| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a The Sperm Physiology Laboratory,
b Department of Obstetrics and Gynecology, W.M. Keck Foundation Biotechnology Resource Laboratory, Yale University School of Medicine, New Haven, Connecticut 06510
c Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711
ABSTRACT
We previously described a putative creatine kinase M isoform in human sperm that is developmentally regulated and expressed during late spermiogenesis, simultaneous with cytoplasmic extrusion. We have now identified this protein as the testis-expressed 70-kDa heat shock protein chaperone known as HspA2 (the human homologue of mouse Hsp70-2). We have isolated and characterized HspA2 (formerly CK-M) by amino acid sequencing and have localized it by immunocytochemistry to spermatocytes at low levels, to spermatids, and in the tail of mature sperm. The specificity of the CK-M/HspA2 antiserum to HspA2 was demonstrated on immunoblots of one- and two-dimensional SDS-PAGE. In agreement with our earlier biochemical data, immunocytochemistry of testicular tissue indicated that HspA2 is selectively expressed in mature spermatids and in sperm about to be released in the seminiferous tubuli. The identity of HspA2 has been further confirmed by cross-absorption of the mouse HSP70-2 antibody by the HspA2/CK-M fraction, and by identical immunostaining patterns of human testicular tissue using either the anti-CK-M/HspA2 or an anti-mouse Hsp70-2 antisera. During spermiogenesis, both cytoplasmic extrusion and plasma membrane remodeling, which facilitate the formation of the zona pellucida binding site, involve major intrasperm protein transport, which may be chaperoned by HspA2. Accordingly, in immature human sperm, which fail to express HspA2, there is cytoplasmic retention and lack of zona pellucida binding. The present findings provide the biological rationale for the role of the human HspA2 as an objective biochemical marker of sperm function and male fertility, which we have established in earlier clinical studies.
fertilization, gametogenesis, meiosis, sperm, sperm maturation, spermatid, spermatogenesis
INTRODUCTION
The primary interest of our laboratory has been the development of objective biochemical markers of human sperm maturity and function that would predict male fertility, independently from the traditional semen criteria of sperm concentration and motility. In measurements of sperm creatine-N-phosphotransferase or creatine kinase (CK), we found what appeared to be significantly higher sperm CK activities in men with diminished fertility [1–3]. There was also a proportional increase of cytoplasmic protein content in diminished-fertility sperm heads. The combination of increased CK and other cytoplasmic protein concentrations, along with diminished fertility, suggested to us that we had identified a sperm developmental defect in the last phase of spermiogenesis, when the cytoplasm is normally extruded and remains in the adluminal area as "residual bodies" [4]. Thus, we hypothesized that the diminished fertility of men with high sperm CK concentrations was related to the arrest of sperm maturation; sperm released from the germinal epithelium carrying surplus residual cytoplasm represent impaired sperm production, yielding immature and functionally defective gametes.
Using the analogy of muscle development, where in addition to the B-type CK isoform, an M-type isoform is also expressed [5], we studied the potential presence of a CK-M isoform. Indeed, upon electrophoretic analysis of human sperm extracts, there appeared to be two types of CK isoforms with retardation factors corresponding to those of the muscle CK-B and CK-M [6]. The inverse correlation between proportions of sperm with cytoplasmic retention and CK-M expression, as measured by sperm CK activity and CK-M ratios [%CK-M/(CK-M + CK-B)], respectively (in three independent studies r = -0.69, -0.71, and -0.76, P < 0.001, n = 159, 134, and 194), indicated that cytoplasmic extrusion and the commencement of CK-M isoform synthesis are related, developmentally regulated spermiogenetic events [6–8].
In studying the sperm CK-M versus muscle CK-M, we noted several differences: 1) the conventional CKs in muscle occur as homodimer species of CK-BB and CK-MM and as a heterodimer of CK-MB. In sperm extracts we found the CK-BB and an apparent CK-MM peak, but there was no heterodimer [6]. When we dissociated the sperm CK homodimers and allowed reassociation of sperm isoforms in the presence of muscle CK-M, muscle CK-M, unlike sperm CK-M, formed heterodimers with the sperm CK-B isoform, indicating that sperm CK-M and muscle CK-Ms were different proteins ([6] Fig. 1). 2) Using a spectrophotometric ATP detection assay, similarly to sperm CK-B, sperm CK-M exhibited ATP synthesis in the presence of creatine phosphate and ADP. However, it did not show the reverse reaction of generating creatine phosphate and ADP from creatine and ATP, as conventional CKs and sperm CK-B do. 3) Sperm CK-M activity was detected at 120–140 kDa in fractions of gel filtration columns, whereas conventional CK homodimer or heterodimer activities elute at about 80 kDa. 4) Sperm CK-M did not cross-react with antibodies that recognize conventional CK-B and CK-M or mitochondrial CK isoforms, which show high sequence homology [9]. Until now, we assumed that the sperm CK-M is a unique CK specific to sperm or germ cells.
|
Regardless of the uncertain identity of the protein, in the past few years we have utilized CK-M as an objective biochemical marker, demonstrating that mature and immature sperm are different in their degree of cytoplasmic retention [3], fertility [8, 10], morphological and morphometrical attributes [11], and zona pellucida-binding properties [12]. Furthermore, we have established that in spermiogenesis, simultaneously with cytoplasmic extrusion and commencement of CK-M synthesis, the sperm plasma membrane also undergoes a maturation-related remodeling [13]. We have also shown, along with another laboratory, that immature sperm have increased rates of lipid peroxidation and DNA fragmentation [14, 15] and a higher incidence of chromosomal aneuploidy [16] (see Fig. 2, a summary of our concept on sperm maturation). Finally, we established that all sperm maturational events that are detected by CK-M expression are completed by the time sperm enter the caput epididymis [17].
The predictive value of CK-M ratios in the assessment of fertility in men was tested in two blinded studies of couples undergoing in vitro fertilization (IVF). In the first, we classified 84 men (without any information on their semen parameters or reproductive history) on the basis of only their CK-M ratios into high likelihood (>10% CK-M ratio) and low likelihood (<10% CK-M ratio) for fertility [10]. All pregnancies occurred in the high likelihood group. More recently, we re-examined the utility of CK-M ratios in predicting IVF failure in 194 couples treated at Yale University. Similar to the 1992 study, none of the 15 men with <10% CK-M ratios achieved pregnancy whether they had low or high sperm concentrations [8].
In the present work, we set out to conclusively identify the sperm CK-M. This goal was potentially important for two reasons: 1) because this protein is expressed under developmental regulation, its identity would provide further insights into the cell biology of spermiogenesis; and 2) the development of monoclonal antibodies to sperm CK-M would allow us to convert the sperm CK-M assay from an electrophoretic to an immunological approach. The data from amino acid sequencing, one- and two-dimensional SDS-PAGE, and immunocytochemistry of testicular tissue demonstrate that CK-M is in fact HspA2, the human homologue of mouse Hsp70-2 chaperone protein. The mouse Hsp70-2 is preferentially expressed in spermatogenic cells and, similarly to HspA2 in men, affects male fertility [18].
MATERIALS AND METHODS
Sperm Collection and Extraction for Isolationof CK-M/HspA2
Typically, 4–6 x 109 sperm were used for each CK-M/HspA2 preparation attempt. Because sperm originate in the leftover portion of semen analysis samples, we performed sperm extraction and preservation daily. Fresh semen, after liquifaction, was washed in at least 10 volumes of ice cold 0.15 M NaCl, 30 mM imidazole, pH 7.0 by centrifugation at 5000 x g for 20 min. The resulting pellet was extracted in ice cold 30 mM imidazole pH 7.0, 5 mM DL-dithiothreitol (DTT), 0.1% Triton X-100, and 10% glycerol (all chemicals used were from Sigma Chemical Co., St. Louis, MO). The extract was cleared with centrifugation at 5000 x g for 20 min, and the supernatant was stored at -70°C.
Purification of Sperm CK-M/HspA2 by AccellIon-Exchange Chromatography
All procedures were carried out at 4°C. The sperm extracts were thawed, vortexed, and cleared by centrifugation at 5000 x g for 10 min. The supernatant was then dialyzed overnight (2 changes) against 20–25 volumes of Accell column buffer (50 mM sodium acetate pH 7.3, 3 mM DTT, 1 mM sodium azide, 0.2 mM EGTA, 0.01% Triton X-100). The extract was then centrifuged at 10 000 x g for 10 min. An aliquot of the cleared extract was removed for determination of total protein, CK activity, and CK-M%. The remainder was applied to an Accell column.
Accell resin (Accell CM-cation exchanger; Millipore Corp., Milford, MA) was swollen overnight in column buffer with 0.01% Triton. After pouring and settling, the column (typically 12 x 0.9 cm) was washed with 50 ml of column buffer. The sperm extract was applied to the column, and after the flow-through peak was collected, the column was washed with another 30 ml of buffer. The CK-M/HspA2-containing peak was eluted stepwise by changing the elutrient to 0.35 M NaCl in column buffer. Subsequently, 0.5 ml fractions were collected and monitored for protein content at OD280, and for CK activity in the flow-through and salt-eluted peaks, respectively. For further purification of CK-M/HspA2, fractions with the highest CK-M activity/protein ratio were dialyzed overnight against column buffer (0.01% Triton in order to remove NaCl), and were rechromatographed on a smaller (typically 6 x 0.9 cm) Accell column, which was developed similarly to the first column.
CK Activity and CK-M Isoform Ratio Determinations
CK parameters of the sperm extract in the flow-through (where the CK-B isozyme was located) and in the salt-eluted CK-M-containing fractions were determined by using a spectophotometric assay (kit UV-47; Sigma, St. Louis, MO) and agarose gel electrophoresis, followed by fluorescent detection of the ATP generated by the CK enzymes, as described previously [8, 13].
Immunization of Rabbits
Rabbits used for immunization were first bled for preimmune serum. For primary immunization, the rabbits were shaved at four sites on the back and leg. The CK-M/HspA2-containing column fractions were emulsified in complete Freunds adjuvant (1:1 ratio), and inoculated at these four sites using approximately 40 µg inoculum per site. Rabbits were boosted 3 and 6 wk later. All procedures were carried out with the assistance of Yale Animal Care personnel according to approved protocols.
Immunostaining of Sperm
Fresh sperm were washed in at least 10 volumes of human tubal fluid (HTF) medium (Irvine Scientific, Irvine, CA) at 400 x g for 18 min. The supernatant was aspirated and the pellet suspended in phosphate buffer (PB)-sucrose (0.1 M sodium phosphate buffer, 0.45% sucrose, pH 7.0) to a final concentration of 10 to 15 million sperm per ml. Sperm were allowed to settle overnight in a humidity chamber at 5°C onto polylysine-treated slides. All subsequent steps were also carried out in a humidity chamber. The residual fluid was carefully aspirated without disturbing the attached sperm, and 3.7% formalin in PB-sucrose was added. After 20 min at 37°C, the fluid was aspirated and the slide was allowed to dry.
Prior to immunostaining, the fixed sperm were blocked using albumin at 3% in PB-sucrose for 1 h at 37°C, followed by three subsequent washes with PB-sucrose. For CK-M/HspA2 or Hsp70-2 immunostaining, the first antibody was applied to the fixed sperm at, typically, a 1 to 1000–5000 dilution in PB-sucrose, 0.1% albumin. The slide was kept overnight at 5°C, and the ABC process was carried out with the respective second antibodies using the Vector kit (Vector Laboratories Inc., Burlingame, CA) followed by color development using a diaminobenzidine (DAB) kit. Upon completion of the procedure, the slides were washed with distilled water, air dried, dipped in xylene, and mounted.
We have used three antisera for these immunotechnical experiments: our CK-M/HspA2 antisera raised in rabbits; the anti-mouse Hsp70-2 rabbit polyclonal antisera, which was a generous gift from Dr. E.M. Eddy of the National Institute for Environmental Health Sciences (NIEHS; Research Triangle Park, NC); and a commercial Hsp70 antiserum from StressGen Biotech (BC, Canada). Because the mouse testis work [18–19] was directly relevant to our experiments, for comparison, in Figures 4, 7, and 8 we present data developed by CK-M/HspA2 and anti-mouse Hsp70-2 antibodies. Use of perimmune serum or omission of the first antibody in various experiments caused a loss of the respective immunostaining pattern.
Immunostaining of Human Testicular Sections
Slides containing sections of human testicular tissue were prepared from biopsy specimens fixed with Buoins medium. Slides cut from blocks were heated at 60°C for about 1/2 h followed by deparaffinization in three 5-min xylene baths followed by three 2-min dips in 100% ethanol [20]. Slides were then washed in running water for 10 min. After drying, a pap pen circle was drawn around the tissue section. Slides were then incubated in quenching solution (0.6% hydrogen peroxide in absolute methanol) for 15 min and washed in running water, in PBS, and then blocked in 1.5% serum (same species as second antibody) for 1 h. After draining, the primary antiserum (anti-CK-M/HspA2, anti-HSP70-2, or StressGen) was applied in PBS/BSA, typically at 1:1000 to 1:5000 dilution. The slides were left overnight at 5°C, and the further staining steps were carried out using the ABC procedure. Omission of the first antibody caused a loss of the characteristic immuno pattern.
Immunostaining of the HspA2 Peptides
Three peptides, arising from different distances from the N-terminal of the chaperone protein (see Fig. 4, residues with green overlay 38–50, 114–127, 140–156) were synthesized and immunostained with the CK-M/HspA2 antibody and with the anti-HSP70 mouse antisera from NIEHS and StressGen Biotechnologies Corp. (Victoria, BC, Canada). The three HspA2 peptides were spotted in lines of circles on nitrocellulose paper (Schleicher & Schuell, Inc., Keene, NH), three for the peptides; and the fourth spot, which was used for control, treated only with the second antibody. The paper was blocked with 1.0% gelatin for 1 h, the CK-M/HspA2, HSP70, and Stressgen antisera (1:1000 dilutions) were applied to the respective circles, and the nitrocellulose paper was immunostained according to a procedure similar to that used for staining testicular sections.
PAGE-SDS Electrophoresis and Western Blotsof Sperm Proteins
Extraction of sperm proteins Fresh semen samples, after liquefaction, were processed through 40%/80% (2/2 ml) Percoll gradients at 500 x g for 18 min in order to eliminate seminal fluid. The pellets of purified sperm were further washed with 10 ml HTF (without albumin) by centrifugation at 5000 x g for 20 min. The sperm pellet was resuspended and briefly extracted in 0.2% SDS (1 million sperm/µl), the extract was centrifuged at 5000 x g for 10 min, the supernatant was adjusted to 1% in SDS, and placed in boiling water for 2 min.
One-dimensional PAGE-SDS electrophoresis The PAGE-SDS gel electrophoresis procedures were carried out under standard conditions using 10% or 7–10% SDS-PAGE gradient gels according to the procedure of Laemmli [21]. Proteins from the gels were electrophoretically transferred to nitrocellulose, and after blocking with 1% albumin, were treated with CK-M/HspA2 or NIEHS antisera, or with the antisera cross-absorbed with Accel-purified CK-M/HspA2 (1:1000 dilution) for 6 h. The immunoblots were developed according to the ABC method (using the avidin-biotin-horseradish peroxidase system [3]). Omission of the first antibody caused the loss of staining.
Two-Dimensional Gels
Two-dimensional electrophoresis was performed according to the method of O'Farrell [22] as follows: 1) Isoelectric focusing was carried out in a glass tube with an inner diameter of 2 mm using 2% pH 4–8 ampholines (BDH, Gallard-Schlesinger, Long Island, NY) for 9600 volt-h. The pH gradient for the ampholines was measured by a surface pH electrode. 2) For PAGE-SDS electrophoresis in the second dimension, the tube gel was incubated in equilibration buffer (10% glycerol, 50 mM dithiothreitol, 2.3% SDS, and 0.0625 M Tris, pH 6.8) for 10 min, sealed to the top of a 10% acrylamide slab gel (0.75 mm thick) with stacking gel, and the electrophoresis was then carried out at 12.5 mA/gel. The slab gel was fixed in a solution of 10% acetic acid/50% methanol overnight and was silver-stained [23]. The gel was dried between sheets of cellophane with the acid edge to the left. Immunoblots were developed as described earlier for the one-dimensional gels.
RESULTS
The isolation of sperm HspA2 presented a problem because of the uncertain identity of the enzyme and because its concentration was only 20–30 µg/108 sperm. In order to unambiguously identify and characterize the protein, the following strategy was developed (Fig. 3): 1) We devised long-term preservation conditions for the sperm extracts. This was necessary because the 6–8 x 109 sperm needed as starting material for each isolation represents the unused portion of 100–120 semen analysis samples. 2) We developed the separation of CK-B and CK-M/HspA2 activities by Accell anion-exchange chromatography, which allowed us to monitor CK-M/HspA2 by its activity and not by its very low optical density. Typically, an extract of 6 x 109 sperm was passed through the Accell ion-exchange column. In the CK-M/HspA2-containing peak, the recoveries of protein and activity were about 9% and 48%, respectively, with a 20-fold enhancement of the CK-M/HspA2 activity/OD280. Three or four of the 0.5-ml fractions, representing the peak of activity, were rechromatographed on a smaller Accell column in order to gain a better resolution. 3) We selected the column fractions with the highest activity/protein ratio, containing, typically, 300–500 µg protein. PAGE-SDS gel of the pooled fractions showed one major band, with no other band detectable without overloading the gels. 4) We combined the pooled fractions with adjuvant and injected rabbits, generating a polyclonal antiserum. 5) Western blots of sperm extracts separated by 7–10% PAGE-SDS gradient gels were developed with the anti-CK-M/HspA2 antiserum in order to identify the protein band with the CK-M/HspA2 activity. The CK-M/HspA2, NIEHS and StressGen HSP70 antibodies stained a single band at about 70 kDa in the SDS sperm extract (Fig. 4 demonstrates the data developed with the CK-M/HspA2 and NIEHS antisera.)
|
The primary structure of HspA2 was determined in four independent sperm extracts. The CK-M/HspA2 protein bands, identified by immunostaining neighboring lanes, were excised from the gels, digested using trypsin, and analyzed using reflectron matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) on a Mirromass Tof-Spec SE instrument (Micromass, Beverly, MA) [24]. Monoisotopic masses observed on this instrument have an accuracy better than ± 0.015%, which enables accurate database searches to be performed. The two programs used for searching were ProFund [25] and PeptideSearch [26]. Both programs identified the top significant match as human HspA2 with greater than 50% coverage of the known 70-kDa sequence [27, 28]. To further confirm the identity of the protein, peptides from the digests were isolated by reverse phase HPLC using a Vydac microbore C18 column (Separations Group, Hesperia, CA). Amino acid sequence was directly determined for 5 peptides (73 residues, or 11% of the molecule) and aligned with the primary structure of HspA2 ([24], residues 38–50, 114–127, 140–156, 161–172, and 329–345). Thus, the putative sperm CK-M was identified as the HspA2 chaperone protein (Fig. 5).
In addition to the data in Figure 4, for further independent confirmation of the identity of the 70-kDa protein, we carried out three experiments. 1) We absorbed the CK-M/HspA2, NIEHS, and StressGen HSP70 antisera with the CK-M/HspA2 protein fraction; the absorbed antibodies did not stain the Western blots of sperm extracts. Thus, all three antisera specifically recognized the HspA2. 2) We synthesized three peptides representing residues 38–50, 114–127, and 140–156 (Fig. 4, see the green segments of the HSP70-2), which arise from various distances from the N-terminal of Hsp70-2. After immunostaining of the three peptides, all three antisera in an identical pattern reacted strongly with the peptide of residues 38–50, stained lightly peptide 140–156, while none of the antibodies stained the 114–127 peptide. 3) The specificity of our HspA2 antiserum was also demonstrated on two-dimensional gels of sperm extracts (Fig. 6). The immunoreactive proteins are localized exclusively at 70 kDa, providing evidence for the specificity of our HspA2 antiserum. The staining pattern of multiple 70-kDa spots between pI 5 and 6 could be consistent with the presence of isoelectric variants, alternative splicing, or post-translational modification of HspA2. Another likely explanation is the cross-reactivity of other homologous HSP70s with the CK-M/HspA2 antisera. Very similar patterns of electrophoretic mobility on two-dimensional SDS PAGE and of multiple immunostaining patterns with anti-HSP70 antibodies have been reported for Hsp70-2 in mouse testis [29].
We also examined the expression pattern of HspA2 in human testicular tissue. Our HspA2 and the mouse HSP70-2 antisera showed identical staining patterns (Figs. 7 and 8). Varying levels of immunostaining by both HspA2 and HSP70-2 antisera was evident in spermatocytes and spermatids, but the staining was particularly striking in the cytoplasm of elongating spermatids and mature sperm that were about to be released from the adluminal compartment. This pattern is consistent with the biochemical data that indicated that HspA2 is developmentally regulated during spermatogenesis and expressed simultaneously with cytoplasmic extrusion in late spermiogenesis [6, 13].
The staining of viable sperm with CK-M/HspA2 antiserum indicates that HspA2 is expressed in the entire tail from the middle piece to the end piece of human sperm. In a typical experiment, 0.5 ml normospermic semen was subjected to 40%/80% Percoll centrifugation at 500 x g for 18 min at room temperature. The pellet was resuspended in HTF-albumin at a concentration of 13 x 106 sperm/ml. Sperm motility was 68%. The sperm suspension was divided, and anti-HspA2 antiserum or preimmune serum was added in a 1:1000 dilution. The suspensions were then incubated at room temperature for 3 h. Subsequently, 0.2-ml aliquots of the antiserum and control suspensions were washed twice in 5 ml of HTF-albumin at 500 x g for 10 min at room temperature in order to remove the unbound antibodies. Sperm motility was determined before and after centrifugation (antiserum fraction: 59% and 54%; preimmune fraction: 61% and 58%). We also determined the percent of viable sperm using the FertiLight stain (Molecular Probes, Eugene, OR). The percent of viable sperm in the antiserum and control fractions were 72% and 69% (the stain also identifies nonmotile but viable sperm). In the experiments with preimmune serum there was no immunostaining detectable (Fig. 9). These experiments suggest that HspA2 is located on the plasma membrane of the tail.
DISCUSSION
Our data demonstrate that the presumptive germ cell-specific CK-M isoform, which is a marker of sperm maturity and function, is in fact the HspA2 chaperone protein. The previous identification of HspA2 as an unusual CK-M variant was due to several coincidences. First, our assay system for CK activity detects ATP synthesis, thus the ATP that had bound the HspA2 chaperone was perceived as new ATP produced; and second, the electrophoretic properties of HspA2 are similar to those of muscle CK-M on the native agarose gels we use for the CK isoform analysis [6, 10]. However, based on our early work ([6] and Fig. 1), we have always been aware that we were not dealing with a conventional CK isoform.
Following isolation of the protein and development of a polyclonal antiserum, we unambiguously identified the antigen as the 70-kDa chaperone, HspA2. Identity was established by Western blots, by excising the respective protein band, and determining the primary structure of the protein by MALDI-MS analysis, and by direct amino acid sequencing of five HPLC-isolated peptides at different distances from the N-terminal. Experiments with our own HspA2 antisera as well as the two Hsp70 antisera from other sources indicated that immunoreactivity is localized to a 70-kDa band. The specificity of our HspA2 antisera was confirmed on two-dimensional Western blots of human sperm protein extracts. It is of interest that there were several 70-kDa spots within the pI 5–6 range. These likely represent other Hsp70 chaperones constitutively expressed in spermatogenic cells (e.g., Hsc70; [30]). At least five different Hsp70s are constitutively expressed in testis [29] and the high degree of homology between different Hsp70s makes slight cross-reactivity of antiserum very likely [29, 30]. However, in human spermatogenic cells and sperm, the predominant Hsp70 is HspA2, as has been reported in mouse [31]. We plan to study the relative proportions of other Hsp70s in sperm of fertile and diminished-fertility men, and relate these results to the quantity of HspA2 and to the type of male infertility.
A further confirmation of the identity of CK-M and HspA2 is based on comparative immunological studies using multiple antisera. Three synthetic HspA2 peptides arising various distances from the N-terminal of HspA2 (Fig. 5) showed an identical pattern of staining or lack of staining with the HspA2, NIEHS, and StressGen antisera. Similar patterns of testicular expression of HSPA2 were also detected by our HspA2 and the NIEHS mouse HSP70-2 antisera. The NIEHS mouse Hsp70-2 antisera [19] has been used extensively to characterize the expression and function of HSP70-2 in mouse [18, 29–31] and rat [32] spermatogenic cells and sperm. The genetic homologue of mouse Hsp70-2, the human hspA2 gene, has been fully sequenced and its expression characterized [27, 28, 33]. However, until this present study, we know of no reports that have immunolocalized HspA2 protein in human testis and sperm and correlated its expression to sperm function. Results from these studies are consistent with earlier biochemical data that demonstrated an inverse correlation (r = -0.69 to -0.76) between cytoplasmic extrusion (as evidenced by a decline of sperm CK activity) and increase in HspA2 concentrations ([6–8], Fig. 2), as well as between the expression of HspA2 and plasma membrane ß-1,4-galactosyltransferase (r = -80 [13]).
The Hsp70 family of chaperones facilitate the assembly, folding, and transmembrane transport of protein complexes by an ATP hydrolysis-mediated process [34]. Of the five different Hsp70s expressed during spermatogenesis, the HspA2 chaperone is prominent during both meiosis and spermiogenesis in germ cells [18, 30, 31], suggesting a specific function during spermatogenic differentiation. Indeed, the expression of HspA2 is simultaneous with major sperm protein translocations underlying cytoplasmic extrusion and remodeling of the human sperm plasma membrane. This in turn facilitates the development of the zona pellucida-binding site [12, 13]. The retention of cytoplasm and the lack of zona-binding sites in immature sperm are likely related to the diminished expression of HspA2 in sperm with arrested maturation. This would explain our finding, in blinded clinical studies, that sperm that lack HspA2 show diminished fertilizing potential in conventional conception or in IVF, both of which depend on sperm-zona pellucida interaction [8, 10].
In mouse testis, Hsp70-2 protein is expressed in pachytene spermatocytes during the meiotic phase of spermatogenesis and in spermatids, and is present in mature sperm [35–37]. The apparent functions of Hsp70-2 in mice are maintaining the synaptonemal complexes and assisting chromosome crossing-over during meiosis and spermatocyte differentiation [31]. Accordingly, in mice with targeted disruption of the hsp70-2 gene, arrested spermatocyte differentiation and azoospermia occur [18, 31]. Another consequence of knocking out the hsp70-2 gene in mice was increased spermatocyte apoptosis [38]. This increase could be related to faulty meiotic recombination in spermatocytes [31], disruption of the meiotic cell cycle regulatory machinery [39], or perhaps a more direct perturbation of apoptosis regulation in spermatocytes. In somatic cells, inducible Hsp70s protects cells from stress-induced apoptosis [40] and interact with antiapoptotic proteins such as BAG-1 [41] and signal tranducing stress kinases [42]. It is possible that HSP70-2/HspA2 plays a similar role in modulating apoptosis in spermatocytes or even in spermatids or ejaculated immature sperm [43].
HSP70 chaperone proteins in the mouse are also synthesized postmeiotically in round spermatids and may be found in the residual bodies. Immunostaining placed mouse HSP70s in the mouse sperm middle piece but, unlike human HspA2, these Hsp70s were absent from the principal and end pieces of the tail [44]. Because of the known association between sperm immaturity and increased rate of lipid peroxidation and fragmentation of DNA in men [14, 15], we hypothesize that both lipid peroxidation and DNA fragmentation are related to apoptosis in immature sperm. We will test this by studying apoptosis in mature and diminished-maturity human sperm. A related line of research concerns the apparent differences in localization of HspA2 in human sperm versus localization of Hsp70s in mouse sperm. We intend to further examine the localization and function of HspA2 and the other Hsp70s (Fig. 6) in human sperm.
We have established that there are immature human spermatozoa that have little or no HspA2 chaperone expression. A question that arises concerns the mechanism by which these immature germ cells proceed with spermiogenic differentiation, including formation of the flagella, acrosome, and other morphological characteristics. These aspects have not yet been investigated in men; however, there is evidence in hsp70-2 null mice that spermatocytes do not complete the meiotic process [18] and that spermatid differentiation can proceed to formation of the acrosome [45]. Similarly, in Drosophila mutants, the formation of the nucleus, acrosome, and flagellum can also occur without completion of the meiotic divisions [46]. In salamanders, secondary spermatocytes that were cultured in the presence of cycloheximide, which inhibits the meiotic process, nevertheless underwent flagella formation [47]. These examples from other species suggest that spermatogenesis in men may also allow the formation of compromised, immature sperm in the absence of adequate HspA2.
In summary, we identified the expression of a 70-kDa chaperone protein, HspA2, in human testis and sperm; more specifically, throughout the tail and on the surface of the tail plasma membrane. The presence of HspA2 correlates well with sperm cellular maturity, function, and fertility. HspA2 is expressed during the terminal phase of spermiogenesis, simultaneous with cytoplasmic extrusion and remodeling of the sperm plasma membrane. We have previously shown that the remodeling of the plasma membrane, coincident with HspA2 expression, facilitates the formation of zona pellucida binding sites in human sperm [12, 13]. Because cytoplasmic extrusion and plasma membrane remodeling involve major protein transport between cellular compartments, it is possible that this process is facilitated by the HspA2 chaperone. Accordingly, in sperm with arrested maturation, which do not express HspA2, there is cytoplasmic retention and a failure of binding to the zona pellucida. In extensive clinical studies we have demonstrated that low levels of HspA2 in human sperm predicts diminished fertility by natural conception or IVF, outcomes that depend on sperm-zona pellucida interaction.
ACKNOWLEDGMENTS
We thank Dr. E.M. Eddy for the generous gift of the anti-Hsp70-2 antiserum and for valuable discussions. We are grateful to Dr. David De Kretser for the gift of human testicular sections, and to Dr. Harvey Kliman for his assistance in establishing the immunocytochemistry of paraffin-embedded sections in our laboratory.
FOOTNOTES
1 Supported by NIH grant HD-32902, and by the CONRAD program. Part of this research was presented at the 1999 annual meeting of the European Society of Human Reproduction, Tours, France. 
2 Correspondence: FAX: 203 737 1200; gabor.huszar{at}yale.edu
Accepted: April 20, 2000.
Received: February 14, 2000.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  T. F. Wu and D. S. Chu Sperm Chromatin: Fertile Grounds for Proteomic Discovery of Clinical Tools Mol. Cell. Proteomics, October 1, 2008; 7(10): 1876 - 1886. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.P. Cedenho, S.B. Lima, M.A. Cenedeze, D.M. Spaine, V. Ortiz, and S. Oehninger Oligozoospermia and heat-shock protein expression in ejaculated spermatozoa Hum. Reprod., July 1, 2006; 21(7): 1791 - 1794. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W.C.L. Ford Glycolysis and sperm motility: does a spoonful of sugar help the flagellum go round? Hum. Reprod. Update, May 1, 2006; 12(3): 269 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Celik-Ozenci, A. Jakab, T. Kovacs, J. Catalanotti, R. Demir, P. Bray-Ward, D. Ward, and G. Huszar Sperm selection for ICSI: shape properties do notpredict the absence or presence of numericalchromosomal aberrations Hum. Reprod., September 1, 2004; 19(9): 2052 - 2059. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Huszar, C. Celik-Ozenci, S. Cayli, T. Kovacs, L. Vigue, and E. Kovanci Semen Characteristics After Overnight Shipping: Preservation of Sperm Concentrations, HspA2 Ratios, CK Activity, Cytoplasmic Retention, Chromatin Maturity, DNA Integrity, and Sperm Shape J Androl, July 1, 2004; 25(4): 593 - 604. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Kuster, R. A. Hess, and G. C. Althouse Immunofluorescence Reveals Ubiquitination of Retained Distal Cytoplasmic Droplets on Ejaculated Porcine Spermatozoa J Androl, May 1, 2004; 25(3): 340 - 347. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Cayli, D. Sakkas, L. Vigue, R. Demir, and G. Huszar Cellular maturity and apoptosis in human sperm: creatine kinase, caspase-3 and Bcl-XL levels in mature and diminished maturity sperm Mol. Hum. Reprod., May 1, 2004; 10(5): 365 - 372. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Celik-Ozenci, J. Catalanotti, A. Jakab, C. Aksu, D. Ward, P. Bray-Ward, R. Demir, and G. Huszar Human Sperm Maintain Their Shape Following Decondensation and Denaturation for Fluorescent In Situ Hybridization: Shape Analysis and Objective Morphometry Biol Reprod, October 1, 2003; 69(4): 1347 - 1355. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Jakab, T. Kovacs, Z. Zavaczki, A. Borsos, P. Bray-Ward, D. Ward, and G. Huszar Efficacy of the swim-up method in eliminating sperm with diminished maturity and aneuploidy Hum. Reprod., July 1, 2003; 18(7): 1481 - 1488. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Sakkas, O. Moffatt, G. C. Manicardi, E. Mariethoz, N. Tarozzi, and D. Bizzaro Nature of DNA Damage in Ejaculated Human Spermatozoa and the Possible Involvement of Apoptosis Biol Reprod, April 1, 2002; 66(4): 1061 - 1067. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Kovanci, T. Kovacs, E. Moretti, L. Vigue, P. Bray-Ward, D. C. Ward, and G. Huszar FISH assessment of aneuploidy frequencies in mature and immature human spermatozoa classified by the absence or presence of cytoplasmic retention Hum. Reprod., June 1, 2001; 16(6): 1209 - 1217. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
