Human Reproduction Vol.21, No.12 pp. 3171–3177, 2006
doi:10.1093/humrep/del281
Advance Access publication July 22, 2006.
Synaptic defects at meiosis I and non-obstructive
azoospermia
Daniel Topping1,6, Petrice Brown1,2, LuAnn Judis2, Stuart Schwartz3, Allen Seftel4,
Anthony Thomas5 and Terry Hassold1
1
School of Molecular Biosciences, Washington State University, Pullman, WA, 2Department of Genetics and The Center for Human
Genetics, Case Western Reserve University and University Hospitals of Cleveland, Cleveland, OH, 3Department of Genetics, University
of Chicago, Chicago, IL, 4Department of Urology, Case Western Reserve University and University Hospitals of Cleveland, Cleveland
VA Medical Center and 5Glickman Urological Institute, Cleveland Clinic Foundation, Cleveland, OH, USA
6
To whom correspondence should be addressed at: School of Molecular Biosciences, Washington State University, 540 Fulmer Hall,
Pullman, WA 99164, USA. E-mail: [email protected]
BACKGROUND: Recent advances in immunofluorescence methodology have made it possible to directly monitor
protein localization patterns in germ cells undergoing meiosis. We used this technology to examine the early stages of
meiosis in testicular material obtained from men presenting for evaluation at infertility clinics. METHODS: Specifically, we compared meiotic progression, synapsis and recombination in 34 individuals with obstructive azoospermia
(‘controls’) to 26 individuals with non-obstructive azoospermia (NOA) (‘cases’). RESULTS: In 9 of the 26 cases, no
germ cells were identified, but in the remaining 17, there was at least some progression through meiosis. Most of these
individuals appeared to have normal levels of spermatogenic activity, with little evidence of meiotic impairment.
However, in three individuals, we observed either complete or partial meiotic arrest associated with abnormalities in
synapsis. CONCLUSIONS: This suggests that >10% of cases of unexplained NOA may be attributable to severe meiotic defects. The characterization of these meiotic arrest phenotypes may guide further research into the molecular
basis of unexplained infertility.
Key words: meiotic arrest/MLH1/non-obstructive azoospermia/recombination/SCP3
Introduction
A number of genetic abnormalities that contribute to human
male infertility have now been identified, including constitutional chromosome abnormalities involving either aneuploidy
or structural rearrangements (Chandley, 1988; Van Assche
et al., 1996), microdeletions of Yq-containing spermatogenesis
genes (Krausz et al., 2003) and cystic fibrosis transmembrane
regulator (CFTR) mutations that result in congenital absence of
the vas deferens (Claustres et al., 2000). However, most cases
of non-obstructive azoospermia (NOA) remain unexplained,
and we and others (Martin et al., 2003; Egozcue et al., 2005;
Sun et al., 2005a,b) have suggested that many of these are
attributable to mutations in genes controlling meiotic synapsis
or recombination. There are at least two lines of evidence to
support this supposition. First, null mutations have been generated for several mouse meiotic genes, including those involved
in the initiation or processing of double-strand breaks into
functional chiasmata (e.g. Spo11, Dmc1, Pms2 and Mlh1)
(Cooke and Saunders, 2002), formation of the synaptonemal
complex (SC) (e.g. Scp3) (Cooke and Saunders, 2002) or
establishment of sister chromatid cohesion (e.g. Rec8 and
Smc1β) (Revenkova et al., 2004; Xu et al., 2005). The meiotic
phenotypes of the different null mutant males vary subtly, but
meiotic disruption and/or arrest and infertility are invariant
features (Hunt and Hassold, 2002). Although similar meiotic
mutations have only rarely been described in humans, presumably, this reflects the fact that null mutants are infrequent and
appropriate tests to detect them have not yet been conducted.
Second, previous cytogenetic studies of meiosis in infertile
males suggest alterations in recombination by comparison with
normal males. For example, in analyses of diakinesis-stage preparations, both low chiasma frequency (Micic et al., 1982) and
alterations in synapsis (Vendrell et al., 1999) have been linked to
male infertility. Also, cases of asynaptic meioses in association
with consanguineous matings have been reported (Chaganti
et al., 1980), suggesting an autosomal recessive defect in recombination as the cause of some cases of male infertility. Taken
together, these two sets of observations suggest that additional
analyses focused on human male meiosis will uncover previously unrecognized genetic causes of male infertility.
Application of recently developed immunofluorescence
methodology should aid in this search. That is, by immunostaining material from testicular biopsies, we can now monitor
the expression of specific recombination machinery proteins
© The Author 2006. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: [email protected]
3171
D.Topping et al.
and examine the formation of meiosis-specific structures in
meiosis I spermatocytes. Preliminary studies that have used
this approach to analyse meiosis in infertile males have been
promising. Specifically, different groups have now identified
either meiotic arrest phenotypes or significant reductions in
meiotic recombination in a proportion of males with NOA
(Gonsalves et al., 2004; Judis et al., 2004; Sun et al., 2004,
2005a,b); presumably, mutational analyses of loci involved in
synapsis or recombination will uncover the molecular basis for
infertility in at least some of these individuals.
As part of ongoing studies of human meiosis, our laboratory
has been interested in applying immunostaining technology to
the study of human spermatogenesis. For the most part, we have
focused on the ‘normal’ situation, examining meiotic progression, synapsis and recombination in azoospermic individuals
whose infertility was unrelated to meiotic errors (e.g. cases of
obstructive azoospermia). These analyses have provided initial
data on the number and distribution of crossovers in the human
male (Hassold et al., 2004), the relationship between the length
of the SC and the number of exchanges (Lynn et al., 2002) and
the number and location of synaptic initiation sites on human
chromosomes (Brown et al., 2005). However, these observations
have also provided important baseline data for investigations of
idiopathic infertility. That is, by comparing results from these
‘control’ males to males with NOA of unknown origin, it should
be possible to examine the contribution of meiotic errors to the
genesis of male infertility. In the present report, we summarize
initial observations using this approach, comparing meiotic progression and recombination rates in a series of control males to
males with NOA or severe oligozoospermia. Our results indicate
that, in the vast majority of cases of NOA, meiosis proceeds normally, at least until the first meiotic division. However, in >10%
of cases, we observed partial-to-complete meiotic arrest during
zygotene or pachytene, providing evidence that meiotic errors
are an important contributor to male infertility.
Materials and methods
Study population
Participants were solicited from individuals attending the Urology
Department of The University Hospitals of Cleveland or the Glickman
Urological Institute of the Cleveland Clinic Foundation for treatment
of infertility. In each instance, informed consent was obtained according to protocols approved by the Institutional Review Boards of the
participating institutions. Controls (n = 34) were defined as individuals who were diagnosed with known causes of infertility not associated with obvious meiotic defects; that is, previous vasectomy (n = 17),
CF or CFTR variants (n = 9), unilateral varicocele (n = 3), testicular
tumour (n = 2) and others (n = 3; childhood injury, surgical scarring
and unspecified obstructive azoospermia, one case each). In all
instances, histological examination revealed active spermatogenesis
in most, if not all, seminiferous tubules. Routine karyotypic analyses
were conducted on 23 of the controls, and sequence tagged site (STS)based assays for the detection of Yq microdeletions were conducted
on 19 of the controls; no abnormalities were detected on either assay.
Results of studies of meiotic recombination on 24 of these individuals
have been presented previously (Hassold et al., 2004).
Cases (n = 26) were defined as individuals with NOA of unknown
aetiology. Nine of the 26 had no evidence of germ cells on histological
3172
examination or immunostaining analysis and were not further processed. Of these, two had non-mosaic 47,XXY chromosome constitutions (Klinefelter syndrome), one had a Yq microdeletion and the
other six were simply described as having Sertoli cell only syndrome.
The remaining 17 cases displayed germ cells that progressed at least
partway through meiosis I, and these individuals form the basis for the
present report. One of the 17 was a mosaic for a ring Y chromosome
[46,X,r(Y)(?p11.1?q11.2).ishY(DYZ1+,DYZ3,DXYS129/S153+)[24]/
45,X[8]; however, no other karyotypic abnormalities or Yq microdeletions were identified in these individuals.
SC preparations and immunostaining
The methods utilized for preparing the surface-spread SC slides and
the immunostaining have been described elsewhere (Hassold et al.,
2004) but with the following modifications. Following maceration,
the testicular tissue was incubated in hypotonic solution for 45 min,
and the immunostaining was performed no longer than 20 h after slide
preparation. The primary antibodies were rabbit polyclonal antibody
against human MLH1 (Calbiochem, San Diego, CA, USA; #PC 56T)
and goat antibody raised against rat SCP3 (kindly provided by
Dr Terry Ashley); CREST antiserum was kindly provided by Dr Sue
Varmuza. The secondary antibodies used were fluorescein-conjugated
donkey anti-rabbit, rhodamine-conjugated donkey anti-goat and aminomethylcoumarin-conjugated anti-human (Jackson ImmunoResearch;
West Grove, PA, USA). Surface-spread SC preparations were incubated with primary antibodies overnight in a 37°C humid chamber, in
antibody dilution buffer (ADB) (10× stock consisted of 10 ml of normal donkey serum (Jackson ImmunoResearch), 3% bovine serum
albumin (BSA) [Sigma, St. Louis, MO, USA), 50 ml of Triton X-100,
brought to a final volume of 100 ml with phosphate-buffered saline
(PBS)], MLH1 diluted 1:75, SCP3 1:50 and CREST anti-sera diluted
1:1000. Slides were washed overnight at 4°C in ADB after an initial
10-min wash in ADB at room temperature. Secondary antibodies were
diluted 1:100 and incubated on slides for 90 min at 37°C in ADB.
Slides were then washed in PBS for 10 min and treated with Antifade
(Bio-Rad; Hercules, CA, USA). Slides were stored in the dark at 4°C.
Fluorescence microscopy, imaging and analysis
The number and location of MLH1 foci in pachytene spermatocytes
are known to agree well with results of genetic linkage analyses and
analyses of chiasmata in diakinesis-stage spermatocytes and thus
appear to reliably localize to sites of exchanges in male meiosis (for
review, see Lynn et al., 2004). Thus, to assess the number of crossovers per cell, we counted the number of MLH1 foci in pachytene spermatocytes. To avoid scoring cells with a partial complement of MLH1
foci (i.e. either because they were just entering or exiting pachytene or
because the cells had synaptic defects), we restricted our analyses to
normally synapsed cells in which at least the haploid number of
MLH1 foci (i.e. 23) was identified (Lynn et al., 2002). For each individual, we attempted to score at least 50 cells.
Immunostained slides were evaluated on a Zeiss Axiophot epifluorescence microscope and imaged with a CCD camera and computer using
Quips Pathvysion smartCapture VP 1.4 software (Digital Scientific).
Results
Meiosis proceeds normally in most individuals with NOA
We analysed meiotic progression and recombination rates in
the 34 controls and the 17 individuals with NOA (‘cases’)
whose biopsies contained at least some germ cells. Two of the
cases exhibited a complete meiotic arrest and one a partial meiotic arrest, and these are discussed separately below. However,
Meiosis in non-obstructive azoospermia
300
250
Number of Cells
200
Controls
Cases
150
100
50
69
67
65
63
61
59
57
55
53
51
49
47
45
43
41
39
37
35
33
31
29
27
0
25
all 34 of the controls and the remaining 14 cases exhibited normal progression through meiotic prophase, with mature spermatozoa being present in all testicular samples. Furthermore,
pachytene-stage cells predominated in all analysed controls
and cases, with little evidence for variation between the two
groups.
We analysed recombination in cases and controls by scoring
the number and location of MLH1 foci in pachytene spermatocytes (see Materials and methods). Table I provides a summary of observations on the cases, and Figure 1 shows the
distributions of MLH1 foci per cell in controls and cases. For
the 34 controls, we analysed a total of 2932 cells; the overall
mean number of MLH1 foci per cell was 49.2 ± 4.9, with a
range of 30–66 foci per cell. For cases, we analysed 1087 cells
from the 14 individuals, with an overall mean of 49.0 ± 5.0
MLH1 foci per cell and a range of 27–67 foci per cell.
We compared recombination levels between the two groups
in several ways. First, we simply pooled all data from each of
the two groups and compared the overall means using Student’s t-test methodology; the difference between the groups
was non-significant (P = 0.26). Second, to account for the fact
that just two individuals (i.e. Sp380 and Sp1001) accounted for
∼40% of all ‘case’ group cells, we conducted a second t-test in
which the means of each of the 14 cases were compared with
the 34 individual means for the control group; the difference
between the two groups was not significant (P = 0.115).
Finally, to assess the possibility that subtle differences between
MLH Foci/cell
Figure 1. Distribution of MLH1 counts per cell in 2932 pachytene
spermatocytes from 34 controls and 1087 pachytene cells from 14
cases.
the two groups might result in slight reductions in MLH1
scores in most cases, we conducted a Mann–Whitney U-test;
the difference was not significant (P = 0.09). Thus, using various approaches, we were unable to detect any obvious difference between controls and cases in the total number of
exchanges per cell. Similarly, there were no apparent differences between cases and controls in the location of exchanges
on the chromosomes or in the number of individual bivalents
with 0, 1, 2 or 3 or more exchanges (data not shown).
Table I. Summary of observations on 26 individuals with non-obstructive azoospermia
Patient
Age
Karyotype
Results of Yq deletion analysis
No germ cells (n = 9)
Sp353
Sp384
Sp390
Sp391
Sp392
Sp394
Sp415
Sp417
Sp1004
30
45
46
31
32
29
36
48
32
46,XY
46,XY
46,XY
46,XY
47,XXY
46,XY
47,XXY
46,XY
46,XY
Negative
No information
Positive
Negative
Negative
No information
Negative
Negative
Negative
0
0
0
0
0
0
0
0
0
46,XY
46,XY
No information
Negative
Negative
No information
0
0
42.0
0
0
8.6
0
0
50
No information
46,XY
46,XY
See text
46,XY
46,XY
46,XY
46,XY
46,XY
46,XY
46,XY
46,XY
46,XY
46,XY
No information
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
Negative
47.4
49.0
46.8
47.5
49.8
48.9
46.3
47.4
50.8
50.2
47.5
51.8
49.2
45.2
49.0
3.6
6.1
4.3
4.5
4.3
5.0
4.5
1.8
5.5
5.2
3.6
3.2
4.1
4.1
5.0
57
43
111
95
207
109
43
7
76
225
43
24
23
24
1087
Meiotic arrest (n = 3)
Sp356a
29
Sp411
28
Sp1016
33
Meiotically normal (n = 14)b
Sp352
34
Sp354
33
Sp358
32
Sp361
32
Sp380
26
Sp381
35
Sp408
48
Sp409
38
Sp412
43
Sp1001
34
Sp1002
35
Sp1003
29
Sp1009
31
Sp1017
No information
Total
Mean number of MLH1 foci/cell
SD
0
0
0
0
0
0
0
0
0
Number of cells
0
0
0
0
0
0
0
0
0
a
Previously presented in Judis et al. (2004).
Previously presented in Hassold et al. (2004).
b
3173
D.Topping et al.
A proportion of individuals with NOA exhibit meiotic arrest
phenotypes
Of the 17 cases whose tissue samples contained at least some
germ cells, three (17.6%) exhibited meiotic arrest phenotypes.
In two cases, we were unable to locate immature or mature testicular spermatozoa. Instead, each sample contained apparently
normal leptotene and early zygotene cells but no normal pachytene cells; these were replaced by abnormal ‘end-stage’ cells
that were subtly different from case to case. In the third case,
we were unable to locate any testicular sperm, but rarely occurring, atypical sperm were identified on histological examination. Each of the three cases (Sp356, Sp411 and Sp1016) is
described below; as Sp356 was described in a previous report
(Judis et al., 2004), only the cellular phenotype is briefly summarized here.
Specifically, of 45 SCP3-positive cells, 24 were scored as
being at leptotene and 21 at zygotene. However, as is apparent
from Figure 3, the end-stage cells appeared more advanced
than those of Case 1. That is, the vast majority of these cells
exhibited at least partial synapsis between bivalents, and in a
few instances, the cells appeared to have near-normal pachytene configurations. Thus, meiosis appeared to arrest at the
zygotene/pachytene boundary, with partial synapsis between
homologs but with no evidence for crossing-over between
homologs.
Case 3 (Sp1016)
This case involved a 33-year-old man who presented for testicular biopsy for severe oligozoospermia. Routine pathology
Case 1 (Sp356)
A total of 102 SCP3-positive cells were identified, with 15
scored as being at leptotene and the remaining 87 at zygotene.
Representative ‘end-stage’ cells are shown in Figure 2; they
contained 46 full-length axial elements, but there was no evidence of transverse filament development and no evidence of
any association between homologous chromosomes. Additionally, MLH1 foci were not observed. Thus, meiotic arrest
appeared to occur at zygotene, with no evidence for synapsis or
subsequent crossing-over between homologs.
Case 2 (Sp411)
This case involved a 28-year-old man who presented for testicular biopsy for azoospermia. Cytogenetic analysis was normal
(46,XY), and the assay for a Yq chromosome deletion was
negative.
Similar to Case 1, no cells were found past the zygotene
stage, and there was no evidence for the deposition of MLH1.
Figure 2. Example of a representative ‘end-stage’ cell from Case 1
(Sp356), showing total failure of synapsis between homologous chromosomes. SCP3 [detecting axial/lateral elements of the synaptonemal
complex (SC)] is in red; CREST (detecting centromeric regions) is in
blue; MLH1 (detecting sites of exchanges) was not identified.
3174
Figure 3. Examples of end-stage cells from Case 2 (Sp411), showing
variation in the extent of synapsis in different cells. In (A), none of the
homologs are synapsed, as indicated by the presence of 46 axial elements. In (B), most bivalents appear to be fully synapsed, although
with several synaptic defects at 7 to 9 o’clock. SCP3 [detecting axial/
lateral elements of the synaptonemal complex (SC)] is in red; CREST
(detecting centromeric regions) is in blue; MLH1 (detecting sites of
exchange) was never identified.
Meiosis in non-obstructive azoospermia
revealed a maturation arrest with a few seminiferous tubules
exhibiting a small number of atypical sperm. Karyotype analysis and Yq chromosome deletion analysis were not done.
In initial studies of meiotic progression in 102 SCP3-positive cells, we identified three categories of cells: 60 appeared
to be normal leptotene cells, 17 were normal zygotene cells
and 25 were abnormal ‘pachytene-like’ cells with MLH1 localization but with severe defects in synapsis (Figure 4). The
extent of the synaptic defects was extremely variable, from
cells in which synapsis was nearly normal (Figure 4A) to cells
in which asynaptic ‘bubbles’ or ‘forks’ were evident on most,
if not all, bivalents (Figure 4B). Very little tissue was available
in this case, and thus, we were unable to examine additional
proteins to aid in specifying the timing of the arrest; however,
in most cells, the morphology of the XY bivalent was consistent with an early pachytene-stage cell (Solari, 1980).
In subsequent studies, we scored MLH1 foci in 50 of the
abnormal ‘pachytene-like’ cells. Surprisingly, the overall number
and distribution of exchanges were relatively normal. That is, the
mean number of MLH1 foci per cell was 42.0 ± 8.6, reduced by
only 15% from the overall mean for controls. Additionally, the
foci displayed interference, and as expected from studies of controls, distally located foci predominated (data not shown).
Thus, in this case, meiotic arrest appeared to occur in the
zygotene/pachytene boundary or early in pachytene. ‘Endstage’ cells were characterized by multiple synaptic defects,
but the recombination pathway was largely intact, with nearnormal numbers of MLH1 foci. However, although our sample
contained no post-pachytene cells, a small number of atypical
sperm were identified on histological examination. Thus, in at
least some seminferous tubules, a small number of cells were
able to escape this arrest and finish meiosis.
Discussion
Figure 4. Examples of end-stage cells from Case 3 (Sp1016), showing variation in the extent of synapsis in different cells. In (A), all but
two bivalents (arrows) appear to be completely synapsed. MLH1 foci
are apparent on all bivalents. In contrast, in (B), most bivalents have
failed to synapse. Examples of bivalents with distal and interstitial
regions of asynapsis are indicated by arrows and the arrowhead,
respectively; in this cell, the XY bivalent is not readily apparent.
Despite the synaptic defects, MLH1 foci are visible on all bivalents.
SCP3 [detecting axial/lateral elements of the synaptonemal complex
(SC)] is in red; CREST (detecting centromeric regions) is in blue;
MLH1 (detecting sites of meiotic exchanges) is in green.
The purpose of this study was 2-fold. First, in NOA cases in
which spermatogenesis proceeded past meiosis I, we wanted to
know whether defects in recombination were a common occurrence. If so, it would imply an increased risk of aneuploidy in
sperm of such individuals, because abnormalities in recombination are a well-established correlate of human meiotic nondisjunction (Lamb et al., 2005). Second, we were interested in
determining the proportion, if any, of cases of unexplained
infertility that were due to complete breakdown of spermatogenesis during meiosis I.
The results for each of the two aims were intriguing. First,
our studies of 14 cases in which post-pachytene stages were
identified provided remarkably little evidence of meiotic disturbances. Synaptic defects were uncommon, recombination
levels were similar to those in controls and testicular sperm
were evident in all cases. These results are somewhat surprising, because they are at variance with two recent immunostaining studies of NOA (Table II). Specifically, both Gonsalves
et al. (2004) and Sun et al. (2005a,b) reported significant
reductions in recombination in association with NOA, and in
cases studied by Martin and colleagues and reported in both
articles, synaptic defects were more common in NOA than in
controls. The reason for the discrepancy between their reports
and ours is not clear but could conceivably reflect methodological differences. That is, in our initial studies of recombination
in ‘control’ males (Lynn et al., 2002), we arbitrarily restricted
our analyses to normally synapsed cells with 23 or more MLH1
foci. We reasoned that this would prevent us from scoring cells
that were just entering or exiting pachytene or cells that had
severe synaptic defects and provided us the potential to detect
at least one MLH1 focus per bivalent. We continued this practice in the present study, because any changes in scoring could
have introduced an artefactual difference between cases and
controls. Neither Gonsalves et al. (2004) nor Sun et al.
3175
D.Topping et al.
Table II. Comparison of cytological findings of other investigators in individuals with non-obstructive azoospermia (NOA)
Investigator
Sun et al. (2005a)
Gonsalves et al. (2004)
Topping et al. (2006)
Total number
of patients
18a
40
26
Number of patients with
Sertoli cell-only [n (%)]
8 (44)
21 (53)
9 (35)
Number of patients with
meiotic arrest [n (%)]
1 (6)
4 (10)
3 (12)
Patients with post-pachytene cells
Number of patients [n (%)]
Mean MLH1 foci/cell ± SD
9 (50)
15 (37)
14 (54)
40.4 ± 4.7
45.9 ± 3.9
49.0 ± 5.0
a
Eight of these 18 individuals are represented in the data set collected by Gonsalves et al.
(2005a,b) took this approach in their studies, and as a large
number of their NOA cases had very low MLH1 scores, the
overall means were lower for NOA than the controls. Thus, if
we had taken their approach (e.g. scoring MLH1 foci in cells
with synaptic defects), it is possible that we also would have
identified a significant difference in recombination between
cases and controls. However, although this is possible, we
think it unlikely. That is, we identified a similarly low level of
synaptic defects in our cases and controls, suggesting that we
would have excluded a similar proportion of cells in both subject categories. Thus, we think it more likely that other factors
(e.g. differences in the sample populations), not methodological differences, are the basis for the among-study variation.
Clearly, additional studies will be needed to resolve this discrepancy and to determine the level of impairment in recombination, if any, in individuals with NOA. Second, we identified
meiotic arrest phenotypes in 3 of the 26 cases, meaning that
>10% of all NOA was attributable to meiotic defects. Furthermore, if cases without evidence of germ cells were excluded,
the frequency increased to ∼20% (i.e. 3/17 cases). Importantly,
these values are similar to those of the two other recently
reported immunostaining studies of NOA (Table II). Specifically, Gonsalves et al. (2004) analysed 40 cases of NOA and
identified two cases with complete arrest at zygotene and two
others in which only a few pachytene cells were present; similarly, Sun et al. (2005a,b) reported complete zygotene arrest in
1 of 18 cases of NOA. Taken together, these studies suggest
that as many as one in five cases of NOA in which germ cells
are present are due to abnormalities in meiotic prophase. This
has obvious clinical implications. For example, in instances in
which testicular biopsies are ascertained for diagnostic purposes, immunostaining analysis is more likely to uncover the
cause of infertility than are other standard tests (e.g. routine
karyotyping). Furthermore, in instances in which a complete
meiotic arrest is identified on immunostaining and in which
multiple regions of the testis are examined without any evidence for spermatogenesis, it seems reasonable to consider
options other than ICSI (e.g. donor sperm or adoption).
The cases of meiotic arrest are also instructive with regard to
the possible mutational sources of male infertility. That is, the
three meiotic arrest phenotypes in our study were similar superficially but on closer examination were clearly distinguishable: one
case exhibited no synapsis, one partial synapsis but no MLH1
localization and the third partial synapsis and MLH1 deposition.
The first two individuals (Sp356 and Sp411) displayed a complete
meiotic arrest because there were no sperm visible in the samples.
The third individual displayed a partial meiotic arrest phenotype
3176
because a small proportion of cells were able to progress through
the ‘pachytene checkpoint’ (Roeder and Bailis, 2000) to produce
sperm, thereby avoiding the arrest and apoptosis seen in models
such as Atm-deficient mice (Xu et al., 1996). This is reminiscent
of the subtle differences in meiotic phenotype observed in mice
homozygous for different null mutations. Thus, it may be that
there is no single major cause of meiotic arrest in human males but
rather rarely occurring mutations involving any of a number of
different meiotic genes. It will be important to confirm or refute
this suggestion on additional series of cases because, if true, it will
complicate attempts to determine the underlying molecular causes
of male infertility.
Acknowledgements
This research was supported by NIH grant HD42720.
References
Brown PW, Judis L, Chan ER, Schwartz S, Seftel A, Thomas A and Hassold
TJ (2005) Meiotic synapsis proceeds from a limited number of subtelomeric
sites in the human male. Am J Hum Genet 77,556–566.
Chaganti RS, Jhanwar SC, Ehrenbard LT, Kourides IA and Williams JJ (1980)
Genetically determined asynapsis, spermatogenic degeneration, and infertility in men. Am J Hum Genet 32,833–848.
Chandley AC (1988) Meiosis in man. Trends Genet 4,79–84.
Claustres M, Guittard C, Bozon D, Chevalier F, Verlingue C, Ferec C, Girodon
E, Cazeneuve C, Bienvenu T, Lalau G et al. (2000) Spectrum of CFTR
mutations in cystic fibrosis and in congenital absence of the vas deferens in
France. Hum Mutat 16,143–156.
Cooke HJ and Saunders PT (2002) Mouse models of male infertility. Nat Rev
Genet 3,790–801.
Egozcue J, Sarrate Z, Codina-Pascual M, Egozcue S, Oliver-Bonet M, Blanco J,
Navarro J, Benet J and Vidal F (2005) Meiotic abnormalities in infertile
males. Cytogenet Genome Res 111,337–342.
Gonsalves J, Sun F, Schlegel PN, Turek PJ, Hopps CV, Greene C, Martin RH
and Pera RA (2004) Defective recombination in infertile men. Hum Mol
Genet 13,2875–2883.
Hassold T, Judis L, Chan ER, Schwartz S, Seftel A and Lynn A (2004) Cytological studies of meiotic recombination in human males. Cytogenet
Genome Res 107,249–255.
Hunt PA and Hassold TJ (2002) Sex matters in meiosis. Science 296,2181–2183.
Judis L, Chan ER, Schwartz S, Seftel A and Hassold T (2004) Meiosis I arrest
and azoospermia in an infertile male explained by failure of formation of a
component of the synaptonemal complex. Fertil Steril 81,205–209.
Krausz C, Forti G and McElreavey K (2003) The Y chromosome and male fertility and infertility. Int J Androl 26,70–75.
Lamb NE, Sherman SL and Hassold TJ (2005) Effect of meiotic recombination
on the production of aneuploid gametes in humans. Cytogenet Genome Res
111,250–255.
Lynn A, Koehler KE, Judis L, Chan ER, Cherry JP, Schwartz S, Seftel A, Hunt
PA and Hassold TJ (2002) Covariation of synaptonemal complex length and
mammalian meiotic exchange rates. Science 296,2222–2225.
Lynn A, Ashley T and Hassold T (2004) Variation in human meiotic recombination. Annu Rev Genomics Hum Genet 5,317–349.
Meiosis in non-obstructive azoospermia
Martin RH, Greene C, Rademaker AW, Ko E and Chernos J (2003) Analysis
of aneuploidy in spermatozoa from testicular biopsies from men with nonobstructive azoospermia. J Androl 24,100–103.
Micic M, Micic S and Diklic V (1982) Low chiasma frequency as an aetiological factor in male infertility. Clin Genet 22,266–269.
Revenkova E, Eijpe M, Heyting C, Hodges CA, Hunt PA, Liebe B, Scherthan H
and Jessberger R (2004) Cohesin SMC1 beta is required for meiotic chromosome dynamics, sister chromatid cohesion and DNA recombination. Nat
Cell Biol 6,555–562.
Roeder GS and Bailis JM (2000) The pachytene checkpoint. Trends Genet
16,395–403.
Solari AJ (1980) Synaptosomal complexes and associated structures in microspread human spermatocytes. Chromosoma 81,315–337.
Sun F, Kozak G, Scott S, Trpkov K, Ko E, Mikhaail-Philips M, Bestor TH,
Moens P and Martin RH (2004) Meiotic defects in a man with non-obstructive azoospermia: case report. Hum Reprod 19,1770–1773.
Sun F, Greene C, Turek PJ, Ko E, Rademaker A and Martin RH (2005a)
Immunofluorescent synaptonemal complex analysis in azoospermic men.
Cytogenet Genome Res 111,366–370.
Sun F, Trpkov K, Rademaker A, Ko E and Martin RH (2005b) Variation in
meiotic recombination frequencies among human males. Hum Genet
116,172–178.
Van Assche E, Bonduelle M, Tournaye H, Joris H, Verheyen G, Devroey P,
Van Steirteghem A and Liebaers I (1996) Cytogenetics of infertile men.
Hum Reprod 11(Suppl 4),1–24; discussion 25–26.
Vendrell JM, Garcia F, Veiga A, Calderon G, Egozcue S, Egozcue J and Barri
PN (1999) Meiotic abnormalities and spermatogenic parameters in severe
oligoasthenozoospermia. Hum Reprod 14,375–378.
Xu Y, Ashley T, Brainerd EE, Bronson RT, Meyn MS and Baltimore D (1996)
Targeted disruption of ATM leads to growth retardation, chromosomal fragmentation during meiosis, immune defects, and thymic lymphoma. Genes
Dev 10,2411–2422.
Xu H, Beasley MD, Warren WD, van der Horst GT and McKay MJ (2005)
Absence of mouse REC8 cohesin promotes synapsis of sister chromatids in
meiosis. Dev Cell 8,949–961.
Submitted on February 6, 2006; resubmitted on April 24, 2006 and May 25,
2006; accepted on June 2, 2006
3177