The stimulation of uterine complement component C3 gene expression by antiestrogens

SWI/SNF Antagonism of PRC2 Mediates Estrogen-Induced Progesterone Receptor Expression

Endometrial cancer (EC) is characterized by high estrogen levels unopposed by progesterone. Treatment with progestins is standard for early EC, but the response to progestins is dependent on progesterone receptor (PGR) expression. Here, we show that the expression of PGR in endometrial epithelial cells is dependent on ARID1A, a DNA-binding subunit of the SWI/SNF chromatin-remodeling complex that is commonly mutated in EC. In endometrial epithelial cells with estrogen receptor overexpression, we find that ARID1A promotes estrogen signaling and regulates common gene expression programs. Normally, endometrial epithelial cells expressing estrogen receptors respond to estrogen by upregulating the PGR. However, when ARID1A expression is lost, upregulation of PGR expression is significantly reduced. This phenomenon can also occur following the loss of the SWI/SNF subunit BRG1, suggesting a role for ARID1A- and BRG1-containing complexes in PGR regulation. We find that PGR is regulated by a bivalent promoter, which harbors both H3K4me3 and H3K27me3 histone tail modifications. H3K27me3 is deposited by EZH2, and inhibition of EZH2 in the context of ARID1A loss results in restoration of estrogen-induced PGR expression. Our results suggest a role for ARID1A deficiency in the loss of PGR in late-stage EC and a therapeutic utility for EZH2 inhibitors in this disease.

Estrogens regulate neuroinflammatory genes via estrogen receptors α and β in the frontal cortex of middle-aged female rats

Background

Estrogens exert anti-inflammatory and neuroprotective effects in the brain mainly via estrogen receptors α (ERα) and β (ERβ). These receptors are members of the nuclear receptor superfamily of ligand-dependent transcription factors. This study was aimed at the elucidation of the effects of ERα and ERβ agonists on the expression of neuroinflammatory genes in the frontal cortex of aging female rats.

Methods

To identify estrogen-responsive immunity/inflammation genes, we treated middle-aged, ovariectomized rats with 17β-estradiol (E2), ERα agonist 16α-lactone-estradiol (16α-LE2) and ERβ agonist diarylpropionitrile (DPN), or vehicle by Alzet minipump delivery for 29 days. Then we compared the transcriptomes of the frontal cortex of estrogen-deprived versus ER agonist-treated animals using Affymetrix Rat230 2.0 expression arrays and TaqMan-based quantitative real-time PCR. Microarray and PCR data were evaluated by using Bioconductor packages and the RealTime StatMiner software, respectively.

Results

Microarray analysis revealed the transcriptional regulation of 21 immunity/inflammation genes by 16α-LE2. The subsequent comparative real-time PCR study analyzed the isotype specific effects of ER agonists on neuroinflammatory genes of primarily glial origin. E2 regulated the expression of sixteen genes, including down-regulation of complement C3 and C4b, Ccl2, Tgfb1, macrophage expressed gene Mpeg1, RT1-Aw2, Cx3cr1, Fcgr2b, Cd11b, Tlr4 and Tlr9, and up-regulation of defensin Np4 and RatNP-3b, IgG-2a, Il6 and ER gene Esr1. Similar to E2, both 16α-LE2 and DPN evoked up-regulation of defensins, IgG-2a and Il6, and down-regulation of C3 and its receptor Cd11b, Ccl2, RT1-Aw2 and Fcgr2b.

Conclusions

These findings provide evidence that E2, 16α-LE2 and DPN modulate the expression of neuroinflammatory genes in the frontal cortex of middle-aged female rats via both ERα and ERβ. We propose that ERβ is a promising target to suppress regulatory functions of glial cells in the E2-deprived female brain and in various neuroinflammatory diseases.

Rat uterine complement C3 expression as a model for progesterone receptor modulators: characterization of the new progestin trimegestone

Progestins have a wide variety of activities in female reproduction. There are also pharmacological applications for progestins, including hormone replacement therapy and contraception. Here we report the development and characterization of the rat uterine complement component C3 mRNA as a molecular target for the evaluation of the antiestrogenic activity of progestins in the uterus. In this assay, ethinyl estradiol (EE) is used to stimulate C3 expression and progestins are then evaluated for their ability to inhibit this expression. The three reference progestins, progesterone (P4), levonorgestrel (LNG), and medroxyprogesterone acetate (MPA) blocked the increase in C3 mRNA levels induced by EE. Dexamethasone (DEX) and 17alpha-methyl testosterone did not inhibit the estrogen induced C3 mRNA levels; in fact, DEX caused a further increase in C3 mRNA levels. Finally, the antiprogestin RU486 was able to block the MPA inhibition of C3 message. RU486, like DEX, caused an increase in C3 mRNA levels above that of estrogen treatment alone. The model was also used to evaluate trimegestone (TMG), a new steroidal progestin, that has been shown to be a potent and selective progesterone receptor agonist. The activity of TMG in the rat uterine decidualization and ovulation inhibition assays was similar to MPA. However, in the C3 model, TMG caused a dose-dependent inhibition of the EE induced C3 message and was approximately five-fold more potent in this model than MPA (EC(50) of 4.7 microg/kg and 26.5 microg/kg, respectively). Therefore, TMG was a more potent antagonist of estrogenic activity in the uterine endometrium than any of the reference progestins tested and therefore may be more effective in protecting the endometrium in hormone replacement therapy.

Regulation by retinoic acid of acylation-stimulating protein and complement C3 in human adipocytes

Acylation-stimulating protein (ASP), a product of complement C3, stimulates triacylglycerol synthesis in adipocytes. Previous studies have identified transthyretin, associated with chylomicrons, as a stimulator of C3 and ASP production. Since both transthyretin and chylomicrons transport retinyl ester/retinol, our goal was to investigate whether retinoic acid (RA) could be a potential hormonal mediator of the effect. Inhibitors of protein synthesis and protein secretion eliminated the stimulatory effects of chylomicrons on both C3 and ASP production in human differentiated adipocytes, suggesting that de novo protein synthesis and secretion are both required. Incubation with chylomicrons increased C3 mRNA levels (37+/-1.5%). RA alone or with chylomicrons had a stimulatory effect on C3 production (29-fold at 16.6 nM RA) and ASP production. An RA receptor antagonist blocked stimulation of C3 mRNA and C3 secretion by both RA and chylomicrons. Finally, RA and chylomicrons activated a 1.8 kb C3-promoter-luciferase construct transfected into 3T3-F442 and 3T3-L1 cells (by 41+/-0.2% and 69+/-0.3% respectively), possibly via RA receptor half-sites identified by sequence analysis. This is the first evidence documenting stimulation by RA of the C3 gene. Thus we propose RA as a novel cellular trigger in chylomicrons that subsequently results in increased ASP production by adipocytes after a meal.

Degradation of bovine complement C3 by trichomonad extracellular proteinase

Bovine trichomoniasis is a local infection of the reproductive tract making interaction with mucosal host defenses crucial. Since the parasite is susceptible to killing by bovine complement, we investigated the role of the third component of complement (C3) in host parasite interactions. Bovine C3 was purified by anionic and cationic exchange chromatography. The purified protein was characterized by immunoreactivity, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and peptide sequencing of the amino terminus of the beta chain. When purified bovine C3 was incubated for varying time periods with trichomonad extracellular proteinases, SDS-PAGE gels revealed digestion of the alpha chain to small fragments. Such degradation in vivo would prevent formation of C3b and completion of the complement cascade, resulting in evasion of killing. To evaluate the relevance of this data, we determined whether C3 was present in bovine genital secretions. With a quantitative ELISA assay, C3 could be demonstrated in both uterine and vaginal washes. To our knowledge, this is the first demonstration of bovine C3 in genital secretions. The C3 concentration increased significantly in vaginal secretions by 8 and 10 weeks in heifers infected with Tritrichomonas foetus. An increase was also seen in uterine secretions of infected heifers, but sample numbers were insufficient for statistical analysis. Transcription of the major extracellular cysteine proteinase (TFCP8) was demonstrated in T. foetus cells from uterine secretions of infected heifers by RT-PCR and Southern blotting. The results indicate that C3 may be important in genital defense and that trichomonad extracellular proteinases may play a role in evasion of complement-mediated killing.

The oestrogen-like effect of 4-hydroxytamoxifen on induction of transforming growth factor alpha mRNA in MDA-MB-231 breast cancer cells stably expressing the oestrogen receptor

Oestrogens and antioestrogens modulate the synthesis of transforming growth factor alpha (TGF-alpha) in breast cancer cells. The purpose of the present report was to examine regulation of TGF-alpha gene expression by oestradiol (E2) and antioestrogens in MDA-MB-231 breast cancer cells transfected with either the wild-type or mutant oestrogen receptor (ER). We recently reported the concentration-dependent E2 stimulation of TGF-alpha mRNA in MDA-MB-231 ER transfectants (Levenson et al, 1997). We now report that 4-hydroxytamoxifen (4-OHT) shows oestrogen-like effects on the induction of TGF-alpha gene expression in our transfectants. Accumulation of TGF-alpha mRNA in response to both E2 and 4-OHT but not in response to the pure antioestrogen ICI 182,780 suggests that E2-ER and 4-OHT-ER complexes can bind to an oestrogen response element (ERE), located in the promoter region of the TGF-alpha gene and can activate transcription of the gene. Surprisingly, no activation of luciferase expression was observed after transient transfection of the TGF-alpha ERE/luciferase reporter constructs. Possible activation of an alternative ER-mediated pathway responsible for the regulation of TGF-alpha gene expression in the ER transfectants is discussed.

Estrogen- and antiestrogen-responsiveness of HEC1A endometrial adenocarcinoma cells in culture

HEC1A endometrial cancer cells express the wild-type form of the estrogen receptor (ER) and 17beta-estradiol (E2) induces proliferation of these cells. In contrast, tamoxifen only causes a minimal increase (<20%) in cell proliferation. In HEC1A cells transiently transfected with the C3-Luc plasmid derived from the complement C3 gene, both E2 and tamoxifen exhibited ER agonist activity and tamoxifen was also a partial antagonist for this response. The relative ER agonist/antagonist activities of E2, tamoxifen and ICI 182,780 were also investigated in HEC1A1 cells transiently transfected with two E2-responsive plasmids, pCATHD-CAT and pCKB-CAT which contain 5'-promoter inserts from the cathepsin D and creatine kinase B genes, respectively. The results showed that E2 and tamoxifen induced reporter gene activity in cells transiently transfected with both constructs. ICI 182,780 exhibited partial ER agonist activity only in cells transiently transfected with pCKB-CAT and antagonized E2-induced reporter gene activity using both the CKB- and CATHD-derived constructs. These results demonstrate that HEC1A endometrial cancer cells are E2-responsive and represent a useful cell culture model for understanding hormone/antihormone-induced endometrial cell responses.

Steroid-induced conformational changes at ends of the hormone-binding domain in the rat glucocorticoid receptor are independent of agonist versus antagonist activity

The underlying molecular mechanism for the expression of agonist versus antagonist activity for a given receptor-steroid complex is still not known. One attractive hypothesis, based on data from progesterone receptors, is that agonist versus antagonist binding induces unique conformations at the C terminus of receptors, which can be detected by the different fragments produced by partial proteolysis. We now report that the determinants of glucocorticoid receptor (GR)-antagonist complex activity are more complex. Steroid binding did cause a conformational change in the GR that was detected by partial trypsin digestion, as described previously (Simons, S. S., Jr., Sistare, F. D., and Chakraborti, P. K. (1989) J. Biol. Chem. 264, 14493-14497). However, there was no uniformity in the digestion patterns of unactivated or activated receptors bound by a series of six structurally different antagonists including the affinity labeling antiglucocorticoid dexamethasone 21-mesylate. A total of four resistant bands were observed on SDS-polyacrylamide gels in the range of 30-27 kDa. Using a series of point mutations and epitope-specific antibodies, it was determined that the 30-kDa species represented the entire C-terminal sequence of amino acids 518-795, whereas the other bands arose from additional N-terminal and/or C-terminal cleavages. Bioassays with GRs containing various point and deletion mutations failed to reveal any C-terminal alterations that could convert antagonists into biologically active agonists. Thus, the presence or absence of C-terminal amino acids of the GR did not uniquely determine either the appearance of smaller trypsin-resistant fragments or the nature of the biological response of receptor-bound antisteroids. When compared with the current model of the ligand-binding domain, which is based on the x-ray structures of the comparable region of thyroid and retinoic acid receptors, the present results suggest that sequences outside of the model structure are relevant for the binding and biological activity of GRs.

Regulation of gene expression of various phase I and phase II drug-metabolizing enzymes by tamoxifen in rat liver

The objective of the present investigation was to evaluate the effect of tamoxifen (TAM) on the gene expression of different phase I and phase II drug-metabolizing enzymes. Groups of male and female F344/NCr rats were administered either corn oil or TAM (2.8 to 45 mg/kg body wt x 14 days) dissolved in corn oil by gavage. An additional group of rats received a diet supplemented with phenobarbital (PB, 500 ppm). Northern blot analyses of total liver RNA were conducted using [32P]-labeled cDNA or oligonucleotide probes coding for different sulfotransferase (ST); UDP-glucuronosyltransferase (UGT), glutathione S-transferase (GST), epoxide hydrolase (EPH) or cytochrome P450 (CYP) mRNA transcripts. In male rats, TAM increased the levels of STel, STa and STpl mRNAs, whereas PB increased only the STel mRNA. In female rats, there was no expression of STel and STHA mRNA in either control or TAM-treated animals. TAM and PB increased UGTBe/p mRNAs in all rats, whereas UGTml mRNA was elevated only in PB-treated animals. EPH mRNA was elevated markedly in all rats treated with TAM and PB, whereas GSTya/ye mRNA was highly increased by PB, but only marginally increased by TAM. Finally, TAM increased CYP3A1 mRNA, and slightly increased CYP2B1 mRNA, whereas PB highly elevated mRNAs for both of these CYP genes. In conclusion, treatments of rats with TAM increased the mRNA levels of many phase I and phase II drug-metabolizing enzymes, and this pleiotypic response to TAM seems to be different from other prototype inducers such as PB or dioxin (TCDD).

Inhibition of estrogen-stimulated growth of uterine leiomyomas by selective estrogen receptor modulators

Uterine leiomyoma is the most frequent gynecologic neoplasm in women. By using a panel of cell lines derived from spontaneous Eker rat leiomyomas, we examined the estrogen-responsive phenotype of these tumor cells. Leiomyoma-derived ELT cell lines proliferated in response to estrogen, and estrogen-induced cell proliferation could be inhibited by the estrogen antagonist ICI 182780 and the selective estrogen-receptor modulators (SERMs) raloxifene and tamoxifen. In addition to inhibiting cell growth, these antagonists also inhibited estrogen-induced increases in progesterone-receptor expression. These data indicate that SERMs such as raloxifene and tamoxifen act as estrogen antagonists in uterine myometrial cells and suggest that this class of compounds may be effective for treatment of this important gynecologic neoplasm.

The rat endometrial adenocarcinoma cell line RUCA-I: a novel hormone-responsive in vivo/in vitro tumor model

The recently established, estrogen receptor positive rat endometrial adenocarcinoma cell line RUCA-I was tested for estrogen responsiveness in vivo and in vitro. In vivo, 10(6) RUCA-I cells were injected subcutaneously into intact, ovariectomized, or ovariectomized, estradiol-substituted syngenic DA/Han rats. All animals developed well differentiated endometrial adenocarcinoma, that had metastasized to peripheral lymph nodes and into the lung. Ovariectomy reduced tumor and lymph node weight, as well as number of lung metastases significantly compared to controls. In another series of experiments, treatment with the pure anti-estrogen ZK 119010 basically gave the same results as seen in ovariectomized animals, whereas tamoxifen treatment had no effect on metastasis of RUCA-I cells. These findings clearly demonstrate the estrogen dependency of growth and metastasis of RUCA-I cells in vivo. In vitro, we assessed the estrogenic and anti-estrogenic potency of various anti-estrogens, thereby investigating their effects on the expression of components of the complement C3 complex as an estradiol-induced protein and on the expression of fibronectin as an estrogen-repressed protein. Evaluating the relative anti-estrogenic potency of these anti-estrogens we found that ICI 164384 and ICI 182780 behaved as complete antagonists in vitro. Tamoxifen, like estradiol, stimulated complement C3 production and repressed fibronectin expression and has to be regarded as an agonist in this particular in vitro system. ZK 119010 if given alone had no significant influence on the biosynthesis of complement C3 and of fibronectin if compared to the unstimulated control. In addition, another estrogen dependent parameter was identified. Estrogen and anti-estrogen treatment affected glycosylation of complement C3 components. After estradiol treatment predominantly the higher glycosylated epitope of complement C3 became detectable, which could be transformed into the low molecular weight epitope by treatment with hyaluronidase. Finally, we compared the anti-proliferative effects of ICI 164384 and of tamoxifen in vitro. Both anti-estrogens slightly inhibited the growth of RUCA-I rat endometrial adenocarcinoma cells. In conclusion, RUCA-I cells represent a powerful, endometrial derived experimental model to test the agonistic and antagonistic properties of anti-estrogens on growth and metastasis in vivo and on gene expression in vitro. The effects of the tested anti-estrogens on gene expression of RUCA-I cells were found to be useful in predicting their effectiveness on tumor growth in vivo.

Contrasting effects of tamoxifen and ICI 182 780 on estrogen-induced calbindin-D 9k gene expression in the uterus and in primary culture of myometrial cells

Antiestrogens have a large range of tissue- and promoter-specific actions, many of which still remain unclear, particularly in the uterus. Thus, we have analyzed the effects of two antiestrogens, tamoxifen (TAM) and ICI 182 780 (ICI) on the uterine estrogen-responsive gene calbindin-D9k (CaBP9k), in the ovariectomized rat uterus, and in primary cultures of myometrial cells. In the ovariectomized rat uterus, estradiol (E2) or E2 plus TAM induced CaBP9k mRNA to the same levels in 6h. Rats given TAM alone had the same mRNA concentration, but maximal induction was obtained later, 12h after injection. ICI alone did not induce CaBP9k gene expression. Rats given E2 plus ICI had low uterine CaBP9k mRNA levels at 6-12h that became undetectable at 24h. Thus ICI has a full antagonistic effect on E2-induced CaBP9k gene. Estradiol receptor (ER) assays showed that TAM had a partial antagonist effect, while ICI had a full antagonist effect on the ER. We also analyzed the effect of TAM and ICI on CaBP9k gene expression in primary cultures of myometrial cells. The effects were similar to those observed in whole uterus. Thus, TAM has mixed effects, being an agonist for CaBP9k gene induction, and an antagonist for ER. ICI antagonizes the effects of E2 on the CaBP9k gene in myometrial cells and in the intact uterus, but in a way that does not involve a decrease in the cellular content of ER. Instead, it interferes with at least one of the events leading to transcriptional activation.

Fibronectin is an estrogen-repressed protein in RUCA-I rat endometrial adenocarcinoma cells

We recently established and characterized two rat endometrial adenocarcinoma cell lines which we called RUCA-I and RUCA-II. Despite high estrogen receptor levels neither cell line responded to estradiol in conventional cell culture conditions on plastic and in the presence of charcoal stripped fetal calf serum. We further demonstrated that culturing of these cells on a reconstituted basement membrane induced the estrogen responsiveness for both proliferation and gene expression. Particularly, the expression of components of the complement C3 system, which represent major estradiol inducible proteins in the rat uterus in vivo, were found to be under the control of estrogens and antiestrogens. In this paper the search for estrogen repressed proteins is reported. For this purpose secretory proteins of RUCA-I cells were metabolically labelled with 35S-methionine and tested for the presence of estrogen-repressed, antiestrogen-inducible protein species. Analyzing cell culture supernatants of RUCA-I cells by polyacrylamide gel electrophoresis under reducing conditions a protein with an apparent size of approx. 250-270 kDa became conspicuous. The formation and secretion of this protein was suppressed by estradiol and induced by the antiestrogen ICI 164384. Gel electrophoresis performed under non-reducing conditions and hyaluronidase digestion showed that this estrogen-repressed protein represents a dimeric glycoprotein. By immunoprecipitation this glycoprotein was identified as fibronectin. Investigations of steady state mRNA levels of fibronectin by rtPCR suggested a post-transcriptional regulation of this molecule by estradiol. This is the first report on repression of components of the extracellular matrix by estradiol and induction by the complete antiestrogen ICI 164384. The consequences of this finding in regard to growth and invasion of endometrial tumors are discussed.

Extracellular matrix induces hormone responsiveness and differentiation in RUCA-I rat endometrial adenocarcinoma cells

We recently described the establishment and the characterization of two rat endometrial adenocarcinoma cell lines which we called RUCA-I and RUCA-II. Despite fairly high estrogen receptor levels neither cell line responded to estradiol in conventional cell culture conditions on plastic and in the presence of serum. A limited hormonal response to the antiestrogen tamoxifen was detectable in RUCA-I but not in RUCA-II cells. To advance our cell culture conditions we plated RUCA-I cells on a layer of reconstituted basement membrane (Harbor Matrix) in the presence of a serum-free defined medium. These cell culture conditions induced hormone responsiveness of RUCA-I cells and permitted a stimulation of proliferation by estradiol. Further, two estradiol-induced secretory proteins with an apparent molecular weight of 115 kD and 60 kD could be identified by SDS-gelelectrophoresis if analyzed under reducing conditions. These proteins migrated as a single band in a non-reducing electrophoresis gel and were identified as components of the complement C3 system. Additionally, our results suggest that the effects of extracellular matrix and hormones on the expression of these proteins are additive. We conclude that processes of functional differentiation are most likely to occur in this in vitro model, particularly since the expression of components of the complement C3 system was under estrogenic control. Complement C3 proteins represent major estradiol-inducible secretory protein of the immature rat uterus in vivo. Culturing RUCA-I cells on top of a layer of reconstituted basement membrane provides a novel tool to study the importance of the extracellular environment on the hormone-induced gene expression in endometrial carcinogenesis in vitro.

Progesterone and dexamethasone inhibition of estrogen-induced synthesis of DNA and complement in rat uterine epithelium: effects of antiprogesterone compounds

Progesterone (P) blocks estrogen induction of cell proliferation and synthesis of complement C3 in the epithelium of the immature rat uterus. Here it is shown that dexamethasone (Dex) exerts a similar inhibitory effect on these two parameters. Furthermore, analysis of the newly synthesized, secreted proteins produced during a 20 h explant culture period showed that not only does the uterus synthesize complement C3 but it is capable of proteolytically cleaving complement into its biologically active peptides. Since large doses of P are required for its inhibitory effects and since P can interact with the glucocorticoid receptor (GR), the role of the GR in mediating the P effect was questioned. Two antiprogesterone compounds with reportedly differing antiglucocorticoid activity, ZK98.734 and RU486, were tested for their ability to antagonize the inhibitory actions of P and Dex. Attenuation of estrogen-induced epithelial DNA synthesis by either P or Dex was fully overcome by concurrent administration of either ZK98.734 or RU486. On the other hand, while either antagonist was effective against the inhibitory action of P on estrogen-induced complement C3 synthesis, only ZK98.734 was fully effective in blocking inhibition by Dex. Thus, unlike its activity in rat hepatic cells, ZK98.734 is a potent antiglucocorticoid in the immature rat uterus. Because of this antiglucocorticoid activity, differential antagonism of the P response was not possible using these two steroid analogs. Although these observations support the notion of GR mediated inhibition of estrogen action in the uterus, they do not answer the question of whether P might act through a GR mediated pathway.

Human estrogen receptor regulation in a yeast model system and studies on receptor agonists and antagonists

An expression system that utilized yeast copper metallothionein promoter and ubiquitin fusion technology to express the human estrogen receptor gene in yeast is described. We have studied the biochemical and transcriptional regulatory properties of the human estrogen receptor. The biochemical properties of the yeast expressed receptors are identical to the receptors isolated from human tissue. Estradiol mediated activation of transcription by the receptor was studied by a reporter beta-galactosidase gene where expression was under the control of estrogen response elements. Using this expression system and a hyperpermeable yeast strain we have studied the effects of various antiestrogens on the regulation of estrogen receptor function. We demonstrate that tamoxifen and ICI 164,384 are capable of binding to the receptor but neither antiestrogen was able to block the estradiol mediated increase in transcription. In fact, both antiestrogens exerted weak agonist activity in this system.

Antagonism of estrogen- and antiestrogen-induced uterine complement component C3 expression by ICI 164,384

The uterus of the immature rat synthesizes and secretes complement component C3 in response to estradiol treatment. This response occurs in the uterine epithelial cells and is also stimulated by several antiestrogens including tamoxifen and LY117018. The administration of a new antiestrogen ICI 164,384 blocked the estradiol as well as the antiestrogen-stimulated increases in uterine weight, epithelial cell height, C3 synthesis and C3 mRNA. ICI 164,384 demonstrated no agonist properties in terms of epithelial cell response as determined by C3 expression.