Computerized cell-scanning system for evaluating human spermatogenesis in non-obstructive azoospermic patients

Transcription

1 Reproductive BioMedicine Online (2012) 24, ARTICLE Computerized cell-scanning system for evaluating human spermatogenesis in non-obstructive azoospermic patients Deborah Strassburger a, *,1, Alisa Komsky-Elbaz a,1, Malka Reichart b, Arieh Raziel a,1, Esti Kasterstein a,1, Daphna Komarovsky a,1, Orna Bern a,1, Shevach Friedler a,1, Raphael Ron-El a,1 a Infertility and IVF Unit, Assaf Harofeh Medical Center, Zerifin , Israel; b BioView Ltd., Nes Ziona, Israel * Corresponding author. address: embriology@asaf.health.gov.il (D Strassburger). 1 Affiliated to the Sackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel. Deborah Strassburger received her PhD from the Sackler Faculty of Medical Sciences in 1994 for her thesis on immunology and pregnancy loss. Since then she has served as Head of the IVF laboratory at the Infertility and IVF Unit of Assaf Harofeh Medical Center, Tel Aviv University, Israel. Deborah is a lecturer in human reproduction and embryology at the Sackler Faculty of Medicine. She has been a member of the executive committee of the Israel Fertilization Association and the organizing committees of international meetings in Israel. Her main interests include male infertility and in-vitro maturation. Abstract There may be incompatibility between testicular histopathological evaluation and testicular sperm extraction (TESE) outcome. Assessment for sperm presence and different pathological disturbances of non-obstructive azoospermia (NOA) remains challenging. An assay for maximal sampling and accurate identification of testicular cells from NOA patients undergoing TESE and autopsied fertile controls was developed. Testicular cells stained and scanned automatically for morphology underwent fluorescence in-situ hybridization using centromeric probes for chromosomes X, Y and 18 after destaining. Cells were automatically classified according to ploidy, and ratios of haploid cells and autosomal (18) and sex-chromosome bivalent rates were calculated. Identification of testicular cells in suspension enabled prediction of spermatogenesis in seven of eight Sertoli-cell-only syndrome patients. Haploid/diploid cell ratios were 67.6:32.2 for controls and 9.6:90.4 for patients. Both autosomal (18) and sex-chromosome bivalents were present in patients (4.1 ± 5.82%) and controls (19.7 ± 8.95%). Few tetraploid pachytene spermatocytes were observed. More secondary spermatocytes with NOA showed two distinct signals for chromosome 18 (27.9 ± 32.69%) compared with controls (0.4 ± 0.35%). The computerized cell-scanning system enables simultaneous application of morphology and chromosome analysis of testicular cells, which enhance assessing different pathological disturbances and estimating the likelihood of a successful second TESE procedure. RBMOnline ª 2011, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. KEYWORDS: FISH, germ cells, pairing, SCOS, TESE, testis /$ - see front matter ª 2011, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. doi: /j.rbmo

2 102 D Strassburger et al. Introduction Assessment for the presence of spermatozoa and the different pathological disturbances of non-obstructive azoospermia (NOA) during testicular sperm extraction (TESE) remains a challenge. Classical histopathological evaluations (Tournaye et al., 1996) are limited to a minute amount of tissue specimen that may not be representative for the whole testes. Moreover, the slide sections which are used and which represent a restricted area in the biopsy may be incompatible with TESE outcome (Bremner et al., 1981; Ramasamy and Schlegel, 2007; Weller et al., 2005). Wet preparation of testicular tissue from the biopsies during TESE serves as a representative sample of the whole biopsy and gives information about the presence of spermatozoa and spermatogenesis in the extracted tissue used for treatment. However, identification by Nomarski or Hoffman modulation of individual immature germ cells in suspension after they have lost their typical position within the seminiferous tubules is imprecise. Stage-specific sorting of spermatogenic cells using comparative morphology both in live and fixed cells (Johnson et al., 2001) and velocity sedimentation under unit gravity or fluorescent-activated cell sorting have provided limited information on the different germ-cell types within the heterogeneous population of testicular cells in a wet cell suspension (Angelopoulos et al., 1997; Aslam et al., 1998; Holstein et al., 2003; Tesarik and Mendoza, 1996). The aim of this study was to develop a biotechnological assay that would enable a more accurate identification of the different cells in maximal sampling of the whole extracted tissue available for treatment and thus to offer supplementary information for the pathological diagnosis. Materials and methods Testicular-tissue specimens were obtained from 12 NOA patients undergoing TESE (mean age 30.4 years; range 26 39). Only one patient had an abnormal karyotype. The mean testicular volume was 14.8 ± 7.93 ml. Mean FSH, LH and testosterone concentrations were 20.3 ± 6.69 IU/l, 7.7 ± 2.47 IU/l and 18.3 ± 7.41 nmol/l, respectively. Testicular tissue from the autopsy of three fertile road-accident victims (mean age 23.7 years, range 18 28) served as normal controls. The study was approved by the Institutional Review Board (code 37/06) and signed consent from all patients and from families of the controls was obtained. Testicular biopsies were excised and collected in a Petri dish filled with 3-ml modified HEPES-buffered Earle s medium with heparin (Sigma-Aldrich, Israel) (Figure 1). One small tissue specimen per testis was fixed in Bouin s solution for histological examination and the rest was shredded to obtain a testicular cell suspension. Initial examination of the wet preparation was performed at 400 magnification with an inverted Hoffman modulation microscope. The procedure was terminated if mature spermatozoa were identified; otherwise, another biopsy was taken. The second testicle was opened only if no mature spermatozoa were observed in the specimens of the first one. The specimen was centrifuged at 300g for 7 min and the pellet was suspended with 2-ml erythrocyte-lysing buffer for 10 min. Intracytoplasmic sperm injection (ICSI) was performed with selected spermatozoa. The rest of the tissue was frozen. Samples with no mature spermatozoa were taken for computerized analysis. Enzymatic treatment of the small testicular-tissue specimen with 1000 IU/ml collagenase type IV (C5138; Sigma-Aldrich) was carried out (Crabbe et al., 1998; Verheyen et al., 1995). The obtained cells were affixed on glass slides (Super Frost Plus, mm; Menzel-Glaser, Germany) using a Cytospin centrifuge (Shandon Cytospin 4) at 600g for 5 min and stained with May Grünwald Giemsa. Each slide was analysed and scanned automatically by bright field microscopy using 20 or 40 objectives of the BioView DUET computerized cell-scanning system (CCSS; BioView, Nes Ziona, Israel; Strassburger et al., 2007). Co-ordinates and images of cells captured for morphology confined to an area of a circle with a 900-lm radius were digitally recorded for future analysis. Staining was removed after immersion in ice-cold methanol/acetic acid 3:1 solution for 1 h and washed with phosphate-buffered saline. Fluorescence in-situ hybridization (FISH) using chromosome X, Y and 18 centromeric sequence-probes spectra (green, orange and aqua, respectively; Vysis, Downer s Grove IL, USA) was performed on the same slide according to Harper et al. (1994). Slides were screened once again by CCSS and target cells with fluorescence signals were automatically searched. The system has been validated to identify the FISH signals automatically while capturing each testicular cell individually. Each cell was represented by a pair of images on a screen, one for morphology and one for FISH results. Cells were observed and identified based on a correlation between the result of FISH staining and the assessment of cell morphology. Identification of the euploid cells, which were analysed in the current study, was performed manually. Aneuploid cells will be the subject of a separate analysis. The corresponding morphological images were not calculated in cases of over- or under-detection of signals. Ploidy was based on chromosomal number (n), DNA content or chromatids (c) and centromere number (s). A diploid cell contained 2n2c2s, a haploid cell contained n, c, s or n, 2c, s, and a tetraploid cell contained 4n4c4s or 4n4c2s. Only euploid cells were counted in this study. Since the DNA probes are very close to the centromere, it was impossible to distinguish the number of chromatids that comprised the chromosome unless chromatid separation had occurred. Nomarski optics were utilized to sort the germ cells according to size, shape, chromatin pattern of nuclei and presence and shape of the acrosome and flagella (Aponte et al., 2005; Holstein et al., 2003). Diploid Leydig cells were large in diameter (15 20 lm) and had a highly refractile surface. The cytoplasm was strongly acidophilic and finely granular. The nucleus was large, round and often located in an eccentric pattern in the cell. Diploid Sertoli cells had an irregular cell boundary: the nucleus was ovoid or angular, large and lightly stained and often contained a large nucleolus. Diploid spermatogonia were identified by a rounded nucleus with chromatin granules and one or two nucleoli. Primary spermatocytes were identified as the largest cells (19 24 lm in diameter). They had a low cytoplasm/nuclear ratio and a granular surface with clumped chromatin masses. Haploid secondary spermatocytes

3 Morphology and FISH of spermatogenic cells 103 and/or early round spermatids were smaller ( lm in diameter). The nuclei showed a homogeneous, finely granular karyoplasm. Round spermatids were spherical cells and 6 8 lm in diameter. They had clear cytoplasm, a smaller but distinctive round nucleus that stained darker and a prominent central nucleolus. Elongating spermatids had Figure 1 Flow chart showing the different steps of the computerized cell-scanning system. CCS = computerized cell scanning; CCSS = computerized cell-scanning system; ELB = erythrocyte-lysing buffer; FISH = fluorescent in-situ hybridization; ICSI = intracytoplasmic sperm injection.

4 104 D Strassburger et al. diameters of 4 6 lm and were asymmetric in appearance. During enzymatic dissociation both the developing flagella and some of the cytoplasmic lobes were severed from the elongating spermatids. Spermatids that had just begun condensation were larger than the more fully condensed late-stage spermatids. Statistical analysis was performed using Stats Direct statistical software (version ; Cheshire, UK). The Mann Whitney U-test was applied to compare different parameters between and within the control and study groups. P-values <0.05 were considered to be significant. Results The routine histopathological findings divided the study group patients into those with hypospermatogenesis (HS; n = 3), maturation arrest (MA; n = 1) and Sertoli-cell-only syndrome (SCOS; n = 8). Two biopsies were taken from one testicle of the HS patients: mature spermatozoa were found in all three patients and the excess testicular suspension samples that were negative for the presence of spermatozoa after a meticulous search were used for the CCSS study. No spermatozoa were detected in any of the samples of the biopsies in the MA and SCOS patients and all these specimens were used for CCSS study purposes. After enzymatic digestion and meticulous searching, a total of 9422 out of 12,001 cells were analysed following bright-field and FISH cell screening. The rest of the cells (21.5%) had over-detection, under-detection of signals or low Giemsa staining quality. Altogether, 7711 euploid cells were assessed, with a mean of ± cells per NOA patient, and 1711 cells, with a mean of ± per control. Sertoli and Leydig cells were present in all subjects but their percentage varied among them, which precluded establishing statistical significance. The frequency of spermatogonia per subject was higher in the study group than in the controls (41.7 ± 21.90% versus 6.6 ± 5.27%; P < 0.02; Table 1), whilst the difference in frequency of primary spermatocytes did not reach statistical significance (33.0 ± 20.49% versus 14.7 ± 10.82% for controls). In the later stages of spermatogenesis, the controls exhibited a higher frequency of secondary spermatocytes and round spermatids >8 lm (19.1 ± 15.12% versus 3.6 ± 3.0% for patients; P < 0.003), round spermatids 8 lm (33.97 ± 3.97% versus 4.9 ± 8.6% for patients; P < ), elongated spermatids (7.6 ± 7.21% versus 1.1 ± 3.6% for patients; P < 0.04) and sperm cells (6.98 ± 5.25% versus 0.02 ± 0.06% for patients; P < ) (Table 1). Haploid secondary spermatocytes and early round spermatids were present in seven of the eight SCOS patients. No round spermatids <8 lm were found in three of the eight SCOS patients, while elongated spermatids were observed in two of the three HS patients and in one SCOS patient. Mature spermatozoa were detected by CCSS in the testicular suspension sample from one HS patient although no spermatozoa had been found in this specimen after an intensive conventional search in the suspension during the TESE procedure. The haploid/diploid cell ratio was 2:1 (67.6:32.2) in the samples from the controls, while the study group had an opposite ratio of 1:9 (9.6:90.4). There were 21.5%, 5.6% and 5.5% haploid cells and 78.5%, 94.5% and 95.5% diploid cells in the HS, MA and SCOS patients, respectively. Various signatures were found by FISH for primary spermatocytes (Table 2; Figure 2). Most of the euploid primary spermatocytes had paired chromatids (4n4c2s) in both the control and study groups (98.9% and 97.7%, respectively). The simultaneous presence of a sex-chromosome bivalent and an autosomal (18), as detected by one signal for X and Y in proximity (XY body) and both homologous chromosomes 18, occurred in 19.7 ± 8.95% of the controls and 4.1 ± 5.82% of the patients (P < 0.05). Paired FISH signals for two homologous chromosomes 18 only were higher among controls (45.9 ± 2.68%) than the patients (20.9 ± 12.88%; P < 0.02). Sex bivalency with autosomal univalency was rare in both control and study groups (1.5 ± 1.07% and 4.9 ± 2.36%, respectively), while univalency for chromosomes X, Y and 18 was higher for patients than for controls (70.1 ± 16.92% versus 32.8 ± 10.56%, Table 1 Distribution of testicular cells according to cell types. Group Cells (n) Somatic cells (Sertoli and Leydig) Spermatogonia Primary spermatocytes Secondary spermatocytes and round spermatids (>8 lm) Round spermatids (8 lm) Elongated spermatids Sperm cells Control ± ± ± 5.27 a 14.7 ± ± b ± 3.97 c 7.6 ± 7.21 d 6.98 ± 5.25 e Study ± ± ± 21.9 a 33.0 ± ± 3.0 b 4.9 ± 8.6 c 1.1 ± 3.6 d 0.02 ± 0.06 e HS ± ± ± ± ± ± ± ± 0.11 MA * SCOS ± ± ± ± ± ± ± Values are mean ± SD percentage (cell amount from total cell number), unless otherwise stated. HS = hypospermatogenesis; MA = maturation arrest; SCOS = Sertoli-cell-only syndrome. a P < b P < c P < d P < e P < * MA, a single case.

5 Morphology and FISH of spermatogenic cells 105 Table 2 Rate of bivalent formation of two chromosome pairs XY and 18. Group Primary spermatocytes 4n4c4s 4n4c2s 4n4c2s XYbody/1818 XY/1818 XYbody/18 XY/18 Control ± ± ± ± 1.07 a 32.8 ± b,c 19.7 ± 8.95 d 45.9 ± 2.68 e,f Study ± ± ± ± ± b 4.1 ± 5.82 d 20.9 ± e HS ± ± ± ± ± ± ± MA * SCOS ± ± ± ± 2.09 a 71.1 ± c 4.3 ± ± f Values are mean or ± SD percentage (cell amount from total cell number). c = DNA content or chromatids; HS = hypospermatogenesis; MA = maturation arrest; n = human chromosomal number; s = centromere number; SCOS = Sertoli-cell-only syndrome; XYbody = two proximal signals for paired XY chromosomes; 1818 = two separate signals for unpaired chromosomes 18; XY = two separate signals for unpaired XY chromosomes; 18 = one signal for paired chromosomes 18. a P < b P < c P < d P < e P < f P < * MA, a single case. Figure 2 Morphology of spermatocyte I (left panels) and chromosome locations (right panels) in the spermatocyte I nucleus. Green and red signals represent X and Y centromeres, respectively, and aqua signals represent chromosome 18 centromeres. Presentation of chromosome statuses as bivalents: XY (A), one signal for the two 18 homologous chromosomes (A, B), univalency for X and Y (B, C), the two 18 chromosomes (C) and four sister chromatids of each two homologous chromosomes (D). Bars = 13 lm (left) and 9.6 lm (right). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) respectively; P < 0.02). A few pachytene primary spermatocytes with tetraploid signals (4n4c4s) were identified in both groups (Figure 2). As mentioned earlier (Table 1), the rates of secondary spermatocytes and early round spermatids were higher for controls than for patients. The majority of the cells in both the study group and the controls had a single FISH signal for chromosome 18 and a single FISH signal for the sex chromosome 2n2c1s or 1n1c1s (Figure 3). The remaining cells had 2n2c2s (X/1818 or Y/1818) (0.4 ± 0.35% in the study group versus 27.9 ± 32.69% in the controls; P < 0.02). Discussion Pathological disturbances in NOA men are assessed by identifying and quantifying the germ cells in the testicular tissue. Histopathology of slide sections that represent a restricted area in a scant random testicular-tissue biopsy does not provide adequate information for those purposes (Johnson et al., 2001). While testicular cell suspensions derived from the testes appear to provide the appropriate source for evaluating an entire biopsy, the evaluation is

6 106 D Strassburger et al. Figure 3 Morphology of spermatocyte II (left panels) and chromosome (right panels) locations in the spermatocyte II nucleus. Green and red signals represent X and Y centromeres, respectively, and aqua signals represent chromosome 18 centromeres. Presentation of haploid spermatocytes II: one signal for chromosome 18 (A, B), Two attached cells: two signals for sister chromatids of chromosome 18 and one signal for chromosome X (left cell) or chromosome Y (right cell). Bars = 8 lm (A and B left) and 6 lm (A and B right). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) complicated and limited to cells with a distinct appearance, e.g. Sertoli cells, pachytene spermatocytes and testicular spermatozoa. The availability of a biotechnological assay enabling identification of the different germ cells throughout the spermatogenesis process and their ploidy would be highly valuable. The CCSS technique previously described for the combined analysis of morphology and FISH of ejaculated sperm cells (Strassburger et al., 2007) enabled simultaneous analysis of morphology and FISH of somatic and germ cells from testicular tissue of NOA patients. The system has been validated to identify the FISH signals automatically while capturing each testicular cell individually. Since it is a time-consuming procedure, it could be more suitable for application when the TESE procedure takes place 1 day before oocyte retrieval. Chiang, Yeh, Lin, Fang and Wei (2002) performed Giemsa and FISH staining of the same testicular sections of 26 azoospermic men and identified haploid cells in five cases (19%) after FISH, while only diploid cells were identified using Giemsa staining. The use of CCSS on sperm suspensions in these SCOS cases provided supplementary information on disturbed spermatogenesis. Germ cells from different stages were identified in seven out of eight patients (specificity 89%), while conventional histopathology in the same patients showed only Sertoli cells and partially atrophic tubules. Indeed, the incidence of 41.5% spermatogonia and 32.6% primary spermatocytes is very unlikely for patients with SCOS. Hence, these results support the possibility that the diagnosis of SCOS, which was based on pathological sections from these seven patients, was not correct. This study observed a large difference in the frequency of haploid cells between the azoospermic patients and the normospermic subjects. These differences were, however, not significant, probably due to the small number of participants and individual spermatogenesis expression. In the current study, the haploid/diploid cell ratio was 2:1 in the controls and 1:9 in the study group. Larger populations of diploid cells (pre-meiotic and meiotic) spermatogonia and primary spermatocytes, together with smaller populations of haploid post-meiotic cells (secondary spermatocytes and spermatids), are characteristic of abnormal spermatogenesis. Scarce and incomplete data on isolation and purification of spermatogenic cells from testicular biopsies of azoospermic men following FISH are available. Huang et al. (1999) reported 55.2% haploid cells in obstructive cases compared with % in five HS and nine early and late MA cases. Tanaka et al. (1999) stained over 1000 cells of 48 SCOS patients for X and Y probes and found 5.7% haploid cells and 94.2% diploid cells. No information on normal controls was available. The latter two studies could not differentiate between the two kinds of diploid (spermatogonia or primary spermatocyte) and haploid (spermatid or secondary spermatocyte) cells since dithiothreitol (DTT), which deforms the cells and prevents their proper identification, was used for DNA decondensation. The current study, therefore, did not use DTT and could compare FISH results with cytogenetic structure, enabling it to distinguish between the various spermatogenic cell populations. Diploid cells, visualized as XY/1818, may represent non-germinal cells. With this study s method, primary spermatocytes or spermatogonia can be identified according to the size and morphology of the cells with a high level of confidence. Likewise, one of the enigmatic aspects was the appearance of a few pachytene primary spermatocytes with tetraploid signals (4n4c4s) among the study and control samples. Although it is not unusual to find metaphase I nuclei that are apparently tetraploid in human meiotic chromosome preparation, it is impossible to differentiate them from the mix-up of bivalents from two normal metaphases, since it is well known that meiosis occurs in waves on a given Sertoli cell. The Giemsa staining of a specific spermatocyte showed a single cell with one pachytene nucleus. Thus far, only one tetraploid pachytene cell has been reported in

7 Morphology and FISH of spermatogenic cells 107 mice (Solari and Moses, 1977) and another in humans (Codina-Pascual et al., 2006). Endoreduplication before meiosis has been suggested as a possible origin for these tetraploid meiocytes, which, in turn, might enter meiosis and produce diploid spermatozoa. Most (97%) euploid primary spermatocytes of all men in the current study had paired chromatids (4n4c2s). Various signals, such as XY body/1818, XY/1818, XY body/18 and XY/18 which indicate the existence of different recombination configurations, were also detected. The question of whether defects in recombination were a common occurrence in NOA patients was addressed by some groups who immunolabelled synaptonemal complex spreads by monoclonal antibodies (scp3, mlh1 and crest). Ma, Arsovska, Moens, Nigro and Chow (2006) concluded that frequencies of XY bivalents with at least one recombination focus were similar in NOA patients and controls. Topping et al. (2006) reported that there was little evidence of meiotic impairment in 17 NOA cases. Others showed significant reductions of recombination in association with NOA (Gonsalves et al., 2004; Guichaoua et al., 2005; Martin et al., 2003; Sun et al., 2007; Tassistro, Ghalamoun-Slami et al., 2009). Yogev et al. (2000,2004,2006) used centromeric DNA probes for X, Y, 9, 15 and 18 and detected the paired XY chromosome (SV) in 84% of the cells in men with normal spermatogenesis compared with 23% in men with spermatocyte arrest. In their study, the percentage of autosomal bivalent chromosomes was higher than the percentage of sex bivalent chromosomes. Their explanation was that the sex chromosomes of the spermatocytes seem to be in a special structure that markedly differs from that of the paired recombined autosomal chromosomes during the meiotic prophase (Handel and Hunt, 1992). In the current study, the frequency of paired signals for two homologous chromosomes 1818 as well as for XY (XY body), which indicate normal pairing (bivalents), was higher for the controls than for the azoospermic group, particularly the SCOS and MA patients. Similarly to other studies, sex bivalency was regularly more extensively impaired than autosomal bivalency in both azoospermic patients and controls. By following genetic signatures during spermatogenesis, this study was able to identify secondary spermatocytes. In general, it is especially difficult to differentiate between secondary spermatocytes and very immature spermatic cells. Secondary spermatocytes usually possess centrally located nuclei >7.5 lm in diameter and stain in exactly the same way as immature spermatids. These cells are only seen occasionally because of their short lifespan (Barratt, 1995). The ability to follow different chromatin colours and structures and simultaneously identify a double-dotted FISH signal for one chromosome helped the current study to observe secondary spermatocytes. However, four different dotted FISH signals could also be displayed when the chromatids separated before completion of the second meiotic division, leading to the generation of round spermatids. In conclusion, using CCSS offers the advantage of the simultaneous application of morphology and chromosome analysis of testicular cells and the examination of meiotic pairing and post-meiotic chromosomal constitution. The results of this comprehensive investigation should enhance the assessment of different pathological disturbances and help in estimating the likelihood of a successful second TESE procedure when no spermatozoa have been detected in the current TESE. If CCSS shows immature sperm cells in the specimen, or if no spermatozoa are detected in the remaining analysed testicular tissue, there is no basis for recommending that the couple undergo another TESE procedure, in which case, alternative treatments should be offered. Acknowledgements The authors thank Mrs Tal Kaplan from BioView, Nes Ziona, Israel for her professional assistance, and Mrs Fredrica Gendler from the Assaf Harofeh Medical Centre, Zerifin, Israel, for her linguistic editing and commitment to the study. They also thank Esther Eshkol for editorial assistance. References Angelopoulos, T., Krey, L., McCullough, A., Adler, A., Grifo, J.A., A simple and objective approach to identifying human round spermatids. Hum. Reprod. 12, Aponte, P.M., van Bragt, M.P., de Rooij, D.G., van Pelt, M., Spermatogonial stem cells: characteristics and experimental possibilities. Apmis 113, Aslam, I., Robins, A., Dowell, K., Fishel, S., Isolation, purification and assessment of viability of spermatogenic cells from testicular biopsies of azoospermic men. Hum. Reprod. 13, Barratt, C.L.R., Spermatogenesis. In: Grudzinsky, J.G., Yovich, J.L. (Eds.), Gamete: the Spermatozoon. Cambridge University Press, Cambridge, pp Bremner, W.J., Matsumoto, A.M., Sussman, A.M., Paulsen, C.A., Follicle-stimulating hormone and human spermatogenesis. J. Clin. Invest. 68, Chiang, H.S., Yeh, S.D., Lin, W.M., Fang, C.L., Wei, H.J., Correlation between fluorescence in situ hybridization and testicular biopsy for the prediction of spermatogenesis in 37 patients with nonobstructive azoospermia. Urology 60, Codina-Pascual, M., Navarro, J., Egozcue, J., Benet, J., A human tetraploid pachytene spermatocyte as the possible origin of diploid sperm: a case report. Hum. Reprod. 21, Crabbe, E., Verheyen, G., Silber, S., Tournaye, H., Van de Velde, H., Goossens, A., Van Steirteghem, A.C., Enzymatic digestion of testicular tissue may rescue the intracytoplasmic sperm injection cycle in some patients with non-obstructive azoospermia. Hum. Reprod. 13, Gonsalves, J., Sun, F., Schlegel, P.N., Turek, P.J., Hopps, C.V., Greene, C., Martin, R.H., Reijo Pera, R.A., Defective recombination in infertile men. Hum. Mol. Genet. 13, Guichaoua, M.R., Perrin, J., Metzler-Guillemain, C., Saias-Magnan, J., Giorgi, R., Grillo, J.M., Meiotic anomalies in infertile men with severe spermatogenic defects. Hum. Reprod. 20, Handel, M.A., Hunt, P.A., Sex-chromosome pairing and activity during mammalian meiosis. BioEssays 14, Harper, J.C., Coonen, E., Ramaekers, F.C., Delhanty, J.D., Handyside, A.H., Winston, R.M., Hopman, A.H.N., Identification of the sex of human preimplantation embryos in two hours using an improved spreading method and fluorescent in-situ hybridization (FISH) using directly labelled probes. Hum. Reprod. 9, Holstein, A.F., Schulze, W., Davidoff, M., Understanding spermatogenesis is a prerequisite for treatment. Reprod. Biol. Endocrinol. 1, 107.

8 108 D Strassburger et al. Huang, W.J., Lamb, D.J., Kim, E.D., de Lara, J., Lin, W.W., Lipshultz, L.I., Bischoff, F.Z., Germ-cell nondisjunction in testes biopsies of men with idiopathic infertility. Am. J. Hum. Genet. 64, Johnson, L., Staub, C., Neaves, W.B., Yanagimachi, R., Live human germ cells in the context of their spermatogenic stages. Hum. Reprod. 16, Ma, S., Arsovska, S., Moens, P., Nigro, M., Chow, V., Analysis of early meiotic events and aneuploidy in nonobstructive azoospermic men: a preliminary report. Fertil. Steril. 85, Martin, R.H., Greene, C., Rademaker, A.W., Ko, E., Chernos, J., Analysis of aneuploidy in spermatozoa from testicular biopsies from men with nonobstructive azoospermia. J. Androl. 24, Ramasamy, R., Schlegel, P.N., Microdissection testicular sperm extraction: effect of prior biopsy on success of sperm retrieval. J. Urol. 177, Solari, A.J., Moses, M.J., Synaptonemal complexes in a tetraploid mouse spermatocyte. Exp. Cell Res. 108, Strassburger, D., Reichart, M., Kaufman, S., Kasterstein, E., Komarovsky, D., Bern, O., Friedler, S., Schachter, M., Ron-El, R., Raziel, A., Morphology assessment and fluorescence in situ hybridization of the same spermatozoon using a computerized cell-scanning system. Hum. Reprod. 22, Sun, F., Turek, P., Greene, C., Ko, E., Rademaker, A., Martin, R.H., Abnormal progression through meiosis in men with nonobstructive azoospermia. Fertil. Steril. 87, Tanaka, A., Tanaka, I., Nagayoshi, M., Awata, S., Mawatari, Y., Kusunoki, H., FISH analysis on human spermatogenic cells in patients diagnosed as Sertoli cell only histologically. In: In Vitro Fertilization and Human Reproductive Genetics. Monduzzi Editore, Sydney, Australia, pp Tassistro, V., Ghalamoun-Slami, R., Saias-Magnan, J., Guichaoua, M.R., Chronology of meiosis and synaptonemal complex abnormalities in normal and abnormal spermatogenesis. Indian J. Med. Res. 129, Tesarik, J., Mendoza, C., Spermatid injection into human oocytes. I. Laboratory techniques and special features of zygote development. Hum. Reprod. 11, Topping, D., Brown, P., Judis, L., Schwartz, S., Seftel, A., Thomas, A., Hassold, T., Synaptic defects at meiosis I and non-obstructive azoospermia. Hum. Reprod. 21, Tournaye, H., Liu, J., Nagy, P.Z., Camus, M., Goossens, A., Silber, S., Van Steirteghem, A.C., Devroey, P., Correlation between testicular histology and outcome after intracytoplasmic sperm injection using testicular spermatozoa. Hum. Reprod. 11, Verheyen, G., De Croo, I., Tournaye, H., Pletincx, I., Devroey, P., van Steirteghem, A.C., Comparison of four mechanical methods to retrieve spermatozoa from testicular tissue. Hum. Reprod. 10, Weller, O., Yogev, L., Yavetz, H., Paz, G., Kleiman, S., Hauser, R., Differentiating between primary and secondary Sertoli-cell-only syndrome by histologic and hormonal parameters. Fertil. Steril. 83, Yogev, L., Gamzu, R., Kleiman, S., Botchan, A., Hauser, R., Yavetz, H., Evaluation of meiotic impairment of azoospermic men by fluorescence in situ hybridization. Fertil. Steril. 74, Yogev, L., Zeharia, E., Kleiman, S.E., Maymon, B.B., Hauser, R., Botchan, A., Yavetz, H., Paz, G., Use of sex chromosome bivalent pairing in spermatocytes of nonobstructive azoospermic men for the prediction of successful sperm retrieval. Fertil. Steril. 86, Yogev, L., Segal, S., Zeharia, E., Gamzu, R., Maymon, B.B., Paz, G., Botchan, A., Hauser, R., Yavetz, H., Kleiman, S.E., Sex chromosome alignment at meiosis of azoospermic men with azoospermia factor microdeletion. J. Androl. 25, Declaration: The author reports no financial or commercial conflicts of interest. Received 9 January 2011; refereed 13 September 2011; accepted 15 September 2011.