Regulation of multidrug resistance gene mdr1b/mdr1 expression in isolated mouse uterine epithelial cells

A new aspect on glutathione-associated biological function of MRP/GS-X pump and its gene expression

The biological function as well as gene expression of the MRP/GS-X pump is closely linked with cellular GSH metabolism. This article describes two important aspects, i.e., 1) a role of the MRP/GS-X pump in the modulation of cell cycle arrest induced by anticancer prostaglandins; 2) coordinated up-regulation of gamma-glutamylcysteine synthetase gamma-GCS) and MRP1 genes. The A and J series of prostaglandins (PGs) accumulate in the nuclei to suppress the proliferation of cancer cells. Delta(7)-Prostaglandin A(1) (Delta(7)-PGA(1)) methyl ester, a synthetic anticancer PG, increased the mRNA level of the cyclin-dependent kinase inhibitor p21(Sdi1/CIP1/WAF1) in human leukemia HL-60 cells. The induction of p21(Sdi1/CIP1/WAF1) was associated with the accumulation of hypophosphorylated retinoblastoma protein (pRB) and the suppression of c-myc gene expression. Unlike HL-60 cells, cisplatin-resistant HL-60/R-CP cells were insensitive to Delta(7)-PGA(1) methyl ester. While c-myc expression was transiently suppressed, neither G1 arrest nor hypophosphorylation of pRB was observed with the anticancer PG. Plasma membrane vesicles from HL-60/R-CP cells showed an enhanced level of GS-X pump activity toward the glutathione S-conjugate of Delta(7)-PGA(1) methyl ester. GIF-0019, a potent inhibitor of the GS-X pump, dose-dependently enhanced the cellular sensitivity of HL-60/R-CP cells to Delta(7)-PGA(1) methyl ester, resulting in G1 arrest. The GS-X pump is suggested to play a pivotal role in modulating the biological action of the anticancer PG. The expression of MRP1 and gamma-GCS genes can be coordinately up-regulated by cisplatin, 1-[5-(4-amino-2-methyl)pyrimidyl]methyl-3-(2-chloroethyl)-3-nitrosourea (ACNU), and heavy metals in human cancer cells. For the up-regulation of these genes, both transcriptional and posttranscriptional regulations are considered to be involved.

Increased expression of MDR1 mRNAs and P-glycoprotein in placentas from HIV-1 infected women

P-glycoprotein transports several compounds including protease inhibitors, actually used in the clinical management of HIV-1 infection. Since P-glycoprotein is expressed in placental trophoblasts, its efflux activity could interfere with placental transfer of antiretrovirals. The purpose of this study was to investigate the expression of P-gp-encoding MDR1 gene and P-gp itself in full-term placentas from uninfected (n=35) and HIV-1 infected women (n=24). MDR1 transcripts were quantified by real-time PCR using relative (MDR1 normalized upon 28S levels) and absolute (copy number) determinations. P-glycoprotein localization and expression were evaluated by immunohistochemistry and western blot analysis, respectively. Relative or absolute PCR quantification showed a significant 3.3-fold (p<0.0009) or 3.7-fold (p<0.0002) mean increase in MDR1 placental transcription in HIV-infected compared to non-infected women, respectively. Ratios of individual HIV-positive values to HIV-negative mean ranged from 0.1 to 21.8. Moreover a significant 2.5-fold increased expression of immunoreactive P-glycoprotein was evidenced in placentas from HIV-infected women (p<0.0001). This MDR1 overexpression was observed in a similar extent in placentas from pregnant women treated with Zidovudine alone or in combination with Nelfinavir and/or Lamivudine. Our findings suggest that P-glycoprotein in placentas from HIV-infected women would contribute to modulate the materno-fetal transport of antiretrovirals across the placental barrier and consequently diminish fetal exposure to these compounds.

Identification of in vivo P-glycoprotein mRNA decay intermediates in normal liver but not in liver tumors

Post-transcriptional regulation at the level of mRNA stability is one important mechanism for over-expression of P-glycoprotein (Pgp) genes observed in cultured cells and in animals. A previous study has shown that mRNA half-lives for Pgp genes in normal liver were less than 2 h, in contrast to greater than 12 h measured in a transplantable liver tumor line. This lower turnover rate of Pgp mRNA may, in large part, contribute to the abundance of Pgp mRNA in liver tumors. The current study sought to investigate the underlying mechanism for the lower turnover rate of Pgp2 mRNA previously determined in liver tumors. As a first approach, we set out to understand the Pgp2 mRNA decay in both normal liver and liver tumors by first identifying and characterizing Pgp2 mRNA degradation intermediates. In this study, we showed that the sensitive ligation-mediated polymerase chain reaction (LM-PCR) method can be used to detect a homogenous pool of in vitro transcribed RNA down to 0.4 ng. By employing gene-specific primers in the LM-PCR method, we successfully identified four Pgp2 mRNA decay intermediates in normal liver. All four decay intermediates detected correspond to the 5' coding region of Pgp2 mRNA, and surprisingly no decay intermediates which correspond to 3' untranslated region, 3' coding region or middle coding region were found using LM-PCR. The identified decay intermediates are unique to the normal liver as they were absent or present at very low level in all three liver tumor samples analyzed. This observation supports our previous findings that the Pgp mRNA turnover rate is lower in liver tumors than in normal liver. These findings have implications for our understanding of the regulation of Pgp mRNA turnover in normal and malignant tissues.

Superinduction of P-glycoprotein messenger RNA in vivo in the presence of transcriptional inhibitors

P-Glycoprotein (P-gp) is comprised of a small family of plasma membrane proteins, and its presence in high amounts often correlates with multidrug resistance in cultured cells. Dramatically increased levels of a single member of P-gp mRNA (pgp2) have been observed in experimental liver carcinogenesis models, during liver regeneration, upon culturing of hepatocytes and in the uterus of pregnant animals. In all cases, the increase in mRNA level appears to be the result of an increase in mRNA half-life (stability). Previously, we have used transcriptional inhibitors alpha-amanitin and actinomycin D to measure P-gp mRNA half-life in normal liver and in liver tumors. We showed that the level of all three P-gp mRNAs decreased with time in the presence of transcriptional inhibitors, yielding measured half-lives of less than 2 h in liver but greater than 12 h in liver tumors. This observation raised the possibility that regulation of P-gp mRNA stability plays a role in liver carcinogenesis. In the present study, we measured P-gp mRNA half-life in other normal tissues to determine if a short P-gp mRNA half-life is unique to the liver. Our study reveals that in contrast to liver, measured P-gp mRNA half-lives in most tissues examined are greater than 12 h. Moreover, we observed an unexpected, marked increase in the level of pgp2 mRNA with time after injection of transcriptional inhibitors. This can only be explained if the transcriptional inhibitors directly or indirectly inhibit the normally high degradation rate of pgp2 mRNA, resulting in the superinduction of this mRNA. These findings have implications for our understanding of the regulation of P-gp gene expression and drug resistance in vivo.

Induction of P-glycoprotein mRNA transcripts by cycloheximide in animal tissues: evidence that class I Pgp is transcriptionally regulated whereas class II Pgp is post-transcriptionally regulated

P-glycoprotein (Pgp) are a small family of plasma membrane proteins capable of transporting substrates across cell membranes. Class I and class II Pgp are able to transport drugs and have been shown to mediate multidrug resistance (MDR). Class III Pgp is a long chain phospholipid transporter and does not mediate MDR. The expression and regulation of Pgp genes in animal tissues are not well understood. In this study, the protein synthesis inhibitor cycloheximide was used as a tool to understand Pgp gene expression and regulation in animal tissues. The sensitive RNase protection assay was used to detect changes in Pgp mRNA levels and nuclear run-on assay was used to determine whether transcription or post-transcription is important. The results showed that cycloheximide significantly induced class II Pgp expression in all tissues examined. This was predominantly through post-transcriptional effect. In contrast, the relatively modest increase in class I Pgp expression by cycloheximide was found to be mainly due to increased transcriptional activity. On the other hand, cycloheximide induced class III Pgp expression in some tissues while caused decay of class III Pgp mRNA in other tissues. The transcriptional and post-transcriptional mechanisms exerted by cycloheximide on Pgp genes are discussed. These findings have implications for our understanding of gene regulation in animal tissues and MDR reversal strategies in vivo.

Repression of chick multidrug resistance-associated protein 1 (chMRP1) gene expression by estrogen

Although a number of genes have been identified whose transcriptional activities are stimulated by estrogen, relatively few have been discovered that are repressed. In an effort to determine whether estrogen can directly repress gene expression, attempts were made to identify genes that are direct targets of the estrogen receptor and whose activities are repressed by it. Because the development and differentiation of the chick oviduct are exquisitely dependent upon estrogen, this seemed an appropriate model system for testing this hypothesis. RNA was isolated from estrogen-treated and estrogen-withdrawn chick oviducts and was subjected to differential display analysis. Surprisingly, one of the products repressed by estrogen encoded the chick homolog of the multidrug resistance-associated protein 1 (MRP1) gene. Further cloning resulted in a chick MRP1 (chMRP1) cDNA clone that is 72% identical with human MRP1. Translation of the chMRP1 sequence indicates a 77% amino acid identity with both the human and mouse MRP1 proteins. Treatment of estrogen-withdrawn chicks with 17beta-estradiol decreased chMRP1 mRNA levels to 50% within 30 min and to 70% by 1h, which is comparable to the level observed with chronic repression by estrogen. ChMRP1 mRNA is present in many other tissues, including the heart, lung, brain, kidney, skeletal muscle, and intestine, but is undetectable in the liver. This study indicates that in estrogen-responsive tissues such as chick oviduct, the regulation of chMRP1 gene expression is controlled by estrogen.

Conditional expression of RNA polymerase II in mammalian cells. Deletion of the carboxyl-terminal domain of the large subunit affects early steps in transcription

The carboxyl-terminal domain (CTD) of the large subunit of mammalian RNA polymerase II contains 52 repeats of a heptapeptide that is the target of a variety of kinases. The hyperphosphorylated CTD recruits important factors for mRNA capping, splicing, and 3'-processing. The role of the CTD for the transcription process in vivo, however, is not yet clear. We have conditionally expressed an alpha-amanitin-resistant large subunit with an almost entirely deleted CTD (LS*Delta5) in B-cells. These cells have a defect in global transcription of cellular genes in the presence of alpha-amanitin. Moreover, pol II harboring LS*Delta5 failed to transcribe up to the promoter-proximal pause sites in the hsp70A and c-fos gene promoters. The results indicate that the CTD is already required for steps that occur before promoter-proximal pausing and maturation of mRNA.

Increased P-glycoprotein messenger RNA stability in rat liver tumors in vivo

P-glycoproteins (Pgp) are comprised of a small family of plasma membrane proteins whose abundance in cultured cells is often associated with the multidrug resistance phenotype. Overexpression of Pgp has been observed in many types of human cancers, but the molecular basis for this overexpression has not been established. We have used primary monolayer cultures of adult rat hepatocytes and a stepwise model of rat liver carcinogenesis to study the regulation of Pgp gene expression. We observed a marked overexpression of Pgp, specifically the class II Pgp, in both systems. In addition, we observed that a number of unrelated genes including alpha-tubulin, beta-actin, gamma-actin, cytokeratin 8, cytokeratin 18, and c-myc are overexpressed in cultured hepatocytes, and they are also overexpressed during liver carcinogenesis and in transplantable tumors. Nuclear run-on assays showed no increase in the transcriptional activity of Pgp genes in transplantable liver tumors compared to normal liver. Studies of in vivo mRNA stability, however, revealed that all three Pgp mRNAs were relatively stable in transplantable liver tumors (t(1/2) > 12 h), in contrast to what was found in normal liver (t(1/2) < 2 h). In addition, mRNA for several other genes, including alpha-tubulin, c-myc, and cyclin D1, all appear to be stabilized in the tumors. These findings suggest that the overexpression of Pgp genes in rat liver tumors may be the result of a mechanism involving stabilization of a diverse group of mRNAs.

Transcriptional regulation of the murine multidrug resistance gene mdr1b by progesterone occurs via an indirect mechanism

The murine multidrug resistance gene mdr1b is highly induced in the endometrium during pregnancy. Evidence suggests that induction occurs mainly as a result of progesterone action. To study the molecular mechanisms involved in this induction, 5'-flanking sequences between -540 and +97 of the mdr1b gene were fused to the reporter gene, bacterial chloramphenicol acetyltransferase (p540CAT). Unlike most progesterone-responsive genes, mdr1b is preferentially activated by the A form of the progesterone receptor. We now report that activation is not observed with a DNA-binding domain mutant of progesterone receptor A (PRA) suggesting that induction occurs at the transcriptional level. Time course experiments demonstrated that induction was first observed 12 hr after hormone addition, suggestive of a secondary (or late) response gene. Sequence comparison highlighted the region M1 (-234 to -206), which contains a partially conserved progesterone response element. Its functional significance was evaluated by expression assays and gel shift analysis. Reporter plasmids with modifications of this element were transfected into HeLa cells. Constructs containing the native M1 element, or a mutated element (M1mt) that eliminated any similarity to a progesterone response element, were induced four-fold by progesterone whereas an element containing a consensus progesterone response element (M1PRE) was induced eight-fold. In addition, by gel shift analysis, the M1 element did not bind the progesterone receptor or any other factors. This suggested that the M1 region does not participate in the response to progesterone. 5' Nested deletion analysis, used to identify other regions of the upstream regulatory region that contributed to induction by progesterone, demonstrated that enhancer sequences between -122 and -65, which contain binding sites for C/EBPbeta and NF-Y, were important. Mutations in the binding sites for these factors decreased induction by progesterone. On the basis of our studies using 540 bp of upstream sequence, mdr1b is activated transcriptionally by progesterone, in an indirect manner dependent on basal factors.

Bromocriptine transcriptionally activates the multidrug resistance gene (pgp2/mdr1b) by a novel pathway

The P-glycoprotein (Pgp) reversing agent, reserpine, induces MDR1 mRNA and PGP protein in human colon carcinoma cells (Schuetz, E. G., Beck, W. T., and Schuetz, J. D. (1996) Mol. Pharmacol. 49, 311-318) and in H35 rat hepatoma cells. Reserpine's interference with cellular dopamine utilization suggested that dopamine and dopaminergics might be important physiological regulators of PGP expression. Initial studies demonstrated that the H35 cells express the D2 dopamine receptor. Pgp protein and pgp2/mdr1b mRNA was increased (maximum of 10- and 8-fold, respectively) by the potent D2 dopamine receptor agonists bromocriptine, R(-)-propylnorapomorphine hydrochloride, and quinpirole, and Pgp protein induction was blocked by D2 receptor antagonists spiperone and clozapine. D2 receptor agonist induction of pgp2/mdr1b mRNA was paralleled by transcriptional activation of the pgp2/mdr1b promoter but blocked by pretreatment with the D2 dopamine receptor antagonists, spiperone, eticlopride, and clozapine. Co-transfection of a D2 dopamine receptor expression vector enhanced bromocriptine's transcriptional activation of the pgp2/mdr1b promoter. The G-protein, Galphai2, is required for bromocriptine transcriptional activation because the G-protein inhibitor, pertussis toxin, suppressed bromocriptine's activation of pgp2/mdr1b transcription and co-transfection of a dominant negative Galphai2 abrogated bromocriptine activation of pgp2/mdr1b. Gi proteins can transduce signals by activation of mitogen-activated protein kinases (MAPKs), and because Raf-1 is a known activator of MDR1, we tested for Raf-1 involvement. Co-transfection of a dominant negative Raf-1 failed to block bromocriptine induction of pgp2/mdr1b, and bromocriptine treatment caused no phosphorylation of the MAP kinase kinase substrates p42 and p44, demonstrating that the MAP kinase pathway was not involved. These are the first studies demonstrating transcriptional activation of an MDR gene by dopamine receptor agonists and that this activation occurs by a signal transduction pathway requiring the D2 dopamine receptor coupled to a functional G-protein.