Y chromosome and male infertility: update, 2006.

Abstract

Male factor infertility accounts for about half the cases of couple infertility and in around 50% of cases its etiology remains unknown. Molecular genetic techniques have unveiled a number of etiopathogenetic factors, including microdeletions of the Yq. Y chromosome microdeletions removing the AZoospermia Factor (AZF) regions are the most frequent molecular genetic causes of oligo/azoospermia. The intense effort of many laboratories contributed to a better understanding of the clinical significance of this genetic anomaly and to the identification of fertility candidate genes in the AZF regions. Important progress has been made on the structure of the Y chromosome and the mechanism of deletion. Studies aimed to define a predisposing genetic background for Yq deletions were not successful, perhaps due to the low number of patients analyzed so far. The screening for Yq deletions became a routine diagnostic test that provides an etiology for spermatogenic disturbances, and assess in the prognosis for testicular sperm retrieval according to the type of deletion. Assisted reproductive techniques represent an efficient symptomatic therapy for men bearing Y microdeletions, however, this genetic defect is transmitted to the male offsprings, affecting their fertility. Future studies should focus on understanding the biological function of AZF genes which is an essential step for the development of more appropriate and knowledge-based therapies.

Full text

Page 1

[Frontiers in Bioscience 11, 3049 -3061, September 1, 2006]
3049
Y chromosome and male infertility: Update, 2006

Csilla Krausz and Selene Degl’Innocenti

Department of Clinical Physiopathology-Center for research, Transfer and High Education, DENOthe, Andrology Unit,
University of Florence, Viale Pieraccini , 6, 50139 Florence, Italy

TABLE OF CONTENTS

1. Abstract
2. Introduction
3. Y chromosome structure
4. Yq deletions and spermatogenesis
5. Genes and gene families of the AZF regions
5.1. AZFa region
5.2. AZFb region
5.3. AZFc region
6. Mechanism of AZF deletions
7. AZF deletions and Y chromosome background
8. Clinical aspects of Yq deletions
8.1. Y deletions are specific for spermatogenic disturbances
8.2. Genotype/phenotype correlations
8.3. Incidence of AZF and AZF gene specific deletions. Indications for screening.
8.4. Yq deletions and genetic counselling
9. Conclusions and perspectives
10. Acknowledgements
11. References






1. ABSTRACT

Male factor infertility accounts for about half the
cases of couple infertility and in around 50% of cases its
etiology remains unknown. Molecular genetic techniques
have unveiled a number of etiopathogenetic factors,
including microdeletions of the Yq. Y chromosome
microdeletions removing the AZoospermia Factor (AZF)
regions are the most frequent molecular genetic causes of
oligo/azoospermia. The intense effort of many laboratories
contributed to a better understanding of the clinical
significance of this genetic anomaly and to the
identification of fertility candidate genes in the AZF
regions. Important progress has been made on the structure
of the Y chromosome and the mechanism of deletion.
Studies aimed to define a predisposing genetic background
for Yq deletions were not successful, perhaps due to the
low number of patients analyzed so far. The screening for
Yq deletions became a routine diagnostic test that provides
an etiology for spermatogenic disturbances, and assess in
the prognosis for testicular sperm retrieval according to the
type of deletion. Assisted reproductive techniques represent
an efficient symptomatic therapy for men bearing Y
microdeletions, however, this genetic defect is transmitted
to the male offsprings, affecting their fertility. Future
studies should focus on understanding the biological
function of AZF genes which is an essential step for the
development of more appropriate and knowledge-based
therapies.






2. INTRODUCTION

The first review on Y chromosome and male
infertility appeared in the Frontiers in Bioscience in 1999
(1), 3 years after the molecular definition of the three AZF
regions (2). At that time, there were already about 20
publications that described large scale screening for Yq
deletions in infertile men. The clinical definition of patients
and the markers used for the screening were heterogeneous
and consequently, many important clinical questions were
still unresolved. At that time, it was still unclear what the
real frequency of the AZF deletions was and if there were
any ethnic or geographic differences or any predisposing Y
background. Concerns were raised about the existence of a
genotype-phenotype correlation and about the specificity of
AZF deletions in spermatogenic failure.

However, despite controversies, we pointed out at
that time that the diagnosis of Yq deletions has an
important clinical value since it provides the etiology of
spermatogenic failure and enables one to avoid unnecessary
medical or surgical treatments. Moreover, since a
progressive decrease of sperm count over time was
observed in these patients, we proposed sperm
cryoconservation for both the patient and for his future son,
as a preventive therapy. In the same year, the first EAA
guidelines for Y chromosome deletion screening was also
published which has clearly indicated the need for using a
reliable set of markers for routine analysis (3). At the time

Page 2

Y chromosome and male infertility
3050


Figure 1. Schematic representation of the Y chromosome
showing the 7 deletion intervals. The three AZF regions
and the candidate AZF
PseudoAutosomial Region 1 and 2 (PAR1, PAR2); Yp:
short arm of the Y chromosome; Yq: long arm of the Y.

that the majority of the AZF genes were identified (4), we
were confident that insights into the function of the AZF
genes would have been achieved very soon.

By the beginning of 2005, many of the above
mentioned clinical questions were answered and thanks to
the EAA guidelines and EAA/EMQN external quality
control programme (5, 6) Yq testing has become more
homogeneous and reliable in different laboratories. The
complete Y sequence and gene /transcript content was
published and the mechanism by which AZF deletions take
place was fully understood (7). Recently, new deletion
types with yet uncertain clinical meaning have also been
reported opening a new area of research in this field (8, 9
and references therein).

Unfortunately, our knowledge concerning the
biological function of the AZF genes has only slowly
progressed during the last 6 years and the time for a
knowledge-based aetiological therapy seems to be still very
far away. An updated analysis of the available data on the
Y chromosome and male infertility including both clinical
and basic research is the subject of the present review.

3. Y CHROMOSOME STRUCTURE

The Y chromosome comprises only 2-3% of the
haploid genome. It is an acrocentric chromosome and
consequently contains a short arm (Yp) and long arm (Yq),
genes are indicated.
demarcated by a centromeric region essential for
chromosome segregation (Figure 1). The human Y
chromosome is classically divided into two functionally
distinct regions: i) the pseudoautosomal regions (PAR1 and
PAR2), which are homologous with X chromosome
sequences and are responsible for correct pairing between
the two sex chromosomes during male meiosis; ii) the male
specific region Y (MSY), previously called the “Non-
recombining region Y” (NRY), in which, in certain parts,
instead of classical recombination “intrachromosomal gene
conversion” (non-reciprocal transfer) takes place (10).
Thanks to the presence of this conversion–based system of
gene copy “correction”, a certain number of Y genes have
been preserved from the gradual accumulation of
deleterious mutations ensuring their continuity in time
(7,10). This region comprises 95% of the length of the
chromosome.

The extent of the MSY region is roughly 63Mb,
but only 23Mb are transcriptionally active (euchromatic
portion). The heterochromatic region comprises distal Yq.
This region is polymorphic (variation in length) within
males, constituting almost half the chromosome in some
men while being undetectable in some others. This part of
the chromosome is assumed to be genetically inert and it is
mainly composed of two highly repetitive sequences
families, DYZ1 and DYZ2, containing about 5000 and
2000 copies of each respectively.

The euchromatic region contains three classes of
sequences, named by Page and collegues X-transposed, X-
degenerate and ampliconic (7). The X transposed sequences
are 99% identical to DNA sequences in Xq21 and derive
from a massive X to Y transposition that occurred about 3-
4 million years ago. These sequences contain the highest
density of repeats (interspersed repeat elements) and only
two genes with homologues in the Xp21. The X-degenerate
segments are relatively rich of single copy gene or
pseudogene homologues of X linked genes. The third class
of sequences, ampliconic segments, are composed by long
MSY specific repeat units named “amplicons” which show
as much as 99,9% of intrachromosomal identity. These
segments exhibit the highest density of genes, both coding
and non-coding, among the three sequence classes.

The MSY euchromatin contains a total of 156
transcriptions units with 78 protein coding units and 78
putative non coding units (7). The protein coding units are
the product of 27 genes, 12 of which are expressed
ubiquitously and 11 are exclusively, or predominantly,
expressed in testes. All 12 ubiquitously expressed genes of
the MSY reside in the so called X-degenerate sequence
whereas the majority of genes with testis specific
expression are located in the ampliconic segments.

The majority of the MSY genes are involved in
male specific functions, such as male sex determination
(SRY) and spermatogenesis (i.e. genes of the AZoospermia
Factor regions of the long arm of the Y chromosome).
Consequently mutations/deletions of these genes lead to
sex reversal or spermatogenic failure (1, 12). In addition,
loss or rearrangements of the Y are also associated to a

Page 3

Y chromosome and male infertility
3051
Figure 2. Schematic representation of the mechanism of AZF deletions. Boxes with motifs represent repeated homologous
sequences situated at the border of recurrently deleted regions on the Yq. AZFa deletions occur between two homologous
retroviral sequences (HERV15q). AZFb* deletions partially overlap with the AZFc region and may take place between different
palindromic sequences (P5, P4, P1). The classical AZFc deletion occurs between two repeated sequences b2/b4 entirely situated
in the AZFc region. The extent of the deleted segment is indicated in each type of AZF deletion (figure not in scale).

number of other phenotypes, such as Turner stigmata (13),
skeletal anomalies (14) and the development of a rare type
of gonadal cancer (gonadoblastoma) (15). The MSY also
contains genes with housekeeping cellular activities (7,16)
and are probably involved in functions other than male
reproduction. The same genes may be responsible of the
sexual dimorphism observed for certain pathologies such as
hypertension, autism etc. (for review see 17 and references
therein).

4. Yq DELETIONS AND SPERMATOGENESIS

The first association between azoospermia
(absence of spermatozoa in the ejaculate) and
microscopically detectable deletions of the long arm of the
Y chromosome has been demonstrated by Tiepolo and
Zuffardi in 1976 (18). Since in 4 patients the deletion was
de novo (their fathers were tested and found to carry intact
Y chromosome) they proposed the existence of a
spermatogenesis factor, the AZoospermia factor (AZF)
encoded by a gene on distal Yq. Successively, the
development of molecular genetic tools, firstly Y specific
DNA probes (19) than sequence tagged sites (STS: short
tracts of DNA that acts as a landmark to define position on
a physical map) (20), permitted to narrow the extension of
the AZF region. Using the above mentioned molecular
tools the breakpoints of a large panel of infertile men
displaying a cytogenetically visible abnormality in Yq lead
to the construction of molecular interval maps. Many STS
based screening have been undertaken in patients affected
by azoo/severe oligozoospermia in order to define the AZF
locus and isolate candidate genes for AZF. Vogt et al. (1)
observed that Yq microdeletions follow a certain deletion
pattern, with three recurrently deleted nonoverlapping
subregions in proximal, middle and distal Yq11, designated
AZFa, AZFb and AZFc, respectively. Only several years
after the discovery of the three AZF regions, with the
precise knowledge of the Y structure in Yq11, is became
evident that the AZFb and AZFc regions are overlapping
(10). New deletion patterns in the AZFb/AZFc regions
have been proposed, however since data in the literature are
based on markers which are unable to distinguish between
these subtypes, it is more convenient to refer to the three
“classical” deletion intervals. (Figure 1)

5. GENES AND GENE FAMILIES OF THE AZF
REGIONS

5.1. AZFa region
AZFa region is located in proximal Yq within
deletion interval 5. In 1999, it was estimated that the region
is between 1 and 3 Mb in size and contains four single copy
genes, namely USP9Y (former DFFRY), DBY, UTY and
TB4Y. After the discovery of the mechanism of the AZFa
deletion formation (see below) it became evident that its
size is of 0,8 Mb (Figure 2) and contains only two single
copy genes: USP9Y and DBY (21,22). The USP9Y protein
is an ubiquitinyl hydrolase and belongs to the family C19
(cystein peptidases) which are intracellular peptidases that
remove ubiquitin molecules from polyubiquitinated
peptides. It may therefore play an important role at the
level of protein turnover by preventing degradation of
proteins through the removal of conjugated ubiquitin.

The DBY gene is predicted to encode an RNA
helicase and thus is involved in RNA metabolism. Both
genes are transcribed in multiple tissues and have a
homologue on the X chromosome which escapes X
inactivation. Protein expression analysis for the DBY gene
allowed the identification of a translational control

Page 4

Y chromosome and male infertility
3052
mechanism. Vogt and coworkers reported (23) that despite
the ubiquitous transcription of the DBY gene, its mRNA is
translated specifically in the testis tissue and in particular in
spermatogonia and pachytene spermatocytes.

5.2. AZFb region
Similar to the AZFa region, the discovery of the
deletion mechanisms in concert with the fine definition of
the Yq structure led to a fundamental change of our view
concerning the extension of the AZFb region. It became
clear that the classical AZFb region is partially overlapping
with the AZFc region. Previously its estimated size was
around 1-3 Mb, now we know that there are two possible
recombination sites and the extent may vary from 6,2 to
7,7Mb (10) (Figure 2).

It contains genes which are present only in the
AZFb region (HSFY, eIF-1Y, SMCY) and others which
are shared with the AZFc region (one copy of BPY2 and
CDY, two copies of DAZ). Moreover, two gene families
with multiple copies on the Y have their active copies in
the AZFb region: RBMY and PRY. More than 30 RBM
genes and pseudogenes are spread over both arms of the
Y chromosome (24-26). The AZFb region contains at
least one functional RBM copy, located in the distal
portion of this deletion interval (27) RBMY (RNA
Binding Motif Y) is a testis specific splicing factor and
it is homologous to RBMX on the X chromosome
(hnRNPG) and to a functional retrogene on chromosome
11 (11p15) hnRNPG-T. Both the X and the Y derived
protein are highly expressed in mitotically active cells
and thus may promote
spermatogonia (28, for review see 29,30 and references
therein). Beside spermatogonia, the RBMY protein is
expressed also in the nuclei of spermatocytes and round
spermatids.

Two functional copies of the PRY gene are
present in the AZFb region and encodes a protein
phosphatase which is probably involved in apoptosis of
defective spermatozoa (31). The expression pattern of this
protein (in spermatids and spermatozoa) indicates a role in
the post-meiotic phases of spermatogenesis.

The HSFY (Heat Shock Factor Y) gene is
situated in the palindromic sequence P4 in proximal
AZFb and there are two copies. This gene encodes a
novel heat shock protein which is specifically expressed
in Sertoli and germ cells which exhibits a stage-
dependent translocation from the cytoplasm to the
nucleus in spermatogenic cells during spermatogenesis
(32). The gene has a homologue HSP2 on 6p22 which
codes for a heat shock transcription factor. The eIF-1AY
encodes a Y isoform of a eIF-1A, an essential
translation initiation factor. eIF-1AY has been mapped
to interval 5q and subsequently defined between sY127-
sY129 (33). This gene has an X homologue (eIF-1AX)
which escapes inactivation. Both X and Y homologoues
are ubiquitously expressed, with high level of
expression in the testis (4). The biological function of
SMCY is unknown but it encodes a male specific
histocompatibility antigen (H-Y).
mitotic divisions in
5.3. AZFc region
The region is 3,5 Mb and in analogy to the other
two AZF regions it is delimited by two repeated sequences
(b2 and b4) (Figure 2).

The first gene identified in the AZFc region is
the DAZ gene (34). It is a multicopy gene with two copies
situated in P2 (DAZ1/DAZ2) and two located 1.47 Mb
distant in the large P1 palindrome (DAZ3/DAZ4). The
DAZ gene has been transposed and pruned (amplified)
from chromosome 3 (3p25; DAZL1 locus) after divergence
of the Old world monkey line of primates (35, 36). The
structure of the 4 copies is variable especially in the
number of exon 7 variants which is also known as the
“DAZ-repeat” (37, 38). The DAZ gene family includes two
autosomal homologues, DAZL on chromosome 3 and
BOULE on chromosome 2 (39). The finding that the
human DAZ transgene is able to partially rescue the
spermatogenic failure of mice homozygous for a null allele
of the Dazl gene (40) suggests a possible interplay between
DAZL and DAZ in humans.

All DAZ gene family members are encoding
testis specific RNA binding proteins and thus are probably
involved in the translational control of transcripts with
spermatogenic function (for review see 41). Studies in flies
demonstrate that the BOULE gene encodes a key factor of
meiosis in male germ cells, regulating the expression of
twine, a cdc25 phosphatase, which promotes progression
through meiosis (42). In analogy, a translational control of
CDC25 transcripts by BOULE has been reported also in
human (43) and human BOULE transgene is able to rescue
the meiotic block in boule mutant flies (44). DAZL and
DAZ both interact with the DAZAP1 protein (Deleted in
Azoospermia Associated Protein 1) which shuttles between
the nucleus and the cytoplasm and may play a role in
mRNA transport and/or localization (45). Expression
analysis of the DAZ protein gave discordant results
probably due to the use of different polyclonal DAZ-
antisera. In one study, DAZ mRNA was detected in early
spermatids whereas the protein was observed in late
spermatids and spermatozoa. (46). In an other study DAZ
expression has been found in multiple cell compartments at
multiple points in male germ cell development (47). In this
study, human DAZ and human and mouse DAZL proteins
were present in both the nuclei and cytoplasm of foetal
gonocytes and in spermatogonial nuclei. Interestingly, the
proteins relocate to the cytoplasm during male meiosis.
Human DAZL protein (but not DAZ) persists in spermatids
and even spermatozoa. These results would suggest that
DAZ proteins may act during meiosis and much earlier,
when spermatogonial stem cell populations are established.
The observed protein-protein interaction between DAZL
and Pumilio-2 (PUM2) also suggest an important role in
fetal germ cell development (48).

The CDY gene family has a similar origin like
the DAZ gene i.e. it was retrotransposed from the CDYL
locus on chromosome 6 (4). While the autosomal
homologue is expressed ubiquitously, the CDY gene is
specifically expressed in the testis. Four alternative spliced
transcripts encoding three different protein isoforms have

Page 5

Y chromosome and male infertility
3053
been described (49). CDY are nuclear proteins observed in
spermatids and show histone acetyl transferase activity
with a strong preference for H4 (50). The Basic Protein Y 2
(BPY2) gene is expressed exclusively in the testis. The
BPY2 protein represent singleton without detectable
sequence homologues. The analysis of the 3D protein
structure predicted a possible role of this highly charged
protein in DNA or RNA binding (51).

A part from protein coding genes the AZF
regions contain two gene families with unknown protein
coding potential: i) the three members of the CYorf family
(CYorf14, CYorf15A, CYorf15B) which are all expressed
in multiple tissues and have homologues on the X (cXorf
14, cXorf15, cXorf15). ii) The TTY (Testis Specific
Transcription Y) family which are considered transcription
units with no significant open reading frame. Members of
this family are present also outside of the AZF regions on
both the short and the long arm of the Y.

For the moment it is not known whether all genes
situated in the AZF regions are important for
spermatogenesis. Since deletions occur in block- removing
more than one gene- the role of a single AZF gene cannot
be extrapolated from the deletion phenotype. To date only
one confirmed gene specific deletion has been reported in
the literature in which the breakpoint of the mutation has
been defined (52). The associated phenotype i.e.
hypospermatogenesis indicates that this gene, USP9Y, is
probably not essential for the initiation and completion of
spermatogenesis and it has probably a role in enhancing
the quality and efficiency of spermatogenesis. Other
authors have also reported single gene deletions (DBY,
USP9Y, HSF-Y) however these studies did not define the
deletion breakpoints (53,54). Since polymorphisms,
rearrangements may lead to false deletions these data await
further confirmation.

6. MECHANISM OF AZF DELETIONS

The mechanism by which classical AZF deletions
take place is through intrachromosomal recombination
between homologous sequences (7, 55). The long arm is
indeed composed of numerous Y specific repetitive
sequence blocks (called “amplicons”). These amplicons are
assembled in eight palindrome structures (P1-P8) and are
highly symmetrical. The homologous amplicons mapped in
the different palindrome arms revealed extensive
homologies between 99.940 and 99.997%. The AZFb and
AZFc deletions occur between palindromes whereas the
AZFa deletion is the result of intrachromosomal
recombination events between two repetitive specific
HERV15 (Human Endogenous Retroviral #15) sequence
blocks (22, 23). (Figure 2)

7. AZF DELETIONS AND Y CHROMOSOME
BACKGROUND

Polymorphic markers on the MSY region define
Y chromosome lineages (or haplogroups) which are
monophyletic groups of Y chromosomes sharing allelic
states at slowly mutating binary markers. Until now, more
than 200 Y specific single nucleotide polymorphisms
(SNPs) has been identified providing, together with a
number of indels (insertions/deletions), a detailed and
robust phylogenetic tree of Y chromosomal lineages
(56,57). The absence of recombination on the MSY region
makes that polymorphisms located in this region are in
tight association with potential functional variations
associated with Y-linked phenotypes.

A predisposing Y background to AZF deletions
has been hypothesized on the basis of a previous study
reporting a Y haplogroup that is susceptible to aberrant
X/Y exchange during male meiosis, leading to the XX male
syndrome (58). Since the mechanism by which AZF
microdeletions originate can be traced to intrachromosomal
recombination between repetitive elements (22, 23, 55),
sequence differences or different orientation of the repeated
blocks may predispose or protect against specific type of
deletions. An example is the polymorphic deletion of
L1PA4 element which confers a higher homology between
the two retroviral blocks in the flanking sequences of the
AZFa region and thus may, in theory, facilitate homologous
recombination. This deletion is considered as a
polymorphism and reach high frequencies in populations
from the Middle East. Similarly to AZFa region, there
could be differing susceptibilities to the other AZF
deletions depending on the Y chromosome structure. Men
carrying Y chromosome microdeletions from Ireland,
Scotland, Germany, Italy, Spain, Holland and Denmark
were haplotyped and the haplotype distribution was then
compared to infertile men without microdeletions (59) or
compared to an unselected male control population (60). Y
chromosome microdeletions were found to occur on
different Y chromosome backgrounds and there was no
significant difference between Y chromosome haplogroup
distribution in the study and control groups. However, due
to the relative rarity of AZFa deletions (5% of the total
deletions), the number of this type of deletions included in
both studies was small and, despite the apparent lack of
association, the relationship
background and microdeletion formation cannot be ruled out.

Different Y chromosome backgrounds may also
be responsible for the inter-individual variation in the
phenotypic expression of a given AZF deletion such as the
complete AZFc deletion which can be associated with both
severe oligospermia or with the total absence of
spermatogenic cells in the testis. Future clinical-molecular
combined studies are needed to accredit this attractive
hypothesis.

A part from AZF deletions other Y related factors
such as partial gene copy deletions of multicopy genes or
rearrangements/inversions occurring in non coding
sequences but with possible functional effects on gene
expression may influence spermatogenesis. These type of
alterations may segregate with certain Y backgrounds,
therefore a role for Y chromosome background in
determining spermatogenic potential has also been
proposed. This hypothesis was confirmed in the Danish and
Japanese population whereas not in others (for review see
17 and references therein).
between Y-chromosome