The Aurora Kinase C c.144delC mutation causes meiosis I arrest in men and is frequent in the North African population

(1)

The Aurora Kinase C c.144delC mutation causes meiosis I arrest in men and is frequent

in the North African population

Klaus Dieterich

^1,2,3

, Raoudha Zouari

⁴

, Radu Harbuz

¹

, Franc¸ois Vialard

⁵

, Delphine Martinez

¹

, Hanane Bellayou

⁶

, Nadia Prisant

⁷

, Abdelali Zoghmar

⁸

, Marie Roberte Guichaoua

⁹

,

Isabelle Koscinski

¹⁰

, Mahmoud Kharouf

⁴

, Mehrdad Noruzinia

¹¹

, Sellama Nadifi

⁶

,

Abdelaziz Sefiani

¹²

, Jacqueline Lornage

¹³

, Mohamed Zahi

¹⁴

, Ste´phane Viville

¹⁰

, Bernard Se`le

^1,2

, Pierre-Simon Jouk

^1,2

, Marie-Christine Jacob

¹⁵

, Denise Escalier

¹⁶

, Yorgos Nikas

¹⁷

,

Sylviane Hennebicq

^1,2

, Joe¨l Lunardi

^1,2,18

and Pierre F. Ray

^1,2,3,

1De´partement de Geńe´tique et Procreátion, CHU de Grenoble, Grenoble cedex 9, France,²Faculte´ de Me´decine- Pharmacie, Domaine de la Merci, Universite´ Joseph Fourrier, La Tronche, France,³Laboratoire TIMC-IMAG, UMR 5525, Domaine de la Merci, 38710 La Tronche, France,⁴Clinique de la reproduction les Jasmins, 23, Av. Louis BRAILLE, 1002 Tunis, Tunisie,⁵Laboratoire de Cytogeńe´tique, CHI Poissy-Saint Germain, 78303 Poissy, France,

6Laboratoire de Geńe´tique et Pathologie Molećulaire, Faculte´ de Me´decine et Pharmacie Casablanca, Maroc,⁷UF de Biologie de la Reproduction, Groupe Hospitalier Pitie´ Salpetrie`re, APHP, Paris, France,⁸Centre de Me´decine de la Reproduction, No. 58 I Cite´ Boussouf, 25000 Constantine, Alge´rie,⁹Laboratoire de biologie de la reproduction, Hoˆpital de la Conception, 147, Bd Baille, 13385 Marseille, France,¹⁰Service de Biologie de la Reproduction, CHU de Strasbourg, 67303 Schiltigheim, France,¹¹Department of Hematology, School of Medical Sciences, Tarbiat Modares University, Tehran, Iran,¹²De´partement de geńe´tique me´dicale, Institut National d’Hygie`ne, B.P. 769, Rabat 11400, Maroc,¹³Labo de biologie de la reproduction, Hoˆpital Edouard Herriot, Lyon, France,¹⁴Centre prive´ de Fećondation In vitro, 13 rue A. Doukkali, les Orangers, Rabat, Maroc,¹⁵Laboratoire d’Immunologie Cellulaire, EFS Rhoˆne Alpes, site de Grenoble, France,¹⁶De´partement d’Andrologie, CHU Le Kremlin-Biceˆtre, F-94275, INSERM U654, Hoˆpital Armand Trousseau, Paris F-75571, France,¹⁷Athens Innovative Microscopy, Skra 36, Voula 16673, Greece and¹⁸Grenoble Institut des Neurosciences, INSERM U836, BP170, 38042 Grenoble, France

Received November 25, 2008; Revised January 6, 2009; Accepted January 13, 2009

Infertility concerns a minimum of 70 million couples worldwide. An important proportion of cases is believed to have a genetic component, yet few causal genes have been identified so far. In a previous study, we demonstrated that a homozygous mutation (c.144delC) in theAurora Kinase C(AURKC)gene led to the production of large-headed polyploid multi-flagellar spermatozoa, a primary infertility phenotype mainly observed in North Africans. We now want to estimate the prevalence of the defect, to improve our under- standing ofAURKCphysiopathology in spermatogenesis and assess its implication in oogenesis. A carrier frequency of 1/50 was established from individuals from the Maghrebian general population, comparable to that of Y-microdeletions, thus far the only known recurrent genetic event altering spermatogenesis. A total of 62 patients were genotyped, all who had a typical phenotype with close to 100% large-headed spermatozoa were homozygously mutated (n532), whereas no AURKC mutations were detected in the others. Two homozygous females were identified; both were fertile indicating that AURKC is not indispensible in oogenesis.

Previous FISH results had showed a great chromosomal heterogeneity in these patient’s spermatozoa. We demonstrate here by flow cytometry that all spermatozoa have in fact a homogeneous 4C DNA content and are thus all blocked before the first meiotic division. Our data thus indicate that a functional AURKC protein is necessary for male meiotic cytokinesis while its absence does not impair oogenesis.

To whom correspondence should be addressed at: UF de Biochimie et Geńe´tique Molećulaire, De´partement de Geńe´tique et Procreátion, CHU de Grenoble, 38043 Grenoble cedex 9, France. Tel:þ33 476765573; Fax:þ33 476765837; Email: pray@chu-grenoble.fr

For Permissions, please email: journals.permissions@oxfordjournals.org

Advance Access published on January 15, 2009

at Lakehead University on March 18, 2015http://hmg.oxfordjournals.org/Downloaded from

(2)

INTRODUCTION

A recent survey carried out on 172 413 women from 25 countries indicated a 9% prevalence of couple infertility, a figure noticeably lower than what is typically cited in the literature. Even with this lower estimate, the authors calculated that 70 million couples at least are concerned by this pathology (1).

It is now accepted that male and female factors contribute almost equally to this problem and that an important proportion of male infertility cases is caused by genetic defects. Yet since the discovery of the Y microdeletions in the 70s (2), only two genes have formally been associated with impaired spermatogenesis in the human (3,4) andAURKCis the only one found mutated in several unrelated individuals (4). Affected men present with primary infertility caused by 100% teratozoospermia with a majority of large-headed spermatozoa with up to four flagella (5) (Figs 1 and 2). This characteristic phenotype was described 30 years ago (6) and new cases have been reported regularly since (7–15) (OMIM 243060). Several FISH studies demonstrated that almost all the spermatozoa from patients with a typical form with 100% teratozoospermia showed polyploidy or aneuploidies (8–14) while ‘mosaic’

forms had a percentage of euploid spermatozoa proportional to their ratio of normal-sized gametes (15).

In our previous study, we carried out a whole genome scan on 10 men of North African origin presenting with typical homogeneous large-headed spermatozoa. We localized a candidate region and subsequently identified the same homozygous mutation on the Aurora Kinase C gene (c.144delC) in all the patients tested (4).AURCis preferentially expressed in the testes and is involved in chromosomal segregation and cytokinesis (16 – 18). We demonstrated that the identified frameshift mutation led to the premature termination of translation yielding a non-functional truncated protein that lacks its kinase domain. All patients shared a common haplo- type allowing us to conclude that they had a common ancestor and that the original mutational event took place approximately 15 centuries ago (4).

Here we investigate the frequency of this genetic defect by analysing patients and control individuals from the Maghrebian population. We also genotyped other family members to try and identify females homozygous for the c.144delC deletion to assessAURKCfunction in female fertility. Finally, we deve- loped a flow cytometry assay to directly measure the quantity of DNA present in the patient’s spermatozoa to verify if haploid gametes are present in the patient’s ejaculate and whether ICSI can be legitimately be proposed to these patients.

RESULTS

AURKCmolecular analysis of patients and family members

Sixty-two patients with various percentages of large-headed spermatozoa were analysed (Table 1). Thirty-two patients presented with a typical profile of .95% teratozoospermia, .35% large headed and .20% multiflagellar spermatozoa and a concentration comprised between 0.4 and 28 M/ml.

All carried two AURKC mutations. Thirty one were c.144delC homozygous and one was heterozygous for

c.144delC and p.C229Y, a novel false sense mutation in exon 6. All AURKC þ/þ patients either lived or originated from North Africa (Morocco Tunisia and Algeria).

Thirty patients did not fit the criteria defined for typical large-headed patients. No mutations were identified in any of these 30 patients after double strand sequencing of the entire coding sequence and intron boundaries of the AURKC gene. These patients can be separated into three separate groups: (i) 19 patients came from Iran. They had a good concentration and a low percentage of large-headed spermatozoa (2 – 5%). (ii) Nine patients had variable percentage of macrocephalic spermatozoa (5 – 75%) associated or not with flagellum abnormalities and are referred to as ‘atypical’ in Table 1. (iii) Two patients had extreme oligozoospermia with only a few spermatozoa—all macrocephalic—in the entire ejaculate. Because of the paucity of data they do not appear in Table 1.

Sperm from typical and mosaic large-headed had a comparable Multiple Anomalies Index.2 for the typical forms and .1.7 for the mosaic as well as acrosome abnormalities (average 80% for both groups).

Ten sisters of AURKC mutated men could be recruited for the study: four were homozygous wild-type, four were heterozygous and two were homozygous for c.144delC, both had conceived naturally two and six children, respectively, and did not report any miscarriages or any difficulties in conceiving.

AURKCc.144delC mutation analysis in control populations A total of 8 heterozygotes were identified in 440 unrelated individuals coming from 6 subgroups as follows:

(i) 1 out of 50 fertile men living near Grenoble and originating from North Africa.

(ii) 2/34 fertile men living near Paris and originating from North Africa.

(iii) 1/71 infertile men with oligozoospermia (n¼61) or azoospermia (n¼10) and without the large-headed phenotype, living near Paris and originating from North Africa.

(iv) 3/150 individuals living near Casablanca, Morocco.

A fourth individual had an abnormal DHPLC profile different from the usual c.144delC heteroduplex peak.

Sequence analysis revealed that he was heterozygous for another mutation: p.Ile79Val (c.235A.G) (Fig. 3).

(v) 1/80 individuals living in the Rabat region, Morroco.

(vi) 0/55 individuals living in the Arabic Peninsula (from Saudi Arabia, Qatar and the United Arab Emirate).

Electron microscopy study

By scanning electron microscopy (SEM), all observed spermatozoa from one homozygous c.144delC patient with a typical phenotype showed important morphological abnormalities (Fig. 1). The heads were usually approximately four times larger than the normal size (Fig. 1B –E) and often appeared with an irregular outline with protrusions (Fig. 1B and C).

Most spermatozoa showed multiple flagella (Fig. 2B–E), often up to four flagella (Fig. 1C). Anomalies of implantation of 1302 Human Molecular Genetics, 2009, Vol. 18, No. 7

(3)

these multiple flagella on a too narrow nuclear basis (Fig. 1D and E) often led to their separation from the sperm head (Fig. 1F).

Transmission electron microscopy (TEM) (Fig. 2) revealed that 90% of the sperm heads were mononucleated, the remain- ing 10% being binucleated. These enlarged irregular sperm heads resulted from a giant nucleus with a very irregular outline associated to a misshapen acrosome (Fig. 2A and B) and a cytoplasmic remnant (Fig. 2A). Several flagella could be seen in longitudinal sections (Fig. 2B). Transverse sections allowed seeing that their number often reached up to four flagella (Fig. 2C). Multiple flagella were seen in 80% of the middle piece sections.

Flow cytometry analysis

A control DNA index M2/M1 was calculated using the ratio of the mean fluorescence intensity (MFI) of propidium iodide

Figure 1.SEM photos (X 9000) of representative spermatozoa from a c.144delCþ/þpatient (B–F) compared with a normal spermatozoa from a fertile donor (A). Normal head (A), large head (B – E), irregular head outline (B and C), multiple flagella (B – E), implantation of multiple flagella as seen at the neck level (D) flagellar angulation (E), decapitated multiple flagella (F).

Figure 2.TEM of spermatozoa from a c.144delCþ/þpatient. The spermatozoa exhibit a giant nucleus, a mishapped acrosome (AandB) and cytoplasmic rem- nants (A). Two flagella (A and B) to four flagella (C) can be seen in sections.

(4)

(PI) fixed by leukocytes (M2¼MFI of DNA level 2C) and control spermatozoa (M1¼MFI of DNA level 1C). Control cells were analysed in all experiments and an M2/M1 ratio comprised between 1.9 and 2.1 validated the efficiency of sperm decondensation and the ensuing proportionality of dye fixation.

For patients, the DNA index was established by calculating the ratio between the MFI of PI fixed by the patient’s spermatozoa (Mp) and that fixed by control spermatozoa (Mp/M1).

Four semen from c.144delCþ/þpatients could be analysed (Fig. 4). All showed an Mp/M1 of 4+0.3 with a coefficient of variation ,10% thus indicating that the overwhelming

majority of the patient’s spermatozoa had a 4C DNA content and thus were tetraploid.

DISCUSSION

TwoAURKCmutations were identified in all patients presenting with a typical phenotype. The relatively large variations observed in the proportion of morphological abnormal spermatozoa within these patients (Table 1) probably reflect, at least partly, inter-laboratory scoring variability (19). All

Table 1. Sperm parameters of all analysed patients^a

Typical phenotype^b(n¼32) Atypical phenotype^c(n¼9) Iranian cohort highly atypical (n¼19)

Average Range Average Range Average Range

Large headed (%) 76 34 – 100 30 5 – 75 3 2 – 5

Multiflagellar (%) 40.3 20 – 100 8.5 0 – 28 N.A. N.A.

Abnormal acrosome (%) 81.8 62 – 100 81 67 – 100 N.A. N.A.

Multiple Anomalies Index 2.8 2 – 3.7 2.7 1.7 – 3.2 N.A. N.A.

Normal morphology (%) 0.6 0 – 15 6.5 0 – 19 27 10 – 37

Sperm volume (ml) 3.6 1.6 – 9.5 4.2 2.1 – 8.3 3.7 3 – 4.5

Nb spz10⁶per ml 7.8 0.4 – 28 8.6 0.01 – 45 168 72 – 280

Round cells (10⁶cells) 1.2 0 – 3.2 8 1.4 – 29 1.6 0 – 4

Motility AþB, 1 h (%) 7.9 0 – 20 17 0 – 40 N.A. N.A.

Vitality (%) 36 5 – 70 63 50 – 84 N.A. N.A.

N.A., not available.

aDoes not include two patients with near azoospermia.

bIncludes 14 patients described in (4).

cIncludes 4 patients described in (15).

Figure 3.DHPLC and sequence analysis results for AURKC exon 3: (A) homozygous for the wild-type sequence, (B) heterozygous for the mutation c.144delC, (C) homozygous for mutation c.144delC, (D) heterozygous for the mutation c.235A.G.

1304 Human Molecular Genetics, 2009, Vol. 18, No. 7

(5)

patients but one were c.144delC homozygous (n¼31) and a novel mutation in exon 6, p.Cys229Tyr (c.686G.A), was identified in a compound heterozygote [c.144delC;

c.686G.A]. This substitution takes place in the kinase domain of the protein and concerns a cystine and a tyrosine,

two amino acids with significant physicochemical difference.

It affects nucleotide 686 which is conserved through 12 species up to the Tetraodon and was not observed in 100 control chromosomes analysed. A total of 30 patients were analysed who had large-headed spermatozoa, but who did

Figure 4.Flow cytometry histograms obtained after decondensation and PI staining of (A) normal spermatozoa from a fertile anonymous donor; (B) leukocytes;

(C–F) spermatozoa from c.144delCþ/þpatients. Thex-axis (FL2-A) represents the intensity of fluorescence on a linear scale. It is directly proportional to the quantity of DNA fixed by each cell analysed (10 000 – 100 000). They-axis (counts) represents the numbers of cells/events analysed on a linear scale. In each case, the first peak (between 0 and 100) corresponds to cell debris. The DNA index for each patient is calculated using M1 from histogram (A). The small peak around 400 in (A) corresponds to a small proportion of diploid white cells present in the ejaculate of the donor. The small peak at 800 in (B) corresponds to a minority of leukocyte in mitosis. In (C – E) a single peak is observed [sometime with an embedded secondary peak (D – E) indicating an unequal access of the dye to the DNA of the oversized nuclei] with a DNA index around 4. Patient F spermogram showed the presence of approximately 1/3 of white cells, they appear as a single narrow peak at 449.

(6)

not present with a typical phenotype. Two patients were almost azoospermic with only a few detectable spermatozoa, all large-headed and multi-flagellar. Nineteen Iranian patients had a good numeration with low percentages of large-headed spermatozoa (2 – 5%), we had hypothesized that they could carry a less severeAURKCmutations. The last nine were prin- cipally of French origin and had between 5 and 75% of large- headed spermatozoa. FISH analysis had previously been realized on four of these patient’s spermatozoa and had showed a high, albeit not total, rate of chromosomal abnormalities (30 – 70%) (15). No mutations were found after sequencing of the whole AURKC coding sequence of these 30 ‘atypical’

patients. No semen samples were available to test the ploidy of these patient’s spermatozoa by FACS. Overall, all typical patients with 100% abnormal spermatozoa with a majority of large-headed spermatozoa frequently associated with several flagella carried two AURKC mutations, while no mutations were identified in the milder forms with fewer abnormal gametes or in patients with near azoospermia.

All mutated patients originated from Tunisia (n¼9), Morocco (n¼10) and Algeria (n¼9), and four had an unspe- cified North African origin. Only four North African reproductive centres participated in the study. We can thus expect that numerous patients did not consult, either because of the taboo still frequently associated with male infertility or for lack of accessible/affordable reproductive medicine.

To estimate the frequency of the mutation, we genotyped four series of subjects originating from North Africa. There were 8 carriers out of 385 individual tested (frequency .1/50). From this figure, the expected prevalence is of 1/10 000 men, thus ranking AURKC infertility in North Africa among the most frequent single gene defect before spinal muscular atrophy or myotonic dystrophy (20). One of the groups of patients analysed was constituted by French residents originating from North Africa undergoing infertility treatment for abnormal sperm parameters (but not for large-headed spermatozoa) in a centre near Paris. Only one heterozygote was identified from the 71 tested indicating that the c.144delC frequency was not higher in infertile men, suggesting that AURKC is probably not implicated in other (milder) forms of male infertility. This was confirmed by the fact that no mutations were identified in patients with less severe forms of large-headed spermatozoa. Also, the only compound heterozygote identified presented with a typical large- headed phenotype despite carrying a false sense mutation (p.Cys229Tyr) that could be expected to be less severe than a truncating mutation located early in the gene-like c.144delC.

Other pathological variants can be expected to be found as a mutation in the exon 3 conserved amino acid 79 (p.Ile79Val) was identified in one (of 385) control individual.

All the affected men interrogated (n¼15) thought to be of Arab rather than of Berber origin. We thus genotyped 55 individuals from the Arabian Peninsula. There were no c.144delC carriers in this small cohort indicating that, if the mutation is present in this population it is probably at a lower frequency than in North Africans. We can speculate that the mutation either preceded or succeeded the Arabic invasion of North Africa (seventh century), but was probably not brought by the invading troops. A larger epidemiological study is underway to better assess the origin of the mutation.

To our knowledge, AURKC is the first gene in which a recurrent mutation has been shown to cause male infertility by impairing spermatogenesis. We estimate the frequency of heterozygotes in the North African population to be of 1 in 50. This indicates that AUKRC deficiency could be among the most frequent single cause of male infertility in Maghre- bian men. This high frequency is surprising for mutation clearly inducing a reproductive handicap. One can wonder if AURKC heterozygosity could bring a selective advantage that would explain this high frequency, but we presently do not have any cue to what it might be.

AURKCinstrumental role in male meiosis and its reported expression in ovary (21) and oocytes (22,23) suggested a prob- able role in oogenesis. None of our patients, however, reported the occurrence of female infertility or recurrent miscarriages in their relatives. We screened the patient’s sisters who volunteered for the study to identify c.144delC homozygous females. Two of them were homozygotes and had fostered two and six children, respectively, indicating that AURKC therefore is not indispensable for oogenesis. This information highlights fundamental differences in meiotic basic mechanisms between male and female. AURKC transcripts have also been reported to be present in 16 somatic tissues (21).

In our experience, apart from infertility, AURKC mutated men do not show any other health problem. We can thus conclude that AURKC expression in somatic tissues and ovaries is either ectopic or redundant and that in human it is only absolutely necessary for spermatogenesis.

Aurora Kinases are key mitotic agents and their inhibitors appear as promising anticancerous molecules (24). The molecules being currently tested inhibit all three kinases or at least Aurora B and C (25). Chemiotherapies have always affected fertility and greatly increase the risks of producing aneuploid gametes. Never before however, have they specifi- cally disrupted male meiosis. Animal studies should therefore be carried out to assess the effect of anti-Auroras on spermatogenesis and establish their reversibility. If these drugs prove their efficiency, more than ever before, it will be critical for male patients to organize gamete cryopreservation before beginning the treatment.

Sperm samples could be obtained from four c.144delC homozygous patients and analysed by flow cytometry after PI staining (Fig. 4). All presented a single peak corresponding to a DNA index of 4 indicating that all spermatozoa had a 4C DNA content instead of the 1C expected from normal gametes. These data indicate that spermatogenesis was blocked after DNA synthesis but prior to the first meiotic division (Fig. 5). Surprisingly, spermiogenesis is not blocked and the presence of four haploid genomes is detected and leads to the frequent formation of four flagella. Previous reports of FISH analyses realized on spermatozoa from affected men showed a great variability in the number of counted chromosomes, even within spermatozoa from the same patients. In these reports, the number of spermatozoa interpreted as haploid varied from 0 to 8%, as diploid from 20 to 60%, as triploid from 10 to 62% and as tetraploid from 5.1 to 36%

(4,8 – 14,26). We demonstrated here that there is no such chromosomal heterogeneity and that all patient’s spermatozoa are in fact tetraploid. It is thus likely that in the spermatozoa scored as haploid after FISH analysis, the bivalent had not 1306 Human Molecular Genetics, 2009, Vol. 18, No. 7

(7)

been separated leading to the visualization of a single (large) fluorescent signal due to probe hybridization on closely adja- cent homologous bivalents and chromatids. Diploidy was most frequently reported indicating that a majority of bivalents had segregated, but that their two chromatids remained in close proximity and appeared as a single FISH signal. Some tetraploid spermatozoa were also detected by FISH indicating that in these cases, both bivalents and chromatids had, at least partly segregated, and could lead to the identification of four distinct FISH signals. The discrepancy observed between FISH and cytometry results clearly highlights the limits of the FISH technique by showing that the superposition of signals can lead to a great undermining of the real number of chromosomes/chromatids.

Spermatozoa fromAURKCknock out mice show a greater variability as only 20% of spermatozoa have an enlarged head. Male mice, although subfertile, can foster litters of reduced size (27), thus indicating that in rodents spermatogenesis, both chromosome segregation and cytokinesis could take place without Aurora C. It has been demonstrated in human cells thatAURKCfunction can overlap with and complement AURKB during mitosis (28). We can hypothesize that the reverse may be possible in mouse testis: Aurora B could compensate for the absence of Aurora C during meiosis.

Alternatively, a negative effect of the mutant protein cannot be ruled out; however, the apparent absence of phenotype in

heterozygous men does not favour this hypothesis. The phenotype of knock-in animals with the equivalent of the c.144delC mutation would have to be investigated to confirm or eliminate this possibility.

Taken together, our observations indicate that in men without a functional Aurora kinase C protein, a perturbed chromosomal segregation can occur, whereas meiotic cytokinesis is always blocked. The division blockage is likely to be secondary to the segregation defects that would trigger a spindle checkpoint signal leading to a forced exit from the meiotic M-phase. This is supported by the effect of ZM447439, a small molecule Aurora inhibitor, that was showed to perturb signal checkpoint signalling in rat semini- ferous tubulesin vitro(29).

Our data show that, although typical forms of large-headed spermatozoa syndrome are genetically homogeneous, other mechanisms and probably other genes lead to the production of such abnormal gametes. It can sometime be difficult to dis- tinguish between the ‘true’ and ‘mosaic’ forms of the syndrome. We demonstrate here that there can be no hope of autologous fertilization forAURKCþ/þmen as all spermatozoa are tetraploid. Although low, the chances of pregnancy cannot be ruled out for patients with atypical forms (15,30).

It is therefore important to realize a molecular diagnosis for all patients. Homozygous men can only be oriented towards gamete donation or adoption, whereas FISH analysis can be performed on AURKC negative patient’s spermatozoa to provide them with a more accurate prognosis and propose PGD if appropriate (31).

MATERIALS AND METHODS Patients and controls

A total of 62 patients were included in this study. Patients were divided into four groups:

(i) Thirty two patients presented with a typical profile with .95% teratozoospermia, .35% large headed and .20% multiflagellar spermatozoa, and a concentration comprised between 0.4 and 28 M/ml. The 14 patients genotyped and described in (4) are included in this group.

(ii) Nine patients had variable percentage of macrocephalic spermatozoa associated or not with flagellum abnormalities and did not fit the criteria for the typical profile.

Among these patients, four had been described previously in (15).

(iii) Nineteen patients had been recruited in Iran for infertility of unknown origin. All had low rates of large-headed spermatozoa (2 – 5%), .35% head defect and a normal concentration and sperm volume.

(iv) Two patients were analysed who had near azoospermia with less than 10 spermatozoa observed after centrifu- gation of the whole ejaculate. All observed spermatozoa were large headed with several flagella.

None of the 62 patients had chromosomal abnormalities.

None of the patients from groups 1, 2 and 4 were positive for Y microdeletions. This information was not available from the Iranian cohort. None of the patients were related apart from two brothers included in group 1. Fifteen additional

Figure 5.Schematic representation of spermatogenesis in (A) fertile men; (C) c.144delCþ/þpatients. In fertile males, germ cells undergo two meiotic divisions (M1, M2) leading to the production of haploid cells with 1c of DNA (A). This is followed by spermiogenesis, the acquisition of sperm specific characteristics (A). In c.144delCþ/þpatients, we demonstrated that all spermatozoa have a 4c DNA content (Fig. 4) indicating that the first cleavage did not occur (B). Germ cells from c.144delCþ/þpatients undergo DNA synthesis and reach the spermatocyte I stage and remain blocked between pro- phase and anaphase I with a quantity 4c of DNA (C). There are no divisions, hence the cells maintain a large size (as seen in Fig. 1). Surprisingly, spermiogenesis takes place with an elongation of the head, a reduction of the cytoplasmic volume and the acquisition of several flagella. The absence of cell division, the possibility of spermatozoa sequestration in the testis and/or cell death can explain the reduced sperm count of most affected men.

(8)

family members were genotyped in an effort to identified homozygous female. Patients and controls were recruited from 10 centres in France, Morocco, Algeria, Tunisia and Iran. All men referred as ‘of North African origin’ were French residents originating from Algeria, Morocco or Tunisia.

To estimate the frequency of the mutation, six cohorts of non-related anonymous control individuals were genotyped for AURKC c.144delC mutation: (i) 50 fertile men living near Grenoble and originating from North Africa; (ii) 34 fertile men living near Paris and originating from North Africa; (iii) 71 infertile men with oligozoospermia (without the large headed phenotype) (n¼61) or azoospermia (n¼10) living near Paris and originating from North Africa, (iv) 150 individuals living in or near Casablanca, Morocco; (v) 80 individuals living in the Rabat region, Morroco; (vi) 55 individuals living in the Arabic Peninsula (from Saudi Arabia, Qatar and the United Arab Emirate).

All patients, family members and anonymous donors gave their written informed consent. This work was approved by the Grenoble teaching hospital ethical committee.

Sperm analysis

Sperm analysis was carried out in the source laboratories.

World Health Organization (WHO) (32) guidelines were respected although small protocol variations might be observed between the different laboratories.

Mutation analysis

DNA preparation and PCR amplification. Genomic DNA was extracted either from peripheral blood leucocytes using a guanidium chloride extraction procedure or from saliva using Oragene DNA Self-Collection Kit (DNAgenotech, Canada). AURKC PCR amplification was carried out as described previously (4).

DHPLC. Exon 3 of the 440 control subjects was amplified and analysed by DHPLC. The PCR products were analysed both individually and pooled 2 by 2 to reduce the risk of not detecting homozygous mutants. PCR products were loaded on a DNA separation column (Transgenomic, Santa Clara, CA, USA) and ran over a 4.5 min period through a linear acetonitrile gradient (54 – 63% buffer B) at a temperature of 61.28C. All abnormal profiles were sequenced to confirm presence and nature of the mutation. DHPLC profiles with their corresponding sequences are shown in Figure 3.

Sequencing analysis. All analyses were carried out using the BigDye Terminator v3.1 sequencing kit and an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). Primers and protocols are as published previously (4).

AURKC exon 3 double-strand sequencing was realized for all the control DNA having yielded an abnormal DHPLC profile (Fig. 3) and for all patients with large-headed spermatozoa. When no mutation was identified in exon 3, double strand sequencing of the other six exons and intron boundaries was realized.

Electron microscopy

After liquefaction, the whole ejaculate was washed in PBS and fixed in 2.5% glutaraldehyde in PBS. The two halves of the sample were used for scanning and transmission microscopy.

Scanning electron microscopy. SEM was carried out on one patient and no statistical data could be obtained. Figure 1 provides an illustration of the obtained results. The spermatozoa were rinsed in PBS buffer, postfixed in 1% osmium tetroxide in buffer, dehydrated in a graded ethanol series, critical point dried and sputter-coated with gold. The samples were then examined using a Jeol JSM5600LV scanning electron microscope (Fig. 1).

Transmission electron microscopy. TEM was carried out on one patient and no statistical data could be obtained.

Figure 2 provides an illustration of the obtained results. The spermatozoa were rinsed in 4% sucrose and embedded in 2% agar. Postfixation was carried out for 1 h in 1% osmic acid in 0.1 M Sorensen buffer with 4% sucrose. After dehy- dration, small pieces were embedded in araldite. Sections were cut, stained with uranyl acetate (4% in 70% ethanol for 20 min) and lead citrate and examined under a JEOL JEM 100CX II transmission electron microscope (Fig. 2).

Flow cytometry

Spermatozoa and leukocytes preparation. The spermatozoa were frozen and stored in liquid nitrogen according to the routine clinical protocols followed by the source laboratory.

During thawing, 12 ml of PBS buffer (GIBCO^w) were added drop by drop to eliminate the cryopreservation medium. The spermatozoa were then centrifuged for 10 min at 800 g at RT and the supernatant eliminated. The resulting pellets were resuspended in PBS by gentle manual shaking.

Leukocytes were obtained on the day of the cytometric analysis from total blood by standard Ficoll gradient and stored at 48C before use.

Nuclear decondensation, PI staining and analysis. Cell mem- branes were permeabilized by a trypsin-detergent procedure based on Vindelov et al. (33). Several modifications to the published protocol were necessary to obtain an adequate degree of DNA decondensation. One hundred microlitres of cell preparation (10⁵– 10⁶cells) was incubated at room temperature for 3 h with 100ml of solution containing 1 g/l of citric acid; 0.52 g/l spermin, 0.06 g/l Tris, 1% v/v NP40, 30 mg/l Trypsin, 1 g/l proteinase K and 10 mM 2-Mercaptoethanol.

Proteases were then degraded by a 10 min incubation with 750ml of solution with 0.1 g/l RNAse, 0.5 g/l trypsin inhibitor and 2 mMphenylmethylsulfonyl fluoride. Finally, 250ml of PI (1 g/l) was added and incubated at 48C in the dark for 2 h prior to cytometric analysis.

All the samples were analysed on a 488 nm Argon laser FACSCalibur flow cytometer (Becton-Dickinson) with a 585 nm band-pass filter used for emission. The data were analysed using CellQuest Pro software (Becton-Dickinson).

Baseline levels for normal diploid and haploid cells were measured on leukocytes and normal spermatozoa, respectively.

1308 Human Molecular Genetics, 2009, Vol. 18, No. 7

(9)

ACKNOWLEDGEMENTS

We thank all patients, family members and controls for their participation in this study.

Conflict of Interest statement. All authors had full access to the data and have no conflicts of interests nor any competing finan- cial interests.

FUNDING

This work was supported in part by grants from the Direction de la Recherche Clinique (DRC) of Grenoble CHU and from Organon FARO.

REFERENCES

1. Boivin, J., Bunting, L., Collins, J.A. and Nygren, K.G. (2007) International estimates of infertility prevalence and treatment-seeking:

potential need and demand for infertility medical care.Hum. Reprod., 22, 1506– 1512.

2. Tiepolo, L. and Zuffardi, O. (1976) Localization of factors controlling spermatogenesis in the nonfluorescent portion of the human Y chromosome long arm.Hum. Genet.,34, 119 – 124.

3. Dam, A.H., Koscinski, I., Kremer, J.A., Moutou, C., Jaeger, A.S., Oudakker, A.R., Tournaye, H., Charlet, N., Lagier-Tourenne, C., van Bokhoven, H. and Viville, S. (2007) Homozygous mutation in SPATA16 is associated with male infertility in human globozoospermia.

Am. J. Hum. Genet.,81, 813 – 820.

4. Dieterich, K., Soto Rifo, R., Faure, A.K., Hennebicq, S., Ben Amar, B., Zahi, M., Perrin, J., Martinez, D., Sele, B., Jouk, P.S.et al.(2007) Homozygous mutation of AURKC yields large-headed polyploid spermatozoa and causes male infertility.Nat. Genet.,39, 661 – 665.

5. McKusick, V.A. (1986) Infertility associated with multi-tailed spermatozoa and excessive DNA. OMIM Online Mendelian Inheritance in man.

6. Nistal, M., Paniagua, R. and Herruzo, A. (1977) Multi-tailed spermatozoa in a case with asthenospermia and teratospermia.Virchows Arch. B Cell Pathol.,26, 111 – 118.

7. Escalier, D. (1983) Human spermatozoa with large heads and multiple flagella: a quantitative ultrastructural study of 6 cases.Biol. Cell,48, 65 – 74.

8. Benzacken, B., Gavelle, F.M., Martin-Pont, B., Dupuy, O., Lievre, N., Hugues, J.N. and Wolf, J.P. (2001) Familial sperm polyploidy induced by genetic spermatogenesis failure: case report.Hum. Reprod.,16, 2646– 2651.

9. In’t Veld, P.A., Broekmans, F.J., de France, H.F., Pearson, P.L., Pieters, M.H. and van Kooij, R.J. (1997) Intracytoplasmic sperm injection (ICSI) and chromosomally abnormal spermatozoa.Hum. Reprod.,12, 752 – 754.

10. Pieters, M.H., Speed, R.M., de Boer, P., Vreeburg, J.T., Dohle, G. and In’t Veld, P.A. (1998) Evidence of disturbed meiosis in a man referred for intracytoplasmic sperm injection.Lancet,351, 957.

11. Guthauser, B., Vialard, F., Dakouane, M., Izard, V., Albert, M. and Selva, J. (2006) Chromosomal analysis of spermatozoa with normal-sized heads in two infertile patients with macrocephalic sperm head syndrome.Fertil.

Steril.,85, 750 e755 – 750 e757.

12. Devillard, F., Metzler-Guillemain, C., Pelletier, R., DeRobertis, C., Bergues, U., Hennebicq, S., Guichaoua, M., Sele, B. and Rousseaux, S.

(2002) Polyploidy in large-headed sperm: FISH study of three cases.Hum.

Reprod.,17, 1292– 1298.

13. Perrin, A., Morel, F., Moy, L., Colleu, D., Amice, V. and Braekeleer, M.D. (2007) Study of aneuploidy in large-headed, multiple-tailed spermatozoa: case report and review of the literature.Fertil. Steril.,90, 1201.e13– 1201.e17.

14. Mateu, E., Rodrigo, L., Prados, N., Gil-Salom, M., Remohi, J., Pellicer, A.

and Rubio, C. (2006) High incidence of chromosomal abnormalities in large-headed and multiple-tailed spermatozoa.J. Androl.,27, 6 – 10.

15. Achard, V., Paulmyer-Lacroix, O., Mercier, G., Porcu, G., Saias-Magnan, J., Metzler-Guillemain, C. and Guichaoua, M.R. (2007) Reproductive failure in patients with various percentages of macronuclear spermatozoa:

high level of aneuploid and polyploid spermatozoa.J. Androl.,28, 600 – 606.

16. Kimura, M., Matsuda, Y., Yoshioka, T. and Okano, Y. (1999) Cell cycle-dependent expression and centrosome localization of a third human aurora/Ipl1-related protein kinase, AIK3.J. Biol. Chem.,274, 7334 – 7340.

17. Tang, C.J., Lin, C.Y. and Tang, T.K. (2006) Dynamic localization and functional implications of Aurora-C kinase during male mouse meiosis.

Dev. Biol.,290, 398 – 410.

18. Yan, X., Cao, L., Li, Q., Wu, Y., Zhang, H., Saiyin, H., Liu, X., Zhang, X., Shi, Q. and Yu, L. (2005) Aurora C is directly associated with Survivin and required for cytokinesis.Genes Cells,10, 617 – 626.

19. Keel, B.A. (2004) How reliable are results from the semen analysis?

Fertil. Steril.,82, 41 – 44.

20. Orphanet. (2008) Pre´valence des maladies rares: Donne´es

bibliographiques. Mai 2008.Les Cahiers d’Orphanet, http://www.orpha.

net/orphacom/cahiers/docs/FR/Prevalence_des_maladies_rares_par_

prevalence_decroissante_ou_cas.pdf.

21. Yan, X., Wu, Y., Li, Q., Cao, L., Liu, X., Saiyin, H. and Yu, L. (2005) Cloning and characterization of a novel human Aurora C splicing variant.

Biochem. Biophys. Res. Commun.,328, 353 – 361.

22. Assou, S., Anahory, T., Pantesco, V., Le Carrour, T., Pellestor, F., Klein, B., Reyftmann, L., Dechaud, H., De Vos, J. and Hamamah, S. (2006) The human cumulus – oocyte complex gene-expression profile.Hum. Reprod., 21, 1705 – 1719.

23. Uzbekova, S., Arlot-Bonnemains, Y., Dupont, J., Dalbies-Tran, R., Papillier, P., Pennetier, S., Thelie, A., Perreau, C., Mermillod, P., Prigent, C. and Uzbekov, R. (2008) Spatio-temporal expression patterns of aurora kinases a, B, and C and cytoplasmic polyadenylation-element-binding protein in bovine oocytes during meiotic maturation.Biol. Reprod.,78, 218 – 233.

24. Keen, N. and Taylor, S. (2004) Aurora-kinase inhibitors as anticancer agents.Nat. Rev. Cancer,4, 927 – 936.

25. Gautschi, O., Heighway, J., Mack, P.C., Purnell, P.R., Lara, P.N. Jr and Gandara, D.R. (2008) Aurora kinases as anticancer drug targets.Clin.

Cancer Res.,14, 1639– 1648.

26. Yurov, Y.B., Saias, M.J., Vorsanova, S.G., Erny, R., Soloviev, I.V., Sharonin, V.O., Guichaoua, M.R. and Luciani, J.M. (1996) Rapid chromosomal analysis of germ-line cells by FISH: an investigation of an infertile male with large-headed spermatozoa.Mol. Hum. Reprod.,2, 665 – 668.

27. Kimmins, S., Crosio, C., Kotaja, N., Hirayama, J., Monaco, L., Hoog, C., van Duin, M., Gossen, J.A. and Sassone-Corsi, P. (2007) Differential functions of the aurora-B and aurora-C kinases in Mammalian spermatogenesis.Mol. Endocrinol.,21, 726 – 739.

28. Sasai, K., Katayama, H., Stenoien, D.L., Fujii, S., Honda, R., Kimura, M., Okano, Y., Tatsuka, M., Suzuki, F., Nigg, E.A.et al.(2004) Aurora-C kinase is a novel chromosomal passenger protein that can complement Aurora-B kinase function in mitotic cells.Cell. Motil. Cytoskeleton,59, 249 – 263.

29. Wang, Y., Toppari, J., Parvinen, M. and Kallio, M.J. (2006) Inhibition of Aurora kinases perturbs chromosome alignment and spindle checkpoint signaling in rat spermatocytes.Exp. Cell. Res.,312, 3459 – 3470.

30. Kahraman, S., Akarsu, C., Cengiz, G., Dirican, K., Sozen, E., Can, B., Guven, C. and Vanderzwalmen, P. (1999) Fertility of ejaculated and testicular megalohead spermatozoa with intracytoplasmic sperm injection.

Hum. Reprod.,14, 726 – 730.

31. Kahraman, S., Sertyel, S., Findikli, N., Kumtepe, Y., Oncu, N., Melil, S., Unal, S., Yelke, H. and Vanderzwalmen, P. (2004) Effect of PGD on implantation and ongoing pregnancy rates in cases with predominantly macrocephalic spermatozoa.Reprod. Biomed. Online,9, 79 – 85.

32. WHO. (1999)World Health Organization WHO laboratory manual for the examination of human semen and sperm – cervical mucus interaction., Cambridge University Press, Cambridge.

33. Vindelov, L.L., Christensen, I.J. and Nissen, N.I. (1983) A detergent- trypsin method for the preparation of nuclei for flow cytometric DNA analysis.Cytometry,3, 323 – 327.