This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Foresta, C. Articles by Ferlin, A. Search for Related Content PubMed PubMed Citation Articles by Foresta, C. Articles by Ferlin, A.

Y Chromosome Microdeletions and Alterations of Spermatogenesis¹

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

	Abstract

Top Abstract I. Introduction II. Structure and Gene... III. The Azoospermia Factors IV. AZFb Region and... V. AZFc Region and... VI. AZFa Region and... VII. Y Chromosome Microdeletions... VIII. Conclusions References

 
Three different spermatogenesis loci have been mapped on the Y chromosome and named "azoospermia factors" (AZFa, b, and c). Deletions in these regions remove one or more of the candidate genes (DAZ, RBMY, USP9Y, and DBY) and cause severe testiculopathy leading to male infertility. We have reviewed the literature and the most recent advances in Y chromosome mapping, focusing our attention on the correlation between Y chromosome microdeletions and alterations of spermatogenesis. More than 4,800 infertile patients were screened for Y microdeletions and published. Such deletions determine azoospermia more frequently than severe oligozoospermia and involve especially the AZFc region including the DAZ gene family. Overall, the prevalence of Y chromosome microdeletions is 4% in oligozoospermic patients, 14% in idiopathic severely oligozoospermic men, 11% in azoospermic men, and 18% in idiopathic azoospermic subjects. Patient selection criteria appear to substantially influence the prevalence of microdeletions. No clear correlation exists between the size and localization of the deletions and the testicular phenotype. However, it is clear that larger deletions are associated with the most severe testicular damage. Patients with Y chromosome deletions frequently have sperm either in the ejaculate or within the testis and are therefore suitable candidates for assisted reproduction techniques. This possibility raises a number of medical and ethical concerns, since the use of spermatozoa carrying Y chromosome deletions may produce pregnancies, but in such cases the genetic anomaly will invariably be passed on to male offspring.

I. Introduction

II. Structure and Gene Content of the Y Chromosome

III. The Azoospermia Factors

IV. AZFb Region and Candidate Genes

V. AZFc Region and Candidate Genes

VI. AZFa Region and Candidate Genes

VII. Y Chromosome Microdeletions in Infertile Men

A. Methods of detection

B. Selection of patients, phenotype-genotype relation, and origin of deletions

C. The concern of assisted reproduction techniques

VIII. Conclusions

	I. Introduction

Top Abstract I. Introduction II. Structure and Gene... III. The Azoospermia Factors IV. AZFb Region and... V. AZFc Region and... VI. AZFa Region and... VII. Y Chromosome Microdeletions... VIII. Conclusions References

 
THE Y chromosome is not essential for life and until recently most regions of it were assumed to be functionally inert. Sex determination (controlled by the SRY gene) (1) has long been viewed as the sole function related to the Y chromosome (2), but this theory changed in recent years when another important function (the control of spermatogenesis) (3) was discovered and many genes were mapped to the Y chromosome.

Spermatogenesis is a long and complex process requiring about 70 days and involving an elaborate succession of distinct cell types (4, 5) generated by mitotic and meiotic divisions. In the initial stages, spermatogonia divide via mitoses, giving rise to primary spermatocytes, which in turn undergo the first meiotic division leading to secondary spermatocytes. Through the second meiosis these cells produce haploid cells (round spermatids), which elongate during the spermiogenesis process (elongated spermatids) and finally differentiate into mature spermatozoa, by condensation of the chromatin, substitution of histones with protamines, and formation of the acrosome and the other sperm components. However, our knowledge of the mechanisms regulating spermatogenesis is still poor, and only recently has research focused on the identification of genes specifically involved in its regulation.

Nevertheless, infertility is a major health problem affecting 10–15% of couples seeking to have children, and a male factor can be identified in about half of these cases (6). A significant proportion of infertile males are affected either by oligozoospermia (reduced sperm production) or azoospermia (lack of any sperm in the ejaculate). Such alterations in sperm production may be related, in turn, to different underlying testicular histological pathologies, ranging from the complete absence of germ cells (Sertoli cell-only syndrome) to hypospermatogenesis and maturation arrest. The alteration of spermatogenesis can be the consequence of many causes, such as systemic diseases, cryptorchidism, endocrinological disorders, obstruction/absence of seminal pathways, or infections. However, the cause of male infertility is unknown in up to 50% of cases, and until recently relatively little research focused on the possible genetic etiologies. The explosive growth of assisted reproduction techniques and, in particular, of intracytoplasmic sperm injection (ICSI) (7) has contributed to the development of such research. The study of Y chromosome microdeletions is particularly important because of the potential for transmission of genetic abnormalities to the offspring, as these techniques bypass the physiological mechanisms related to fertilization.

	II. Structure and Gene Content of the Y Chromosome

Top Abstract I. Introduction II. Structure and Gene... III. The Azoospermia Factors IV. AZFb Region and... V. AZFc Region and... VI. AZFa Region and... VII. Y Chromosome Microdeletions... VIII. Conclusions References

 
The Y chromosome is the smallest human chromosome and consists of a short (Yp) and a long (Yq) arm. The pseudoautosomal regions (PARs), which pair with the X chromosome during meiosis, are located at both ends. The region outside the PARs that does not recombine is called the nonrecombining region of the Y chromosome (NRY) (Fig. 1). This part consists of several repetitive sequences that may be homologs to regions on the X chromosome or Y-specific. The Yp and the proximal part of Yq consist of euchromatin, while the distal part of the long arm is made of heterochromatin, and this region may vary in length to constitute about one-half to two-thirds of Yq (Fig. 1). Therefore, the Y chromosome long arm may be cytogenetically divided in an euchromatic proximal region (Yq11, subdivided into Yq11.1, 11.21, 11.22, and 11.23) and a heterochromatic distal region (Yq12), whereas the euchromatic short arm is called Yp11 (Fig. 2a).

View larger version (10K): [in this window] [in a new window]   Figure 1. Schematic representation of the Y chromosome. The PARs are represented in black, the euchromatic region in white, and the heterochromatic region is striped. Outside the PARs is the nonrecombining region of the Y chromosome (NRY). Below are schematically represented the phenotypes associated with alterations of the Y chromosome (sex determination, risk of gonadoblastoma, small stature, and the three spermatogenic failure loci AZFa, b, and c). Yp and Yq, Short and long arm, respectively.

View larger version (35K): [in this window] [in a new window]   Figure 2. a, Representation of the cytological bands of the Y chromosome. The short arm is called Yp11, and the long arm is Yq11 (euchromatic region) and Yq12 (heterochromatic region, striped). b, The seven intervals of the Vergnaud map of the Y chromosome (8 ), where intervals 1–4 span the short arm and the centromere, intervals 5 and 6 span the euchromatic region, and interval 7 spans the heterochromatic region. c, The 43 interval map of the Y chromosome (9 ). On the right are represented a list of genes mapped to the Y chromosome, the localization of AZF regions and the corresponding candidate genes.

Initial attempts to draw a physical map of the Y chromosome led to the isolation of overlapping yeast artificial chromosome (YAC) contigs (10, 11). More recent efforts in sequencing came from a systematic approach within the Human Genome Project. With the major contribution of the two leading centers involved in this project (Washington University and Whitehead Institute), nearly 40% of the euchromatic region of the Y chromosome (about 13 Mb out of the estimated 35 Mb) have been sequenced, and more than 40 contigs have been isolated (data updated at the end of June 2000; official site www.ncbi.nlm.nih.gov/genome). Even if only half of them have been physically mapped to date, a finished sequence of the entire Y chromosome will be available in the near future. Together with this sequence map, more than 300 sequence-tagged sites (STSs) have been generated and mapped. STSs are known sequences of genomic DNA that can be amplified by PCR. These STSs may be specific for a gene or may overlap anonymous regions of the Y chromosome, and their use in Y deletion screening will be discussed in Section VII.

Many genes on the Y chromosome have been identified only recently, and perhaps novel genes will be described once sequencing has been completed; so far, more than 30 genes and gene families are known. As summarized by Yen (12), these genes can be classified into three groups on the basis of their location on the Y, their copy number, and their pattern of expression: 1) pseudoautosomal genes (such as ASMTL, MIC2, and IL9R): their sequences are identical on the Y and X chromosomes and, with few exceptions, they are expressed in different tissues; 2) genes located within X-Y homologous regions on the NRY (such as USP9Y, DBY, and UTY): these have homologs on the X chromosome encoding for proteins with very high amino acid identity. These genes are ubiquitously expressed, although some have testis- specific transcripts in addition to ubiquitous transcripts; 3) Y-specific gene families (such as DAZ, CDY, and TSPY); these are multicopy genes, widely distributed on the Y chromosome or clustered within a small region, and they are expressed only in the testis. One exception to this classification is SRY, the gene that determines testis development: it is Y-specific, but it is in single copy and has a different pattern of expression, limited to the genital ridge and in fetal and adult Sertoli cells and germ cells (13, 14, 15).

	III. The Azoospermia Factors

Top Abstract I. Introduction II. Structure and Gene... III. The Azoospermia Factors IV. AZFb Region and... V. AZFc Region and... VI. AZFa Region and... VII. Y Chromosome Microdeletions... VIII. Conclusions References

 
Tiepolo and Zuffardi in 1976 (3) were the first to hypothesize a correlation between Y chromosome deletions and male infertility. These authors examined the karyotype of 1,170 men; in six sterile males with azoospermia they observed large deletions including the entire heterochromatic region (Yq12) and an undefined amount of the adjacent euchromatic part (Yq11). In two cases they demonstrated that the fathers of the patients with deletions carried a normal Y chromosome, indicating that these mutations were de novo events. This suggested that the deletions were the cause of the azoospermia and they postulated that a genetic factor located in Yq11 was important for male germ cell development. This gene or gene cluster was defined as "azoospermia factor" (AZF). However, the genetic complexity of the AZF locus could be revealed only with the development of STS- and YAC-based mapping. These analyses permitted the detection of interstitial submicroscopic deletions not visible at the cytogenetic level and detectable only by STS-PCR or Southern hybridization. Such deletions are called microdeletions. Molecular mapping analyses on patients with microdeletions have complicated the original hypothesis of a single locus for spermatogenesis on Yq, suggesting that three nonoverlapping regions in deletion intervals 5 and 6 may be deleted in infertile men. These spermatogenesis loci are termed AZFa, AZFb, and AZFc (16) from proximal to distal Yq. Furthermore, a fourth region (AZFd) has been proposed between AZFb and AZFc (17), but this finding must be confirmed. According to Vogt et al. (18) (Fig. 2c), AZFa is located at the proximal portion of deletion interval 5 (subinterval 5C), AZFb spans from the distal portion of deletion interval 5 to the proximal end of deletion interval 6 (subinterval 5O–6B), and AZFc is located at the distal part of deletion interval 6 (subintervals 6C–6E). Several genes located in AZF regions are expressed in the testis and could therefore be viewed as "AZF candidate genes." However, based on studies of infertile patients, only a few genes can actually be considered responsible for the AZF phenotype. The first AZF-candidate gene was isolated in 1993 from a region subsequently shown to correspond to AZFb. Two years later the second AZF-candidate gene was identified from the AZFc region. Both genes have been quite well studied both at the molecular level and in terms of deletions in infertile patients. The structure of AZFa and its gene content have only recently been described, and analyses of this region in infertile males suggested that two genes may be considered AZFa-candidate. However, it should be noted that recent findings of many genes or gene families outside the proposed AZF regions (19) suggest that even this classification may be an oversimplified picture.

	IV. AZFb Region and Candidate Genes

Top Abstract I. Introduction II. Structure and Gene... III. The Azoospermia Factors IV. AZFb Region and... V. AZFc Region and... VI. AZFa Region and... VII. Y Chromosome Microdeletions... VIII. Conclusions References

 
A detailed sequence and gene map of AZFb is still not available, and only nonoverlapping YAC and BAC clones have been described. Even if the AZFb interval, as usually defined, spans the subintervals 5O–6B, the precise boundaries can differ and its exact extension is unclear. This depends on different deletion events between individuals and differences in screening procedures. Microdeletions can remove AZFb alone, parts of AZFb, or also include flanking regions (e.g., AZFc). The extent of deletions will obviously affect which genes are removed: for example, deletions may remove the block SMCY-XKRY-CDY2 (19), or the block PRY-TTY2 (19) if they extend proximally or distally, respectively.

To date, two genes have been mapped to subintervals 5O–6B: EIF1AY (translation-initiation factor 1A, Y isoform) and RBMY (RNA binding motif on the Y). EIF1AY encodes a Y isoform of eIF-1A, an essential translation initiation factor, which has an X homolog and is ubiquitously expressed (19). Its role in spermatogenesis is completely unknown, and no deletion specifically removing this gene has been reported. Therefore, it is not considered an AZFb-candidate gene. However, EIF1AY possesses abundant testis-specific transcripts in addition to ubiquitous transcripts (19), suggesting that this gene may contribute to the AZFb phenotype.

RBMY was the first among the AZF candidate genes to be identified and was cloned using patient DNA with a deletion in the proximal region of interval 6 (20). Initially, two similar cDNAs were isolated and named YRRM1 and 2 (Y-specific RNA recognition motif) as they were predicted to encode a protein with an RNA-binding motif. Subsequently it was shown that, in fact, there is a family of 20–50 genes and pseudogenes spread over both arms of the Y chromosome, including a cluster within the AZFb region (21, 22), and YRRM was renamed "RBMY gene family" (18). Such copies belong to at least six subfamilies (23, 24), but RBMY-I is the only actively transcribed gene, and the most functional copies are located on interval 6B (22), thus making it a major candidate for the AZFb region (18). RBMY is present as multiple copies in all eutherian ("placental") mammals (25) and has an active X-borne homolog recently discovered both in humans and marsupials (26, 27). It has been proposed that RBMX retained a widespread function while RBMY evolved a male-specific function in spermatogenesis.

The RBMY proteins (Fig. 3) consist of a single RNA-binding domain of the RRM (RNA recognition motif) type at the N terminus and an auxiliary C-terminal domain containing four 37-amino acid (aa) repeats. This domain is known as the SRGY box, since it contains a serine-arginine-glycine- tyrosine sequence (20, 28). The gene consists of 12 exons, and the 37-residue repeats are encoded by single exons (exons 7–10) (Fig. 3), which show more than 85% homology between their nucleotide sequences (28).

View larger version (9K): [in this window] [in a new window]   Figure 3. Genomic structure of the RBMY1 gene and relative protein. Exons 2–4 (black) encode for the RNA-binding domain of the protein, while each of exons 7–10 (111 bp) encode for a single 37-aa SRGY box (striped). Identity at the amino acid level between RBMY1 and mouse Rbm is shown in the lower part of the figure.

RBMY is considered the major AZFb candidate gene given its testis specificity, its absence in a fraction of infertile patients, and its homology with the mouse Rbm, deletion of which causes male sterility (30, 31, 32). However, the multicopy nature of this gene has complicated attempts to prove its role in human spermatogenesis, as detrimental mutations in patients have not yet been identified.

	V. AZFc Region and Candidate Genes

Top Abstract I. Introduction II. Structure and Gene... III. The Azoospermia Factors IV. AZFb Region and... V. AZFc Region and... VI. AZFa Region and... VII. Y Chromosome Microdeletions... VIII. Conclusions References

 
In an attempt to correlate severe spermatogenic defects with frequent and consistent de novo microdeletions of the Y chromosome, Reijo and colleagues (33) cloned a novel gene from the distal portion of deletion interval 6 shown to be deleted in men with azoospermia. This gene was named DAZ (deleted in azoospermia) and was originally thought to be single copy, whereas it is now clear that it is a member of a multigene family with more than one copy on the Y chromosome, clustered in the AZFc region (18, 34, 35, 36). Therefore, DAZ was renamed "DAZ gene family" (18). The exact number of genes in the family is still not clear, although at least three copies have been described by Southern blotting or restriction mapping (34, 36, 37), and seven copies, representing either active genes or pseudogenes, may be seen by fiber-fluorescent in situ hybridization (35).

The structure of the DAZ gene is somewhat similar to that of RMBY (Fig. 4). DAZ encodes a protein with a single RNA-binding domain at the N terminus and a C-terminal domain containing an internally repeated sequence of 24 aa, the so-called DAZ repeats (33, 34). The DAZ transcription unit appears to contain at least 16 exons and to span about 42 kb. Exon 1 consists of the initiator codon, exons 2–5 encode the RNA-binding domain, and each of exons 7a–7 g encodes a single DAZ repeat (34).

View larger version (13K): [in this window] [in a new window]   Figure 4. Genomic structure of the DAZ gene and relative protein. Exons 2–5 (black) encode for the RNA-binding domain of the protein, while each of exons 7a–7g (72 bp) encode for a single 24-aa DAZ repeat (striped). Identity at the amino acid level among DAZ, autosomal human DAZL1, mouse Dazl, and Drosophila boule is shown in the lower part of the figure.

DAZ is found on the Y chromosome only in humans, Old World monkeys, and apes (34, 42, 43, 44). In all other mammals it is represented as an autosomally located, single copy gene (42, 45, 46, 47). DAZ was acquired by the Y chromosome from an autosomal homolog DAZL1 located on chromosome 3p24 and with a single DAZ repeat (34, 48, 49, 50, 51) (Fig. 4). A complete DAZL1 copy was transposed to the Y chromosome, a 2.4-kb genomic sequence encompassing exons 7 and 8 was tandemly repeated, and, finally, the whole transcription unit was amplified, giving rise to a multicopy gene family (34, 37).

Although DAZ is not the only gene present in the distal Yq interval 6 (19, 36, 52), its high prevalence of deletions in infertile men makes it the major AZFc candidate. This possibility is further strengthened by the high homology of DAZ with a Drosophila male infertility gene, boule, mutation of which causes spermatogenic arrest (53, 54). Furthermore, more recent proof of the spermatogenic role of the DAZ gene product arises from the observation that a human DAZ transgene is capable of partially rescuing the sterile phenotype of a mouse knockout for the homologous gene Dazl (55). However, most difficulties in understanding the biological function of DAZ and the genotype-phenotype relation probably arise from the multicopy nature of this gene. Like RBMY, deletions of DAZ in infertile patients are generally screened by PCR on genomic DNA extracted from peripheral leukocytes. Therefore, only deletions removing the whole of the DAZ gene cluster can be detected, and intragenic deletions or deletions not involving all the DAZ copies, as well as de novo point mutations in affected patients, have yet to be discovered. Therefore, there is still no definitive proof for a requirement of DAZ in spermatogenesis.

Although a sequence map of AZFc is not yet available, several genes other than DAZ have been mapped to this region (19, 36): CDY1 (chromodomain Y 1), BPY2 (basic protein Y 2), PRY (PTA-BL related Y), and TTY2 (testis transcript Y 2). The function of these genes is unknown, but they share similar characteristics: they are in multiple copies on the Y chromosome, they are expressed in the testis only, and they are Y specific (19). In particular, three PRY and TTY2 genes have been identified in the proximal part of AZFc by restriction mapping (36), and therefore they are probably not involved in the spermatogenic disruption observed in patients with deletion limited to DAZ. Two CDY1 genes map in the AZFc region, one within the DAZ cluster and the other one at the distal end (36). This finding is intriguing since at least one CDY1copy in invariably absent in patients with DAZ deletion. Therefore, CDY1 can be considered an AZFc-candidate gene, but deletions removing this gene specifically should be identified in patients to confirm this hypothesis. No mapping and deletion data are available for BPY2, which, therefore, is not included at present among the AZFc- candidate genes.

	VI. AZFa Region and Candidate Genes

Top Abstract I. Introduction II. Structure and Gene... III. The Azoospermia Factors IV. AZFb Region and... V. AZFc Region and... VI. AZFa Region and... VII. Y Chromosome Microdeletions... VIII. Conclusions References

 
The most recent data suggest that more than one gene may be responsible for the AZFa phenotype. However, the characterization of the critical interval is still under way, and the structure and gene constitution of this region have only recently been described (19, 56, 57, 58, 59).This region contains three genes and has a syntenic homology with the mouse Sxr^b interval (56) (Fig. 5) located on the Y chromosome short arm, deletion of which causes a severe spermatogenic impairment (60), very similar to that observed in patients with the AZFa deletion.

View larger version (16K): [in this window] [in a new window]   Figure 5. The human AZFa region and the mouse Sxrb interval. a, The AZFa region is included in two nonoverlapping contigs, spanning more than 1.2 Mb. b, Ten overlapping clones covering the AZFa region. c, Localization of the three genes, USP9Y, DBY, and UTY, with respect to the clones. The arrows indicate the 5'–3' orientation of the genes. Note that UTY is not completely included in the clone sequences and therefore the complete genomic structure of this gene is still unknown. d, Transcription map of the mouse Sxrb interval (spanning more than 900 kb), showing the syntenic homology with the human AZFa interval: the block Dffry, Dby, and Uty is highly conserved between human and mouse. The numbers below the clones and contigs represent the GenBank accession numbers.

	VII. Y Chromosome Microdeletions in Infertile Men

Top Abstract I. Introduction II. Structure and Gene... III. The Azoospermia Factors IV. AZFb Region and... V. AZFc Region and... VI. AZFa Region and... VII. Y Chromosome Microdeletions... VIII. Conclusions References

 
A. Methods of detection
Once a map of PCR markers covering the Y chromosome was assembled (9, 69), the first searches for interstitial deletions in infertile men were published (33, 70, 71). Such PCR markers are known as STSs, and to date more than 300 have been physically mapped. Each STS detects known sequences of genomic DNA, and its normal amplification by PCR indicates the presence of this DNA sequence in the Y chromosome, while its absence may indicate a deletion. An STS may be specific for a gene or gene family or may detect anonymous sequences.

Most problems with the STS-PCR technique in screening for microdeletions derive from the intrinsic nature of the Y chromosome, which largely consists of repetitive elements and gene families widely dispersed along the chromosome. As described by Kostiner et al. (72), STS markers fall into three categories: 1) markers that are single copy, such as those for SRY, USP9Y, DBY, or UTY; these STSs are the most informative given their specificity, and their absence indicates the loss of the specific gene sequence; 2) markers that are multicopy but clustered in a small region of the Y chromosome, such as those for the DAZ gene family; in such a case a negative PCR result indicates the loss of all members of the gene family. However, the presence of a PCR band is less informative, since a normal amplification merely indicates the presence of at least one gene copy containing the marker sequence; 3) markers that are multicopy and dispersed across large regions of the Y chromosome, such as those for RBMY, CDY, or TSPY or other multicopy gene families widely distributed on the chromosome; such repetitive markers may be informative only when their absence indicates that very large deletions of the Y chromosome have occurred, e.g., encompassing the entire Yq11.

Some markers may represent normal polymorphisms, as they may be absent in both fertile and infertile patients. Polymorphic markers (such as sY207 and sY272) are not useful in screening since their absence does not represent significant deletions (73).

The STS-PCR technique is performed on genomic DNA extracted from peripheral leukocytes; although rapid and simple, certain precautionary measures should be applied in Y chromosome deletion screening before assuming that a deletion exists; these measures include the use of high quality DNA and internal and external positive and negative controls (SRY or ZFX/ZFY gene, fertile male, normal woman, and blank controls). Furthermore, European guidelines for the molecular diagnosis of Y chromosome microdeletions (74) suggest that the screening should be performed by multiplex PCR amplification in which an internal control is amplified together with the selected STS(s). The number of STSs to be used for a first screening is not well established, and it varies substantially among the authors. However, as a general rule, at least two or three STSs for each AZF region should be used; if a deletion is found and confirmed, the number of STSs should be increased to determine the deletion breakpoints.

Given the repetitive nature of some STSs and the quite frequent finding of noncontiguous deletions, the results obtained by PCR should be sometimes confirmed by Southern blotting experiments, especially if a deletion not encompassing known spermatogenesis loci is found. Furthermore, a deletion may be assumed to have a pathogenic role only if it is demonstrated that it is of de novo origin and not present in other fertile family members. Therefore, when available, male relatives of patients should be analyzed.

The rapid progress in molecular biology technologies will allow us, in the near future, to improve both the quality of the diagnosis and the time needed for a Y deletion screening. For example, fluorescent PCR will probably take the place of the conventional PCR method, since it is 1,000-fold more sensitive, less time consuming, and would allow the accurate testing of a small amount of DNA (e.g., for single cell analysis). The analysis of Y chromosome deletions may also shift to other more automated methods of detection, such as the microarrays technique. However, time is needed to increase our knowledge of Yq genes and mapping before this method may completely replace traditional methods.

B. Selection of patients, genotype-phenotype relation, and origin of deletions
The recent progress in molecular biology and Y chromosome mapping has rendered the analysis of Y chromosome microdeletions in infertile patients a routine diagnostic step. As a result, numerous studies on this topic have been published: more than 4,800 infertile men have been studied. Taken together, these data suggest that Y chromosome microdeletions constitute one of the most common specific causes of male infertility. In fact, based on studies from 1992, the overall prevalence in infertile men is 8.2% (401/4,868) (16, 17, 33, 66, 70, 73, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102) (Fig. 6). In contrast, only 12 of 2,663 (0.4%) fertile males were found to carry a deletion, probably reflecting polymorphisms, as discussed below. However, among the various studies, remarkable differences in the prevalence of microdeletions exist, ranging from 1% (85) to 35% (63), reflecting above all different patient selection criteria. Therefore, the actual incidence of clinically relevant microdeletions in infertile men is still unclear. In this review we have summarized the studies published up to May 2000, grouping them on the basis of the various patient classifications used. In fact, male infertility is a heterogeneous diagnostic category that may be classified only on clinical and historical data, or on seminological data (normo-, oligo-, azoospermia) and/or on testicular structure (Sertoli cell-only syndrome, hypospermatogenesis, spermatogenic arrest, obstructive forms) (103). Figure 6 shows the categories of infertile men, as described in the original studies: in some reports other or nonuniform classifications are used, and therefore only 3,640 of 4,868 patients may be clearly grouped (16, 17, 33, 62, 63, 73, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98). To clarify a genotype-phenotype relation, the different classification of infertile patients is even more crucial.

View larger version (28K): [in this window] [in a new window]   Figure 6. Summary of the literature on Yq microdeletions in infertile males from 1992 to May 2000 (16 17 33 63 64 66 70 73 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 ). The first category at the top of the figure represents all patients published, while the other categories have been grouped from homogeneous studies in which a clear description of patients was reported. The prevalence of Yq microdeletions was calculated regardless of the number of STSs used or genes analyzed.

The variable phenotype observed both in patients with AZFb and in patients with AZFc deletions may be explained by different hypotheses:

1. Different extension of the deletion. Deletions may completely remove AZFb or AZFc or they may be smaller, extending, for example, only one gene or gene cluster or few STS markers. It can be speculated that larger deletions may be associated with a more severe phenotype. However, neither a review of the literature nor our experience support this hypothesis. For example, deletions removing only DAZ and not extending to flanking regions may cause both azoospermia and severe oligozoospermia. Furthermore, the combined deletion of DAZ and CDY1 do not seem to cause a phenotype worse than that observed in patients who retained the distal CDY1 copy (A. Ferlin, E. Moro, A. Rossi, and C. Foresta, submitted). However, it should be kept in mind that apparently identical deletions, as assessed by STS-PCR, may be actually slightly different in size and this may, at least in part, justify a different phenotype.

2. Role of homologous genes and genetic background. Each of the AZF candidate gene has homologs on the X chromosome (DBX, USP9X, RBMX) or on autosomes (DAZL1), as many other genes on the Y chromosome have (e.g., EIF1AY has a homolog on the X chromosome, and CDY has a homolog on chromosome 6). Even if no direct evidence of a role of these genes in spermatogenesis exists, their status may modify the phenotypic expression of Y-deleted patients. This is particularly suggestive for DAZL1/DAZ, since DAZL1 is expressed exclusively in the gonads and may therefore synergistically act in combination with DAZ during spermatogenesis. Therefore, the genetic background may modulate the phenotypic effect of a given deletion, and the absence of an AZF gene may be differently compensated by other genes of the family.

3. Progression of the spermatogenic failure. It has been suggested that Yq microdeletions could result in progressive worsening of sperm production (88, 104, 116), and that oligozoospermic men may become azoospermic with time. This is an intriguing hypothesis that, if confirmed, has important prognostic consequences, since cryoconservation of spermatozoa in these cases could avoid more invasive techniques, such as ICSI, in the future.

From the analysis of the literature the only clear correlation that has been found is with very large deletions involving more than one AZF locus. These deletions are associated with the most severe phenotype, and this is exemplified by the invariable finding of azoospermia and Sertoli cell-only in patients with deletions of AZFa-c. Such large deletions probably show the additional effects of each single gene deletion. These data are also in agreement with the report that smaller deletions are associated with finding some sperm at TESE (testicular sperm extraction) during ICSI treatments, while larger deletions are not (92).

For 401 Y-deleted patients, 136 male relatives (father and/or brothers) were analyzed, and in 95% of cases (394/401) no microdeletions were found (Fig. 6). In the other seven cases a deletion was found both in the father and in his infertile son (16, 17, 70, 73, 105); in some cases the deletion involved STS markers subsequently shown to represent polymorphisms, but in other descriptions the father was found to carry a clinically relevant deletion of apparently identical size to that of the son, or even a smaller deletion (105). These findings demonstrate that such nonpolymorphic deletions can, in rare cases, be passed on from father to son, and may become larger (with possible additional negative effects). In most cases, these naturally occurring transmissions involved the AZFc region, and this was confirmed by a recent study that reported an identical AZFc deletion including DAZ in four infertile brothers and their father (104). The father was azoospermic at the time of analysis, but he evidently possessed some degree of fertility when he fathered, suggesting that the loss of germ cells caused by DAZ deletions may be progressive over the years. However, as seen above, it is possible that apparently identical deletions, as determined by STS analysis, are actually different in size, and that the father, for example, could have a smaller deletion than his infertile sons. Alternatively, a different exposure to the environment or a different genetic background may have determined the phenotypic variation.

Apart from the few inherited cases described above, the major part of Yq deletions are of de novo origin, which means that they are not present in the father’s DNA and probably originated in the germ line of the father (106). The spermatogenic stage at which the deletion occurs is not known: even if primary spermatocytes in meiotic prophase are probably the most sensitive cells, deletions may arise during other steps of DNA replication or even during spermiogenesis (106). The mechanism by which a Y deletion arises is not clear, since unlike autosomes, only limited parts of the Y chromosome pair with the X chromosome and no recombination occurs within the AZF regions. Therefore, it has been proposed that Y deletions are likely to be the consequence of the presence of many highly repeated DNA elements causing illegitimate intrachromosomal recombination. Specific genetic or environmental factors may predispose certain individuals to produce higher proportions of sperm with de novo deletions that may compete successfully with nondeleted spermatozoa to fertilize an egg and give rise to a Y-deleted child. This hypothesis could only be demonstrated by deletion analysis on single spermatozoa of the fathers. The other stage at which a Y deletion could originate is in the fertilized egg or during the first embryonic developmental phases of the infertile son (106). This should give rise to a mosaicism (normal Y chromosome in leukocytes, deleted Y chromosome in the germ line) and also this hypothesis should be analyzed in the future. Whatever is the case, the de novo origin of a Y deletion in an infertile patient is fundamental to assess its pathogenic role in determining the spermatogenic disruption.

The causative role of Y chromosome microdeletions in spermatogenic impairment is also supported by the evidence that male infertility caused by well known etiologies, such as Klinefelter’s syndrome, previous chemo-radiotherapy, orchitis, or testicular trauma, are rarely associated with Y deletions (63, 107). However, a recent study (83) reported a high prevalence of deletion also in azoospermia and severe oligozoospermia associated with other apparent causes (5/72, 6.9%). These data suggest that such severe testiculopathies may actually be due to Y chromosome deletions. In particular, sporadic observations showed that patients with cryptorchidism or varicocele and azoospermia or severe oligozoospermia could carry deletions (17, 73, 80, 88, 93). We reported a high incidence of deletion in unilateral ex-cryptorchid patients affected by azoospermia or severe oligozoospermia due to severe bilateral testicular damage (11/40, 27.5%), while we found no deletions in ex-cryptorchid men with normal function of the descended testis (108). Furthermore, it is possible that the severe testiculopathy caused by Y chromosome microdeletion may have rendered the testis unresponsive to the normal stimuli that regulate testicular descent. These results suggest that the bilateral testiculopathy observed in patients with unilateral cryptorchidism may be related to the deletion of the Y chromosome and not to the abnormal location of the testis. Similarly, seven of 40 (17.5%) patients affected by left varicocele and severe oligozoospermia sustained by severe hypospermatogenesis were found to have a Y deletion, while no abnormalities were detected in patients with varicocele and mild oligozoospermia (109). Also in this group of patients, the bilateral testicular damage is due to the underlying genetic anomaly and not to varicocele itself. The finding of Y chromosome microdeletions in such highly selected patient groups strongly suggests that the phenotype associated with a Y chromosome microdeletion may be also cryptorchidism and varicocele other than idiopathic infertility. All patients affected by severe testiculopathies should be screened for microdeletions, regardless of other concomitant causes of testicular damage (110).

From the clinical point of view, hormonal values and testicular volumes in Y-deleted patients indicate a severe testiculopathy involving only the spermatogenic system. In fact, the testes are generally reduced in size, FSH concentrations are high, and LH and testosterone plasma levels are within the normal range. The major part of studies reported no significant differences in testicular volumes and FSH levels between patients affected by severe testiculopathy with and without Y deletions. However, one study reported that FSH levels in patients with deletions, although higher than controls, were lower with respect to patients with similar tubular alterations but without deletions (80), and another study showed that Y-deleted patients had normal FSH plasma concentrations (87). If a difference in hormonal concentrations between deleted and nondeleted patients really exists, further studies are necessary to understand the mechanisms responsible for such differences.

As previously noted (111), there is no correlation between the frequency of microdeletions detected and the number of STSs analyzed. For example, if we consider only homogenous studies in which idiopathic azoospermic and oligozoospermic patients are reported (63, 64, 73, 80, 81, 82, 83, 87, 94, 95, 101), the number of STSs used varies from eight (80) to 85 (73), but the prevalence of deletions does not increase if more STSs are used (P = 0.9, not significant). Therefore, at the present time, the diagnostic performance is not improved by using large sets of STSs, and good results can be obtained using two to three STS markers for each AZF region, as recently suggested (74). What is really important for a careful diagnosis is that the panels of STS chosen amplify tracts of the Y chromosome containing genes that are known to be deleted specifically in infertile men and are not polymorphic. Furthermore, research in this field is growing and if a gene of the Y chromosome becomes a strong spermatogenesis candidate it should be included in the screening; in general, it is better to choose STS markers that amplify specific regions of a gene than those amplifying anonymous tracts of the Y chromosome.

C. The concern of assisted reproduction techniques
ICSI is the direct introduction of a spermatozoon into an oocyte to achieve fertilization and pregnancy when the number of spermatozoa in the ejaculate is very low or even absent. In the latter case, ICSI can be performed using spermatozoa obtained from the epididymis or directly extracted from testicular tissue. Furthermore, techniques of spermatid injection into oocytes may be performed and first term pregnancies have been achieved in humans (112, 113). Despite the world-wide diffusion of ICSI in recent years, the possible risks that might ensue from its indiscriminate use have been considered only recently. These concerns arose especially with the recent advances in genetically determined male infertility (114, 115). ICSI arouses more fears of the transmission of genetic abnormalities to offspring than other forms of assisted reproduction techniques because it bypasses all the physiological mechanisms related to fertilization, which need an active motile spermatozoon to undergo normal capacitation and acrosome reaction and to start all mechanisms required to penetrate the oocyte. By bypassing these steps, ICSI allows an altered spermatozoon to fertilize an oocyte, thus increasing the risk of genetic defects in the offspring. In other words, a genetic defect giving rise to abnormal spermatogenesis that can be surmounted by ICSI could be transmitted to the children produced. The concerns are most remarkable for male infertility related to Y chromosome microdeletions, since Y-deleted patients are strong candidates for ICSI, as in most cases spermatozoa or spermatids suitable for the procedure can be recovered from semen or the testis, but all male offspring will invariably inherit the deleted Y chromosome from the father.

Mulhall et al. (90) first reported the fertilization and pregnancy achieved utilizing ICSI with testicular spermatozoa from azoospermic patients presenting deletions in the DAZ region, suggesting that Y-deleted spermatozoa are fully competent for fertilization. Subsequently, the same group reported the birth via ICSI of male offspring from an AZFc-deleted man (116), and other authors reported similar findings (117, 118). Following Mendelian expectations, all the boys inherited their fathers’ deleted Y chromosome. These observations provide a concrete foundation for alerting couples to the likelihood of transmitting infertility-causing Y deletions by ICSI. Furthermore, since Y microdeletions are the most common molecularly defined causes of spermatogenic failure, one might expect that significant numbers of Y-deleted boys will be fathered through ICSI. From the analysis of the literature (Fig. 6), the prevalence of Y chromosome microdeletions in the ICSI candidate group is relatively low (3.8%), but it should be kept in mind that all severely oligozoospermic and azoospermic men with spermatozoa or spermatids in the testis represent the best candidates for this procedure, and the prevalence in these groups of patients is much higher. These findings provide a compelling rationale for the screening of all infertile men before ICSI.

Y-deleted spermatozoa may also transmit the deletion to male offspring via in vitro fertilization (IVF). In fact, it has been demonstrated that spermatozoa from an oligozoospermic subject carrying a Yq deletion are able to fertilize oocytes in vitro (119), suggesting that sperm carrying a deletion possess all the characteristics required to regulate capacitation, acrosome reaction, and the ability to penetrate and fertilize the oocyte. These findings are also supported by the evidence that Y deletion may be transmitted from father to sons also by spontaneous conception (16, 105, 106), even if very rarely (see above). In our opinion, Yq microdeletion analysis should be considered not only in patients undergoing ICSI, but also when undergoing other IVF programs.

The actual consequence of inheriting a Y deletion is still not clear, and we will have to wait until babies with the Y deletion conceived by IVF/ICSI become adult. Since descriptions of larger deletions in the sons with respect to the fathers have been reported (105), careful counseling of patients is mandatory, especially as to the risk of inheriting a deletion the clinical consequences of which are presently unknown, but, at the very least, include infertility. An identical deletion may be associated with different phenotypes, and therefore we cannot foresee the actual defects in the sons. Of greater concern is the possibility that additional unknown genetic problems may be present in infertile men whom nature has deemed unable to reproduce until now. Obviously, after extensive counseling, the couple must make the final decision about further treatment, but important ethical considerations arise since severe male infertility has now become a hereditary disorder. The clinician should be aware that the use of such techniques may lead to an increase in the number of infertile subjects in future populations. The primary rationale of medically assisted reproduction is the opportunity for parents to have a healthy baby, and a mandatory condition for achieving this goal is that the genetic characteristics of the gametes must be normal.

	VIII. Conclusions

Top Abstract I. Introduction II. Structure and Gene... III. The Azoospermia Factors IV. AZFb Region and... V. AZFc Region and... VI. AZFa Region and... VII. Y Chromosome Microdeletions... VIII. Conclusions References

 
The rapid growth of molecular biology has determined that microdeletions of the Y chromosome represent an important cause of male infertility and the most frequent genetic etiology of severe testiculopathy. Such findings are fundamental both for a careful diagnosis of male infertility and for its treatment, and Y chromosome screening is now a reality in the major andrological and infertility centers. The detection of a deletion in an infertile man provides a proper diagnosis of the disease, allows the clinician to avoid empirical, unnecessary, and often expensive treatments to improve fertility (e.g., hormonal treatments), and has important ethical consequences if the patient is a candidate for assisted reproduction techniques. Furthermore, it is now clear that a molecular diagnostic test of Y chromosome microdeletions should be at least performed in all men with a sperm concentration of less that 5 x 10⁶/ml, regardless of the presence of other apparent concomitant causes of testicular damage, such as varicocele or cryptorchidism.

The identification of the actual role played by the AZF-candidate genes in spermatogenesis will provide significant advances to our understanding of the biology of spermatogenesis, as well as the analysis of novel Y-chromosomal genes with a potential role in male germ cell development will clarify other important features of this important chromosome.

	Footnotes

 
Address reprint requests to: Carlo Foresta, M.D., University of Padova, Department of Medical and Surgical Sciences, Clinica Medica 3, Via Ospedale 105, 35128 Padova, Italy.

¹ This work was supported by Telethon Grant E.C0988 (Italy) and Ministry of University and Scientific and Technological Research (MURST) 1999.

	References

Top Abstract I. Introduction II. Structure and Gene... III. The Azoospermia Factors IV. AZFb Region and... V. AZFc Region and... VI. AZFa Region and... VII. Y Chromosome Microdeletions... VIII. Conclusions References

 

1. Sinclair AH, Berta P, Palmer MS, Hawkins JR, Griffiths BL, Smith MJ, Foster JW, Frischauf AM, Lovell-Badge R, Goodfellow PN 1990 A gene from the human sex-determining region encodes a protein with homology to a conserved DNA-binding motif. Nature 346:240–244[CrossRef][Medline]
2. Burgoyne PS 1998 The mammalian Y chromosome: a new perspective. Bioessays 20:363–366[CrossRef][Medline]
3. Tiepolo L, Zuffardi O 1976 Localization of factors controlling spermatogenesis in the nonfluorescent portion of the human Y chromosome long arm. Hum Genet 34:119–124[CrossRef][Medline]
4. Clermont Y 1966 Renewal of spermatogonia in man. Am J Anat 118:509–524[CrossRef][Medline]
5. Dym M 1994 Spermatogonial stem cells of the testis. Proc Natl Acad Sci USA 91:11287–11289[Free Full Text]
6. De Kretser DM, Baker HWG 1999 Infertility in men: recent advances and continuing controversies. J Clin Endocrinol Metab 84:3443–3450[Free Full Text]
7. Palermo G, Joris H, Devroey P, Van Steirteghem AC 1992 Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 340:17–18[CrossRef][Medline]
8. Vergnaud G, Page DC, Simmler MC, Brown L, Rouyer F, Noel B, Botstein D, de la Chapelle A, Weissenbach J 1986 A deletion map of the human Y chromosome based on DNA hybridization. Am J Hum Genet 38:109–124[Medline]
9. Vollrath D, Foote S, Hilton A, Brown LG, Beer-Romero P, Bogan JS, Page DC 1992 The human Y chromosome: a 43-interval map based on naturally occurring deletions. Science 258:52–59[Abstract/Free Full Text]
10. Jones MH, Khwaja OS, Briggs H, Lambson B, Davey PM, Chalmers J, Zhou CY, Walker EM, Zhang Y, Todd C, Ferguson-Smith MA, Affara NA 1994 A set of ninety-seven overlapping yeast artificial chromosome clones spanning the human Y chromosome euchromatin. Genomics 24:266–275[CrossRef][Medline]
11. Affara N, Bishop C, Brown W, Cooke H, Davey P, Ellis N, Graves JM, Jones M, Mitchell M, Rappold G, Tyler-Smith C, Yen P, Lau Y-FC 1996 Report of the Second International Workshop on Y-Chromosome Mapping 1995. Cytogenet Cell Genet 73:33–72[Medline]
12. Yen PH 1999 Advances in Y chromosome mapping. Curr Opin Obstet Gynecol 11:275–281[CrossRef][Medline]
13. Koopman P 1999 Sry a