Inflammation is known to markedly impair hepatic detoxification pathways, and thereby to alter pharmacokinetics of drugs (Slaviero et al., 2003). This has been linked to changes in liver expression of drug-metabolizing enzymes such as cytochromes P450 (Renton, 2004) and more recently to that of drug transporters (Geier et al., 2007; Petrovic et al., 2007; Morgan et al., 2008; Teng and Piquette-Miller, 2008). Indeed, treatment by the proinflammatory factor lipopolysaccharide (LPS) has been shown to result in a pronounced alteration of hepatic transporter expression in rodents: sinusoidal influx solute-carrier (SLC) transporters such as sodium-taurocholate cotransporting polypeptide (Ntcp) (Slc10a1), organic anion-transporting polypeptides (Oatp) 1a1 (Slco1a1) and Oatp1a2 (Slco1a2) and organic cation transporter (Oct) 1 (Slc22a1), and canalicular ATP-binding cassette (ABC) transporters such as multidrug resistance (Mdr) 1a gene/P-glycoprotein (Abcb1a), bile salt export pump (Bsep) (Abcb11), and multidrug resistance-associated protein (Mrp) 2 (Abcc2) thus are markedly down-regulated in liver of endotoxemic rats or mice (Bolder et al., 1997; Piquette-Miller et al., 1998; Vos et al., 1998; Cherrington et al., 2004). Expression of NTCP (SLC10A1) and MRP2 (ABCC2) is similarly decreased in human liver in response to LPS (Elferink et al., 2004).

Proinflammatory cytokines such as interleukin (IL)-1β, tumor necrosis factor-α (TNF-α), and IL-6 probably contribute to altered expression of hepatic drug transporters occurring during inflammation (Hartmann et al., 2002). Indeed, administration of IL-1β, TNF-α, or IL-6 to rodents resulted in reduced expression of various sinusoidal or canalicular drug transporters (Green et al., 1996; Siewert et al., 2004; Geier et al., 2005). In the same way, treatment of human hepatocytes by IL-1β has recently been shown to alter expression of several major hepatic transporters (Le Vee et al., 2008). However, whether TNF-α and IL-6 also affect human hepatic transporters remains poorly characterized, even if MDR1 (ABCB1), MRP2, and MRP3 (ABCC3) have been shown to constitute targets for TNF-α or IL-6 in some human hepatoma cell lines (Bohan et al., 2003; Lee and Piquette-Miller, 2003).

Therefore, the present study was designed to analyze the effects of TNF-α and IL-6 treatment on expression of major human hepatic transporters using primary human hepatocytes, which are well recognized as a convenient cellular model to study regulatory ways of drug detoxification proteins (Gómez-Lechón et al., 2004; Jigorel et al., 2006; Hewitt et al., 2007). Our data indicate that exposure to these cytokines markedly changes the expression profile of drug transporters, thus highlighting the probable contributing role played by these cytokines to the alteration of drug hepatic detoxification commonly observed in humans during inflammation.

Materials and Methods

Chemicals and Reagents. Recombinant human TNF-α and IL-6 were provided by R&D Systems (Minneapolis, MN). [³H(G)]Taurocholic acid (sp. act. 1.19 Ci/mmol), [6, 7-³H(N)]estrone-3-sulfate (sp. act. 57.3 Ci/mmol), and [1-¹⁴C]tetra-ethylammonium (TEA) (sp. act. 2.4 mCi/mmol) were purchased from PerkinElmer Life and Analytical Sciences (Waltham, MA). Probenecid and verapamil were from Sigma-Aldrich (Saint-Quentin Fallavier, France). Antibodies against MDR1/P-glycoprotein and breast cancer resistance protein (BCRP) (ABCG2) were from Alexis Corporation (Lausen, Switzerland), whereas those against MRP2 and MRP3 were provided by Millipore Bioscience Research Reagents (Temecula, CA). Tetramethyl rhodamine isothiocyanate-labeled secondary antibody against mouse IgG was from Jackson ImmunoResearch Laboratories Inc. (West Grove, PA). All of the other compounds and reagents were commercial products of the highest purity available. Vehicle for TNF-α and IL-6 was phosphate-buffered saline; control cultures received the same dose of vehicles as treated counterparts.

Cell Isolation and Culture. Human hepatocytes were obtained from adult donors undergoing hepatic resection for primary and secondary tumors via the Biological Resource Center (Rennes, France). Cells were prepared by perfusion of histologically normal liver fragments using a collagenase solution (Jigorel et al., 2005). They were primary cultured on plastic dishes in Williams' E medium as already reported (Chouteau et al., 2001; Jigorel et al., 2005; Le Vee et al., 2008). All of the experimental procedures complied with French laws and regulations and were approved by the National Ethics Committee.

RNA Isolation and Analysis. Total RNA was isolated from cells using the TRIzol reagent (Invitrogen, Cergy-Pontoise, France). RNA was then subjected to reverse transcription-real-time quantitative polymerase chain reaction (RT-qPCR) using the fluorescent dye SYBR Green methodology and an ABI Prism 7000 detector (Applied Biosystems, Foster City, CA) as already reported (Jigorel et al., 2005). Gene primers were exactly as described previously (Jigorel et al., 2006; Le Vee et al., 2008). Relative quantification of the steady-state target mRNA levels was calculated after normalization of the total amount of cDNA tested to an 18S endogenous reference.

Western Blot Analysis. Crude membrane extracts were prepared from primary human hepatocytes as described previously (Lecureur et al., 2005; Jigorel et al., 2006). Proteins were then separated on polyacrylamide gels and electrophoretically transferred to nitrocellulose membranes. After blocking in Tris-buffered saline containing 4% bovine serum albumin, membranes were incubated overnight at 4°C with primary antibodies directed against NTCP (Kullak-Ublick et al., 1997), OATP1B1 (SLCO1B1) (Zollner et al., 2003), MDR1, MRP2, MRP3, or BCRP. Thereafter, peroxidase-conjugated monoclonal antibodies were used as secondary antibodies. After washing, immunolabeled proteins were visualized by chemiluminescence. Gel loading and transfer were checked by staining membranes with Ponceau Red. The intensities of stained bands were measured by densitometry using ImageJ 1.40g software (National Institutes of Health, Bethesda, MD).

Immunofluorescence Studies. Primary hepatocytes cultured on glass coverslips were first fixed in ice-cold acetone for 10 min. Cells were next incubated for 2 h with mouse monoclonal antibodies used at 5 μg/ml and directed against MRP2 or MRP3 or with corresponding isotypic IgG_2a or IgG₁ controls. After washing, a tetramethyl rhodamine isothiocyanate-labeled secondary antibody directed against mouse IgG was added for 1 h, and nuclei were subsequently stained with 4,6-diamidino-2-phenylindole. Immunofluorescent images were finally detected with an inverted laser scanning confocal microscope DM IRE2 SP2 (Leica, Rueil Malmaison, France) equipped with a 63× NA 1.4 objective lens using the LCS three-dimensional software (Leica).

Transport Assays. Transport activities caused by NTCP, OATPs, or OCT1 (SLC22A1) were analyzed through measuring sodium-dependent intracellular accumulation of the NTCP substrate taurocholate, probenecid-sensitive uptake of the OATP substrate estrone-3-sulfate, and verapamil-sensitive uptake of the OCT1 substrate TEA, as described previously (Jigorel et al., 2005). In brief, cells were incubated at 37°C for 5 min with 0.17 μM[³H]taurocholate in the presence or absence of sodium, with 1.7 nM [³H]estrone-3-sulfate in the presence or absence of the OATP inhibitor probenecid used at 2 mM, or with 40 μM[¹⁴C]TEA in the presence or absence of the OCT1 inhibitor verapamil used at 50 μM. After washing in phosphate-buffered saline, cells were lysed, and accumulation of radiolabeled substrates was determined through scintillation counting. Taurocholate accumulation values in the presence of sodium minus accumulation values in the absence of sodium, estrone-3 sulfate uptake values in the absence of probenecid minus uptake values in the presence of probenecid, and TEA uptake values in the absence of verapamil minus uptake values in the presence of verapamil are thought to represent NTCP, OATP, and OCT1 activities (Jigorel et al., 2005; Le Vee et al., 2006).

Statistical Analysis. Quantitative data were usually expressed as mean ± S.E.M. They were statistically analyzed using the paired Student's t test, analysis of variance followed by the Newman-Keuls test, or the nonparametric Spearman's rank correlation method. The criterion of significance was p < 0.05.

Results

Effects of Cytokine Treatments on Expression of Referent Inflammation Markers. Primary human hepatocytes from six liver donors were exposed to 100 ng/ml TNF-α or 10 ng/ml IL-6 for 8, 24, or 48 h. None of these cytokine treatments exerted toxicity as shown by phase-contrast microscopic examination of the cultures and analysis of cell viability using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide colorimetric assay (data not shown). As shown in Table 1, TNF-α and IL-6 treatments were shown to markedly increase mRNA expression of IL-8 or C-reactive protein (CRP), respectively, whatever the exposure time, even if the level of induction varied according to hepatocyte populations, especially for CRP up-regulation. Because IL-8 and CRP are considered as prototypical inflammation markers inducible by TNF-α (Osawa et al., 2002) and IL-6 (Zhang et al., 1996), respectively, this most likely indicated that primary human hepatocytes were fully responsive to the cytokines and thus were suitable for investigating TNF-α and IL-6 effects toward drug transporter expression.

Effects of TNF-α and IL-6 Treatment on Drug Transporter mRNA Expression. We first analyzed the effects of TNF-α treatment on SLC and ABC transporter mRNA expression by RT-qPCR. As shown in Fig. 1, TNF-α-treated hepatocytes from six individuals exhibited lower mRNA expression of NTCP, OATP2B1 (SLCO2B1), OATP1B1, OATP1B3 (SLCO1B3), OCT1, organic anion transporter (OAT) 2 (SLC22A7), and BSEP when compared with untreated counterparts, whatever the time of treatment (8, 24, or 48 h), whereas mRNA levels of MDR1, MRP2, MRP3, MRP4 (ABCC4), and BCRP remained unchanged. With respect to NTCP, OATP2B1, and OAT2, the time of exposure did not affect the level of expression repression in a major way, i.e., similar reduced levels of transporter expression were found in hepatocytes exposed to TNF-α for 8 or 48 h (Fig. 1); by contrast, expressions of OATP1B1, OATP1B3, OCT1, and BSEP in hepatocytes treated with TNF-α for 48 h were lower than those found in counterparts exposed to the cytokine for only 8 h (p < 0.05), probably indicating that repression increased with exposure time. When considering the repression factor after a 24-h exposure to TNF-α, i.e., the ratio of mRNA levels in untreated hepatocytes versus those found in cytokine-exposed counterparts, BSEP was the most repressed (by an 8-fold factor) among the transporters affected by TNF-α, followed by OAT2 (repression by a 4.7-fold factor) and OATP1B3 (repression by a 3.6-fold factor) (Table 2). The dose-response relationship for TNF-α effects toward transporter expression was then characterized. For this purpose, we have only analyzed transporters whose mRNA level was decreased by at least 50% in hepatocytes exposed to 100 ng/ml TNF-α for 48 h, i.e., NTCP, OATP1B1, OATP1B3, OCT1, OAT2, and BSEP (Figs. 1 and 2). These transporters were similarly markedly down-regulated by 10 ng/ml TNF-α (Fig. 2). TNF-α used at 1 ng/ml was also found to significantly repress transporter expression, although in a more moderate manner (Fig. 2). Lower concentrations of TNF-α, such as 0.01 or 0.1 ng/ml, failed to alter transporter expression; they also did not increase IL-8 mRNA levels, which were, by contrast, hugely induced by TNF-α used at 10 or 100 ng/ml (Fig. 2).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

Effects of TNF-α on drug transporter mRNA expression in human hepatocytes. Primary human hepatocytes were either untreated or exposed to 100 ng/ml TNF-α for 8, 24, or 48 h. SLC (top) and ABC (bottom) transporter mRNA expression was then determined by RT-qPCR, as described under Materials and Methods. Data are expressed for each transporter as percentage of expression found in untreated hepatocytes, arbitrarily set at the value of 100%. They are the means ± S.E.M. of values from six independent hepatocyte populations. *, p < 0.05 when compared with untreated cells.

We next studied the effects of IL-6 treatment on drug transporter mRNA expression. As shown in Fig. 3, IL-6-exposed human hepatocytes from six liver donors exhibited lower mRNA expression of NTCP, OATP2B1, OATP1B1, OATP1B3, OCT1, OAT2, MRP2, and BCRP when compared with untreated counterparts, whatever the time of treatment (8, 24, or 48 h). With respect to OAT2, MRP2, and BCRP, the time of exposure did not affect the level of expression repression in a major way, i.e., similar reduced levels of transporter expression were found in hepatocytes exposed to IL-6 for 8 or 48 h (Fig. 3); by contrast, expressions of NTCP, OATP2B1, OATP1B1, OATP1B3, and OCT1 in human hepatocytes treated with IL-6 for 48 h were lower than those found in counterparts exposed to the cytokine for only 8 h (p < 0.05), probably indicating that repression increased with exposure time. IL-6 treatment was also found to moderately repress MDR1 mRNA expression but only after a 24- or 48-h treatment; it also reduced MRP4 mRNA levels but only for an 8- and 24-h treatment; by contrast, it did not significantly alter mRNA expression of BSEP and MRP3, whatever the time of treatment (Fig. 3). When considering the repression factor after a 24-h exposure to IL-6, NTCP was the most repressed (by an 18.8-fold factor) among the transporters affected by IL-6, followed by OATP1B1 (repression by a 6.5-fold factor) and OATP2B1 (repression by a 3.8-fold factor) (Table 2). The dose-response relationship for IL-6 effects toward transporter expression was then characterized. For this purpose, we have only analyzed transporters whose mRNA level was decreased by at least 50% in hepatocytes exposed to 10 ng/ml IL-6 for 48 h, i.e., NTCP, OATP2B1, OATP1B1, OATP1B3, OCT1, and BCRP (Figs. 3 and 4). IL-6 used at 1 ng/ml also down-regulated these transporters, but in a more moderate manner, and these repressions reached significant level for only NTCP, OATP2B1, OATP1B1, and OCT1 (Fig. 4); similarly, CRP up-regulation was weaker in response to 1 ng/ml IL-6 than in response to 10 ng/ml IL-6 (Fig. 4). The use of a lower IL-6 concentration, such as 0.1 ng/ml, failed to alter transporter expression; it also did not alter CRP mRNA levels (Fig. 4).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Dose-response of TNF-α effects toward transporter expression. Primary human hepatocytes were either untreated or exposed to 0.01, 0.1, 1, 10, or 100 ng/ml TNF-α for 48 h. IL-8 (top) and transporter (bottom) mRNA expression was then determined by RT-qPCR, as described under Materials and Methods. Data are expressed for IL-8 and each transporter as percentage of expression found in untreated hepatocytes, arbitrarily set at the value of 100%. They are the mean ± S.E.M. of values from three independent hepatocyte populations. *, p < 0.05 when compared with untreated cells.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

Effects of IL-6 on drug transporter mRNA expression in human hepatocytes. Primary human hepatocytes were either untreated or exposed to 10 ng/ml IL-6 for 8, 24, or 48 h. SLC (top) and ABC (bottom) transporter mRNA expression was then determined by RT-qPCR, as described under Materials and Methods. Data are expressed for each transporter as percentage of expression found in untreated hepatocytes, arbitrarily set at the value of 100%. They are the mean ± S.E.M. of values from six independent hepatocyte populations. *, p < 0.05 when compared with untreated cells.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

Dose-response of IL-6 effects toward transporter expression. Primary human hepatocytes were either untreated or exposed to 0.1, 1, or 10 ng/ml IL-6 for 48 h. CRP (top) and transporter (bottom) mRNA expression was then determined by RT-qPCR, as described under Materials and Methods. Data are expressed for CRP and each transporter as percentage of expression found in untreated hepatocytes, arbitrarily set at the value of 100%. They are the mean ± S.E.M. of values from three independent hepatocyte populations. *, p < 0.05 when compared with untreated cells.

Effects of TNF-α and IL-6 Treatment on Drug Transporter Protein Expression. To determine whether some of the changes in transporter mRNA levels induced by TNF-α or IL-6 treatment also occur at the protein level, we next performed Western blot analysis of crude membranes from cytokine-treated primary human hepatocytes and untreated counterparts. As indicated in Fig. 5, both TNF-α and IL-6 treatments for 48 h failed to obviously alter P-glycoprotein expression, whereas they induced MRP3 expression and reduced NTCP and OATP1B1 levels. Moreover, IL-6 was shown to reduce MRP2 expression, which remained unaffected by TNF-α treatment. IL-6 also decreased BCRP expression, which was, by contrast, induced in response to TNF-α (Fig. 5).

Immunofluorescence studies were next performed to determine whether cytokine treatment may alter cellular distribution of transporters. A characteristic pattern of MRP2-related fluorescence, restricted to canalicular networks of cultured human hepatocytes as previously reported (Hoffmaster et al., 2004), was similarly found in untreated and TNF-α- and IL-6-exposed human hepatocytes (Fig. 6A). In agreement with previous studies (Konig et al., 1999), MRP3 was shown to be localized to basolateral membranes of untreated hepatocytes, and such a cellular location was not altered by TNF-α or IL-6 treatment (Fig. 6B).

Effects of TNF-α and IL-6 Treatment on Transporter Activities. Owing to the limited availability of human hepatocytes, we focused on the effects of cytokine treatments on activities of NTCP, OATPs, and OCT1, whose mRNA expressions were among the most repressed by both TNF-α and IL-6 (Table 2). As shown in Fig. 7, exposure to TNF-α or IL-6 for 48 h resulted in decreased NTCP, OATP, and OCT1 transport activity in human primary hepatocytes, when compared with untreated counterparts.

Correlation Analysis of Cytokine-Repressing Effects on Drug Transporter mRNA Expression in Human Hepatocytes. To search for a putative correlation between the alterations of drug transporter expression occurring in primary human hepatocytes exposed to TNF-α, IL-6, or IL-1β, drug transporters were ranked according to the down-regulation of their mRNA expression in response to a 24-h treatment by cytokines, from data from Table 2. Data reported for IL-1β effects were from Le Vee et al. (2008); they have been obtained using primary human hepatocytes cultured in the same conditions as those used in the present study. For each treatment, transporters were ranked from the most repressed transporter to the less repressed according to mRNA expression levels. Correlations were analyzed using the Spearman's rank correlation method. Results indicated that the effects of IL-1β were significantly correlated with those of TNF-α and IL-6 (Fig. 8). By contrast, TNF-α effects on drug transporter expression were not correlated with those of IL-6 (Fig. 8).

Discussion

The data reported in the present study show that the proinflammatory cytokines TNF-α and IL-6 can markedly alter drug transporter expression in human hepatocytes. These cytokines were found to act in a dose-dependent manner: concentrations such as 100 ng/ml (for TNF-α) and 10 ng/ml (for TNF-α and IL-6) were fully active on transporter mRNA levels, whereas the use of a lower dose of cytokines (1 ng/ml) remained also active, but in a more moderate manner.

Global effects of the cytokines toward transporters after a 48-h treatment are summarized in Table 3. In particular, mRNA levels of hepatocyte uptake transporters NTCP, OATP2B1, OATP1B1, OATP1B3, OCT1, and OAT2 are decreased in response to both TNF-α and IL-6 in primary human hepatocytes; moreover, some of these transporters, i.e., NTCP and OATP1B1, are also down-regulated at protein levels, whereas NTCP, OATP, and OCT1 transport activity are concomitantly reduced. Taken together, these data indicated that these sinusoidal influx transporters, which are well known—for at least some of them—to be down-modulated by TNF-α and IL-6 in rodent liver (Hartmann et al., 2002; Geier et al., 2005), also constitute major targets of these proinflammatory cytokines in human hepatocytes. The down-regulation of NTCP, OATP1B1, and OCT1 mRNA levels in IL-6-exposed primary human hepatocytes from two individuals, recently reported as preliminary data (Nakai et al., 2008), fully supports this conclusion. However, OAT2 mRNA expression, down-regulated in human hepatocytes exposed to either TNF-α or IL-6, has been shown to remain unaffected in response to LPS in rat liver (Cherrington et al., 2004), thus suggesting that this anion transporter may be differentially regulated by inflammatory cytokines in human and rodent hepatocytes.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 5.

Effects of TNF-α and IL-6 on drug transporter protein expression in human hepatocytes. Primary human hepatocytes were either untreated (UNT) or exposed to 100 ng/ml TNF-α or 10 ng/ml IL-6 for 48 h. Transporter protein content was then determined by Western blot analysis. For each transporter, data were quantified by densitometric analysis and expressed relatively to transporter expression found in untreated cells, arbitrarily set at the value of 100%; they are the mean ± S.E.M. of values from four independent hepatocyte populations (top). A representative blot from one hepatocyte population is also shown for each transporter (bottom). *, p < 0.05. P-gp, MDR1/P-glycoprotein.

Besides influx SLC transporters, TNF-α and IL-6 also regulate expression of efflux ABC transporters in human hepatocytes. In contrast to SLC transporters, ABC transporters are rather differentially affected according to the nature of the cytokine (Table 3). Indeed, exposure to IL-6 for 48 h was found to down-regulate mRNA and protein levels of MRP2 and BCRP, whereas TNF-α failed to alter MRP2 mRNA and protein expression and induced protein expression of BCRP. IL-6 was also shown to down-regulate MDR1 mRNA levels in response to a 24- or 48-h treatment; moreover, it diminished MRP4 mRNA expression after an 8- or 24-h exposure. By contrast, TNF-α failed to alter MDR1 and MRP4 mRNA expression, whatever the exposure time. In addition, BSEP mRNA levels were markedly reduced by TNF-α but remained unaffected by IL-6. MRP3 was finally the unique ABC transporter whose regulation was similar in both TNF-α- and IL-6-exposed human hepatocytes, i.e., it was up-regulated at protein levels and remained unaffected at mRNA levels. It is interesting to note that MRP3 has also been shown to be induced at protein levels in human hepatoma HepG2 cells exposed to TNF-α (Bohan et al., 2003). By contrast, MRP2 expression, down-regulated in IL-6-exposed primary human hepatocytes, remained unchanged in IL-6-treated human hepatoma HepG2 cells, whereas MDR1/P-glycoprotein expression was decreased in response to TNF-α in HepG2 cells (Lee and Piquette-Miller, 2003) but not in primary human hepatocytes. Comparison of ABC transporter regulation in cytokine-exposed human hepatocytes with that observed in rodent counterparts also reveals some similarities and some differences: for example, IL-6 decreased MRP2 mRNA levels and failed to significantly alter MRP3 mRNA expression in both primary human hepatocytes and mouse liver, whereas it down-regulated BSEP mRNA levels in mice (Hartmann et al., 2002) but not in human hepatocytes; IL-6 also reduced P-glycoprotein expression in primary rat hepatocytes (Sukhai et al., 2001) but not in human counterparts. In addition, TNF-α reduced BSEP mRNA levels in both human hepatocytes and mouse liver (Geier et al., 2005), whereas it down-regulated MRP2 and MRP3 mRNA expression in mice (Hartmann et al., 2002) but not in human hepatocytes. Taken together, these data, highlighting some differences with respect to transporter regulatory pathways between human hepatocytes and human hepatoma cells and also between human and rodent hepatocytes, underline the interest of the use of primary human hepatocyte cultures for investigating regulation of drug detoxification proteins in humans, as already reported (Gómez-Lechón et al., 2004).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 6.

Effects of TNF-α and IL-6 on MRP2 and MRP3 localization in human hepatocytes. Primary human hepatocytes were either untreated (UNT) or exposed to 100 ng/ml TNF-α or 10 ng/ml IL-6 for 48 h. MRP2 (A) and MRP3 (B) localizations were then analyzed through immunofluorescence and confocal microscopy using monoclonal antibodies directed against MRP2 and MRP3 and corresponding isotypic controls (IgG_2a and IgG₁), as described under Materials and Methods. Arrows indicate MRP2-related red fluorescence of canalicular membranes of primary hepatocytes (A) or MRP3-related red fluorescence of basolateral membranes of primary hepatocytes (B). Blue fluorescence corresponds to 4,6-diamidino-2-phenylindolestained nuclei. Data shown are representative from immunofluorescence studies performed with three independent hepatocyte populations. Bar, 10 μm.

Cellular and molecular mechanisms responsible for TNF-α- and IL-6-related regulation of transporters in human primary hepatocytes remain to be determined. Transcriptional mechanisms may be involved, as already suggested (Sukhai et al., 2000; Geier et al., 2005). In addition, post-transcriptional mechanisms probably have to be considered, especially for the up-regulation of MRP3 expression in response to TNF-α and IL-6 and for that of BCRP in response to TNF-α, which both occur in primary human hepatocytes at protein levels, without concomitant changes in mRNA levels. However, cytokine treatment failed to alter cellular localization of transporters such as MRP2 and MRP3, indicating that putative post-transcriptional regulatory mechanisms triggered by cytokines do not interfere with location of these transporters.

In addition to TNF-α and IL-6, the proinflammatory cytokine IL-1β has been shown to markedly alter expression of major SLC and ABC drug transporters in both human and rodent hepatocytes (Geier et al., 2005; Le Vee et al., 2008) (Table 2). This probably indicates that IL-1β, TNF-α, and IL-6 may act redundantly on at least some hepatic drug transporters. In the same way, in wild-type and knockout mouse strains lacking TNF receptor 1, IL-1 receptor, or IL-6, LPS administration gives a similar pattern of altered hepatic expression of transporters, thus showing that inhibiting one cytokine-related signaling pathway failed to modify transporter regulation in response to LPS, owing to the putative redundant nature of TNF-α, IL-1β, and IL-6 effects toward transporters in mice (Lickteig et al., 2007). It is interesting to note that using rank correlation analysis, IL-1β effect toward transporter expression profile was found to be significantly correlated with those of TNF-α and IL-6. This may suggest some interrelationships or cross-talks between IL-1β signaling pathways and those linked to TNF-α or IL-6, with respect to transporter regulation. However, it is noteworthy that TNF-α and IL-6 effects were, by contrast, not correlated, thus favoring the idea that they mobilized independent signaling pathways and that regulatory pathways shared by TNF-α and IL-1β therefore are different from those shared by IL-6 and IL-1β.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 7.

Effects of TNF-α and IL-6 on drug transporter activities in human hepatocytes. Primary human hepatocytes were either untreated (UNT) or exposed to 100 ng/ml TNF-α or 10 ng/ml IL-6 for 48 h. NTCP, OATP, and OCT1 activities were then determined using radiolabeled substrates as described under Materials and Methods. Data are expressed relatively to transporter activity found in untreated cells, arbitrarily set at the value of 100%; they are the mean ± S.E.M. of values from three independent hepatocytes populations. *, p < 0.05 when compared with untreated cells.

Global repressing effects of TNF-α and IL-6 effects toward influx transporters suggest that in vivo exposure to each of these cytokines is likely to decrease uptake of drugs by the liver. Moreover, TNF-α, unlike IL-6, may impair bile acid secretion through down-regulating BSEP. Finally, it should be kept in mind that liver inflammation is usually associated with increased production of TNF-α, IL-6, and IL-1β. Therefore, the specific effects of these cytokines toward hepatic transporter expression are likely to add up and to overlap, thus putatively resulting in a pronounced down-regulation of hepatobiliary transport of drugs and also of bile acids, and thereby contributing in a notable way to alteration of pharmacokinetics and development of cholestasis occurring during liver inflammation in humans (Moseley, 1997; Slaviero et al., 2003). Down-regulation of SLC and ABC transporters in response to LPS in rats is similarly thought to participate in the impaired hepatic uptake and secretion of drugs and the decreased bile salt secretion occurring in endotoxemic rats (Whiting et al., 1995; Bolder et al., 1997). In this context, up-regulation of the sinusoidal efflux transporter MRP3 by TNF-α and IL-6 may be considered as a protective mechanism contributing to prevent accumulation of toxic compounds, including bile acids, into human hepatocytes during inflammatory processes. A similar beneficial role for induction of Mrp3 during cholestasis in rodents has already been proposed (Donner and Keppler, 2001).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 8.

Rank correlation analysis of cytokine repressing effects on drug transporter mRNA expression in human hepatocytes. Drug transporters were ranked according to the down-regulation of their mRNA expression in response to a 24-h treatment by cytokines, from data from Table 2. For each treatment, transporters were ranked from the most repressed transporter to the less repressed, through considering the repression factor for each transporter, i.e., the ratio of mRNA levels in untreated hepatocytes versus those found in cytokine-exposed cells. Correlations were analyzed using the Spearman's rank correlation method. Spearman's rank coefficients (ρ) and p values are provided on the right of the correlation graphs.

In conclusion, the current data show that the proinflammatory cytokines TNF-α and IL-6 can markedly impair expression of both sinusoidal and canalicular drug transporters in human hepatocytes. Such changes, associated with those triggered by IL-1β (Le Vee et al., 2008) and also with the well established down-regulation of liver drug-metabolizing enzymes such as cytochromes P450 in response to inflammatory cytokines (Aitken and Morgan, 2007), are likely to contribute to alterations of pharmacokinetic features of drugs caused by inflammation in humans.