully acknowledged.


    Notes
 
1 To whom correspondence should be addressed Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Agulnik, A.I., Zharkikh, A., Boettger-Tong, H. et al. (1998) Evolution of the DAZ gene family suggests that Y-linked DAZ plays a little, or a limited, role in spermatogenesis but underlines a recent African origin for human population. Hum. Mol. Genet., 7, 1371–1377.[Abstract/Free Full Text]

Brown, G.M., Furlong, R.A., Sargent, C.A. et al. (1998) Characterisation of the coding sequence and fine mapping of the human DFFRY gene and comparative expression analysis and mapping to the Sxrb interval of the mouse Y chromosome of the Dffry gene. Hum. Mol. Genet., 7, 97–107.[Abstract/Free Full Text]

Chai, N.N., Salido, E.C. and Yen, P.H. (1997) Multiple functional copies of the RBM gene family, a spermatogenesis candidate on the human Y chromosome. Genomics, 45, 355–361.[ISI][Medline]

Chai, N.N., Zhou, H., Hernandez, J. et al. (1998) Structure and organization of the RBMY genes on the human Y chromosome: translocation and amplification of an ancestral autosomal hnRNPG gene. Genomics, 49, 283–289.[ISI][Medline]

Chandley, A.C. and Cooke, H.J. (1994) Human male fertility – Y linked genes and spermatogenesis. Hum. Mol. Genet., 3, 1449–1452.[Abstract]

Cooke, H. and Elliott, D.J. (1997) RNA-binding proteins and human male infertility. Trends Genet., 13, 87–89.[ISI][Medline]

Delbridge, M.L., Harry, J.L., Toder, R. et al. (1997) A human candidate spermatogenesis gene, RBM1, is conserved and amplified on the marsupial Y chromosome. Nature Genet., 15, 131–136.[ISI][Medline]

Duell, T., Mathews, S., Wunderlich, B. et al. (1998) Interstitial and terminal deletion of chromosome Y in a male individual with cryptozoospermia. Mol. Hum. Reprod., 4, 325–331.[Abstract]

Eberhart, C.G., Maines, J.Z. and Wasserman, S.A. (1996) Meiotic cell cycle requirement for a fly homologue of human Deleted in Azoospermia. Nature, 381, 783–785.[ISI][Medline]

Elliott, D.J., Millar, M.R., Oghene, K. et al. (1997) Expression of RBM in the nuclei of human germ cells is dependent on a critical region of the Y chromosome long arm. Proc. Natl Acad. Sci. USA, 94, 3848–3853.[Abstract/Free Full Text]

Elliott, D.J., Oghene, K., Makarov, G. et al. (1998) Dynamic changes in the subnuclear organisation of pre-mRNA splicing proteins and RBM during human germ cell development. J. Cell Sci., 111, 1255–1265.[Abstract/Free Full Text]

Foresta, C. and Varotto, A. (1992) Assessment of testicular cytology by fine needle aspiration as a diagnostic parameter in the evaluation of the oligozoospermic subject. Fertil. Steril., 58, 1028–1033.[ISI][Medline]

Foresta, C., Varotto, A. and Scandellari, C. (1992) Assessment of testicular cytology by fine needle aspiration as a diagnostic parameter in the evaluation of the azoospermic subject. Fertil. Steril., 57, 858–865.[ISI][Medline]

Foresta, C., Ferlin, A., Bettella, A. et al. (1995) Diagnostic and clinical features in azoospermia. Clin. Endocrinol., 43, 537–543.[ISI][Medline]

Foresta, C., Ferlin, A., Garolla, A. et al. (1997) Y-chromosome deletions in idiopathic severe testiculopathies. J. Clin. Endocrinol. Metab., 82, 1075–1080.[Abstract/Free Full Text]

Foresta, C., Ferlin, A., Garolla, A. et al. (1998) High frequency of well-defined Y-chromosome deletions in idiopathic Sertoli cell-only syndrome. Hum. Reprod., 13, 302–307.[ISI][Medline]

Girardi, S.K., Mielnik, A. and Schlegel, P.N. (1997) Submicroscopic deletions in the Y chromosome of infertile men. Hum. Reprod., 12, 1635–1641.[Abstract]

Gläser, B., Herl, T., Taylor, K. et al. (1997) High resolution fluorescence in situ hybridization of human Y-linked genes on released chromatin. Chrom. Res., 5, 23–30.[ISI][Medline]

Habermann, B., Mi, H.F., Edelmann, A. et al. (1998) DAZ (Deleted in Azoospermia) genes encode proteins located in human late spermatids and in sperm tails. Hum. Reprod., 13, 363–368.[ISI][Medline]

Kobayashi, K., Mizumo, K., Hida, A. et al. (1995) PCR analysis of the Y chromosome long arm in azoospermic patients: evidence for a second locus required for spermatogenesis. Hum. Mol. Genet., 3, 1965–1967.[Abstract]

Lahn, B.T. and Page, D.C. (1997) Functional coherence of the human Y chromosome. Science, 278, 675–680.[Abstract/Free Full Text]

Lee, J.H., Lee, D.R., Yoon, S.J. et al. (1998) Expression of DAZ (deleted in azoospermia), DAZL1 (DAZ-like) and protamine-2 in testis and its application for diagnosis of spermatogenesis in non-obstructive azoospermia. Mol. Hum. Reprod., 4, 827–834.[Abstract]

Liow, S.L., Ghadessy, F.J., Ng, S.C. et al. (1998) Y chromosome microdeletions, in azoospermic or near-azoospermic subjects, are located in the AZFc (DAZ) subregion. Mol. Hum. Reprod., 4, 763–768.[Abstract]

Ma, K., Inglis, J.D., Sharkey, A. et al. (1993) A Y chromosome gene family with RNA-binding protein homology: candidates for the azoospermia factor AZF controlling human spermatogenesis. Cell, 75, 1287–1295.[ISI][Medline]

Mazeyrat, S., Sant, N., Sargent, C.A. et al. (1998) The mouse Y chromosome interval necessary for spermatogonial proliferation is gene dense with syntenic homology to the human AZFa region. Hum. Mol. Genet., 7, 1713–1724.[Abstract/Free Full Text]

Menke, D.B., Mutter, G.L. and Page, D.C. (1997) Expression of DAZ, an Azoospermia factor candidate, in human spermatogonia. Am. J. Hum. Genet., 60, 237–240.[ISI][Medline]

Mulhall, J.P., Reijo, R., Alagappan, R. et al. (1997) Azoospermic men with deletion of the DAZ gene cluster are capable of completing spermatogenesis: fertilization, normal embryonic development and pregnancy occur when retrieved testicular spermatozoa are used for intracytoplasmic sperm injection. Hum. Reprod., 12, 503–508.[ISI][Medline]

Najmabadi, H., Huang, V., Yen, P. et al. (1996a) Substantial prevalence of microdeletions of the Y-chromosome in infertile men with idiopathic azoospermia and oligozoospermia detected using a sequence-tagged site-based mapping strategy. J. Clin. Endocrinol. Metab., 81, 1347–1352.[Abstract]

Najmabadi, H., Chai, N.N., Kapali, A. et al. (1996b) Genomic structure of a Y-specific ribonucleic acid binding motif-containing gene: a putative candidate for a subset of male infertility. J. Clin. Endocrinol. Metab., 81, 2159–2164.[Abstract]

Pryor, J.L., Kent-First, M., Muallem, A. et al. (1997) Microdeletions in the Y chromosome of infertile men. N. Engl. J. Med., 336, 534–539.[Abstract/Free Full Text]

Qureshi, S.J., Ross, A.R., Ma, K. et al. (1996) Polymerase chain reaction screening for Y chromosome microdeletions: a first step towards the diagnosis of genetically-determined spermatogenic failure in men. Mol. Hum. Reprod., 2, 775–779.[Abstract]

Reijo, R., Lee, T., Salo, P. et al. (1995) Diverse spermatogenic defects in humans caused by Y chromosome deletions encompassing a novel RNA-binding protein gene. Nature Genet., 10, 383–393.[ISI][Medline]

Reijo, R., Alagappan, R.K., Patrizio, P. et al. (1996) Severe oligozoospermia resulting from deletions of azoospermia factor gene on Y chromosome. Lancet, 347, 1290–1293.[ISI][Medline]

Rossato, M., Ferlin, A., Garolla, A. et al. (1998) High fertilization rate in conventional in-vitro fertilization utilizing spermatozoa from an oligozoospermic subject presenting microdeletions of the Y chromosome long arm. Mol. Hum. Reprod., 4, 473–476.[Abstract]

Saxena, R., Brown, L., Hawkins, T. et al. (1996) The DAZ gene cluster on the human Y chromosome arose from an autosomal gene that was transposed, repeatedly amplified and pruned. Nature Genet., 14, 292–299.[ISI][Medline]

Simoni, M., Gromoll, J., Dworniczak, B. et al. (1997) Screening for deletions of the Y chromosome involving the DAZ (Deleted in AZoospermia) gene in azoospermia and severe oligozoospermia. Fertil. Steril., 67, 542–547.[ISI][Medline]

Stuppia, L., Mastroprimiano, G., Calabrese, G. et al. (1996) Microdeletions in interval 6 of the human Y chromosome detected by STS-PCR in 6 of 33 patients with idiopathic oligo-azoospermia. Cytogenet. Cell. Genet., 72, 155–158.[ISI][Medline]

Stuppia, L., Gatta, V., Mastroprimiano, G. et al. (1997) Clustering of Y chromosome deletions in subinterval E of interval 6 supports the existence of an oligozoospermia critical region outside the DAZ gene. J. Med. Genet., 34, 881–883.[Abstract]

Tiepolo, L. and Zuffardi, O. (1976) Localization of factors controlling spermatogenesis in the nonfluorescent portion of the human Y chromosome long arm. Hum. Genet., 34, 119–124.[ISI][Medline]

Vereb, M., Agulnik, A.I., Houston, J.T. et al. (1997) Absence of DAZ gene mutations in cases of non-obstructed azoospermia. Mol. Hum. Reprod., 3, 55–59.[Abstract]

Vogt, P.H., Edelmann, A., Kirsch, S. et al. (1996) Human Y chromosome azoospermia factors (AZF) mapped to different subregions in Yq11. Hum. Mol. Genet., 5, 933–943.[Abstract/Free Full Text]

Vogt, P.H., Affara, N., Davey, P. et al. (1997) Report of the third international workshop on Y chromosome mapping 1997. Cytogenet. Cell. Genet., 79, 1–20.[Medline]

Vollrath, D., Foote, S., Hilton, A. et al. (1992) The human Y chromosome: a 43-interval map based on naturally occurring deletions. Science, 258, 52–59.[ISI][Medline]

Word Health Organization (1992) WHO Laboratory Manual for the Examination of Human Semen and Sperm–Cervical Mucus Interaction. Cambridge University Press, Cambridge.

Yen, P.H. (1998) A long-range restriction map of deletion interval 6 of the human Y chromosome: a region frequently deleted in azoospermic males. Genomics, 54, 5–12.[ISI][Medline]

Yen, P.H., Chai, N.N. and Salido, E.C. (1997) The human DAZ genes, a putative male infertility factor on the Y chromosome, are highly polymorphic in the DAZ repeat regions. Mamm. Genome, 8, 756–759.[ISI][Medline]

Submitted on November 30, 1998; accepted on March 18, 1999.