Flow cytometry isolation and reverse transcriptase–polymerase chain reaction characterization of human round spermatids in infertile patients

Ahmed Ziyyat1, Bruno Lassalle1, Jacques Testart1,2, Pascal Briot2, Edouard Amar2, Catherine Finaz1 and Annick Lefèvre1,3

1 Unité Maturation Gamètique et Fécondation, INSERM U 355 and Institut Fédératif de Recherche sur les Cytokines, IFR 13, 32 rue des Carnets, 92140 Clamart and 2 Centre de Médecine de la Procréation, Hôpital Américain, 63, Boulevard Victor Hugo, 92202 Neuilly, France


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Flow cytometry coupled to cell sorting is proposed as a method to isolate round spermatids from testicular biopsies in obstructive azoospermic patients. The cells were separated on the basis of their size and density only. We obtained homogenous populations of alive round spermatids free of lymphocytes and diploid germ cells. The detection of protamine 1 gene (PRM1) and PRM2 expression in the sorted cells proves that these cells are round spermatids. On the contrary, neither the expression of CD3-{delta}, which is specific to lymphoid cells, nor that of MAGE1, which has been demonstrated in diploid germ cells, could be observed in the round spermatid population even after using a nested polymerase chain reaction (PCR) assay. The flow cytometry procedure failed to isolate round spermatids from ejaculates in non-obstructive azoospermic patients. In >39 ejaculates tested by reverse transcriptase–PCR, only nine revealed the presence of some round spermatids, as demonstrated by the expression of PRM1. However, these round spermatids did not express PRM2.

Key words: flow cytometry/obstructive azoospermia/protamine/round spermatids


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The round spermatid is the result of the second meiotic division in males. It is the youngest male germ cell with a set of haploid chromosomes that differentiates into a spermatozoon to acquire its fertilizing ability. Sofikitis et al. (1994) were the first to report a completed pregnancy in mammals achieved from the injection of a round spermatid into a rabbit oocyte. Ogura et al. (1994) and Kimura et al. (1995) obtained the birth of normal fertile mice of both sexes after electrofusion or injection of mouse oocytes with round spermatids respectively. Although the success rates were very low, these experiments showed that there is no genetic barrier to fertilization by round spermatids. Studies on electrical oocyte activation before round spermatid nuclear injection showed that a simple technical amelioration alone have led to a considerably improved success rate (Sofikitis et al., 1996Go). The authors of these pioneering experiments concluded that the nucleus of the round spermatid is genetically ready to participate in the normal fertilization process.

These results have offered new prospects for the treatment of male infertility due to defective spermiogenesis in humans. Vanderzwalmen et al. (1995) were the first to succeed in fertilizing a human oocyte by a late-stage spermatid (obtained from testicular biopsy) that was microinjected. All oocytes cleaved further to 4-cell embryos. Fishel et al. (1995) reported the implantation of such embryos after uterine transfer. The first birth of a healthy child after round spermatid injection into human oocytes (Tesarik et al., 1995Go) confirmed the feasibility of this novel approach in the treatment of non-obstructive azoospermia. In another paper, Tesarik et al. (1996) provided a complete documentation of the series of 11 cases using spermatids retrieved from ejaculated semen. In four cases, elongated spermatids were used, whereas in the other seven cases, round spermatids were injected. Two pregnancies were achieved, resulting in the birth of two normal boys. These first results have encouraged many clinics to introduce in their intracytoplasmic sperm injection (ICSI) programme the treatment of azoospermic patients using spermatids. Thus, Araki et al. (1997) reported three cases of successful paternity, achieved by intracytoplasmic injection of late spermatids leading to the birth of four healthy boys and girls. Several other pregnancies and births were then obtained after injection of either elongated or round spermatids, both retrieved from testicular biopsies (Antinori et al., 1997aGo; Vanderzwalmen et al., 1997Go; Kahraman et al., 1998Go; Sofikitis et al., 1998Go).

Currently, it is not clear which source is preferable to obtain spermatids in testicular or ejaculated samples. If the testis is the source of spermatids, several biopsies may be necessary before a pregnancy could be finally achieved even when the recovered spermatids are frozen. One successful fertilization and pregnancy has been obtained after injection of freeze–thawed round spermatids (Antinori et al., 1997bGo). On the other hand, Tesarik et al. (1996) advocate the use of spermatids recovered from the ejaculate, as the least invasive approach sparing the residual foci of spermatogenesis eventually present in the testis. Antinori et al. (1997a,b) reported the complete absence of spermatids after the extensive examination of the ejaculate of all the patients entering their spermatid ICSI programme.

The main point clinicians are very much concerned with is the difficulty of identifying with certainty round spermatids even under a Hoffman modulation contrast microscope (Vanderzwalmen et al., 1997Go). As reported by Yamanaka et al. (1997), the qualitative criteria retained, i.e. a regular zone of cytoplasm surrounding a round nucleus and a developing acrosome structure, are highly susceptible of intra- and extra-observer variations. It may result in recovery of round cells rather than round spermatids, and this may suggest one reason for the low fertilization rate with round spermatids.

Therefore, the present study has been designed to answer the following puzzling questions: are we able to propose a fast and reliable procedure to isolate pure populations of viable well-characterized round spermatids, and are ejaculates from non-obstructive azoospermic patients a possible alternative source of round spermatids? We have used flow cytometry/cell sorting to purify human round spermatids from testicular biopsies and ejaculates from infertile patients, previously optimized with mouse spermatids (Lassalle et al., 1999Go). The presence of round spermatids has been demonstrated by the expression of protamine 1 gene (PRM1) and PRM2. During the time we were writing this article, Aslam et al. (1998) have published a comparative study between velocity sedimentation under unit gravity and fluorescent activated cell sorting but they have identified the cells only on the basis of their morphological characteristics.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Patients
A total of 51 patients was involved in this study. They were consulted and informed about the nature of this study. All gave their consent.

Testicular biopsies
Testicular tissues used for round spermatid isolation were obtained from patients undergoing testicular biopsy retrieval for infertility treatment by ICSI. The seven cases retained for the study were obstructive azoospermia, showing normal spermatogenesis with all types of germ cells, motile spermatozoa included. Biopsies were performed through standard open surgical technique and the retrieved tissue was immersed and rinsed twice in PBS (Dulbecco's phosphate-buffered saline; Sigma; Saint Quentin, Fallavier, France), minced into small pieces using a pair of fine scissors and placed in 1 ml PBS at room temperature. Continuous pipetting for 1 min enabled spermatogenic cells to be released into the medium. The resulting suspension was left for 15 min at room temperature to allow the sedimentation of the remaining large fragments of intact tubules. The supernatant was centrifuged at 600 g for 5 min and the pellet was resuspended in 1 ml PBS. To increase the proportion of round spermatids, prior to flow cytometry and cell sorting, the cell suspension was centrifuged on a discontinuous Percoll (Pharmacia, Uppsala, Sweden) gradient (45, 30, 22, and 15%) as previously described (Lassalle et al., 1999Go). After centrifugation, the 22% Percoll fraction, where most of the round spermatids sedimented, was analysed by flow cytometry.

Ejaculate samples
Ejaculates from 44 patients with non-obstructive azoospermia were analysed in order to look for spermatids. The mean age of these patients was 33.8 years (range, 26–48 years). Chromosome analysis was performed on 28 patients and all their karyotypes were normal (46,XY). Lysis of erythrocytes was performed as previously described (Sofikitis et al., 1994Go). The cells were washed twice with PBS, resuspended in 1 ml of PBS and analysed via flow cytometry (14 of them) and/or kept at –80°C for further mRNA extraction and RT–PCR treatment.

Lymphoid cells recovery
Haemolysed human blood was centrifuged over a discontinuous gradient (90, 70, 45 and 22%) of Percoll and lymphoid cells were recovered in the 70% Percoll fractions as already described by Kolb et al. (1993). Monoclonal mouse anti-human CD 45 fluoroscein isothiocyanate (FITC)-conjugated antibody (PharMingen, San Diego, USA) was used (dilution: 1/100 in PBS; incubation 30 min at 37°C) to confirm the presence of lymphoid cells.

Microscope observations and cell size measurement
Ten microlitres of the cell suspensions were placed between a slide and coverslip. Small amounts of vaseline were applied on the slide at four points around the droplet containing cells before the coverslip was placed on top and pressed slightly to limit cell crushing. The diameter of cells was measured with x1000 magnification using the micrometer in one eyepiece of a phase-contrast microscope. One graduation of the ocular micrometric scale was estimated to 0.97 µm.

Flow cytometric analysis and cell sorting
Cell analysis and sorting were performed on a FACScalibur flow cytometer (Becton Dickinson, San Jose, CA, USA) equipped with a cell sorting system. The cell size (forward angle light scatter, FSC) and the cell density (90° light scatter, SSC) were simultaneously measured. The FACScalibur instrument settings were adjusted as already described by Lassalle et al. (1999). Ten thousand events were analysed per sample. The sorted cells were recovered in 50 ml culture tubes previously coated with bovine serum albumin or fetal calf serum (PBS plus 4% BSA or FCS, 24 h at 4°C). BSA or FCS concentration (~4%) must be maintained during the entire cell sorting procedure to limit the adhesion of spermatids on the surface of the plastic tube.

Assessment of viability
After cell sorting, viability of round spermatids was determined by the Trypan Blue exclusion test (Talbot and Chacon, 1981Go).

RT–PCR analysis
Extraction of mRNA
PolyA+ mRNA were isolated from either all ejaculated cells or FACSCalibur purified round spermatids originating from biopsies, using the mRNA capture kit (Boehringer Mannheim), as previously described (Lassalle et al., 1999Go).

One-step reverse transcription and PCR amplification
Reverse transcription and PCR were done in one step using the Titan RT–PCR system from Boehringer Mannheim, according to the manufacturer information as previously descibed (Lassalle et al., 1999Go). Outer and nested primer sequences for glucose-6-phosphate dehydrogenase (G6PDH), protamine 1 (PRM1), protamine 2 (PRM2), MAGE-1 and CD3-{delta}, annealing temperatures and sizes of PCR products are shown in Table IGo.


View this table:
[in this window]
[in a new window]
 
Table I. Details of primers
 
In an attempt to increase the sensitivity for the detection of these gene transcripts, 2.5 µl of cDNA obtained from the first amplification served as template for a second DNA amplification reaction, using inner nested primers (Table IGo). The nested PCR conditions were as previously described (Lassalle et al., 1999Go).

A simultaneous reaction in which reverse transcriptase enzyme was heat-inhibited (for 5 min at 94°C) prior to the PCR was run as a control for the presence of DNA. Amplification of cDNA from whole testis served as a positive control. For all amplifications, negative controls (water only) were included (data not shown). Primers for the amplification of the ubiquitous G6PDH gene were used as a control for the synthesis of cDNA. The PCR products (10 µl of each) were analysed on 2% agarose gel stained with ethidium bromide and molecular sizes were determined with the molecular weight marker f X 174 Hae Digest (Sigma).

Statistical analysis
Cell size data were analysed using Statview software. Means (± SEM) were compared using Student's t-test.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Testicular biopsies from obstructive azoospermia
Isolation of round spermatids
After manual mincing and moderate pipetting, the cell suspension obtained from the testicular biopsies of obstructive azoospermic patients consisted mainly of spermatozoa, elongating and round spermatids and primary spermatocytes. This cell suspension was first enriched 2-fold in round spermatids by centrifugation on a discontinuous Percoll gradient. Round spermatids sedimented mainly in the 22% fraction according to the results of Meistrich et al. (1981) in the rat and Lassalle et al. (1999) in the mouse, while lymphocytes were found preferentially in the 70% fraction as expected (Kolb et al., 1993Go).

When the 22% Percoll fraction was subjected to flow cytometry coupled with cell sorting, round spermatids were recovered as a homogeneous population in a delimited area localized between 400 and 600 on the FSC axis (arbitrary units). The dot plot diagrams were very similar from one patient to another (Figure 1Go), except that the round spermatid populations were more or less abundant. Lymphocytes obtained from human blood, which served as control, were found in a well-defined area at ~200 on the FSC axis (Figure 2AGo). This was corroborated by the use of a mouse monoclonal anti-human CD45 antibody conjugated with FITC (Figure 2BGo). Round spermatids and lymphocytes were sorted in distinct areas as a consequence of their different sizes. Round spermatid populations were very homogeneous with an average size of 10.6 ± 0.5 µm (Figure 3CGo). Typical characteristics of round spermatids were observed: a round shape with a smooth outline, a round nucleus showing a distinct eccentric nucleolus, a well-formed acrosomal cap, an acrosomal granule and a continuous zone of cytoplasm surrounding the nucleus. More than 99% of the sorted spermatids retained their viability as assessed by the Trypan Blue exclusion test. Figure 3Go shows three stages of haploid cells observed under a phase-contrast microscope. Figure 3AGo shows an elongating spermatid from the 22% Percoll gradient fraction. Figure 3B and CGo represents round spermatids isolated by the FACScalibur procedure; on spermatid B the acrosomal cap is clearly visible, whereas on spermatid C the early acrosomal vesicle is present as a bright white spot. The lymphocytes were much smaller in size, 7.1 ± 0.7 µm, and exhibited a larger nucleocytoplasmic ratio (Figure 3DGo).



View larger version (59K):
[in this window]
[in a new window]
 
Figure 1. Flow cytometric analysis of spermatogenetic cell populations obtained from human testicular biopsies (four representative cases). The round spermatid population is easily recognizable on the dot plot diagram as indicated by arrows.

 


View larger version (21K):
[in this window]
[in a new window]
 
Figure 2. Flow cytometric analysis of lymphoid cells isolated from human blood. Lymphoid cells were stained with a mouse monoclonal anti-human CD45 antibody conjugated with fluorescein isothiocyanate (FITC), then analysed by flow cytometry on the basis of (A) FSC/SSC (cell size/cell density) and (B) FSC/FL1 (cell size/relative FITC fluorescence) parameters.

 


View larger version (47K):
[in this window]
[in a new window]
 
Figure 3. Phase-contrast microscopy of human spermatid cells. (A) Elongating spermatid. (B) Round spermatid with a well-formed acrosomal cap. (C) Round spermatid showing an acrosomal vesicle. (D) Human lymphoid cell. Scale bar = 10 µm.

 
RT–PCR characterization of the sorted cells
mRNA corresponding to 3000 cells of the purified round spermatid population were analysed for the presence of PRM1 and PRM2, MAGE1 and CD3-{delta} transcripts. All primers were designed to span one or more introns (Table IGo) so that the PCR products could be distinguished from possible genomic amplified products by size on an agarose gel. No products deriving from genomic DNA templates were obtained. In addition, experiments omitting the reverse transcription step confirmed the specificity of the products to mRNA. Blanks consisting of PCR reaction mixture without added template were always included to ensure absence of contamination.

Both PRM1 and PRM2 transcripts were detected in the round spermatid population (Figure 4Go). mRNA isolated from a single cell gave a positive signal when performing a nested PCR assay with PRM1 primers (14 independent experiments were done; data not shown). After a first round of PCR, mRNA representing MAGE1, which is expressed in spermatogonia and primary spermatocytes (Takahashi et al., 1995Go), were detected in total testis, but not in the round spermatid population, even after performing a nested primer assay. As expected, transcripts for CD3-{delta} were detected in lymphocytes after one round of PCR, but not in round spermatids as confirmed by performing a second round of PCR with nested primers.



View larger version (24K):
[in this window]
[in a new window]
 
Figure 4. Reverse transcriptase–polymerase chain reaction (RT–PCR) analysis of PRM1, PRM2, MAGE1 and CD3-{delta} expression. RT–PCR was performed on poly A+ mRNA from 3000 round spermatids or lymphocytes and on 1 µg of total mRNA from whole testis. Lane 1 and 2, RT+ and RT– round spermatid samples with PRM1 primers. Lanes 3 and 4, RT+ and RT– round spermatid samples with PRM2 primers. Lanes 5, RT+ round spermatid sample with MAGE1 nested primers. Lane 6, RT+ whole testis sample with MAGE1 nested primers. Lane 7, RT+ round spermatid sample with CD3-{delta} nested primers. Lanes 8 and 9, RT+ lymphocyte samples with CD3-{delta} outer and nested primers respectively. M, f Hae III marker. PRM1 PCR product was 150 bp; PRM2 PCR product was 301 bp; MAGE1 nested PCR product was 213 bp; CD3-{delta} product was 225 bp; CD3-{delta} nested PCR product was 189 bp.

 
Ejaculates from non-obstructive azoospermy
Figure 5Go shows the dot plot diagrams obtained when ejaculates of non-obstructive azoospermic patients were analysed by flow cytometry. For the 14 ejaculates tested, no spot appeared in the area where round spermatids were expected. Seven patients exhibited a spot of lymphocytes. Most of the ejaculates were heavily contaminated with bacteria and cell debris. This absence of spot in the area where round spermatids are expected was also observed when normospermic ejaculates were subjected to flow cytometry/cell sorting.



View larger version (48K):
[in this window]
[in a new window]
 
Figure 5. Flow cytometric analysis of ejaculate cells from eight representative non-obstructive azoospermic patients. No round spermatid spots could be visualized. Arrows indicate the lymphoid cell spots.

 
Thirty-nine crude ejaculates were analysed using a nested PCR protocol for the expression of PRM1 and PRM2. Primers for the amplification of the ubiquitous G6PDH gene were used as a control for the synthesis of cDNA and the 207 bp fragment attended was generated in all the samples (Figure 6Go). Two samples only showed the 150 bp fragment corresponding to the outer primers PRM1 amplicon, while seven others were weakly positive after a second round of PCR with semi-nested primers. Only two of these positive samples in the semi-nested PCR assay belong to the group of 14 patients previously assayed with flow cytometry, the 12 others being negative. mRNA corresponding to PRM2 was not detected in any sample, even after nested PCR (data not shown).



View larger version (22K):
[in this window]
[in a new window]
 
Figure 6. Analysis of PRM1 transcription in non-obstructive azoospermic patient ejaculates. Reverse transcriptase–polymerase chain reaction (RT–PCR) was performed on poly A+ mRNA from whole ejaculates. (A) Oligonucleotide primers for G6PDH were included in each analysis as a control for the presence of cDNA. Lanes 1–10, patient samples with G6PDH inner primers. Lane 11, whole testis sample with outer and nested G6PDH primers. G6PDH PCR product was 338 bp; nested G6PDH PCR product was 207 bp. (B) Lanes 1 and 2, patient samples with PRM1 outer primers. Lane 2', patient 2 sample with PRM1 nested primers. Lanes 3–9, patient samples with PRM1 nested primers. Lane 10, one of the 39 ejaculates tested that do not express PRM1 in a nested PCR assay. Lane 11, whole testis sample with outer and nested PRM1 primers. PRM1 PCR product was 150 bp; nested PRM1 PCR product was 128 bp. M, f Hae III marker.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
When microinjected into mammalian oocytes, spermatids were able to fertilize and to produce live offspring (Ogura et al., 1994Go; Sofikitis et al., 1994Go; Kimura et al., 1995). It has been suggested that spermatid injection may be the treatment of choice for patients having defective spermatogenesis (Edwards, 1994; Sofikitis et al., 1994Go). In cases of azoospermia caused by testicular failure including Sertoli cell-only syndrome, maturation arrest, cryptorchid testicular atrophy or Klinefelter's syndrome, areas of some persisting spermatogenesis can be detected eventually after multiple testicular biopsies (Devroey et al., 1995Go; Silber et al., 1996Go). The few spermatozoa retrieved can result in normal pregnancy after ICSI (Silber et al., 1996Go). However, in 40% of azoospermic men with germinal failure, no mature spermatozoa could be found, despite the presence of tubules with complete spermatogenesis in a previous biopsy for some of them. In fact, fertilization and pregnancies can be obtained using elongating and elongated spermatids with an acceptable implantation rate, as compared to conventional ICSI (Fishel et al., 1995Go, 1997Go; Vanderzwalmen et al., 1995Go, 1997Go; Antinori et al., 1997aGo; Araki et al., 1997Go; Kahraman et al., 1998Go; Sofikitis et al., 1998Go).

On the other hand, even though many attempts have been made, the capacity of round spermatids for achieving fertilization and pregnancy is disappointing, and the birth of healthy babies after ROSI (round spermatid injection) remains exceptional (Tesarik et al., 1995Go; Vanderzwalmen et al., 1997Go; Antinori et al., 1997bGo; Kahraman et al., 1998Go). The causes of such a relative failure are puzzling. It is generally agreed that it is very difficult to differentiate with certainty a round spermatid from the variety of `round cells' present either in testicular biopsies or in ejaculates. The resulting confusion may account, in part, for the inefficiency of round spermatids to achieve pregnancy (Silber and Johnson, 1998Go). None of the techniques utilized so far have provided homogeneous populations of round spermatids. Tesarik and Mendoza (1996) admitted that the round spermatid enriched fraction utilized in their ROSI programme contained only 1–5% spermatids.

Therefore, the development of a reliable technique for the isolation of round spermatids appears to be essential if microinjection is to be proposed as a new treatment for male infertility. The aim of our work was to find a reliable method of obtaining populations of human round spermatids from either testicular biopsies or ejaculates, with the exclusion of other round cell types, via flow cytometry coupled to cell sorting.

Even though a greater number of cells can be retrieved by using enzymatic digestion (Crabbé et al., 1997), we have favoured the mechanical dissociation of the testicular biopsy in order to avoid a possible alteration of the cell membrane. As previously described for the mouse (Lassalle et al., 1999Go), the centrifugation of the cell suspension obtained on a discontinuous Percoll gradient prior to flow cytometry resulted in a significant enrichment in round spermatids. In agreement with results in rat (Meistrich et al., 1981Go) and in mouse (Lassalle et al., 1999Go), most of the human round spermatids were recovered in the 22% Percoll fraction, while other workers attested their presence in the 70% Percoll fraction (Angelopoulos et al., 1997Go; Vanderzwalmen et al., 1997Go; Kahraman et al., 1998Go). In our own experience, and according to Kolb et al. (1993), lymphocytes prepared from blood sedimented in that 70% Percoll gradient. The testis 22% Percoll fraction could be resolved in a very homogeneous population, showing all the characteristics assigned to round spermatids: a round shape, a round nucleus with a clearly visible nucleolus and surrounded by a continuous zone of cytoplasm and an acrosomal structure. Indeed, the sorted round spermatids clearly showed the characteristic cape covering half the nucleus surface, surmounted by the acrosomal granule which is the typical feature of a round spermatid in an advanced stage of development (i.e. step 5 to 7 of the cap phase, just before its elongating process starts) as initially described by Clermond and Leblond (1955). The diameter of the round spermatids obtained via the flow cytometry procedure is 10.6 ± 0.5 µm. It is close to the observations of Dadoune (1996), Fishel et al. (1997) Kahraman et al. (1998) and Verheyen et al. (1998), but it differs from the descriptions of Angelopoulos et al. (1997), Vanderzwalmen et al. (1997) and Tesarik et al. (1998) who attributed to round spermatids an average size of 5–8 µm. It is noteworthy that Mendoza and Tesarik (1996) ascribed to the cells they have isolated the features of round spermatids in an early stage of development, i.e. the Golgi phase of spermiogenesis.

When subjected to flow cytometry, the lymphocyte population (average size, 7.1 ± 0.7 µm) was sorted in a well-defined area, which cannot be confused with that of spermatids. In cases of activation (inflammatory process, lymphoma), the size of lymphocytes could be slightly increased. Nonetheless, the size of the lymphoid cells we have observed in the ejaculate of non-obstructive azoospermic patients was similar to that of circulating blood lymphocytes.

The discrepancies concerning the size of the round spermatids coupled with the remaining uncertainty affecting the recognition of round spermatids admitted by various authors (Yamanaka et al., 1997Go; Silber et al., 1998; Vanderzwalmen et al., 1998Go) enabled us to establish, without doubt, the identity of the cells we have sorted, using uncontroversial parameters such as the presence of mRNA corresponding to genes whose expression is restricted to the haploid germ cells, i.e. PRM1 and PRM2. Such a proof would not be susceptible to the intra- and extra-observer variations resulting from visual identification under a microscope. The results of RT–PCR experiments clearly demonstrate that the sorted cell population expressed both PRM1 and PRM2.

It was also of great importance to demonstrate that the population of round spermatids sorted was not heavily contaminated with other cell species. At the time ROSI is to be performed, the round spermatid chosen for microinjection can only be identified by eye. It is therefore essential that the source of spermatids should be as free of contaminating cells as possible, so that there will be no risk of selecting a wrong cell. To evaluate a possible contamination with either lymphocytes or diploid germ cells, we performed nested RT–PCR with primers specific to the two following genes: CD3-{delta}, which codes for an antigen expressed ubiquituously at the surface of lymphoid cells (Van Den Elsen et al., 1986); MAGE1, which codes for a tumour rejection antigen expressed in various cancers and also in spermatogonia and primary spermatocytes, with the exception of haploid germ cells (Takahashi et al., 1995Go). Taking advantage of the extreme sensitivity of the nested one-step RT–PCR technique which can reveal the expression of a gene in a single cell, we proved that the round spermatid populations obtained via the flow cytometry procedure was free of either lymphocytes, spermatogonia or spermatocytes.

Once the standard parameters to isolate homogenous populations of spermatids from testicular biopsy cell suspensions has been established, we applied the procedure to the isolation of round spermatids from azoospermic patient ejaculates. Actually, the presence of spermatids in the ejaculates of these patients is somewhat controversial. Mendoza and Tesarik (1996) attested their presence in 69% of the 124 observed azoospermic patients, whereas, after an extensive examination of the ejaculate of 36 patients, Antinori et al. (1997a) concluded that there was a complete absence of spermatogenic cells. Nevertheless, it is tempting to consider ejaculates as a favourable source of spermatids as this would overcome the risks associated with extensive sampling of testicular tissue in search of germ cells. Unfortunately, cell sorting coupled to flow cytometry did not permit the isolation of a spermatid population in the 14 ejaculates from the non-obstructive azoospermic patients analysed. In addition, among the 39 crude ejaculates analysed by RT–PCR, only nine were positive with the primers for PRM1. This means that only 23% of the non-obstructive azoospermic patients recruited in our ICSI centre exhibited some spermatids in their ejaculates and that these spermatids were too few to be isolated via the flow cytometry procedure.

Furthermore, these spermatids did not express protamine 2. These results are to be brought together with those of Yebra et al. (1998), who showed a marked reduction of the PRM2 protein content in sperm cells of infertile patients. This poses the question of spermatid quality in ROSI programmes. Is this failure in expressing PRM2 exceptional or widely distributed among non-obstructive azoospermic patients? Does it affect only ejaculated spermatids? Could it account in part for the very low fertilizing efficiency of spermatids? Are there other genes that are not normally expressed in the haploid germ cells of these patients?

To sum up, flow cytometry coupled to cell sorting seems to be the more powerful technique to provide homogeneous populations of spermatids from biopsies of men with obstructive azoospermia. The procedure offers decisive advantages, as it is fast, reliable, requires little tissue, as compared with elution which is time consuming and needs much more starting material.


    Acknowledgments
 
Ahmed Ziyyat was supported by a fellowship from the American Hospital of Paris (Neuilly sur Seine, 92202, France) and by a grant from the Ministère de l'Education Nationale et de la Recherche Scientifique (ACC SV4).


    Notes
 
3 To whom correspondence should be addressed at: INSERM unité 355, 32 rue des Carnets, 92140 Clamart, France Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Angelopoulos,T., Krey, L., Mc Cullough, A. et al. (1997) A simple and objective approach to identifying human round spermatids. Hum. Reprod., 12, 2208–2216.[Abstract]

Antinori, S., Versaci, C., Dani, G. et al. (1997a) Fertilization with human testicular spermatids: four successful pregnancies. Hum. Reprod., 12, 286–291.[Abstract]

Antinori, S., Versci, C., Dani, G. et al. (1997b) Successful fertilization and pregnancy after injection of frozen–thawed round spermatids into human oocytes. Hum. Reprod., 12, 554–556.[ISI][Medline]

Araki, Y., Motoyama, M., Yoshida, A. et al. (1997) Intracytoplasmic injection with late spermatids: a successful procedure in achieving childbirth for couples in which the male partner suffers from azoospermia due to deficient spermatogenesis. Fertil. Steril., 67, 559–561.[ISI][Medline]

Aslam, I., Robins, A., Dowell, K. et al. (1998) Isolation, purification and assessment of viability of spermatogenic cells from testicular biopsies of azoospermic men. Hum. Reprod., 13, 639–645.[Abstract]

Chen, E.Y., Cheng, A., Lee, A. et al. (1991) Sequence of human glucose-6-phosphate dehydrogenase cloned in plasmids and a yeast artificial chromosome. Genomics, 10, 792–800.[ISI][Medline]

Clermont, Y. and Leblond, C.P. (1955) Spermiogenesis of man, monkey, ram and other mammals as shown by the periodic acid–Schiff technique. Am. J. Anat., 96, 229–254.[ISI]

Crabbe, E., Verheyen, G., Tournaye, H. et al. (1997) The use of enzymatic procedures to recover testicular germ cells. Hum. Reprod., 12, 1682–1687.[Abstract]

Dadoune, J. P. (1997) La spermatide, cette méconnue. Andrologie, 7, 22–33.

Devroey, P. Liu, J., Nagy, Z. et al. (1995) Pregnancies after testicular sperm extraction (TESE) and intracytoplasmic sperm injection (ICSI) in non obstructive azoospermia. Hum. Reprod., 10, 1457–1460.[Abstract]

Domenjoud, L., Nussbaum, G., Adham, I. et al. (1990) Genomic sequences of human protamines whose genes, PRM1 and PRM2, are clustered. Genomics, 8, 127–133.[ISI][Medline]

Edwards, R., Tarin, J., Dean, N. et al. (1994) Are spermatid injections into human oocytes now mandatory? Hum. Reprod., 9, 2217–2219.[ISI][Medline]

Fishel, S., Green, S., Bishop, M. et al. (1995) Pregnancy after intracytoplasmic injection of spermatid. Lancet, 245, 1641–1642.

Fishel, S., Green, S., Hunter, A. et al. (1997) Human fertilization with round and elongated spermatids. Hum. Reprod., 12, 336–340.[Abstract]

Kahraman, S., Polat, G., Samli, M. et al. (1998) Multiple pregnancies obtained by testicular spermatid injection in combination with intracytoplasmic sperm injection. Hum. Reprod., 13, 104–110.[Abstract]

Kimura, Y. and Yanagimachi, R. (1995) Mouse oocytes injected with testicular spermatozoa or round spermatids can develop into normal offspring. Development, 121, 2397–2405.[Abstract/Free Full Text]

Kolb, J.P., Abadie, A., Paul-Eugène, N. et al. (1993) Ligation of CD23 triggers cyclic AMP generation in human B lymphocytes. J. Immunol., 150, 4798–4809.[Abstract/Free Full Text]

Lassalle, B., Ziyyat, A., Testart, J. et al. (1999) Flow cytometric method to isolate round spermatids from mouse testis. Hum. Reprod., 14, in press.

Meistrich, M., Longtin, J., Brock, W. et al. (1981) Purification of rat spermatogenic cells and preliminary biochemical analysis of these cells. Biol. Reprod., 25, 1065–1077.[ISI][Medline]

Mendoza, C. and Tesarik, Y. (1996) The occurrence and identification of round spermatids in the ejaculate of men with nonobstructive azoospermia. Fertil. Steril., 66, 826–829.[ISI][Medline]

Ogura, A., Matsuda, J. and Yanagimachi, R. (1994) Birth of normal young after electrofusion of mouse oocytes with round spermatids. Proc. Natl. Acad. Sci. USA, 91, 7460–7462.[Abstract]

Silber, S. and Johnson, L. (1998) Are spermatid injections of any clinical value? ROSNI and ROSI revisited. Hum. Reprod., 13, 509–515.[Free Full Text]

Silber, S., Van Steirteghem, A.C., Nagy, Z. et al. (1996) Normal pregnancies resulting from testicular sperm extraction and intracytoplasmic sperm injection for azoospermia due to maturation arrest. Fertil. Steril., 66, 110–117.[ISI][Medline]

Sofikitis, N., Miyagawa, I. Agapitos, E. et al. (1994) Reproductive capacity of the nucleus of the male gamete after completion of meiosis. J. Assist. Reprod. Genet., 11, 325–341.[ISI][Medline]

Sofikitis, N., Toda, T., Miyagawa, I. et al. (1996) Beneficial effects of electrical stimulation before round spermatid nuclei injections into rabbit oocytes on fertilization and subsequent embryonic development. Fertil. Steril., 65, 176–185.[ISI][Medline]

Sofikitis, N., Yamamoto, Y., Miyagawa, I. et al. (1998) Ooplasmic injection of elongating spermatid for the treatment of non-obstuctive azoospermia. Hum. Reprod., 13, 709–714.[Abstract]

Takahashi, K., Shichijo, S., Noguchi, M. et al. (1995) Identification of MAGE-1 and MAGE-4 proteins in spermatogonia and primary spermatocytes of testis. Cancer Res., 16, 3478–3482.

Talbot, P. and Chacon, R.S. (1981) A triple-stain technique for evaluating normal accrosome reaction of human sperm. J. Exp. Zool., 215, 201–208.[ISI][Medline]

Tesarik, Y. and Mendoza, C. (1996) Spermatid injection into oocytes. I. Laboratory techniques and special features of zygote development. Hum. Reprod., 11, 772–779.[Abstract]

Tesarik, J., Mendoza, C. and Testard, J. (1995) Viable embryos from injection of round spermatids into oocytes. N. Engl. J. Med., 333, 525.[Free Full Text]

Tesarik, J., Rolet, F., Brami, C. et al. (1996) Spermatid injection into human oocytes. II. Clinical application in the treatment of infertility due to non-obstuctive azoospermia. Hum. Reprod., 11, 780–783.[Abstract]

Tesarik, Y., Greco, E. and Mendoza, C. (1998) ROSI, instructions for use: 1997 update. Hum. Reprod., 13, 519–523.[ISI][Medline]

Van den Elson, P., Georgopoulos, K., Shepley, B.A. et al. (1986) Exon/intron organization of the genes coding for the delta chains of the human and murine T-cell receptor/T3 complex. Proc. Natl. Acad. Sci. USA, 83, 2944–2948.[Abstract]

Van der Bruggen, P., Traversari, C., Chomez, P. et al. (1991) A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma. Science, 254, 1643–1647.[ISI][Medline]

Vanderzwalmen, P., Lejeune, B., Nijs, M. et al. (1995) Fertilization of an oocyte microinseminated with a spermatid in an in-vitro fertilization programme. Hum. Reprod., 10, 502–503.[ISI][Medline]

Vanderzwalmen, P., Zech, H., Birkenfeld, A. et al. (1997) Intracytoplasmic injection of spermatids retrieved from testicular tissue: influence of testicular pathology, type of selected spermatids and oocyte activation. Hum. Reprod., 12, 1203–1213.[ISI][Medline]

Vanderzwalmen,P., Nijs, M., Schoysman, R. et al. (1998) The problems of spermatid microinjection in the human: the need for an accurate morphological appoach and selective methods for viable and normal cells. Hum. Reprod., 13, 515–519.[ISI][Medline]

Verheyen, G., Crabbé, E., Joris, H. et al. (1998) Simple and reliable identification of the human round spermatid by inverted phase contrast microscopy. Hum. Reprod., 13, 1570–1577.[Abstract]

Yamanaka, K., Sofikitis, N., Miyagawa, I. et al. (1997) Ooplasmic round spermatid nuclear injection procedures as an experimental treatment for nonobstructive azoospermia. J. Assist. Reprod. Genet., 14, 55–62.[ISI][Medline]

Yebra de, L., Ballesca, J.L., Vanrell, J. et al. (1998) Detection of P2 precursors in the sperm cells of infertile patients who have reduced protamine P2 levels. Fertil. Steril., 69, 755–759.[ISI][Medline]

Submitted on July 31, 1998; accepted on October 29, 1998.