Transcriptional regulation of the interleukin 2 gene by glucocorticoid hormones. Role of steroid receptor and antigen-responsive 5'-flanking sequences

Dexamethasone treatment of the Jurkat T77 cell clone inhibited the enhancing effect of 12-O-tetradecanoylporbol-13-acetate (TPA) and the calcium ionophore A23187 on the interleukin 2 (IL2) mRNA levels and gene transcription from intact nuclei. Dexamethasone treatment of Jurkat T77 cells inhibited the TPA/A23187-dependent activation of the transcription from the transfected pIL2CAT, containing 600 base pairs of the genomic sequences upstream of the coding region of IL2 gene, including the TPA/calcium responsive cis-regulatory elements and promoter sequences, driving the expression of the chloramphenicol acetyltransferase (CAT) gene. Transfection of either Jurkat T77 cell clone or glucocorticoid-resistant Jurkat cells with a human glucocorticoid receptor cDNA under the transcriptional control of the Rous sarcoma virus long terminal repeat (LTR) (pRShGR alpha) significantly increased or induced, respectively, the dexamethasone-mediated inhibition of the TPA/A23187-dependent expression of pIL2-CAT as well as the enhancing effect on the expression of the cotransfected CAT gene under the control of the mouse mammary tumor virus LTR, as a marker of glucocorticoid receptor action. This suggests a role for the glucocorticoid receptor in mediating the dexamethasone action on IL2 gene expression. To study the cis-regulatory sequence specificity of the dexamethasone-induced interference with the TPA/A23187-mediated T cell activating signals, we studied the effect of the hormone on the regulatory elements contained in the Rous sarcoma virus and human T lymphotropic virus 1 long terminal repeats and the SV40 promoter, which are known to be transcriptionally enhanced by those activating agents. Dexamethasone was unable to interfere with the TPA/A23187-mediated enhancement of these cis-regulatory elements, suggesting that the hormone effect is specific for IL-2 gene sequences. Our data suggest that the dexamethasone-mediated transcriptional inhibition of the IL2 gene is mediated by an interference with the protein kinase C and calcium-mediated trans-activation of the antigen-responsive and T cell-specific elements lying in the 5'-flanking region of the gene.

Post-transcriptional control of c-myc proto-oncogene expression by glucocorticoid hormones in human T lymphoblastic leukemic cells

We have studied the regulation of the human c-myc proto-oncogene by glucocorticoid hormones in T lymphoblastic leukemic cells. A significant decrease (50%) of the steady state levels of c-myc mRNA was observed as early as 3 h after dexamethasone treatment of CEM-1.3 human lymphoma cells, reaching less than 5% values, with respect to untreated cells, 24 h after hormone administration. Nuclear run-on experiments showed no modifications of the transcriptional rate from the first exon. However, a slight decrease (15%) of the transcript elongation from the first exon/first intron boundary was observed in the dexamethasone-treated cells. Using actinomycin D to block gene transcription, we have observed a significant increase in the rate of c-myc RNA specific decay after dexamethasone treatment. Furthermore, cycloheximide was able to overcome completely the dexamethasone-induced down-regulation of the c-myc gene expression. Our data suggest that dexamethasone is able to inhibit human c-myc gene expression primarily at the post-transcriptional level, through the synthesis of hormone-induced regulatory protein(s) controlling c-myc transcript stability.

Enhancement of c-erbA proto-oncogene expression by glucocorticoid hormones in S49.1 lymphoma cells

The modifications of the mRNA levels of the c-myc and c-erbA proto-oncogenes during the dexamethasone-induced decrease of S49.1 cell proliferation have been studied. The levels of c-myc mRNA decreased significantly between 3 and 18 h after dexamethasone (1 microM) treatment. In contrast, a significant increase in the levels of a 2.6 kb c-erbA mRNA was observed between 6 and 18 h after hormone treatment. Cycloheximide treatment of S49.1 cells increased the levels of c-erbA RNA and overcome the enhancing effect of dexamethasone on the expression of this proto-oncogene, suggesting that ongoing protein synthesis is necessary to elicit this hormone effect. The associated decrease of cell proliferation and changes in c-myc and c-erbA mRNA levels after dexamethasone treatment suggest that such oncogenes might be involved in the dexamethasone-mediated control of lymphoid cell growth.

Tumor-promoting phorbol ester and ras oncogene expression inhibit the glucocorticoid-dependent transcription from the mouse mammary tumor virus long terminal repeat

Oncogene activation has been suggested to play some role in determining the hormone independency of tumors. In order to study the role of protein kinase C in mediating the inhibition of the glucocorticoid-dependent transcription from the Mouse Mammary Tumor Virus (MMTV)-Long Terminal Repeat induced by overexpressed activated ras oncogene, we studied the effects of protein kinase C activators [the tumor promoting phorbol ester 12-O-tetradecanoyl-phorbol-13-acetate (TPA)] and inhibitors [1-(5-isoquinolinesulfonyl)-2-methylpiperazine (H-7)] on the dexamethasone (DEX)-mediated activation of a MMTV-Long Terminal Repeat-chloramphenicol acetyltransferase (pMMTV-CAT) chimeric reporter gene transiently transfected into NIH-3T3 cells and in Ha-ras-transformed fibroblasts (T24-NIH-3T3). TPA (30 ng/ml) together with DEX (0.1 microM) treatment of NIH-3T3 cells resulted in a significant decrease of CAT activity from pMMTV-CAT, compared to DEX treatment alone. The addition of H-7 (40 microM) was able to overcome the TPA-induced inhibition of DEX-dependent transcription from pMMTV-CAT. DEX-dependent expression of pMMTV-CAT was significantly reduced in T24-NIH-3T3 with respect to wild-type NIH-3T3 cells. Treatment of T24-NIH-3T3 cells with either H-7 or TPA significantly enhanced or decreased, respectively, the DEX-dependent expression of pMMTV-CAT. TPA and/or H-7 did not affect CAT activity from either pMMTV-CAT in the absence of DEX or from CAT gene under the control of the SV40 promoter. Similar glucocorticoid receptor sites and binding affinities were observed in T24-NIH-3T3 or TPA-treated NIH-3T3 cells compared to wild-type untreated cells. Our data suggest that activation of PKC is involved in the reduced transcriptional regulatory activity of glucocorticoid hormone induced by overexpressed Ha-ras oncogene in NIH-3T3 fibroblasts.

Steroid sensitivity of thymocyte subpopulations during intrathymic differentiation. Effects of 17 beta-estradiol and dexamethasone on subsets expressing T cell antigen receptor or IL-2 receptor

We have studied the effects of the steroid hormones, 17 beta-estradiol and dexamethasone, on the relative proportion of thymocyte expression of CD4 (L3T4), CD8 (Ly-2), TCR and IL-2R, used to identify different stages of thymocyte differentiation. After short-term in vivo steroid treatment, a significant decrease in the number and proportion of the CD4+/CD8+, double positive subpopulation was observed in parallel with a proportional increase in the percentage of the CD4+/CD8- single positive, of the CD4-/CD8-, double negative and, to a lesser extent, of the CD8+/CD4- subsets. Either steroid treatment increased the proportion of cells expressing either the epsilon-chain of the CD3 complex and/or the beta-chain of the TCR (beta-TCR) (TCR+/CD3+) and the 55 kDa protein of the IL-2R (IL-2R+), related to the increase of CD4+ SP thymocytes and of DN cells, respectively. Furthermore, the increased proportion of CD3+ cells could also be partially related to the increase of both the CD4+ and DN subsets. A decrease of the effect on either DN/IL-2R+ cells or CD4+ SP cells was selectively observed after long-term treatments with 17 beta-estradiol or DEX, respectively. It is concluded that after short-term administration, different steroid hormones mediate a similar selective depletion of DP, TCR-/CD3-, IL2R- cells presumably in an intermediate stage of differentiation. However, either steroid effects evolve differently in long-term treatment schedules, resulting in different effects on early (DN/IL2R+) and late (SP/IL2R-) steps of thymocyte differentiation.

Steroid sensitivity of thymocyte subpopulations during intrathymic differentiation. Effects of 17 beta-estradiol and dexamethasone on subsets expressing T cell antigen receptor or IL-2 receptor

We have studied the effects of the steroid hormones, 17 beta-estradiol and dexamethasone, on the relative proportion of thymocyte expression of CD4 (L3T4), CD8 (Ly-2), TCR and IL-2R, used to identify different stages of thymocyte differentiation. After short-term in vivo steroid treatment, a significant decrease in the number and proportion of the CD4+/CD8+, double positive subpopulation was observed in parallel with a proportional increase in the percentage of the CD4+/CD8- single positive, of the CD4-/CD8-, double negative and, to a lesser extent, of the CD8+/CD4- subsets. Either steroid treatment increased the proportion of cells expressing either the epsilon-chain of the CD3 complex and/or the beta-chain of the TCR (beta-TCR) (TCR+/CD3+) and the 55 kDa protein of the IL-2R (IL-2R+), related to the increase of CD4+ SP thymocytes and of DN cells, respectively. Furthermore, the increased proportion of CD3+ cells could also be partially related to the increase of both the CD4+ and DN subsets. A decrease of the effect on either DN/IL-2R+ cells or CD4+ SP cells was selectively observed after long-term treatments with 17 beta-estradiol or DEX, respectively. It is concluded that after short-term administration, different steroid hormones mediate a similar selective depletion of DP, TCR-/CD3-, IL2R- cells presumably in an intermediate stage of differentiation. However, either steroid effects evolve differently in long-term treatment schedules, resulting in different effects on early (DN/IL2R+) and late (SP/IL2R-) steps of thymocyte differentiation.

Estrogen and antiestrogen modulation of the levels of mouse natural killer activity and large granular lymphocytes

We investigated the time course of the 17 beta-estradiol effect on mouse natural killer (NK) activity and the number of splenic large granular lymphocytes (LGL), a cell population recently associated with natural cytotoxicity and enriched in low density fractions of Percoll discontinuous density gradients. After 7 days of in vivo treatment with estrogen, an increased cytotoxicity against YAC-1 lymphoma cells was observed using only as effectors cells recovered from higher density fractions, usually devoid of NK activity. In contrast, after a 30-day treatment, augmented NK activity and an increase in LGL number were observed in the lower density Percoll fractions. Similar results were observed after a 30-day treatment with the antiestrogen tamoxifen. The cytotoxicity of both low density and high density splenocyte fractions was totally abrogated by treatment with antiserum to asialo GM1 plus complement, whereas anti-Thy 1.2 antibody treatment only partially decreased the reactivity. Further estrogen administration up to 60 days decreased both NK activity and LGL number. It is concluded that estradiol can induce opposite effects on NK activity depending on the time of treatment, with stimulation of NK activity during the first 30 days after treatment followed by depressed NK activity 1 month later.

Calmodulin antagonism and growth-inhibiting activity of triphenylethylene antiestrogens in MCF-7 human breast cancer cells

The triphenylethylene antiestrogen tamoxifen has been shown previously to inhibit both calmodulin and protein kinase C activities, which are involved in the control of cell proliferation. We have studied the effect of several derivatives of the triphenylethylene antiestrogen family on the inhibition of both calmodulin-dependent cyclic adenosine 3':5'-monophosphate-phosphodiesterase activity and proliferation of breast cancer cells cultured with 0.5 microM estradiol in order to prevent interaction of these drugs with the estrogen receptor. We have observed that hydroxylation of the triphenylethylene molecule significantly decreases its ability to inhibit the calmodulin-dependent phosphodiesterase activity in vitro. Furthermore, the growth-inhibiting activity of several antiestrogens and other calmodulin antagonists [R24571, trifluoperazine, N-(6-aminohexyl)-5-chloronaphthalene-1-sulfonamide, and N-(6-aminohexyl)-1-naphthalenesulfonamide] correlated with their antagonistic effects on calmodulin activity. The level of activity was determined as follows: R24571 greater than tamoxifen = N-demethyltamoxifen = nafoxidine greater than 4-hydroxytamoxifen greater than 3,4-dihydroxytamoxifen = trifluoperazine greater than N-(6-aminohexyl)-5-chloronaphthalene-1-sulfononamide greater than metabolite A greater than N-(6-aminohexyl)-1-naphthalenesulfonamide. On the other hand both protein kinase C-activating and -inhibiting drugs (phorboltetradecanoate-13-acetate and tamoxifen, respectively) have a synergistic inhibitory effect on the growth of MCF-7 cells. Our data suggest that antiestrogen interactions with calmodulin and not protein kinase C may play a role in mediating the drug-induced estrogen-independent inhibition of breast cancer cell growth.

Subcellular and extracellular localization of specific binding sites for triphenylethylene antiestrogens in human breast cancer

MCF-7 human breast cancer cell homogenates and subcellular organelles were submitted to isopycnic centrifugation on Percoll gradients to investigate the subcellular localization of triphenylethylene antiestrogen specific binding sites (AEBS). Electron microscopy revealed that gradient fractions coincident with the migration of [3H]tamoxifen-AEBS complexes were homogeneously represented by rough and smooth endoplasmic reticulum. Eighty percent of AEBS were localized in the endoplasmic reticulum [45,000 +/- 4,000 sites/cell, mean +/- S.D.), while 20% of these sites were also found in the nuclear fraction (12,000 +/- 1,000 sites/cell, mean +/- S.D.). A similar subcellular distribution of AEBS was observed in human breast cancer bioptic specimens. No differences in [3H]tamoxifen binding affinity between microsomal and nuclear AEBS were observed in MCF-7 and bioptic breast cancer. No major differences in microsomal AEBS levels were observed in the limited number of estrogen receptor-positive or -negative breast cancer specimens we have studied, whereas estrogen receptor-negative samples had higher levels of nuclear AEBS with respect to estrogen receptor-positive tumors. The presence of AEBS was also detected in the human serum of healthy and tumor-bearing subjects. The affinity and the binding specificity of serum AEBS were similar to those of intracellular AEBS. No differences in the levels of serum AEBS were observed between healthy and tumor-bearing subjects [19 +/- 4 and 22 +/- 4 pmoles/ml (mean +/- S.D.) respectively. Human serum AEBS did not appear to be associated to lipoproteins, whereas it migrated as a 5.5 S sedimenting molecule.

Estrogen receptors and estrogen sensitivity of fetal thymocytes are restricted to blast lymphoid cells

We have observed [3H]estradiol binding sites in the cytosol of lymphoid cell suspensions obtained from the fetal thymus of guinea pig. Scatchard analysis [dissociation constant (Kd), 0.5 +/- 0.02 (SE) nM], binding specificity, and diethylaminoethyl chromatography of these [3H]estradiol binding sites are similar to those described for estrogen receptors. Estrogen receptor levels in the thymic lymphoid cell population have been studied after fractionating cells by Percoll discontinuous density gradients into large and low density cells (alpha-cells) and small and high density cells (beta-cells). Estrogen receptor levels are higher in alpha-cells [1002 +/- 200 (SE) sites per cell] than beta-cells [61 +/- 6 (SE) sites per cell]. Electron microscopy shows that the ultrastructural characteristics of 95% alpha-cells are consistent with their lymphoblastoid nature, whereas 90% of the beta-cells represented here can be categorized as typical small lymphocytes. To study whether cellular estrogen receptor expression was related to blastogenesis, fetal thymocyte suspensions were cultured throughout 48 h in the presence of the mitogen Concanavalin A. A significant increase in estrogen receptor levels was observed in thymic lymphoid cells cultured in the presence of the mitogen [1500 +/- 193 (SE) sites per cell] with respect to cells cultured in the absence of the mitogen [400 +/- 54 (SE) sites per cell]. One-day and 3-day in vivo estrogen treatments decrease significantly the [3H]thymidine uptake [by 40 +/- 1% (SE) and 66 +/- 5% (SE), respectively], the mitotic index [by 82 +/- 8% (SE) and 96 +/- 2% (SE), respectively] and the frequency of lymphoblastoid cells [by 15 +/- 4% (SE) and 50 +/- 2% (SE), respectively] in fetal thymocyte suspensions. A 29 +/- 0.7% (SE), 31 +/- 0.8% (SE), and 35 +/- 2% (SE) decrease of [3H]thymidine incorporation in the DNA of alpha-cells cultured throughout 24 h, 48 h, and 72 h, respectively, in the presence of estradiol (10 nM) was observed with respect to untreated alpha-cells (P less than 0.01), while no effect of estradiol was observed on [3H]thymidine uptake by beta-cells. The mitotic index of cultured alpha-cells was significantly decreased [by 67 +/- 5% (SE)] 48 h after the addition of estradiol to the culture medium. It is concluded that: 1) estrogen receptors and estrogen sensitivity are restricted to lymphoblastoid thymic cells, and 2) estrogen receptor levels are increased after lymphoid blast transformation.

Differential estrogen and antiestrogen responsiveness of the uterus during development in the fetal, neonatal and immature guinea pig

After 2-day estradiol treatments, wet weight increases in fetuses, newborns and immature guinea pigs by (means +/- SEM) 75 +/- 4%, 170 +/- 16% and 234 +/- 25%, respectively; while after 3-day tamoxifen treatments they are 83 +/- 11%, 157 +/- 35% and 127 +/- 9%, respectively. During the same periods, estradiol increases the uterine content of DNA while the effect of tamoxifen on uterine DNA decreases throughout development. Histologically, both estradiol and tamoxifen induce in the fetus an increase in the size of the stroma and myometrium. Estradiol or tamoxifen, respectively, increase the luminal epithelial cell height by (means +/- SEM) 95 +/- 2% and 67 +/- 2% in fetuses, 286 +/- 20% and 100 +/- 2% in newborns and 260 +/- 10% and 138 +/- 4% in immature animals. Luminal epithelial cell number increases in fetuses, newborns and immature animals by (means +/- SEM) 167 +/- 10%, 248 +/- 50% and 76 +/- 15%, respectively, after estradiol treatments and 160 +/- 20%, 69 +/- 15% and 17 +/- 5%, respectively, after tamoxifen treatments. Uterine epithelial growth invading the stroma was observed in both estradiol- and tamoxifen-treated fetuses. In neonatal or immature animals, estradiol increases the size and the number of endometrial glands, while tamoxifen has progressively less effects on endometrial glands and on the myometrium. It is concluded that: 1) the estradiol-induced uterotropic effect increases progressively in fetal, neonatal and immature animals; and 2) throughout development, tamoxifen has progressively weaker estrogenic properties than estradiol.

Anti-estrogens in fetal and newborn target tissues

The antagonistic effects of progesterone and of the anti-estrogens, tamoxifen and nafoxidine, to estrogen responses were studied in the target tissues of fetal and newborn guinea pigs. In the fetal uterus, progesterone inhibits the stimulatory effect provoked by estradiol on uterine growth, on progesterone receptor and on the acetylation of nuclear histones. Progesterone also blocks the synthesis of new progesterone receptor protein in organ culture. Tamoxifen or nafoxidine (1 or 10 mg/kg/day injected to the mother for 3 days) provoke a uterotrophic effect similar to that of estradiol (1 mg/kg/day injected to the mother for 3 days) but these anti-estrogens have a limited effect on the progesterone receptor. Tamoxifen given together with estradiol antagonizes the effect of the estrogen on the acetylation of histones but the anti-estrogens do not block the effect of estradiol on uterine growth. Histological studies show that both estradiol and tamoxifen provoke a dramatic hypertrophic and hyperplastic effect particularly in the uterine epithelium. In the newborn uterus (6-day old), tamoxifen (s.c. injection of 0.6 micrograms/g body weight) and estradiol (injection of 30 ng/g body weight) provoke a similar uterotrophic effect and both have a limited effect on the progesterone receptor. In the fetal thymus estradiol provokes a selective decrease in the larger and actively proliferating lymphoid cells of the cortical zone. Tamoxifen has a similar effect but to a much lesser extent than estradiol. On the other hand, tamoxifen antagonizes the effect of estradiol on this fetal tissue. It is concluded that during fetal life progesterone antagonizes the effect of estradiol but tamoxifen can act as an agonist or an antagonist of estrogen action which is a function of the type of response or organ considered.

Estrogen and antiestrogen effects on different lymphoid cell populations in the developing fetal thymus of guinea pig

Estrogen responsiveness in the developing fetal thymus of guinea pig has been investigated. One-day estradiol (E2) treatment (1 mg/kg BW) of pregnant animals selectively decreases the number of larger lymphoid cells of the outer cortex of the fetal thymus, without affecting the global number of cortical lymphocytes. E2 treatments for longer periods (from 2 until 6 days) provoke a further impairment in the pattern of outer cortical lymphocytes, selectively reducing larger lymphoid cells, and also decrease the global number of cortical lymphocytes without affecting medullary lymphocytes. Treatment with the antiestrogen tamoxifen (2 mg/kg BW . day for 6 days) results in a transfer of cytoplasmic estrogen receptor to the nuclei of fetal thymus and also affects the histology of the fetal organ, although to a lesser extent than E2. However, when tamoxifen is administered with E2 (1 mg/kg BW . day during the last 3 days of tamoxifen treatment), it antagonizes all of the effects induced by E2 alone on the fetal thymus. The 6-day E2 treatment decreases the weight of the fetal thymus in an age-dependent manner; this effect is progressively more intense from 42 days of gestation at the time of treatment (mean +/- SE, 25 +/- 5% decrease) up to the end of gestation (63 +/- 7% decrease). The effect of E2 on large lymphoid cells also increases in an age-dependent manner. The extent of the estrogen responses observed is correlated with the levels of cytoplasmic estrogen receptors, which increase from 38 days to the end of gestation. The intracellular compartmentalization of estrogen receptor shows that cytoplasmic estrogen receptors are transferred to the nuclei by E2 at both younger and older ages. It is concluded that E2 has differential actions on the lymphoid cell population of the fetal thymus of guinea pig by events antagonized by the antiestrogen tamoxifen and related to the availability and development of cytoplasmic estrogen receptors.

The fetal thymus of guinea pig as an estrogen target organ

[3H]Estradiol ([3H]E2) shows a saturable, high affinity binding (Kd = 0.18 nM) in the cytosol of the guinea pig fetal thymus. The association rate constant at 4 C of [3H]E2 to cytoplasmic receptor is 1.0 X 10(5) M-1 sec-1. The dissociation rate constant of this complex is 4.4 X 10(-6) sec-1 at 4 C and 3.3 X 10(-4) sec-1 at 26 C. Other physicochemical characteristics, such as binding specificity, which is limited to natural and synthetic estrogens, and sedimentation coefficient, which is 8S under low salt conditions and 4S under high salt conditions, are typical for estrogen receptor. The [3H]E2-cytoplasmic receptor complex chromatographed in DEAE-Trisacryl columns is eluted in a continuous 0- to 0.5-M KCl gradient by 0.15- to 0.2-M KCl concentrations. The cytoplasmic [3H]E2-receptor complex can be transferred to the nucleus in vitro by a temperature-dependent process; it can be extracted from the nucleus by 0.4 M KCl as a 5S sedimenting macromolecule. The levels of available cytoplasmic receptors increase during fetal development from 9.7 fmol/mg protein at 36 days of gestation to 31 fmol/mg protein at the end of gestation. Three or six daily injections of E2 or estrone (1 mg/kg BW) to pregnant animals (58-64 days gestation) significantly decreased the weight of the fetal thymus by 50% with respect to values in vehicle-injected animals. Histologically, a reduction in the size of thymic lobules with a decrease in the width of the cortical lymphoid area and an increase in the medulla-cortex ratio was observed. The in vitro incorporation of [3H]thymidine into thymic DNA was reduced by 50% after E2 treatment. The same estrogen treatment also induced a decrease in cytoplasmic estrogen receptors by 63-72% and a concomitant 5- to 6-fold increase in nuclear estrogen receptors, with respect to values in vehicle-injected animals. It is concluded that estrogen receptors are present in the fetal thymus of the guinea pig, and that this fetal organ is estrogen responsive.