Experimental validation with the reproductive toxicant vinclozolin
Jacques Auger , , a, Corinne Lesaffrea, Alexandre Bazireb, Damien Schoevaert-Brossaultc and Florence Eustached
a Service d’Histologie-Embryologie, Biologie de la Reproduction/CECOS, GREFH, EA1752, H?pital Cochin, Faculté de Médecine, Université René Descartes, Paris, France
b Centre Anti-Poison, H?pital F. Widal, Paris, France
c Laboratoire d’Analyse d’Images en Pathologie Cellulaire, IUH, H?pital Saint-Louis, Paris, France
d CECOS, H?pital Necker-Enfants Malades, Faculté de Médecine, Université René Descartes, Paris, France
Abstract
Recent studies have shown that the complex inter-related processes of sperm chromatin organization and nuclear morphogenesis, both of which are important fertility determinants, may be disrupted by chemicals. A high-resolution image cytometry method has been developed, using the fluorescent dye bisbenzimide, for the measurement of 20 features of the sperm nucleus related to size, form and chromatin status in the rat. For the complete set of features measured and from a total of 150 spermatozoa assessed per sample, the overall coefficient of reproducibility was 5%. Then, an experimental validation of the method was carried out in rats chronically exposed to the antiandrogenic reproductive toxicant vinclozolin and control animals. Univariate statistics revealed significant vinclozolin-induced changes for 19 out of 20 morphometric and chromatin features. Stepwise linear discriminant analysis classified correctly 84.3% of the sperm nuclei with only four features selected. The accuracy and reproducibility of the cytometry assessment of the sperm nuclei together with the results of the experimental validation suggest this method may be a new powerful tool for use in reproductive toxicology.
Author Keywords: Chromatin; Image analysis; Morphometry; Multivariate statistics; Reproductive toxicology; Sperm nucleus; Spermatogenesis; Texture; Vinclozolin
1. Introduction
Sperm morphology is a major determinant of human fertility [1] and an increased incidence of abnormal sperm after exposure to reproductive toxicants has been reported [2 and 3]. However, the conventional assessment of sperm morphology, which is based on microscopy, has several inherent difficulties. An important source of intra- and inter-individual variability in the traditional assessment of sperm morphology is the subjective nature of visual perception [4]. This recognized difficulty has provided many incentives for replacing visual assessment by objective approaches for the analysis of a number of sperm characteristics. For example, computer-aided sperm analysis (CASA) has been proposed and is routinely used in clinical practice and in reproductive toxicology, to evaluate sperm motion and sperm count [2 and 5]. By contrast, this technology is currently inadequate for the analysis of sperm morphology for routine clinical applications [6]. However, it is suitable for sperm morphometry [7, 8, 9, 10 and 11]. The relevance of this approach for sperm morphometry has been demonstrated in several human and animal fertility studies [7, 12, 13, 14, 15, 16 and 17] and in rare studies on chemical or physical factors that may affect spermatogenesis [18, 19, 20 and 21].
Computerized image cytometry of morphometric and textural nuclear features is an established, powerful tool in cyto- and histopathology for the assessment of normal and abnormal cells types [22 and 23] and for monitoring the cell cycle [24 and 25]. In contrast, this approach has been rarely used for studying sperm nuclei. To our knowledge, it was only used to follow sperm nuclear formation and maturation quantitatively in humans [26], to demonstrate nuclear differences according to fertility status [27, 28 and 29] and for revealing radiation-induced defects [21].
Several studies have reported that the intensity of staining with various DNA-specific fluorescent dyes in mammalian sperm nuclei depends strongly on chromatin condensation and/or DNA damage [30, 31 and 32]. Therefore, it can be postulated that the sperm chromatin appearance related to condensation, distribution and organization of chromatin within the nucleus [25] can be indirectly and quantitatively assessed by analyzing the intensity of staining and distribution of a DNA-specific fluorescent dye by high-resolution image cytometry.
The present study reports a new image cytometry method for assessing jointly sperm nuclear shape, size and chromatin appearance in the rat followed by an experimental validation of the method in control unexposed and rats chronically exposed to the reproductive toxicant vinclozolin. We chose to study this fungicide because it has been reported to disrupt formation of the male reproductive tract, with persistent consequences for spermatogenesis in the rat [33] when administered in utero, its main mechanism of action being mediated by inhibition of androgenic receptors [34].
2. Materials and methods
2.1. Animals
The image cytometry method was developed in 90-day-old male Wistar rats purchased from Harlan France Sarl (Gannat, France).
2.2. Sperm collection and preparation for cytometry
Each rat was killed by inhalation of an overdose of anesthetic and immediately excised to collect the spermatozoa stored in the vas deferens. A ~0.8 cm portion of the proximal part of the vas deferens was excised and placed in a Petri dish containing 1 ml of prewarmed Hank’s medium (Sigma-Aldrich Chimie Sarl, Lyon, France) and incubated for 5 min at 37 °C. The Petri dish was swirled to facilitate the spontaneous release of sperm from the vas deferens, which resembled an expanding filament at the two ends of the fragment, rendering the medium cloudy. The liquid phase was removed by aspiration and vortexed. The DNA-specific dye was then added immediately (see below). The protocol for sperm staining was adapted from Strader et al. [35]. Briefly, 0.2 ml of the sperm suspension was taken up by aspirated and immediately deposited in a tube containing the fluorescent DNA-specific dye bisbenzimide (Ident stain, Hamilton-Thorn Research; Danvers, MA). Bisbenzimide permeates into cells and becomes brightly fluorescent upon binding to DNA, binding preferentially to contiguous AT base pairs. The preparation was vortexed for 30 s and kept in the dark for equal amounts of time for each sample tested (15 min). We took up 20 l of the well-mixed preparation by aspiration, deposited it on a clean glass slide, added two drops of Vectashield mounting medium for fluorescence H-1000 (Vector Laboratories Inc, Burlingame, CA) to prevent rapid photobleaching, covered the slide with a 24 mm × 60 mm coverslip and sealed the perimeter with nail polish. Slides were observed immediately.
2.3. Sperm nuclear morphometry and chromatin texture analysis
Sperm nuclei were analyzed using a Samba IPS2 image analysis system, version 4.07 (Samba Technologies, Meylan, France) connected to an Axioskop Zeiss fluorescence microscope (Carl Zeiss S.A.S., Le Pecq, France). The microscope is equipped with a cooled black and white CCD camera connected to an image intensifier unit (Hamamatsu Photonics, Massy, France). Briefly, the Samba instrument incorporates a PC computer with a Meteor II Matrox digitizer board, integrates subroutines for binary, gray-scale level or color image processing and allows the user to link these subroutines for specific applications.
The rat sperm karyometry (RSK) macro written for this study combined manual and automatic steps and linked appropriate subroutines with the aim to measure relevant morphometric features including the nuclear curvature and chromatin features related to the condensation, distribution and organization of the chromatin within the nucleus.
Briefly, the operator loaded a stained slide onto the stage, selected the objective 100× and the fluorescent light source. The use of the DNA fluorescent dye offered the possibility to dissect the nucleus with a sharp outline and the 100× objective was found to give the best dynamics of the gray-scale levels of the nuclear pixels after optimizing the set-up of the light intensification device (Fig. 1A and B). Then, the operator provided information about the specimen to the program, positioned the sperm nucleus in the centre of the microscopic field, adjusted the set-up for light level and the grey-level threshold in order to detect the nucleus and to define its outline accurately (checked by alternate superimposition of the native image and the binary image of the nucleus after thresholding). Afterwards, the program passed into automatic mode with the following steps: (i) measures of morphometric features from the nuclear outline; (ii) measures of gray-scale level values of all the pixels included in the area delimited by the nuclear outline and computation of derived chromatin features; (iii) measures of gray-scale level values of all the pixels included in three nuclear subregions after successive openings (erosion-dilations) allowing the computation of an index of chromatin organization, and (iv) measure of the nuclear skeleton (nuclear curvature) from successive thinning steps. The abbreviations, names and descriptions of the 20 reported features are presented in Fig. 1C and D and corresponding legend. Finally, the alternate manual and automatic steps are repeated for each new nucleus until the appropriate number of nuclei per sample has been examined (see Section 2.5) using a standardized method of slide screening to prevent repeated analysis of the same nucleus. For each sperm nucleus, all measurements are stored in the specimen data file.
Fig. 1. (A) Light microscopy of a typical rat spermatozoon (100× magnifications). (B) The same spermatozoon under epifluorescence after bisbenzimide staining and an optimal adjustment of the light intensification level; the combined use of a DNA-specific fluorescent dye, a 100× objective and an image intensifier results in a dissection of the nucleus with an outline standing out clearly, a major advantage for image segmentation and morphometry, and a chromatin appearance showing regional variations in the gray-scale levels suitable for texture analysis. (C) Morphometric features measured; A: area in m2, P: perimeter in m, FF: form factor defined as P2/4 A, equal to 1 for a circle, it increases with the complexity of the outline, BE: bending energy, a feature summarizing the curvature changes of the form; it decreases with the curvature radius along the outline, W: width of the nucleus, the diameter of the highest inscribed circle in m, L: length of the nucleus, the diameter of the lowest circumscribed circle in m, S: skeleton, length of the medial axis of the nucleus in m; C: chord, length of the segment joining the ends of the skeleton, in m; two other features, the ratios W/L and C/S are not shown on the figure. (D) Several features pertaining to chromatin condensation, distribution and organization are measured from the gray-scale levels of the pixels within the area delimited by the nuclear outline and within three restricted regions of the nucleus, R1, R2 and R3 obtained by two successive openings (using 13 × 13 and 17 × 17 structuring elements) combined with XOR operations; SGL: sum of the gray-scale level values of all the pixels within the outline of the nucleus, arbitrary units (256 gray-scale levels, from 0: black to 255: white), MGL: mean gray-scale level for all the pixels of the nucleus, reflecting the overall level of chromatin condensation, arbitrary units, CVGL: coefficient of variation of the gray-scale level values of all the pixels of the nucleus, reflecting chromatin distribution or the level of homogeneity of sperm condensation, %, MGL1 and CVGL1, MGL2 and CVGL2 and MGL3 and CVGL3: mean gray-scale levels and corresponding coefficient of variations for the regions R1, R2 and R3, IGL1/3: chromatin condensation differences in the anterior, R1, and posterior, R3, regions of the nucleus, an index of chromatin organization along the major direction of the nucleus equal to MGL1/MGL3, %.
2.4. Experimental validation: effects of vinclozolin on sperm nuclear features
An experimental validation of the method was undertaken on subsamples of Wistar rats from a study aiming to investigate the long term effects of low and high doses endocrine disrupters alone or in association on the reproductive function (manuscript in preparation). The subsamples of rats used for the experimental validation of the study were an unexposed control group and a group exposed orally to 30 mg/kg per day of the reproductive toxicant vinclozolin from conception to PND90 (n = 10 for both). Vinclozolin was extracted from the commercial formulation Ronilan (BASF, Levallois-Perret, France), and the final preparation was >95% pure. Due to the protocol of the study, the spermatozoa were retrieved from cauda epididymis preparations and not from vas deferens fragments. Briefly, one epididymis was excised from each rat immediately after that rat was killed by inhalation of an overdose of anesthetic. The epididymides were then stored at ?80 °C. The cauda epididymis was dissected out just below the point at which the vas deferens joins the epididymis at the distal corpus and at the boundary between the corpus and the proximal end of the cauda epididymis. It was placed in a 50 ml plastic conical tube, chopped up finely with scissors and mixed with 25 ml phosphate-buffered saline + 0.05 g/100 ml Triton X-100 (Sigma-Aldrich Chimie Sarl, Lyon, France). The cauda epididymis preparation was kept frozen at ?20 °C, thawed the day of the nucleus examination and vortexed immediately before adding the DNA-specific dye bisbenzimide as described above.
2.5. Statistical analysis
Statistical analysis was carried out with BMDP statistical software [36]. In the development of the method, the minimum number of nuclei to be assessed for obtaining stable results was evaluated by analyzing 50, 75, 100, 125, 150, 200 and 250 nuclei from the same sample. Then, the coefficient of reproducibility (%) of the measures for two samples, each analyzed in triplicate was calculated.
In the experimental validation study, the samples were analyzed blindly. Differences between exposed and unexposed rats were assessed by non parametric Mann–Whitney rank-sum tests for each nuclear feature measured and the distributions are presented using box-plots. We investigated which nuclear features discriminated most clearly between exposed and unexposed rats by carrying out stepwise linear-discriminant analysis with the 7M-BMDP program. This program ranked the nuclear features according to their discriminatory power and calculated the percent correct classification at each step.
3. Results
3.1. Technical evaluation
The minimum number of nuclei to be assessed was estimated to be ~150 nuclei per sample (Fig. 2). We therefore assessed 150 nuclei per sample for each rat of the experimental validation study. For the complete set of features measured on the 150 nuclei assessed per sample, the mean coefficient of reproducibility of the measurements was 5% (minimum: 3% maximum: 8%). Typically, the time required for the complete analysis of 150 sperm was 60–90 min.
Fig. 2. Stability of the features measured according to the number of sperm nuclei analyzed. Relative levels of the median value of four features, area (A), form factor (FF), mean gray-scale level (MGL), coefficient of variation of the mean gray-scale level (CVGL), 50–250 sperm nuclei analyzed; on average, the median values of the features increased moderately between 50 and 150 sperm nuclei analyzed while it tended to stabilize when comparing 150, 200 and 250 sperm nuclei analyzed. In the subsequent experiments, 150 sperm nuclei per specimen have been analyzed.
3.2. Experimental validation
The distributions of the 10 morphometric features differed significantly between the exposed and unexposed rats as did chromatin features (Fig. 3A and B). Polar plots illustrate the overall difference in size, shape and chromatin status between unexposed rats and rats that were chronically exposed to vinclozolin (Fig. 4). The values obtained for morphometry and chromatin features indicated that the nuclei of the exposed rats were less mature than those of the control rats: they covered a larger area, had a lower form factor, a higher width to length ratio, a higher mean gray-scale level value or a smaller difference in chromatin condensation between the anterior and posterior part of the nucleus.
Fig. 3. Distributions (box-plots) and statistical differences in the morphometric features (A) and the chromatin features (B) of the sperm nuclei from control animals (white box; n = 1529 nuclei for 10 rats) and animals chronically exposed to 30 mg/kg per day vinclozolin (gray box; n = 1588 nuclei for 10 rats); lower and upper margins of the boxes indicate 25th and 75th percentiles, and horizontal bars on the whiskers mark 10th and 90th percentiles. The lines within the box indicate the mean value (bold line) and the median; the differences were analyzed by the non parametric Mann–Whitney rank-sum test.
Fig. 4. Polar plots illustrating the relative differences in the main morphometric and chromatin features of the sperm nuclei in control animals (A) and animals chronically exposed to 30 mg/kg per day vinclozolin (B). The values obtained for morphometric and chromatin features indicated that the nuclei of the exposed rats were less mature than those of the control rats: they covered a larger area, had a lower form factor, a higher width to length ratio, a higher mean gray-scale level value or a smaller difference in chromatin condensation between the anterior and posterior part of the nucleus.
The homogeneity of chromatin condensation (CVGL, the coefficient of variation of the gray-scale levels within the nucleus) was the nuclear feature that best discriminated between the nuclei of spermatozoa from exposed and control populations. Stepwise discriminant analysis gave a correct classification rate of 84.3% with a single canonical variable combining the four most discriminating features (Table 1 and Fig. 5).
Table 1. Ranking of nuclear features and correct classification of nuclei by stepwise discriminant analysis
Fig. 5. Automated classification of the sperm nuclei of control and animals chronically exposed to 30 mg/kg per day vinclozolin by discriminant analysis. The method automatically categorized the 3117 sperm nuclei analyzed for the 20 exposed and unexposed rats studied. A single variable combining the four most discriminating features (canonical variable) was found to express at best the cytometric differences between the sperm nuclei of the exposed and unexposed rats with an overall correct classification rate of more than 80%. Horizontal lines represent the canonical variable values for group means.
4. Discussion
The results reported herein provide evidence of the accuracy, reproducibility and discriminating power of the image cytometry method jointly assessing sperm nuclear morphometry and chromatin texture. The method offers noticeable advantages over the cytometric approaches previously reported for the assessment of the sperm head dimensions or DNA condensation thanks to the use of a DNA-specific fluorescent dye combined with high-resolution image cytometry.
4.1. Sperm nuclear morphometry
Unlike other methods for sperm head morphometry, the present approach does not require the complex pre-analytical steps essential for accuracy when using standard sperm preparations and stains [11 and 37], or processing of the sperm suspension by washing, fixing or sonication to separate head and tail [19]. The use of a fluorescent DNA-specific dye that makes the outline of the nucleus stand out clearly is a major advantage for image segmentation and, consequently, accurate morphometric measurements. To our knowledge, only one study has reported the rat sperm head dimensions using an automated morphometric method [19]. The method was based on a separation of the sperm heads by sonication followed by density gradient, a conventional staining of the smear and the use of automated sperm morphology analysis. Taking in account the differences in sperm preparation and staining, the morphometric values found in the control adult rats were remarkably closed in our validation study and in this study.
4.2. Sperm chromatin texture
The sperm DNA content, damage or chromatin condensation can be assessed at the population level by means of various specific DNA stains and flow cytometry [38, 39, 40 and 41]. This technique which makes it possible to assess several aspects of sperm structure and function on thousands of spermatozoa reliably, reproducibly and rapidly [38] has demonstrated the detrimental impact of reproductive toxicants on rodent sperm nuclei [40 and 41]. However, because of its principle, flow cytometry does not offer the possibility to assess the distribution and organization of the chromatin within the nucleus. In contrast, image cytometry is an appropriate technique for jointly assessing the chromatin/DNA condensation, distribution and organization within the nucleus both at the cellular and population levels [21, 25, 26, 27 and 28] as in the method reported herein. However, standard staining methods, low magnifications and conventional microscopy are inadequate for specifically assessing the sperm nuclear texture. The present study indicated that sperm DNA-specific imaging combined with high-resolution cytometry was especially relevant for accurately depicting the highly condensed status of the sperm nucleus. The fluorescent DNA dye giving a sharp transition between the nucleus and the background ensured that the detected nuclear outline represented the relevant window for an accurate measurement of the chromatin features derived from a grey-scale analysis of all the pixels within the window. Also, an accurate texture (and morphometric) analysis required a careful adjustment of the light intensification device from control sperm nuclei, both for the subsequent correct detection of the nuclear outline and for the optimization of the dynamic of the grey levels of the pixels composing the nucleus. Thus, it is imperative that this optimal light threshold should be incorporate in the set-up to avoid bias in comparative/exposed–unexposed studies.
The technical and experimental validation of the method indicated that it was not necessary to analyze thousands of nuclei but only 150 nuclei per specimen for providing quite stable and reproducible results or for showing significant differences when comparing exposed and unexposed animals. On the other hand, the time required for the complete analysis of 150 sperm—typically, 60–90 min—most of which relates to the manual positioning and focusing of the nuclei in the microscopic field represented the main disadvantage of the method for routine applications. However, this problem will be solved in the near future by upgrading the system by addition of a motorized XY-stage and a Z-autofocus for the automated slide screening and recording of nuclear images to check the correct orientation of the nucleus before fully automated image processing.
4.3. Experimental validation and usefulness of multivariate statistics
The experimental validation of the method showed for the first time that a chronic exposure, by the oral route, to 30 mg/kg per day of the reproductive toxicant vinclozolin, from conception to adulthood, disrupted both the sperm nuclear morphology (morphometry features) and the chromatin texture. Such deleterious effects could result in an impairment of the fertilizing ability and/or the embryo development as suggested by a number of studies [42, 43, 44 and 45]. The molecular and cellular mechanisms underlying these effects are probably numerous, complex and interrelated and warrant further studies. It should be pointed out that these significant effects were found despite a wide interindividual variation both in exposed and unexposed animals as indicated by the distribution ranges (see Fig. 3A and B). The experimental validation also gave an illustration of the usefulness of multivariate statistics for determining which feature(s) among the complete set of features was primarily impaired. It was found that, above all, a chronic exposure to vinclozolin had a deleterious impact on the homogeneity of chromatin condensation (76.2% of correct classification of the nuclei with this feature solely, see Table 1). Certainly, the multivariate cytometric approach applied to other reproductive toxicants (or different hazardous or pathological conditions) could result in different conclusions and a different ranking of the nuclear features. Therefore, the joint assessment of the nuclear morphology and texture followed by multivariate statistics could represent one of the most attractive aspects of the method, this sensitive approach being also useful to generate hypotheses on the mechanisms involved in physiopathology, environmental or toxicology studies.
In conclusion, this high-resolution image cytometry method makes it possible to measure jointly, in a reproducible, precise manner, nuclear morphology and chromatin status, as demonstrated here with original data obtained from rats chronically exposed to the reproductive toxicant vinclozolin. We believe that this new approach is a potentially powerful tool for use in the field of reproductive toxicology.
References
1. J.P. Bonde, E. Ernst, T.K. Jensen et al., Relation between semen quality and fertility: a population-based study of 430 first-pregnancy planners. Lancet 352 (1998), pp. 1172–1177.
2. S.D. Perreault and A.M. Cancel, Significance of incorporating measures of sperm production and function into rat toxicology studies. Reproduction 121 (2001), pp. 207–216.
3. A.J. Wyrobek and W.R. Bruce, Chemical induction of sperm abnormalities in mice. Proc. Natl. Acad. Sci. USA 72 (1975), pp. 4425–4429.
4. F. Eustache and J. Auger, Inter-individual variability in the morphological assessment of human sperm: effect of the level of experience and the use of standard methods. Hum. Reprod. 18 (2003), pp. 1018–1022.
5. J. Verstegen, M. Iguer-Ouada and K. Onclin, Computer-assisted semen analyzers in andrology and veterinary practice. Theriogenology 57 (2002), pp. 149–179.
6. ESHRE Andrology Special Interest Group. Guidelines on the application of CASA technology in the analysis of spermatozoa. Hum Reprod 1998;13:142–5.
7. D.F. Katz, J.W. Overstreet, S.J. Samuels, P.W. Niswander, T.D. Bloom and E.L. Lewis, Morphometric analysis of spermatozoa in the assessment of human male fertility. J. Androl. 7 (1986), pp. 203–210.
8. S.M. Schrader, T.W. Turner and S.D. Simon, Longitudinal study of semen quality of unexposed workers: sperm head morphometry. J. Androl. 11 (1990), pp. 32–39.
9. R.O. Davis, D.E. Bain, R.J. Siemers, D.M. Thal, J.B. Andrew and C.G. Gravance, Accuracy and precision of the CellForm-human automated sperm morphometry instrument. Fertil. Steril. 58 (1992), pp. 763–769.
10. F. Perez-Sanchez, J.J. de Monserrat and C. Soler, Morphometric analysis of human sperm morphology. Int. J. Androl. 17 (1994), pp. 248–255.
11. C. Garrett and H.W. Baker, A new fully automated system for the morphometric analysis of human sperm heads. Fertil. Steril. 63 (1995), pp. 1306–1317.
12. I.C. Macleod and D.S. Irvine, The predictive value of computer-assisted semen analysis in the context of a donor insemination programme. Hum. Reprod. 10 (1995), pp. 580–586.
13. M.F. Vine, R. Woodrow Setzer, Jr., R.B. Everson and A.J. Wyrobek, Human sperm morphometry and smoking, caffeine, and alcohol consumption. Reprod. Toxicol. 11 (1997), pp. 179–184.
14. C. Garrett, D.Y. Liu, G.N. Clarke, D.D. Rushford and H.W. Baker, Automated semen analysis: ‘zona pellucida preferred sperm morphometry and straight-line velocity are related to pregnancy rate in subfertile couples. Hum. Reprod. 18 (2003), pp. 1643–1649.
15. M. Hirai, A. Boersma, A. Hoeflich, E. Wolf, J. Foll, T.R. Aumuller and J. Braun, Objectively measured sperm motility and sperm head morphometry in boars (Sus scrofa): relation to fertility and seminal plasma growth factors. J. Androl. 22 (2001), pp. 104–110.
16. C.G. Gravance, Z.J. Champion and P.J. Casey, Computer-assisted sperm head morphometry analysis (ASMA) of cryopreserved ram spermatozoa. Theriogenology 49 (1998), pp. 1219–1230.
17. B.L. Sailer, L.K. Jost and D.P. Evenson, Bull sperm head morphometry related to abnormal chromatin structure and fertility. Cytometry 24 (1996), pp. 167–173.
18. L. Tablado, F. Perez-Sanchez, J. Nunez, M. Nunez and C. Soler, Effects of exposure to static magnetic fields on the morphology and morphometry of mouse epididymal sperm. Bioelectromagnetics 19 (1998), pp. 377–383.
19. R.O. Davis, C.G. Gravance, D.M. Thal and M.G. Miller, Automated analysis of toxicant-induced changes in rat sperm head morphometry. Reprod. Toxicol. 8 (1994), pp. 521–529.
20. Y. Ide, M. Kaido and O. Koide, Changes in spermatozoa due to large doses of pyridoxine (vitamin B6). Acta Pathol. Jpn. 42 (1992), pp. 861–869.
21. M. Aubele, U. Jütting, K. Rodenacker, P. Gais, G. Burger and U. Hacker-Klom, Quantitative evaluation of radiation-induced changes in sperm morphology and chromatin distribution. Cytometry 11 (1990), pp. 586–594.
22. T. Cremer, G. Kreth, H. Koester et al., Chromosome territories, interchromatin domain compartment, and nuclear matrix: an integrated view of the functional nuclear architecture. Crit. Rev. Eukaryot. Gene Exp. 10 (2000), pp. 179–212.
23. A. Sarkar, S. Eroglu, M.G. Poirier, P. Gupta, A. Nemani and J.F. Marko, Dynamics of chromosome compaction during mitosis. Exp. Cell Res. 277 (2002), pp. 48–56.
24. W. Sawicki, J. Rowinski and R. Swenson, Change of chromatin morphology during the cell cycle detected by means of automated image analysis. J. Cell Physiol. 84 (1974), pp. 423–428.
25. F. Giroud, Cell nucleus pattern analysis: geometric and densitometric featuring, automatic cell phase identification. Biol. Cell 44 (1982), pp. 177–188.
26. J. Auger and J.P. Dadoune, The nuclear status of human sperm cells by TEM image cytometry: analysis of nuclear shaping and chromatin texture changes during spermiogenesis and epididymal transit. Biol. Reprod. 49 (1993), pp. 166–175.
27. J.F. Moruzzi, A.J. Wyrobek, B.H. Mayall and B.L. Gledhill, Quantification and classification of human sperm morphology by computer-assisted image analysis. Fertil. Steril. 50 (1988), pp. 142–152.
28. J. Auger, D. Schoevaert, I. Negulesco and J.P. Dadoune, The nuclear status of human sperm cells by TEM image cytometry: relation of nuclear shape and chromatin texture to fertility. J. Androl. 14 (1993), pp. 456–463.
29. L. Ramos, J.C. Hendriks, P. Peelen, D.D. Braat and A.M. Wetzels, Use of computerized karyometric image analysis for evaluation of human spermatozoa. J. Androl. 23 (2002), pp. 882–888.
30. D. Evenson, Z. Darzynkiewicz, L. Jost, F. Janca and B. Ballachey, Changes in accessibility of DNA to various fluorochromes during spermatogenesis. Cytometry 7 (1986), pp. 45–53.
31. P.G. Bianchi, G.C. Manicardi, D. Bizzaro, U. Bianchi and D. Sakkas, Effect of deoxyribonucleic acid protamination on fluorochrome staining and in situ nick-translation of murine and human mature spermatozoa. Biol. Reprod. 49 (1993), pp. 1083–1088.
32. G.R. Aravindan, J. Bjordahl, L.K. Jost and D.P. Evenson, Susceptibility of human sperm to in situ DNA denaturation is strongly correlated with DNA strand breaks identified by single-cell electrophoresis. Exp. Cell Res. 236 (1997), pp. 231–237.
33. L.E. Gray, Jr., J. Ostby, E. Monosson and W.R. Kelce, Environmental antiandrogens: low doses of the fungicide vinclozolin alter sexual differentiation of the male rat. Toxicol. Ind. Health 15 (1999), pp. 48–64.
34. W.R. Kelce, E. Monosson, M.P. Gamcsik, S.C. Laws and L.E. Gray, Jr., Environmental hormone disruptors: evidence that vinclozolin developmental toxicity is mediated by antiandrogenic metabolites. Toxicol. Appl. Pharmacol. 126 (1994), pp. 276–285.
35. L.F. Strader, R.E. Linder and S.D. Perreault, Comparison of rat epididymal sperm counts by IVOS HTM-IDENT and hemacytometer. Reprod. Toxicol. 10 (1996), pp. 529–533.
36. Dixon WJ. BMDP statistical software manual. Berkeley: University of California Press; 1988.
37. R.O. Davis and C.G. Gravance, Standardization of specimen preparation, staining, and sampling methods improves automated sperm-head morphometry analysis. Fertil. Steril. 59 (1993), pp. 412–417.
38. M. Spano and D.P. Evenson, Flow cytometric analysis for reproductive biology. Biol. Cell 78 (1993), pp. 53–62.
39. I. Dobrinski, H.P. Hughes and A.D. Barth, Flow cytometric and microscopic evaluation and effect on fertility of abnormal chromatin condensation in bovine sperm nuclei. J. Reprod. Fertil. 101 (1994), pp. 531–538.
40. M. Spano, C. Bartoleschi, E. Cordelli et al., Flow cytometric and histological assessment of 1,2:3,4-diepoxybutane toxicity on mouse spermatogenesis. J. Toxicol. Environ. Health 47 (1996), pp. 423–441.
41. D.P. Evenson, F.C. Janca, L.K. Jost, R.K. Baer and D.S. Karabinus, Flow cytometric analysis of effects of 1,3-dinitrobenzene on rat spermatogenesis. J. Toxicol. Environ. Health 28 (1989), pp. 81–98.
42. A. Agarwal and T.M. Said, Role of sperm chromatin abnormalities and DNA damage in male infertility. Hum. Reprod. Update 9 (2003), pp. 331–345.
43. D.P. Evenson, L.K. Jost, D. Marshall et al., Utility of the sperm chromatin structure assay as a diagnostic and prognostic tool in the human fertility clinic. Hum. Reprod. 14 (1999), pp. 1039–1049.
44. M.V. Filatov, E.V. Semenova, O.A. Vorob’eva, O.A. Leont’eva and E.A. Drobchenko, Relationship between abnormal sperm chromatin packing and IVF results. Mol. Hum. Reprod. 5 (1999), pp. 825–830.
45. K.L. Larson, C.J. DeJonge, A.M. Barnes, L.K. Jost and D.P. Evenson, Sperm chromatin structure assay parameters as predictors of failed pregnancy following assisted reproductive techniques. Hum. Reprod. 15 (2000), pp. 1717–1722.