5Correspondence: Lawrence Espey, Ph.D.

Current evidence supports the hypothesis that the biochemical events of mammalian ovulation are analogous to an acute inflammatory reaction. This study reveals that tumor necrosis factor-stimulated gene-6 (TSG-6), which encodes a member of the superfamily of hyaluronan-binding proteins that is specifically translated in inflammatory reactions, is expressed in ovarian follicles that have been induced to ovulate. Immature Wistar rats were primed with 10 IU eCG sc, and 48 h later the 12-h ovulatory process was initiated by 10 IU hCG, sc. Ovarian RNA was extracted at 0, 2, 4, 8, 12, and 24 h after the primed animals were injected with hCG. The RNA extracts were used for RT-PCR differential display of amplified cDNAs that represented gene expression in the stimulated ovarian tissue. Northern analysis of one of the differentially amplified cDNAs confirmed that it was part of a gene that was substantially up-regulated at 4–8 h after the ovaries had been stimulated by hCG. Sub-cloning and sequence analysis revealed that the cDNA matched the gene for TSG-6. In situ hybridization indicated that the TSG-6 mRNA was primarily located in the cumulus mass and the antral granulosa cells of large ovarian follicles. In conclusion, the data show that expression of TSG-6 is an integral part of the cascade of inflammatory-like changes that occur in an ovulatory follicle in response to a trophic hormone that couples with LH/hCG receptors.

For the past two decades, mammalian ovulation has been characterized as an acute inflammatory reaction that is induced in mature ovarian follicles by a surge in pituitary gonadotropin(s) (1–3). The pro-inflammatory cytokines interleukin-1 (IL-1) and tumor necrosis factor-a (TNF-a ), which are primary mediators of the acute phase response (4), have both been firmly implicated in the ovulatory process (5-7). These two cytokines activate a number of inflammation-related genes, including TNF-stimulated gene-6 (TSG-6), which is reportedly translated only in inflammatory reactions (4, 8). The present report describes the detection by RT-PCR differential display of TSG-6 gene expression in rat ovarian follicles that had been induced to ovulate by treatment with gonadotropin.

TSG-6 is a "link protein" that binds rather specifically to hyaluronan (HA) (4), a glycosaminoglycan that has a central role in the formation and stability of extracellular matrix (9, 10). Interestingly, when HA and TSG-6 exist independently of one another, HA reportedly has a pro-inflammatory effect while the TSG-6 protein exerts an opposite effect (11). Nevertheless, in spite of its potent anti-inflammatory properties, TSG-6 expression is considered to be an integral part of inflammatory processes—presumably affecting the early stages of an acute inflammatory response (4, 11).

The distribution of HA in ovarian follicles has been studied extensively (10, 12). This glycosaminoglycan is thought to have an especially significant role in the expansion of the cumulus cell-oocyte complex (COC) shortly after the ovulatory process has been initiated by a surge in gonadotropin(s). Such hormone action induces a transient, but detectable, increase in the synthesis of HA by mural granulosa cells, along with a substantial increase in HA production by the cumulus granulosa cells (13, 14). In the latter instance, the accompanying expansion of the COC is thought to be important for ovulation and fertilization (12).

It has been reported recently that TSG-6 is also expressed in the COC of mice at approximately the same time as the ovulatory increase in HA synthesis (9, 15). In the present study, this member of the link module superfamily was discovered by PAGE display of RT-PCR products of mRNA extracts from rat ovaries. The results from in situ hybridization show that ovarian expression of TSG-6 mRNA extends beyond the cumulus oopherous and includes the mural granulosa. In addition, the report analyzes the temporal pattern of expression of this link-protein gene, and it assesses the affect of ovulation-inhibiting doses of indomethacin and epostane on ovarian expression of the gene.

Immature Wistar rats were induced to superovulate by equine chorionic gonadotropin (eCG) and hCG treatment as described previously (16). Ovarian RNA was extracted at the peri-ovulatory intervals of 0, 2, 4, 8, 12, and 24 h after hCG. These nucleic acid extracts were used for differential display and for Northern blotting. Epostane (courtesy of Sanofi~ Synthelabo Research, Malvern, PA) and indomethacin (Sigma Chemical Company, St. Louis, MO) were injected s.c., also as described previously (16). These anti-ovulatory agents were administered at 3 h after hCG in doses of 5.0 mg and 1.0 mg, respectively. The ovulation rate in the various experimental animals was determined by a procedure that also has been described previously (16). For the determination of ovulation rate and the extraction of ovarian RNA, rats were killed by exposure to CO₂. The animals were acquired, retained, and used in compliance with the NIH Guide and with the approval of the institutional animal care review committee.

The steps of the differential display were carried out as described previously (16). In brief, RNA was extracted by a standard guanidine isothiocyanate/cesium chloride procedure. RT- PCR was performed using primers from an RNAimage Kit (G506, GenHunter Corporation, Nashville, TN). The specific primer set that yielded differentially expressed cDNA for TSG-6 was 5'-HTTTTTTTTTC-3' and 5'-HGGCTGAC-3', where "H" represents a HindIII restriction site attached to the primers. After extraction and re-amplification of the differentially expressed cDNA, a standard Northern analysis was performed to confirm the ovulation-specific expression of the parent mRNA for TSG-6. The unique cDNA fragment was cloned using a pCR-TRAP Cloning System (P404, GenHunter), and a cloning colony containing the TSG-6 cDNA was identified by secondary Northern analysis. Manual sequencing of the cDNA was performed using a Sequenase Version 2.0 DNA Sequencing Kit (US70770, Amersham Pharmacia biotech, Inc., Piscataway, NJ). In situ hybridization was performed as described previously (16).

Densitometric analysis of the intensity of the signals from the Northern blots were analyzed by the NIH-image program as described previously (16). Numerical data are presented as means ± SEM. The significance of the differences among the six principal time points of 0, 2, 4, 8, 12, and 24 h after hCG were determined by Duncan's multiple range tests after a completely randomized one-way analysis of variance of the means of the groups. The probability value used as the cutoff between "significant" and "not significant" was P = 0.05.

Following RT-PCR, the sub-populations of radioactively labeled cDNAs that were generated from RNA extracts at each of the six stages of the peri-ovulatory period were separated from one another by electrophoresis on a polyacrylamide gel. The autoradiograph of these PAGE results revealed differentially expressed cDNA bands that were only moderately conspicuous, with strongest amplification in the lanes containing PCR products from mRNA that was extracted at 4 and 8 h after hCG, (Figure 1). Therefore, the most intense cDNA band (i.e., the band in the 8-h lane) was excised from the acrylamide gel and re-amplified for use as a probe in Northern analysis.

The Northerns revealed a pattern of mRNA expression during ovulation that was similar to the pattern on the differential display autoradiograph (Figure 2). The strongest signals on the Northerns were from hybridization with mRNA that had been extracted at 4 and 8 h after hCG. By a discretionary precedent in our laboratory, the intensity of the signal from the 8-h lane was arbitrarily set at 100%, and the densities at the other points during the peri-ovulatory period were expressed as fractions of that maximum. Accordingly, the NIH-image program was used to digitize all of the bands on the Northerns, and the ratio of the density of each experimental band to its corresponding b -actin control band was calculated for each lane. Based on five Northern blots, the signal densities at 0, 2, 4, 8, 12, and 24 h after hCG were 0%, 17.0% ± 6,7%, 113.2% ± 33.3%, 100%, 22.7% ± 2.9%, and 8.9% ± 1.3%, respectively. Thus, TSG-6 gene expression was at a maximum at 4-8 h into the ovulatory process, and this expression declined significantly before the follicles began rupturing at 12-14 h after hCG.

After the ovulation-specific expression of the TSG-6 gene had been confirmed by Northern analysis, the cDNA fragment of this gene was cloned and sequenced. The length of the sequence between the primers was 233 bp. The National Center for Biotechnology Information (NCBI) accession number for this fragment is #AF159103. The cDNA fragment is highly homologous to a segment of a gene (NCBI accession #U83903) that has been cloned recently from hCG-stimulated murine COC tissue (Fulop, 1997). Also, the TSG-6 gene fragment is homologous with a gene (NCBI accession #M86381) that is expressed following serum stimulation of quiescent vascular smooth muscle cells from the rabbit (Feng, 1993). Likewise, the fragment is homologous with a gene (NCBI accession #M31165) that is expressed by TNF-treated human FS-4 fibroblasts (Lee, 1992).

For these tests, Northern blots were prepared from RNA that was extracted from control ovaries at 0 and 8 h into the ovulatory process, or extracted from experimental ovaries that were taken at 8 h after hCG from rats that had been treated 5 h earlier with ovulation-inhibiting doses of epostane or indomethacin. As in the Northern blotting tests at the six different intervals during ovulation, the signal density (normalized against the b -actin control) of the 8-h lane was arbitrarily set at 100% (Figure 3). There was no detectable expression of TSG-6 mRNA at 0 h, but substantial expression at 8 h. In animals treated with the anti-ovulatory agent epostane, which blocks progesterone synthesis, the signal density of 74.1 ± 6.3% was not significantly different from the 8-h control value. However, animals treated with the anti-ovulatory agent indomethacin, which blocks prostanoid synthesis, had a TSG-6 mRNA level of 49.5% ± 11.8%, which was significantly below the 8 h control value, but also was significantly above the 0 h control value.

In situ hybridization confirmed the temporal pattern of TSG-6 mRNA expression that was observed in the differential display autoradiograph and the Northern analysis. There was negligible signal at 0-2 h into the ovulatory process, a strong signal at 4-8 h, and only a trace of signal by 12 h after hCG (Figure 4). Hybridization was limited to the granulosa layer (including the cumulus mass) of the follicles, but the intensity of the signal was not distributed evenly. The largest follicles (i.e., the follicles that were sectioned on a plane closest to the maximum diameter of mature follicles) exhibited a more intense signal for TSG-6 mRNA along the innermost granulosa cells located adjacent to the follicular antrum (Figure 5).

A decade ago, TSG-6 was first discovered during a search for TNF-activated genes by differential screening of a cDNA library prepared from TNF-treated human FS-4 fibroblasts (19). Since then, it has been recognized that the TSG-6 gene is activated by TNF-a and IL-1 through a promoter region that includes one AP-1 and two NF-IL6 binding sites (20, 21). Also, the structure of the TSG-6 protein has been characterized in detail (4, 9, 22, 23). TSG-6 is now recognized as a 35 kDa glycoprotein that is one of the shortest members of a superfamily of hyaluronan-binding proteins that include link protein, aggrecan, versican, brevican, neurocan, and CD44, along with TSG-6. The TSG-6 protein is composed of two structural domains: an N-terminal hyaluronin-binding domain consisting of about 100 residues that are characteristic of proteins in the hyaladherin family (4, 9), and a C-terminal CUB domain consisting of about 110 residues that are characteristic of a number of diverse proteins involved in development and differentiation (24, 25). Thus, a basic function of TSG-6 is presumed to be linkage to hyaluronin and stabilization of the extracellular matrix in various tissues during morphogenic processes

In the present study, ovarian TSG-6 mRNA expression increased to a peak within 4 h after the ovulatory process had been initiated by hCG. It remained elevated for an additional 4 h, but then declined by 12 h after hCG to a level that was not significantly different from the 0-h control value. This temporal pattern of expression coincides with the increase in IL-1 and IL-1 receptor transcripts in rat ovarian follicles in vitro (26). In assessing the spatial distribution of TSG-6 mRNA expression, this transcript followed a pattern most comparable to the increase in ovarian HA synthesis after injection of hCG into PMSG-primed mice (13). Not only was TSG-6 expression observed in COC, it was also intense in the mural granulosa cells that were most adjacent to the antrum, but was minimal in the outermost layers of the mural granulosa cells.

The function of TSG-6 in ovulatory follicles has not been established. It is clear that this member of the link-module superfamily is commonly transcribed in response to local elevation in the inflammatory cytokines IL-1 and TNF-a (4, 9). It also appears likely that TSG-6 is only produced in instances of inflammation (8). In view of this information, there is an inclination to deduce that TSG-6 functions to mediate acute inflammatory reactions and participates in some way in extracellular matrix (ECM) degradation associated with inflammatory processes. However, it is not yet clear whether TSG-6 has a destabilizing or a stabilizing effect on the ECM during local tissue remodeling that is characteristic of acutely inflamed tissues (9, 11). The preponderance of the evidence unexpectedly supports a stabilizing role for TSG-6. This glycoprotein binds to HA and thereby stabilizes the ECM (4, 9). Furthermore, there is substantial evidence to show that TSG-6 readily forms a stable complex with inter-a -inhibitor (Ia I), a Kunitz-type serine protease inhibitor in the plasma (4). This combination of TSG-6 and Ia I inhibits plasmin, a serine protease that activates matrix metalloproteinases that degrade ECM during inflammatory reactions (4, 11). Thus, the synergistic action of TSG-6 and Ia I has an anti-inflammatory effect, which suggests a negative feedback role in moderating acute inflammatory responses.

The role of TSG-6 in ovulatory follicles is not yet clear. Since metalloproteinases such as ADAMTS-1 are produced in substantial amounts in the COC and in the stratum granulosum in response to an ovulatory dose of gonadotropin (27, 28), the hypothesis that TSG-6 binds to hyaluronan and interacts with Ia I to protect the COC matrix from degradation is a reasonable deduction (15). Still, the present results show that TSG-6 is also expressed in mural granulosa cells, even though the apical portion of this follicular layer is significantly degraded by the time the follicle ruptures. Therefore, the function of TSG-6 expression by mural granulosa cells has not been determined. The present data show that it is transcribed relatively early during the ovulatory process, and it declines to non-significant levels even before ADAMTS-1 expression has reached its peak at 12 h after hCG (27). Also, the data show that TSG-6 expression is not affected by inhibition of follicular progesterone synthesis, and it is only moderately reduced by inhibition of follicular eicosanoid synthesis. Therefore, TSG-6 appears to have a role in ovulation that is independent of these two well-known mediators of follicular rupture. In any event, the distinct increase in TSG-6 mRNA in the follicle in response to an ovulatory dose of gonadotropin provides additional support for the hypothesis that the biochemical events of ovulation are comparable to an acute inflammatory reaction.

1. Espey LL 1980 Ovulation as an inflammatory reaction: a hypothesis. Biol Reprod 22:73-106

2. Espey LL 1994 Current status of the hypothesis that mammalian ovulation is comparable to an inflammatory reaction. Biol Reprod 50:233-238

3. Espey LL Ovulation. In: Knobil E, Neill JD (eds), Encyclopedia of Reproduction, volume 3. Academic Press, San Diego, pp 605-614

4. Wisniewski HG, Vilcek J 1997 TSG-6: an IL-1/TNF-inducible protein with anti-inflammatory activity. Cytokine Growth Factor Rev 8:143-156

5. Adashi EY 1998 The potential role of interleukin-1 in the ovulatory process: an evolving hypothesis. Mol Cell Endocrinol, 140:77-81

6. Terranova PF, Rice VM 1997 Review: cytokine involvement in ovarian processes. Am J Reprod Immunol 37:50-63

7. Ujioka T, Matsukawa A, Tanaka N, Matsuura K, Yoshinaga M, Okamura H 1998 Analysis of the cytokine interaction among interleukin-1beta, interleukin-8, and interleukin-1 receptor antagonist in the rabbit ovulatory process. Fertil Steril 70:759-765

8. Parkar AA, Kahmann JD, Howat SL, Bayliss MT, Day AJ 1998 TSG-6 interacts with hyaluronan and aggrecan in a pH-dependent manner via a common functional element: implications for its regulation in inflamed cartilage. FEBS Lett 428:171-176

9. Day AJ 1999 The structure and regulation of hyaluronan-binding proteins. Biochem Soc Trans 27:115-121

10. Salustri A, Camaioni A, Di Giacomo M, Fulop C, Hascall VC. 1999 Hyaluronan and proteoglycans in ovarian follicles. Hum Reprod Update 5:293-301

11. Wisniewski HG, Hua JC, Poppers DM, Naime D, Vilcek J, Cronstein BN 1996 TNF/IL-1-inducible protein TSG-6 potentiates plasmin inhibition by inter-alpha-inhibitor and exerts a strong anti-inflammatory effect in vivo. J Immunol 156:1609-1615

12. Yanagishita M 1994 Proteoglycans and hyaluronan in female reproductive organs. In: Proteoglycans. Jolles P, (ed), Birkhauser Verlag, Basel, pp 179-190

13. Salustri A, Yanagishita M, Underhill CB, Laurent TC, Hascall VC 1992 Localization and synthesis of hyaluronic acid in the cumulus cells and mural granulosa cells of the preovulatory follicle. Dev Biol 151:541-551

14. Tirone E, D'Alessandris C, Hascall VC, Siracusa G, Salustri A 1997 Hyaluronan synthesis by mouse cumulus cells is regulated by interactions between follicle-stimulating hormone (or epidermal growth factor) and a soluble oocyte factor (or transforming growth factor beta1). J Biol Chem 272:4787-4794

15. Fulop C, Kamath RV, Li Y, Otto JM, Salustri A, Olsen BR, Glant TT, Hascall VC 1997 Coding sequenece, exon-intron structure and chromosomal localization of murine TNF-stimulated gene 6 that is specifically expressed by expanding cumulus cell-oocyte complexes. Gene 202:95-102

16. Espey LL, Yoshioka S, Russell D, Ujioka T, Vladu B, Skelsey M, Fujii S, Okamura H, Richards JS 2000 Characterization of ovarian carbonyl reductase gene expression during ovulation in the gonadotropin-primed immature rat. Biol. Reprod. 62:390-397

17. Feng P, Liau G 1993 Identification of a novel serum and growth factor-inducible gene in vascular smooth muscle cells. J Biol Chem 268:9387-9392

18. Lee TH, Wisniewski HG, Vilcek J 1992 A novel secretory tumor necrosis factor-inducible protein (TSG-6) is a member of the family of hyaluronate binding proteins, closely related to the adhesion receptor CD44. J Cell Biol 116:545-557

19. Lee TH, Lee GW, Ziff EB, Vilcek J 1990 Isolation and characterization of eight tumor necrosis factor induced gene sequences from human fibroblasts. Mol Cell Biol 10:1982-1988

20. Klampfer L, Lee TH, Hsu W, Vilcek J, Chen-Kiang S 1994 NF-IL6 and AP-1 cooperatively modulate the activation of the TSG-6 gene by tumor necrosis factor alpha and interleukin-1. Mol Cell Biol 14:6561-6569

21. Klampfer L, Chen-Kiang S, Vilcek J 1995 Activation of the TSG-6 gene by NF-IL6 requires two adjacent NF-IL6 binding sites. J Biol Chem 270:3677-3682

22. Neame PJ, Barry FP 1994 The link proteins. EXS 70:53-72

23. Banerji S, Day AJ, Kahmann JD, Jackson DG 1998 Characterization of a functional hyaluronan-binding domain from the human CD44 molecule expressed in Escherichia coli. Protein Expr Purif 14:371-381

24. Bork P, Beckmann G 1993 The CUB domain. A widespread module in developmentally regulated proteins. J Mol Biol 231:539-545

25. Chen L, Zhang H, Powers RW, Russell PT, Larsen WJ 1999 Covalent linkage between proteins of the inter-alpha-inhibitor family and hyaluronic acid is mediated by a factor produced by granulosa cells. J Biol Chem 271:19409-19414

26. Kol S, Donesky BW, Ruutiainen-Altman K, Ben-Shlomo I, Irahara M, Ando M, Rohan RM, Adashi EY 1999 Ovarian interleukin-1 receptor antagonist in rats: gene expression, cellular localization, cyclic variation, and hormonal regulation of a potential determinant of interleukin-1 action. Biol Reprod 61:274-282

27. Espey LL, Yoshioka S, Russell DL, Robker RL, Fujii S, and Richards JS 2000 Ovarian expression of a disintegrin and metalloproteinase with thrombospondin motifs during ovulation in the gonadotropin-primed immature rat. Biol Reprod 62:1090-1095

28. Robker RL, Russell DL, Espey LL, Lydon JP, O'Malley BW, Richards JS 2000 Progesterone-regulated genes in the ovulation process: ADAMTS-1 and Cathepsin L proteases. Proc Natl Acad Sci, USA 97:4689-4694

(Figure 1). Autoradiograph of differentially displayed TSG-6 cDNA (arrow). Note that the cDNA is not visible in the 0-2 h RT-PCR product, and the greatest amplification is at 4-8 h.

(Figure 2). Amount of ovarian TSG-6 mRNA as determined by intensity of Northern blot signals at six intervals of the periovulatory period after hCG administration. Solid dots represent the mean value of three Northern blots prepared from two separate RNA extractions from two batches of rats. The signal density at 8 h was arbitrarily set at 100%. An actual Northern analysis of the TSG-6 cDNA along with its b -actin control are shown below the graph. Note that the pattern of signal intensity is comparable to the RT-PCR amplification shown in Figure 1. "a", significantly different from 0-h control.

(Figure 3). Comparison of the TSG-6 mRNA signal from Northern blots with data on ovulation rate in parallel groups of animals that were treated with either 5 mg epostane (Epo) or 1 mg indomethacin (Indo), administered at 3 h after hCG. Ovarian RNA was extracted at 8 h after hCG treatment because this was near the peak level, and because the data could be compared with the pattern of expression of other ovulation-related genes (16, 27). The bar graphs that quantitate Northern blot data are based on NIH Image analyses of three different Northern blots that were prepared from one RNA extraction from experimental groups consisting of seven rats each, i.e. the RNA extracts were pooled from seven pairs of ovaries. The signal from the 8-h control lane was arbitrarily set at 100% OD to compare the intensities of signals from the four different Northern blots. In parallel groups of rats, the ovulation rate was determined at the optimal time of 24 h after hCG by counting ova in the oviducts. "a", Significantly different from 0-h control; "b", significantly different from the 8 and 24 h controls.

(Figure 4). Change in intensity of the in situ hybridization signal for TSG-6 mRNA during the six periovulatory intervals after hCG administration. Lightfield micrographs on the left show the histology of ovarian sections stained with hematoxylin and eosin, whereas the darkfield micrographs of the same sections show the localization of TSG-6 mRNA as detected by hybridization of a ³⁵S-labeled antisense probe derived from the TSG-6 cDNA. Magnification, ~ X7.0.

(Figure 5). Closer view of the distribution of TSG-6 mRNA probe in the 8-h ovary. Note that the signal is particularly strong in the vicinity of the cumulus mass (white arrow) and the antral granulosa cells. Magnification, ~ X50.0.