Human Reproduction Update Advance Access originally published online on April 29, 2005
Human Reproduction Update 2005 11(4):337-349; doi:10.1093/humupd/dmi011
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Spermatozoal nuclear determinants of reproductive outcome: implications for ART
Department of Obstetrics, Gynecology and Reproductive Sciences, Yale University School of Medicine, New Haven, CT 06520, USA
1 To whom correspondence should be addressed. Email: denny.sakkas{at}yale.edu
 |
Abstract |
|---|
 TOP Abstract IntroductionAneuploidy of paternal originY chromosome microdeletionsEpigenetic factorsDNA Strand breaks in...ConclusionReferences |
|---|
A male factor is implicated in more than 50% of couples treated with IVF. However, neither the routine testing of male fertility potential nor its treatment address the specific mechanisms by which spermatozoal factors may impact upon reproductive outcome. An important function of spermatozoa is to deliver the paternal genome to the oocyte. Recently, a number of acquired spermatozoal nuclear factors that may have implications on reproductive outcome have been described. These include non-specific DNA strand breaks, numerical abnormalities in spermatozoal chromosome content, Y chromosome microdeletions and alterations in the epigenetic regulation of paternal genome. The exact mechanisms by which these factors affect reproduction are unknown and their implications for assisted reproduction technology outcome need to be further investigated. These recent findings point to the need for novel and more personalized approaches to test and treat male factor infertility.
Key words: epigenetic regulation / non-specific DNA strand breaks / reproductive outcome / sperm chromosome abnormalities / Y chromosome microdeletions
|
Introduction |
|---|
|
TOPAbstractIntroductionAneuploidy of paternal originY chromosome microdeletionsEpigenetic factorsDNA Strand breaks in...ConclusionReferences |
|---|
Infertility has been commonly defined as the inability to conceive after 12 months of regular intercourse, in the absence of contraceptives. Based on this definition, the prevalence of infertility has been calculated to be 10–15% and a multicentre study conducted by the World Health Organization (WHO) (WHO, 1987
) concluded that in 20% of infertile couples, the predominant cause of infertility is the male factor, while in another 27% anomalies in both partners contribute. In the same study 38% of couples were found to have infertility that originates predominantly from the female partner, while 15% were unexplained. Similarly, in 2001, 53.4% of the IVF cycles completed in the USA, were diagnosed with male factor infertility either as a single (19%) or combined (18.2%) diagnosis (www.cdc.gov/reproductivehealth/ART 01). While these statistics underlie the importance of male factor in reproduction, the clinical and analytical methodology used to diagnose male infertility has pitfalls (Jequier, 2004).
Semen analysis is routinely used to evaluate the male partner of the infertile couple. Although widely used thresholds for normal semen measurements have been published by the WHO (WHO, 1999), the current norms for sperm concentration, motility and morphology fail to meet rigorous clinical, technical and statistical standards. In recognition of these limitations, the terminology in the WHO manual for semen evaluation was changed from ‘normal’ to ‘reference’ values (WHO, 1999). Three recent prospective trials concluded that the current WHO criteria should be reconsidered (Bonde et al., 1998; Zinaman et al., 2000; Guzick et al., 2001). In parallel with these observations, many clinical trials use proven fertile men as controls rather than normozoospermic men identified using WHO criteria.
In this review, we will summarize the evidence suggesting that the influence of the paternal genome on reproduction goes beyond that which can be appreciated by simple quantitative and morphologic evaluation of spermatozoa. Indeed, problems that arise at different levels of spermatozoal genomic organization may impact upon reproductive function (Figure 1). Moreover, simple circumvention of fertilization using ICSI may not overcome all the deleterious effects arising from imperfect spermatozoal DNA, and other reproductive parameters such as embryo development, implantation, pregnancy loss and live birth rate may be affected. Therefore, it is necessary to identify molecular and cellular mechanisms that cause impaired spermatozoal function in order to develop personalized diagnostic and therapeutic interventions.
|
|
Aneuploidy of paternal origin |
|---|
|
TOPAbstractIntroductionAneuploidy of paternal originY chromosome microdeletionsEpigenetic factorsDNA Strand breaks in...ConclusionReferences |
|---|
Paternal origin of aneuploidy in human
Sperm chromosomal analysis has initially been investigated using the zona-free hamster ova penetration test (Kamiguchi and Mikamo, 1986), a costly, labour-intensive method that requires sperm-fertilizing ability. Although still costly, and labour-intensive, introduction of fluorescence in-situ hybridization (FISH) made possible the cytogenetic analysis of a larger number of spermatozoa in a shorter period of time without the need for sperm-fertilizing ability. While maternal metaphase I (MI) errors are the predominant etiology (Hassold et al., 1996) based on FISH studies, paternal errors account for 5–10% of autosomal trisomies. Paternal effect on sex chromosome trisomies is higher since 100% of 47,XYY, and nearly 50% of 47,XXY are paternal in origin (MacDonald et al., 1994; Hassold et al., 1996). Seventy to eighty per cent of 45,X have single maternally derived X chromosome and the paternal chromosome (X or Y) is lost either in meiosis or in an early stage of embryo development (Jacobs et al., 1997).
Klinefelter's syndrome occurs in 1 in 500 newborn males and is the most common chromosomal disorder associated with male infertility. A 47,XXY chromosome complement is present in approximately 95% of cases. The 47,XXY karyotype has a high incidence among infertile men, detected in 11% of men with azoospermia and in 0.7% of men with oligozoospermia. The ability of the germ cells from a 47,XXY male to proceed through meiosis and to generate XX or XY hyperhaploid gametes is not known. While the presence of more than one X chromosome in a male germ cell is believed to be detrimental, studies using FISH in spermatozoa of mosaic Klinefelter's patients with 47,XXY, 46,XY karyotype suggest that this may not be the case (Cozzi et al., 1994; Chevret et al., 1996). An increased frequency of hyperhaploid 24,XY in these patients argues that such cells may indeed possess a meiotic capacity. Meanwhile, an elevated rate of meiotic non-disjunction is yet to be ruled out.
Incidence of sperm chromosome aneuploidy and implications on reproduction
Hansen et al. (2002) reported compiled data from the registries in Western Australia, involving 301 infants conceived with ICSI, 837 infants conceived with IVF and 4000 naturally conceived controls between 1993 and 1997. They found the incidence of major birth defects to be more than 2-fold higher for ICSI and IVF groups (8.6 and 9%, respectively) compared to normal controls (4.2%). Their data show an increased incidence of chromosomal abnormalities in the ICSI group (1% for all infants and 1.6% for singletons only) compared to IVF (0.7% for all infants and 0.6% for singletons only; the difference not statistically significant) and normal controls (0.2% for all infants and 0.2% for singletons only; P<0.05).
While some other studies failed to show an increase in major birth defects in children conceived with IVF and/or ICSI (Van Steirteghem, 1998), these were criticized for methodologic problems such as inadequate sample sizes, lack of appropriate data for comparison, use of different criteria to define major birth defects in infants conceived with assisted reproduction technology (ART) and those conceived naturally (Kurinczuk and Bower, 1997).
Van Steirteghem et al. (2002) summarized data from seven studies reporting karyotype analyses performed for prenatal diagnosis in a total of 2139 pregnancies conceived with ICSI. In comparison with the general population, they calculated a slight but significant increase in de novo sex chromosomal aneuploidy (0.6 versus 0.2%), and structural autosomal abnormalities (0.4 versus 0.07%), and an increased number of inherited (mostly from the father) structural aberrations.
Possible causes of elevated chromosome aneuploidy following ICSI include the technique itself, or an increased risk from the sperm used for ICSI. Carrell et al. (2004) found the mean aneuploidy rate for five chromosomes (X, Y, 13, 18, 21) to be 1.2% for fertile controls, 1.4% for the general population, 2.6% for men with moderately diminished semen quality (count, motility, or morphology were diminished but greater than 50% of normal value), 4% for men with severe teratoasthenooligozoospermia (count, motility and morphology were all less than 50% of normal value) and 15.9% for men with rare ultrastructural defects. Their findings are consistent with previous reports and suggest that the incidence of chromosome aneuploidy in spermatozoa relates with the severity of sperm defects. The same group also found the sperm chromosome aneuploidy rate to be higher in partners of women with recurrent pregnancy loss (2.77%) compared to the general population (1.48%) or fertile controls (1.19%) (Carrell et al., 2003). This is consistent with a previous report by Rubio et al. (1999). More recently, studies using PGD in embryos derived from men with meiotic abnormalities reported an elevated incidence of chromosomal abnormalities (Aran et al., 2004; Platteau et al., 2004).
In summary, available evidence suggest a relationship between abnormal semen parameters and an increased incidence of sperm chromosomal aneuploidy as well as chromosomal abnormalities in the embryo. Moreover, there may be a relationship between recurrent pregnancy loss and elevated sperm chromosomal aneuploidy. However, the relevance of these findings, beyond their use in counselling is currently unclear, because even in men with increased sperm chromosomal aneuploidy, only less than 5% of spermatozoa seem to have aneuploid chromosome content.
At present we are unable to offer our patients a selection technique whereby spermatozoa with normal chromosome number can be used for fertilization. Very recently, Jakab et al. (2005) reported a new method based on the hyaluronic acid binding ability of spermatozoa. Using this method, they were able to simulate ICSI techniques and select a spermatozoa population with a 4–5-fold decreased frequency of disomy or diploidy compared to unselected spermatozoa. While further validation is necessary, their findings are encouraging in providing a test that allows selection of spermatozoa that may be used in fertility treatment.
|
Y chromosome microdeletions |
|---|
|
TOPAbstractIntroductionAneuploidy of paternal originY chromosome microdeletionsEpigenetic factorsDNA Strand breaks in...ConclusionReferences |
|---|
The human Y chromosome consists of a short and a long arm, termed Yp and Yq, respectively (Figure 2). The areas of sequence identity to the X chromosome that allows pairing and recombination during meiosis, called pseudoautosomal pairing regions (PARs), are located at the distal portions of Yp and Yq. Because of the absence of meiotic recombination within most regions of the human Y chromosome impeding linkage analysis, mapping has been based on naturally occurring deletions.
|
Y chromosome microdeletions are the second most frequent genetic cause of male infertility after Klinefelter's syndrome. Almost 30 years ago, Tiepolo and Zuffardi (1976)
were first to report microscopically detectable deletions in the distal portion of Yq in six azoospermic men. Their findings suggested the presence of genes regulating spermatogenesis in the distal portion of Yq later referred to as the azoospermia factor (AZF). More recent studies of men with non-obstructive severe oligozoospermia or azoospermia (Vogt et al., 1992; Ma et al., 1993; Vogt et al., 1996) suggested the existence of a multitude of loci within Yq that may be associated with infertility, and lead to the characterization of three regions required for spermatogenesis referred to as AZFa, AZFb and AZFc (Figure 2) (Vogt et al., 1996; Kuroda-Kawaguchi et al., 2001; Repping et al., 2002). A fourth AZF subregion, positioned between AZFb and AZFc has also been proposed and named AZFd (Kent-First et al., 1999).
Studies of Y chromosome deletions have also demonstrated that, although the genes within the non-recombining portion play an important role in male fertility, their loss is not exclusively linked with azoospermia (Reijo et al., 1996). Indeed, microdeletions within the AZF regions have been associated with a multitude of clinical findings including Sertoli cell-only syndrome, spermatogenic arrest and morphological abnormalities of post-meiotic germ cells.
Naturally occurring infertility-associated deletions may be restricted to any one of the AZF regions or they may extend beyond a given region to encompass multiple subregions of AZF. In a recent review, Simoni et al. (2004) described four clinically relevant Y chromosome microdeletions found in men with oligo- or azoospermia: AZFa, AZFb, AZFbc and AZFc. The most frequently deleted region is AZFc (approximately 60% of cases), followed by deletions of the AZFb and AZFbc or AZFabc regions (35%), while isolated AZFa deletions are rare (5%) (Krausz et al., 2003). Other types of deletions were described (Fernandes et al., 2002; Repping et al., 2003); although their clinical significance has yet to be explored. Isolated gene deletions were only reported by a single group (Ferlin et al., 1999; Foresta et al., 2000) for AZFa genes. This has not yet been confirmed in more than 1000 infertile males tested by different groups (Krausz et al., 2003).
Phenotypes associated with Y chromosome microdeletions
Complete deletions removing the entire AZFa, or AZFb regions are associated with Sertoli cell-only syndrome and spermatogenic arrest, respectively (Krausz et al., 2000). Partial deletions of AZFa or AZFb regions, or complete or partial deletions of the AZFc region are associated with a variable phenotype ranging from oligozoospermia to Sertoli cell-only syndrome. The broad range of phenotypes observed is not surprising in that several potentially functional genes or gene families lie in each AZF region. These include USP9Y (ubiquitin-specific protease 9, Y chromosome; initially identified as DFFRY, Drosophila fat facets related Y), and DBY (dead box on the Y) single genes in AZFa, RBM (RNA binding motif) gene family in AZFb, and DAZ (deleted in azoospermia) gene family in AZFc (Kent-First, 2000). Another explanation for the variable phenotype is the progressive regression of the germinal epithelium reported in patients with AZFc microdeletions (Warchol et al., 2000; Colegero et al., 2001), although the findings of Oates et al. (2002) does not support this hypothesis.
Y chromosome microdeletions in fertile and infertile man
The pathogenic role of Y chromosome microdeletions in male infertility has been questioned by reports of naturally transmitted deletions removing the entire AZFc region (Vogt et al., 1996; Pryor et al., 1997; Chang et al., 1999; Saut et al., 2000). Sperm analysis was available in only two cases. Two men, one with reduced and the other with unknown sperm count, were able to father a single child (Vogt et al., 1996; Pryor et al., 1997), whereas the father of four infertile sons was found to be azoospermic many years after the natural conception of his sons (Chang et al., 1999).
Most studies investigating Y chromosome microdeletions, used ‘proven fertile men’ as controls, rather than ‘normozoospermic men’. Krausz and McElreavey summarized the results of 26 studies reported between 1995 and 2001 and calculated that only 12 of 2295 proven fertile men, and none of 392 normospermic men had Y chromosome microdeletions (Krausz and McElreavey, 2001). In their elegant discussion, they pointed to the fact that the fertility potential of a man also depends on his partner's fertility potential, and that fertility is not a synonym of normozoospermia (Krausz and McElreavey, 2001). Indeed, the 12 fertile men with Y chromosome microdeletions reported in two studies (Pryor et al., 1997; Kent-First et al., 1999) could have been oligozoospermic. In summary, Y chromosome microdeletions have been found almost exclusively in patients with less than 1x106 spermatozoa/ml, and are extremely rare (approximately 0.7%) among men with sperm concentrations greater than 5x106 spermatozoa/ml (Krausz et al., 2003), arguing for an important role for Y chromosome microdeletions in spermatogenic failure.
The incidence of Y chromosome microdeletions reported in infertile men varies between 1 (van der Ven et al., 1997) and 55% (Foresta et al., 1998). This variability may be caused by several factors including patient selection criteria, lack of rigorous testing of negative results and ethnic variances among study populations (Krausz et al., 2003).
Y chromosome microdeletions and ART outcome
The introduction of ICSI combined with sperm retrieval techniques such as testicular sperm extraction (TESE), now allows men with Y chromosome microdeletions to reproduce. Mulhall et al. (1997) compared eight azoospermic men with AZFc deletions with 28 controls with normal Y chromosomes. All patients were treated with TESE and subsequent ICSI. While fertilization rates seemed to be lower in the AZFc deletion group compared to controls (36 versus 45%), there was no statistically significant difference. Pregnancy rates did not differ between the two groups.
In a subsequent study, van Golde et al. (2001) retrospectively compared the success rate of 19 ICSI treatments in eight couples with AZFc microdeletions to a control group of 239 ICSI treatments in 107 couples. Ejaculated spermatozoa were used in both study groups. Although they found significantly lower fertilization rates (55 versus 71%, P<0.01) and embryo scores in couples with AZFc microdeletions, overall pregnancy rates did not differ.
Oates et al. (2002) evaluated 713 men with non-obstructive azospermia or severe oligospermia and identified 42 (5.9%) with microdeletions strictly confined to the AZFc region. These were classified into four subgroups based upon spermatogenic capability: severe oligospermia (16 men); azoospermia with sperm detected on TESE or quantitative histological analysis (14 men); azoospermia with no sperm detected on TESE or quantitative histological analysis (seven men); azoospermia but no TESE or quantitative histological analysis was performed, leaving unanswered the question of whether testicular sperm might be present (five men). A total of 48 cycles of ICSI were performed in 26 of these couples. The overall fertilization rate was 47% and the overall term pregnancy rate was 27%. The fertilization and term pregnancy rates for those cycles in which ejaculated sperm served as the gamete source were 64 and 47%, respectively. Fertilization rate and term pregnancy rate for the group who had testicular sperm used were 36 and 14%, respectively. Comparing fertilization rates from the ejaculated sperm group with the testicular sperm group revealed a statistically significant difference (P<0.0001).
Most recently, Choi et al. (2004) reported their experience with 17 men with different types of Y chromosome microdeletions. Consistent with previous reports, they were unable to obtain spermatozoa with TESE in men with complete deletion of AZFa or AZFb or AZFbc regions. Spermatozoa were obtained from men with AZFc deficiencies and one with partial AZFb deficiency. Patients with Y chromosome microdeletions were studied in two groups depending on whether TESE or ejaculated spermatozoa was used. These were compared to matched controls. Although there was a tendency towards decreased fertilization and pregnancy rates, the differences were not statistically significant.
Overall, studies of ART outcome in patients with AZFc deletions suggest a tendency toward decreased fertilization rates but not a significant change in overall pregnancy and delivery rates compared to matched controls (Mulhall et al., 1997; van Golde et al., 2001; Oates et al., 2002; Choi et al., 2004). Men with ejaculated spermatozoa seem to do significantly better that those who need TESE (Oates et al., 2002; Choi et al., 2004).
Indications for Y chromosome microdeletion testing in a clinical setting and the need for standardization
As summarized above, clinically relevant deletions are found in patients with azoospermia or a sperm concentration of less than 1x106 spermatozoa/ml, and very rarely, in patients with sperm concentration between 1 and 5x106 spermatozoa/ml (Simoni et al., 2004). Although the incidence of microdeletions is higher when patients are selected by testicular histology (e.g. Sertoli cell-only syndrome), no absolute criteria exist for the selection of patients for molecular analysis. In general, Y chromosome testing is not indicated for patients with chromosomal abnormalities, obstructive azoospermia, or hypogonadotropic hypogonadism (Simoni et al., 2004).
Patients with azoospermia or severe oligospermia who may be candidates for TESE combined with ICSI, should be offered Y chromosome testing because TESE may not be beneficial in cases of complete AZFa or AZFb or AZFbc deletions (Simoni et al., 2004). In addition, microdeletions of the AZFc region are transmitted to the male offspring if ART is performed (Page et al., 1999; Oates et al., 2002). Therefore, information obtained from Y chromosome testing may be used in genetic counselling of patients contemplating ART (Simoni et al., 2004).
With the aim of offering an additional diagnostic tool for male infertility, many laboratories across the world provide PCR-based deletion analysis using in-house methods. This occurred in the absence of a consensus on methodology. The European Academy of Andrology and European Molecular Genetics Quality Network supported studies evaluating methodology used in different laboratories. These studies formed the basis of the 1999 Laboratory guidelines for molecular diagnosis of Y chromosomal microdeletions (Simoni et al., 1999) that were updated recently following the best practice meeting held in Florence, Italy in October 2003 (Simoni et al., 2004). We hope that similar regulatory approaches will be accepted by other countries worldwide.
|
Epigenetic factors |
|---|
|
TOPAbstractIntroductionAneuploidy of paternal originY chromosome microdeletionsEpigenetic factorsDNA Strand breaks in...ConclusionReferences |
|---|
Epigenetics refers to covalent modifications of DNA or core histones that regulate gene expression without affecting DNA sequence (Figure 1). Currently, methylation of cytosine residues within CpG dinucleotides, called DNA methylation, constitutes the best understood mechanism by which gene activity is modulated by epigenetic factors. CpG methylation, especially within the promoter region of genes, is associated with repression of transcription and provides a means to control gene expression. DNA methylation has been implicated in the regulation of a number of functions including allele-specific gene expression also called genomic imprinting and X chromosome inactivation (Bestor, 2000
). Human disorders associated with epigenetic abnormalities include rare imprinting diseases, molar pregnancies and childhood cancers.
Germ cell development and early embryo development are critical times when epigenetic patterns are initiated or maintained (Lucifero et al., 2004). Striking modulations in methylation during gametogenesis and embryogenesis occur for imprinted genes that are expressed from only the maternal or the paternal genome. DNA methylation is the best-studied epigenetic mark that distinguishes the maternal and paternal alleles of imprinted genes. Most imprinted genes contain sequences that are differentially methylated in the gametes, and in most cases, the two parental alleles have different levels of DNA methylation. As a heritable, reversible, epigenetic mark, DNA methylation of imprinted genes can be stably propagated after DNA replication and can maintain monoallelic gene expression throughout life.
DNA methylation in gametes and embryos
Early primordial germ cells (PGCs) are believed to carry somatic epigenetic patterns that are erased prior to and soon after their entry into the future gonads. In the mouse, erasure of DNA methylation in PGCs occurs around the time when these cells enter the gonad (Szabo and Mann, 1995; Szabo et al., 2002) and results in a transformation from monoallelic to biallelic expression of imprinted genes (Szabo et al., 2002). After an almost genome-wide demethylation in PGCs, sex- and sequence-specific methylation is re-established in the male and female gametes (Reik et al., 2001). Although a second phase of genomic demethylation has been described in the preimplantation embryo, some sequences, especially imprinted genes seem to retain their methylation status acquired at the gamete stage. It has been proposed that the gamete methylation that is retained during the wave of preimplantation demethylation may be important for embryo development following implantation.
Methylation at CpG dinucleotides is catalysed by DNA (cytosine-5)-methyltransferase (DNMT) enzymes. DNMT1 is the predominant mammalian DNMT, while DNMT2, DNMT3a, DNMT3b and DNMT3L have also been characterized. Currently, the only known catalytically active DNMTs are DNMT1, DNMT3a and DNMT3b (Kelly and Trasler, 2004). The expression of these three enzymes in both female and male germlines seems to be tightly regulated (La Salle et al., 2004). The functions of these enzymes have been investigated using a knockout approach. However, mice homozygous for targeted deletions in Dnmt1 (Li et al., 1992) and Dnmt3b (Okano et al., 1999) are embryonic lethal. Therefore, determining the role of DNMT1 and DNMT3b during spermatogenesis will require germ cell-specific knockout or knock-down studies. Dnmt3a-deficient mice are not embryonic lethal although they are underdeveloped, and die 3–4 weeks after birth. These mice show defects in spermatogenesis (Okano et al., 1999).
In the mouse and the rat, DNA methylation in male germ cells may be disrupted by chronic administration of cytosine analogues 5-azacytidine and 5-aza-2'-deoxycytidine. This causes severe adverse effects including decreased sperm counts, decreased fertility and increased preimplantation loss (Doerksen and Trasler, 1996; Doerksen et al., 2000; Kelly et al., 2003). These findings argue for an important role for epigenetic mechanisms in the control of gamete and embryo development.
Implications of altered gamete and early embryo DNA methylation
Gestational trophoblastic disease
One type of molar pregnancy, named complete hydatiform mole is classically described as carrying two paternal haploid genomes. Interestingly, recent studies report complete hydatiform moles arising from normal fertilization events (uniting a maternal and a paternal haploid genome) where the maternal alleles carry paternal imprints or the paternal epigenotype (Helwani et al., 1999; Moglabey et al., 1999; Fisher et al., 2000). Hayward et al. (2003) failed to detect mutations in the known DNMTs that might account for the failure of methylation in the female germline. Their findings suggest that maternal imprinting during oogenesis may involve additional factors.
Imprinting diseases and ART
Animal studies suggest that epigenetic marks, especially DNA methylation, are unstable and can be altered by culture conditions (Doherty et al., 2000; Young et al., 2001). ART rely on manipulation and culture of gametes and embryos at times when epigenetic programmes are being acquired and modified. An initial study found no evidence of altered methylation at 15q11–13, the locus linked to the pathogenesis of the imprinting disorders Angelman and Prader–Willi syndromes, in samples from 92 children conceived using ICSI (Manning et al., 2000). However, recently six studies have reported two imprinting disorders, Beckwith–Wiedemann syndrome (DeBaun et al., 2003; Gicquel et al., 2003; Maher et al., 2003; Halliday et al., 2004) and Angelman syndrome (Cox et al., 2002; Orstavik et al., 2003) in association with ART.
Angelman syndrome has an incidence of approximately 1 in 15 000, and is characterized by mental retardation, ataxia, epilepsy, hypotonia and absence of speech. It results from a loss of function of the maternal allele or duplication of the paternal allele within a region of 15q11–13. In approximately 5% of affected children, the syndrome is caused by loss of methylation within the SNRPN imprinting region, which then assumes a hypomethylated, paternal profile. Three children with Angelman syndrome, conceived using ICSI have been reported (Cox et al., 2002; Orstavik et al., 2003). In all three cases, a loss of methylation on the maternal SNRPN allele was found.
Beckwith–Wiedemann syndrome is characterized by prenatal or post-natal overgrowth, macroglossia, abdominal wall defects, neonatal hypoglycaemia, hemihypertrophy, ear abnormalities and an increased risk of embryonal tumours. Analysis of Beckwith–Wiedemann syndrome registries from three centres has shown the proportion of affected progeny conceived with IVF to be 3/65 (DeBaun et al., 2003), 6/149 (Maher et al., 2003) and 6/149 (Gicquel et al., 2003). These data suggest that overall 4% of individuals with Beckwith–Wiedemann syndrome registered in these centres were conceived using IVF, a figure greater than that expected. Further interpretation of these results has been limited due to methodological limitations of these studies.
More recently, Halliday et al. (2004) reported the first case–control study in an Australian population. Among 1 316 500 live births in Victoria between 1983 and 2003 they identified 37 cases of Beckwith–Wiedemann syndrome. For each Beckwith–Wiedemann syndrome case, they randomly selected four live-born controls. IVF was the method of conception in four Beckwith–Wiedemann syndrome cases and in one control. Their results indicated that if a child has Beckwith–Wiedemann syndrome, the odds that the child was conceived using IVF was 18 times greater than that for a child without Beckwith–Wiedemann syndrome. The calculated risk of Beckwith–Wiedemann syndrome in the IVF population was 1/4000, or nine times greater than the general population. Although this study has shortcomings including a large confidence interval, its results are concerning.
Beckwith–Wiedemann syndrome is linked to a loss of function of the maternal allele at 11p15 (Weksberg et al., 2003). Two epigenetic DNA methylation defects have been associated with Beckwith–Wiedemann syndrome. The most common one involves loss of methylation at the maternal KVDMR1/LIT1 (46% of cases) (Halliday et al., 2004), and the other a gain of methylation at H19 DMD on the maternal allele such that it assumes a paternal (methylated) profile (7% of cases). The rest is due to uniparental disomy of chromosome 11 (16%) and to an unidentified mutation (31%) (Halliday et al., 2004).
A total of 15 cases with Beckwith–Wiedemann syndrome were evaluated for methylation defects in the four studies. Of those, 14 showed hypomethylation of maternal KVDMR1/LIT1. The preponderance of Beckwith–Wiedemann syndrome cases conceived by IVF that show hypomethylation of maternal KVDMR1/LIT1 suggests that collection or in vitro culture conditions may disturb the methylation in the oocyte or the embryo, predisposing the maternal allele to demethylation. While spermatozoa do not seem to be the source of altered methylation in the case of Beckwith–Wiedemann and Angelman syndromes, the same mechanisms may be effective in causing altered methylation of spermatozoal genes.
Imprinting in spermatozoa of men with abnormal semen parameters
An initial study by Manning et al. (2001a,b) using PCR-based techniques to analyse DNA extracted from spermatozoa of men with normal semen analysis (n=30) and from men with medium (n=30, 5–20x106 spermatozoa/ml) and high grade semen pathology (n=30, <5x106 spermatozoa/ml) undergoing ICSI, failed to detect a difference in methylation status. More recently, Marques et al. studied two oppositely imprinted genes in spermatozoan DNA from normozoospermic and oligozoospermic patients. In the mesodermal specific transcript gene (MEST), bisulphite genomic sequencing showed that maternal imprinting was correctly erased in all 123 patients. However, methylation of the H19 gene did not change in any of 27 normozoospermic individuals (0%) compared with methylation changes in eight moderate (17%, P=0.026) and 15 severe (30%, P=0.002) oligozoospermic patients. Their findings suggest an association between abnormal genomic imprinting and hypospermatogenesis, and that spermatozoa from oligozoospermic patients carry a raised risk of transmitting imprinting errors (Marques et al., 2004).
TESE and ICSI (TESE–ICSI) is a frequently used therapeutic option in azoospermic males. Manning et al. used a hemi-nested methylation-specific PCR for the target region 15q11–13 to analyse imprinting at the single-cell level in cells from different stages of spermatogenesis. They analysed spermatozoa, elongated spermatids and round spermatids and found completed establishment of the correct paternal imprint in these three developmental stages. However, the high rate of amplification failure in round spermatids in this study is a factor of uncertainty (Manning et al., 2001a,b).
It should be emphasized that our knowledge on the epigenetic regulation of gene expression is in its infancy. In addition to DNA methylation, other mechanisms including covalent modification of core histones seems to be involved. The role of these different epigenetic regulatory pathways in controlling gene expression in gametes and embryos, as well as their modulation by ART remains to be investigated.
|
DNA Strand breaks in spermatozoa |
|---|
|
TOPAbstractIntroductionAneuploidy of paternal originY chromosome microdeletionsEpigenetic factorsDNA Strand breaks in...ConclusionReferences |
|---|
Apoptosis as a regulator of spermatogenesis
Spermatogenesis is a complex process of proliferation and differentiation transforming spermatogonia into mature spermatozoa. This unique process involves a series of mitoses and a meiotic division followed by marked changes in cell structure. In addition to proliferation and differentiation, the outcome of spermatogenesis is affected by the extent of germ cell death. During spermatogenesis, germ cell death occurs normally and continuously, and is estimated to result in the loss of up to 75% of the potential number of spermatozoa (Huckins, 1978).
Germ cell death during mammalian spermatogenesis occurs mainly via apoptosis (Blanco-Rodriguez and Martinez-Garcia, 1996, 1998; Rodriguez et al., 1997). Evidence from animal models show that the active proliferation of early germ cells is balanced by selective apoptosis of their progeny (Wang et al., 1998). As such, testicular germ cell apoptosis seems to occur physiologically and continuously throughout life. Additional etiologic factors have been found to cause testicular germ cell apoptosis. These include gonadotropin withdrawal (Hikim et al., 1995), cryptorchidism (Shikone et al., 1994), irradiation (Henriksen et al., 1996), heat exposure (Yin et al., 1997) and vasectomy (Lue et al., 1997).
Apoptosis may play two roles during normal spermatogenesis. Logically the first is limitation of the germ cell population to numbers that can be supported by the Sertoli cells; however, we have proposed that it plays a secondary role in the selective depletion of abnormal spermatozoa.
In mouse models, numerous pro- and anti-apoptotic proteins have been found to play key roles in spermatogenesis. The Bcl-2 family includes both pro-survival and pro-apoptotic members, and provides a signalling pathway that seems to be necessary for maintaining male germ cell homeostasis (Huynh et al., 2002). Bcl-2 and Bcl-xL are pro-survival members of the Bcl-2 family. Transgenic overexpression of Bcl-2 or Bcl-xL results in blockage of cell death at a critical stage, and results in disruption of normal spermatognenesis and infertility in male mice (Furuchi et al., 1996; Rodriguez et al., 1997).
Bax is a multidomain, pro-apoptotic member of the Bcl-2 family. Despite its pro-apoptotic function, Bax-deficient mature male mice demonstrate increased germ cell death and testicular atrophy. This is because, Bax is required for normal developmental germ cell death in the dividing A(2), A(3) and A(4) spermatogonia, at a time when the number of spermatogonia is regulated in a density-dependent manner (Russell et al., 2002). A massive male germ cell hyperplasia initially occurs in Bax-deficient mice, and subsequently results in Bax independent cell death that may be triggered by overcrowding of the seminiferous epithelium (Russell et al., 2002). These studies show that an increase in pro-apoptotic or anti-apoptotic proteins can result in disruption of normal spermatogenesis and suggest that apoptosis plays an important role in male gametogenesis by regulating the size of the spermatogenic cell population. Meanwhile, although it seems plausible, the role of apoptosis in selective depletion of abnormal spermatozoa is yet to be proven.
Apoptosis and DNA strand breaks in spermatozoa
A key indicator of apoptosis is believed to be the presence of DNA strand breaks. Due to the extensive nuclear remodelling undergone by the spermatozoa, DNA strand breaks detected in these cells could also be due to factors unrelated to apoptosis. Indeed, McPherson and Longo (1993b) demonstrated the presence of endogenous DNA strand breaks in elongating rat spermatids, when chromatin structure and nucleoproteins are modified. They proposed (McPherson and Longo, 1993a) that the presence of endogenous nicks in ejaculated spermatozoa might be indicative of incomplete maturation during spermiogenesis. They also postulated (McPherson and Longo, 1992, 1993a,b) that chromatin packaging might involve endogenous nuclease activity in order to create and ligate nicks during the replacement of histones by protamines, and that an endogenous nuclease, topoisomerase II, may play a role. Topoisomerase II functions by transiently introducing DNA double strand breaks, allowing the passage of one double helix through another, and resealing the double strand break (Wang et al., 1990). While the role of topoisomerase II in spermatogenesis is yet to be clarified, it is expressed in human testis (Seli et al., 2003) and the transient presence of DNA breaks has been reported in both mouse and human (Sakkas et al., 1995; Marcon and Boissonneault, 2004).
Although it is debatable whether DNA strand breaks detected in male germ cells are indicative of ongoing apoptosis, tests of DNA integrity are being used to investigate male reproductive function. This approach is justified by the fact that only less than 5% of germ cells in normal testis have DNA strand breaks (Hikim et al., 1998; Seli et al., 2003), and that there is an association between increased number of spermatozoa with DNA strand breaks and impaired male fertility parameters.
Testing for DNA strand breaks in spermatozoa
A test commonly used to detect DNA strand breaks is terminal deoxynucleotidyl transferase [TdT]-mediated dUTP nick end-labelling (TUNEL). TUNEL technique labels single or double-stranded DNA breaks, but does not quantify DNA strand breaks in a given cell. False-positive staining with this technique has been reported in certain tissues due to endogenous endonucleases (Stahelin et al., 1998). The length of time that tissues are left before fixation can have an effect; and 2 h appears to be the limit before spurious results are obtained (Tateyama et al., 1998).
Another test now commercially available to evaluate spermatozoal DNA integrity is the sperm chromatin structure assay (SCSA). The SCSA, is a fluorescence-activated cell sorter test that measures the susceptibility of sperm nuclear DNA to heat- or acid-induced DNA denaturation in situ, followed by staining with acridine orange (Evenson and Jost, 2000; Larson-Cook et al., 2003). Acridine orange is a metachromatic dye that fluoresces red when associated with denatured (fragmented) DNA and green when bound to double-stranded (normal) DNA. Therefore, an increase in the percentage of cells with a high ratio from red to green fluorescence indicates an overall increase in DNA fragmentation in the spermatozoa from that ejaculate. Because the SCSA is a quantitative (on a continuous scale), as opposed to a qualitative measurement, it has the potential to better define thresholds associated with reproductive outcome (Larson-Cook et al., 2003). SCSA parameters relate with DNA strand breaks detected using the TUNEL technique (Sailer et al., 1995; Aravindan et al., 1997). Other tests of sperm nuclear DNA integrity include in situ nick translation (Tomlinson et al., 2001) and the comet assay (Morris et al., 2002).
DNA strand breaks in spermatozoa and reproductive outcome
Higher levels of nuclear DNA strand breaks are detected in the testicular germ cells and ejaculated spermatozoa of men with abnormal semen parameters (Irvine et al., 2000; Seli et al., 2002). At present, it is not known whether this results from an increased induction of DNA strand breaks or a defective apoptotic mechanism, previously labelled by us as abortive apoptosis (Sakkas et al., 1999), failing to complete cell death prior to ejaculation, or both. It is also unknown if there is a difference between different testicular pathologies in their propensity to initiate and complete apoptotic cell death. Whatever the cause of increased DNA strand breaks in the ejaculate, it may affect the quality of the ejaculated spermatozoa and may have implications on reproduction, in particular the outcome of ART where the natural selection during fertilization is by-passed.
Reproductive parameters that could be affected by an increased apoptotic activity in ejaculated spermatozoa include fertilization, blastocyst development and pregnancy rates. Investigation of the possible association between apoptotic activity in spermatozoa and fertilization rates in patients undergoing ART found no relation between DNA integrity of ejaculated spermatozoa and IVF and ICSI fertilization rates using in situ nick translation (Tomlinson et al., 2001), Comet assay (Morris et al., 2002), TUNEL assay (Ahmadi and Ng, 1999), or SCSA (Larson et al., 2000; Larson-Cook et al., 2003). In contrast to these reports, a negative relation between sperm DNA strand breaks and IVF (Sun et al., 1997) and ICSI (Lopes et al., 1998) fertilization rates have been reported, using the TUNEL assay. Activation of embryonic genome expression occurs at the 4–8-cell stage in human embryos (Braude et al., 1988), suggesting that the paternal genome may not be effective until that stage. Therefore, the lack of relation between elevated DNA strand breaks in spermatozoa and fertilization rates is not surprising.
As expected, a negative relation between the extent of nuclear DNA damage in ejaculated spermatozoa and blastocyst development after IVF and ICSI was observed using both the TUNEL assay to evaluate spermatozoa processed for IVF (Figure 3 and Table I) (Seli et al., 2004) and the SCSA to evaluate unprocessed spermatozoa (Virro et al., 2004). In addition, pregnancy rates after IVF are reduced in couples who have higher percentages of spermatozoa with DNA strand breaks detected by in situ nick translation (Tomlinson et al., 2001). Similarly, there is strong evidence for a relationship between sperm nuclear DNA integrity, as assessed with the SCSA and fertility after both normal intercourse (Evenson et al., 1999; Spano et al., 2000) and ART (Larson et al., 2000).
|
|
Interestingly, Evenson et al. (1999)
found cases where the classical criteria (concentration, motility and morphology) were within the normal ranges, but the SCSA values were poor and not compatible with good fertility after intercourse. Indeed, SCSA parameters are not strongly related with World Health Organization (WHO) parameters including concentration, motility and morphology (Evenson et al., 1991). Based on these results Evenson et al. speculated that SCSA parameters may be independent predictors of reproductive outcome beyond WHO parameters. On the other hand, in a recent study, Gardner et al. (2004) did not find a difference in the implantation and pregnancy rates between two groups that had 16 versus 40% fragmentation rates detected by SCSA. This result contradicts previous findings using this technique. As SCSA is performed in the raw semen sample prior to processing, these results may reflect a selective elimination of spermatozoa with DNA fragmentation during sperm preparation for ART. If that is the case, revalidation of SCSA in spermatozoa processed for use in IVF may become necessary.
Finally, Carrell et al. (2003) found that the percentage of sperm with DNA fragmentation detected using the TUNEL assay is significantly increased in men whose wives suffer recurrent pregnancy loss (38±4.2) compared to donor sperm (11.9±1) or the general population (22±2). In the recurrent pregnancy loss group, no relation was observed between semen quality parameters and TUNEL positivity.
Post-testicular changes in spermatozoal DNA integrity and reactive oxygen species
Further to the events impacting on sperm DNA in the testes are that of post-testicular DNA damage and the possible implication of reactive oxygen species (ROS). Greco et al. (2005) recently reported that DNA fragmentation in ejaculated spermatozoa detected by the TUNEL assay is significantly higher compared to that in testicular spermatozoa (23 versus 4.5%). They also found higher pregnancy rates using testicular sperm compared to ejaculated spermatozoa (Greco et al., 2005). Steele et al. (1999) found similar results when they compared epididymal sperm to testicular sperm using the Comet assay. In 20 men with obstructive azoospermia, they found the percentage of undamaged DNA to be significantly higher in testicular spermatozoa compared to spermatozoa retrieved from the proximal epididymis (83 versus 75.4%) (Steele et al., 1999). One of the theories attempting to explain the differences in DNA damage between testicular and epididymal spermatozoa implicates ROS. Gil-Guzman et al. (2001) showed that there is significant cell-to-cell variation in ROS production in subsets of spermatozoa at different stages of maturation. Ollero et al. (2001) suggested that high levels of ROS production and DNA damage observed in immature spermatozoa may be indicative of derangements in the regulation of spermiogenesis and that DNA damage in mature spermatozoa may be the result of oxidative damage by ROS-producing immature spermatozoa during sperm migration from the seminiferous tubules to the epididymis.
Although there are indications that post-testicular events impact on sperm DNA integrity, it is still unclear whether it is a testicular failure to generate normal spermatozoa that makes them more susceptible to the effects of ROS.
The relevance of sperm DNA integrity testing
Animal models show clear indications that sperm DNA integrity impacts on reproductive outcome. In this area, some of the most interesting studies have been those performed by Robaire and colleagues using cyclophosphamide in rodents to induce DNA damage in spermatozoa and examining the impact on embryos, fetuses and future generations (Trasler et al., 1985; Hales et al., 1992; Harrouk et al., 2000; Hales and Robaire, 2001; Robaire and Hales, 2003). The aim in the human is to ascertain whether testing of sperm DNA integrity has predictive value for fertility in general, and in particular ART outcomes (Spano et al., 2005).
Although we are still far from establishing a strong relation in the human as seen in animals, the current data on sperm DNA testing would suggest the following:
- The presence of a high fraction of spermatozoa showing DNA damage is a negative trait that reduces the chances to father a child.
- A specific threshold or percentage of DNA damaged sperm that is incompatible with pregnancy is not yet established (Gandini et al., 2004).
- The ability of the sperm DNA integrity tests to be predictive diminishes when spermatozoa are prepared using techniques such as density gradient centrifugation (Larson et al., 1999; Sakkas et al., 2000; Tomlinson et al., 2001; O'Connell et al., 2003). In light of this, tests such as the SCSA, may have a greater strength to predict natural conception when compared to ART where the sperm is prepared and the DNA damaged sperm fraction can be reduced.
|
Conclusion |
|---|
|
TOPAbstractIntroductionAneuploidy of paternal originY chromosome microdeletionsEpigenetic factorsDNA Strand breaks in...ConclusionReferences |
|---|
The use of ICSI has increased the likelihood of an abnormal spermatozoon to be selected for fertilization and participate in the development of an embryo. An important function of spermatozoa is to deliver the paternal genome to the oocyte. Therefore, abnormalities at different levels of paternal genomic organization may affect reproductive potential and ART outcome. Recent evidence suggests that spermatozoa containing fragmented DNA, Y chromosome microdeletion, abnormal chromosome number, or altered genetic imprint may be associated with impaired fertilization, embryogenesis, or fetal development.
This review has highlighted acquired spermatozoal nuclear determinants; however, a number of non-nuclear factors such as the centriole and mitochondria have also been linked to male infertility and shown to have an influence on fertilization and embryogenesis (Hewitson et al., 1997; Navara et al., 1997; Cummins, 2000, 2004; St John et al., 2000, 2004; Hewitson et al., 2002; Terada et al., 2004). Further investigation of specific spermatozoal factors will contribute to the development of novel and more personalized approaches to test and treat male factor infertility.
|
References |
|---|
|
TOPAbstractIntroductionAneuploidy of paternal originY chromosome microdeletionsEpigenetic factorsDNA Strand breaks in...ConclusionReferences |
|---|
Ahmadi A and Ng S-C (1999) Fertilizing ability of DNA-damaged spermatozoa. J Exp Zool 284, 696–704.[CrossRef][ISI][Medline]
Aran B, Veiga A, Vidal F, Parriego M, Vendrell JM, Santalo J, Egozcue J and Barri PN (2004) Preimplantation genetic diagnosis in patients with male meiotic abnormalities. Reprod Biomed Online 8, 470–476.[ISI][Medline]
Aravindan GR, Bjordahl J, Jost LK and Evenson DP (1997) Susceptibility of human sperm to in situ DNA denaturation is strongly correlated with DNA strand breaks identified by single-cell electrophoresis. Exp Cell Res 236, 231–237.[CrossRef][ISI][Medline]
Bestor TH (2000) The DNA methyltransferases of mammals. Hum Mol Genet 9, 2395–2402.
Blanco-Rodriguez J and Martinez-Garcia C (1996) Spontaneous germ cell death in the testis of the adult rat takes the form of apoptosis: re-evaluation of cell types that exhibit the ability to die during spermatogenesis. Cell Prolif 29, 13–31.[CrossRef][ISI][Medline]
Blanco-Rodriguez J and Martinez-Garcia C (1998) Apoptosis precedes detachment of germ cells from the seminiferous epithelium after hormone suppression by short-term oestradiol treatment of rats. Int J Androl 21, 109–115.[Medline]
Bonde JP, Ernst E, Jensen TK, Hjollund NH, Kolstad H, Henriksen TB, Scheike T, Giwercman A, Olsen J and Skakkebaek NE (1998) Relation between semen quality and fertility: a population-based study of 430 first-pregnancy planners. [see comment]. Lancet 352, 1172–1177.[CrossRef][ISI][Medline]
Braude P, Bolton V and Moore S (1988) Human gene expression first occurs between the four- and eight-cell stages of preimplantation development. Nature 332, 459–461.[CrossRef][Medline]
Carrell DT, Wilcox AL, Lowy L, Peterson CM, Jones KP, Erickson L, Campbell B, Branch W and Hatasaka HH (2003) Elevated sperm chromosome aneuploidy and apoptosis in patients with unexplained recurrent pregnancy loss. Obstet Gynecol 101, 1229–1235.
Carrell DT, Emery BR, Wilcox AL, Campbell B, Erickson L, Hatasaka HH, Jones KP and Peterson CM (2004) Sperm chromosome aneuploidy as related to male factor infertility and some ultrastructure defects. Arch Androl 50, 181–185.[CrossRef][ISI][Medline]
Chang PL, Sauer MV and Brown S (1999) Y chromosome microdeletion in a father and his four infertile sons. Hum Reprod 14, 2689–2694.
Chevret E, Rousseaux S, Monteil M, Usson Y, Cozzi J, Pelletier R and Sele B (1996) Increased incidence of hyperhaploid 24,XY spermatozoa detected by three-colour FISH in a 46,XY/47,XXY male. Hum Genet 97, 171–175.[CrossRef][ISI][Medline]
Choi JM, Chung P, Veeck L, Mielnik A, Palermo GD and Schlegel PN (2004) AZF microdeletions of the Y chromosome and in vitro fertilization outcome. Fertil Steril 81, 337–341.[CrossRef][ISI][Medline]
Colegero AE, Garofalo MR, Barone N, Palma AD, Vicari E, Romeo R, Tumino S and D'Agata R (2001) Spontaneous regression over time of the germinal epithelium in a Y chromosome-microdeleted patient: case report. Hum Reprod 16, 1845–1848.
Cox GF, Burger J, Lip V, Mau UA, Sperling K, Wu BL and Horsthemke B (2002) Intracytoplasmic sperm injection may increase the risk of imprinting defects. Am J Hum Genet 71, 162–164.[CrossRef][ISI][Medline]
Cozzi J, Chevret E, Rousseaux S, Pelletier R, Benitz V, Jalbert H and Sele B (1994) Achievement of meiosis in XXY germ cells: study of 543 sperm karyotypes from an XY/XXY mosaic patient. Hum Genet 93, 32–34.[ISI][Medline]
Cummins JM (2000) Fertilization and elimination of the paternal mitochondrial genome. Hum Reprod 15 (Suppl 2), 92–101.[Medline]
Cummins JM (2004) Mitochondria in reproduction. Reprod Biomed Online 8, 14–15.[ISI][Medline]
DeBaun MR, Niemitz EL and Feinberg AP (2003) Association of in vitro fertilization with Beckwith–Wiedemann syndrome and epigenetic alterations of LIT1 and H19. Am J Hum Genet 72, 156–160.[CrossRef][ISI][Medline]
Doerksen T and Trasler JM (1996) Developmental exposure of male germ cells to 5-azacytidine results in abnormal preimplantation development in rats. Biol Reprod 55, 1155–1162.[Abstract]
Doerksen T, Benoit G and Trasler JM (2000) Deoxyribonucleic acidhypomethylation of male germ cells by mitotic and meiotic exposure to 5-azacytidine is associated with altered testicular histology. Endocrinology 141, 3235–3244.
Doherty A, Mann M, Tremblay K, Bartolomei M and Schultz R (2000) Differential effects of culture on imprinted H19 expression in the preimplantation mouse embry