Hossein Sadeghi-Nejada and Robert D. Oatesb
Purpose of review
Therapies for the treatment of severe male factor infertility have advanced well beyond our knowledge of the conditions we are treating. An intact Y chromosome is necessary for optimal spermatogenesis. It is imperative for the clinician to understand the molecular basis and clinical implications of anomalies that may afflict the Y chromosome.
The molecular geography of the Y chromosome has recently been described, allowing correlations to be made to various clinical states of severe spermatogenic compromise. Microdeletions of parts of the Y chromosome are found in a small number of men with spermatogenic failure involving, predominantly, three regions termed AZFa, AZFb, and AZFc.
It is necessary that a Y chromosomal microdeletion assay be carried out prior to any intervention using ejaculated sperm or prior to any surgical procedure to try and find spermatozoa in an azoospermic man.
AZFc, male infertility, Y chromosome
Long ago it was hypothesized that there may be ‘factors’ on the long arm of the Y (Yq) that were necessary for optimal spermatogenesis. Today, as the molecular anatomy of the Y chromosome has been clarified and refined, we recognize that, indeed, the regulation of spermatogenesis is at least partly controlled by multiple genes spread along proximal Yq. Early work in this area led to the belief that the azoospermia factor (AZF) region consisted of three spatially and topographically distinct areas (AZFa, AZFb, and AZFc) on Yq, each further away and distanced from the centromere. However, even better definition of this complicated stretch of the Y chromosome has revealed an unusual and unique architecture and allowed for the realization that while the AZFa region is truly separate and distinct, the AZFb and AZFc regions actually overlap one another and are simply different stretches of Yq within one much longer, encompassing expanse. Clinically, severe spermatogenic compromise or complete failure is manifest when AZFa, AZFb, or AZFc is singly, or in combination, deleted from the genome. The term used is ‘microdeletion’ as the mishap is not able to be visualized on karyotype but must be discerned through molecular biology techniques. In men with severe oligospermia or nonobstructive azoospermia (NOA), an assay to identify microdeletions must be carried out prior to any intervention so as to fully inform the couple of the basis of the sperm production compromise that assists them in formulating both near and long-term plans.
Molecular anatomy of the Y chromosome
Some of the many genes located in the proximal segment of the long arm of the Y chromosome play critical roles in the spermatogenic process. Therefore, a review of the molecular geography of the Y chromosome is necessary prior to describing the clinical ramifications of the known microdeletions that occur in this area.
The Y chromosome is 60 million base pairs (Mb) in length and contains heterochromatic and euchromatic regions in equal proportions . The much smaller pseudoautosomal regions (3Mb) are located at the ends of both the short (Yp) and the long (Yq) arms; these areas pair and recombine with similar regions on the X chromosome. The vast majority of the 60Mb is composed of the nonrecombining region, a male-specific area that occupies fully 95% of the chromosome, has an unusually repetitive sequence composition, and does not undergo sexual recombination . This unique, nonrecombining area between the terminal pseudoautosomal regions is termed the ‘male-specific Y’ (MSY) and is replete with a number of genes (i.e., DAZ, USP9Y, DDX3Y, RBMY1, and BPY2) involved in spermatogenesis that await full characterization.
MSY is made up of a combination of three classes of euchromatic (X-transposed, X-degenerate, and ampliconic) and heterochromatic sequences. Amplicons are stretches of nucleotides that are nearly identical, large repeats reading in the same (direct) or opposite (inverted) directions. They may be located immediately adjacent to each other with their ‘proximal’ ends ‘touching’ one another while their exact, duplicate sequences go in opposite directions (an inverted arrangement) or they may be spatially quite separate, located either upstream or downstream from one another. Most Y chromosome genes exclusively expressed in the testes are located in the ampliconic regions. These amplicons, however, are the building blocks of even more complex structures called palindromes .
Each palindrome has two arms that radiate outwards from a central core of a variable, but small number of nucleotides. Each arm is a mirror image of the other (hence the designation of these structures as ‘palindromes’). Each arm is constructed from either 1, 2, or more amplicons in sequential order. P8–P1 (closest to farthest from the centromere) represent the eight palindromic sequences within the euchromatic portion of Yq. The P5–P1 expanse is prone (statistically) to nonallelic homologous recombination (NAHR) due to this unusual molecular geography . As there is no counterpart in the genome for mitotic pairing and meiotic recombination of the MSY, this repetitive, interesting molecular architecture may have evolved to protect the long-term genetic integrity of the Y chromosome by allowing the MSY to pair with and repair itself. However, a consequence of this adaptive, innovative solution to having no molecular mate in the remaining genome is that, on rare occasions, NAHR may go awry when two spatially separate ampliconic regions ‘permanently stick’ together during Y chromosomal replication with resultant loss of all chromosomal material in the intervening portion. These deletions are not cytogenetically visible, hence the term ‘microdeletion’. NAHR is the proximate cause, therefore, of almost all Y chromosomal microdeletions. Costa et al.  have reported on rare microdeletions in this area of the Y resulting from nonhomologous recombination events and mechanisms.
The azoospermia factor regions of the Y chromosome
Microdeletions of the AZF regions of the Y chromosome are frequently observed genetic causes of NOA and severe oligospermia. As above, the molecular etiology of AZFa, AZFb, and AZFc microdeletions results from intrachromosomal recombination events between homologous repetitive sequence blocks in Yq11.
An AZFa microdeletion is responsible for failure of spermatogenesis in approximately 1% of men with NOA. The AZFa region is 792 kilobases (kb) in length, is located in proximal Yq, and contains two candidate genes, DDX3Y (16.3 kb, also known as DBY) and USP9Y that are believed play a crucial role in spermatogenesis. The exact role of DDX3Y has not been fully elucidated. It is most likely involved in the latter stages of spermatogenesis by coding for an ATP-dependent RNA helicase that shuttles between nucleus and cytoplasm. USP9Y encodes a protease involved in the regulation of protein metabolism. USP9Y probably plays a less critical role than DDX3Y as deletions of USP9Y only have been associated with a less pronounced effect on spermatogenesis. The AZFa region is not palindromic, as discussed above. Instead, NAHR between 10 kb retroviral elements (HERV15yq1 and HERV15yq2) flanking the two AZFa genes and thereby defining the endpoints of the AZFa region, results in complete loss of both DDX3Y and USP9Y.
Further along Yq, intrachromosomal ectopic homologous recombination (NAHR) in the P5–P1 interval can result in microdeletions of variable length with different proximal and distal endpoints. Depending on the length and the location of the microdeletion, various names (i.e., AZFb or AZFc) are used to designate the specific microdeletion. An AZFb microdeletion is 6.2Mb long that begins in the P5 palindrome and ends in the proximal portion of the P1 palindrome. The AZFb microdeletion may also be termed in molecular vernacular as the P5/ Proximal P1 microdeletion. The so-called AZFb/AZFc (P5/Distal P1) microdeletion also starts in P5, but spans a larger area (7.7 Mb) and ends in distal P1. The AZFc region stretches from the distal portion of the P3 palindrome to the distal portion of the P1 palindrome and is 3.5Mb in length. It was previously believed that AZFb and AZFc were completely separate, nonoverlapping areas along Yq that were independently necessary for normal sperm production. Subsequent exact definition of these intervals demonstrated that both the AZFb and AZFb/AZFc regions overlap the AZFc region and that genomic material lost during any of these three types of microdeletion is not separate and distinct from the other two. In other words, the stretch of Y chromosomal material from the P5 palindrome proximally to the P1 palindrome distally is replete with possible sites of NAHR leading to many possible microdeletions of variable length and position. Three are clinically significant and are still referred to by their original acronyms but could also be described by their molecular endpoints, 2 Andrology, sexual dysfunction and infertility as above: AZFb (P5/Proximal P1), AZFb/AZFc (P5/Distal P1), and AZFc (b2/b4).
AZFb (P5/Proximal P1) or AZFb/AZFc (P5/Distal P1) microdeletions are found in approximately 1–2% of men with NOA. RBMY was the first proposed candidate gene involved with spermatogenesis within the AZFb region. However, Ferlin et al.  suggested that other genes within this expanse may have important roles in spermatogenesis when they reported severe spermatogenic failure in men with partial AZFb microdeletions that had removed various genes within the AZFb area, but not RBMY.
AZFc microdeletions are the most frequently observed microdeletion in NOA and occur in up to 13% of men with azoospermia and 6% of men with severe oligospermia. There are 11 families of transcription units (including BPY2, CDY, DAZ, CSPG4LY, and GOLGAZLY) within the AZFc region that include three copies of BPY2, two copies of CDY1, and four copies of DAZ . AZFc is a complex of three palindromes beginning in distal P3 and extending into the P1 palindrome (Fig. 1). Six distinct families of amplicons make up the three palindromes in the AZFc area. They range in length from 115 to 678 kb. The 3.5Mb AZFc interval is defined and bounded by two amplicons, 229 kb in length and in direct orientation, that undergo NAHR, ultimately causing the loss (i.e., ‘microdeletion’) of the AZFc area including all four DAZ genes. DAZ encodes an RNA-binding protein exclusively expressed in early germ cells and felt to be responsible for activation of silent mRNAs, perhaps during premeiosis stages . Geoffroy-Siraudin et al.  studied meiosis in three infertile patients with complete AZFc microdeletions and concluded that AZFc is not critical for meiotic recombination, but that the absence of the AZFc region results is extension of the transient zygotene stage and reduction of chromosomal condensation. Although most AZFc microdeletions are canonical and 3.5Mb in length, intra-AZFc microdeletions of much shorter length can be predicted based on the ampliconic structure of the region and the mechanisms involved in microdeletion. Several have been reported and are rare such as b2/b3 (no known clinical relevance), b1/b3 (no known clinical relevance), and gr/ gr. The gr/gr microdeletion removes one of the DAZ gene pairs and its relationship to spermatogenic failure is still debated and unclear. The phenotypic outcome may be, in addition, based on the genetic background that the man comes from, which determines his Y chromosomal haplotype grouping [9,10].
Studies on the evolution of the DAZ gene cluster and the AZFc region [11,12] in particular are fascinating and shed light on why and how this incredibly complex and unusual molecular arrangement came to be. For example, the DAZ and CDY families made their way to the Y chromosome via transposition and retroposition of autosomal copies, respectively . As McElreavey et al.  point out, there is tremendous variability in the Y chromosome of different lineages and there should not be a single reference Y chromosome, as the discussion above may imply, but several based on Y chromosomal haplogroups with each being slightly structurally different. Finally, Vogt et al.  have recently revived the hypothesis that there are distinct chromatin regions in this area of the Y chromosome that are involved in premeiotic X–Y pairing – microdeletion could impact on this function as well – not a loss of genes but a loss of structure.
Clinical correlates of Y chromosome microdeletions
In infertile men with NOA or severe oligospermia, the results of a Y chromosomal microdeletion assay are critical for patients to know prior to embarking on a testicular sperm retrieval (TESE) and/or assisted reproduction journey. Information about the presence of and type of microdeletion that may exist helps in prognosis as regards the success of TESE and the transmission to offspring. Quite simply stated, complete AZFa, AZFb, and AZFb/ AZFc microdeletions are found in the NOA population and predict that TESE will be unsuccessful, sperm will not be found, and no treatment is presently available . Men with an AZFc microdeletion may have sperm in their ejaculate at very low levels (severe oligospermia), may have sperm found in the testis tissue, or may have no ongoing spermatogenesis at all (see below) . Testis biopsy histological correlates range from a complete absence of germ cells to severe hypospermatogensis. Men with Y chromosomal microdeletions rarely have sperm density greater than 5 million per milliliter. The vast majority of AZFc microdeletions are de novo (i.e., the father of the patient is not microdeleted), but rare cases of natural transmission have been reported .
Oates et al.  fully characterized 42 infertile men with AZFc microdeletions and reported on 18 children conceived through the use of intracytoplasmic sperm injection (ICSI). Among the 42 men evaluated in this study, 38% were found to be severely oligospermic and 62% were azoospermic, but during TESE and or testis biopsy, some degree of spermatogenesis was seen in 67% of this latter group and the results of ICSI were unaffected by the AZFc microdeletion. Sperm retrieval was impossible in 19% of the 42 men. Simoni et al.  reported a 60% rate of sperm harvesting upon TESE for the AZFc azoospermic group. Paternal age was not increased in the fathers of the patients compared with non-AZFc men with severe spermatogenic compromise. All men were somatically healthy without any other general systemic or testicular disorder. There were no factors in the patients’ history, physical examination, or hormonal/laboratory parameters that could predict sperm presence either in the ejaculate or testis (the NOA subgroup). When sperm were used for ICSI, function was adequate as fertilization, embryo development, and term pregnancy rates were excellent. The children were all somatically healthy but, of course, the reproductive potential of the sons could not be known. Prior suggestions had been posited in the literature that there may be a progressive and unrelenting decline in spermatogenesis over time. However, Oates et al. reported an overall stable spermatogenic potential noting fluctuation, but not a sustained decrease in sperm production. Successful assisted reproduction/ICSI can be achieved in men with AZFc microdeletions, but the male offspring will inherit the microdeletion and will be infertile or sterile when of reproductive age . When an AZFc microdeletion exists and sperm are available for use in combination with ICSI, the couple may choose to avoid use of the husband’s sperm altogether, use of his sperm for ICSI with transfer of embryos regardless of potential sex, or avoidance of AZFc microdeletion propagation by use of preimplantation genetic testing to allow for transfer of only female embryos.
The gr/gr intra-AZFc microdeletion has been studied extensively and it is still unclear whether it is a risk factor for infertility or sperm production compromise but it can clearly be passed from father to son in vertical fashion [21,22]. Some debate remains, as well, in regard to a gr/gr microdeletion and a causative role in cryptorchidism and testicular cancer, though the weight of the evidence suggests no biological connection [23,24]. Could a partial AZFc microdeletion, such as gr/gr, predispose this region to a full AZFc microdeletion and could either lead to an increased rate of sex chromosomal aneuploidy that may have further, more devastating consequences on the offspring are also areas of recent research .
In summary, the enigmatic and convoluted structure of the Y chromosome is coming into focus. It is incumbent upon those who treat male factor infertility to be well versed in the intricacies of the Y chromosome, the mechanisms of microdeletions, and the clinical consequences that arise when they are present. Education of our patients in regards to the genetic reasons for their sperm production deficiency allows them to make reasonable, well educated decisions for themselves in both the short term (their own particular situation of infertility) and the long term (infertility or sterility in a son who inherits his father’s AZFc microdeletion).
- Li Z, Haines CJ, Han Y. ‘Microdeletions’ of the human Y chromosome and their relationship with male infertility. J Genet Genomics 2008; 35:193–199.
- Skaletsky H, Kuroda-Kawaguchi T, Minx PJ, et al. The male-specific region of the human Y chromosome is a mosaic of discrete sequence classes. Nature 2003; 423:825–837.
- Kuroda-Kawaguchi T, Skaletsky H, Brown LG, et al. The AZFc region of the Y chromosome features massive palindromes and uniform recurrent deletions in infertile men. Nat Genet 2001; 29:279–286.
- Repping S, Skaletsky H, Lange J, et al. Recombination between palindromes P5 and P1 on the human Y chromosome causes massive deletions and spermatogenic failure. Am J Hum Genet 2002; 71:906–922.
- Costa P, Goncalves R, Ferras C, et al. Identification of new breakpoints in AZFb and AZFc. Mol Hum Reprod 2008; 14:251–258.
- Ferlin A, Moro E, Rossi A, et al. The human Y chromosome’s azoospermia factor b (AZFb) region: sequence, structure, and deletion analysis in infertile men. J Med Genet 2003; 40:18–24.
- Huang WJ, Lin YW, Hsiao KN, et al. Restricted expression of the human DAZ protein in premeiotic germ cells. Hum Reprod 2008; 23:1280–1289.
- Geoffroy-Siraudin C, Aknin-Seiffer I, Metzler-Guillemain C, et al. Meiotic abnormalities in patients bearing complete AZFc deletion of Y chromosome. Hum Reprod 2007; 22:1567–1572.
- Navarro-Costa P, Pereira L, Alves C, et al. Characterizing partial AZFc deletions of the Y chromosome with amplicon-specific sequence markers. BMC Genomics 2007; 8:342.
- Wu B, Lu NX, Xia YK, et al. A frequent Y chromosome b2/b3 subdeletion shows strong association with male infertility in Han-Chinese population. Hum Reprod 2007; 22:1107–1113.
- Tyler-Smith C. An evolutionary perspective on Y-chromosomal variation and male infertility. Int J Androl 2008; 31:376–382.
- Yu YH, Lin YW, Yu JF, et al. Evolution of the DAZ gene and the AZFc region on primate Y chromosomes. BMC Evol Biol 2008; 8:96.
- Bhowmick BK, Satta Y, Takahata N. The origin and evolution of human ampliconic gene families and ampliconic structure. Genome Res 2007; 17:441–450.
- McElreavey K, Chantot-Bastaraud S, Ravel C, et al. Y chromosome and male infertility: what is a normal Y chromosome? J Soc Biol 2008; 202:135–141.
- Vogt PH, Falcao CL, Hanstein R, Zimmer J. The AZF proteins. Int J Androl 2008; 31:383–394.
- Ferlin A, Arredi B, Speltra E, et al. Molecular and clinical characterization of Y chromosome microdeletions in infertile men: a 10-year experience in Italy. J Clin Endocrinol Metab 2007; 92:762–770.
- Kuhnert B, Gromoll J, Kostova E, et al. Case report: natural transmission of an AZFc Y-chromosomal microdeletion from father to his sons. Hum Reprod 2004; 19:886–888.
- Oates RD, Silber S, Brown LG, Page DC. Clinical characterization of 42 oligospermic or azoospermic men with microdeletion of the AZFc region of the Y chromosome, and of 18 children conceived via ICSI. Hum Reprod 2002; 17:2813–2824.
- Simoni M, Tuttelmann F, Gromoll J, Nieschlag E. Clinical consequences of microdeletions of the Y chromosome: the extended Munster experience. Reprod Biomed Online 2008; 16:289–303.
- Mau Kai C, Juul A, McElreavey K, et al. Sons conceived by assisted reproduction techniques inherit deletions in the azoospermia factor (AZF) region of the Y chromosome and the DAZ gene copy number. Hum Reprod 2008; 23:1669–1678.
- Minor A, Wong EC, Harmer K, Ma S. Molecular and cytogenetic investigation of Y chromosome deletions over three generations facilitated by intracytoplasmic sperm injection. Prenat Diagn 2007; 27:743–747.
- Stouffs K, Tournaye H, Van der Elst J, et al. Do we need to search for gr/gr deletions in infertile men in a clinical setting? Hum Reprod 2008; 23:1193– 1199.
- Giachini C, Nuti F, Marinari E, et al. Partial AZFc deletions in infertile men with cryptorchidism. Hum Reprod 2007; 22:2398–2403.
- Linger R, Dudakia D, Huddart R, et al. A physical analysis of the Y chromosome shows no additional deletions, other than Gr/Gr, associated with testicular germ cell tumour. Br J Cancer 2007; 96:357–361.
- Zhang F, Lu C, Li Z, et al. Partial deletions are associated with an increased risk of complete deletion in AZFc: a new insight into the role of partial AZFc deletions in male infertility. J Med Genet 2007; 44:437– 444.