Human Reproduction Update Advance Access originally published online on September 28, 2006
Human Reproduction Update 2007 13(1):63-75; doi:10.1093/humupd/dml047
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Globozoospermia revisited
1 Department of Obstetrics and Gynaecology and 2 Department of Human Genetics, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
3 To whom correspondence should be addressed at: Radboud University Nijmegen Medical Centre, 791, Department of Obstetrics and Gynaecology, PO Box 9101, 6500 HB Nijmegen, The Netherlands. E-mail: j.kremer{at}obgyn.umcn.nl
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and methodsResultsConclusions and discussionReferences |
|---|
Globozoospermia is a rare (incidence <0.1%) but severe disorder in male infertility. Total globozoospermia is diagnosed by the presence of 100% round-headed spermatozoa lacking an acrosome. It is still unclear whether patients whose ejaculate contains both normal and globozoospermic cells (partial globozoospermia) suffer from a variation of the same syndrome. Apart from the fact that affected males suffer from reduced fertility or even infertility, no other physical characteristics can be associated with the syndrome. ICSI is a treatment option for these patients, although low fertilization rates after ICSI show a reduced ability to activate the oocyte. In globozoospermic cells, the use of acrosome markers has demonstrated an absent or severely malformed acrosome. Chromatin compaction appears to be disturbed but is not consistently over- or undercondensed. In some cases, an increased number of cells with DNA fragmentation have been observed. The analysis of the cytogenetic composition revealed an increased aneuploidy rate in some cases. Nonetheless, no increased number of spontaneous abortions or congenital defects has been reported in pregnancies conceived after ICSI. The pathogenesis of globozoospermia most probably originates in spermiogenesis, more specifically in acrosome formation and sperm head elongation. In several knockout mouse models, a phenotype similar to that in humans was found. Together with the occurrence of affected siblings, these findings indicate a genetic origin, which makes globozoospermia a good candidate for genetic analysis. More research is needed to elucidate the pathogenesis of human globozoospermia to further understand globozoospermia as well as (abnormalities in) spermiogenesis and spermatogenesis in general.
Key words: acrosome / genetics / globozoospermia / male infertility / round-headed sperm cells
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsConclusions and discussionReferences |
|---|
Globozoospermia is a rare but severe disorder causing male infertility. Schirren et al. (1971)
were the first to properly describe this syndrome of round-headed spermatozoa that lacked an acrosome. In the following three decades, a limited number of additional case reports were published. In these case reports, the morphological and aetiological aspects of globozoospermia were emphasized. However, the underlying causes of the syndrome still remain to be elucidated. A genetic contribution was postulated by Kullander and Rausing, which was supported by additional case reports of affected brothers (Kullander and Rausing, 1975; Florke-Gerloff et al., 1984; Dale et al., 1994; Carrell et al., 1999, 2001; Kilani et al., 2004). The gene(s) responsible or the mode(s) of inheritance remains obscure.
It is clear that male infertility in general is based on a multifactorial aetiology and tends to cluster in families. Diverse inheritance patterns have been described (Gianotten et al., 2004; van Golde et al., 2004). Genetic analysis has been complicated by the lack of a distinct correlation between genotype and (a highly variable) phenotype. The first description of globozoospermia, presenting as a disorder that affected all spermatozoa of a patient in a very specific way (Schirren et al., 1971), suggested that globozoospermia was an exception. The apparently very distinct phenotype suggested a monogenetic trait, thereby rendering this syndrome into an attractive target for genetic studies. Subsequent reports, however, presented a more diverse phenotype, obscuring the initial clear-cut phenotype (Holstein et al., 1973; Anton-Lamprecht et al., 1976; Florke-Gerloff et al., 1984).
Morphological defects in human sperm cells raise suspicion for further anomalies within the sperm cell and could have implications in clinical practice. Therefore, in this review we have attempted to summarize the phenotypic manifestations, and possible causes of the condition as well as implications for clinical practice of globozoospermia, as a prerequisite for genetic analysis.
|
Materials and methods |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsConclusions and discussionReferences |
|---|
For this systematic review, we collected all papers published on globozoospermia from 1965 to January 2006, using the following keywords [globozoosperm* OR (round AND headed AND sperma*) OR (round AND headed AND sperm) OR acrosomeless].
The initial literature search was performed using PubMed, Scirus and Medline. English was used as a limit, except for the Scirus database, because language selection is not possible in this database. Regular updates were performed using the same keywords in PubMed up until January 2006.
The initial search resulted in 88 hits in PubMed, 72 hits in Medline and 148 journal hits in Scirus. Subsequent searches in PubMed revealed another 11 papers. Papers dealing with megalohead sperm cells, round spermatids, macronuclear spermatozoa or acrosome malformation were excluded. In addition, we excluded papers in which cases of globozoospermia were used or proposed for research purposes, without specifically focusing on globozoospermia, nor providing additional information on globozoospermia (Kaufmann et al., 1987; von Bulow et al., 1995; Dimitrova-Dikanarova et al., 1998; Lefievre et al., 2003). Scirus appeared to be less accurate and produced a lot of hits that did not deal with globozoospermia or formed duplicates; these papers were excluded as well. After this selection, 80 PubMed, 60 Medline and 69 Scirus papers remained. The hits generated by PubMed included the 60 Medline papers. Scirus found seven papers that were not found by PubMed, resulting in 87 papers selected for review.
From the reference list of the selected papers, four German papers were selected to be included in our review, because they either mentioned the occurrence (Meyhöfer, 1965), features and pathogenesis of round-headed sperm cells (Schirren et al., 1971; Holstein et al., 1973) or suggested the term globozoospermia for the first time (Wolff et al., 1976). For background information see the Handbook of Andrology (http://www.andrologysociety.com/resources/handbook.aspx).
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsConclusions and discussionReferences |
|---|
Morphological description
The first to mention ‘Rundkopfspermatozoen’ (round-headed spermatozoa in German) after light microscopic analysis was Meyhöfer (1965). In 1971, Schirren et al. described the fine structure of these round-headed sperm cells as determined by electron microscopy (Figure 1) and discovered that their round shape was caused by a round nucleus lacking an acrosome. Other groups reported similar findings (Pedersen and Rebbe, 1974; Kullander and Rausing, 1975; Wolff et al., 1976; Baccetti et al., 1977, 1981), of whom Wolff et al. first suggested the term ‘globozoospermia’. Table I summarizes the reported morphological characteristics of globozoospermia.
|
|
Globozoospermia is normally diagnosed by the detection of round-headed sperm heads during routine light microscopic examination of a semen sample. The absence of an acrosome is another major feature, which is best visualized by transmission electron microscopy. Furthermore, sperm cells tend to have multiple defects involving the cytoskeleton such as a round nucleus, absence of the post-acrosomal sheath, separation of the nuclear membranes and frequently coiled tails. Also maturation defects such as a persisting residual cytoplasmic body/droplet surrounding the nucleus or the midpiece have frequently been reported.
Most of the papers based on describing this morphology deal with patients who produced ejaculates showing 100% round sperm heads lacking an acrosome. Holstein et al. (1973), however, reported a number of patients who only had 20–60% round-headed spermatozoa in their ejaculate. Whether this ‘partial’ globozoospermia was a different aspect of the same syndrome or a separate disorder could not be determined. The matter was further complicated by a case report by Anton-Lamprecht et al. (1976), who reported two types of patients with round-headed sperm cells. The first patient was of the classic ‘Schirren–Holstein type’, with exclusively round-headed acrosomeless sperm cells as determined by light and electron microscopy. The ejaculate of the second patient showed 80% of round-headed sperm cells, as determined by light microscopy, but electron microscopy showed that the round-headed phenotype was due to a residual cytoplasmic body that surrounded the nucleus and acrosome. They suggested the terms ‘globozoospermia type I’ for the classic Schirren–Holstein phenotype, and ‘globozoospermia type II’ for the phenotype observed in the second patient. Although their first suggestion was adopted in literature, only two globozoospermia type II patients have been reported since (Christensen et al., 2006). However, patients with <100% round-headed sperm cells in their ejaculate have frequently been described (Pedersen and Rebbe, 1974; Weissenberg et al., 1983; Florke-Gerloff et al., 1984; Syms et al., 1984; Tyler et al., 1985; Singer et al., 1986; Lanzendorf et al., 1988; Rybouchkin et al., 1996, 1997; Carrell et al., 1999, 2001; Coetzee et al., 2001; Larson et al., 2001).
To complicate matters even further, a number of cases were described in which electron microscopic examination showed that the sperm cells in these patients that did not appear to be round-headed by light microscopy could nonetheless lack an acrosome (Pedersen and Rebbe, 1974; Tyler et al., 1985; Lanzendorf et al., 1988; Coetzee et al., 2001). In other cases, only the round-headed sperm cells were found to be acrosomeless (Syms et al., 1984; Singer et al., 1986; Carrell et al., 1999). This was confirmed by immunofluorescent techniques using antibodies against (pro-) acrosin (Florke-Gerloff et al., 1984). Even if sperm cells were of normal or otherwise abnormal shape, an acrosome often appeared to be missing or anomalous (Florke-Gerloff et al., 1984; Rybouchkin et al., 1996; Carrell et al., 2001; Larson et al., 2001). According to these findings, it is clear that these cases do not belong to type II as described by Anton-Lamprecht et al. (1976). Hence, this nomenclature could be misleading. We therefore suggest the term ‘partial globozoospermia’ for patients in which <100% of the sperm cells show a round-headed form without an acrosome. Whether classic/total globozoospermia and partial globozoospermia are part of the same syndrome remains to be elucidated, today as urgently as >30 years ago.
The first case of globozoospermia, described by Schirren et al. (1971), was found after the examination of 2200 patients undergoing routine andrological screening, indicating an incidence of <0.05%. Later, the same group corrected the incidence to 0.1% among andrological patients (Holstein et al., 1973). Subsequent reports adopted this estimated incidence (Schill, 1991; Coetzee et al., 2001; Kim et al., 2001; Kalahanis et al., 2002). The documentation of globozoospermia, however, did not go beyond the level of case reports, so this incidence might be overestimated.
Round-headed cells also occur in otherwise normal semen samples (Florke-Gerloff et al., 1984). In the ejaculate of a proven fertile man, Florke-Gerloff et al. (1984) observed up to 6% of round-headed sperm cells. These cells lacked acrosin and the outer acrosomal membrane. Andrade-Rocha examined 233 suspected infertile men. The percentage of round-headed sperm cells in their ejaculates varied from 0.1 to 0.8%. This percentage was, in contrast to the percentage of abnormal sperm cells in general, not significantly related to the sperm count (Andrade-Rocha, 2001). Finally, Kalahanis et al. (2002) examined semen from 114 subfertile (
2 years involuntary childlessness without a female cause) and 60 proven fertile men. The percentage of round-headed sperm cells was slightly, but significantly higher in subfertile men (2.3 0.5%) compared to fertile men (0.5 0.1%) (Kalahanis et al., 2002). In relation to environmental factors, Rubes et al. (1998) concluded that smokers show a significantly increased amount of round-headed sperm cells in their ejaculate, in comparison with non-smokers. Although this was a relatively small study (20 smokers versus 15 non-smokers), these results might form an indication for an environmental factor in partial globozoospermia.
According to our search, 99 cases of globozoospermia have been reported since the first Schirren case. In this review, we summarize the clinical parameters of the patients involved.
The mean age of the patients, as reported in 70 of 99 cases, was 34.7 years, and 68% of the men were between 30 and 40 years old when examined.
Semen parameters were described in detail in 72 cases and showed great variation. Figure 2 shows the reported values in volume, concentration and motility, respectively. Besides a general observation of slight asthenozoospermia and the typical morphological characteristics, no abnormalities in semen parameters could be characterized as a typical feature of globozoospermia. Closer investigations by a computerized digital image analysis system revealed no abnormalities in sperm movement characteristics in globozoospermia (Aitken et al., 1990).
|
In 16 of 99 cases, a somatic karyotype of the affected patient was determined. A normal male somatic karyotype was found in 15 cases, one case displayed a mosaic trisomy 21 in 17.5% of the somatic cells (Kim et al., 2001
). On the level of DNA, a microdeletion of the Y-chromosome was found in another patient (Zeyneloglu et al., 2002). Gunalp et al. (2001), however, did not find any Y-chromosome microdeletions in 12 cases of globozoospermia. In summary, these data indicate that patients with globozoospermia in general carry a normal karyotype. The incidence of Y-chromosome microdeletions in globozoospermia does not appear to exceed the incidence in male subfertility in general.
The results of andrological history or examination were reported in 31 cases. In 22 cases, a normal health, normal masculine development and/or a normal aspect of the genitalia were reported (Pedersen and Rebbe, 1974; Kullander and Rausing, 1975; Anton-Lamprecht et al., 1976; Weissenberg et al., 1983; Jeyendran et al., 1985; Santi de et al., 1985; Lalonde et al., 1988; von Bernhardi et al., 1990; Dale et al., 1994; Bourne et al., 1995; Kilani et al., 1998, 2004; Viville et al., 2000; Nardo et al., 2002; Vicari et al., 2002; Pirrello et al., 2005). The remaining nine cases included four (treated) varicoceles (Florke-Gerloff et al., 1985; Jeyendran et al., 1985; Pirrello et al., 2005), two orchidectomies (Jeyendran et al., 1985; Pirrello et al., 2005), two (gonorrhoeal) infectious epidydimitis (Kullander and Rausing, 1975; Singer et al., 1986) and one unspecified penile ulcer in the past (Pedersen and Rebbe, 1974).
Aberrant details were sporadic, including a globozoospermic patient with bronchiectasis (Santi de et al., 1985), two cases of toxic exposition (Jeyendran et al., 1985; Carrell et al., 1999), two cases of consanguinity (Kilani et al., 2004; Pirrello et al., 2005) and two cases of opiate addiction in the past (Florke-Gerloff et al., 1985; Singer et al., 1986). Parotitis was mentioned in three cases, although orchitis did not occur in any case (Schirren et al., 1971; Kullander and Rausing, 1975), thereby excluding parotitis as a causing factor.
The inconsistent occurrence of these physical characteristics leads to the conclusion that next to reduced fertility, there are no other characteristics correlated with the globozoospermia phenotype.
(In)Fertility
In 64 cases, the patient was reported to be infertile. When Schirren et al. (1971) discovered that in case of globozoospermia all spermatozoa were acrosomeless, this was considered to be the cause of the infertility. Remarkably, proven fertility was noted in two separate cases. In the first case, a pregnancy was reported from a couple of whom the male partner presented with ‘type II’ globozoospermia (Anton-Lamprecht et al., 1976). The second case suffered from a 3-year involuntary childlessness, but this episode was preceded by two subsequent spontaneous abortions. No cause, explanation or evidence for acquired total globozoospermia was formulated (Arrighi et al., 1980). It should be noted that in neither case a paternity test was performed. Therefore, the involvement of a third party cannot be excluded.
The introduction of ICSI (Palermo et al., 1992) provided a solution in fertility treatment for patients suffering from globozoospermia (Hamberger et al., 1998). In 1994, Lundin et al. reported the first pregnancy and delivery using globozoospermic, acrosomeless spermatozoa in a second ICSI cycle. Since then, numerous reports have described successful attempts to achieve either fertilization or pregnancy following ICSI with globozoospermic sperm cells (Bourne et al., 1995a,b; Liu et al., 1995; Trokoudes et al., 1995; Battaglia et al., 1997; Rybouchkin et al., 1997; Kilani et al., 1998, 2004; Stone et al., 2000; Coetzee et al., 2001; Kim et al., 2001; Nardo et al., 2002; Tesarik et al., 2002; Zeyneloglu et al., 2002; Heindryckx et al., 2005). All reports are summarized in Table II. In general, ICSI with globozoospermic cells is less successful compared with ICSI in general. Several authors reported a low to absent fertilization (Liu et al., 1995; Battaglia et al., 1997; Rybouchkin et al., 1997; Stone et al., 2000; Kilani et al., 2004). Rybouchkin et al. (1996) discovered that fertilization was improved by the addition of a calcium ionophore. They suggested that a sperm-associated oocyte-activating factor that normally causes the Ca2+ flux required for fertilization might be absent or down-regulated in globozoospermic sperm, a suggestion that was subsequently confirmed by several studies (Rybouchkin et al., 1996; Gomez et al., 2000; Schmiady et al., 2005).
|
As for the pregnancy outcome, Nagy et al. (1998)
reported that although the fertilization rate was decreased in cases in which morphologically abnormal spermatozoa were used, no increase in the number of spontaneous abortions or congenital defects occurred. In accordance with these findings, no increase in such outcome failures has been reported in globozoospermia. Nevertheless, abnormal sperm morphology might be associated with genetic alterations in the sperm cells, which theoretically could have consequences on the long term for the offspring from artificial reproduction techniques. Sperm constitution in case of globozoospermia will be evaluated in the chapter ‘molecular description’. In globozoospermia, one or more of the sperm-remodelling mechanisms in spermiogenesis appear to be impaired. Especially, acrosome formation and nucleus elongation were studied in detail in globozoospermia patients. Four possible mechanisms have been postulated to explain the absence of the acrosome.
First, the acrosome may develop separately from the nucleus to be lost in the Sertoli cell.
Electron microscopic studies by Schirren et al. (1971) showed that the acrosome is present in early spermatids but develops in the cytoplasm, independently from the nucleus. In addition, the nucleus remained round, but no condensation abnormalities were reported (Schirren et al., 1971). Several authors confirmed these findings (Baccetti et al., 1977; Castellani et al., 1978; Nistal and Paniagua, 1978). The same group (Holstein et al., 1973) observed that the abnormally developed acrosome was removed with the residual body and left in the Sertoli cell, where it degenerated. This has been confirmed by the identification of remnants of abnormal acrosomal vesicles in the Sertoli cell cytoplasm (Castellani et al., 1978).
Second, the acrosomal vesicles do not fuse and even detach from the nuclear membrane. Results obtained with immunohistochemical techniques (Florke-Gerloff et al., 1985) are in accordance with the electron microscopic findings described above. These authors demonstrated the presence of three acrosomal markers (acrosin, intra-acrosin inhibitor and purified outer acrosomal membrane) in early globozoospermic spermatids. These markers were located adjacent to, but separated from, the nuclear membrane. Instead, they appeared to move into the cytoplasm and to be discarded together with the residual body.
Third, the acrosomal granules may be formed but degenerate subsequently. Baccetti et al. (1977) postulated this theory after the observation that the acrosomic vesicle does attach to the nuclear membrane but degenerates in the late spermatid stage, leaving a pouch of membranes and small vesicles located on top of the nucleus. Moreover, hypoplasia of the Golgi apparatus was found in most spermatids in case of globozoospermia. A malfunctioning Golgi apparatus was therefore postulated as a possible cause of this malformation of the acrosome (Baccetti et al., 1977; Castellani et al., 1978; Nistal and Paniagua, 1978).
Finally, the caudal manchette may be absent or malfunctioning (Baccetti et al., 1977; Castellani et al., 1978; Nistal and Paniagua, 1978). Naturally, a combination of these cytoskeleton impairments is possible as well (Lalonde et al., 1988; Escalier, 1990).
With regard to the round shape of the nucleus, Schirren et al. (1971) suggested that the nuclear shape was only determined by the acrosome, leaving the nucleus round-shaped in its absence. Subsequent studies, however, showed that the caudal manchette was often missing in developing spermatids of a globozoospermic male, indicating that the shape of the nucleus might be influenced by aplasia or hypoplasia of the caudal manchette as well (Baccetti et al., 1977; Castellani et al., 1978; Nistal and Paniagua, 1978).
Longo et al. (1987) showed that calicin, a basic protein that is almost exclusively located to the posterior part or calyx of the sperm nuclear theca, appeared to be absent in globozoospermic cells. This finding indicates an impaired development of the sperm-specific skeleton (Escalier, 1990; Courtot, 1991) which may influence both the formation of the acrosome and the shape of the nucleus.
Research in genetically altered mice has succeeded in elucidating some aspects of spermiogenesis in globozoospermia together with other kinds of teratozoospermia. This will be discussed in the last section of this article.
Acrosomal markers
The apparent absence of the acrosome in globozoospermia initiated several studies to demonstrate a disturbance of the functional properties of the acrosome and its membranes. First, the location of acrosin, its precursor proacrosin and the outer acrosomal membrane at the acrosomal region was investigated in normal spermatozoa using an indirect immunofluorescent staining technique with antibodies specific to these acrosomal components. In normal spermatids, the acrosomal developmental pattern was in accordance with data obtained using the electron microscope. In contrast, both acrosin and the outer acrosomal membrane were absent in mature globozoospermic spermatozoa, and the developmental pattern of globozoospermic spermatids was disturbed compared with normal spermatids. These findings were confirmed by testing acrosome function with the gelatinolysis test, which showed no proteolytic activity of the mature sperm-head of globozoospermic cells (Florke-Gerloff et al., 1983). Spectrophotometrically evaluated acrosin activity showed severely diminished to absent activity in globozoospermia (Florke-Gerloff et al., 1984, 1985; Lalonde et al., 1988). Jeyendran et al. (1985) found an eight times decreased amount of proacrosin, but the level of active acrosin after induced conversion appeared to be normal (Jeyendran et al., 1985). This suggests that the level of proacrosin was diminished, but yet capable of conversion.
Phospholipase A2, which is believed to play a role in the acrosome reaction by hydrolysing fatty acids, which are linked to membrane phospholipids overlying the acrosome. Its activity in globozoospermic cells was significantly lower compared with that of spermatozoa from donors of proven fertility (Lalonde et al., 1988).
Fluorescein-labelled lectins were used as acrosome-membrane markers in developing spermatids. The binding of peanut agglutinin (PNA), a marker for acrosome differentiation, to the developing acrosome was absent. Ricinus communis agglutinin II (RCA II), which displays a cap-like fluorescence pattern in normal spermatids, presented a dot-like pattern in globozoospermic spermatids. The authors concluded that the adhesion and penetration failure observed in globozoospermia might be caused by membrane defects (Wollina et al., 1989).
In accordance with these findings, vesicle-associated membrane protein (VAMP) or synaptobrevin was only found in rudiment form on globozoospermic cells when used as an acrosome marker, again pointing at a defect acrosome formation (Ramalho-Santos et al., 2002).
In summary, these studies show a present, but disturbed acrosome genesis, resulting in either a severely malformed or an absent acrosome in globozoospermia.
Fertilization capacity
Literature indicates that no spontaneous fertilization occurs in case of total globozoospermia. Several authors tested the fertilization capacity of human round-headed spermatozoa in animal models. In summary, these spermatozoa were capable of penetrating the cervical mucus (Jeyendran et al., 1985) but did not succeed in fertilizing (zona-free) hamster oocytes, unlike spermatozoa of men of proven fertility (Weissenberg et al., 1983; Syms et al., 1984; Jeyendran et al., 1985; Sutherland et al., 1985; Lalonde et al., 1988; von Bernhardi et al., 1990). This was most probably because of the incapacity to bind to the zona pellucida, as well as to problems during the fusion process of the spermatozoon with the oocyte membrane. Globozoospermic spermatozoa also failed to adhere to human zona pellucida. As elevated intracellular Ca2+ levels were supposed to be required for binding of the spermatozoon to the oocyte, binding capacity was also studied in the presence of the calcium ionophore A23187.
[GenBank]
Even then, however, the spermatozoa failed to adhere (Aitken et al., 1990; von Bernhardi et al., 1990; Carrell et al., 1999), which led to the conclusion that sperm fusion capacity cannot be triggered by increasing intracellular calcium (Dale et al., 1994).
Interestingly, the nuclei of round-headed spermatozoa do decondense to a similar degree as do normal spermatozoa when incubated with crushed hamster ova, indicating a normal fertilization capacity once the barrier of the oocyte membrane is overcome (Syms et al., 1984). Lanzendorf et al. (1988) elaborated on these findings by injecting human globozoospermic sperm cells into zona-intact hamster oocytes. They showed that the abnormal sperm was capable of apparently normal chromatin decondensation and pronucleus formation after injection (Lanzendorf et al., 1988). After the introduction of ICSI in humans (Palermo et al., 1992), fertilization and even pregnancy was achieved in several cases of globozoospermia (Lundin et al., 1994; Sathananthan, 1994; Trokoudes et al., 1995). The success of ICSI as a treatment option for globozoospermia, however, was not universal, as successive case reports reported low to absent fertilization in cases of ICSI. A down-regulation of an oocyte-activating factor was suspected (Liu et al., 1995; Battaglia et al., 1997; Rybouchkin et al., 1997; Nagy et al., 1998; Gomez et al., 2000; Stone et al., 2000; Coetzee et al., 2001). Poor oocyte activation capability was confirmed by the observation of premature chromosome condensation of globozoospermic sperm cells (Edirisinghe et al., 1998; Schmiady et al., 2005). Interestingly, performing ICSI in the presence of the calcium ionophore A23187
[GenBank]
appeared to be able to overcome the failure in oocyte activation. (Liu et al., 1995; Battaglia et al., 1997; Rybouchkin et al., 1997; Nagy et al., 1998; Gomez et al., 2000; Stone et al., 2000; Coetzee et al., 2001).
Several heterologous ICSI models have been developed to test the oocyte-activating capacity of spermatozoa after injection. In a mouse model, the human globozoospermic cells indeed failed to activate and fertilize the oocytes. When the oocytes were artificial activated with ethanol 8%, fertilization did increase to normal levels (Rybouchkin et al., 1996; Battaglia et al., 1997; Kim et al., 2001; Tesarik et al., 2002). A similar mouse oocyte activation test has been used in a clinical setting to predict the necessity of oocyte activation with a CaCl2/Ca2+ ionophore, and showed again poor fertilization of globozoospermic cells, overcome by the CaCl2/Ca2+ ionophore oocyte activation (Heindryckx et al., 2005).
Poor oocyte-activation capacity is thought to originate from the sperm centrosome as well, in its function as a microtubule-organizing centre. The function of the centrosome was examined in bovine oocytes and showed low sperm aster formation rates in globozoospermia after ICSI, supposedly because of a dysfunctional centrosome. Surprisingly, ethanol-induced oocyte activation increased fertilization rates, suggesting a centrosome-independent sperm ability to activate the oocyte (Terada et al., 2001, 2004; Nakamura et al., 2002; Terada, 2004).
Viability
Sperm viability is normally tested by eosin staining, as recommended by the World Health Organization (WHO). Jeyendran et al. (1984), however, developed the hypo-osmotic swelling test, based on the observation that exposing living sperm with intact membranes to hypo-osmotic conditions (150 mOsm/l) results in fluid flux into the cell, thereby increasing the sperm volume and resulting in a visible swelling of the tail membrane. This hypo-osmotic swelling test (HST) was used in combination with eosin staining to test the viability of round-headed sperm cells. In two globozoospermia patients, no decreased percentage of viable sperm cells were encountered (Jeyendran et al., 1985; Check et al., 1993).
Morphological deformities are indicative of structural chromatin and DNA abnormalities and cytogenetic defects in spermatozoa. Chromatin condensation and stability, DNA fragmentation and aneuploidy rates have been investigated and evaluated in round-headed sperm cells, although it should be noted that in general the number of patients investigated in each study was low, resulting in rather anecdotal data sets.
Nuclear chromatin
Sperm chromatin is condensed during spermiogenesis, in which process, amongst others, histones are replaced by protamines (for a review, see Dadoune, 2003). Abnormal chromatin condensation is a strong indication of a maturation defect. The morphological aspects of chromatin condensation in globozoospermia have been described repeatedly in various (electron microscopic) studies. These reports, however, differ in their findings. Chromatin condensation was classified as normal, granular or fine treaded in early reports (Pedersen and Rebbe, 1974; Anton-Lamprecht et al., 1976; Castellani et al., 1978). Later descriptions report abnormally condensed chromatin, without further specification (Kullander and Rausing, 1975; Wolff et al., 1976; Tyler et al., 1985; Lalonde et al., 1988; Escalier, 1990; Singh, 1992; Battaglia et al., 1997; Carrell et al., 1999; Vicari et al., 2002). In addition, the evaluation of several factors involved in chromatin condensation was performed. Again, various results have been published. Arrighi et al. (1980) observed high heterogeneity in Feulgen DNA contents within one patient, with low mean Feulgen stainability in globozoospermia, indicating overmaturity (Arrighi et al., 1980). Baccetti et al. (1977) based their conclusions on their previous findings that chromatin immaturity is characterized by high zinc, low phosphorus and variable lysine levels (Baccetti et al., 1977). Varying results were obtained within one patient. Although most cells showed immature chromatin by their definition, several cells were compact and mature as defined by low zinc and high phosphorus levels and by severely reduced lysine levels.
The presence of lysine was believed to indicate poor substitution of histones by protamines. Later reports indeed showed a disrupted replacement of histones by protamines (Blanchard et al., 1990). In one of two globozoospermia patients, a high histone rate in combination with a low protamine 1 (P1) percentage was found. The protamine 2 (P2) percentage was low but not related to the histone content. In both patients, a decreased P1/P2 ratio (0.51 and 0.52, respectively in comparison with 0.59 in fertile controls) was found. Contradictorily, an increased P1/P2 ratio (2.15) was found in one of two siblings suffering from globozoospermia (Carrell et al., 1999). In the other sibling, a normal ratio (0.79) in comparison with fertile controls was found. Chromatin structure measured by the flow cytometric sperm chromatin structure assay (SCSA), which is based on the susceptibility of sperm nuclear DNA to acid-induced denaturation, showed no differences in the chromatin stability of globozoospermic cells in comparison with highly fertile patients (Larson et al., 2001). Chromatin maturation can also be measured by propidium iodide staining, as immature chromatin is more accessible to this compound and is therefore more strongly stained. Vicari et al. (2002) reported an elevated number of globozoospermic sperm cells with immature chromatin in a case report, as determined by fluorescence-activated cell sorter (FACS) analysis of propidium iodide-stained semen samples (Vicari et al., 2002). Lalonde et al. (1988) investigated whether the absence of acrosomal components influenced nuclear chromatin decondensation. Globozoospermic sperm cells were therefore exposed to either dithiothreitol (DTT) or EDTA, which are known to induce nuclear chromatin decondensation. No difference was found between globozoospermic and normal spermatozoa (Lalonde et al., 1988). In contrast, Carrell et al. (1999) showed an increased decondensation rate in two affected siblings when exposed to heparin, from which they concluded that chromatin is less stable in globozoospermia (Carrell et al., 1999).
Taken together, these findings indicate that chromatin condensation is disturbed in globozoospermia, with a high heterogeneity in the degree of maturity. The aetiology of these findings remains to be revealed.
DNA fragmentation
DNA strand breaks can be visualized by labelling the 3'-OH ends using the TdT (terminal deoxynucleotidyl transferase)-mediated dUTP-biotin nick-end labelling (TUNEL) assay, which is indicative of DNA fragmentation. Both Baccetti et al. (1996) and Vicari et al. (2002) found an increased percentage of DNA fragmentation in globozoospermic sperm cells compared with fertile controls (10 versus 0.1% and 37 versus 22.5%, respectively). The neutral single-cell gel electrophoresis or COMET assay also quantifies double-stranded DNA breaks. Larson et al. (2001) found no elevation of DNA fragmentation in round-headed sperm cells compared with fertile controls, which was in accordance with their SCSA results, described in the previous section.
The sperm-ubiquitin tag immunoassay (SUTI), which detects DNA damage as well, was developed to identify abnormal sperm cells (Sutovsky et al., 2001). Round-headed sperm cells were highly ubiquinated, indicating defective DNA.
Cytogenetics of sperm cells
Whether morphological sperm deformities are linked to sperm cell chromosomal abnormalities has been investigated extensively but is still controversial. Carrell et al. (1999) published two case reports in which siblings with globozoospermia were evaluated by fluorescence in-situ hybridization (FISH). After the evaluation of >5000 cells, they found increased aneuploidy rates for chromosomes 13, 21 and XY in one of two affected siblings and a mildly increased aneuploidy rate for chromosome 21 in the other sibling (Carrell et al., 1999). In a similar study, three siblings, of whom two brothers were affected, were tested. These authors found no significant increase in aneuploidy rates for chromosomes 18, X and Y in either sibling but did detect a significant increase in chromosome 15 aneuploidy in one of the affected siblings (4.03 versus <0.4% in fertile controls) and his fertile brother (1.18%). The other affected brother, however, did not show a significant increase in aneuploidy rates (Carrell et al., 2001). Martin et al. (2003) also tested chromosomes 15, X and Y in over 10000 cells. They did not detect an elevated rate of chromosome 15 aneuploidy but did report an increase in XY disomy (Martin et al., 2003). Recently, these findings were supplemented with the observation of increased aneuploidy rates of chromosomes 13, 16 and 21 in a globozoospermic man compared to a normozoospermic man (Ditzel et al., 2005). Other studies could not confirm a higher aneuploidy rate in globozoospermic cells compared with aneuploidy rates in sperm cells of fertile controls (Rybouchkin et al., 1996; Viville et al., 2000; Vicari et al., 2002; Morel et al., 2004), although Morel et al. did observe a significantly higher disomy rate of chromosomes 13 and 21 in one patient compared with another globozoospermic man. Machev et al. (2005) have reviewed these papers, who concluded that increased aneuploidy rates occurred mostly in the acrocentric (13, 14, 15, 18 and 21) and sex chromosomes and that these findings do not differ from other types of infertility.
A genetic basis for globozoospermia was suspected and is supported by several case reports of families with two or more affected siblings (Kullander and Rausing, 1975; Florke-Gerloff et al., 1984; Dale et al., 1994; Carrell et al., 1999, 2001; Kilani et al., 2004). The mode of inheritance remains obscure. By investigating the occurrence of sperm defects in consanguinity, no additive evidence was found for an autosomal recessive condition in case of globozoospermia (Baccetti et al., 2001). Also, evidence for the participation of Y chromosome microdeletions in globozoospermia could not be established (Gunalp et al., 2001).
As early as 1975, Moutschen and Colizzi designated acrosomelessness as an interesting and efficient tool in mammalian mutation research. Up to now, four mouse genes have been associated with globozoospermia. The first gene, the so-called blind-sterile (bs) mutation, was identified in 1986. This was an autosomal recessive mutation on chromosome 2 that was shown to result in a failure to assemble an acrosome, a severely impaired spermatogenesis leading to a strongly reduced sperm count, and absence of motility and bilenticular cataract (Sotomayor and Handel, 1986).
In a more recent publication, globozoospermia was reported in mice with a homozygous deletion of the Csnk2a2 gene* (Xu et al., 1999; Rocha and Affara, 2000; Truong et al., 2003). This gene encodes a substrate of protein kinase casein kinase II, which is preferentially expressed in the late stages of spermatogenesis and is associated with the nuclear matrix. In these mice, the nucleus of spermatozoa was deformed. In spermatids, the acrosome often detached from the nucleus and disappeared during the next stages of spermatogenesis. Also, oligozoospermia due to Sertoli cell phagocytosis and apoptosis was observed (Xu et al., 1999). Recently, the Csnk2a2* gene has been examined for mutations in human globozoospermia patients. Despite expectations, no mutations were detected in any of six investigated patients (Pirrello et al., 2005).
Globozoospermia has also been described in Hrb–/– mice. Besides the characteristic round-headed and acrosomeless sperm cells, also sperm cells with a midpiece lacking a mitochondrial sheath in combination with reduced sperm motility were observed (Kang-Decker et al., 2001). Additional characteristics included loss in cell polarity, intracellular flagellar coiling, multinucleation, supernumerary centrioles and multiflagellation and nuclear vacuolization (Juneja and Van Deursen, 2005). The Hrb gene encodes for the HIV-1 Rev binding protein, an important factor in proacrosomic vesicle fusion, which is abundantly transcribed during spermiogenesis. As a consequence, proacrosomic vesicles are formed in the medulla of the Golgi apparatus in Hrb–/– mice spermatids but fail to fuse and transform into an acrosome (Kang-Decker et al., 2001).
Kierszenbaum et al. (2004) investigated the factors involved in sperm nuclear morphogenesis and identified a structure they previously designated the acroplaxome. This cytoskeletal scaffold, which, amongst others, contains F-actin and keratin 5, develops in the subacrosomal space and anchors the developing acrosome to the nuclear envelope and is involved in acrosome formation. When they studied the acroplaxome in Hrb–/– mice, they observed a deficiency in keratin 5 and the formation of a pseudo-acrosome (Kierszenbaum et al., 2004). The latter result was definitely reminiscent of the observations described by Baccetti et al. (1977), which would make Hrb an interesting candidate gene for human globozoospermia. Unfortunately, the first results on human globozoospermia patients are yet less promising (Christensen et al., 2006).
The most recently described mouse gene that is known to result in globozoospermia when knocked out is called the Golgi-associated PDZ and coiled-coil motif containing protein (GOPC) gene#. In spermatids, this protein is localized in the trans-Golgi cisternae and the trans-Golgi network. It plays a role in Golgi-to-membrane vesicle transport. In early spermatids, the fragmentation of the acrosomal cap was observed, as well as abnormal proacrosomic vesicles that failed to fuse (Yao et al., 2002).
Characteristics that remind of the observations made in human globozoospermia patients in early reports (Baccetti et al., 1977; Castellani et al., 1978; Nistal and Paniagua, 1978; Florke-Gerloff et al., 1985) were observed in later studies. These concluded that in GOPC–/– mice larger proacrosomic vesicles are formed and do attach to the poorly developed acroplaxome but that the vesicles detach and are lost when the spermatozoon is released into the lumen. Multiple failures regarding the nucleus, caudal manchette, post-acrosomal sheath and posterior ring were noted, resulting in impaired spermatid nuclear elongation (Ito et al., 2004). In detail, the defective posterior ring did not influence tail formation, but during epididymal passage, the tail was coiled around the nucleus, with the dislocation of the implantation fossa and a disorganized mitochondrial sheath (Suzuki-Toyota et al., 2004).
The last three candidate genes (Table III) have recently been screened in two men with (partial) globozoospermia. This study led to the identification of polymorphisms, but no mutation with a clear link to partial globozoospermia was found (Christensen et al., 2006). These results have yet to be confirmed in a larger group of patients with partial or total globozoospermia, before conclusions on these candidate genes can be drawn. Undoubtedly, the identification of proteins and protein complexes that are involved in spermiogenesis will substantially facilitate the search for additional candidate genes (Luo et al., 2003).
|
|
Conclusions and discussion |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsConclusions and discussionReferences |
|---|
Globozoospermia can be diagnosed by an extended semen analysis, which is likely to show the typical feature of round-headed, acrosomeless spermatozoa. A slight asthenozoospermia might also be found. No peculiarities in case history or physical examination of affected males have been associated with globozoospermia. The pathogenesis occurs during spermiogenesis and probably originates in acrosomic vesicle fusion and cytoskeleton disorders, which remains to be elucidated. Genetically manipulated mice show that the knockout of different genes can lead to approximately the same phenotype. Moreover, these observations show similarities with the results reported several decades ago (Baccetti et al., 1977
; Castellani et al., 1978; Nistal and Paniagua, 1978; Florke-Gerloff et al., 1985), which indicates that there might indeed be different genetic pathways of pathogenesis that lead to this sperm morphology disorder (Sotomayor and Handel, 1986; Xu et al., 1999; Kang-Decker et al., 2001; Yao et al., 2002; Kierszenbaum et al., 2004).
This could also explain the variable results in the cases of globozoospermia as described above. It is clear that a Schirren–Holstein or type I globozoospermia is a common term for 100% of round-headed and acrosomeless spermatozoa per ejaculate. The term type II globozoospermia for round-headed spermatozoa with an acrosome but with maturation defects was not broadly adopted in literature (Anton-Lamprecht et al., 1976; Singh, 1992; Christensen et al., 2006). Also, the use of the current nomenclature, namely type I and II, is not consistent and therefore misleading, as well in literature as in daily practice. We therefore suggest replacing these terms by total globozoospermia in case of 100% round-headed, acrosomeless spermatozoa and of partial globozoospermia if <100% of the spermatozoa are affected. Because globozoospermia is a rare disorder, only a relatively small number of patients have been described in literature. Moreover, various methods have been used to study these patients. These factors make an overall analysis very difficult. In our opinion, systematic collection and citation of all the facts found was the best solution, severe selection on quality would have excluded too much information. This implicates that no firm conclusions can be drawn. Instead, an overview of what has been written on globozoospermia in the past three decades has been provided to outline the syndrome globozoospermia.
Next to these academical considerations, physicians in their daily practice are in need of treatment options for these patients. No treatment was available until ICSI was introduced, which seemed to be a promising solution. Unfortunately, fertilization rates after ICSI often appear to be severely diminished in globozoospermia (Table II). A solution to this problem could be to use diagnostic heterologous ICSI to evaluate the fertilization capacity for each case, in order to predict the need of oocyte activation with a calcium ionophore (Heindryckx et al., 2005).
An important factor to consider is the possible effect on offspring of the abnormalities in chromatin structure and DNA integrity that, although inconsistently, have been described in globozoospermic cells. Whenever possible, these factors should be evaluated for each case in specialized centres. Obviously, preferably normally shaped sperm cells should be used in case of partial globozoospermia.
A lot of questions on globozoospermia remain to be answered. Additional studies are needed to elucidate whether cases of partial globozoospermia and total globozoospermia are variations of the same syndrome. Genetic analysis could be of great distinctive value in this matter, although environmental factors involved in globozoospermia aetiology cannot be ruled out completely. Nevertheless, genetic analysis in globozoospermia could be easier than in male subfertility in general, because of its distinct morphological characteristics, which places its origin in spermiogenesis. Therefore, studies should be focused on genes that are involved in the Golgi apparatus and cytoskeleton structures during spermiogenesis, like the candidate genes found in mice. Although mouse models may not be directly applicable to the human situation, these models may nevertheless help us better understand mammal spermatogenesis, giving us a perspective on the cause(s) of human globozoospermia. The fact that similar pathophysiologies are found in both mice and men does not necessarily mean that the homologous genes are involved in both species but surely gives an indication on what kind of genes we are looking for. In addition, the rarity of the syndrome is suggestive for a homozygous inheritance pattern, although Baccetti et al. (2001) did not find correlation between consanguinity and globozoospermia. Candidate genes could be identified by linkage analysis in consanguineous families, or families with more than one affected member, as the syndrome is too rare to occur in first-degree family members by chance. Also, the identification of proteins that are involved in spermiogenesis and the aetiology of globozoospermia by testing testis mRNA on expression microarrays could provide candidate genes that could subsequently be screened in patients and their families.
In conclusion, genetic research on globozoospermia is still in its initial stages but will undoubtedly prove to be invaluable in elucidating the processes of spermiogenesis and spermatogenesis in general and of the aetiology of globozoospermia in particular.
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsConclusions and discussionReferences |
|---|
The authors thank Leonie van den Hoven and Hannie Robben for their expert technical assistance and Henry Dijkman for his expertise on electron microscopy and the provided images.
|
Notes |
|---|
* Literature searches in another context revealed an article (Escalier, Silvius et al., 2003
), in which the authors of the same group that discovered the relation between Csnk2A2 and globozoospermia (Xu, Toselli et al., 1999) concl