The rate of aneuploidy is altered in spermatids from infertile mice

L. Oppedisano1, G. Haines1, C. Hrabchak1, G. Fimia3, R. Elliott2, P. Sassone-Corsi3 and S. Varmuza1,4

1 Department of Zoology, University of Toronto, 25 Harbord St., Toronto, Ontario, Canada M4S 3G5, 2 Rosewell Park Cancer Institute, Department of Molecular and Cellular Biology, Buffalo, NY 14263-0001, USA and 3 Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS-INSERM, Université Louis Pasteur, Illkirch-Strasbourg, France, 67404


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: It is now possible for infertile males to father their own genetic children through the technique of ICSI. This prospect has consequently prompted several investigations into the quality of sperm being retrieved from infertile males. One potential risk is the use of aneuploid sperm or spermatids, which might then be transferred to the fertilized oocyte. METHODS: In this investigation, aneuploidy of spermatids was assessed through immunocytochemistry using antibodies directed against chromosome centromeric regions and complexes. Three different types of infertile male mice with phenotypes closely resembling those described in human non-obstructive azoospermia [PP1c{gamma}-deficient mice, CREM-deficient mice and C57BL/6J.MAC-170–23 mice] were examined for chromosome numbers by counting the number of kinetochores in round spermatids using a CREST antiserum. RESULTS: PP1c{gamma}-/- and CREM-/- spermatids from infertile mice showed highly significant elevated levels in the rate of aneuploidy compared with wild-type animals (P < 0.0001). Thus infertile males with independent genetic mutations resulting in different histopathologies showed a high risk in the level of aneuploidy in their spermatids. CONCLUSIONS: These results suggest that impaired spermatogenesis may lead to production of aneuploid gametes. Analysis of aneuploidy in gametes from infertile men, coupled with appropriate genetic counselling, is recommended prior to ICSI.

Key words: aneuploidy/immunocytochemistry/spermatid/spermatogenesis/testis


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Recent insights into the molecular mechanisms regulating fertility and pregnancy have revolutionized the field of reproductive biology and produced advances in assisted reproductive techniques. It has become standard practice for couples with infertility problems to consider the technologies offered by IVF (Andrews and Elster, 1998Go; Givens, 2000Go). New developments in this field have allowed many couples, who were previously unable to conceive naturally, the opportunity to bear their own genetic children (Sofikitis et al., 1998Go; Prapas et al., 1999Go; Givens, 2000Go).

Approximately half of couples who present themselves at infertility clinics have problems that can be attributed to the male partner (Aitken and Stewart, 1996Go; Cooke et al., 1998Go; Huang et al., 1999Go; Rives et al., 1999Go). Causes of male factor infertility include prior infections, primary endocrine failure and immunological disorders (Diemer and Desjardins, 1999Go). However, the largest single class of male factor infertility cannot be attributed to any of these disorders and is classified as idiopathic (Elliott and Cooke, 1997Go; Diemer and Desjardins, 1999Go). Idiopathic infertile men routinely show symptoms of severe oligozoospermia or azoospermia (Diemer and Desjardins, 1999Go).

While the underlying aetiologies of idiopathic azoospermia are unknown, attempts are being made to shed light on this issue. A significant proportion of azoospermia cases have been associated with microdeletions of the Y chromosome (Oliva et al., 1998Go), and many numerical or constitutional chromosome abnormalities can lead to a block in meiosis, presumably through pairing dysfunction (Diemer and Desjardins, 1999Go). However, a large proportion of idiopathic azoospermia remains unexplained, leading some reviewers to speculate that as yet unidentified X-linked or autosomal mutations in genes involved in spermatogenesis may be the underlying cause (Elliott and Cooke, 1997Go; Tuerlings et al., 1997Go; Chandley, 1998Go; McLachlan et al., 1998Go; Okabe et al., 1998Go). A vertical genetic study (Liliford et al., 1994Go) has provided statistical support for this view.

Despite its success, concern over the use of immature germ cells in ICSI and round spermatid injection (ROSI) procedures has been raised, and long-term implications of this technique are still unclear. Success rates with ROSI and elongated spermatid injection are significantly lower than that with ICSI (Prapas et al., 1999Go), possibly due to the immature nature of the gametes being used (Sousa et al., 1998Go). Another risk factor associated with these procedures is the possibility that gametes with chromosome abnormalities may be used for fertilization, thus generating embryos with unbalanced chromosome complements (Bernardini et al., 1998Go). Chromosomally unbalanced embryos can result in early spontaneous abortion, stillbirth, or birth with severe developmental and/or congenital problems (Giltay et al., 1999Go). Many studies have reported elevated levels of aneuploidy in sperm from infertile men (Colombero et al., 1999Go; Pfeffer et al., 1999Go; Huang et al., 1999Go; Rives et al., 1999Go; Ushijima et al., 2000Go). Further, paternally derived aneuploidy has been reported to be a leading cause of pregnancy loss, particularly in the case of (45,X) embryos (Hassold, 1998Go). Studies have been published which attempt to assess the effects and success of these procedures specifically in the context of these risk factors in humans (Feichtinger et al., 1995Go; In't Veld et al., 1995Go; Kurinczuk and Bower, 1997Go; Mansour et al., 1997Go; Bonduelle et al., 1998Go; Munné et al., 1998Go).

We report here the results of an analysis performed on spermatids retrieved from testicular cell suspensions from three different types of infertile male mice whose sterility is genetically based. PP1c{gamma}-/- males are completely sterile, with reduced production of condensing and elongating spermatids, very few mature testicular sperm, and no sperm in the epididymides (Varmuza et al., 1999Go). CREM -/- males show complete arrest at the round spermatid stage (Blendy et al., 1996Go; Nantel et al., 1996Go). C57BL/6J.MAC170–23 males are sterile with a histopathological profile similar to PP1c{gamma} -/- males (R.Elliott, unpublished data). The absence of sperm in all three types of males led us to devise an alternative methodology for assessing haploid chromosome numbers based on immunocytochemical detection of centromeres in round spermatids. We show that the levels of numerical chromosome abnormalities in the spermatids from two of the three different sterile males are elevated. Our observations are consistent with the hypothesis that a compromised testicular environment may play a role in the increased levels of chromosomal abnormalities (Mroz et al., 1998Go), but that not all types of sterility increase the frequency of meiotic non-disjunction.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Mice
Mice of various genotypes were bred using standard animal husbandry. Individual males were genotyped as described (Nantel et al., 1996Go; Varmuza et al., 1999Go). C57BL/6J.MAC-170–23 (hereafter referred to as B6.MAC) mice were genotyped using chromosome specific markers (R.Elliott, unpublished data). Testes of all mice were processed and analysed fresh each time as described below. The PP1c{gamma} mice are from a genetically mixed outbred colony, while the CREM mutants are from a mixed C57BL/6x129 colony, and the B6.MAC mice are pure C57BL/6J with a portion of chromosome 17 derived from Mus macedonicus. We used the wild-type littermates of the PP1c{gamma}-/- males as controls for all experiments.

Immunocytochemistry
The CREST antiserum, obtained from patients with Calcinosis Raynaud's phenomenon Esophageal dysmotility Sclerodactyly Telangiectasia, a syndrome of scleroderma, recognizes an epitope in kinetochores and produces a distinctive, and highly reproducible, punctate signal in immunocytochemical preparations. The number of signals reflects indirectly the number of chromosomes (Figure 1Go). The reliability of the methodology makes it possible to obtain large amounts of data with very small amounts of material. Because round spermatids are the most abundant cell type in testicular cell suspensions, it is possible to perform extensive analyses on small amounts of testicular tissue.



View larger version (102K):
[in this window]
[in a new window]
 
Figure 1. Immunofluorescent labelling of testicular round spermatids using antibodies directed against centromeres (CREST), histones (H1T), and the synaptonemal complex (COR-1). CREST and COR-1 are visualized with rhodamine secondary antibodies. The autofluorescent acrosome (arrow) and size of CREST signals (large arrowhead) distinguish the spermatid on right from pachytene spermatocyte on the left, which stains positively for the distinctive COR-1 protein (small arrowhead). (B) The same field of cells as in (A) visualized with the FITC filter display labelling of Histone H1T, which is strong in the spermatid and weak in the pachytene spermatocyte. Spermatids positive for H1T, CREST and the autofluorescent acrosomal marker, and negative for COR-1 were used for counting. The autofluorescent acrosome marker was specific for the particular CREST antiserum used throughout this study. Other CREST antisera do not possess this feature. Bar = 10 µm.

 
Sexually mature male mice (between 9 and 24 weeks of age) were killed by cervical dislocation and both testes were dissected. Following removal of the tunica albuginea, seminiferous tubules were dissociated, and testis cell suspensions were processed for immunocytochemistry as described (Moens et al., 1987Go; Moens and Earnshaw, 1989Go; Dobson et al., 1994Go). Slides were incubated with a mixture of primary antibodies (human CREST 1/500, rabbit-anti histone H1T 1/16, rabbit-anti COR1 1/500) overnight at room temperature. After washing, slides were incubated with a mixture of secondary antibodies [Cy3-conjugated goat anti-human, fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit] for 1 h at 37°C. Just before mounting, slides were incubated briefly in 1 µg/ml 4',6-diamidino-2-phenylindole, air-dried and coverslips were mounted with an antifade mounting solution. The same batch of CREST antiserum was used for all experiments. In a few experiments, mouse anti-COR1 at a concentration of 1/500 was used instead of rabbit anti-COR1. In these experiments, slides were incubated with a mixture of secondary antibodies including Rhodamine-conjugated goat-anti human, Rhodamine-conjugated goat anti-mouse and FITC-conjugated goat anti-rabbit antibodies.

Data analysis
Slides were examined under fluorescent light at x1000 using oil immersion objectives on an Olympus BX60 microscope. Spermatids were identified as those cells that stained positive for Histone H1T and CREST, negative for COR-1, and positive for an autofluorescent acrosome signal (see Figure 1Go). Spermatids represent approximately half of all non-condensed cells in the testicular cell suspension, with the other half comprising diploid, tetraploid and octaploid germ cells in various stages of mitosis and meiosis, and somatic cells. Spermatids with indistinct or excessively clumped centromeres, or overlapping or very closely apposed cells were not recorded. Slides were systematically scanned until 100 images per subject had been captured. All scorable spermatids in a field of view were recorded to reduce observer bias in sampling. The images were captured and recorded using a CCD camera and stored on writeable compact discs as TIFF files. Some images contained several well-separated spermatids in the field of view. These spermatids were scored individually. Following transformation of images into greyscale, centromeres were counted manually and recorded using Quantity One image analysis software (BioRad Laboratories, Hercules, CA, USA). The number of mice assayed in each group was 9 (+/+), 9 (+/-), 14 (-/-), four (CREM-/-) and five (B6.MAC).

Scoring criteria were based on the high degree of uniformity of CREST signals from cell to cell, and experiment to experiment. Many cells contained centromeres that were close together. However, shape and intensity of the aggregate signals were used as cues for scoring clumped centromeres (Figure 2Go). Singletons in each field acted as internal controls for size and intensity. No correlation between clumping and chromosome scores was detected (P = 0.712).



View larger version (133K):
[in this window]
[in a new window]
 
Figure 2. Scoring criteria. Most spermatids had at least one clump of overlapping centromere signals. Intensity and shape of clumps were both used to score the number of centromeres within a clump, with intensity of singletons acting as a guide to intensity and size of one signal. Only cells with multiple single signals were recorded. In the cell shown in (A), the boxed segment is highlighted twice in (B) and (C), with triangles indicating the scores assigned to the clump on the right. Signals that were too weak, or that did not have the typical rounded shape, were not counted (arrow). Bar = 10 µm.

 
The entire set of images was counted independently by two people in a semi-blind fashion. Since mutant phenotypes in some cases made identification of mutant cells obvious, it was not possible to do a double-blind image analysis. Comparison of the two data sets by analysis of variance revealed that between-set variation was much smaller than within-set variation. Only one data set is presented here. Data were subjected to Fisher's exact test and analysed to determine the significance of differences in variation of chromosome numbers, comparing the wild-type spermatid data set derived from the wild-type littermates of PP1c{gamma}- mutant males, to each of the PP1c{gamma}+/- PP1c{gamma}-/-, CREM-/- and B6.MAC spermatid data sets. While chromosome numbers are meristic traits with discrete values, and therefore have the potential to display a skewed distribution, in practice they appeared to be distributed normally (see Figure 3Go). This validated the use of the various statistical tests in our study.



View larger version (46K):
[in this window]
[in a new window]
 
Figure 3. Chromosome numbers in PP1c{gamma} mutant spermatids. (A) Chromosome numbers from wild-type, PP1c{gamma}+/-, and PP1c{gamma}-/- spermatids, raw data; (B) chromosome numbers from all mice used are plotted to illustrate the range of values. Categories with values of zero are left blank. SD are shown by horizontal bars, for wild-type spermatids (green; SD: ± 1.2), and for mutant spermatids (red;SD: ± 4.71).

 
Comparative rates of aneuploidy were determined by calculating the frequency of abnormal chromosome numbers for each data set. Some degree of variability in scoring among different observers was seen (average SD across 563 images scored by three observers = 0.72). To account for scoring and other experimental errors, we arbitrarily set a `normal' range of the mean ± SD of the wild-type data set (i.e. cells with 18–22 centromere signals were classified as normal, those with fewer or greater numbers were classified as abnormal). The SD in the wild-type data set is 1.205. Twice this value is an integer >2; however, the outcome does not change if a range of 17–23 is used to define `normal'. This conservative approach was deemed necessary to account for some of the experimental error in the procedure, and to take into account the fact that our approach involved scoring the entire genome, and not just a subset of chromosomes.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
PP1c{gamma} mutant males produce round spermatids with elevated numerical chromosome abnormalities
The wild-type data set (shown in Figure 3Go), composed of 1010 cells from nine different mice, yielded a mean chromosome number of 20.19 ± 1.24 (n = 1010, Table IGo). The variation from the expected haploid chromosome number of 20 could reflect experimental error (extra signals caused by debris or stray chromosomes from burst nuclei; reduced signals representing overlapping centromeres or loss of chromosomes following rupture of nuclei, inconsistency in applying scoring criteria), a base-line level of aneuploidy in normal males, or a combination of the two. We cannot distinguish these explanations with our approach, and therefore conservatively favour the latter. This robust data set is the standard against which our experimental data sets were compared, without making assumptions about absolute rates of chromosome abnormalities.


View this table:
[in this window]
[in a new window]
 
Table I. F-Test analysis of the variance in numerical chromosome numbers in spermatids from various strains of infertile mice
 
Spermatids from PP1c{gamma}-/- males displayed a significantly increased variation in chromosome numbers compared to wild-type controls (mean 21.35 ± 4.709; n = 1598) (Table IGo and Figure 3BGo). In the initial study of the effect of the PP1c{gamma} mutation on spermatogenesis (Varmuza et al., 1999Go), it was reported that polyploid spermatids were observed, based on the presence of multiple acrosomes or doublet elongating spermatids. The presence of diploid spermatids could therefore be solely responsible for the increased variation in chromosome number. There were indeed a number of spermatids with 40 or close to 40 chromosomes (an example is shown in Figure 1AGo) and the distribution of chromosome numbers for the entire data set is clearly right skewed (Figure 3Go). In order to exclude diploid cells from the data set, we eliminated cells with >35 or >30 chromosomes. In both cases, variation in chromosome numbers in the truncated data set was significantly greater than that observed in the wild-type spermatids (Table IGo), suggesting that diploidy alone is not responsible for the increased variation in chromosome number.

Some of the observed variation may be skewed by the high numbers of centromere signals in some cells. Therefore it was informative to look at frequencies of cells with numerical chromosome abnormalities rather than variation in centromere numbers. However, the actual rate of aneuploidy cannot be determined by our approach because of the uncertainty in distinguishing experimental error from biological chromosome abnormality. Only comparative rates of numerical chromosome abnormalities can be estimated in relation to the wild-type group, if it is assumed that all groups are subject to the same degree of experimental error (debris on the slide, addition or loss of chromosomes from burst nuclei, overlapping centromeres, counting errors).

We adopted a conservative approach to establishing `normal' numbers of centromere signals by arbitrarily setting a range of the mean ± 2 SD of the wild-type data. This convention revealed rates of numerical chromosome abnormality of 4.16% (wild type), 3.78% (PP1c{gamma}+/-, and 15.39% (PP1c{gamma}-/-). {chi}2-Analysis revealed that only the mutant PP1c{gamma}-/- numerical chromosome abnormality rate was statistically significant, even after accounting for diploidy by removing cells with >=35 centromere signals, or cells with >=30 centromere signals from the data set (Table IIGo). Interestingly, the rate of chromosome abnormality declined markedly after removing potentially diploid cells from the data set. This observation is consistent with the presence of both aneuploid and diploid spermatids in the PP1c{gamma}-/- gamete pool.


View this table:
[in this window]
[in a new window]
 
Table II. Frequency of chromosomally abnormal spermatids
 
Other mutations causing sterility have different effects on aneuploidy
The observation of such a marked increase in numerical chromosome abnormalities in PP1c{gamma}-/- mice prompted us to investigate whether similar abnormalities are observed in other mouse strains that exhibit a sterile phenotype. We examined chromosome numbers in round spermatids of two other sterile mouse models, CREM-/- and B6.MAC, each of which closely resembles the phenotype observed in PP1c{gamma}-/- males. The germ cells in males derived from each of these genetic backgrounds appear to progress through meiosis normally within the tubules, but then arrest at some stage during spermiogenesis. CREM-/- mice possess a targeted mutation in the CREM transcription factor gene. Males are sterile, with germ cells arrested at the round spermatid stage (Blendy et al., 1996Go; Nantel et al., 1996Go). Animals from the congenic strain B6.MAC contain the proximal region of chromosome 17 from position 0–23 derived from Mus macedonicus in a C57BL/6J background (R.Elliott, unpublished data). Males with this chromosome configuration are sterile, again with an arrest during spermiogenesis. It appears that these males make sperm heads, albeit largely abnormal in appearance, but no tails (R.Elliott, unpublished data). The variation in chromosome numbers of round spermatids was significantly elevated in both types of males (Table IGo). In order to eliminate the possibility that diploid spermatids might contribute most of the variation in chromosome numbers, CREM-/- and B6.MAC spermatids with >=35 chromosomes, or with >=30 chromosomes, were separately eliminated from the pool analysed (Table IGo). Statistical significance in differences in variation of chromosome numbers becomes marginal for the B6.MAC spermatids when cells with greater than 30 centromere signals are eliminated from the data set, but remains high for the CREM-/- spermatids.

The frequency of numerical chromosome numbers in CREM-/- spermatids was 13.06%. This elevated frequency is statistically significantly different from wild type, even after removal of potentially diploid cells (Table IIGo). The latter exercise had a much less dramatic effect on the proportion of chromosomally abnormal cells in CREM-/- spermatids than in PP1c{gamma}-/- spermatids, suggesting that there are far fewer diploid cells in these mice. This may be an artefact of image capture if diploid or polyploid spermatids do not display the same features (histone H1T positive, presence of autofluorescent acrosome) as haploid spermatids. Interestingly, spermatids from B6.MAC mice displayed a low frequency of numerical chromosome numbers (4.47%) that was not statistically different from wild type (Table IIGo).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Male mice bearing a targeted mutation in PP1c{gamma} (PP1c{gamma}-/-) are completely sterile with failure of spermiogenesis beginning at the round spermatid stage (Varmuza et al., 1999Go). These mice show pathological and histological similarities to human males suffering from idiopathic non-obstructive azoospermia. Thus, these mice may provide us with a suitable model to study certain forms of human infertility. The experiments described in this paper were aimed at determining whether the germ cells from PP1c{gamma}-/- male mice contained elevated levels of numerical chromosome abnormalities, as has been observed in infertile human males. Further, we investigated whether this condition is correlated with the PP1c{gamma} mutation alone, or whether it can be demonstrated in other types of sterile mice and might therefore be attributed to testicular pathology affecting meiosis adversely.

The variation in chromosome numbers in PP1c{gamma}-/- spermatids was significantly elevated when compared with wild-type controls. These results indicate that a large proportion of germ cells in PP1c{gamma}-/- males is either aneuploid or diploid. While our approach did not allow us unequivocally to measure absolute rates of aneuploidy, the frequency of spermatids in the PP1c{gamma}-/- sample with numerical chromosome abnormalities was 3.7-fold higher than the frequency in the wild-type sample (4.16 versus 15.39%). Even after removal of potentially diploid cells from the data set, a significantly elevated frequency of cells with abnormal chromosome numbers was observed. Heterozygous PP1c{gamma}+/- spermatids were indistinguishable from wild type when the frequency of abnormal cells was compared.

The increased level of aneuploidy/diploidy observed in PP1c{gamma} mutants may have been a direct consequence of the mutation, since it is known that type 1 protein phosphatases are involved in both meiosis and mitosis in yeast and fruit flies (Ohkura et al., 1989Go; Axton et al., 1990Go; Baksa et al., 1993Go; Tu et al., 1996Go). Alternatively, altered testicular environment may have contributed to numerical chromosome abnormalities by adversely affecting some aspects of meiosis. We investigated this latter hypothesis by examining two other strains of sterile mice that appear to progress through meiosis, and arrest or become dysfunctional during spermiogenesis. Both CREM-/- and B6.MAC spermatids displayed increased variation in chromosome numbers when compared with wild type. However, when frequencies of abnormal cells were analysed in these two data sets, only CREM-/- spermatids continued to display statistically significant differences from wild type. These results argue that testis pathology plays an important role in maintaining normal meiotic divisions, but that different mutations may have different effects on the general pathology.

It was evident from both the quantitative data and the visual data that some proportion of PP1c{gamma} spermatids was diploid. In contrast, very few spermatids from CREM-/- and B6.MAC mice appeared to be diploid. This observation argues further for different responses of testicular pathology to different genetic lesions. In the case of the PP1c{gamma} mutation, the presence of both diploid and aneuploid spermatids argues for two defective responses—one involving failure of one meiotic division in a small proportion of spermatocytes, and another involving non-disjunction in a somewhat larger proportion of spermatocytes.

A large number of studies have recently applied the technique of multicolour FISH using centromeric probes specific for a small subset of chromosomes to assess aneuploidy in sperm of infertile men. The majority of these studies suggest that germ cells derived from infertile males have a higher incidence of chromosomal abnormalities compared to normal (fertile) men (Colombero et al., 1999Go; Pfeffer et al., 1999Go; Huang et al., 1999Go; Rives et al., 1999Go; Ushijima et al., 2000Go). However, other studies have reported no statistically significant difference (Damri et al., 2000Go; Martin et al., 2000Go). These studies may reflect variation in the aetiology of infertility among the subjects, rather than inherent methodological differences in the analyses, although the impact of the latter cannot be ignored. Nevertheless, the potential risk of using aneuploid gametes for ICSI when the male partner displays abnormal spermatogenesis cannot be underestimated. One of the limiting factors in these studies is availability of suitable material for analysis (i.e. sperm). For men with azoospermia, FISH studies on sperm are not possible. It may prove useful therefore to adapt the procedure we have described here, since it is possible with a very small testicular biopsy to obtain sufficient round spermatids for an extensive analysis (S.Varmuza, unpublished data). Suitable markers that would identify early stage round spermatids would need to be used. The autofluorescent acrosome marker that we found so very useful in our studies of mice was a specific feature of the particular batch of CREST antiserum used throughout the study, and is not seen with other CREST antisera (S.Varmuza, unpublished data).

Some studies have shown that children born from ICSI may be at risk for increased levels of sex chromosome aneuploidy (In't Veld et al., 1995Go; Andrews and Elster, 1998Go; Bonduelle et al., 1998Go; Givens, 2000Go), whereas others report no significant increase in risk of aneuploid embryos (Feichtinger et al., 1995Go; Munné et al., 1998Go). However, a prenatal chromosomal analysis performed on ICSI fetuses (Van Opstal et al., 1997Go) determined that all of the sex chromosome aneuploidy (six out of 71 ICSI-derived fetuses) was of paternal origin. None of these studies, however, distinguished between male partners with normal spermatogenesis and those with abnormal spermatogenesis. The possibility therefore exists that germ cells acquired from males suffering from non-obstructive azoospermia are less capable of supporting development and producing normal children than ICSI procedures performed on obstructive azoospermic patients where spermatogenesis appears normal (Chandley and Hargreave, 1996Go; Lange et al., 1997Go; Zech et al., 2000Go). Further, Hassold (1998) reports that paternally derived aneuploidy involving the X chromosome (45,X) is one of the leading known causes of spontaneous pregnancy loss.

It should be noted that the centromeres in CREM-/- round spermatids appeared slightly more ragged than those in spermatids from wild-type or PP1c{gamma} mutant testes. This is a function of the early block in spermiogenesis; very early spermatids in wild-type testicular cell preparations with a similar appearance can be found at much lower frequency. However, these cells are passed over for image capture in favour of the more advanced spermatids with a strong autofluorescent acrosome and more distinct centromeres. In recording CREM mutant spermatids, attempts were made to choose those cells with the most distinct centromeres. Inter-observer variation in scoring was slightly higher for the CREM images than for the PP1c{gamma} images, but was lower than for the B6.MAC images, which had clearer centromere signals.

Quantitative data produced by visual scoring of cytological characteristics such as our centromere signals, or FISH signals, are inherently error-prone, especially where subjective evaluation of spots on a slide is required. Even with guidance by strict scoring criteria, it is inevitable that some error will creep in. This point is well illustrated by the high variability in aneuploidy rates reported by various laboratories (Shi and Martin, 2000Go). We would argue that cytological approaches such as ours, and FISH, are most useful in comparative studies, and perhaps less useful in establishing absolute rates of aneuploidy.

We have described a series of experiments in which chromosome numbers in round spermatids from infertile male mice were quantified using immunofluorescent localization of centromeres. An elevated level of numerical chromosome abnormalities was observed in two of three different sterile strains. Our results are consistent with the hypothesis that abnormal testicular environment can adversely affect meiosis, leading to elevated non-disjunction. This observation suggests that the use of ICSI for treatment of male factor infertility may carry an increased risk of producing a chromosomally unbalanced zygote. Adaptation of the methodology described here to analysis of human testicular biopsies may provide an additional level of understanding of the risk of aneuploidy.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
We would like to thank Dr Peter Moens for providing the primary antibodies, the protocol for processing testicular cells and for identifying the autosomal fluorescent marker. We would also like to thank Ian Dworkin for assistance with statistical analysis of the data, and Dr Terry Hassold and three anonymous reviewers for helpful comments on the manuscript. S.Varmuza was funded by NSERC #OGP0138636.


    Notes
 
4 To whom correspondence should be addressed. E-mail: svarmuza{at}zoo.utoronto.ca Back

Submitted on June 6, 2001


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Aitken, J.R. and Stewart, I.D. (1996) Reproductive biology: fertilization without sperm. Nature, 379, 493–495.[ISI][Medline]

Andrews, L. and Elster, N. (1998) ART into science: regulation of fertility techniques. Science, 281, 651–652.[Free Full Text]

Axton, J.M., Dombradi, V., Cohen, P.T. et al. (1990) One of the protein phosphatase 1 isoenzymes in Drosophila is essential for mitosis. Cell, 63, 33–46.[ISI][Medline]

Baksa, K., Morawietz, H., Dombradi, V. et al. (1993) Mutations in the protein phosphatase 1 gene at 87B can differentially affect suppression of position-effect variegation and mitosis in Drosophila melanogaster. Genetics, 135, 117–125.[Abstract/Free Full Text]

Bernardini, L., Borini, A. and Preti, S. (1998) Study of aneuploidy in normal and abnormal germ cells from semen of fertile and infertile men. Hum. Reprod., 13, 3406–3413.[Abstract]

Blendy, J.A., Kaestner, K.H., Weinbauer, G.F. et al. (1996) Severe impairment of spermatogenesis in mice lacking the CREM gene. Nature, 380, 162–165.[ISI][Medline]

Bonduelle, M., Aytoz, A., Van Assche, E. et al. (1998) Incidence of chromosomal aberrations in children born after assisted reproduction through intracytoplasmic sperm injection. Hum. Reprod., 13, 781–782.[Free Full Text]

Chandley, A.C. (1998) Genetic contribution to male infertility. Hum. Reprod., 13 (Suppl. 3), 76–83.

Chandley, A.C. and Hargreave, T.B. (1996) Genetic anomalies and ICSI. Hum. Reprod., 11, 930–932.[ISI][Medline]

Cooke, H.J., Hargreave, T. and Elliott, D.J. (1998) Understanding the genes involved in spermatogenesis: a progress report. Fertil. Steril., 69, 989–995.[ISI][Medline]

Colombero, L.T., Hariprahad, J.J., Tsai, M.C. et al. (1999) Incidence of sperm aneuploidy in relation to semen characteristics and assisted reproductive outcome. Fertil. Steril., 72, 90–96.[ISI][Medline]

Damri, L.E., Vutyavanich, T. and Fishel, S. (2000) Comparison of sex chromosome aneuploidy in spermatozoa of fertile men and those requiring ICSI treatment detected by fluorescence in situ hybridization. J. Obstet. Gynaecol. Res., 26, 181–188.[Medline]

Diemer, T. and Desjardins, C. (1999) Developmental and genetic disorders in spermatogenesis. Hum. Reprod. Update, 5, 120–140.[Abstract/Free Full Text]

Dobson, M.J., Pearlman, R.E., Karaiskakis, A. et al. (1994) Synaptonemal complex proteins: occurrence, epitope mapping and chromosome disjunction. J. Cell Sci., 107, 2749–2760.[Abstract/Free Full Text]

Elliott, D.J. and Cooke, H.J. (1997) The molecular genetics of male infertility. BioEssays, 19, 801–809.[ISI][Medline]

Feichtinger, W., Obruca, A. and Brunner, M. (1995) Sex chromosomal abnormalities and intracytoplasmic sperm injection. Lancet, 346, 1566.[ISI][Medline]

Giltay, J.C., Kastrop, P.M.M., Tuerlings, J.H.A.M. et al. (1999) Subfertile men with constitutive chromosome abnormalities do not necessarily refrain from intracytoplasmic sperm injection treatment: a follow-up study on 75 Dutch patients. Hum. Reprod., 14, 318–320.[Abstract/Free Full Text]

Givens, C.R. (2000) Intracytoplasmic sperm injection: what are the risks? Obstet. Gynecol. Surv., 55, 58–62.[ISI][Medline]

Hassold, T.J., (1998) Nondisjunction in the human male. Curr. Top. Dev. Biol., 37, 383–406.[ISI][Medline]

Huang, W.J., Lamb, D.J., Kim, E.D. et al. (1999) Germ-cell nondisjunction in testes biopsies of men with idiopathic infertility. Am. J. Hum. Genet., 64, 1638–1645.[ISI][Medline]

In't Veld, P.A., Brandenburg, H., Verhoeff, A. et al. (1995) Sex chromosomal abnormalities and intracytoplasmic sperm injection. Lancet, 346, 773.[ISI][Medline]

Kurinczuk, J.J. and Bower, C. (1997) Birth defects in infants conceived by intracytoplasmic sperm injection: an alternative interpretation. Br. Med. J., 315, 1260–1265.[Abstract/Free Full Text]

Lange, R., Krause, W. and Engel, W. (1997) Analysis of meiotic chromosomes in testicular biopsies of infertile patients. Hum. Reprod., 12, 2154–2158.[Abstract]

Liliford, R., Jones, A.M., Bishop, D.T. et al. (1994) Case–control study of whether subfertility in men is familial. Br. Med. J. 309, 570–573.[Abstract/Free Full Text]

Mansour, R.T., Kamal, A., Fahmy, I. et al. (1997) Intracytoplasmic sperm injection in obstructive and non-obstructive azoospermia. Hum. Reprod., 12, 1974–1979.[Abstract]

Martin, R.H., Greene, C., Rademaker, A. et al. (2000) Chromosome analysis of spermatozoa extracted from testes of men with non-obstructive azoospermia. Hum. Reprod., 15, 1121–1124.[Abstract/Free Full Text]

McLachlan, R.I., Mallidis, C., Ma, K. et al. (1998) Genetic disorders and spermatogenesis. Reprod. Fertil. Dev., 10, 97–104.[ISI][Medline]

Moens, P.B. and Earnshaw, W.C. (1989) Anti-topoisomerase II recognizes meiotic chromosome cores. Chromosoma, 98, 317–322.[ISI][Medline]

Moens, P.B., Heyting, C., Dietrich, A.J.J. et al. (1987) Synaptonemal complex antigen localization and conservation. J. Cell Biol., 105, 93–103.[Abstract]

Mroz, K., Hassold, T.J. and Hunt, P.A. (1998) Meiotic aneuploidy in the XXY mouse: evidence that a compromised testicular environment increases the incidence of meiotic errors. Hum. Reprod., 14, 1151–1156.[Abstract/Free Full Text]

Munné, S., Marquez, C., Reing, A. et al. (1998) Chromosome abnormalities in embryos obtained after conventional in vitro fertilization and intracytoplasmic sperm injection. Fertil. Steril., 69, 904–908.[ISI][Medline]

Nantel, F., Monaco, L., Foulkes, N.S. et al. (1996) Spermiogenesis deficiency and germ-cell apoptosis in CREM-mutant mice. Nature, 380, 159–162.[ISI][Medline]

Ohkura, H., Kinoshita, N., Miyatani, S. et al. (1989) The fission yeast dis2+ gene required for chromosome disjoining encodes one of two putative type 1 protein phosphatases. Cell, 57, 997–1007.[ISI][Medline]

Okabe, M., Ikawa, M. and Ashkenas, J. (1998) Male infertility and the genetics of spermatogenesis. Am. J. Hum. Genet.62, 1274–1281.[ISI][Medline]

Oliva, R., Margarit, E., Ballesca, J. et al. (1998) Prevalence of Y chromosome microdeletions in oligospermic and azoospermic candidates for intracytoplasmic sperm injection. Fertil. Steril., 70, 506–510.[ISI][Medline]

Pfeffer, J., Pang, M., Hoegerman, S.F. et al. (1999) Aneuploidy frequencies in semen fractions from ten oligoasthenoteratozoospermic patients donating sperm for intracytoplasmic sperm injection. Fertil. Steril., 72, 472–478.[ISI][Medline]

Prapas, Y., Chatziparasidou, A. and Vanderzwalmen, P. (1999) Reconsidering spermatid injection. Hum. Reprod., 14, 2186–2188.[Free Full Text]

Rives, N., Saint Clair, A., Mazurier, S. et al. (1999) Relationship between clinical phenotype, semen parameters and aneuploidy frequency in sperm nuclei of 50 fertile males. Hum. Genet., 105, 266–272.[ISI][Medline]

Shi, Q. and Martin, R.H. (2000) Aneuploidy in human sperm: a review of the frequency and distribution of aneuploidy, effects of donor age and lifestyle factors. Cytogenet. Cell. Genet., 90, 219–226.[ISI][Medline]

Sofikitis, N., Miyagawa, I., Yamamoto, Y. et al. (1998) Micro- and macro-consequences of ooplasmic injections of early haploid male gametes. Hum. Reprod. Update, 4, 197–212.[Abstract/Free Full Text]

Sousa, M., Barros, A. and Tesarik, J. (1998) Current problems with spermatid conception. Hum. Reprod., 13, 255–258.[Free Full Text]

Tu, J., Song, W. and Carlson, M. (1996) Protein phosphatase type 1 interacts with proteins required for meiosis and other cellular processes in Saccharomyces cerevisiae. Mol. Cell. Biol., 16, 4199–4206.[Abstract]

Tuerlings, J.H., Kremer, J.A. and Meulman, E.J. (1997) The practical application of genetics in the male infertility clinic. J. Androl., 18, 576–581.[Free Full Text]

Ushijima, C., Kumasako, Y., Kihaile, P.E. et al. (2000) Analysis of chromosomal abnormalities in human spermatozoa using multi-colour fluorescence in-situ hybridization. Hum. Reprod., 15, 1107–1111.[Abstract/Free Full Text]

Van Opstal, D., Los, F.J., Ramlakhan, S. et al. (1997) Determination of the parent of origin in nine cases of prenatally detected chromosome aberrations found after intracytoplasmic sperm injection. Hum. Reprod., 12, 682–686.[Abstract]

Varmuza, S., Jurisicova, A., Okano, K. et al. (1999) Spermiogenesis is impaired in mice bearing a targeted mutation in the Protein Phosphatase 1c{gamma} gene. Dev. Biol., 205, 98–110.[ISI][Medline]

Zech, H., Vanderswalmen, P., Prapas, Y. et al. (2000) Congenital malformations after intracytoplasmic injection of spermatids. Hum. Reprod., 15, 969–971.[Abstract/Free Full Text]

Submitted on June 6, 2001; accepted on November 2, 2001.