Influence of isolation procedure, extracellular matrix and dexamethasone on the regulation of membrane transporters gene expression in rat hepatocytes

Abstract

The influence of the isolation procedure of hepatocytes, extracellular matrix (ECM) configuration and incubation medium supplementation by dexamethasone (DEX) on the cell morphology and on the gene expression of membrane transporters was examined in rat hepatocytes. The mRNA levels were determined using oligonucleotide microarrays, in liver, in suspension and in primary culture in monolayer (CPC), and in collagen gels sandwich (SPC) in absence and presence of DEX (100 and 1000 nM). The results indicated pronounced morphological differences between CPC and SPC in response to DEX demonstrating that the hepatocytes re-formed, as in vivo, multicellular arrays with extensive bile canalicular network only in SPC in presence of DEX. The mRNA levels of membrane transporters were not affected significantly during isolation procedure. However, plating hepatocytes in CPC resulted in a decrease of major basolateral transporters mRNA level whereas mRNA levels of mdr1b and mrp3 were increased (>100-fold). Similar observations were made in SPC in the absence of DEX demonstrating that the ECM configuration alone did not play a critical role in the regulation of membrane transporters. However, adding DEX to the incubation medium in SPC resulted in an up-regulation of mdr2, oatp2 and mrp2 in a concentration-dependent way for the two latter genes, whereas mdr1b and mrp3 expression were maintained to their baseline liver levels. These data suggested therefore that the combination of ECM and DEX supplementation is essential for the formation of the bile canalicular network and is a determinant factor in the regulation of membrane transporters in cultured rat hepatocytes.

Introduction

In vivo, hepatocytes are polarized and express different transport systems located at their basolateral, lateral and canalicular domains. On the basolateral membrane, several active transporters such as Na⁺-taurocholic acid transporting polypeptide (ntcp), organic anion transporting polypeptide (oatp1 and oatp2), and organic cation transporter (oct1) are expressed and play an important role in the Na⁺-dependent uptake of bile acids, Na⁺-independent uptake of organic anions and organic cations, respectively [1], [2]. In addition, bile acids are taken up in a Na⁺-independent way by the liver-specific rat organic anion transporter rlst1 [3]. The bile canalicular membrane of the mammalian hepatocyte contains several primary active transporters that couple ATP-hydrolysis to the transport of specific substrates into the bile canaliculus [4]. These transmembrane transporters are members of the ATP-binding cassette (ABC) transporters and currently include the multidrug resistance associated protein transporter (mrp2 or cmoat), the P-glycoproteins (mdr1, mdr2) and the sister P-gp (spgp) or cbsep that have been shown to act as export pumps for their ligands into bile [2]. The mdr1 sub-family (P-glycoprotein), encoded by two different genes in rodent (mdr1a and mdr1b), is thought to be responsible for the excretion of cationic and neutral amphipathic compounds whereas mdr2 acts as an export pump of phospholipids [5], [6]. The canalicular multispecific organic anion transporter (cmoat or mrp2) is largely involved in the biliary excretion of many organic anions including conjugated xenobiotics [7]. Some other transporters identified at the basolateral membrane with similar substrate specificity as the canalicular mrp2 can also act as efflux pumps by exporting drugs from cytosol to blood [6], [8], [9], [10], [11]. In fact, at least three new members of the mrp family that have been identified, mrp1, mrp3 and mrp6 are expressed in rat liver. Mrp6 is localized at the lateral and, to a lesser extent at the canalicular membrane of hepatocyte and is supposed to fulfill a housekeeping transport function involved in regulation of para- or trans-cellular solute movement from blood to bile [10]. The localization of mrp3 is controversial and needs further investigations. It is thought to be present at the basolateral membrane and/or at the canalicular membrane of the hepatocytes [9], [12]. Another member of this transporter family (mrp1) was identified and present at a very low level in rat liver at the lateral membrane [13].

The regulation of sinusoidal uptake and canalicular secretion occurs at different levels. It was shown that the regulation is dependent on the physiological environment, osmolarity, transporter gene expression and transporter degradation [14], [15], [16]. On a short-term basis, the level of substrate availability, the covalent modification of transporters, and their regulated exocytic insertion into or retrieval from the membrane may also contribute to their regulation [17]. It became evident that the expression of some of these proteins, particularly involved in transport processes, is extensively regulated in vivo. Therefore, the secretory functions of bile acids and drugs may be hormonally modulated by either vesicle-mediated retrieval or insertion of transport proteins into the canalicular domain, thus regulating their surface density [18].

Strong efforts are made to mimic the in vivo situation in various in vitro models. Several strategies have been pursued to maintain the morphology and the liver-specific properties of hepatocytes. After isolation from liver, the hepatocytes can be maintained in different extracellular matrices (ECM) configuration (single or double collagen matrices, Matrigel, Vitrogen) in customized culture media supplemented with different nutrients including mainly glucocorticoids, growth factors, insulin or hydrocortisone. Nevertheless, “de-differentiation” is well known to occur in primary monolayer culture, where hepatocytes lose many of their specific properties such as reduced synthesis of serum proteins; a progressive fall in levels of glucose-6-phosphatase; a decrease in cytochromes P450, NADPH cytochrome P450 reductase [19], [20]. In contrast, it has been shown that the hepatocytes cultured between two layers of hydrated collagen in a medium containing dexamethasone (DEX), a synthetic glucocorticoid, retrieve their membrane polarity [21], [22], [23], a variety of cell functions and especially their excretion capacity after 4 days [24].

Various liver functions have been demonstrated to be also strongly modulated both in vivo and in vitro by soluble factors, and for instance corticosteroid hormones. There is a general consensus that glucocorticoids markedly improve the attachment, survival, morphology, and overall performance of hepatocytes seeded on single substrata [25], [26]. As suggested by the effects of DEX on the overall pattern of protein synthesis, glucocorticoids directly influence a wide range of activities in culture. DEX has been found to increase fibronectin secretion, induce tyrosine aminotransferases, promote an ordered arrangement of the cytoskeleton, enhance gap junction expression and function, regulate the P-gp expression, support P450 activity, and curtail the decrease in protein synthesis observed in hepatocytes during the initial 24 hr [16], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36]. The formation of bile canalicular networks is enhanced in the presence of dexamethasone, especially in conjunction with an overlay of ECM [33], [34]. Recent data suggest that the ECM plays an important role in the maintenance of differentiated characteristics of primary hepatocytes by inducing the transcription of liver-specific genes and, also, by destabilizing the mRNAs of ubiquitously-expressed genes [37].

Taken together, the extracellular matrix addition and/or the medium composition may have a significant impact on expression of the different transport proteins [38], [39], [40]. However, there is still a lack of fundamental understanding of all necessary factors to culture and maintain differentiated hepatocytes. The present study was therefore designed to investigate (i) the impact of the isolation process, (ii) the influence of the configuration of the extracellular matrix, and (iii) the effect of dexamethasone treatment on the cell morphology and on the expression at the mRNA level of genes encoding membrane-specific transport proteins in hepatocytes. The gene expression of the different liver-specific sinusoidal, lateral and canalicular transport proteins was therefore determined using rat GeneChip^®, a high density oligonucleotide microarrays, in male rat liver, in suspension of freshly isolated hepatocytes and at different culture stage under conventional and sandwich configurations. In order to evaluate the impact of dexamethasone, the mRNA levels of specific hepatic transporters was evaluated in hepatocytes in sandwich configuration in response to dexamethasone.