HUMAN OVIDUCTAL FLUID PROTEINS*

Transcription

1 FERTILITY AND STERILITY Copyright o 1981 The American Fertility Society Vol. 36, No. 5, November 1981 Printed in U.SA. HUMAN OVIDUCTAL FLUID PROTEINS* JACK LIPPES, M.D.t JOSEPH KRASNER, PH.DJ LYDIA A. ALFONSO, M.D. EMILIA D. DACALOS, M.D. REBECCA LUCERO, M.D. Department of Obstetrics and Gynecology, State University of New York at Buffalo, Immunology Department, Roswell Park Memorial Institute, Buffalo, New York 1423, and Department of Obstetrics and Gynecology, Southwestern University, Cebu City, Philippines The popularity of mini-laparotomy provided an opportunity to easily collect human oviductal fluid (HOF). Volumes of HOF produced by two oviducts per 24 hours correlated positively with serum estradiol determinations, while protein concentration in HOF was inversely proportional to estrogen levels. Estrogen appeared to stimulate the production of oviductal fluid in women. The greatest volumes of HOF were observed at midcycle, coincident with the estrogen peak. Protein concentration was lowest at the time of ovulation and highest immediately before and after menstruation. In a large percentage of patients, certain proteins made their appearance at the time of ovulation and receded or disappeared within three to five days thereafter. By electrophoresis, these proteins were seen in the albumin region and in the beta globulin region. Fertil Steril36:623, 1981 During the normal 28-day menstrual cycle of women, important physiologic events occur at ovulation and for 3 days thereafter. During these 3 days, the oviduct secretes material that promotes capacitation, fertilization, and sustinence of the conceptus. Finally, the oviduct propels the morula downward into the uterus. Previous pa- Received September 22, 198; revised and accepted July 15, *Presented at the Thirty-Sixth Annual Meeting of the American Fertility Society, March 18 to 22, 198, Houston, Texas. treprint requests: Jack Lippes, M.D., Professor, Department of Obstetrics and Gynecology, School of Medicine, State University of New York at Buffalo, Deaconess Hospital, 11 Humboldt Parkway, Buffalo, New York :j:cancer Research Scientist II, Immunology Department, Roswell Park Memorial Institute, Buffalo, New York Department of Obstetrics and Gynecology, Southwestern University, College of Medicine, Cebu City, Philippines. 623 pers have suggested that secretions of the human oviduct are probably under endocrine control, 1 2. perhaps mediated through shifts in prostaglandin from the mucosa before ovulation to the lamina propria after ovulation. 3 This study was undertaken to examine the proteins of human oviductal fluid (HOF) at various times of the menstrual cycle to determine whether such proteins correlate with the dominant, gonadal endocrine environment. If protein electrophoretic patterns of HOF were different than patient's serum protein electrophoresis, this finding would suggest that HOF possibly contained unique proteins. MATERIALS AND METHODS Thirty-one patients who desired elective sterilization by tubal ligation because of multiparity were asked to participate in this study. Guidelines of the National Institutes of Health and the

2 624 LIPPES ET AL. Helsinki Convention for the protection of patient's rights were scrupulously followed. Successful collection of some HOF occurred in 28 of 31 cases, by the use of surgical techniques already described. 1 4 The majority of tubal ligations were performed near days of ovulation. HOF samples were removed every 12 hours from plastic collection bags, attached to Foley catheters, whose tips and inflated balloons were retained within the oviductal lumen. Collection of HOF continued for 48 hours after surgery, when the Foley catheters were deflated and removed. Volumes of fluid were carefully recorded upon withdrawal from the plastic bags. Samples were immediately centrifuged. The sediment was examined microscopically for leukocytes and cultured for bacterial contamination. HOF was forced through a millipore filter,.45 J.m, to be certain of sterility. These samples were stored in sterile tubes at -8 C until several were available for analysis. Two samples, which were bacteriologically contaminated and/or had leukocytes in the sediment were not used in this study. Serum was obtained from patients on the night before surgery and at 48 hours postoperatively. Only preoperative hormonal serum levels were used for correlation with HOF volumes and HOF protein values. Postoperative hormonal levels confirmed correlations and preoperative steroid determinations. Luteinizing hormone (LH), progesterone, and estradiol levels were measured to ascertain the endocrine effect, if any, on HOF volumes and proteins. LH levels helped to decide which HOF determinations were periovulatory. Such LH, progesterone, and estradiol serum levels were determined by radioimmunoassay (RIA), using commercially available kits (LH by BIO RIA of Lexington, Kentucky; progesterone by Nuclear Medical Systems, Inc., of Newport Beach, California and estradiol by Diagnostics, Biochem International, Inc., of Tampa, Florida). Protein concentrations of HOF were examined by the Lowry Method. 5 Protein electrophoresis was accomplished by two techniques: (1) For electrophoretic analysis of HOF samples, a "spectrophor" (Bausch and Lomb) instrument was used. An HOF sample was applied to agarose in a barbitol buffer, ph 8.6, layered in a quartz tray. Electrophoresis was then carried out for 2 minutes at 36 volts. Separation patterns were recorded from a scanning photoelectric cell, which measured the 25-nm absorbance without any further manipulation or staining of the supporting gel. (2) Acrylamide gel electrophoresis staining with Coomas- November 1981 sie blue was carried out on a different set of HOF samples. The samples were dialyzed against 2 liters of distilled water at 4 C for 24 hours and then lyophilized. The proteins were then resuspended in the exact amount of.1 M phosphate buffer, ph 7.5, to make each sample's protein concentration equal to 1 mg/ml, which facilitated making comparisons. Acrylamide gel electrophoresis was performed on the samples using the method of Weber and Osborn, 6 without the addition of sodium dodecyl sulfate. The gels were placed in small tubes and proteins stained with Coomassie blue (R-25). Initially, gels were cleared with a mixture of7.5% acetic acid and 5% methanol solution and were allowed to stand at room temperature for 3 to 4 hours. Further clearing of the gels was concluded by several changes of a 5% acetic acid solution. Correlation coefficients were calculated for serum levels of estradiol and progesterone against HOF daily volume, protein concentrations, and daily protein production. The significance of correlation coefficients was determined by a t-test. Patient Acceptance and Complications. Thirtyone patients who desired sterilization by minilaparotomy tubal ligation agreed to the catheterization of their oviducts. No major complications - ' Total protein 18. I\ 15. I \ 15. I \ 135. I 12.. E I c:: \ "E 15. ;; 8 1 I Human oviductal a; ' 1'5... fluid protein 9. :s. Q) I concentration fti :::J Eo 75. E.c:: I e>oo:l" ' c:: 6 ' 6. ;; "E oc:: 45. = -. ;E4 c: :::J Q)- 3. u 2 u 15. c:: a; 1'5....o a Cycle day FIG. 1. Human oviductal fluid volumes, protein concentration, and total protein by cycle day.

3 Vol. 36, No.5 HUMAN OVIDUCTAL FLUID PROTEINS 625 were observed with any patient. A few minor problems were seen. Of 31 patients, 3 were treat-. " ed for cystitis and 2 developed wound infections. "" tl ", +I.a?,_ t- 1 Except for the 2 women with wound infections, all co "" 1 " patients left the hospital within 72 hours. HOF Volumes. In this series, collection ofhof was successful in 28 of 31 cases. In 1972, 1 we ;::.... <»...::j reported collecting HOF in 28 of 37 attempts. "" <» tl 1:: 'CJ) Data from our 1972 paper were incorporated to e "" determine average daily HOF volume and aver- <» age HOF protein concentration. By combining data from the 1972 publication with the current Q., >... ",...;,...; series, it was possible to plot daily HOF volume, ] tl +I tl J total protein, and protein concentration against CJ).g... CJ),...j.,; the time of the menstrual cycle, correlating these ]S... observations with serum hormone levels, espe- 1:: c t- cially estradiol and progesterone (Tables 1 and 2; fi::l Fig. 1). The largest volumes of HOF were col- i: E tl t- tl ;!;... d lected in preovulatory days 12 and 13, (average 1:: co tlq., 9.48 ml/24 hours), which coincided with the " higher serum estradiol levels (r = +.694; p < CJ) ::,S 'Oj4 1.1). With one exception, daily HOF volumes of tl :::! " -g :!l,...; t- " ' ;::. 3.1 ml or more were obtained when serum es-. e 'S tl +I coj +I" +ltd... <»... :::!: CJ) t- tradiollevels were at least 2 pg/ml. In contrast, """: " " Q.,;.,; t- serum progesterone values were inversely pro- c portional to HOF daily volumes, but these data - 8E 'Oj4 were not statistically significant (r = -.22, P.,; ",...; 1:: t- 1 t- 1:: )1 " tl >.5) (Tables 1 and 2). :I:-!11 1 " Ol. Average daily HOF volumes of3.89 ml on cycle Q..'E-... a) days 14 and 15 were noted, which decreased to an s average of2.18 ml/day for days 16 and 17. Prior to, day 12 and after day 17, volumes of HOF were co "... CJ) " smaller. :::! tl1 1 tl... oq ti<:dc'i tl - Protein Production. On individual samples,... "E t-,...; CJ) " protein concentrations varied from.1 gm/dl :[.g (1972 observation) to 8.6 gm/dl. The greatest con-.g 1!::rl.l 'Oj4 centrations of protein in HOF were observed in tl:l. 1 >. CJ) samples collected premenstrually and postmen- tl tl, 1 strually. The most dilute protein concentrations.,; " t- were seen prior to ovulation and for 3 days there- r:.. tl -Q;: after. Thus, on days 12 and 13, average protein.g concentration was 1.9 gm/dl. These low protein ::!,_ ;::. concentrations coincided with large volumes of +1.-td tl... oq 1:: HOF samples; e.g., in a 24-hour period, we ob-.,; tained from one patient a 24-ml sample of HOF collected from two oviducts. The protein concen-,...; tration was.4 gm/dl. This sample was the larg- rz:l,..;j est ever collected in one day. On days 14 and 15 of 1:: :<:j bb the menstrual cycle, protein concentrations rose as s bb to 3.21 gm/dl and continued to rise during days 17 Em Q) Q) Q).,... Q) - as... -:. through 2 to an average of 3.93 gm/dl. Similar 1:: Q., 1:: Q., s s as 8 e levels were observed in HOF samples obtained during the immediate postmenstrual period of Q) 1:: ' Q) " ]l, ".s:: 1:: days 7 and 8. The highest protein concentrations z&! J.lz&!... Q) Obi)

4 "---.._.;; 626 LIPPES ET AL. TABLE 2. Correlation Coefficient between Serum Estradiol, Progesterone, and Human Oviductal Fluid Data Estradiol concentrations HOF daily volumes HOF daily protein concentrations HOF daily total protein production Progesterone correlations HOF daily volumes -.22 HOF daily protein concentrations HOF daily total protein production "r = Correlation coefficient. bp = 'Significance. en = Number of cases where two determinants were available for correlation. HOF = Human oviductal fluid. ph n' <.1 19 <.5 13 <.1 13 >.5 13 >.5 13 >.5 13 were found on cycle days 27 and 28, when the average protein concentration was 7.4 gm/dl. HOF volumes and protein concentrations were inversely proportional (Fig. 1). HOF daily protein concentrations were inversely proportional to serum estradiol levels (r = -.616, P <.5) (Tables 1 and 2; Fig. 1). By multiplying the average protein concentration of a designated cycle day times that same day's average HOF volume, we calculated the average daily total quantities of HOF protein (Tables 1 and 2; Fig. 1). The largest quantities of HOF protein in a single 24-hour period were found on days 12 and 13, coinciding with the highest serum estradiol levels. This was true although the lowest protein concentrations of the cycle were seen on these same days. The volumes of HOF were so great at midcycle that total oviductal protein per 24 hours was also greatest at midcycle. Conversely, although HOF protein concentration was highest on days 27 and 28, total proteins produced per day by the human oviduct at this time of the cycle was low, because HOF volumes were very small. Average total protein concentration for one day varied from a low of 1.8 gm/dl to a high of 7.4 gm/dl. Total protein per day varied from a low of 11.4 mg/24 hours to a high of mg/24 hours. The daily quantity ofhof protein correlated positively with the daily HOF volume (r =. 778, P <.1). Hormonal-HOF Data Coefficients of Correlation. Table 2 delineates the coefficients of correlation between serum estradiol levels, serum pro- November 1981 gesterone levels, and HOF data. Serum estradiol levels correlated positively with the daily HOF volumes and the daily total protein production by the human oviducts. There was a negative correlation between the HOF daily protein concentrations and serum estradiol determinations. These three correlations with estradiol determinations were statistically significant. In contrast, progesterone displayed negative correlations both with HOF daily volumes and HOF daily protein production. There was a positive correlation between HOF daily protein concentrations and serum progesterone levels. However, the above three correlations with serum progesterone determinations were not statistically significant. Electrophoresis ofhof. In 6 of 15 HOF samples, which were subjected to electrophoresis with the "spectrophor," a split albumin could be seen. An example of the split albumin can be seen in Figure 2. This was observed only with HOF samples that had been collected around the time of ovulation and during 3 days following ovulation. Split albumin peaks were not observed with HOF samples that had been collected in the premenstrual period or during the immediate postmenstrual period (Fig. 3). Ovulatory or periovulatory samples of HOF, demonstrating this albumin augmentation, were colorless and clear. HOF ovulatory samples that were yellow in color revealed a single albumin peak (Fig. 4). It was also observed that prealbumin was elevated in 13 of 15 samples. Electrophoresis of Periovulatory HOF Demonstrating a Split Albumin FIG. 2. Electrophoresis of periovulatory human oviductal fluid demonstrating a split albumin. The bottom curve is the electrophoretic pattern of the human oviductal fluid. The top line is an integration of the relative amounts of protein in each peak.

5 Vol. 36, No. 5 HUMAN OVIDUCTAL FLUID PROTEINS 627 A When these same samples were subjected to electrqphoresis on acrylamide gel and stained with Coomassie blue, the split albumin could not be seen (Fig. 5). With acrylamide gel electrophoresis, ovulatory or periovulatory samples of HOF almost always displayed a prominent band of albumin and a second band in the rl-globulin region (Fig. 6). With some periovulatory HOF samples, the albumin and rl-globulin bands were frequently the only bands seen. At times of the cycle away from ovulation, electrophoretic patterns of HOF, both on agarose, using the "electrophor" and on acrylamide gel, appeared to be similar to serum (Figs. 3 and 5). Figure 5 displays electrophoretie changes on acrylamide gels throughout the menstrual cycle. The periovulatory split albumin was never seen on acrylamide gels. DISCUSSION Electrophoresis of HOF From A Premenstrual Day B The coefficient of correlation between estradiol and HOF volumes was significant. Thus, the human oviduct behaves in a manner similar to other mammalian oviducts, 7 " 1 in that estrogen apparently stimulates the production of oviductal fluid. With single samples, daily HOF volumes can change dramatically from.1 ml before and after menstruation to a high of 24 ml at ovulation, a 24-fold increase. In monkeys, Mastroianni observed that larger volumes of monkey oviductal fluid were produced around the time of ovula Eiectrophoresis of HOF From A Postmenstrual Day FIG. 3. A, Electrophoresis of human oviductal fiuid (HOF) from a premenstrual day. B, Electrophoresis of HOF from a postmenstrual day. o Eiectrophoresis of HOF Collected at Ovulation FIG. 4. Electrophoresis of human oviductal fiuid collected at ovulation.

6 628 LIPPES ET AL. FIG. 5. Acrylamide gel electrophoresis of preovulatory, ovulatory, and postovulatory human oviductal fluid. tion, 11 a characteristic shared by humans. Lowest HOF protein concentrations occurred with the largest volumes ofhof just before ovulation, and HOF protein concentrations were inversely proportional to serum estradiol levels. The HOF periovulatory volumes were great enough so that in spite of low protein concentrations, the daily oviductal total protein production was at-its maximum one day prior to ovulation and for 3 days thereafter. It is understood that oviductal fluid and proteins collected during these experiments were derived from transudation and secretion. What proportion each process contributes and how these dynamics might vary throughout the cycle is not known. Does the human oviduct produce unique proteins? Electrophoresis revealed augmentations in the albumin and -globulin region. Whether these augmentations represent a protein or proteins unique to the human oviduct awaits further investigation. Split albumin peaks seen in HOF were not seen in serum samples. It was also noted that prealbumin was elevated in most HOF specimens. In clear, colorless HOF samples, these observations were consistent. This was not found with HOF samples that were straw-colored. It is possible that some straw-colored samples had been contaminated by blood or serum, which masked electrophoretic patterns of HOF so that the split albumin could have been present but not seen. The "spectrophor" utilizing agarose is a more sensitive electrophoretic technique than staining acrylamide gels with Coomassie blue. This is a possible explanation of why split albumin in HOF was only seen on agarose analyzed with the "spectrophor" The split albumin in periovulatory HOF is of interest because albumin has been shown to be one of the necessary ingredients for mammalian sperm capacitation and for fertiliza:tion The November 1981 observation of an albumin augmentation in HOF suggests that in humans, as in other mammals, albumin may play an important role in capacitation, fertilization,.and cleavage. Kay and Feigelson, studying rabbit-tubal fluid, found a carbohydrate-containing prealbumin band and an estrogen-dependent postalbumin band Similar proteins were observed in rabbit tubal fluid in early prgnancy by Foley et al. 2 The albumin, prealbumin, and postalbumin bands seen in rabbit oviductal fluids may have a human counterpart represented by the split albumin peaks. It should be noted that around ovulation, the only other visible band besides albumin is a globulin fraction. Moghissi described a -gly coprotein in HOF that migrated behind transferrin and was not seen in human serum. 21 This observation was made on a single biologic sample, behooving cautious interpretation. However, Mastroianni described a similar -glycoprotein in monkey oviductal fluid, 22 which heightens suspicion that such a -glycoprotein may exist in FIG. 6. Acrylamide gel electrophoresis of periovulatory human oviductal fluid showing a prominent albumin and p-globulin band. Albumin, 33.8%; Cit. 3.8%; a2, 6.2%; p, 44.4%; "(, 11.8%.

7 Vol. 36, No.5 HUMAN OVIDUCTAL FLUID PROTEINS 629 HOF. Even if one or more of these HOF proteins are eventually shown to be unique, their physiologic function is not known. With the use of immunofluorescence, immunoglobulins and secretory piece have been localized in the human oviduct. 23 These proteins are part of a mucosal immune system. Receptor sites for estrogen and progesterone have been demonstrated in the oviduct This fact plus the variation ofhof volumes and protein concentrations correlating with estradiol levels throughout the menstrual cycle provide cumulative evidence that the human oviduct's secretory functions are hormonally controlled. HOF electrophoretic patterns, different from serum, are noted around ovulation, when large volumes of HOF are produced, probably stimulated by estrogen. With immunofluorescent techniques, it has been demonstrated that prostaglandin is present in the human oviductal mucosa before ovulation but shifts to the lamina propria after ovulation. 3 By RIA, prostaglandin F 2 a levels have been shown to vary in HOF at different times of the cycle.. This suggests the possibility that prostaglandin may be the intermediary between estrogen and oviductal secretion. For these reasons, the human oviduct should be considered at least an endocrine target organ. Knowledge about HOF proteins may be useful for in vitro fertilization and the transfer of human ova. It may also be of value in the development of new and safer contraceptives. REFERENCES 1. Lippes J, Enders RG, Pragay.DA, Bartholomew WR: The collection and analysis of human fallopian tubal fluid. Contraception 5:85, Lippes J: Applied physiology of the uterine tube. In Obstetrics and Gynecology Annual 1975, Series Edited by RM Wynn. New York, Appleton-Century-Crofts, 1975, p Ogra SS, Kirton KT, Thomasi TB, Lippes J: Prostaglandins in the human fallopian tube. Fertil Steril 25:25, Lippes J: Analysis of human oviductal fluid for low molecular weight compounds. In The Biology of Fluids in the Female Genital Tract, Edited by FK Beller, GFB Schumacher. New York, Elsevier-North Holland, 1979, p Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the folin phenol reagent. J Bioi Chern 193:265, Weber K, Osborn M: The reliability of molecular weight determinations by dodecyl sulphate-polyacrylamide gel electrophoresis. J Bioi Chern 244:446, Hamner CE: The oviduct fluid. In Pathways to Conception. Springfield, Ill, Charles C Thomas, 1971, p 3 8. Restall BJ: The fallopian tube of the sheep. II. The influence of progesterone and oestrogen on the secretory activities of the fallopian tube. Aust J Bioi Sci 19:181, Yoshinaga K, Mahoney WA, Pincus C: Collection of oviduct fluid from unrestricted monkeys. J Reprod Fertil 25:117, Stanke DF, DeYoung DW, Sikes JD, Mather EC: Collection of bovine oviduct secretions. J Reprod Fertil 32:535, Mastroianni L, Shah U, Abdul-Karim R: Prolonged volumetric collection of oviduct fluid in the rhesus monkey. Fertil Steril12:417, Miyamoto H, Chang CM: The importance of serum albumin and metabolic intermediates for capacitation of spermatozoa and fertilization of mouse eggs in vitro. J Reprod Fertil 32:193, Biggers JD: New observations on the nutrition of the mammalian oocyte and the preimplantation embryo. In Biology of the Blastocyst, Edited by RJ Blandau, Chicago, University of Chicago Press, 1971, p Whitten WK, Biggers JK: Complete development in vitro of the preimplantation stages of the mouse in a simple chemically defined medium. J Reprod Fertil 17:399, Onuma H, Maurer RB, Foote RH: In vitro culture of rabbit ova from early cleavage stages to the blastocyst stage. J Reprod Fertil 16:491, Kane MT, Foote RH: Factors affecting blastocyst expansion of rabbit zygotes and young embryos in defined media. Bioi Reprod 4:41, Mukeijee AB: Normal progeny from mouse oocytes matured in culture and fertilized in vitro by sperm capacitated in vitro. Nature, London, 1972, p Kay E, Feigelson M: An estrogen modulated protein in rabbit oviductal fluid. Biochem Biophys Acta 271:436, Feigelson M, Kay E: Protein patterns of oviductal fluid. Bioi Reprod 6:224, Foley CW, Engle CC, Plotka ED, Roberts RC, Johnson AD: Influence of early pregnancy on proteins and amino acids of uterine tubal fluids in rabbits. Am J Vet Res 33:259, Moghissi KS: Human fallopian tube fluid. I. Protein composition. Fertil Steril 21:821, Mastroianni L Jr, Urzua M, Stambaugh R: Protein patterns in monkey oviductal fluid. Fertil Steril 21:817, Lippes J, Ogra SS, Tomasi TB Jr, Tourville DR: Immunohistological localization of Gamma G, Gamma A, Gamma M secretory piece and lactoferrin in the human female genital tract. Contraception 1:163, Brush MG, Taylor RW, King RJ: The uptake of 6,7-3H estradiol by the normal human female reproductive tract. J Endocrinol 39:599, Van Leusden HA: Endocrine activity of the human fallopian tube (abstr). Acta Endocrinol (Suppl) 138:8, 1969