Diaa M. El-Mowafi - Zagagig University, Egypt

Peritoneal fluid embryotoxicity and cytokine (TNF-alpha) level in endometriosis associated infertility

Diaa M. El-Mowafi, M.D.*, Wafaa A. El-Mosallamy, M.D., Emad Basiony, M.D.,
Sayed Abdel Ghany, M.D., Roshdan M. Arafa M.D, and Umnia A. El-Hendy, M.D.
Obstetrics & Gynecology,* and Microbiology & Immunology Departments, Benha Faculty of Medicine, Egypt.

ABSTRACT

Objective: The effect of peritoneal fluid (PF) on development of 2-cell mouse embryos in vitro and the relation between cytokine (TNF-a ) level and infertility were evaluated.

Design and setting : Prospective controlled study

Subjects and interventions: PF samples were collected through laparoscopy from 30 infertile females with endometriosis (IF-E) (test group) and another 30 fertile females without endometriosis (F-NE) (control group). PF samples were subjected to study their effect on 2-cell mouse embryos development, estimation of total protein concentration, detection of different protein bands by sodium dodycyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) and estimation of tumor necrosing factor (TNF-a ) by Western blotting and ELISA.

Results: The embryotoxicity of PF of IF-E group was significantly high (P<0.001), embryos reached the blastocyst stage in 2-cell mouse embryos were 57/306 embryos in IF-E group and 280/300 embryos in F-NE group. The total protein concentration was statistically high in IF-E group (P< 0.001). Analysis of PF by SDS-PAGE showed similar protein bands but the total number was 10-22 bands (mean 11 bands) in IF-E group and 5-8 bands (mean 7 bands) in F-NE group. A peritoneal fluid protein band corresponds to 32 Kd appeared in all PF samples of IF-E group and not detected in PF of F-NE samples. A Band of TNF-a was detected by western blotting in all samples of IF-E and in only 7/30 samples of F-NE. The level of TNF-a was significantly high in IF-E group (P < 0.0001).

Conclusion: The PF from patient with mild endometriosis were embryotoxic to the 2-cell mouse embryos and had a high level of protein concentration and TNF-a and also a foreign protein band at 32Kd. These results suggest the presence of one or more PF factors that are toxic to early embryonic development and thus may play a role in the infertility associated with endometriosis, without pelvic adhesions or other anatomical distortions.

Key words: Infertility, endometriosis, peritoneal fluid, 2-cell mouse embryo, embryotoxicity.

Endometriosis stands as one of the most investigated disorders in gynecology with more than 4500 published articles in the past 25 years. Yet in spite of this intense academic interest, there remain basic holes in understanding this disease. The question is still not clearly answered why endometriotic patients without pelvic adhesions or distorted anatomy are infertile?

The peritoneal fluid (PF) is a dynamic environment that links the reproductive and immune systems. It appears likely that endometriotic tissue is influenced by this environment and in turn, the PF is altered by the presence of endometriosis. The result is a cellular soap rich in stimuli that can assist the growth and maintenance of endometrial implants as well as inhibition of fertility.1

Peritoneal fluid factors act as a mediators of infertility include macrophages2 markers of inflammatory response and prostaglandins3. Accumulated soluble substances such as prostaglandins, cytokines and growth factors probably act as mediators to induce endometriotic support and interference with fertility and early embryonic development. Activated macrophages release a number of biologically active products including tumor necrosing factor (TNF), which may cause variety of effects including cytotoxicity. It is possible that activation of macrophages and hence infertility associated with endometriosis in some way is related to production of TNF4,5 .

PATIENTS AND METHODS

Peritoneal fluid (PF) samples were collected through laparoscopy from 60 patients attending the gynecology department in Benha University Hospital, Egypt. Thirty patients were infertile with mild or minimal endometriosis (IF-E, test group) and 30 patients were fertile without endometriosis (F-NE) and indicated for laporoscopic sterilization (control group). The infertile patients were free from pelvic adhesions, distorted anatomy or blocked tubes. Thorough infertility investigations were done to exclude any cause of infertility. These investigations included semen analysis, post-coital test, premenstrual endometrial biopsy, hysterosalpingoraphy and diagnostic laparoscopy. The fertile patients were non-somking, non-lactating, parous women, they neither taking hormonal contraceptives nor using intrauterine contraceptive device in the last 3 months and were candidates for laparoscopic tubal sterilization.

A complete history was taken from each patient including history of exposure to infection, previous medications or surgical interference. All women were subjected to general and local examination.

Peritoneal fluid samples were collected by the gynecologic surgeon from the anterior and posterior cul-de-sacs by an aspiration needle during laparoscopy before any manipulative procedures or tubal chromopertubation. Samples were transported in ice to the laboratory within 30 minutes. Samples were centrifuged at 1500 rpm for 10 minutes at 4ºC. The supernatant free cells was stored in aliquots of 250 m l volume at - 70ºC till the time of assay for: its effect on 2-cell mouse embryos development, estimation of total protein concentration, analysis by sodium dodycyl sulphate poly acrylamide gel electrophoresis (SDS-PAGE), immunoblotting for detection of tumor necrosis factor alpha (TNF-a ) and quantitative estimation of TNF-a by ELISA.

Materials:

Animal: B6 CBAF1/J female and male mice (Jackson Lab, Ontario, Canada).

Reagents, solutions and kits : pregnant mare serum, human chorionic gonad(hCG) (Sigma, USA), bicinchoninic acid protein assay reagent kit (Pierce Chemical Company, USA). Acrylamide, bisacrylamide, ammonium persulphate, TEMED (N, N, N, N tetramethyl enediamine), glycine, glycerol, beta-mercaptoethanol, bromophenol blue, molecular weight marker (17000 Dalton), nitrocellulose membrane sheet (0.45 microns), polyclonal antihuman TNF-a (Endogen, USA), goat anti-rabbit IgG-alkaline phosphate conjugate substrate kit (Bio-Rad Lab, USA) and Quantikine human TNF-a immunoassay kit (R & D, systems Inc, USA).

Embryotoxicity Assay of Peritoneal Fluid (Hogan et al., 1994)

Female mice of B6 CBAFI/J strain 8 weeks old were injected intraperitoneally by 5 IU of pregnant mare serum gonadotrophin (day 1). Forty-eight hours later, mice were given an intraperitoneal injection of 5 IU of human chorionic gonadotrophin (hCG). Female mice were kept with male mice for overnight. Mating was confirmed by the presence of vaginal plug the next morning. Forty-eight hours after hCG injection, the mouse was sacrificed. Under dissecting microscopy, the oviduct was separated by cutting between the uterine horn and ovary with fine scissors. The oviduct was put into a 35 mm petridish or embryological watch glass containing Ham’s F10 medium supplemented by 10% fetal calf serum, 100 U/ml benzyl penicillin and 100 m g/ml streptomycine sulphate. With one-ml syringe attached to a 32-gauge needle, the oviduct was flushed to separate the 2-cell mouse embryo. Sometimes, if flushing was difficult, the oviduct could be simply torn at several points along its length with the needle tip, so the embryos could be extruded out. Paraffin oil was added to the previous medium, which equilibrated in 5% CO2 incubator at 37° C for overnight. Thirty m L of the equilibrated medium were dispensed on the bottom of the tissue culture dish, which was flooded with equilibrated paraffin oil. The tested PF was added in dilution 1:1 to the culture medium in drops. At least 10 embryos were transferred to each drop and were placed in the CO2 incubator for72 hours. The embryos were examined by dissecting microscopy at 24 hours intervals. Embryos were classified as degenerated if there is dark granular cytoplasm fragments or cell mass was retracted from the zona pellucida. The percentage of noblastocyst to total numbers of embryos in each culture after 24, 48 and 72 hours of development was estimated.

Estimation of Protein Concentration in the PF:

Protein concentration was measured by bicinchoninic acid protein assay reagent kit following the instruction of kit manufacturer and as described by Smith et al., (1985)7.

Analysis of PF by SDS-PAGE:

Analysis of PF by SDS-PAGE was performed as described previously by Laemmli, (1970)8. Pouring 10% separating gel, top of the gel was covered with a layer of deionized water, the gel polymerized in about 30 minutes at ambient temperature. The water layer was removed and the staking gel was poured. A teflon comb was inserted into the layer of the staking gel solution which was allowed to polymerized for 30-45 minutes at room temperature.

PF samples with the same protein concentration were diluted by sample buffer and boiled for 5 min at 100oCfor denaturation. The PF samples and molecular weight marker were loaded into the wells. The power supply was then connected. Lastely, the gel was dissembling from the glass plates where it could be stained with Coomassie brilliant blue solution.

Western Immunoblotting for Detection of TNF-a (Towbin et al., 1979)9:

Following electrophoresis, the gel was rinsed in transfer buffer. The nitrocellulose membrane was cut to the dimensions of the gel. Saturated filter paper with transfer buffer was placed on the top of the filter pan. The gel was placed on the top of the paper. The surface of the gel was flooded with transfer buffer, the membrane was applied and the sandwich was completed by placing a piece of saturated filter paper on top of the membrane. The power supply was turned to initiate transfer at 100 voltage for 1 hour.

At the end the blotting, the membrane was immersed into the blocking solution then washed. The polyclonal anti-human TNF-a was added to the membrane and incubated overnight to increase the sensitivity of detection. Then, the conjugated secondary antibody was added and color development solution according to instructions in the kit, where the purple visible bands of TNF-a protein appeared.

Quantitative Estimation of Human Tumor Necosis Factor (TNF- a ) by ELISA:

The technique was performed according to the instructions of kit manufacturer. TNF-a standard was reconstituted to produce a stock solution of 1000 pg/ml and then making different dilutions from the standard 500, 250, 125, 62.5, 31,2 15.6 pg/ml. To the TNF-a microtiter plate 50 m l of assay diluent was added to the well, then 200 m l of standards and samples were added to each well and incubated for 2 hours at room temperature. The wells were washed 3 times, 200 m l of polyclonal anti-human TNF-a conjugate was added to the well and incubated for 2 hours, washed, then color reagent was added to the well and incubated for 20 minutes, 50 m l of stopping solution was added and the optical density of each well was determined within 30 minutes by microtiter plate reader set to 450 nm. The optical density of the standards versus the concentration of the standards was plotted and the curve was drown using log/log paper.

Student T test was used to analyze our results.

RESULTS

This study was carried out on 60 patients, their clinical characters regarding age, weight and duration of infertility are shown in Table (1).

The embryotoxicity of the peritoneal fluid of IF-E group was extremely higher compared to F-NE group (P < 0.001). Fifty-seven out of 306 embryos (18.62%) reached to the blastocyst stage when cultured with PF from IF-E. This number was compared to 280/300 embryos (93.33%) in F-NE. (Table 2 and Fig.1).

The total protein concentration (mean 32.81± 6.16 mg/ml) in PF of IF-E group was significantly higher than that in PF of F-NE group (mean 21.07± 12.66 mg/ml) (P < 0.001).

Peritoneal analysis by SDS-PAGE revealed several similar protein bands in both groups. PF of IF-E had higher number of protein bands, mean 11 bands, compared with F-NE group, mean 7 bands, (fig. 2). There was a protein band corresponding to 32kd, molecular weight, in the PF of all the cases of IF-E group and it could not be detected in any one of F-NE group (Fig.2).

By alkaline phosphatase immuno-bol kit, the cytokine TNF-a band appeared in the PF. of IF-E group and in only 4 cases of F-NE group.

Detection of the cytokine TNF-a by ELISA revealed that its level in the PF of IF-E group (295.33± 63.41 pg/ml) was significantly higher than that in F-NE group (106.97 ± 20.72 pg/ml) (P < 0.0001) (table 3).

Table (1): The clinical characters of fertile female without endometriosis (F-NE) group and infertile female with endometriosis (IF-E) group.

Groups / Characters	F- NE group	IF-E group
Age/years: mean ± SD (range)	31.3 ± 3.79 (21 - 40)	28.83 ± 4.04 (21 - 35)
Weight /kg: mean ± SD (range)	68.96 ± 7.45 (60 - 83)	71.36 ± 6.32 (60 - 86)
Duration of infertility: mean ± SD (range)	-	5.17 ± 2.87 (1 - 10)

Table (2) : Embryotoxicity assay of PF on the 2-cell mouse embryos development in fertile female without endometriosis (F-NE) group and infertile female with endometriosis (IF-E) group..

Groups / Data	F-NE group	IF-E group
Total number of embryos	300	306
Number of blastocyst	280	57
Percent of blastocyst	93.33 %	18.62
Z		8.89
P		< 0.001

Table (3): The level of peritoneal fluid cytokine TNF-a by ELISA

Groups / TNF-a level pg/ml	F-NE group	IF-E group
Range	80-140	200-400
Mean± SD	106.97± 20.72	295.33± 63.46

P< 0.0001

 DISCUSSION

Infertility is a serious complaint affecting about 15% of married couples and has its social and psychic impacts. The problem becomes amazing when no cause can be detected in spite of thorough routine investigations. This is the situation in what is called unexplained infertility and infertility in endometriosis patients without any anatomical distortion. In such cases, the attention was directed to the peritoneal fluid microenvironment that is surrounding the proximal parts of the tubes and passes easily into the tubal lumen, the portion in which the normal fertilization takes place. Spermatozoa as well as early embryonic stages are exposed to the peritoneal fluiwith its cellular and soluble components10 .

In the current study, the percentage of 2-cell mouse embryos that developed to the blastocyst stage was 18.62% when incubated with PF of endometriotic patients, and 93.33% when incubated with PF of the control group. These results of embryotoxic effect of PF from endometriotic patients agree with those of Macros et al., (1985)11, and Prough et al., (1990)12 but disagree with those of Dodds et al., (1992)13 and Haney et al., (1994)14.

The present study showed that the total protein concentration of PF was significantly higher in endometriotic group (32.81± 6.16mg/ml) than the fertile control group (21.07 ± 12.66mg/ml) (P < 0.0001). These results disagree with that of Joshi et al., (1986)15 and Nothnick et al., (1994)16. Actually, all our PF samples were collected in the proliferative phase of the cycle while those authors did not mention a particular phase for collection of samples in their study. Endometriotic implants differ from its uterine counter part in steroid responsiveness and receptor contents16. These altered characteristics combined with the ectopic location of the implants, may affect the physiological activity and hence the protein synthesis by the endometriotic tissues.

The analysis of PF by SDS-page in our study revealed different number of protein bands in the two groups. The total number of bands in the endometriotic group was higher (range 10-22) bands than those in control group (range 5-8 bands). The detection of 32kd band in all PF samples from the endometriosis, and its absence in the fertile control group make it possible to be used as a marker for diagnosis and follow up of the treated cases of endometriosis. Our results in this point agree with that of Nothnick et al., 199416, who dthe same protein band in PF of endometriotic patients but not in fertile women undergoing tubal ligation.

PF contains a variety of free-floating cells including macrophages, mesothelial cells, lymphocytes, esinophile and mast cells. Macrophare capable of secreting various substances such as growth factors, cytokines, prostanoids, complement components and hydrolytic enzymes17. It is now became evident that cytokines play an important role in reproduction at various levels, including gamete function, fertilization, embryo development, implantation, and post-implantation survival of the conceptus. Furthermore, growth factors and inflammatory mediators produced by peritoneal leucocytes have recently been postulated to participate in the pathogenesis of endometroisis18.

Eisermann et al.,(1988)4 reported that the difference between the mean concentration of PF-TNFa of women with normal pelvic anatomy and women with moderate or severe endometriosis was significant (P<0.01). Also the proportion of PF-TNFa positive women with pelvic inflammatory disease and those with moderate endometriosis was also significantly higher when compared to women with normal pelvic anatomy. In our study, a band of the cytokine TNF-a could be detected by alkaline phosphatase blot on nitrocellulose paper (Western blotting) in the peritoneal fluid of all cases of endometriosis with different density. The level of this cytokines was determined quantitatively by ELISA and was significantly higher in endometriosis (295.33± 63.46 pg/ml) than that in the control group (106.97± 20.72pg/ml) (P < 0.0001). These results were similar to those of Tanaka et al., (1992)19, Rana et al., (1996)20, and Montoya et al., (1997)21, but disagree with Keenan et al., (1995)22.

Conclusion

The peritoneal fluid from patients with mild endometriosis had more embryotoxic effect on 2-cell mouse embryo, contains high protein level, more protein bands, constant foreign band at 32kd and high TNF-a concentration than the control group. This may explain the inability for conception in spite of the absence of adhesions, anatomical changes or other infertility factors in endometriosis.

Immuno-regulatory therapy for such cases may be directed to the inhibition of lymphocytes proliferation and to normalize the plasma concentration of TNF-a and other cytokines as well as blocking the activation of macrophages. In addition, TNF-a and 32kd protein band in the PF can be used as a marker for diagnosis of endometriosis and its response to the different modalities of treatment.

We acknowledge the support and assistance of Michael Diamond, M.D. and Chitranjian Lall, M.D., professors, Department of Obstetrics and Gynecology, Wayne State University, MI, USA.

REFERENCES

1- Vinatier D, Dufour P, Oosterlynck D. Immunological aspects of endometriosis. Hum Reprod Update 1996; 5:371-84.

2- Halme J, White C, Kauma, S. Peritoneal macrophages from patients with endometriosis release growth factor activity in vitro. J Clin Endocrinol Metab 1988; 66: 1044-49.

3- Fakih H, Baggett B, Holtz G. Interleukin-1: A possible role in the infertility associated with endometriosis. Fertil Steril 1987; 47: 213-17.

4- Eisermann J, Gast MJ, Pineda J. Tumor necrosis factor in peritoneal fluid of women undergoing Laparoscopic surgery. Fertil Steril 1988; 50: 573 - 76.

5- Oral E, David L, Arici A. The peritoneal environment in endometriosis. Hum Reprod Update 1996; 2: 385-98.

6- Hogan B, Beddinggtion R, Costantini F, Lacy E. Recovery, culture and transfer of embryos in: Manipulating the Mouse Embryo. Cold spring Harbor Laboratory, New York 1994. p. 91-104.

7- Smith PK, Krohn RL, Hermanson GT Measurement of protein using Bicnchoninic acid. Anal Biochem 1985; 150: 76-85.

8- Laemmli MK. Cleavage of structural proteins during the assembly of the head of bacteriophage. Nature 1970; 227: 680 - 85.

9- Towbin I, Staehelin T, Gordon J. Electrophoresis transfer of proteins from ployacrylamide gels to nitrocellulose sheets: procedures and some application. Proc Nat Acad Sci 1979; 76:4350-56.

10- Casslen KJ. Uterine fluid volume. Cyclic variations and possible extrauterine contributions. J Reprod Med 1986 31: 506 - 10.

11- Macros RN, Gibbons WE, Findley WE. Effect of peritoneal fluid on in vitro cleavage of 2-cell mouse embryos: possible role in infertility associated with endometriosis. Fertil Steril 1985; 44:678-83.

12- Prough GL, Akesl A, Gilmore S. Peritoneal fluid fractions from patients with endometriosis do not promote two-cell mouse embryo growth. Fertil Steril 1990; 54:927-30.

13-Dodds W, Miller F, Friedman C. The effect of preovulatory peritoneal fluid from cases of endometriosis on murine in vitro fertilization, embryo development, oviduct transport and implantation. Am J Obstet Gynecol 1992;166: 219 - 24.

14- Haney AF, Rehta R, Doty E. An elicited intraperitoneal inflammatory response has no effect on the establishment of pregnancy in mouse. Ferlil Steril 1994; 61: 926 -32.

15- Joshi S, Zamab N, Raikar R. Serum and peritoneal fluid proteins in women with and without endometriosis. Fertil Steril 1986; 46:1077-86.

16- Nothnick W, Steve N, Curry T. Detection of a unique 32 Kd protein in the peritoneal fluid of women with endometriosis. Fertil Steril 1994; 61:288-93.

17- Nathan CF. Secretory products of macrophages. J Clin Invest 1987; 79: 319-26.

18- Arici A, Tazuke S, Attar E. Interleukin-8 concentration in peritoneal fluid of patients with endometriosis and modulation of interleukin-8 expression in human mesothelial cells. Mol Hum Reprod 1996; 2: 40 - 45.

19- Tanaka A, Kuo T, Mizuno I. Comparison of cytokine levels and embryotoxicity of peritoneal fluid of fertile women with untreated or treated endometriosis. Am J Obstet Gynecol 1992; 167:265-70.

20- Rana N, Gebel H, Braun D. Basal and stimulated secretion of cytokines by peritoneal macrophages in women with endometriosis. Fertil Steril 1996; 65: 925-30.

21- Montoya L, Doucet R, Monewald L. Tumour necrosis factor-a . increases endometrial stromal cell adhesto extracellular matrix proteins. J Soc Gynecol Invest 1997;1:219-21.

22- Keenan J, Thomas T, Nancy L. TL-IB, TNF-a and Tl-2 in peritoneal fluid and macrop hage - conditioned media of women with endometriosis. Am J Repord Immunol 1995; 34: 381 - 85.