- Email: [email protected]
Fibronectin concentrations correlate with ovarian follicular size and estradiol values in equine follicular fluid
SCIENCE Animal Reproduction Science 45 (1996) 91- 102
Fibronectin concentrations correlate with ovarian follicular size and estradiol values in equine follicular fluid Patricia A. Gentry *, Mehri Zareie, Robert M. Liptrap Depurtment ofBiomedicu1 Sciences, Ontario Veterinary College, University of Guelph, Guelph, Ont. NIG 2W1, Cunadu
Accepted 26 February 1996
Abstract The amounts of total protein, albumin, fibronectin, (Y2-macroglobulin (a 2-M), immunoglobulin G, ceruloplasmin and antithrombin were determined in fluids collected from 53 preovulatory equine follicles and compared with the contents of estradiol-17P, progesterone and androstenedione, with follicle size and the amounts of the equivalent proteins in normal equine plasma. The concentration of fibronectin and the fibronectin/albumin ratios increased significantly with follicle size and with follicular estradiol levels. The (Y2-M levels and a 2-M/albumin ratios correlated with follicle size but not with hormone content. Both fibronectin and a 2-M were present in lower amounts in follicular fluid compared with plasma while the other proteins were present in similar amounts. Among the proteins evaluated, there was a positive correlation between the amount of the protein in the follicular fluid and the molecular weight of the protein. Keywords: Horse; Follicular fluid; Fibronectin;
Estradiol;
Proteins
1. Introduction Constituents
of follicular
fluid
are known
to play
an important
role
in follicular
growth, oocyte maturation and ovulation. It has been suggested that the concentrations of specific hormones and proteins in follicular fluid may reflect the physiological status of the follicle (Edwards, 1974; Nayudu et al., 1983; Gulamali-Majid et al., 1987). While the granulosa and theta cells are the major contributors of the steroid hormones present
* Corresponding author. Tel.: (5 19)-824-4120; fax: (519)-767- 1450. 0378-4320/96/S 15.00Copyright PII SO378-4320(96)01554-O
0 1996 Elsevier Science B.V. All rights reserved.
92
PA. Gentry et al./Animal Reproduction Science 45 (1996) 91-102
in follicular fluid the proteins appear to be derived from both the circulating blood as well as from local secretions, principally by the granulosa cells (Gosden et al., 1988). The function of many of the proteins in follicular fluid is thought to be similar to that of plasma. In both human and porcine preovulatory follicles, the colloid osmotic pressure remains relatively constant despite the increase in size and volume of the follicle that occurs during maturation (Gosden et al., 1988). This may be due, in part, to the amounts of total protein and albumin that have been reported in fluid from those follicles (Shagli et al., 1973; Chang et al., 1976; Suchanek et al., 1990). It has been suggested that, as the follicle grows, the follicular wall becomes more permeable to plasma constituents which permits small molecular weight proteins to be filtered more readily from the circulation (Shagli et al., 1973). This concept is supported by observations from several species, including human, equine, porcine and bovine follicles, that the concentration of albumin in preovulatory follicular fluid is similar to that found in the circulation (Andersen et al., 1976; Widders et al., 1984; Wise, 1987; Nagy et al., 1989; Yamada and Gentry, 1995a; Yamada and Gentry, 1995b). Studies of follicular fluids from several species have shown that while the concentration of some proteins corresponds to that of plasma other proteins may be present in higher or lower amounts (Shagli et al., 1973; Chang et al., 1976; Widders et al., 1984; Nagy et al., 1989; Suchanek et al., 1990; Yamada and Gentry; 1995a; Yamada and Gentry, 1995b). The differences in the relative distribution of various proteins in follicular fluid compared with plasma may be a function of both selective transport processes across the follicle wall as well as the result of local synthesis and secretion of a select group of follicular fluid proteins. For example, plasminogen and plasminogen activators have been found in equine follicular fluid samples in higher amounts than those found in normal adult equine plasma (Yamada and Gentry, 1995b). Cell culture studies have shown that human, bovine, porcine and chicken granulosa cells can synthesize and secrete these proteins (Beers, 1975; Reich et al., 1985; Jones et al., 1989; Politis et al., 1990; Reinthaller et al., 1990; Yamada et al., 1991; Ny et al., 1993). Similarly, human, bovine, rat and chicken granulosa cells have been shown to synthesize fibronectin, a protein associated with cell adhesion, proliferation, differentiation and cell migration (Skinner and Dorrington, 1984; Lobb and Dorrington, 1987; Reinthaller et al., 1990; Asem et al., 1992; Lehwald and Gentry, 1992). The early maturation and growth of the follicle is, in part, controlled by the follicle stimulating hormone (FSH) released from the pituitary gland which produces an increase in the ability of the granulosa cells to convert androgens to estrogen (Hsueh et al., 1984). With rising luteinizing hormone (LH) levels, the theta cells convert more cholesterol to androgens thereby supplying the granulosa cells with the necessary precursors for increased estrogen synthesis (Short, 1962; Hsueh et al., 1984; Ny et al., 1993). The increased estrogen levels in the follicle in turn help stimulate the proliferation of the granulosa cells. The ovulatory LH surge also increases the production of progesterone by the highly differentiated granulosa cells (Richards et al., 1976). Despite the ability of granulosa cells to secret both steroids and proteins into the follicular fluid in the preovulatory follicle, there appears to be little information in the literature correlating the functions of these two groups of constituents in equine preovulatory follicles. Hence, one of the objectives of this study was to determine the levels of progesterone,
P.A. Gentry
et
al./Animal ReproducrionScience 45 (1996) 91-102
93
androstenedione and estradiol in equine follicular fluid samples and to determine whether there was any correlation between hormonal values and those of follicular proteins. The proteins selected for evaluation included: albumin, immunoglobulin (IgG), ceruloplasmin and antithrombin (AT), that have low molecular weights and the higher molecular weight proteins, fibronectin and (Y2-macroglobulin (a 2-M). Based on studies of follicular fluid from other species, these proteins have been detected in follicular fluid, although not all have been reported previously in equine follicular fluid. Using this range of proteins, the association between protein content and molecular size could be examined in samples collected from preovulatory follicles.
2. Materials and methods 2.1. Preparation
of follicular fluid
Ovaries were collected at slaughter from mares of unknown reproductive histories during May and June, that is during the physiological breeding season. The ovaries were transported to the laboratory in 10 mM phosphate buffered saline (PBS, pH 7.2) at 39°C in an insulated container. The ovaries were visually examined and the diameter of each follicle measured. In nonspherical follicles, the width was measured in two dimensions at right angles and the average value reported as the diameter. Follicular fluid from individual follicles was aspirated and transferred to a plastic tube which was then centrifuged at 3000 X g for 15 min at 4°C. The supematant was transferred, in several aliquots, to clean vials and frozen at - 20°C. 2.2. Preparation
of plasma
Blood was collected from 15 healthy mares, by venepuncture of the external jugular vein, into a plastic syringe containing 0.17 M sodium citrate in the proportion nine parts blood to one part anticoagulant. The mares were maintained at the University of Guelph research farm. The titrated blood was transferred to plastic centrifuge tubes, centrifuged at 3000 X g for 20 min at 4°C and the top two-thirds of the resulting platelet poor plasma (PPP> removed. The PPP from each animal was pooled and the resulting sample stored in small aliquots at -20°C. 2.3. Protein determinations The total protein and the albumin content of each of the follicular fluid samples and the pooled equine plasma were evaluated with the Biuret (Peters, 1968) and the Bromcresol Green (Gentry and Lumsden, 1978) calorimetric assays, respectively. For each assay a standard curve was prepared with bovine serum albumin (Sigma Chemical Co., St. Louis, MO). Antithrombin values were determined with the chromogenic
94
P.A. Gentry et al./Animal
Reproduction
Science 45 (1996) 91-102
substrate S-2238 (Chromogenix AB, Molandal, Sweden) as previously described for equine plasma (Gentry et al., 1992). Rocket immunoelectrophoresis was used to assess the fibronectin, (Y2-macroglobulin (o2M), ceruloplasmin and immunoglobulin (IgG) levels in the follicular fluid and plasma samples. For fibronectin and o2M, 0.7% solutions of anti-human fibronectin (Calbiochem Co., La Jolla, CA) and anti-human (Y2-M (Kent Labs Inc., Markham, Ont.) were used while 1.5% solutions of anti-human ceruloplasmin (Sigma Chemical Co.) and anti-horse IgG (Sigma Chemical Co.) were used for ceruloplasmin and IgG, respectively. In preliminary examinations the cross-reactivity of these antibody preparations to the specific proteins in equine plasma and follicular fluid had been confirmed. For fibronectin and a2-M, standard curves were prepared with purified human fibronectin (Calbiochem Co., La Jolla, CA) and human ol2-M (Kent Labs Inc., Markham, Ont.), respectively. For all proteins, standard curves prepared from the pooled equine plasma were run concurrently with the follicular fluid samples. 2.4. Hormone assays Follicular fluid aliquots from individual follicles were extracted twice with diethyl ether (Caledon Laboratories Ltd., Georgetown, Ont.> prior to analysis using standard radioimmunoassay procedures to determine concentrations of estradiol- 17B, the estrogen precursor androstenedione and progesterone. Each sample was assayed in duplicate. All radioisotopes were purchased from Amersham Canada Ltd. (Oakville, Ont.) and all non-radioactive steroid standards from Steranti Research Ltd. (St Albans, UK). Antiserum to estradiol (courtesy Dr. J.J. Pratt, Isotopenlaboratorium Academisch, Groningen, Netherlands) has a cross reactivity of 1.32% and 1.23% with estradiol-17cr and estrone, respectively. Androstenedione antiserum (Cedarlane Laboratories, Homby, Ont.) cross reacts 4.0% with androsterone and 2.05% with both testosterone and dehydroepiandrosterone. Antibody to progesterone (UCB Bioproducts SA, Belgium) cross reacts 20% with pregnenolone and under 0.5% with other steroids. 2.5. Statistical analysis Statistical analyses were carried out using a SigmaStat computer program (Jandel Scientific, San Rafael, CA).
3. Results 3.1. Selection of follicles
Only fluid aspirated from follicles between 3.0 and 5.9 cm in diameter were included in this study since these represent growing follicles with the potential to ovulate (Pierson
Psi. Gentry et al./Animal
Reproduction
Science 45 (1996) 91-102
95
Fig. 1. Relationship of the concentration (mean+ SE) of proteinsin equine follicular fluid to their respective molecular weights: AT, antithrombin III; Alb, albumin; Cerulo, ceruloplasmin; IgG, immunoglobulin G; Fn, fibronectin; cx2-M. o(2-macroglobulin.
and Ginther, 1985). The estradiol-androstenedione ratio of the follicular fluid samples was used to confirm that the samples had been collected from developing preovulatory follicles (Ginther, 1992). The application of this selection process resulted in 53 samples from individual follicles being available for analysis. 3.2. Protein content of follicular jluid For the initial evaluation of the distribution of the proteins in the follicular fluid samples the concentration of each of the proteins was compared with the concentration of the equivalent protein in normal equine plasma. The relative concentrations of the proteins in follicular fluid and plasma were found to range from a high of 122.9% for antithrombin to a low of 13.9% for a2-M. When the relative concentration of the individual proteins was compared with the molecular weight of the protein there was a significant (P < 0.01) inverse relationship (R = 0.983) between the two values (Fig. 1). The smaller proteins, ranging in size from 52 kDa for antithrombin to 150 kDa for IgG, were those present in the follicular fluid samples at similar levels (P > 0.05) to that found in plasma while two of the proteins, a2-M and fibronectin were present in much lower amounts. Despite the variation in the relative amounts of the individual proteins, the total protein content in the follicular fluid was in the same range as that found in the normal equine plasma (Table 1). The mean ( + SE) value for the total protein in the follicular fluid was 64.9 k 3.0 g 1-l compared with 60.7 _+2.7 g 1-l for plasma. As is found in plasma, approximately 50% of the total protein in the follicular fluid was albumin (Table 1).
96
PA. Gentry et al./Animal Reproduction Science 45 (1996) 91-102
Table 1 Protein and hormonal concentrations in fluid obtained from preovulatory equine ovarian follicles Protein
Concentration (g I-‘)
Hormone
Concentration (t.~gml- ‘)
Total protein Albumin Fibronectin (Y2-macroglobulin
64.9k3.0 30. I + 2.8 1.4f0.2 0.6f0.2
Progesterone Androstendione Estradiol- 17p
60.1 f 24.1 64.9 + 8.6 229.7 f 32.6
Values are reported as mean 5 SE; n = 53.
To determine whether there was any significant alteration in the protein content in fluid aspirated from follicles of various sizes, regardless of animal, linear regression analysis was performed comparing the amount of protein in a sample and the size of the follicle from which the sample was obtained. For the smaller molecular weight proteins, i.e. albumin, IgG, antithrombin and ceruloplasmin, there was no correlation (P > 0.05) between the protein content and the size of the follicle. Similarly, no correlation was found between the total protein content of follicular fluid and the size of an individual follicle (P > 0.05). In contrast, for the larger molecular weight proteins there appeared to be a statistically significant inverse relationship (P < 0.05) between the amount of protein and the size of the follicle (Table 2). When the ratio of either fibronectin or o2-M to albumin was compared with follicular diameter, a similar significant (P < 0.05) correlation was found (Table 2). 3.3. Hormonal composition of follicular fluid The mean values (+ SE) for progesterone, androstenedione and estradiol-17P are shown in Table 1. As expected for preovulatory follicles, estradiol was the predominant hormone in the samples, being present at approximately four-fold higher concentrations than was found for either progesterone or androstenedione. There was no correlation (P > 0.05) between the hormonal content and the size of the follicle for any of the hormones measured. When the follicular fluids were divided into three groups on the
Table 2 Correlations between fibronectin (Fn), the fibronectin/albumin ratio, a2-macroglobulin (a2-M) and the u2-M/albumin ratio with follicle size and the logarithmic estradiol concentration (n = 53)
Diameter Estradiol
Fn
Fn/albumin
a2-M
aZM/albumin
R = 0.342 (P < 0.012) R = 0.377 (P < 0.005)
R = 0.357 (P < 0.009) R = 0.295 f P < 0.032)
R = 0.501 (P < 0.001) R = 0.300 (P < 0.029)
R = 0.497 (P < 0.001) R = 0.212
R, correlation coefficient; P, level of significance.
(P < 0.127)
P.A. Gentry et d/Animal
Reproducrion Science 4.5 (1996) 91-102
97
Fig. 2. Steroid hormone concentrations (mean f SE) in equine ovarian follicular fluid from follicles (n = 53) of various sizes: P4, progesterone; A 4, androstenedione; E,, estradiol.
basis of follicle diameter a marked difference (P < 0.05) was observed between the estradiol content of the smallest diameter follicles and the other two groups (Fig. 2). The mean estradiol concentration almost doubled in the fluid from follicles 4.0-4.9 cm in diameter compared with the smaller 3.0-3.9 cm follicles. However, the estradiol values were essentially similar in the two larger groups of follicles as were the estradiol/androstenedione ratios (Fig. 2). 3.4. Correlation between protein and hormone content of follicular
fluid
When linear regression analysis was used to compare the relative concentrations of fibronectin and a2-M in the follicular fluid samples with the values for the individual hormones, a significant correlation was found between each of these proteins and the logarithmic value of estradiol (Table 2). The correlation was greater for fibronectin than a2-M with the levels of significance being P < 0.01 and P < 0.03, respectively. The relationship between fibronectin and estradiol was sustained when the ratio of fibronectin/albumin was used rather than the absolute fibronectin values (Table 2). However, when the ratio of cx2-M/albumin was used to compare the relationship between this protein and estradiol, the regression value indicated that there was no statistically significant correlation (P > 0.05, Table 2). Similar statistical relationships were found when the ratios of fibronectin or albumin to total protein were compared with the logarithmic value for estradiol (data not shown). The relative concentration of albumin, IgG, antithrombin and ceruloplasmin remained constant among the follicular fluids from the various follicles and no correlation with hormonal content was found (P > 0.05, data not shown).
98
P.A. Gentry et al./Animal Reproduction Science 45 (1996) 91-102
4. Discussion Of the proteins evaluated in this study, fibronectin is the only one that has been shown unequivocally to be synthesized by granulosa cells (Skinner and Dorrington, 1984; Lobb and Dorrington, 1987; Reinthaller et al., 1990; Asem et al., 1992; Lehwald and Gentry, 1992). Fibronectin synthesis appears not to have been studied in equine granulosa cells, however, the ability of granulosa cell cultures in species ranging from human beings to chickens to do so suggests that fibronectin production is one of the properties of these cells in the preovulatory follicle. Although a2-M has been found in fluid from rat preovulatory follicles, no mRNA for a2-M could be detected in the granulosa cells prior to ovulation (Curry et al., 1990). In the present study, the finding of a strong correlation between fibronectin levels and both follicular size and estradiol concentration of equine follicular fluid supports the conjecture that fibronectin may act locally on the growth and differentiation of somatic cells (Dorrington and Skinner, 1986). The results suggest that the increase in fibronectin found in equine follicular fluid, like the changes reported in human follicular fluid, are related to the degree of differentiation of granulosa cells (Tsuiki et al., 1988). It has been shown that fibronectin is important in embryogenesis (Samuel et al., 1994). By binding to cells through integrin receptors fibronectin provides positional information for migrating and differentiating cells (Akiyama et al., 1989). Fibronectin is also involved in the organization of the extracellular matrix via its interactions with collagens, proteoglycans and laminins (Ahumada et al., 1981). It is possible that fibronectin may have similar functions in the growing follicle in the mare. The overall values for the estradiol-17l3, progesterone and androstenedione content found for the preovulatory equine follicles were within the range of values previously reported (Short, 1962; Meinecke et al., 1987; Watson and Hinricks, 1988). The changes in concentrations of estradiol-17P found are in general agreement with previous observations that, in the mare, estradiol values increase as the follicle develops and reach maximal values at estrus (Sirois et al., 1990). Similarly, the relatively constant values found for progesterone in this study are compatible with earlier observations and, in contrast to other species, remain relatively constant throughout follicular development (Sirois et al., 1990; Sirois et al., 1991). This absence of change in progesterone may explain, in part, the lack of correlation between fibronectin and progesterone in the equine follicular fluid. A correlation between these two constituents has been found in human follicular fluid where both fibronectin and progesterone increase as the follicle matures (Tsuiki et al., 1988). Ultrasound studies have shown that, in the mare, as the large preovulatory follicles develop they change from a spherical to a conical shape (Pierson and Ginther, 1985). This alteration makes the accurate determination of size based on the measurement of the external diameter of the follicle difficult. Since albumin is a protein that is readily measured and whose concentration was found to be proportional to the follicle size, the relationship between the fibronectin/albumin and oZM/albumin ratios was also compared to follicle size and estradiol concentration. For both fibronectin and o2-M, the relationships between protein content and follicle size were consistent irrespective of
PA. Gentry et al./Animd
Reproduction
Science 45 (1996) 91-102
99
whether the albumin ratios were used. However, when the ratio was compared with estradiol values only the relationship for fibronectin remained consistent with that of the direct fibronectin concentration value. Using the albumin ratio to compare the amount of a specific protein with size and hormone content may provide a more accurate reflection of the interrelationships of constituents in fluid from individual follicles since, in human follicular fluid, it has been shown that there is no correlation between a2-M content and the maturity of the follicle (Gulamali-Majid et al., 1987). The level of antithrombin found in this study is similar to that previously reported for equine and bovine follicular fluids (Yamada and Gentry, 1995a; Yamada and Gentry, 1995b) but appears to be higher than has been found in human follicular fluid (Gulamali-Majid et al., 1987; Nagy et al., 1989; Suchanek et al., 1990). Antithrombin is a member of the serpin superfamily of endogenous proteinase inhibitors that assist in the maintenance of homeostasis by regulating proteolytic enzymes (Pratt and Church, 1991; Potempa et al., 1994). As has been reported for human follicular fluid (Gulamali-Majid et al., 1987; Suchanek et al., 1990), higher concentrations of antithrombin are found in follicular fluid than the other proteinase inhibitor evaluated in this study, cr2-M. Antithrombin has been isolated from porcine follicular fluid and shown to stimulate porcine sperm motility (Lee et al., 1992; Lee et al., 1994). If antithrombin has a similar function in equine reproduction it might be expected to be present in follicular fluid at the relatively high concentrations found in this and the previous study (Yamada and Gentry, 1995b). Although the precise biological role of ceruloplasmin remains to be determined, it has been shown to be an effective scavenger of free radicals and superoxide ions. In this study the amount of ceruloplasmin in the follicular fluid was found to be similar to that of equine plasma. The levels of ceruloplasmin reported in human follicular fluid are about 40% of that found in human plasma (Gulamali-Majid et al., 1987). Whether the difference in the amounts of specific proteins in follicular fluids from different species is related to the species, is the result of the follicles being at different stages of development or is the result of differences in sample handling remains to be resolved. The concentrations of ceruloplasmin, like those of antithrombin and a2-M, have all been reported to be higher in fluids from human follicles containing mature oocytes compared to immature follicles (Gulamali-Majid et al., 1987; Suchanek et al., 1990). In this study no correlation was found between the level of these proteins and either follicle size or hormone content. The relative amounts of albumin and IgG found are consistent with earlier findings that the amounts of albumin, the immunoglobulins and lipoproteins in equine follicular fluid are dependent on the molecular size of the protein (Widders et al., 1984; Legoff, 1994). However, it is important to recognize that factors other than size must also be contributing to the movement of proteins across the follicular wall. For example, only a selective group of plasma coagulation proteins are found in equine follicular fluid. Relative to plasma, the amount of the hemostatic proteins was reported to be 15% for Factor VII, 60% for Factor X while Factor IX was not detected (Yamada and Gentry, 1995b). Factors IX and X are quite similar in molecular structure and size at 56 kDa and 59 kDa, respectively and Factor VII is only marginally smaller at 50 kDa. Hence, it is likely that there may be specific transport mechanisms across the follicle wall for some
100
PA. Gentry et ai./Animal
Reproduction Science 45 (1996) 91-102
proteins and not for others. Clearly, the mechanism controlling the interchange of proteins between mammalian plasma and follicular fluid requires further investigation.
Acknowledgements The expert technical assistance of Eunice Cummings and Michelle Ross is gratefully acknowledged as is the assistance of Dr Peter Ryan for collection of the follicular fluid samples. This research was supported, in part, by the Natural Sciences and Engineering Research Council of Canada and the Ontario Ministry of Agriculture, Food and Rural Affairs.
References Ahumada,G.G.,Rennard,%I., Figueroa, A.A. and Silver, M.H., 1981. Cardiac tibronectin: developmental distribution and quantitative comparison of possible sites of synthesis. J. Mol. Cell. Cardiol., 13: 667-678. Akiyama, S.K., Yamada, S.S., Chen, W.T. and Yamada, K.M., 1989. Analysis of fibronectin receptor function with monoclonal antibodies: roles in cell adhesion, migration, matrix assembly, and cytoskeletal organization. J. Cell Biol., 109: 863-875. Andersen, M.M., Kroll, J., Byskov, A.G. and Faber, M., 1976. Protein composition in the fluid of individual bovine follicles. J. Reprod. Fertil., 48: 109- 118. Asem, E.K., Carnegie, J.A. and Tsang, B.K., 1992. Fibronectin production by chicken granulosa cells in vitro: effect of follicular development. Acta Endocrinol., 127: 466-470. Beers, W.H., 1975. Follicular plasminogen and plasminogen activator and the effect of plasmin on ovarian follicle wall. Cell, 6: 379-386. Chang, S.C.S., Jones, J.D., Ellefson, R.D. and Ryan, R.J., 1976. The porcine ovarian follicle: 1. Selected chemical analysis of follicular fluid at different stages of development. Biol. Reprod., 15: 321-328. Curry, T.E., Mann, J.S., Estes, R.S. and Jones, P.B.C., 1990. cx,-macroglobulin and tissue inhibitor of metalloproteinases: collagenase inhibitors in human preovulatory ovaries. Endocrinology, 127: 63-68. Donington, J.H. and Skinner, M.K., 1986. Cytodifferentiation of granulosa cells induced by gonadotropin-mleasing hormone promotes tibronectin secretion. Endocrinology, 118: 2065-207 1. Edwards, R.G., 1974. Folhcular fluid. J. Reprod. Fertil., 37: 189-219. Gentry, P.A. and Lumsden, J.H., 1978. Determination of serum albumin in domestic animals using the immediate bromcresol green reaction. Vet. Clin. Pathol., 7: 12- 15. Gentry, P.A., Feldman, B.F., O’Neill, S.L., Madigan, J.E. and Zinkl, J.G., 1992. Evaluation of the haemostatic profile in the pre- and post parturient mare, with particular focus on the perinatal period. Equine Vet. J., 24: 33-36. Ginther, O.J., 1992. Reproductive Biology of the Mare: Basic and Applied Aspects. 2nd edn. Equiservices, Wisconsin. Gosden, R.G., Hunter, R.H.F., Telfer, E., Torrance, C. and Brown, N., 1988. Physiological factors underlying the formation of ovarian follicular fluid. J. Reprod. Fertil., 82: 813-825. Gulamali-Majid, F., Ackerman, S., Veeck, L., Acosta, A. and Pleban, P., 1987. Kinetic immunonephelometric determination of protein concentrations in follicular fluid. Clin. Chem., 33: 1185-I 189. Hsueh, A.J.W., Adashi, E.Y., Jones, P.B.C. and Welsh, T.H., 1984. Hormonal regulation of the differentiation of cultured ovarian granulosa cells. Endocrine Rev., 5: 76- 127. Jones, P.B.C., Vernon, M.W., Muse, K.N. and Curry, T.E., 1989. Plasminogen activator and plasminogen activator inhibitor in human preovulatory follicular fluid. J. Clin. Endocrinol. Metab., 68: 1039-1045. Lee, S., Kuo, Y., Kao, C., Hong, C. and Wei, Y., 1992. Purification of a sperm motility stimulator from porcine follicular fluid. Comp. Biochem. Physiol., 101B: 591-594.
P.A. Gentry et al./Animal
Reproduction
Science 45 (1996) 91-102
101
Lee, S., Kao, C. and Wei, Y., 1994. Antithrombin III enhances the motility and chemotaxis of boar sperm. Comp. B&hem. Physiol., 107A: 277-282. Legoff, D., 1994. Foilicular fluid lipoproteins in the mare-evaluation of HDL transfer from plasma to follicular fluid. Biochim. Biophys. Acta, 1210: 226-232. Lehwald, N.A.M. and Gentry, P.A., 1992. Fibronectin production by bovine granulosa cells. Proc. Vth Congress of the International Society of Animal Clinical Biochemistry, 2-6 September 1992, Parma, Italy, p. 131. Lobb, D.K. and Dorrington, J.H., 1987. Human gmnulosa and thecal cells secrete distinct proteins. Fertil. Steril., 48: 243-248. Meinecke, B., Gips, H. and Meinecke-Tillmann, S., 1987. Progestagen, androgen and oestrogen levels in plasma and ovarian follicular fluid during the cestrous cycle in the mare. Anim. Reprod. Sci., 12: 255-265. Nagy, B., Pulay, T., Szarka, G. and Csomor, S., 1989. The serum protein content of human follicular fluid and its correlation with maturity of oocytes. Acta Physiol. Hung., 73: 71-76. Nayudu, P.L., Lopata, A., Leung, P.C.S. and Johnston, W.H.I., 1983. Current problems in human in vitro fertilization and embryo transfer. J. Exp. Zoo]., 228: 203-213. Ny, T., Peng, X.-R. and Ohlsson, M., 1993. Hormonal regulation of the tibrinolytic components in the ovary. Thromb. Res., 71: l-45. Peters, T., 1968. Proposals for standardization of total protein assays. Clin. Chem., 14: 1147-l 159. Pierson, R.A. and Ginther, O.J.. 1985. Ultrasonic evaluation of the preovulatory follicle in the mare. Theriogenology, 24: 259-268. Politis, I., Wang, L., Turener, J.D. and Tsang, B.K., 1990. Changes in tissue-type plasminogen activator-like and plasminogen activator inhibitor activities in granulosa and theta layers during ovarian follicle development in the domestic hen. Biol. Reprod., 42: 747-754. Potempa, J., Korzus, E. and Travis, J., 1994. The serpin superfamily of proteinase inhibitors: structure, function and regulation. J. Biol. Chem., 269: 15957-15960. Pratt, W.S. and Church, F.C., 1991. Antithrombin: structure and function. Semin. Hematol., 28: 3-9. Reich, R., Miskin, R. and Tsafriri, A., 1985. Follicular plasminogen activator: involvement in ovulation. Endocrinology, 116: 516-521. Reinthaller, A., Bieglmayer, C., Kirchheimer, J.C., Christ, G., Deutinger, J. and Binder, B.R., 1990. Plasminogen activator, plasminogen activator inhibitor and tibronectin in human granulosa cells and follicular fluid related of oocyte maturation and intrafollicular gonadotropin levels. Fertil. Steril., 54: 1045-1051. Richards, J.S., Ireland, J.J., Rao, M.C., Bematt, G.A., Midgley, A.R. and Reichert, L.E., 1976. Ovarian follicular development in the rat: hormone receptor regulation by estradiol, follicle stimulating hormone and luteinizing hormone. Endocrinology, 99: 1562-1570. Samuel, J.J., Farhadian, F., Sabri, A., Marotte, F., Robert, V. and Rappaport, L., 1994. Expression of fibronectin during rat fetal and postnatal development: an in situ hybridization and immunohistochemical study. Cardiovasc. Res., 28: 1653- 1661. Shagli, R., Kraicer, P., Rimon, A., Pinto, M. and Soferman, N., 1973. Proteins of human follicular fluid: the blood-follicle barrier. Fertil. Steril., 24: 429-434. Short, R.V., 1962. Steroids in the follicular fluids and the corpus luteum of the mare. A ‘two-cell type’ theory of ovarian steroid synthesis. J. Endocrinol., 24: 59-63. Sirois, J., Kimmich, T.L. and Fortune, J.E., 1990. Developmental changes in steroidogenesis by equine preovulatory follicles: effects of equine LH, FSH and CG. Endocrinology, 127: 2423-2430. Sirois, J., Kimmich, T.L. and Fortune, J.E., 1991. Steroidogenesis by equine preovulatory follicles: relative roles of theta intema and granulosa cells. Endocrinology, 128: 1159- 1166. Skinner, M.K. and Dorrington, J.H., 1984. Control of tibronectin synthesis by rat granulosa cells in culture. Endocrinology, 115: 2029-203 1. Suchanek, E., Mujkic-Klaric, A., Grizeij, V., Simunic, V. and Kopjar, B., 1990. Protein concentration in pre-ovulatory follicular fluid related to ovarian stimulation. Int. J. Gynecol. Obstet., 32: 53-59. Tsuiki, A., Preyer, J. and Hung, T.T., 1988. Fibronectin and glycosaminoglycans in human pmovulatory follicular fluid and their correlation to follicular maturation. Human Reprod., 3: 425-429.
102
Pd.
Gentry et al./Animul
Reproduction
Science 45 (1996) 91-102
Watson, E.D. and Hinricks, K., 1988. Changes in the concentrations of steroids and prostaglandin F in preovulatory follicles of the mare after administration of hCG. J. Reprod. Fertil., 84: 557-561. Widders, P.R., Stokes, CR., David, J.S.E. and Boume, F.J., 1984. Quantitation of the immunoglobulins in reproductive tract secretions of the mare. Res. Vet. Sci., 37: 324-330. Wise, T., 1987. Biochemical analysis of ovine follicular fluid: albumin, total protein, lysosomal enzymes, ions, steroids and ascorbic acid content in relation to follicular size, rank, atresia, classification and day of estrous cycle. J. Anim. Sci., 64: 1153-l 169. Yamada, M. and Gentry, P.A., 1995. Hemostatic profile of bovine ovarian follicular fluid. Can. J. Physiol. Pharmacol., 73: 624-629. Yamada, M. and Gentry, P.A., 1995. The hemostatic profile of equine ovarian follicular fluid. Thromb. Res., 77: 45-54. Yamada, M., Tatemoto, G.L.E.H. and Horiuchi, T., 1991. Fibrinolytic activity of bovine follicular fluid. Bull. Hiroshima Prefectur. Univ.. 3: 210-218.
Fibronectin concentrations correlate with ovarian follicular size and estradiol values in equine follicular fluid Patricia A. Gentry *, Mehri Zareie, Robert M. Liptrap Depurtment ofBiomedicu1 Sciences, Ontario Veterinary College, University of Guelph, Guelph, Ont. NIG 2W1, Cunadu
Accepted 26 February 1996
Abstract The amounts of total protein, albumin, fibronectin, (Y2-macroglobulin (a 2-M), immunoglobulin G, ceruloplasmin and antithrombin were determined in fluids collected from 53 preovulatory equine follicles and compared with the contents of estradiol-17P, progesterone and androstenedione, with follicle size and the amounts of the equivalent proteins in normal equine plasma. The concentration of fibronectin and the fibronectin/albumin ratios increased significantly with follicle size and with follicular estradiol levels. The (Y2-M levels and a 2-M/albumin ratios correlated with follicle size but not with hormone content. Both fibronectin and a 2-M were present in lower amounts in follicular fluid compared with plasma while the other proteins were present in similar amounts. Among the proteins evaluated, there was a positive correlation between the amount of the protein in the follicular fluid and the molecular weight of the protein. Keywords: Horse; Follicular fluid; Fibronectin;
Estradiol;
Proteins
1. Introduction Constituents
of follicular
fluid
are known
to play
an important
role
in follicular
growth, oocyte maturation and ovulation. It has been suggested that the concentrations of specific hormones and proteins in follicular fluid may reflect the physiological status of the follicle (Edwards, 1974; Nayudu et al., 1983; Gulamali-Majid et al., 1987). While the granulosa and theta cells are the major contributors of the steroid hormones present
* Corresponding author. Tel.: (5 19)-824-4120; fax: (519)-767- 1450. 0378-4320/96/S 15.00Copyright PII SO378-4320(96)01554-O
0 1996 Elsevier Science B.V. All rights reserved.
92
PA. Gentry et al./Animal Reproduction Science 45 (1996) 91-102
in follicular fluid the proteins appear to be derived from both the circulating blood as well as from local secretions, principally by the granulosa cells (Gosden et al., 1988). The function of many of the proteins in follicular fluid is thought to be similar to that of plasma. In both human and porcine preovulatory follicles, the colloid osmotic pressure remains relatively constant despite the increase in size and volume of the follicle that occurs during maturation (Gosden et al., 1988). This may be due, in part, to the amounts of total protein and albumin that have been reported in fluid from those follicles (Shagli et al., 1973; Chang et al., 1976; Suchanek et al., 1990). It has been suggested that, as the follicle grows, the follicular wall becomes more permeable to plasma constituents which permits small molecular weight proteins to be filtered more readily from the circulation (Shagli et al., 1973). This concept is supported by observations from several species, including human, equine, porcine and bovine follicles, that the concentration of albumin in preovulatory follicular fluid is similar to that found in the circulation (Andersen et al., 1976; Widders et al., 1984; Wise, 1987; Nagy et al., 1989; Yamada and Gentry, 1995a; Yamada and Gentry, 1995b). Studies of follicular fluids from several species have shown that while the concentration of some proteins corresponds to that of plasma other proteins may be present in higher or lower amounts (Shagli et al., 1973; Chang et al., 1976; Widders et al., 1984; Nagy et al., 1989; Suchanek et al., 1990; Yamada and Gentry; 1995a; Yamada and Gentry, 1995b). The differences in the relative distribution of various proteins in follicular fluid compared with plasma may be a function of both selective transport processes across the follicle wall as well as the result of local synthesis and secretion of a select group of follicular fluid proteins. For example, plasminogen and plasminogen activators have been found in equine follicular fluid samples in higher amounts than those found in normal adult equine plasma (Yamada and Gentry, 1995b). Cell culture studies have shown that human, bovine, porcine and chicken granulosa cells can synthesize and secrete these proteins (Beers, 1975; Reich et al., 1985; Jones et al., 1989; Politis et al., 1990; Reinthaller et al., 1990; Yamada et al., 1991; Ny et al., 1993). Similarly, human, bovine, rat and chicken granulosa cells have been shown to synthesize fibronectin, a protein associated with cell adhesion, proliferation, differentiation and cell migration (Skinner and Dorrington, 1984; Lobb and Dorrington, 1987; Reinthaller et al., 1990; Asem et al., 1992; Lehwald and Gentry, 1992). The early maturation and growth of the follicle is, in part, controlled by the follicle stimulating hormone (FSH) released from the pituitary gland which produces an increase in the ability of the granulosa cells to convert androgens to estrogen (Hsueh et al., 1984). With rising luteinizing hormone (LH) levels, the theta cells convert more cholesterol to androgens thereby supplying the granulosa cells with the necessary precursors for increased estrogen synthesis (Short, 1962; Hsueh et al., 1984; Ny et al., 1993). The increased estrogen levels in the follicle in turn help stimulate the proliferation of the granulosa cells. The ovulatory LH surge also increases the production of progesterone by the highly differentiated granulosa cells (Richards et al., 1976). Despite the ability of granulosa cells to secret both steroids and proteins into the follicular fluid in the preovulatory follicle, there appears to be little information in the literature correlating the functions of these two groups of constituents in equine preovulatory follicles. Hence, one of the objectives of this study was to determine the levels of progesterone,
P.A. Gentry
et
al./Animal ReproducrionScience 45 (1996) 91-102
93
androstenedione and estradiol in equine follicular fluid samples and to determine whether there was any correlation between hormonal values and those of follicular proteins. The proteins selected for evaluation included: albumin, immunoglobulin (IgG), ceruloplasmin and antithrombin (AT), that have low molecular weights and the higher molecular weight proteins, fibronectin and (Y2-macroglobulin (a 2-M). Based on studies of follicular fluid from other species, these proteins have been detected in follicular fluid, although not all have been reported previously in equine follicular fluid. Using this range of proteins, the association between protein content and molecular size could be examined in samples collected from preovulatory follicles.
2. Materials and methods 2.1. Preparation
of follicular fluid
Ovaries were collected at slaughter from mares of unknown reproductive histories during May and June, that is during the physiological breeding season. The ovaries were transported to the laboratory in 10 mM phosphate buffered saline (PBS, pH 7.2) at 39°C in an insulated container. The ovaries were visually examined and the diameter of each follicle measured. In nonspherical follicles, the width was measured in two dimensions at right angles and the average value reported as the diameter. Follicular fluid from individual follicles was aspirated and transferred to a plastic tube which was then centrifuged at 3000 X g for 15 min at 4°C. The supematant was transferred, in several aliquots, to clean vials and frozen at - 20°C. 2.2. Preparation
of plasma
Blood was collected from 15 healthy mares, by venepuncture of the external jugular vein, into a plastic syringe containing 0.17 M sodium citrate in the proportion nine parts blood to one part anticoagulant. The mares were maintained at the University of Guelph research farm. The titrated blood was transferred to plastic centrifuge tubes, centrifuged at 3000 X g for 20 min at 4°C and the top two-thirds of the resulting platelet poor plasma (PPP> removed. The PPP from each animal was pooled and the resulting sample stored in small aliquots at -20°C. 2.3. Protein determinations The total protein and the albumin content of each of the follicular fluid samples and the pooled equine plasma were evaluated with the Biuret (Peters, 1968) and the Bromcresol Green (Gentry and Lumsden, 1978) calorimetric assays, respectively. For each assay a standard curve was prepared with bovine serum albumin (Sigma Chemical Co., St. Louis, MO). Antithrombin values were determined with the chromogenic
94
P.A. Gentry et al./Animal
Reproduction
Science 45 (1996) 91-102
substrate S-2238 (Chromogenix AB, Molandal, Sweden) as previously described for equine plasma (Gentry et al., 1992). Rocket immunoelectrophoresis was used to assess the fibronectin, (Y2-macroglobulin (o2M), ceruloplasmin and immunoglobulin (IgG) levels in the follicular fluid and plasma samples. For fibronectin and o2M, 0.7% solutions of anti-human fibronectin (Calbiochem Co., La Jolla, CA) and anti-human (Y2-M (Kent Labs Inc., Markham, Ont.) were used while 1.5% solutions of anti-human ceruloplasmin (Sigma Chemical Co.) and anti-horse IgG (Sigma Chemical Co.) were used for ceruloplasmin and IgG, respectively. In preliminary examinations the cross-reactivity of these antibody preparations to the specific proteins in equine plasma and follicular fluid had been confirmed. For fibronectin and a2-M, standard curves were prepared with purified human fibronectin (Calbiochem Co., La Jolla, CA) and human ol2-M (Kent Labs Inc., Markham, Ont.), respectively. For all proteins, standard curves prepared from the pooled equine plasma were run concurrently with the follicular fluid samples. 2.4. Hormone assays Follicular fluid aliquots from individual follicles were extracted twice with diethyl ether (Caledon Laboratories Ltd., Georgetown, Ont.> prior to analysis using standard radioimmunoassay procedures to determine concentrations of estradiol- 17B, the estrogen precursor androstenedione and progesterone. Each sample was assayed in duplicate. All radioisotopes were purchased from Amersham Canada Ltd. (Oakville, Ont.) and all non-radioactive steroid standards from Steranti Research Ltd. (St Albans, UK). Antiserum to estradiol (courtesy Dr. J.J. Pratt, Isotopenlaboratorium Academisch, Groningen, Netherlands) has a cross reactivity of 1.32% and 1.23% with estradiol-17cr and estrone, respectively. Androstenedione antiserum (Cedarlane Laboratories, Homby, Ont.) cross reacts 4.0% with androsterone and 2.05% with both testosterone and dehydroepiandrosterone. Antibody to progesterone (UCB Bioproducts SA, Belgium) cross reacts 20% with pregnenolone and under 0.5% with other steroids. 2.5. Statistical analysis Statistical analyses were carried out using a SigmaStat computer program (Jandel Scientific, San Rafael, CA).
3. Results 3.1. Selection of follicles
Only fluid aspirated from follicles between 3.0 and 5.9 cm in diameter were included in this study since these represent growing follicles with the potential to ovulate (Pierson
Psi. Gentry et al./Animal
Reproduction
Science 45 (1996) 91-102
95
Fig. 1. Relationship of the concentration (mean+ SE) of proteinsin equine follicular fluid to their respective molecular weights: AT, antithrombin III; Alb, albumin; Cerulo, ceruloplasmin; IgG, immunoglobulin G; Fn, fibronectin; cx2-M. o(2-macroglobulin.
and Ginther, 1985). The estradiol-androstenedione ratio of the follicular fluid samples was used to confirm that the samples had been collected from developing preovulatory follicles (Ginther, 1992). The application of this selection process resulted in 53 samples from individual follicles being available for analysis. 3.2. Protein content of follicular jluid For the initial evaluation of the distribution of the proteins in the follicular fluid samples the concentration of each of the proteins was compared with the concentration of the equivalent protein in normal equine plasma. The relative concentrations of the proteins in follicular fluid and plasma were found to range from a high of 122.9% for antithrombin to a low of 13.9% for a2-M. When the relative concentration of the individual proteins was compared with the molecular weight of the protein there was a significant (P < 0.01) inverse relationship (R = 0.983) between the two values (Fig. 1). The smaller proteins, ranging in size from 52 kDa for antithrombin to 150 kDa for IgG, were those present in the follicular fluid samples at similar levels (P > 0.05) to that found in plasma while two of the proteins, a2-M and fibronectin were present in much lower amounts. Despite the variation in the relative amounts of the individual proteins, the total protein content in the follicular fluid was in the same range as that found in the normal equine plasma (Table 1). The mean ( + SE) value for the total protein in the follicular fluid was 64.9 k 3.0 g 1-l compared with 60.7 _+2.7 g 1-l for plasma. As is found in plasma, approximately 50% of the total protein in the follicular fluid was albumin (Table 1).
96
PA. Gentry et al./Animal Reproduction Science 45 (1996) 91-102
Table 1 Protein and hormonal concentrations in fluid obtained from preovulatory equine ovarian follicles Protein
Concentration (g I-‘)
Hormone
Concentration (t.~gml- ‘)
Total protein Albumin Fibronectin (Y2-macroglobulin
64.9k3.0 30. I + 2.8 1.4f0.2 0.6f0.2
Progesterone Androstendione Estradiol- 17p
60.1 f 24.1 64.9 + 8.6 229.7 f 32.6
Values are reported as mean 5 SE; n = 53.
To determine whether there was any significant alteration in the protein content in fluid aspirated from follicles of various sizes, regardless of animal, linear regression analysis was performed comparing the amount of protein in a sample and the size of the follicle from which the sample was obtained. For the smaller molecular weight proteins, i.e. albumin, IgG, antithrombin and ceruloplasmin, there was no correlation (P > 0.05) between the protein content and the size of the follicle. Similarly, no correlation was found between the total protein content of follicular fluid and the size of an individual follicle (P > 0.05). In contrast, for the larger molecular weight proteins there appeared to be a statistically significant inverse relationship (P < 0.05) between the amount of protein and the size of the follicle (Table 2). When the ratio of either fibronectin or o2-M to albumin was compared with follicular diameter, a similar significant (P < 0.05) correlation was found (Table 2). 3.3. Hormonal composition of follicular fluid The mean values (+ SE) for progesterone, androstenedione and estradiol-17P are shown in Table 1. As expected for preovulatory follicles, estradiol was the predominant hormone in the samples, being present at approximately four-fold higher concentrations than was found for either progesterone or androstenedione. There was no correlation (P > 0.05) between the hormonal content and the size of the follicle for any of the hormones measured. When the follicular fluids were divided into three groups on the
Table 2 Correlations between fibronectin (Fn), the fibronectin/albumin ratio, a2-macroglobulin (a2-M) and the u2-M/albumin ratio with follicle size and the logarithmic estradiol concentration (n = 53)
Diameter Estradiol
Fn
Fn/albumin
a2-M
aZM/albumin
R = 0.342 (P < 0.012) R = 0.377 (P < 0.005)
R = 0.357 (P < 0.009) R = 0.295 f P < 0.032)
R = 0.501 (P < 0.001) R = 0.300 (P < 0.029)
R = 0.497 (P < 0.001) R = 0.212
R, correlation coefficient; P, level of significance.
(P < 0.127)
P.A. Gentry et d/Animal
Reproducrion Science 4.5 (1996) 91-102
97
Fig. 2. Steroid hormone concentrations (mean f SE) in equine ovarian follicular fluid from follicles (n = 53) of various sizes: P4, progesterone; A 4, androstenedione; E,, estradiol.
basis of follicle diameter a marked difference (P < 0.05) was observed between the estradiol content of the smallest diameter follicles and the other two groups (Fig. 2). The mean estradiol concentration almost doubled in the fluid from follicles 4.0-4.9 cm in diameter compared with the smaller 3.0-3.9 cm follicles. However, the estradiol values were essentially similar in the two larger groups of follicles as were the estradiol/androstenedione ratios (Fig. 2). 3.4. Correlation between protein and hormone content of follicular
fluid
When linear regression analysis was used to compare the relative concentrations of fibronectin and a2-M in the follicular fluid samples with the values for the individual hormones, a significant correlation was found between each of these proteins and the logarithmic value of estradiol (Table 2). The correlation was greater for fibronectin than a2-M with the levels of significance being P < 0.01 and P < 0.03, respectively. The relationship between fibronectin and estradiol was sustained when the ratio of fibronectin/albumin was used rather than the absolute fibronectin values (Table 2). However, when the ratio of cx2-M/albumin was used to compare the relationship between this protein and estradiol, the regression value indicated that there was no statistically significant correlation (P > 0.05, Table 2). Similar statistical relationships were found when the ratios of fibronectin or albumin to total protein were compared with the logarithmic value for estradiol (data not shown). The relative concentration of albumin, IgG, antithrombin and ceruloplasmin remained constant among the follicular fluids from the various follicles and no correlation with hormonal content was found (P > 0.05, data not shown).
98
P.A. Gentry et al./Animal Reproduction Science 45 (1996) 91-102
4. Discussion Of the proteins evaluated in this study, fibronectin is the only one that has been shown unequivocally to be synthesized by granulosa cells (Skinner and Dorrington, 1984; Lobb and Dorrington, 1987; Reinthaller et al., 1990; Asem et al., 1992; Lehwald and Gentry, 1992). Fibronectin synthesis appears not to have been studied in equine granulosa cells, however, the ability of granulosa cell cultures in species ranging from human beings to chickens to do so suggests that fibronectin production is one of the properties of these cells in the preovulatory follicle. Although a2-M has been found in fluid from rat preovulatory follicles, no mRNA for a2-M could be detected in the granulosa cells prior to ovulation (Curry et al., 1990). In the present study, the finding of a strong correlation between fibronectin levels and both follicular size and estradiol concentration of equine follicular fluid supports the conjecture that fibronectin may act locally on the growth and differentiation of somatic cells (Dorrington and Skinner, 1986). The results suggest that the increase in fibronectin found in equine follicular fluid, like the changes reported in human follicular fluid, are related to the degree of differentiation of granulosa cells (Tsuiki et al., 1988). It has been shown that fibronectin is important in embryogenesis (Samuel et al., 1994). By binding to cells through integrin receptors fibronectin provides positional information for migrating and differentiating cells (Akiyama et al., 1989). Fibronectin is also involved in the organization of the extracellular matrix via its interactions with collagens, proteoglycans and laminins (Ahumada et al., 1981). It is possible that fibronectin may have similar functions in the growing follicle in the mare. The overall values for the estradiol-17l3, progesterone and androstenedione content found for the preovulatory equine follicles were within the range of values previously reported (Short, 1962; Meinecke et al., 1987; Watson and Hinricks, 1988). The changes in concentrations of estradiol-17P found are in general agreement with previous observations that, in the mare, estradiol values increase as the follicle develops and reach maximal values at estrus (Sirois et al., 1990). Similarly, the relatively constant values found for progesterone in this study are compatible with earlier observations and, in contrast to other species, remain relatively constant throughout follicular development (Sirois et al., 1990; Sirois et al., 1991). This absence of change in progesterone may explain, in part, the lack of correlation between fibronectin and progesterone in the equine follicular fluid. A correlation between these two constituents has been found in human follicular fluid where both fibronectin and progesterone increase as the follicle matures (Tsuiki et al., 1988). Ultrasound studies have shown that, in the mare, as the large preovulatory follicles develop they change from a spherical to a conical shape (Pierson and Ginther, 1985). This alteration makes the accurate determination of size based on the measurement of the external diameter of the follicle difficult. Since albumin is a protein that is readily measured and whose concentration was found to be proportional to the follicle size, the relationship between the fibronectin/albumin and oZM/albumin ratios was also compared to follicle size and estradiol concentration. For both fibronectin and o2-M, the relationships between protein content and follicle size were consistent irrespective of
PA. Gentry et al./Animd
Reproduction
Science 45 (1996) 91-102
99
whether the albumin ratios were used. However, when the ratio was compared with estradiol values only the relationship for fibronectin remained consistent with that of the direct fibronectin concentration value. Using the albumin ratio to compare the amount of a specific protein with size and hormone content may provide a more accurate reflection of the interrelationships of constituents in fluid from individual follicles since, in human follicular fluid, it has been shown that there is no correlation between a2-M content and the maturity of the follicle (Gulamali-Majid et al., 1987). The level of antithrombin found in this study is similar to that previously reported for equine and bovine follicular fluids (Yamada and Gentry, 1995a; Yamada and Gentry, 1995b) but appears to be higher than has been found in human follicular fluid (Gulamali-Majid et al., 1987; Nagy et al., 1989; Suchanek et al., 1990). Antithrombin is a member of the serpin superfamily of endogenous proteinase inhibitors that assist in the maintenance of homeostasis by regulating proteolytic enzymes (Pratt and Church, 1991; Potempa et al., 1994). As has been reported for human follicular fluid (Gulamali-Majid et al., 1987; Suchanek et al., 1990), higher concentrations of antithrombin are found in follicular fluid than the other proteinase inhibitor evaluated in this study, cr2-M. Antithrombin has been isolated from porcine follicular fluid and shown to stimulate porcine sperm motility (Lee et al., 1992; Lee et al., 1994). If antithrombin has a similar function in equine reproduction it might be expected to be present in follicular fluid at the relatively high concentrations found in this and the previous study (Yamada and Gentry, 1995b). Although the precise biological role of ceruloplasmin remains to be determined, it has been shown to be an effective scavenger of free radicals and superoxide ions. In this study the amount of ceruloplasmin in the follicular fluid was found to be similar to that of equine plasma. The levels of ceruloplasmin reported in human follicular fluid are about 40% of that found in human plasma (Gulamali-Majid et al., 1987). Whether the difference in the amounts of specific proteins in follicular fluids from different species is related to the species, is the result of the follicles being at different stages of development or is the result of differences in sample handling remains to be resolved. The concentrations of ceruloplasmin, like those of antithrombin and a2-M, have all been reported to be higher in fluids from human follicles containing mature oocytes compared to immature follicles (Gulamali-Majid et al., 1987; Suchanek et al., 1990). In this study no correlation was found between the level of these proteins and either follicle size or hormone content. The relative amounts of albumin and IgG found are consistent with earlier findings that the amounts of albumin, the immunoglobulins and lipoproteins in equine follicular fluid are dependent on the molecular size of the protein (Widders et al., 1984; Legoff, 1994). However, it is important to recognize that factors other than size must also be contributing to the movement of proteins across the follicular wall. For example, only a selective group of plasma coagulation proteins are found in equine follicular fluid. Relative to plasma, the amount of the hemostatic proteins was reported to be 15% for Factor VII, 60% for Factor X while Factor IX was not detected (Yamada and Gentry, 1995b). Factors IX and X are quite similar in molecular structure and size at 56 kDa and 59 kDa, respectively and Factor VII is only marginally smaller at 50 kDa. Hence, it is likely that there may be specific transport mechanisms across the follicle wall for some
100
PA. Gentry et ai./Animal
Reproduction Science 45 (1996) 91-102
proteins and not for others. Clearly, the mechanism controlling the interchange of proteins between mammalian plasma and follicular fluid requires further investigation.
Acknowledgements The expert technical assistance of Eunice Cummings and Michelle Ross is gratefully acknowledged as is the assistance of Dr Peter Ryan for collection of the follicular fluid samples. This research was supported, in part, by the Natural Sciences and Engineering Research Council of Canada and the Ontario Ministry of Agriculture, Food and Rural Affairs.
References Ahumada,G.G.,Rennard,%I., Figueroa, A.A. and Silver, M.H., 1981. Cardiac tibronectin: developmental distribution and quantitative comparison of possible sites of synthesis. J. Mol. Cell. Cardiol., 13: 667-678. Akiyama, S.K., Yamada, S.S., Chen, W.T. and Yamada, K.M., 1989. Analysis of fibronectin receptor function with monoclonal antibodies: roles in cell adhesion, migration, matrix assembly, and cytoskeletal organization. J. Cell Biol., 109: 863-875. Andersen, M.M., Kroll, J., Byskov, A.G. and Faber, M., 1976. Protein composition in the fluid of individual bovine follicles. J. Reprod. Fertil., 48: 109- 118. Asem, E.K., Carnegie, J.A. and Tsang, B.K., 1992. Fibronectin production by chicken granulosa cells in vitro: effect of follicular development. Acta Endocrinol., 127: 466-470. Beers, W.H., 1975. Follicular plasminogen and plasminogen activator and the effect of plasmin on ovarian follicle wall. Cell, 6: 379-386. Chang, S.C.S., Jones, J.D., Ellefson, R.D. and Ryan, R.J., 1976. The porcine ovarian follicle: 1. Selected chemical analysis of follicular fluid at different stages of development. Biol. Reprod., 15: 321-328. Curry, T.E., Mann, J.S., Estes, R.S. and Jones, P.B.C., 1990. cx,-macroglobulin and tissue inhibitor of metalloproteinases: collagenase inhibitors in human preovulatory ovaries. Endocrinology, 127: 63-68. Donington, J.H. and Skinner, M.K., 1986. Cytodifferentiation of granulosa cells induced by gonadotropin-mleasing hormone promotes tibronectin secretion. Endocrinology, 118: 2065-207 1. Edwards, R.G., 1974. Folhcular fluid. J. Reprod. Fertil., 37: 189-219. Gentry, P.A. and Lumsden, J.H., 1978. Determination of serum albumin in domestic animals using the immediate bromcresol green reaction. Vet. Clin. Pathol., 7: 12- 15. Gentry, P.A., Feldman, B.F., O’Neill, S.L., Madigan, J.E. and Zinkl, J.G., 1992. Evaluation of the haemostatic profile in the pre- and post parturient mare, with particular focus on the perinatal period. Equine Vet. J., 24: 33-36. Ginther, O.J., 1992. Reproductive Biology of the Mare: Basic and Applied Aspects. 2nd edn. Equiservices, Wisconsin. Gosden, R.G., Hunter, R.H.F., Telfer, E., Torrance, C. and Brown, N., 1988. Physiological factors underlying the formation of ovarian follicular fluid. J. Reprod. Fertil., 82: 813-825. Gulamali-Majid, F., Ackerman, S., Veeck, L., Acosta, A. and Pleban, P., 1987. Kinetic immunonephelometric determination of protein concentrations in follicular fluid. Clin. Chem., 33: 1185-I 189. Hsueh, A.J.W., Adashi, E.Y., Jones, P.B.C. and Welsh, T.H., 1984. Hormonal regulation of the differentiation of cultured ovarian granulosa cells. Endocrine Rev., 5: 76- 127. Jones, P.B.C., Vernon, M.W., Muse, K.N. and Curry, T.E., 1989. Plasminogen activator and plasminogen activator inhibitor in human preovulatory follicular fluid. J. Clin. Endocrinol. Metab., 68: 1039-1045. Lee, S., Kuo, Y., Kao, C., Hong, C. and Wei, Y., 1992. Purification of a sperm motility stimulator from porcine follicular fluid. Comp. Biochem. Physiol., 101B: 591-594.
P.A. Gentry et al./Animal
Reproduction
Science 45 (1996) 91-102
101
Lee, S., Kao, C. and Wei, Y., 1994. Antithrombin III enhances the motility and chemotaxis of boar sperm. Comp. B&hem. Physiol., 107A: 277-282. Legoff, D., 1994. Foilicular fluid lipoproteins in the mare-evaluation of HDL transfer from plasma to follicular fluid. Biochim. Biophys. Acta, 1210: 226-232. Lehwald, N.A.M. and Gentry, P.A., 1992. Fibronectin production by bovine granulosa cells. Proc. Vth Congress of the International Society of Animal Clinical Biochemistry, 2-6 September 1992, Parma, Italy, p. 131. Lobb, D.K. and Dorrington, J.H., 1987. Human gmnulosa and thecal cells secrete distinct proteins. Fertil. Steril., 48: 243-248. Meinecke, B., Gips, H. and Meinecke-Tillmann, S., 1987. Progestagen, androgen and oestrogen levels in plasma and ovarian follicular fluid during the cestrous cycle in the mare. Anim. Reprod. Sci., 12: 255-265. Nagy, B., Pulay, T., Szarka, G. and Csomor, S., 1989. The serum protein content of human follicular fluid and its correlation with maturity of oocytes. Acta Physiol. Hung., 73: 71-76. Nayudu, P.L., Lopata, A., Leung, P.C.S. and Johnston, W.H.I., 1983. Current problems in human in vitro fertilization and embryo transfer. J. Exp. Zoo]., 228: 203-213. Ny, T., Peng, X.-R. and Ohlsson, M., 1993. Hormonal regulation of the tibrinolytic components in the ovary. Thromb. Res., 71: l-45. Peters, T., 1968. Proposals for standardization of total protein assays. Clin. Chem., 14: 1147-l 159. Pierson, R.A. and Ginther, O.J.. 1985. Ultrasonic evaluation of the preovulatory follicle in the mare. Theriogenology, 24: 259-268. Politis, I., Wang, L., Turener, J.D. and Tsang, B.K., 1990. Changes in tissue-type plasminogen activator-like and plasminogen activator inhibitor activities in granulosa and theta layers during ovarian follicle development in the domestic hen. Biol. Reprod., 42: 747-754. Potempa, J., Korzus, E. and Travis, J., 1994. The serpin superfamily of proteinase inhibitors: structure, function and regulation. J. Biol. Chem., 269: 15957-15960. Pratt, W.S. and Church, F.C., 1991. Antithrombin: structure and function. Semin. Hematol., 28: 3-9. Reich, R., Miskin, R. and Tsafriri, A., 1985. Follicular plasminogen activator: involvement in ovulation. Endocrinology, 116: 516-521. Reinthaller, A., Bieglmayer, C., Kirchheimer, J.C., Christ, G., Deutinger, J. and Binder, B.R., 1990. Plasminogen activator, plasminogen activator inhibitor and tibronectin in human granulosa cells and follicular fluid related of oocyte maturation and intrafollicular gonadotropin levels. Fertil. Steril., 54: 1045-1051. Richards, J.S., Ireland, J.J., Rao, M.C., Bematt, G.A., Midgley, A.R. and Reichert, L.E., 1976. Ovarian follicular development in the rat: hormone receptor regulation by estradiol, follicle stimulating hormone and luteinizing hormone. Endocrinology, 99: 1562-1570. Samuel, J.J., Farhadian, F., Sabri, A., Marotte, F., Robert, V. and Rappaport, L., 1994. Expression of fibronectin during rat fetal and postnatal development: an in situ hybridization and immunohistochemical study. Cardiovasc. Res., 28: 1653- 1661. Shagli, R., Kraicer, P., Rimon, A., Pinto, M. and Soferman, N., 1973. Proteins of human follicular fluid: the blood-follicle barrier. Fertil. Steril., 24: 429-434. Short, R.V., 1962. Steroids in the follicular fluids and the corpus luteum of the mare. A ‘two-cell type’ theory of ovarian steroid synthesis. J. Endocrinol., 24: 59-63. Sirois, J., Kimmich, T.L. and Fortune, J.E., 1990. Developmental changes in steroidogenesis by equine preovulatory follicles: effects of equine LH, FSH and CG. Endocrinology, 127: 2423-2430. Sirois, J., Kimmich, T.L. and Fortune, J.E., 1991. Steroidogenesis by equine preovulatory follicles: relative roles of theta intema and granulosa cells. Endocrinology, 128: 1159- 1166. Skinner, M.K. and Dorrington, J.H., 1984. Control of tibronectin synthesis by rat granulosa cells in culture. Endocrinology, 115: 2029-203 1. Suchanek, E., Mujkic-Klaric, A., Grizeij, V., Simunic, V. and Kopjar, B., 1990. Protein concentration in pre-ovulatory follicular fluid related to ovarian stimulation. Int. J. Gynecol. Obstet., 32: 53-59. Tsuiki, A., Preyer, J. and Hung, T.T., 1988. Fibronectin and glycosaminoglycans in human pmovulatory follicular fluid and their correlation to follicular maturation. Human Reprod., 3: 425-429.
102
Pd.
Gentry et al./Animul
Reproduction
Science 45 (1996) 91-102
Watson, E.D. and Hinricks, K., 1988. Changes in the concentrations of steroids and prostaglandin F in preovulatory follicles of the mare after administration of hCG. J. Reprod. Fertil., 84: 557-561. Widders, P.R., Stokes, CR., David, J.S.E. and Boume, F.J., 1984. Quantitation of the immunoglobulins in reproductive tract secretions of the mare. Res. Vet. Sci., 37: 324-330. Wise, T., 1987. Biochemical analysis of ovine follicular fluid: albumin, total protein, lysosomal enzymes, ions, steroids and ascorbic acid content in relation to follicular size, rank, atresia, classification and day of estrous cycle. J. Anim. Sci., 64: 1153-l 169. Yamada, M. and Gentry, P.A., 1995. Hemostatic profile of bovine ovarian follicular fluid. Can. J. Physiol. Pharmacol., 73: 624-629. Yamada, M. and Gentry, P.A., 1995. The hemostatic profile of equine ovarian follicular fluid. Thromb. Res., 77: 45-54. Yamada, M., Tatemoto, G.L.E.H. and Horiuchi, T., 1991. Fibrinolytic activity of bovine follicular fluid. Bull. Hiroshima Prefectur. Univ.. 3: 210-218.