Human oviductal fluid (hof) proteins. IV. Evidence for hof proteins binding to human sperm*

Transcription

1 FERTILITY AND STERILITY Copyright e 1989 The American Fertility Society Printed in U.S.A. Human oviductal fluid (hof) proteins. IV. Evidence for hof proteins binding to human sperm* Jack Lippes, M.D.t Premanand V. Wagh, Ph.D. Department of Gynecology and Obstetrics, School of Medicine, and Department of Biological Sciences, State University of New York at Buffalo, Buffalo, New York In the management of infertile patients, it is well recognized that gamete intrafallopian transfer (GIFT) has a higher success rate than in vitro fertilization and embryo transfer. Oviducts produce unique proteins that may be responsible for the success of GIFT. Unique proteins from human oviductal fluid (hof) have been purified in the authors' laboratory. One of these proteins of 54 kda molecular weight, and containing carbohydrate, was obtained in a highly purified state. In this study, human sperm were incubated with: (1) a mixture ofhof-specific proteins, and (2) the highly purified hof glycoprotein, designated human oviductin I (hov-1). Using indirect immunofluorescence, the authors studied the ability ofhof proteins to bind to human sperm. The mixture ofhof proteins appeared to bind to the surface of the entire sperm. Conversely, hov -1 binding was restricted to the head region only. Studies are in progress to discern the function of the sperm-binding interaction(s) with hof proteins. Fertil Steril51:89, 1989 The management of infertile patients currently includes the sophisticated procedures of in vitro fertilization and embryo transfer (IVF-ET), as well as gamete intrafallopian transfer (GIFT). It is recognized that GIFT has a higher success rate than IVF-ET. Therefore, it appears that human oviductal fluid (hof) possesses dynamic and special characteristics. There may be chemical entities that foster gamete maturation, fertilization, early embryogenesis, and perhaps, nidation. Information about hof might be useful to promote or control fertility. Accordingly, the oviduct and oviductal fluid has been the object of numerous studies, which are well documented in various reviews. 1-5 Previous studies have shown that mammalian Received January 4, 1988; revised and accepted September 29,1988. * Presented in part at the Fifth Annual Meeting of the Society for Advancement of Contraception, October 5 to 8, 1987, Caracas, Venezuela. t Reprint requests: Jack Lippes, M.D., Department of Gynecology and Obstetrics, Erie County Medical Center, 462 Grider Street, Buffalo, New York oviducts produce unique proteins A search of the literature reveals that no role has been determined for any protein specific to hof. Voglmayr and Sawyer/ 3 however, using radiolabeled oviductal fluid, have shown that oviductal proteins from the ewe bind to ram sperm. The identification and partial purification of unique hof proteins has been described. 14 A 54 kda non hof glycoprotein was obtained in a highly purified state. This study was undertaken to discern whether any hof protein(s) bind to human sperm. Collection of hof MATERIALS AND METHODS Human oviductal fluid was collected by a technique previously described with slight modifications and approved annually by the Investigational Review Board of the Erie County Medical Center. 15 Multiparous patients electing to be sterilized volunteered as study subjects. A Pomeroy type tubal ligation was.performed on the isthmic portion of Lippes and Wagh hof proteins and sperm binding 89

2 the oviduct close to the uterus through a minilaparotomy incision. Foley catheters (Pediatric #8 Fr., C. R. Bard, Murray Hill, NJ) were inserted through the tubal ostia. A 3/0 chromic catgut suture was lightly sewn circumferentially around the fimbriated end of the oviduct. Inflation of the Foley catheter balloon maintained the catheter in the oviductal lumen in front of the catgut suture until the catheters were removed. Catheters were connected to transfusion transfer bags. Human oviductal fluid was collected under refrigeration by placing the bags on ice at the patient's bedside. Fluid was removed and bags were changed at 8- to 12-hour intervals. Collection continued for 24 to 36 hours because it has been shown that, at about 48 hours, leukocytes can be found in hof, indicating an oviductal inflammatory response. 15 Usually, catheters were removed at 24 to 36 hours after surgery. Patients were discharged shortly thereafter. Human oviductal fluid was centrifuged 5000 X g for 10 minutes to eliminate debris and cells, and sterilized by filtration through a millipore filter into sterile tubes. Phenylmethylsulfonyl fluoride (PMSF) was added to prevent proteolysis. Samples were stored frozen at -70oC until used. Phenylmethylsulfonyl was removed by dialysis before sperm binding studies were begun. Preparation of hof Protein(s) Antibodies Purification procedures of hof proteins are described in detail in this issue. 14 Briefly, four samples of hof were pooled. An immunoaffinity column consisting of rabbit gamma G immunoglobulin (IgG) antihuman -Sepharose 4B (Sigma, St. Louis, MO) was used to remove proteins from hof. Further purification of this mixture by Diethylaminoethyl (DEAE) cellulose chromatography and G-75 Sephadex gel filtration provided a highly purified protein, designated "human oviductin" (hov-i). Two New Zealand white female rabbits (2.7 kg) were immunized with non hof proteins emulsified in Freund's complete adjuvant. Theresulting antigen mixture represented about 1 mg protein, and was injected intradermally at multiple sites in the animal's flanks at 0, 12, 22, 41, and 67 days. Similarly, two rabbits were immunized with hov-i. Trial bleedings were taken at 0, 7, 19, 29, 48, 74, and 82 days on each pair of rabbits. E_valuation of Antisera Rabbit antisera were evaluated by immunodiffusion. This was carried out in 50 mm dishes 90 Lippes and Wagh hof proteins and sperm binding Figure 1 Ouchterlony double diffusion pattern: The outermost well at the 12 o'clock position contained buffer as a marker. Center well: Rabbit lgg antinon hof proteins. Outer wells: (1) non hof proteins; (2) human liver extract; (3) human kidney extract; (4) human ; (5) human oviduct extract; (6) human ovary extract. containing 3 ml 1% agarose, made up in 0.15 M NaCl. Wells were 3 mm in diameter and 3 mm apart. The patterns were allowed to develop overnight at room temperature. Preimmunization bleeds of normal rabbit sera were evaluated for reactivity with: (1) the non hof proteins and (2) hov-i. No reactions were observed. After three injections, precipitation lines were seen against the corresponding antigens, i.e., non hof proteins and hov-i. They were more pronounced with the 48-day and 82-day bleedings. Rabbit antisera were purified by treatment with DEAE-Affigel blue (BioRad, Richmond, CA). These two protease-free immunoglobulins reacted against the respective parent antigens and extracts of human oviductal tissue prepared in borate buffered saline, ph 8.0, (30% wt/vol). They did not react against extracts of human uterus, liver, kidney, ovary, or human. (Fig. 1) Immunoelectrophoresis (IEP) Immunoelectrophoresis (IEP) was performed as described by van Oss and Bartholomew.16 In Figure 2A, rabbit anti-hof proteins anti reacted against the parent compound, i.e., hof proteins wherein the components were removed by an immunoaffinity gel. 14 Two precipitin lines of non hof proteins that migrated to the albumin and postalbumin regions were seen. There was no reactiob against human. Fertility and Sterility

3 A A B B Figure 2 Immunoelectrophoretic analysis of antisera against non hof proteins and hov -I with schematic diagrams of the IEP gels. (A) Upper well, human ; lower well, non hof protein; trough, anti-hof non proteins. Anode is at the right. (B) Upper well, human ; lower well, hov I; trough, anti-hov-i. Anode is at the right. Furthermore, in Figure 2B, anti-hov-i reacted against the highly purified hov-i with a single precipitin band. Again, there was no reaction against human. Gamma G immunoglobulin purified from rabbit antihuman did not react against hov-i or against non hof proteins, but did react against human (not shown). Other Antibodies Rabbit IgG antihuman and rabbit IgG antihuman albumin were prepared in our laboratory. Goat IgG antirabbit, Fluorescein isothiocyanate (FITC) goat IgG antirabbit, and FITC rabbit IgG antigoat were purchased (Sigma Chemical Company, St. Louis, MO). Treatment of Sperm Semen was obtained from donors of known fertility by self masturbation after 4 days of abstinence. Spermiograms were within normal ranges, i.e., (1) 4 to 4 ml semen volume; (2) at least 60% normal forms; (3) >70 million sperm/ml; (4) total motile sperm count >80 million. Seminal plasma was removed by centrifugation at 500 X g for 5 minutes and washed in a medium of110 mm NaCl, 5 mm KCl, and 10 mm morpholinopropane sulfonic acid (NKM) at 37oc_n Centrifugation and resuspension were repeated three times, giving a concentration of 5 X 10 8 sperm/mi. Sperm Binding Table 1 Tube no Incubation Media Used for Sperm Binding Studies Component Non HOF proteins HOV-I HOV-I Preovulatory hof Periovulatory hof Human Human albumin Rabbit MOPS buffer Sperm binding experiments consisted of two parts. In part I, the non hof proteins, i.e., the unabsorbed fraction from the immunoaffinity gel, were incubated with the sperm suspensions. In part II, the highly purified 54 kda hof protein designated human oviductin-i (hov-i), was used for incubation with similar sperm suspensions. Each sperm suspension was examined for protein binding by indirect immunofluorescence (IIF). Incubation tubes contained 100 ~l of sperm suspension (5 X 10 7 ), which were mixed with various test materials (50 ~l) in a final volume of 150 ~l (Table 1). Protein concentrations in these two sperm binding experiments were chosen to approximate the levels of the respective components in the starting pooled hof sample (Table 1). The ph of the incubation media was 7.4, and the osmolality was 280 mosm. Sperm with test materials were incubated for 3 hours at 37oC in a humidified chamber exposed to 5% C0 2 in air. For IIF, the technique of Wolf et al. was followed with slight modifications. 18 At the end of 3 hours, sperm were washed three times in phosphatebuffered saline (PBS, ph 7.4) by centrifugation and resuspension. They were fixed in ph 7.4 PBS containing 1% paraformaldehyde and 0.2 M gly- Final concen- Component tration volume MOPS* ofprotein (~til (~til (mgfml) *MOPS, 110 rom NaCl, 5 rom KCl, 10 rom morpholinopropane sulfonic acid. Lippes and Wagh hof proteins and sperm binding 91

4 Table2 Slide no. Antibodies Used for Sperm Binding Experiments Incubation component First antibody Second antibody Fluorescence (present or absent) 1. Non hof Rabbit lgg antinon FITC-goat lgg proteins hofprotein anti-rabbit + 2. Non hof Rabbit lgg antihuman Same proteins 3. Preovulatory hof Rabbit lgg antinon Same + hof proteins 4. Preovulatory hof Rabbit lgg antihuman Same 5. Periovulatory hof Rabbit lgg antinon Same + hof proteins 6. Periovulatory hof Rabbit lgg antihuman Same 7. Human Rabbit lgg antinon Same hofprotein 8. Human Rabbit lgg antihuman Same 9. Human abumin Rabbit lgg antinon Same hof proteins 10. Human Rabbit lgg antihuman Same albumin albumin 11. Rabbit Goat lgg antirabbit FITC-rabbit lgg antigoat 12. hov-1 Rabbit lgg anti-hov-1 FITC goat lgg antirabbit hov-1 Rabbit lgg antihuman FITC goat lgg antirabbit 14. NKM Rabbit lgg Same antinon hofprotein cine. 19 Circles, 2 em in diameter, were etched on acetone-cleaned glass slides with a diamond stylet. Sperm fields were prepared by smearing 20 ~l of the cell suspensions within these circles and were airdried. Slides were washed three times with PBS containing 0.01% NP-40 (nonylphenol polyethyleneglycol ether) and 0.1% 2-mercaptoethanol. In part II of the binding experiments, sperm were incubated with hov-1 (tubes 2 and 3, Table 1). Appropriate controls were included in these experiments (tubes 6 to 9, Table 1; and slides 2, 4, 6, 8-11, 13, 14, Table 2). Each sperm field was covered with 20 ~l of the first antibody. Slides were placed in a humidified chamber at room temperature (25 C) for 30 minutes. They were washed two times with phosphatebuffered saline (PBS) and then second antibodies (20 ~l) were applied. Following incubation for 30 minutes at room temperature in a humidified chamber, the slides were washed three times with P:J3S. Details of the primary and secondary antibodies used for indirect immunofluorescence are shown in Table 2. Cover slips were mounted using 50% glycerol containing 0.2 M n-propylgallate and 0.1% p-phenylenediamine to prevent quenching. Cells were examined for fluorescence and by phase contrast using a Zeiss Photomicroscope (Carl Zeiss, Oberkochen, W. Germany) with a 100 W mercury light source. Photomicrographs of representative fields using Ilford XPI film (Ilford, Mobberley, Cheshire, U.K.) were taken at 1600 ASA and push developed to 2000 ASA. In each instance, duplicate pictures of the identical microscopic fields were photographed with phase contrast and epifluorescence, respectively. RESULTS In binding experiment, part I, when sperm were incubated in either non hof proteins or periovulatory hof or preovulatory hof, fluorescence of the intact sperm, including head, neck-piece, and tail, was clearly seen (slides 1, 3, 5; Table 2). In addition, even the sperm fragments fluoresced (Figs. 3A and 3B). Spermatozoa on control slides where the f}.rst antibody was either antihuman se- 92 Lippes and Wagh hof proteins and sperm binding Fertility and Sterility

5 Figure 3 Indirect immunofluorescence of human sperm with a mixture of non hof proteins. (A) phase contrast; (B) epifluorescence. Because controls were blank, they have not been reproduced. rum or antihuman albumin raised in rabbits or antirabbit raised in goats, did not fluoresce (slides 2, 4, 6, 8, 10, 11; Table 2). This indicated that human proteins did not bind to human sperm. The observations of sperm binding in part II contrasted sharply from those in Part I. When sperm were incubated with hov-i, only the heads fluoresced (Figs. 4A and 4B). However, such fluorescence was not seen with all sperm, indicating that hov-i probably binds in a selective manner. Sperm placed in periovulatory or preovulatory hof displayed a marked increase in motility. 20 Incubation of sperm in media containing periovulatory or preovulatory hof revealed similar fluorescent characteristics as those treated with non hof proteins. Control slides were blank. Moreover, cells incubated in hof, or hof non proteins or hov-i, did not fluoresce when the first antibody was rabbit IgG antihuman and the second antibody was FITC goat IgG antirabbit (slides 2, 4, 6, and 13; Table 2). These results strongly indicated that only hof non proteins bind to the surface of human sperm. DISCUSSION.Voglmayr and Sawyer, using radio iodinated pro- teins from ewe oviductal fluid, reported that the surface of ram sperm selectively adsorbed proteins with molecular weights of 140, 95, 78, 66, and 53 kda. 13 Of considerable interest in their study was the fact that one of the major proteins that adhered to the surface of the ram sperm had a molecular weight of 66 kda. They believed this was albumin. Presumably, this family of oviductal fluid proteins sharing similar molecular sizes, but from different mammals, bind to the surface of their respective sperm. This evidence suggests a possible functional homology and a probable structural similarity between the oviductal fluid proteins of women and sheep. Whether the observed process of oviductal fluid proteins binding to sperm is species-specific is not known. However, the homology and similarity may be close enough to present us with an animal model with which to study this phenomenon. Sutton et al. 11 also found that a major component of sheep oviductal fluid was a 66 kda protein. In 1972, this laboratory reported that hof displayed augmentations in the albumin region. 15 Electrophoretic patterns on some hof samples manifested a shoulder next to the albumin peak that was not seen in the patients' sera Binding of non hof proteins and hov-i to human sperm raises new and interesting possibilities. The highly purified hov-i selectively bound to the head region of human sperm. This macromolecule had a molecular weight of 66 kda under reducing conditions and 54 kda under nonreducing conditions, behavior Figure 4 Indirect immunofluorescence of human sperm with hov-1. (A) phase contrast; (B) epifluorescence. Because controls were blank, they have not been reproduced. Lippes and Wagh hof proteins and sperm binding 93

6 similar to albumin. However, hov-1 was found to be a glycoprotein, and therefore could not be albumin. 14 An understanding of the molecular mechanism of hov-1 and sperm interaction should be worthwhile. For example, hov-1 that is found in high concentration in peri ovulatory hof 20 may act as an acrosome stabilizing factor, preventing a premature acrosome reaction. The existence of such a mechanism seems biologically and physiologically reasonable. Our observation that hov-i bound to some sperm and not to others may be a clue to explore functions. It is not known what molecular and/or physiologic events, e.g., capacitation or acrosome reaction, differentiate these sperm populations. When the incubation media contained a mixture of non hof proteins, the entire surface of all sperm or fragments of sperm fluoresced. An hof protein(s) completely covering sperm might prevent these cells from stimulating antisperm antibody production. Oliphant and co-workers have isolated a macromolecular compound from rabbit oviductal fluid that inhibits complement activity within the rabbit oviduct. 22 The complement inhibitor arises from a group of sulfated glycoproteins elaborated by the rabbit oviductal epithelial secretory cells. Human oviductal fluid proteins similarly may be immunologically protective. Further studies on the functions of hof proteins are in progress. Such investigations may provide an understanding of molecular events concerning gamete-hof interactions, which may be used to promote and/or control fertility. Acknowledgments. We thank Morton Rothstein, Ph.D., and Darrell Doyle, Ph.D., for providing laboratory facilities and for their helpful suggestions with this work. The help of Carel J. van Oss, Ph.D., and Paul Bronson with the immunoelectrophoretic analyses is acknowledged with gratitude. REFERENCES 1. Sherman AI: Pathways to Conception. Springfield, Charles C. Thomas, Johnson AD, Foley CW: The Oviduct and Its Functions. New York, London, Academic Press, Lippes J: Applied physiology of the uterine tube. Obstet Gynecol Ann, 4:119, Jansen RPS: Endocrine response in the fallopian tube. Endoer Rev 5:525, Siegler AM: The Fallopian Tube: Basic Studies and Clinical Contributions. Mt. Kisco, New York. Pub-Futura Publishing Company, Mastroianni L Jr, Urzua M, Stambaugh R: Protein patterns in monkey oviductal fluid before and after ovulation. Fertil Steril 21:817, Kay E, Feigelson M: An estrogen modulated protein in rabbit oviductal fluid. Biochim Biophys Acta 271:436, Feigelson M, Kay E: Protein patterns of rabbit oviductal fluid. Bioi Reprod 6:244, Oliplant G, Ross PR: Demonstration of production and isolation of three sulfated glycoproteins from the rabbit oviduct. Bioi Reprod 26:537, Lippes J, VanOss CJ, Bronson PM, Alfonso LA, Dacalos EA, Lucero R: Human oviductal fluid proteins. II. Preparation of an anti to a human oviductal fluid protein: existence of autoantibodies against it in some sera. Fertil Steril 39:824, Sutton R, Nancarrow CD, Wallace ALC, Rigby NW: Identification of an oestrus-associated glycoprotein in oviductal fluid in the sheep. J Reprod Fertil 73:415, Fazleabas AT, Verhage HG: The detection of oviduct-specific proteins in the baboon (Papio anubis). Bioi Reprod 35: 455, Voglmayr JK, Sawyer RF Jr: Surface transformation of ram spermatazoa in uterine, oviduct and cauda epididymal fluids in-vitro: J Reprod Fertil 78:315, Wagh P, Lippes J: Human oviductal fluid proteins. III. Identification and partial purification. Fertil Steril 51:81, Lippes J, Enders RG, Pragay DA, Bartholomew WR: The Collection and analysis of human fallopian tube fluid. Contraception 5:85, van Oss CJ, Bartholomew, WR: Precipitation reactions. In Methods in Immunodiagnosis, Edited by NR Rose, PE Bigazzi. New York, Wiley Interscience, 1980, p San Agustin JT, Hughes P, Lardy HA: Properties and function of caltrin, the calcium-transport inhibitor of bull seminal plasma. FASEB J 1:60, WolfDP, Boldt J, Byrd W, Bechtol KB: Acrosomal status evaluation in human ejaculated sperm with monoclonal antibodies. Bioi Reprod 32:1157, Koehler JD, The mammalian sperm surface: studies with specific labeling techniques. Int Rev Cytol 54:73, Lippes J, Wagh P: Unpublished data 21. Lippes J, Krasner J, Alfonso LA, Dacalos ED, Lucero R: Human oviductal fluid proteins. Fertil Steril36:623, Oliphant G, Cabot C, Ross P, Marta J: Control ofthe humoral immune system within the rabbit oviduct. Bioi Reprod 31:205, Lippes and Wagh hof proteins and sperm binding Fertility and Sterility