ABSTRACT
Proacrosin biosynthesis timing during human spermatogenesis has been studied using the monoclonal antibody 41)4 (mAb 4D4). Frozen and paraffin-embedded sections of testicular biopsies were labelled by standard indirect immunofluorescence and avidin-biotin immunoperoxidase procedures. The labelling specificity was checked by immunochemistry assays on unrelated tissues and by western blotting of testis extracts showing that only the 50–55×103 Mr proacrosin was recognized by mAb 4D4. Proacrosin was first observed in the Golgi region of midpachytene primary spermatocytes. In late pachytene primary spermatocytes, proacrosin was observed in two regions located at opposite nuclear poles. During the subsequent steps of the first meiotic division, the two bodies containing proacrosin were located: (i) on opposite sides of the equatorial plate during metaphase; along the microtubular spindle during anaphase; and close to each chromosomal aggregate during telophase. Two bodies containing proacrosin were still observed in interphasic secondary spermatocytes. The single labelled area observed in early spermatids was found to increase considerably in size during spermiogenesis. Anomalies of proacrosin scattering were observed in patients with Golgi complex partitioning failure. These data’ reveal proacrosin biosynthesis during diploid and haploid phases of human spermatogenesis and the proacrosin partitioning pattern during meiosis.
Introduction
The acrosome is elaborated from vesicles stored in the Golgi apparatus center immediately after the completion of meiosis. These vesicles fuse to give rise to a large proacrosomal vesicle (Sandoz, 1970, 1972) which forms the acrosomal vesicle over the nucleus (Dooher and Bennett, 1973; Burgos and Fawcett, 1955). Additional vesicles and granules originating from the Golgi apparatus contribute to the formation of the acrosome (Susi et al. 1971; Hermo et al. 1979, 1980; Burgos and Gutierrez, 1986). Several hydrolases are sequestered within the acrosomal matrix (Bellvé and O’Brien, 1983), the major one had been identified as proacrosin (Meizel, 1972; Parrish and Polakoski, 1979).
Radioautographic methods have been used in studies on the synthesis of acrosomal glycoproteins during spermatogenesis (Sandoz, 1972; Sandoz and Roland, 1976; Tang et al. 1982; Clermont and Tang, 1985); while recently, germ cells have been investigated by means of antibodies specific for acrosomal hydrolases. It has been suggested that the expression of proacrosin during mammahan spermatogenesis is initiated in spermatids. However, it is not clear whether proacrosin is present from the beginning of spermiogenesis (Flörke et al. 1983a,b; Phi-Van et al. 1983; Mansouri et al. 1983; Kallajoki and Siiominen, 1984; Arboleda and Gerton, 1988), or appears during the course of spermatid differentiation (Kallajoki et al. 1986).
We have previously demonstrated that the monoclonal antibody 4D4 recognizes human proacrosin in mature spermatozoa and in ejaculated spermatids (Gallo et al. 1991). In the present study, labelling of human testis sections with mAb 4D4 using several immunohistochemical procedures, revealed that proacrosin was present as early as the prophase of the first meiotic division. It was detected during all the subsequent steps of spermatogenesis and increased in quantity during acrosomogenesis. These data suggest meiotic as well as postmeiotic synthesis of proacrosin during the human spermatogenesis.
Materials and methods
Tissues
Human testicular biopsies (free from the tunica albuginea) were obtained from: (1) three autopsies and nine patients with an obstructive azoospermia, all presenting a normal germinal epithelium as revealed by light microscopy, (2) one patient with macrocephalic spermatozoa known to originate from meiotic division failure (Escalier, 1983, 1985) and (3) one patient with oligozoospermia and multinucleate sperm cells due to cytokinesis impairment (Holstein et al. 1988). Biopsies from human spleen, liver, kidney, ovary, breast, lung, pancreas, colon and dermoid kyst were used to check the specificity of germ cell immunolabelling.
Antibodies
The monoclonal antibody 4D4 was obtained after immunization of a mouse with a detergent-insoluble fraction from human spermatozoa. mAb 4D4 specifically recognizes primate proacrosin (Gallo et al. 1991). Two other monoclonal antibodies were used as controls: (1) a mAb generated at the same time as mAb 4D4 that specifically recognizes a human flagellar protein expressed during spermiogenesis (unpublished data) and (2) an anti-desmin antibody (Dakopatts, Santa Barbara, CA, USA). Secondary antibodies for immunocytochemistry were either biotinylated horse anti-mouse IgG (Vector Laboratories, Burlingame, CA, USA), or goat anti-mouse IgG conjugated to isothiocyanate fluorescein (GAM/FTTC, Nordic, The Netherlands). Secondary antibodies for western blotting were either goat anti-mouse IgG conjugated to peroxidase (BI 2413, Biosys SA, Compiègne, France) or sheep anti-mouse IgG conjugated to 125I (Amer-sham, Buckinghamshire, UK).
Electrophoresis and immunoblotting
Frozen human testis autopsies and frozen human sperm with normal parameters were thawed and homogenized in ice in phosphate-buffered saline (PBS pH 7.2) containing 0.05% Triton X-100, 1 mM phenyl-methyl-sulfonyl-fluoride (PMSF) and 10 μg ml-1 α2-macroglobulin. After centrifugation at 15000g for 2 min, the pellets were sonicated in a buffer composed of 10 mM Tris, 0.9% TV-lauroyl sarcosine, 2.5 M urea, 0.5% Triton X-100, ImM PMSF and 10 μg ml-1 α2-macroglobulin and incubated in ice for 2h. Suspensions were subsequently centrifuged at 15000g for 2min, and SDS–electrophoresis sample buffer was added to the supernatants. Proteins were separated by electrophoresis on 8.5 % polyacrylamide gels according to Laemmli (1970). Transfer to nitrocellulose sheets and subsequent incubations with antibodies were performed according to Towbin et al. (1979) with mAb 4D4 culture supernatant as primary antibody. Antibody binding was detected by either the immunoperoxidase method using diaminobenzidine as a chromogen or by autoradiography with the anti-mouse IgG conjugated to 125I. Controls were performed by omitting mAb 4D4 or by substituting it with the anti-human flagellum antibody (culture supernatant).
Biopsy processing for immunocytochemistry
For paraffin embedding, tissue specimens were trimmed into 4×2 mm blocks, then fixed in either Bouin’s fixative or 7% formaldehyde at room temperature for a time not exceeding 24 h. After the standard clarification procedures through xylol and alcohol, tissues were embedded in low melting point paraffin (56°C). Paraffin sections (3 μm thick) were placed on clear or neoprene-coated glass slides to avoid loosening of the sections during further manipulations. Sections were then dewaxed in toluene, rehydrated through graded alcohols and washed with PBS at pH 7.2.
For frozen testis sections, unfixed testis specimens were embedded in Tissue-Tek (Myles Diagnostic Division, Elkart, USA), snap frozen in liquid nitrogen and stored at −80°C until use. Frozen sections (3 gm thick) were cut with a Reichert cryomicrotome at −25°C, collected on slides as above, air-dried at room temperature for 20 min, fixed in −20°C acetone for 5 min, then either immediately processed or stored in a cellulose and tinfoil wrapping at −80 °C until use. Prior to immunostaining, sections were warmed to room temperature and then washed in PBS three times for 3 min. Some frozen sections were additionally postfixed with 1 % formaldehyde in PBS, then washed again with PBS.
Immunocytochemistry
Immunolabelling was performed by standard indirect immunofluorescence or the three-step immunoperoxidase technique using the biotin-avidin system (Vector Laboratories, Burlingame,CA, USA). Some sections were first treated with either 7.5 % H2O2 in PBS for 10 min at room temperature or with 0.1% trypsin in 0.05M Tris buffer (pH7.6) with 0.1% calcium chloride for a time not exceeding 15 min at 37 °C. The reaction was stopped with 4°C PBS. To block nonspecific antibody binding, all washed sections were incubated at room temperature for 20 min in either 5 % normal horse serum or 5 % BSA (grade V, Sigma, USA) in PBS. Sections were then incubated with mAb 4D4 culture supernatant (diluted 1:2 and 1:4) in a humid chamber at room temperature for lh. After washing in PBS, sections were incubated for 30 min with the fluorescein-conjugated antibody (diluted 1:40 with PBS/BSA) or the biotinylated antibody (diluted 1:250 with PBS/BSA) and washed again. Sections treated with the biotinylated antibody were then incubated for 30 min with preformed avidin-peroxidase complexes, and amino-ethyl-carbazole (AEC) was used as chromogen. For controls, mAb 4D4 was substituted by PBS/BSA or irrelevant monoclonal antibodies (see Antibodies section).
Sections treated with the immunoperoxidase method were counterstained with Harris’s hematoxylin and mounted in an aqueous medium (Glycergel, Dakopatts, Santa Barbara, USA). Sections stained by indirect immunofluorescence were counterstained with 0.02% Evans blue and mounted in Citifluor (glycerol/PBS, Citifluor Ltd, London, UK). Slides were observed under an Olympus BH-2 photo-microscope fitted with standard light, epi-illumination and filters for FITC. Micrographs were taken at ×1250 and ×2500 using either Kodak Ektar 25 or T-MAX 400 films (Kodak, France) and automatic exposure.
Electron microscopy
Testis biopsy specimens from the patients exhibiting atypical sperm cells were fixed for 3 h in 1.2 % glutaraldehyde in 0.1 M Sorensen’s buffer and washed for 30 min in fresh buffer containing 4 % (w/v) sucrose. Post-fixation was for 1 h in 1 % w/v osmic acid in 0.1M Sörensen’s buffer with 4% sucrose. After dehydration through graded alcohols, biopsy specimens were embedded in Araldite. Sections obtained using a Reichert OmU3 ultramicrotome were stained with uranyl acetate and lead citrate and examined in a Siemens Elmiskop CT 150 transmission electron microscope.
Morphological identification of the spermatogenic cells
Spermatogenic stages were identified according to the criteria of Clermont (1970), Holstein and Roosen-Runge (1981), and Schulze and Rehder (1984).
Spermatocytes were classified according to the following criteria: (1) leptotene stage: densely packed, regularly distributed small round patches of chromatin; nuclei of intermediate density (small, irregularly shaped nucleoli eccentrically positioned); (2) zygotene stage: loosely distributed cords or thread-like chromatin patches with a more or less tuft-like orientation; light nuclear area(s) near the nucleolemma; faded nuclear aspect (in favourable cell sections, one or two small darker masses are seen adjacent to the nuclear margin); (3) early pachytene stage: thicker chromatin cords resulting in a denser nuclear aspect; poorly discernible nuclear margin; (4) mid-pachytene stage: scattered aggregates of intensely stained and irregularly shaped coarse chromatin; marked contrast between the chromatin patches and the surrounding karyoplasm; dense area underlying the nucleolemma (in favourable cell sections, three intensely stained masses including a prominent round central nucleolus and two smaller components are observed); (5) late pachytene stage: more diffuse large patches of intensely stained coarse chromatin (Guichaoua et al. 1986); dense nuclear contents and margin; (6) diplotene stage: intensively stained coarse chromatin patches rather regularly disposed; (7) secondary spermatocyte stage: round nucleus with sharply contrasted nuclear envelope and homogeneous or slightly granular dense chromatin (in favourable cell sections, three nucleoli or more are discernible).
Main criteria for spermatid characterization were (1) early spermatids: small round nucleus with uniformly grainy chromatin, acrosomal bleb; (2) mid-spermatids: more or less elongated nucleus and presence of an acrosomal cap; (3) mature spermatids: highly condensed chromatin, elongated and flattened nucleus, acrosome over the anterior nuclear region.
Results
mAb 4D4 labelling in human testis
In testis sections stained by immunofluorescence or immunoperoxidase, mAb 4D4 positively reacted with spermatocytes and spermatids while the other testicular cells were unstained (Figs 1 and 2). A similar mAb 4D4 labelling pattern was found in spermatocytes and spermatids in sections pretreated with H2O2 to quench endogenous peroxidase activity and stained with the immunoperoxidase method. In controls incubated with PBS/BSA without mAb 4D4 (Fig. 2D) or with irrelevant monoclonal antibodies, no staining was observed in spermatocytes or spermatids. When mAb 4D4 was replaced by a monoclonal antibody directed against a sperm flagellar protein, only spermatids were stained at the flagellar level (data not shown). Sections from a panel of other tissues (see Materials and methods) showed no staining with mAb 4D4 under the same conditions.
mAb 4D4 immunolabelling of human seminiferous epithelium (Bouin’s-fixed, paraffin-embedded testis biopsies labelled by the avidin-biotin-immunoperoxidase procedure). ×740. Bar=10 μm. (A,B) Sections of the same seminiferous tubule showing germ cell associations corresponding to stages VI (left-hand side in A), IV–V (right-hand side in A), I (left-hand side in B) and III–IV (right-hand side in B). Spermatogonia (G); Primary spermatocytes at the zygotene (z), pachytene (P), anaphase (a), metaphase-anaphase (m/a), and telophase (t) steps; Secondary spermatocyte (SII); Spermatids at various steps of maturation (T1–T4).
mAb 4D4 immunolabelling of human seminiferous epithelium (Bouin’s-fixed, paraffin-embedded testis biopsies labelled by the avidin-biotin-immunoperoxidase procedure). ×740. Bar=10 μm. (A,B) Sections of the same seminiferous tubule showing germ cell associations corresponding to stages VI (left-hand side in A), IV–V (right-hand side in A), I (left-hand side in B) and III–IV (right-hand side in B). Spermatogonia (G); Primary spermatocytes at the zygotene (z), pachytene (P), anaphase (a), metaphase-anaphase (m/a), and telophase (t) steps; Secondary spermatocyte (SII); Spermatids at various steps of maturation (T1–T4).
mAb 4D4 labelling of the human seminiferous epithelium according to different biopsies processings. (A) Bouin’s-fixed, paraffin-embedded testis sections labelled by indirect immunofluorescence. Fluorescence is detected in 4 pachytene primary spermatocytes (arrows). (B) Phase-contrast micrograph of the same field’as-in A. Pachytene cells are labelled in the Golgi region (arrows and insert). (C) Bouin’s-fixed, paraffin-embedded testis sections labelled by the avidin-biotin-peroxidase complexes. A prominent labelling is observed in the pachytene primary spermatocytes (arrow) and spermatids (arrowhead). (D) Control testis section treated as in C except that mAb 4D4 has been substituted by PBS/BSA. Germ cells are unlabelled. (E) Testis section processed as in C but treated with 0.1% trypsin during 10min prior to immunolabelling. Pachytene primary spermatocytes are poorly labelled (arrows). (F) Frozen section labelled by indirect immunofluorescence microscopy. The pachytene primary spermatocytes exhibit a faint and spotty fluorescence (arrows). A–D, ×840; E, ×425; F, ×890. A–F, bar=20 μm.
mAb 4D4 labelling of the human seminiferous epithelium according to different biopsies processings. (A) Bouin’s-fixed, paraffin-embedded testis sections labelled by indirect immunofluorescence. Fluorescence is detected in 4 pachytene primary spermatocytes (arrows). (B) Phase-contrast micrograph of the same field’as-in A. Pachytene cells are labelled in the Golgi region (arrows and insert). (C) Bouin’s-fixed, paraffin-embedded testis sections labelled by the avidin-biotin-peroxidase complexes. A prominent labelling is observed in the pachytene primary spermatocytes (arrow) and spermatids (arrowhead). (D) Control testis section treated as in C except that mAb 4D4 has been substituted by PBS/BSA. Germ cells are unlabelled. (E) Testis section processed as in C but treated with 0.1% trypsin during 10min prior to immunolabelling. Pachytene primary spermatocytes are poorly labelled (arrows). (F) Frozen section labelled by indirect immunofluorescence microscopy. The pachytene primary spermatocytes exhibit a faint and spotty fluorescence (arrows). A–D, ×840; E, ×425; F, ×890. A–F, bar=20 μm.
The efficiency of the proacrosin immunolabelling was found to be dependent on the technical procedures used. Staining of germ cells with mAb 4D4 by standard indirect two-step immunofluorescence was faint, especially in primary spermatocytes (Fig. 2A,F). The three-step avidin-biotin-immunoperoxidase technique gave a more conspicuous labelling of both the spermatocytes and spermatids (Fig. 2C). In acetone-fixed frozen testis sections, only a few spermatocytes were labelled with mAb 4D4. Moreover, proacrosin appeared to diffuse around the spermatocyte nuclei and outside the spermatid acrosomes. Postfixation of frozen sections with formaldehyde prior to immunolabelling slightly increased proacrosin preservation following thawing as revealed by the greater number of labelled spermatids and the decrease of labelling diffusion (Fig. 2F). In testis sections fixed with Bouin’s fluid prior to paraffin embedding, numerous spermatocytes and almost all spermatids exhibited a strong and welldefined labelling (Fig. 1). A similar staining pattern was obtained in sections fixed in Bouin’s fluid and pretreated with trypsin (Fig. 2E). When Bouin’s fluid was replaced by formaldehyde, the incidence of labelled spermatocytes and spermatids was variable from one biopsy to another, some of them exhibiting scarce labelling (data not shown).
mAb 4D4 specificity to testis proacrosin determined by Western blotting
As previously determined (Gallo et al. 1991), mAb 4D4 recognized the 50 and 55×103Mr proacrosin in ejaculated human sperm cells (Fig. 3C, lane 3). Detection of the mAb 4D4 binding in testis extracts by the immunoperoxidase method (which gave optimally labelled spermatocytes in sections) (Fig. 3B) or the autoradiographic method (Fig. 3C) showed that only the 50 and 55×10 Mr proacrosin polypeptides, with the exclusion of any other component, were recognized by mAb 4D4 in the human testis.
Identification of the 4D4 epitope-bearing polypeptides in human testis. (A) Coomassie-blue-stained gel showing human testis proteins separated by SDS–PAGE under reducing conditions. Lane 1: fraction insoluble in Triton X-100. Lane 2: fraction solusble in Triton X-100. (B,C) Corresponding nitrocellulose replica incubated with mAb 4D4 and revealed either by the immunoperoxidase method (B) or by autoradiography with anti-mouse IgG conjugated to l25I (C). The 4D4 antibody recognized polypeptides in the 50–55 ×103MrAft range in the testis fraction insoluble in Triton X-100 (lanes 1) while the testis fraction soluble in Triton X-100 were unlabelled by 4D4 (lanes 2). The polypeptides recognized by mAb 4D4 in testis were in the same molecular mass range that those recognized by 4D4 in sperm fraction insoluble in Triton X-100 (lane 3). The 55×103Mrcorresponds to human proacrosin and the 50×103Mr to an intermediate form between proacrosin and acrosin which appears following proacrosin cleavage during protein extraction (Gallo et al. 1991). Lane 4 and left-hand side: relative molecular mass markers.
Identification of the 4D4 epitope-bearing polypeptides in human testis. (A) Coomassie-blue-stained gel showing human testis proteins separated by SDS–PAGE under reducing conditions. Lane 1: fraction insoluble in Triton X-100. Lane 2: fraction solusble in Triton X-100. (B,C) Corresponding nitrocellulose replica incubated with mAb 4D4 and revealed either by the immunoperoxidase method (B) or by autoradiography with anti-mouse IgG conjugated to l25I (C). The 4D4 antibody recognized polypeptides in the 50–55 ×103MrAft range in the testis fraction insoluble in Triton X-100 (lanes 1) while the testis fraction soluble in Triton X-100 were unlabelled by 4D4 (lanes 2). The polypeptides recognized by mAb 4D4 in testis were in the same molecular mass range that those recognized by 4D4 in sperm fraction insoluble in Triton X-100 (lane 3). The 55×103Mrcorresponds to human proacrosin and the 50×103Mr to an intermediate form between proacrosin and acrosin which appears following proacrosin cleavage during protein extraction (Gallo et al. 1991). Lane 4 and left-hand side: relative molecular mass markers.
Cellular location of proacrosin during spermatogenesis
Zygotene and early pachytene primary spermatocytes were unlabelled by mAb 4D4 (Fig. 1). In contrast, midpachytene primary spermatocytes exhibited a prominent more or less elongated labelling (being up to 3 μm in length) (Fig. 4C). Indirect immunofluorescent staining revealed that this labelling was located at the level of the Golgi complex (Fig. 2A,B). Inside this strongly labelled body, clear areas were observed (Fig. 5A). In some other cells, several protrusions seemed to emerge from this stained mass (Fig. 5B) which, in some other cases, assumed the appearance of a network of elongated elements with various thicknesses (Fig. 5C).
mAb 4D4 immunolabelling of human spermatocytes and spermatids (Bouin’s-fixed, paraffin-embedded testis biopsies labelled by the avidin-biotin-immunoperoxidase procedure). ×1250. Bar=10 μm. (A) The various labelling patterns of late-pachytene primary spermatocytes. Two labelled bodies are observed close to one another (arrowhead), or at various distances from oneanother or even at opposite poles. Arrows indicate bodies apparently less stained due to their location in another plane. (B) Late spermatids exhibiting labelling of the whole acrosome. (C–F): mAb 4D4-labelling patterns of the pachytene primary spermatocyte as a function of cell maturation: (C) single labelling being either round-shaped (left-hand side) or elongated (right-hand side); (D) single labelled body in which distinct subunits are discernible; (E) two labelled bodies being distant from one another; (F) two labelled bodies at opposite nuclear poles. (G) Metaphasic primary spermatocyte showing two labelled bodies on opposite sides of the metaphase plate margin. (H) Early anaphase primary spermatocyte exhibiting labelled bodies situated each in one half of the microtubular spindle. (I) More advanced anaphase primary spermatocyte showing labelled bodies close to the spindle poles. (J) Telophase primary spermatocyte exhibiting a labelled body close to each chromatin mass (right-hand side). The arrow indicates the less discernible labelled body due to its location in another plane. (K) Interphasic secondary spermatocytes exhibiting two labelled bodies (arrows). (L) Young spermatids with a single labelled body (left-hand side). They can be distinguished from secondary spermatocytes (right-hand side) due to their smaller nuclear diameter. (M) More differentiated spermatids exhibiting mAb 4D4-labelling extension. Acrosomal vesicles that have released some proacrosin contents exhibit clear centers. (N) Still more differentiated spermatids exhibiting labelling of the whole acrosomal cap which extents over the anterior face of the nucleus.
mAb 4D4 immunolabelling of human spermatocytes and spermatids (Bouin’s-fixed, paraffin-embedded testis biopsies labelled by the avidin-biotin-immunoperoxidase procedure). ×1250. Bar=10 μm. (A) The various labelling patterns of late-pachytene primary spermatocytes. Two labelled bodies are observed close to one another (arrowhead), or at various distances from oneanother or even at opposite poles. Arrows indicate bodies apparently less stained due to their location in another plane. (B) Late spermatids exhibiting labelling of the whole acrosome. (C–F): mAb 4D4-labelling patterns of the pachytene primary spermatocyte as a function of cell maturation: (C) single labelling being either round-shaped (left-hand side) or elongated (right-hand side); (D) single labelled body in which distinct subunits are discernible; (E) two labelled bodies being distant from one another; (F) two labelled bodies at opposite nuclear poles. (G) Metaphasic primary spermatocyte showing two labelled bodies on opposite sides of the metaphase plate margin. (H) Early anaphase primary spermatocyte exhibiting labelled bodies situated each in one half of the microtubular spindle. (I) More advanced anaphase primary spermatocyte showing labelled bodies close to the spindle poles. (J) Telophase primary spermatocyte exhibiting a labelled body close to each chromatin mass (right-hand side). The arrow indicates the less discernible labelled body due to its location in another plane. (K) Interphasic secondary spermatocytes exhibiting two labelled bodies (arrows). (L) Young spermatids with a single labelled body (left-hand side). They can be distinguished from secondary spermatocytes (right-hand side) due to their smaller nuclear diameter. (M) More differentiated spermatids exhibiting mAb 4D4-labelling extension. Acrosomal vesicles that have released some proacrosin contents exhibit clear centers. (N) Still more differentiated spermatids exhibiting labelling of the whole acrosomal cap which extents over the anterior face of the nucleus.
mAb 4D4-Iabelling patterns in pachytene primary spermatocytes. (A) Mid-pachytene primary spermatocyte showing clear areas inside the strongly labelled Golgi area (arrow). (B,C) Pachytene primary spermatocytes exhibiting various labelling patterns of the Golgi region. The labelled areas in B are suggestive of structures from which four protrusions seem to emerge (arrow). In C, labelling evokes a network of elements (arrowhead). (D) Late pachytene primary spermatocyte exhibiting labelling near each nuclear pole: one labelled area is prominent and assumes the same aspect as in C (arrow), the other is smaller and round shaped (arrowhead). (A–D) ×2300. Bar=5 μm.
mAb 4D4-Iabelling patterns in pachytene primary spermatocytes. (A) Mid-pachytene primary spermatocyte showing clear areas inside the strongly labelled Golgi area (arrow). (B,C) Pachytene primary spermatocytes exhibiting various labelling patterns of the Golgi region. The labelled areas in B are suggestive of structures from which four protrusions seem to emerge (arrow). In C, labelling evokes a network of elements (arrowhead). (D) Late pachytene primary spermatocyte exhibiting labelling near each nuclear pole: one labelled area is prominent and assumes the same aspect as in C (arrow), the other is smaller and round shaped (arrowhead). (A–D) ×2300. Bar=5 μm.
Late pachytene spermatocytes as cells at an intermediate stage between mid- and late pachytene showed two distinct labelled bodies of equal size or not (Figs 4A,E,F and 5D). They were located near the nucleus either close to each other (Fig. 4A), or at various distances from one another (Fig. 4E,F) including at opposite nuclear poles (Fig. 5D).
Metaphasic primary spermatocytes exhibited two labelled round bodies close to the metaphasic plate (Fig. 4G). At the anaphase stage, the labelled dots were close to the microtubular spindle and at various distances from the spindle poles (Fig. 4H,I). Finally, in telophase primary spermatocytes a labelled body was observed in the vicinity of each chromatin mass (Fig. 4J). Interphasic secondary spermatocytes exhibited two labelled bodies located close to the nuclear margin (Fig. 4K,L).
Additional smaller labelled dots were sometimes observed in primary spermatocytes during chromosome movement (Fig. 4H) and in secondary spermatocytes (Fig. 4K). Whether they correspond to scattered Golgi vesicles could not be determined at the light microscope level. Several attempts to obtain mAb 4D4 immunogold labelling on Lowicryl-embedded testis sections were performed but the very low probability of observing spermatocytes undergoing meiotic divisions as secondary spermatocytes in electron microscopy have led us to tackle the question of the organelle location of proacrosin by investigating pathological models (see below).
Early spermatids contained a single round labelled body apposed to the nuclear outer border (Fig. 4L). In more differentiated spermatids, the labelled body appeared to be flattened on the side facing the nucleus (Fig. 4M). Still more differentiated spermatids exhibited a crescent-shaped strongly labelled cap over the anterior face of the nucleus (Fig. 4N). As spermatid maturation proceeds, the amount of proacrosin considerably increased as revealed by the enlargement of the labelled area (Fig. 4L–N). In mature spermatids, the whole acrosomal region was labelled (Fig. 4B).
Proacrosin labelling in the case of spermatocyte division anomalies
Since mAb 4D4-stained proacrosin appeared to be located in the Golgi region of primary spermatocytes (Figs 2, 5) and partitioned during meiosis (Fig. 4), biopsies from patients exhibiting different types of meiotic division anomalies were taken as models to elucidate whether proacrosin scattering depends on the Golgi complex partitioning.
In a patient with spermatocyte monopolar halfspindles (Fig. 6C), anaphase I-stage was not encountered, which suggests post-metaphase chromosome movement inability leading to giant spermatozoa (Escalier, 1983, 1985). Staining of testis sections with mAb 4D4 revealed two main anomalies. First, metaphasic primary spermatocytes exhibited a labelled body on only one side of the equatorial plate (Fig. 6A,B) which was twice as large as in normal metaphasic spermatocytes (3 μm vs 1.5 μm in length) (Fig. 4G). This labelled body was comparable to that observed in normal mid-pachytene spermatocytes (Fig. 4C). Second, early spermatids showed a labelled body being up to 6.7 μm in length (Fig. 6D) vs 2.3 μm in normal spermatids (Fig. 4M). As shown in electron microscopy, this spermatid region contained an abnormally prominent Golgi complex extending all over the acrosome in formation (Fig. 6E).
mAb 4D4-labelling pattern in human testis bearing meiotic division anomalies. (A, B, D, F, H) Bouin’s-fixed, paraffin-embedded testis biopsies labelled by the avidinbiotin-immunopcroxidase procedure. (C, E, G, I) Electron micrographs. (A–E) Sections from a biopsy characterized by postmetaphase chromosome movement failure. (A,B) Metaphasic primary spermatocytes exhibiting an abnormally voluminous 4D4 labelled body on only one side of the metaphase plate (arrowheads) as seen in polar (A) and lateral (B) views of the plate. ×2000. Bar=5 μm. (C) Metaphasic primary spermatocyte as seen in electron microscopy showing a monopolar half-spindle (ms). ×6500. Bar=3 μm. (D) Young spermatids exhibiting an abnormally prominent labelled body (arrows). ×1900. Bar=10 μm. (E) Early spermatid exhibiting Golgi units (arrows) which abnormally extends all over the acrosome in formation (A). ×12000. Bar=2 μm. (F–I) Sections from a biopsy characterized by spermatocyte cytokinesis failure. (F) In this early multinucleate spermatid, mAb 4D4 staining reveals a single and prominent round labelled body which is centrally placed (arrow). ×3000. Bar=5 μm. (G) An unscattered round-shaped Golgi complex (GC) is present in the space between the nuclei (N) of multinucleate spermatids. ×13600. Bar=1.5 μm. (H, I) Spermatids more differentiated than in F exhibit a flattened labelled body comprising several extensions (arrows in H). This body corresponds to a single acrosome (a in I) spreading between the nuclei (N). (H) ×2600. Bar=5 μm. (I) ×19 500. Bar=1 μm.
mAb 4D4-labelling pattern in human testis bearing meiotic division anomalies. (A, B, D, F, H) Bouin’s-fixed, paraffin-embedded testis biopsies labelled by the avidinbiotin-immunopcroxidase procedure. (C, E, G, I) Electron micrographs. (A–E) Sections from a biopsy characterized by postmetaphase chromosome movement failure. (A,B) Metaphasic primary spermatocytes exhibiting an abnormally voluminous 4D4 labelled body on only one side of the metaphase plate (arrowheads) as seen in polar (A) and lateral (B) views of the plate. ×2000. Bar=5 μm. (C) Metaphasic primary spermatocyte as seen in electron microscopy showing a monopolar half-spindle (ms). ×6500. Bar=3 μm. (D) Young spermatids exhibiting an abnormally prominent labelled body (arrows). ×1900. Bar=10 μm. (E) Early spermatid exhibiting Golgi units (arrows) which abnormally extends all over the acrosome in formation (A). ×12000. Bar=2 μm. (F–I) Sections from a biopsy characterized by spermatocyte cytokinesis failure. (F) In this early multinucleate spermatid, mAb 4D4 staining reveals a single and prominent round labelled body which is centrally placed (arrow). ×3000. Bar=5 μm. (G) An unscattered round-shaped Golgi complex (GC) is present in the space between the nuclei (N) of multinucleate spermatids. ×13600. Bar=1.5 μm. (H, I) Spermatids more differentiated than in F exhibit a flattened labelled body comprising several extensions (arrows in H). This body corresponds to a single acrosome (a in I) spreading between the nuclei (N). (H) ×2600. Bar=5 μm. (I) ×19 500. Bar=1 μm.
In another patient with spermatocyte cytokinesis failure, mAb 4D4 revealed a large labelled body in the space between nuclei (Fig. 6F). In this cell region, electron microscopy showed the presence of a roundshaped Golgi complex (Fig. 6G) as normally found in pachytene primary spermatocytes. This Golgi complex gave rise to an acrosome with several extensions as revealed by both 4D4 staining (Fig. 6H) and electron microscopy (Fig. 6I).
Discussion
Northern blot analysis of mouse mRNA has suggested a proacrosin biosynthesis in both meiotic and postmeiotic spermatogenic cells (Kashiwabara et al. 1990). Data obtained by means of the monoclonal antibody 4D4 support this notion in human spermatogenesis and show that proacrosin appears as early as the prophase of the first meiotic division. Positive reactivity with mAb 4D4 was first observed in pachytene primary spermatocytes which exhibited an intensely labelled Golgi complex. During the subsequent meiotic steps, proacrosin labelling was distributed among daughter cells. Finally, an increased amount of labelled proacrosin was observed during spermatid maturation.
Immunodetection of testicular proacrosin
Previous studies on proacrosin detection during mammalian spermatogenesis using antibodies were performed on either frozen testis sections (Kallajoki and Suominen, 1984; Kallajoki et al. 1986) or isolated germ cells (Flörke et al. 1983a,b;Phi-Van et al. 1983; Mansoiiri et al. 1983) with a two-step labelling method.
All these studies indicated that only spermatids were labelled, except in the guinea pig, which showed a few labelled primary spermatocytes (Arboleda and Gerton, 1988). Using mAb 4D4, we found that only the three-step immunolabelling on Bouin’s-fluid-fixed, paraffinembedded testis sections enables the analysis of proacrosin distribution in germ cells to be performed.
Other acrosomal constituents have also been found to be first expressed in rodent primary spermatocytes (Fendersen et al. 1984; O’Brien et al. 1988; Toppari et al. 1985a,b), particularly the acrogranin, an acrosomal glycoprotein also present in pachytene spermatocytes (Anakwe and Gerton, 1990).
The immunodetection of proacrosin in germ cells may considerably be influenced by its molecular diversity (Hardy et al. 1987). The anti-acrosin antibodies generated from acrosin (Flörke et al. 1983a,b, Phi-Van et al. 1983) may not recognize the antigenic determinants of the primary proacrosin form or may be directed against new determinants produced during spermiogenesis (Arboleda and Gerton, 1988). It is the case for monoclonal antibody C11H which has been found to recognize proacrosin only in late spermatids in the mouse (Kallajoki et al. 1986). In contrast, this study shows that mAb 4D4 is directed against an epitope present as soon as primary proacrosin is expressed in the human testis.
Proacrosin and spermatogenic events
The mechanism of cytoplasmic organelles partitioning during meiosis remains unknown (Suarez-Quian et al. 1991). mAb 4D4 has allowed us to show how the labelled body containing proacrosin in mid-pachytene spermatocytes gives rise to smaller bodies which are distributed among daughter cells.
It has been suggested that forerunners of the proacrosomal vesicle first appear in zygotene primary spermatocytes (Nicander and Ploën, 1969; Fawcett, 1975; Holstein and Roosen-Runge, 1981). In pachytene spermatocytes, the Golgi body comprises several units (Kerr and de Kretzer, 1981) and its size increases due to an accumulation of trans Golgi during pachytene maturation (Suarez-Quian et al. 1991). The Golgi units scatter during diakinesis (Holstein and Roosen-Runge, 1981). Finally, during the first three steps of spermiogenesis, the dense acrosomal matrix increases in size and electron density due to the incorporation of additional vesicles delivered by the Golgi apparatus (Burgos and Gutierrez, 1986; Thome-Tjomsland et al. 1988).
The ontogeny of proacrosin may, therefore, parallel the biogenesis of acrosome-related Golgi vesicles and reflect the scattering of the Golgi complex units. Data on two types of germ cell division failures are also in favour of this hypothesis, since, in these cases, both the Golgi complex and the 4D4 labelling failed to scatter during spermatogenesis indicating that proacrosin was sequestered in the Golgi body. In one case, the formation of meiotic monopolar half-spindles led to postmetaphase chromosome movement inability (Escalier, 1985). Nevertheless, chromosome replication and postmeiotic differentiation occurred, giving rise to giant spermatids (Escalier, 1983). In the other case, the nuclear events occurred normally but the absence of spermatocyte cytokinesis led to the formation of multinucleate spermatids (Holstein et al. 1988).
Haploid expression of the preproacrosin gene has been demonstrated in mammals (Adham et al. 1989; Klemm et al. 1990). Our data suggest a meiotic and postmeiotic proacrosin synthesis in the human as in the mouse in which acrosin mRNA, first detected in pachytene spermatocytes, has been found to increase in amount in round spermatids (Kashiwabara et al. 1990). Different mechanisms may be involved. First, the gene may be activated during meiosis and reexpressed postmeiotically (Bellvé and O’Brien, 1983; O’Brien, 1987), as is the case for sperm cytochrome C (Golberg et al. 1977) and LDH-X (Hintz and Goldberg, 1977). Alternatively, meiotic and postmeiotic translation products may differ, as is the case for the sperm histone H3 gene (Kennedy et al. 1985). Such stage-specific changes leading to the synthesis of distinct protein isoforms are known to occur also for sperm actin and tubulin (Hecht et al. 1984; Kim et al. 1989; Waters et al. 1985; Slaughter et al. 1989).
In conclusion, the anti-proacrosin monoclonal antibody 4D4 appears to be a useful probe for investigating proacrosin biosynthesis and acrosome biogenesis in man as well as for studying the partitioning pattern of the Golgi apparatus during meiosis. Such properties may be of particular interest in analysing various spermatogenesis disorders or arrests in human infertile patients.
ACKNOWLEDGEMENTS
We would like to thank Professor J. M. LUCIANI and Dr M. R. GUICHAOUA (Laboratoire de Cytogénétique et de Biologie de la Reproduction, CHU Marseille, France) for helpful discussions, Mrs N. PARSEGHLAN for technical assistance, Dr A. VIELLEFOND (Laboratoire d’AnatomoPathologie, CHU Bicêtre, France) and the Deparment of Urology (CHU Bicêtre, France) for providing tissues, Dr M. APPLANAT (U 135, INSERM, CHU Bicêtre, France) for help in autoradiography and Mrs F. SIRYANI for help in translation. This work was supported by grants from the Agence Nationale de Valorisation de la Recherche (No x-84-07-039T-029-0) and from the Ministère de la Recherche et de la Technologie (MRT,No 85T-0873). Dr D. BERMUDEZ was supported by a fellowship from the DGICYT (Ministry of the Education, Spain).