Nuclear envelope acts as a calcium barrier in C6 glioma cells

This study evaluated the expression of the corticosteroid-metabolizing enzyme 11 beta-hydroxysteroid dehydrogenase (11 beta HSD) during in vitro decidualization of human endometrial stromal cells. The cultured stromal cells displayed both NADP(+)-dependent (type 1) and NAD(+)-dependent (type 2) 11 beta HSD activities under basal conditions. Although the cells did not respond to estradiol (E2) added alone, catalytic levels of both isoforms were enhanced by medroxyprogesterone acetate (MPA) and further enhanced by E2 plus MPA. Type I messenger RNA (mRNA) was undetected by Northern analysis of total RNA, but was evident as a 1.5-kilobase band in polyadenylated selected RNA from E2- plus MPA-treated cultures. Use of RT-PCR to augment the sensitivity of mRNA detection revealed the presence of type I mRNA as a faint band in the MPA-treated cultures and as an intense band in the E2- plus MPA-treated cultures. Thus, type I mRNA is present as a low abundance message in the cultured stromal cells whose steady state levels parallel progestin-enhanced enzyme activity. As the expression of several progestin-regulated decidualization markers is also augmented by E2, the results of the present study reveal a correlation between enhanced 11 beta HSD expression and the decidualization reaction. Time-course measurements indicated that elevated 11 beta HSD expression is an early event in the decidualization response, which precedes E2- plus MPA-enhanced PRL production by several days. Clear dose-response effects on both type 1 and type 2 11 beta HSD activities were obtained in cells incubated with 10(-8) mol/liter E2 added together with MPA at concentrations that approximated circulating progesterone levels from the luteal phase (10(-9) mol/liter) through pregnancy (10(-7) mol/liter). Corticosteroids are thought to exert toxic and teratogenic effects on the implanting embryo and could influence trophoblast invasion by regulating extracellular matrix turnover. Therefore, the novel finding that decidualization involves marked enhancement of the corticosteroid-metabolizing capacity of stromal cells suggests a mechanism by which decidual cells could affect the health and invasiveness of implanting trophoblastic cells.     

Biological mechanisms underlying the clinical effects of RU 486: modulation of cultured endometrial stromal cell stromelysin-1 and prolactin expression

During in vitro decidualization of human endometrial stromal cells (HESCs), medroxyprogesterone acetate (MPA) inhibits expression of the potent extracellular matrix (ECM)-degrading protease stromelysin-1 (MMP-3), but enhances PRL expression. Consistent with its priming role in vivo, estradiol (E2) augments these effects. In the current study, immunoblot analysis revealed that coincubation with 10(-6) M RU 486 blocked the inhibition in HESC-secreted MMP-3 levels (50,000 mol wt) evoked by 10(-8) M E2 + 10(-7) M MPA. Although MPA can act as a glucocorticoid, the HESCs were refractory to 10(-7) M dexamethasone added alone or with E2. Because E2 elevates progesterone but not glucocorticoid receptor levels, MPA and RU 486 control MMP-3 expression as a progestin and antiprogestin, respectively. To study RU 486 involvement in steroid withdrawal leading to menstruation, HESCs were decidualized during 10 days incubation with E2 + MPA, and parallel cultures were kept in E2 + MPA or withdrawn to either control or RU 486-containing medium. Compared with E2 + MPA-suppressed HESCs, increases in levels of secreted MMP-3 (2.0-fold), and its 2.1-kilobase messenger RNA (10-fold) were observed in HESCs after 4 days of withdrawal to control medium, with much greater increases seen in RU 486-containing medium (10-fold protein, 100-fold messenger RNA). Previously, we showed that RU 486 up-regulated E2 + MPA-inhibited plasminogen activator expression in the cultured HESCs. Extrapolation of these in vitro observations to endometrial events following RU 486 administration suggests that coordinate enhancement of MMP-3 and plasminogen activator expression promotes proteolysis of the stromal/decidual ECM, which leads to endometrial sloughing. Moreover, destabilization of endometrial microvessels resulting from degradation of their surrounding ECM is consistent with the heavy menstrual bleeding stemming from RU 486 administration. However, in contrast to the marked RU 486-initiated reversal of MMP-3 expression, RU 486 did not significantly reverse E2 + MPA-enhanced PRL secretion by the cultured HESCs. Interestingly, decidual PRL, unlike decidual MMP-3, does not appear to play a role in menstruation. Interleukin-1 beta counteracted E2 + MPA-mediated inhibition of secreted MMP-3 levels, implying that leukocyte/trophoblast-derived cytokines can modulate steroid-regulated MMP-3 expression by stromal/decidual cells during menstruation and pregnancy.     

Plasminogen activator activity during decidualization of human endometrial stromal cells is regulated by plasminogen activator inhibitor 1

Progesterone acts on the estradiol (E2)-conditioned human endometrium to induce decidualization of stromal cells. Consistent with these differential hormone actions in vivo, progestins regulate several end points of decidualization in human endometrial stromal cell monolayers, and E2 augments the effects of progestin. This study shows that in vitro decidualization of the stromal cells is accompanied by diminished plasminogen activator (PA) expression. Polyacrylamide gel electrophoretic separation after immunoprecipitation of biosynthetically labeled PAs revealed that medroxyprogesterone acetate (MPA) lowered levels of secreted tissue type PA (tPA) at 67 kilodaltons and urokinase type PA (uPA) at 55 kilodaltons. These levels were reduced further by E2 plus MPA despite a lack of response to E2 alone. Although tPA activity was readily measured by a chromogenic assay, detection of uPA activity required prior activation, indicating that uPA is released as the pro-uPA zymogen. Comparisons of levels of immunogenic PAs, as measured by specific enzyme-linked immunosorbent assays, with the corresponding catalytic activities revealed selective progestational inhibition of PA activity vs. antigen after 3 days of experimental incubation. Thus, 10(-7) mol/L MPA produced about a 2-fold greater reduction of levels of PA activity than that of its corresponding antigen. More strikingly, 10(-8) mol/L E2 plus 10(-7) mol/L MPA virtually eliminated both tPA activity (99% inhibition; P < 0.005) and uPA activity (93% inhibition; P < 0.005); the reductions in levels of the corresponding antigens were only about 50% of the control levels and did not attain statistical significance. Only after 3-6 days of incubation with E2 plus MPA was statistically significant inhibition achieved for immunogenic levels of both tPA (P < 0.05) and uPA (P < 0.005). Preferential inhibition of levels of PA activities compared with those of the corresponding PA antigens reflects the action of the potent PA inhibitor PAI-1. Thus, the concentration of PAI-1 in the stromal cell-conditioned medium at the end of 0-3 days exceeded those of tPA and uPA, respectively, by 28- and 12-fold in response to MPA and by 52- and 25-fold in response to E2 plus MPA. Extrapolation of these in vitro results to the events of the luteal phase, whose steroidal milieu is mimicked by E2 plus MPA, indicates that decidual cell-derived PAI-1 is a key regulator of proteolytic degradation of extracellular matrix and fibrinolysis during implantation and menstruation.     

Localization and sex steroid regulation of androgen receptor gene expression in rhesus monkey uterus

OBJECTIVE: To characterize the cellular sites and hormonal regulation of uterine androgen receptor gene expression in the monkey. METHODS: Ovariectomized rhesus monkeys (five in each group) were treated with placebo (the control group), estradiol (E2), E2 plus progesterone, or E2 plus testosterone by sustained-release pellets administered subcutaneously. After 3 days of treatment, uteri were removed and uterine sections were analyzed by in situ hybridization for androgen receptor messenger RNA (mRNA). RESULTS: Androgen receptor mRNA was detected in endometrial stromal cells and myometrial smooth muscle cells, with lesser expression in endometrial epithelial cells. Both E2 and E2 plus progesterone treatment doubled androgen receptor mRNA levels in stromal cells (P < .01), whereas E2 plus testosterone treatment increased stromal androgen receptor mRNA levels by about five-fold (P < .001) compared with placebo treatment. In the endometrial epithelium, E2 alone did not increase androgen receptor mRNA levels significantly. However, the E2 plus progesterone and E2 plus testosterone treatments increased epithelial androgen receptor mRNA levels by 4.3 and 5 times, respectively (P = .008 and P < .002, respectively). Androgen receptor mRNA was distributed homogeneously in smooth muscle cells across the myometrium. Estradiol treatment alone did not increase myometrial androgen receptor mRNA levels significantly, but the E2 plus progesterone and E2 plus testosterone treatments increased myometrial androgen receptor mRNA levels by 1.8 and 2 times, respectively (P = .001 and P < .001, respectively). CONCLUSION: Androgen receptor gene expression was detected in all uterine cell compartments where it was subject to significant sex steroid regulation. The fact that androgen receptor mRNA levels were consistently up-regulated by a combined E2 plus testosterone treatment while E2 treatment alone had little or no effect shows that a collaborative action of E2 and testosterone enhances androgen receptor expression in the monkey uterus.     

Transforming growth factor-beta inhibits progesterone-induced enkephalinase expression in human endometrial stromal cells

The specific activity of enkephalinase in endometrial tissue of nonpregnant ovulatory women is correlated in a highly significant, positive manner with the plasma level of progesterone. The specific activity and levels of enkephalinase messenger ribonucleic acid and immunoreactive protein also are increased in human endometrial stromal cells in culture by treatment with a synthetic progestin, medroxyprogesterone acetate (MPA), in a time- and dose-dependent manner. From an analysis of the temporal relationship between the specific activity and half-life of enkephalinase in endometrial tissue and the level of progesterone in plasma, it appeared highly likely that some mechanism, in addition to progesterone withdrawal, was operative to reduce enkephalinase activity in endometrium during the late luteal phase of the ovarian cycle before progesterone levels had declined below those known to be effective for progesterone action. In stromal cells previously (and concurrently) treated with MPA (10(-9) mol/L), the addition of transforming growth factor-beta 1 (TGF beta 1) or TGF beta 2 (1 ng/mL) to the medium caused a decrease in enkephalinase specific activity despite the continued presence of MPA. The half-life of enkephalinase (activity) in stromal cells treated with MPA plus TGF beta 1 was 2.8 days, which is similar to the computed half-life for enkephalinase in endometrial tissue during the mid- to late secretory phase of the endometrial cycle (2.5 days). Simultaneous treatment of endometrial stromal cells with MPA (10(-9) mol/L) and TGF beta 1 (1 ng/ mL) prevented the progestin-induced increase in enkephalinase specific activity and immunoreactive enkephalinase protein. Thus, TGF beta acts to oppose the progesterone-induced increase in enkephalinase expression in endometrial stromal cells, even in the continued presence of MPA.     

Matrix metalloproteinase and matrix metalloproteinase inhibitor expression in endometrial stromal cells during progestin-initiated decidualization and menstruation-related progestin withdrawal

Estradiol (E) primes human endometrial stromal cells (HESCs) for the decidualizing effects of progesterone in vivo and in vitro. Matrix metalloproteinase (MMP) expression was evaluated in confluent HESCs incubated in control medium, and in medium supplemented with either E, or the synthetic progestin medroxyprogesterone acetate (P), or E + P. Measurements with a specific ELISA indicated that basal pro-MMP-1 output was unaffected by E, whereas E + P, which induces the expression of several decidualization-related markers, produced a time-dependent inhibition in HESC-secreted levels of pro-MMP-1. Consistent with progestin inhibition of MMP-1 protein expression in the HESCs, P but not E, reduced steady state levels of MMP-1 messenger RNA (mRNA) as determined by Northern analysis. By contrast, mRNA levels for MMP-2 and the MMP inhibitor TIMP-1 were not altered by either P or E. Steroid withdrawal studies indicated that after MMP-1 expression was suppressed by incubation of the HESCs with E + P, 4 days of exposure to the antiprogestin RU 486 (mifepristone) significantly up-regulated MMP-1 levels in the conditioned medium by severalfold compared with cultures maintained in E + P. The change to steroid-free control medium required a more prolonged period of withdrawal to attain up regulatory effects that were comparable with those evoked by RU 486. The ELISA measurements were validated by immunoblot analysis with a specific MMP-1 antibody, which showed corresponding changes in a band at the expected mobility of about 50 kDa. Moreover, Northern analysis revealed parallel changes in MMP-1 mRNA levels, whereas neither MMP-2 nor TIMP-1 mRNA levels were modulated by adding or withdrawing steroids. The contrast between regulated MMP-1 expression and constitutive MMP-2 expression observed in the cultured HESCs is consistent with the demonstrated presence on the MMP-1 promoter of regulatory elements such as AP-1 and PEA-3 that are absent from the MMP-2 promoter. Extrapolation of these in vitro changes in HESCs to in vivo endometrial events suggests that: 1) inhibition of MMP-1 expression by E and progesterone would stabilize the perivascular endometrial ECM to prevent local hemorrhage during endovascular invasion by the implanting trophoblast; 2) enhanced expression of MMP-1 evoked by steroid withdrawal would mediate endometrial ECM degradation leading to sloughing of the functional layer during menstruation.     

Cytokine-mediated regulation of 92-kilodalton type IV collagenase, tissue inhibitor or metalloproteinase-1 (TIMP-1), and TIMP-3 messenger ribonucleic acid expression in human endometrial stromal cells

Interleukin-1 (IL-1) is expressed in human endometrium and has been shown to play an integral role in local cellular interactions during implantation. In addition, the matrix metalloproteinase (MMP) and its inhibitor, the tissue inhibitor of metalloproteinase (TIMP), are crucial during implantation, mediating in vitro trophoblast penetration, and are regulated by several cytokines expressed by trophoblast cells. We have investigated the roles of IL-1 beta and transforming growth factor-beta (TGF beta) in regulating TIMP-1, TIMP-3, and 92-kDa type IV collagenase messenger ribonucleic acid (mRNA) expression in human endometrial stromal cells using quantitative competitive PCR. Confluent stromal cell cultures treated with progesterone and estradiol for 9 days were stimulated with IL-1 beta, IL-1 beta plus anti-IL-1 beta antibody, TGF beta, and TGF beta plus anti-TGF beta antibody for an additional 24 h. Competitive complementary DNA fragments were constructed by deletion of a defined fragment from each of the target complementary DNA sequences and coamplified in quantitative competitive PCR as an internal standard. TIMP-1 and TIMP-3, but not 92-kDa type IV collagenase mRNA, were expressed in stromal cells. The 92-kDa type IV collagenase mRNA was only expressed after stimulation with IL-1 beta. IL-1 beta both augmented 92-kDa type IV collagenase mRNA expression and decreased TIMP-1 and TIMP-3 mRNA expression in a dose-dependent manner. Conversely, TGF beta augmented TIMP-1 and TIMP-3 mRNA expression, but did not affect 92-kDa type IV collagenase expression. IL-1 and TGF beta-mediated changes were both neutralized by specific antibodies. These results provide indirect evidence that IL-1 and TGF beta may play crucial roles at the embryo-maternal interface during trophoblast invasion by regulating stromal cell expression of TIMP-1, TIMP-3, and 92-kDa type IV collagenase, all of which are known to be important in trophoblast invasion.     

Ovarian steroid regulation of vascular endothelial growth factor in the human endometrium: implications for angiogenesis during the menstrual cycle and in the pathogenesis of endometriosis

The human endometrium undergoes a complex process of vascular and glandular proliferation, differentiation, and regeneration with each menstrual cycle in preparation for implantation. Vascular endothelial growth factor (VEGF) is an endothelial cell-specific angiogenic protein that appears to play an important role in both physiological and pathological neovascularization. To investigate whether VEGF may regulate human endometrial angiogenesis, we examined VEGF messenger ribonucleic acid (mRNA) and protein throughout the menstrual cycle and studied the regulation of VEGF by reproductive steroids in isolated human endometrial cells. By ribonuclease protection analysis, VEGF mRNA increased relative to early proliferative phase expression by 1.6-,2.0-, and 3.6-fold in midproliferative, late proliferative, and secretory endometrium, respectively. In histological sections, VEGF mRNA and protein were localized focally in glandular epithelial cells and more diffusely in surrounding stroma, with greatest VEGF expression in secretory endometrium. Consistent with these in vivo results, the treatment of isolated human endometrial cells with estradiol (E2), medroxyprogesterone acetate (MPA), or E2 plus MPA significantly increased VEGF mRNA expression over the control value by 3.1-, 2.8-, and 4.7-fold, respectively. The VEGF response to E2 was rapid, with steady state levels of VEGF mRNA reaching 85% maximum 1 h after the addition of steroid. E2 also caused a 46% increase in secreted VEGF protein, and the combination of E2 and MPA caused an 18% increase. VEGF expression in endometriosis, an angiogenesis-dependent, estrogen-sensitive disease was similar to that seen in eutopic endometrium. Peritoneal fluid concentrations of VEGF were significantly higher in women with moderate to severe endometriosis than in women with minimal to mild endometriosis or no disease. VEGF, therefore, may be important in both physiological and pathological angiogenesis of human endometrium, as it is an estrogen-responsive angiogenic factor that varies throughout the menstrual cycle and is elevated in women with endometriosis.     

Regulation of cyclooxygenase-2 and prostaglandin F synthase gene expression by steroid hormones and interferon-tau in bovine endometrial cells

Estradiol (E2) and progesterone are responsible for regulating PG synthesis in the endometrium during the estrous cycle and interferon-tau (IFN-tau) alters PG synthesis during early pregnancy in ruminants. In this study, we examined the effects of these steroid hormones and recombinant bovine IFN-tau (rbIFN-tau) on PG production and on cyclooxygenase-2 (COX-2) and PG F (PGF) synthase (PGFS) gene expression in isolated endometrial cells. E2 decreased both PGF2alpha and PG E2 (PGE2) whereas progesterone increased PGF2alpha secretion in epithelial cells. Steroid hormones had no effect on PG production in stromal cells. rbIFN-tau attenuated both PGF2alpha and PGE2 production in epithelial cells and enhanced their production, and the ratio of PGE2 to PGF2alpha, in stromal cells. Northern blot analysis showed that E2 and rbIFN-tau decreased COX-2 messenger RNA (mRNA) levels in epithelial cells. Conversely, rbIFN-tau increased COX-2 mRNA in stromal cells. Furthermore, rbIFN-tau decreased PGFS mRNA in both cell types and this was associated with the increase in PGE2/PGF2alpha ratio. These results show that the regulation of PG synthesis by steroid hormones is different in endometrial epithelial and stromal cells in vitro. The attenuation of PGF2alpha secretion from epithelial cells and increased PGE2 production in stromal cells by rbIFN-tau are modulated by steroid hormones.     

Expression of laminin and nidogen genes during the postimplantation development of the mouse placenta

The expression patterns of laminin A, B1, B2, and nidogen genes were identified by in situ hybridization in postimplantation mouse extraembryonic tissues and maternal decidua during the period when the chorioallantoic placenta is established. Laminin and nidogen genes were not coordinately expressed either in the decidua or in trophoblast cells, indicating that these genes are regulated independently in these cell types during the establishment of the placenta. Laminin A mRNA was absent from the decidua except in the outer layer of cells adjacent to the myometrium and in the central decidual zone adjacent to the remnant of the uterine epithelium on Day 9. At this stage laminin B1, B2, and nidogen genes were strongly expressed in these cells and also in other regions of the decidua. Laminin B1 mRNA was present at higher levels in the decidua capsularis than in the decidua basalis, while nidogen mRNA showed highest expression in the decidua basalis. Laminin B2 mRNA was produced uniformly throughout the decidua at very high levels, suggesting that laminin B2 chains may be an important component of the decidual matrix. By Day 11, the nidogen gene was expressed only in endothelial cells lining the maternal blood spaces within the decidua. Laminin B1 and nidogen mRNAs were found at high levels within trophoblast giant cells at all stages, while laminin A mRNA was detected in trophoblast giant cells at later stages and laminin B2 mRNA was not produced in high levels by these cells. The patterns of gene expression show a very high degree of regional specialization, suggesting that the extracellular matrices in different regions of the decidua and extraembryonic membranes are likely to be composed of quite different ratios of laminin and nidogen polypeptides.(ABSTRACT TRUNCATED AT 250 WORDS)     

Identification of ''tissue'' transglutaminase binding proteins in neural cells committed to apoptosis

Mice in which the gene that encodes the receptor (R) for leukemia inhibitory factor (LIF) has been deleted show abnormal growth and development of the placenta. This indicates that LIF plays an important role in placental development. The expression of LIF-R and LIF was examined in human trophoblast and decidua using in situ hybridization and immunocytochemistry. LIF-R mRNA and immunoreactivity was localized in villous and extravillous trophoblast throughout pregnancy, and in endothelial cells of the fetal villi. Strong expression of mRNA encoding LIF was detected in decidual leukocytes, which are abundant at the implantation site. Extravillous trophoblast, which invades the maternal decidua, therefore expresses LIF-R as it moves past decidual leukocytes, which express LIF mRNA. The effect of LIF on cultured human trophoblast was examined in vitro. Recombinant human LIF had no effect on [3H]thymidine incorporation by purified extravillous trophoblast, nor on expression of integrins alpha1, alpha5, or beta1 by isolated trophoblast. These results identify fetal endothelial cells and all cells of the trophoblast lineage as targets for the action of LIF in human placenta. Although its effects on trophoblast are not yet clear, LIF appears to mediate interactions between maternal decidual leukocytes and invading trophoblast. LIF may also play a critical role in controlling angiogenesis in the placental villi, since human fetal endothelial cells express LIF-R, and mice lacking a functional LIF receptor gene show altered vascular development in the placenta.     

Involvement of trophoblast in embryo implantation: regulation by paracrine factors

In order to investigate the regulatory role of only one endometrial cell type on trophoblastic invasion, we explored the effects of culture medium conditioned by in vitro decidualised stromal cells (DCM) and of insulin-like growth factor binding protein-1 (IGFBP-1, the main secretory product of decidual cells) on the trophoblastic secretion of gelatinases and tissue inhibitor of metalloproteinases (TIMP-1). First trimester cytotrophoblastic cells (CTB) were obtained from abortions and cultured in vitro in presence or absence of DCM or IGFBP-1. Secreted gelatinases were analysed in the culture supernatants by zymography and by measurements of the total gelatinolytic activity. Tissue inhibitor of metalloproteinases (TIMP-1) was measured by a commercially available immunoassay. DCM inhibited the total gelatinolytic activity of CTB but increased trophoblastic MMP-9 and TIMP-1. In contrast, IGFBP-1 increased the total gelatinolytic activity and TIMP-1 and had no effect on MMP-2 and MMP-9. We conclude that a factor secreted by decidual cells (possibly TGFbeta) inhibits the total gelatinolytic activity of trophoblast by increasing TIMP-1 but other factors, as yet unidentified, increase MMP-2 and MMP-9 to an extent which does not shift the equilibrium between the gelatinases and TIMP-1 in favour of the gelatinases. In contrast to DCM, IGFBP-1 increases the total gelatinolytic activity probably by stimulating another gelatinase (stromelysin-1?) as MMP-2 and MMP 9 are unchanged by IGFBP-1. The possibility of an integrin mediated effect of IGFBP-1 on CTB is discussed.     

The expression of stem cell factor and its receptor, c-kit in human endometrium and placental tissues during pregnancy

Stem cell factor (SCF) and its receptor Kit regulate the proliferation and survival of early hematopoietic cell types as well as germ cells and melanocytes. As SCF augments the effects of several hematopoietic growth factors that are produced in reproductive tissues during pregnancy and also plays an important role in cell migration, proliferation, and survival, we studied the expression and localization of this receptor/ligand in human endometrial and placental tissues. Kit was detected by Western blot analysis in early decidual and placental tissues (8-19 weeks gestation) and in term placenta. Immunohistochemistry localized Kit mainly in trophoblast and to a lesser extent in scattered cells in the placental villous core and decidual stroma. Ribonuclease protection assay showed that SCF messenger ribonucleic acid (mRNA) expression increased 3-fold in decidua from early pregnancy compared to proliferative and secretory endometrium (P < 0.05). Placental tissues expressed 4- to 8-fold higher levels of SCF mRNA compared to decidus (P < 0.05). Isolated placental villous core expressed 7-fold higher levels of SCF mRNA than did trophoblast (P < 0.05). Thus, SCF and its receptor Kit are expressed in human endometrium and placental tissues during pregnancy, and the pattern of receptor/ligand expression suggests that endometrial and placental villous core SCF may have a paracrine effect on trophoblast through the receptor Kit.     

Epithelial cells up-regulate matrix metalloproteinases in cells within the same mammary carcinoma that have undergone an epithelial-mesenchymal transition

A metastatic rat mammary carcinoma cell line, BC1, contains cells that have retained epithelial differentiation characteristics and metaplastic cells that have undergone an epithelial-mesenchymal transition. These two subpopulations cooperate to degrade collagen. We have used novel PCR assays to quantitate, for the first time, absolute levels of the mRNAs encoding matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) in cell and tumor samples. BC1 tumors expressed high levels of the collagenase-3, TIMP-2, stromelysin-1, and gelatinase B genes and low levels of the stromelysin-2 and TIMP-1 genes. This pattern of expression was repeated in cultures of BC1 and cultures containing mixed clones of epithelial cells and metaplastic cells. In both BC1 and the biclonal cultures, metaplastic cells were the main source of collagenase-3, stromelysin-1 and stromelysin-2, whereas TIMPs were equally distributed and epithelial cells were the main source of gelatinase B. High levels of all four MMP mRNAs in metaplastic cells were dependent on coculture with epithelial cells, suggesting the production of an inducing factor by the epithelial cells. In contrast, gelatinase B mRNA was produced at a high level by epithelial cells in the absence of metaplastic cells. TIMP-2 mRNA was abundant in both subpopulations grown alone and did not change substantially upon coculture. Thus, the interclonal cooperativity to degrade collagen in BC1 cells required the induction of MMPs in metaplastic cells by epithelial cells. Interclonal cooperativity may be important to the progression of neoplastic tumors, a feature of which is phenotypic heterogeneity.