Macrophage migration inhibitory factor in the human ovary: Presence ...

Vol. 4!, No. 4, April 1997

BIOCHEMISTRYand MOLECULARBIOLOGY INTERNATIONAL Pages 805-8] 4

Macrophage Migration Inhibitory Factor in the Human Ovary: Presence in the Follicular Fluids and Production by Granulosa Cells Shin-ichiro Wada 1, Sei-ichiro Fujimoto 1, Yuka Mizue 3, and Jun Nishihira 2 1Department of Gynecology and 2Centrat Research Institute, Hokkaido University School of Medicine, Sapporo 060, Japan 3Sapporo Immunodiagnostic Laboratory, Sapporo 060, Japan Received November 26, 1996 Received alter revision December 9, 1996

SUMMARY: Cytokines play an important role in ovarian function. We unexpectedly found high expression of macrophage migration inhibitory factor (MIF) mRNA in human ovarian tissues. Hence, we examined the presence of MIF in the follicular fluid because the follicular microenvironment is important for oocyte fecundity. The follicular fluids were collected from ovaries of patients undergoing in vitro fertilization and embryo transfer. A higher amount of MIF was identified in the follicular fluid, 80.3 _+ 4.6 ng/ml (mean + SE), in which the concentration was significantly decreased as the size of the follicles increased. To detect MIF mRNA expression in the granulosa cells, reverse transcription-polymerase chain reaction was carried out and this showed an amplified transcript specific for MIF. Furthermore, the presence of MIF protein in the granulosa cells was confirmed by Western blot analysis. These results suggest the possibility that MIF mediates various immunological events in the process of oocyte development. Keywords: follicular fluid, granulosa cells, macrophage migration inhibitory factor. INTRODUCTION The immune system is a local regulatory mechanism in the reproductive organs, and a variety of cytokines can exert profound effects on embryo development. For instance, various cytokines regulate gonadal steroid secretion, corpus luteum function, and implantation (1). From the data available to date, the follicular fluid obtained from patients for in vitro fertilization-embryo transfer (IVF-ET) is known to contain various cytokines, e.g. tumor necrosis factor-c~ (TNF-ct), interleukin (IL)-I and IL-6, which have the potential to influence oocyte fecundity (2-4). The roles of these macrophage-derived cytokines have been intensively investigated in embryo Correspondence should be addressed to J. Nishihira, Central Research Institute, School of Medicine, Hokkaido University, N15, W7, Kita-Ku, Sapporo 060, Japan. Tel: +81-11-706-6081 Fax: +81-11-706-7864 1039-9712/97/040805-10505.00/0 805

Copyright 9 1997 by Academic Press Australia. All rightx qf reproduction in any,formtexerved,

Vol. 41, No. 4, 1997

BIOCHEMISTRYcmd MOLECULAR BIOLOGY INTERNATIONAL

development because a large population of macrophages is observed in the stroma surrounding the preovulatory follicles and inside the corpus luteum (5, 6); however, their precise pathophysiological functions remain unclear. Macrophage migration inhibitory factor (MIF), originally identified as a tymphokine concentrating macrophages at infection sites, is a potent activator of macrophages and contributes to cell-mediated immunity (7, 8). Moreover, the protein is reported to enhance macrophage adherence, phagocytosis, and turmoricidal activity (9). MIF was considered to be expressed exclusively in activated T-lymphocytes; however, a recent report indicated that macrophages are another major source of the protein (10). Recently, additional novel properties of MIF have been reported. This protein has the potential to bind glutathione (11), and is a pituitary-derived cytokine possessing a hormone-like action (12). These findings suggest that MIF is ubiquitously expressed in various cells and tissues, and plays an important role beyond the immune system. Human MIF cDNA was first cloned from human T-lymphocytes, which revealed that MIF consists of 114 amino acid residues (13). Later, we cloned rat MIF cDNA, and reported its physicochemical properties (14). Recently, we succeeded in the crystallization of both human and rat MIFs, and revealed the tertiary structure of rat MIF at 2.2 A resolution (15, 16). In this study, we explored the presence of MIF in the human ovary and also in follicular aspirates of ovaries obtained from patients undergoing in vitro fertilization-embryo transfer therapy (IVF-ET). Moreover, we examined whether the granulosa cells of the preovulatory follicles could biosynthesize MIF de novo. MATERIALS

AND

METHODS

Materials

The following materials were obtained from commercial sources. Nylon membrane filters were purchased from Schleicher & Schuell (Keene, NH, USA); Isogen from Nippon Gene (Tokyo, Japan); M-MLV reverse transcriptase from Gibco (Grand Island, NY, USA); Taq DNA polymerase from Perkin-Elmer; horseradish peroxidase-conjugated goat anti-rabbit antibody was from Pierce (Rockford, IL, USA); Vector ABC Kit from Vector Laboratories (Burlingame, CA, USA); Konica immunostaining HRP-1000 from Konica (Tokyo, Japan); Percoll from Pharmacia

806

Vol. 41, No. 4, 1997

BIOCHEMISTRY and MOLECULARBIOLOGYINTERNATIONAL

LKB (Uppsala, Sweden); human chorionic gonadotropin (hCG) from Mochida (Tokyo, Japan), human menopausal gonadotropin from Serono (Tokyo, Japan), buserelin acetate from Hoechst (Tokyo, Japan), and human tubal fluid (HTF) from Irvine Scientific (Santa Ana, CA, USA). The anti-human MIF polyclonal antibody was generated by immunizing New Zealand White rabbits with recombinant human MIF. Human recombinant MIF was expressed in E. coil and purified to homogeneity as previously described (17). All other chemicals used were of analytical grade. F o l l i c u l a r fluid c o l l e c t i o n Human follicular fluids were obtained from patients undergoing IVF-ET. The combined therapy for ovarian stimulation included 150 to 300 U of human menopausal gonadotropin and 900 p.g/day of buserelin acetate from the mid-luteal phase of the previous cycle to the day of hCG administration, hCG (10,000 IU) was injected intramuscularly, and transvaginal ultrasound-guided oocyte retrieval was performed 35 hr later. The follicular development was monitored by daily measurements of serum estradiol (E2) and by ultrasonographic measurements of the follicular diameters. Little red blood cell contamination in these samples was detected. The follicular fluids were stored at -80~ until assay. Granulosa cell c u l t u r e Follicular fluids from a single patient were centrifuged at 200 xg for 5 min, and the supernatant was decanted. The cells were washed once with HTF layered onto 50% Percoll and spun at 200 xg for 30 min to pellet red blood cells. The purified granulosa cell population was aspirated from the interface. Cells were again washed with HTF and passed through a sterile nylon mesh to generate a single cell suspension. Cells were seeded in each well containing 1 ml HTF with 10% fetal bovine serum.

Western blot analysis Homogenates of human ovaries and the primary cultured granulosa cells were treated with 4 volumes of SDS-reducing buffer (60 mM Tris/HCI, pH 6.8, 2% SDS, 10% glycerol, 5% ~-mercaptoethanol, and 0.025% bromophenol blue), and subjected to SDS-PAGE as described (18). Proteins on the gel were transferred electrophoretically to a nitrocellulose membrane by the procedure of Towbin et al (19). Proteins were visualized with a Konica immunostaining HRP-1000 kit as recommended in the manufacturer's protocol.

RT-PCR analysis After harvest, cultured human granulosa cells were frozen immediately in liquid nitrogen until use. The total RNA was extracted from the cultured granulosa cells with an Isogen RNA extraction kit. The reverse transcription of RNA was carried out by M-MLV reverse transcriptase using oligo-dT primer and subsequent amplification using Taq DNA polymerase. PCR was carried out for 28 cycles of denaturation at 94~ for 1 min, annealing at 53~ for 2 min and extension at 72~ for 1 min using a thermal cycler (Perkin-Elmer, Model 2400). MIF primers used were 5'-CTCTCCGAGCTCACCCAGCAG-3' (58-78) (forward) and 5'CGCGTTCATGTCGTAATAGTT-3' (292-312) (reverse). I~-Actin primers used were 5'-CGTTCTGGCGGCACCACCAT-3' (936-935) (forward) and 5'GCAACTAAGTCATAGTCCGC-3' (1170-1189) (reverse). After PCR, the amplified products were analyzed by agarose gel electrophoresis.

Enzyme-linked immunoadsorbent assay (ELISA) Anti-human MIF antibody dissolved in 50 ~tl PBS was added to each well of a 96-well microtiter plate. After incubation for 1 hr at room temperature, the plate was washed three times with PBS. All wells were filled with PBS containing BSA (1%) for blocking and left for 1 hr at room temperature. After removal of the blocking

807

Vol. 41, No. 4, 1997

BIOCHEMISTRYond MOLECULAR BIOLOGYINTERNATIONAL

solution followed by washing three times with PBS containing 0.05% Tween 20 (washing buffer), follicular fluid samples were added to individual wells and incubated for 1 hr at room temperature. After the plate was washed three times with the washing buffer, 50 #1 of biotin-conjugated anti-human MIF antibody was added. Following incubation for 1 hr at room temperature, the plate was again washed three times with the washing buffer. Then avidin-conjugated goat anti-rabbit IgG antibody was added to individual wells and further incubated for 1 hr at room temperature. The plate was again washed three times with the washing buffer. Fiftymicroliters of substrate containing 200 lug of o-phenylethylenediamine and 10 #1 of 30% hydrogen peroxide in 10 #1 of citrate-phosphate buffer (pH 5.0), was adjusted by using citric acid (0.05 M) and disodium hydrogen phosphate (0.1 M). After incubation for 20 min at room temperature, the reaction was stopped with 50 ~tl of 1 N sulfuric acid. The absorbance at 492 nm was measured in an ELISA plate reader (Bio-Rad, Model 3550).

Immunohistochemistry Immunohistochemical study of the cultured granulosa cells was carried out using an anti-human MIF antibody. The cultures of cells grown on cover slips were rinsed with PBS and dipped in -20~ acetone for 5 min. The fixed cells were stained with an avidin-biotin-peroxidase complex procedure using a Vector ABC Kit according to the manufacturer's protocol. In brief, the cell specimens were incubated overnight at 4~ with the anti-human MIF antibody. After three washes with PBS, the samples were reacted with biotinylated goat anti-rabbit IgG and avidin-biotin complex at room temperature for 30 min. The reaction was developed by 3,3'-diaminobenzidine tetrahydrochloride containing hydrogen peroxide (0.01%), and the tissue samples were mounted with alkylacrylates. Statistical analysis Comparison between values was performed by Student's t test or by one-way analysis of variance. RESULTS

Identificatior) of MIF protein To examine the presence of MIF protein in the granulosa cells, Western blot analysis was performed using the anti-human MIF antibody. The immunoblot analysis demonstrated that the protein migrated to the corresponding molecular weight of MIF, about 12.5 kDa, visualized by a Konica immunostaining HRP-1000 kit (Fig. 1). As for the control, human recombinant MIF was used. In the immunoblot analysis, pre-immune rabbit IgG did not react with human MIF (data not shown).

Expression of MIF mRNA MIF mRNA expression in the primary cultured human granulosa cells was examined by RT-PCR analysis. The total RNA extracted from the granulosa cells, reverse-transcribed, then treated with PCR reactions containing oligonucleotide

808

Vol. 41, No. 4, 1 9 9 7

BIOCHEMISTRYand MOLECULARBIOLOGYINTERNATIONAL

1

2

3

4

(kD) 175.0 83.0

47.5 32.5 25.0 16.5 6.5

Fig. 1. Western blot analysis of human ovary and granulosa cells for MIF. The tissue samples were collected, electrophoresed, transferred to a nitrocellulose membrane, and visu/~lized with a Konica immunostaining kit as described in Materials and Methods: Lane 1, recombinant human MIF (50 ng); lane 2, human ovary (50 gg); lane 3, human granulosa cells (5 x 104 cells); lane 4, prestained molecular marker (New England Biolabs).

primers for human MIF showed a specific 255-base pair band corresponding to the amplified product of MIF transcripts (Fig. 2). As a positive control, a transcript of human T-lymphocytes (Jurkat) is shown. The level of MIF mRNA expression was similar in granulosa cells and T-lymphocytes as assumed from the similar intensities of their PCR products normalized by the intensities of #-actin. These results indicated that the granulosa cells had the potential to produce MIF de novo, and together with the results of Western blot analysis, indicated that MIF was synthesized de novo and localized in the granulosa cells. Immunohistochemical detection of MIF in the granulosa cells Localization of MIF protein was clearly identified in the granulosa cells by immunohistochemistry. The cultured granulosa cells were stained by the antibody against human MIF (Fig. 3A). Positive staining was diffusely observed in the cytoplasm and nuclei of the cells but not when the cell sample was reacted with pre-immune rabbit IgG (Fig. 3B). Content of MIF in follicular fluids Follicular fluid was obtained from IVF-ET patients. MIF content was measured by ELISA for 118 follicular fluids obtained from 19 patients. Since the serum concentration of MIF is approximately 5 ng/ml, the results indicated that the content

809

Voi. 41, No. 4, 1997


1

2

3

4

Fig. 2. RT-PCR analysis of human ovary and granulosa cells for MIF mRNA, PCR products obtained from the total RNA isolated from cultured human granulosa cells. The PCR products were electrophoresed on 2% agarose gel. Lane 1, ovary; lane 2, granulosa cells; lane 3, human T-lymphocytes (Jurkat); lane 4, the molecular size marker (pBR322 DNA/AIul). The RTPCR products of 13-actin are shown at the bottom of each lane.

B

Fig. 3. Immunohistochemistry of MIF in human granutosa cells. A, the cells obtained from patients for IVF-ET was reacted with anti-human MIF antibody and stained with a Vector ABC immunostaining kit (x 400); B, a control specimen reacted with pre-immune rabbit IgG (x 400).

810

Vol. 41, No. 4, 1 9 9 7

BIOCHEMISTRY and MOLECULARBIOLOGYINTERNATIONAL

in the follicular fluids was highly concentrated. It is of note that the concentration of MIF was significantly decreased as the size of the follicles increased. In detail, the small (4.0 ml) follicles contained 99.4 + 7.4, 75.6 + 6.9, and 56.0 + 6.6 ng/ml (mean + SE), respectively. This suggested the possibility that the production of MIF by the granulosa cells decreased as the maturation of the follicles proceeded.

/

DISCUSSION We demonstrated the presence of a substantial amount of MIF (80.3 + 4.6 ng/ml) in human follicular fluid. It has long been considered that MIF is exclusively expressed in activated T-lymphocytes and macrophages, but the present results clearly demonstrated that the granulosa cells are another source of MIF. Hence MIF secreted from the granulosa cells may play an important role for mononuclear cell recruitment into the follicles and, moreover, participate in the regulation of cytokines and growth factors. Furthermore, we found that the MIF concentration in the follicular fluids depended on follicular size, whereas no such correlation was found for other cytokine levels, such as IL-1, IL-2 or TNF-c~ (20-22). These facts suggest the possibility that MIF is essential for oocyte fecundity and may exert its unique physiological function in the ovulated follicles. Various cytokines have been identified in the follicular fluids, and their physiological functions have been extensively investigated (20, 23). TNF-o~ in human follicular fluid can stimulate production of prostaglandin (PG) in cultured granulosa cells. It was reported that MIF potentiates release of TNF-c~ and IL-1 from macrophages, and MIF may express its physiological function in concert with these proinflammatory cytokines (10). That is, MIF may play a role not only for macrophage recruitment into the ovulating follicle, b u t also for PG production during ovulation. This stimulates oocyte development during the formation of the early corpus luteum. It is of note, on the other hand, that MIF can induce IL-6 production (24). IL-6 produced by granulosa cells in the preovulatory follicle, could

811

Vol. 41, No. 4, 1997

BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL

be an intraovarian regulatory factor concerned with steroidogenesis (22). These findings indicate a novel mechanism for cytokine-mediated endocrinological events in the human ovulatory follicle by MIF and IL-6. As for another possible function of MIF in the follicle, the protein may contribute to a defense mechanism for oocyte deyelopment. Recently, MIF has been shown to have the potential to bind glutathione and may participate in the cellular-mediated immune system and enzymatic detoxification system (11). It is well known that glutathione acts as a radical scavenger to defend the cell membrane. Since the concentration of glutathione in ovulated murine oOcy{es (ca. 10 mM) is higher than that of other tissues, glutathione is consiclered to provide an adequate microenvironment for oocyte fecundity (25). It'is of interest that the addition of glutathione to cultured oocytes can improve their development (26). Recently, it was reported that MIF is expressed in the early stage of embryonic chicken lens, correlating with lens differentiation (27). We also found the presence of MIF in highly proliferative basal cell layers of human epidermis and human cornea (28, 29). Considering these results together, it appears that MIF might be a critical factor for the early-stage oocyte development in the follicle. In conclusion, we showed highly concentrated MIF in human follicular fluids by an enzyme-linked immunoadsorbent assay. However, we found that cultured granulosa cells produce MIF and the present results do not exclude the possibility that monocytes contaminating in the follicular fluid during aspiration could be a source of MIF in the follicular fluids. Nonetheless, based on'the immunological analyses, it is clear that MIF detected in the follicular fluid, if not all of it, originated in the granulosa cells. Very recently, a novel biological function of MIF was discovered. The protein regulates glucocorticoid induction (24). Glucocorticoid potentiates steroidogenesis in the ovary via action of insulin-like growth factor I (30). These facts support the idea that MIF might be involved not only in the immunological events, but also in steroidogenesis by the granulosa cells. Further investigation is currently under way.

812

Vol. 41, No. 4, 1997

BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL

ACKNOWLEDGMENT: This research was supported by a Grant-in-Aid for research (No. 07670162) from the Ministry of Education, Science and Culture of Japan, and grants from Akiyama Foundatio~'and Ohyama Foundation. We are grateful to T. Kudyama and S. Tone for their technical assistance. REFERENCES

1. Adashi, E. Y. (1990) Endocr. Rev. 11,454-464. 2. Droesh, K., Fulgham, D. L., Liu, H. C., Rosenwaks, Z. and Alexander, N. J. (1988) Fertil. Steril. 50, 618-621. 3. Buyalos, R. P., Watson, J. M. and Martinez-Maza, O. (1992) Fertil. Steril. 57, 1230-1234. 4. Zolti, M., Bider, D., Seidman, S. D., Mashiach, S. and Ben-Rafael, S. (1992) Fertil. Sterit. 57, 501-504. 5. Roby, K. F., Lyles, W. R. and Terranova, P. F. (1990) J. Clin. Endocrinol. Metab. 71, 1096-1102. 6. Veldhuis, J. D., Garmey, J. C., Urban, R. J., Demers, L. M. and Aggarwal, B. B. (1991) Endocrinol. 129, 641-648. 7. David, J. R. (1966) Proc. Natl. Acad. Sci. USA 56, 72-77. 8. Weiser, W. Y., Pozzi, L. M. and David, J. R. (1991) J. Immunot. 147, 2006-2011. 9. Bloom, B. R. and Bennett, B. (1966) Science t53, 80-82. 10. Calandra, T., Bernhagen, J., Mitchell, R. A. and Bucala, R. (1994) J. Exp. Med. 179, 1895-1902. 11. Blocki, F. A, Schilievert, P. M. and Wackett, L. P. (1992) Nature 360, 269-270. 12. Bernhagen, J., Calandra, T., Mitchell, R. A., Martin, S. B., Tracey, K. J., Voelter, W., Manogue, K. R., Cerami, A. and Bucala, R. (1993) Nature 365, 756-759. 13. Weiser, W. Y., Temple, P. A., Witek-Giannotti, J. S., Remold, H. G., Clark, S. C. and David, J. R. (1989) Proc. Natl. Acad. Sci. USA. 86, 7522-7526. 14. Nishihira, J., Kuriyama, T., Sakai, M., Nishi, S., Ohki, S. and Hikichi, K. (1995) Biochim. Biophys. Acta. 1247, 159-162. 15. Suzuki, M., Sugimoto, H., Nakagawa, A., Tanaka, I., Nishihira, J. and Sakai, M. (1996) Nature. Struct. Biol. 3, 259-266. 16. Sugimoto, H., Suzuki, M., Nakagawa, A. Tanaka, I. and Nishihira, J. (1996) FEBS LETT. 389, 145-148. 17. Nishihira, J., Kuriyama, T., Nishino, H., Ishibashi, T., Sakai, M. and Nishi, S. (1993) Biochem. Mol. Biol. Int. 31,841-850. 18. Schagger, H. and Von Jagow. (1987) Anal. Biochem. 166, 368-379. 19. Towbin, H., Staehelin, T. and Gordon, J. (1979) Proc. Natl. Acad. Sci, USA 76, 4350-4354. 20. Wang, L. J. and Norman, R. J. (1992) Hum. Reprod. 7, 147-150. 21. Wang, L. J., Brannstrom, M., Robertson, S. A. and Norman, R. J. (1992) Fertil. Steril. 58, 934-940. 22. Machelon, V., Emilie, D., Lefevre, A., Nome, F., Durand-Gasselin, I. and Testart, J. (1994) J. Clin. Endocrinol. Metab. 79, 633-642. 23. Barak, V., Roisman, I., Yanai, P., Simon, A., Treves, A. J. and Laufer, N. (1992) Fertil. Steril. 58, 719-725. 24. Calandra, T., Bernhagen, J., Mets, C. N., Spiegel, L. A., Bacher, M., Donnelly, T., Cerami, A, and Bucala, R. (1995) Nature 377, 68-71. 25. Legge, M. and Sellens, M. H. (1991) Hum. Reprod. 6, 867-871.

813

Vol. 41, No. 4, 1997


26. Calvin, H. I., Grosshans, K. and Blake, E. (1986) Gamete Res. 14, 265-275. 27. Wistow, G. J., Shaughnessy, M. P., Lee, D. C., Hodin, J. and Zelenka, P. S. (1993) Proc. Natl Acad. Sci. USA 90, 1272-1275. 28. Shimizu, T., Ohkawara, A., Nishihira, J. and Sakamoto, W. (1996) FEBS Lett. 381, 199-202. 29. Matsuda, A., Tagawa, Y., Matsuda, H. and Nishihira, J. FEBS Lett. 385,225228. 30. Urban, R. J., Bodenburg, Y. H., Nagamani, M. and Peirce, J. (1994) Am. J. Physiol. 267, 115-123.

814