Abstract
Citrobacter rodentium is the rodent equivalent of human enteropathogenic Escherichia coli infection. This study investigated regulation of hepatic and renal cytochrome P450 (P450) mRNAs, hepatic P450 proteins, cytokines, and acute phase proteins during C. rodentium infection. Female C3H/HeOuJ (HeOu) and C3H/HeJ (HeJ) mice [which lack functional toll-like receptor 4 (TLR4)] were infected with C. rodentium by oral gavage and sacrificed 6 days later. Hepatic CYP4A10 and 4A14 mRNAs were decreased in HeOu mice (<4% of control). CYP3A11, 2C29, 4F14, and 4F15 mRNAs were reduced to 16 to 55% of control levels, whereas CYP2A5, 4F16, and 4F18 mRNAs were induced (180, 190, and 600% of control, respectively). The pattern of P450 regulation in HeJ mice was similar to that in HeOu mice for most P450s, with the exception of the TLR4 dependence of CYP4F15. Hepatic CYP2C, 3A, and 4A proteins in both groups were decreased, whereas CYP2E protein was not. Renal CYP4A10 and 4A14 mRNAs were significantly down-regulated in HeOu mice, whereas other P450s were unaffected. Most renal P450 mRNAs in infected HeJ mice were increased, notably CYP4A10, 4A14, 4F18, 2A5, and 3A13. Hepatic levels of interleukin (IL)-1β, IL-6, and tumor necrosis factor α (TNFα) mRNAs were significantly increased in infected HeOu mice, whereas only TNFα mRNA was significantly increased in HeJ mice. Hepatic α1-acid glycoprotein was induced in both groups, whereas α-fibrinogen and angiotensinogen were unchanged. These data indicate that hepatic inflammation induced by C. rodentium infection is mainly TLR4-independent and suggest that hepatic P450 down-regulation in this model may be cytokine-mediated.
Bacterial infections and/or inflammation suppress both hepatic and extrahepatic cytochrome P450 (P450) expression and metabolism (Morgan, 2001; Renton, 2004), which can result in altered drug responses. Lipopolysaccharide (LPS; endotoxin) is the major constituent of the outer membrane of Gram-negative bacteria and is a principal mediator of the inflammatory response to invading pathogens. Toll-like receptor (TLR) 4 is responsible for LPS signaling (Poltorak et al., 1998; Hoshino et al., 1999). TLR4 specifically responds to LPS in association with several proteins, including LPS-binding protein, CD14, and MD-2 proteins (Beutler et al., 2001). LPS has been used extensively as a sterile model of sepsis to study the down-regulation of hepatic P450s during inflammation. The LPS model has proven invaluable in understanding the importance and mechanisms of hepatic P450 suppression. However, the LPS model may not accurately predict how P450 enzymes will be regulated in other models of inflammation or in live bacterial infection.
As a live bacterial infection model, Citrobacter rodentium is the rodent equivalent of human enteropathogenic Escherichia coli (EPEC) infection. The colonic pathology elicited by C. rodentium is indistinguishable from that of EPEC (Schauer and Falkow, 1993), with characteristic attaching and effacing lesions on intestinal cells (Goosney et al., 2000). EPECs are a specific serotype of E. coli that are the major cause of infantile diarrhea worldwide (Nataro and Kaper, 1998), and contaminate food and water supplies. C. rodentium, a member of the Enterobacteriaceae family and a natural murine pathogen, provides an excellent model for human EPEC infection and colitis (Jurjus et al., 2004; Wales et al., 2005). The colitis caused by C. rodentium infection is also characteristic of inflammatory bowel disease (IBD) in mice and humans, and C. rodentium is considered a model for IBD (Higgins et al., 1999; Caradonna et al., 2000; Goncalves et al., 2001; Gobert et al., 2004).
Chemically induced colitis models have been developed for determination of mechanisms of IBD. Experimental models of inflammation mimicking human ulcerative colitis include treatment with dextran sulfate sodium (DSS) in drinking water for 6 to 10 days or intrarectal treatment with 2,4,6-trinitrobenzene sulfonic acid (Okayasu et al., 1990). However, very little is known about the impact of colitis on hepatic P450 function or expression in humans or animals. Weidenbach et al. (2000) studied hepatic P450 activity in perfused livers from rats with 2,4,6-trinitrobenzene sulfonic acid-induced colitis, finding decreased metabolic capacity at 1 to 2 days. Hepatic P450 activity toward lidocaine was decreased by approximately 30% 1 to 2 days after colitis induction, with recovery at 7 days. More recently, Masubuchi and Horie (2004) studied the effect of DSS-induced colitis on rat hepatic P450 activities. CYP3A2, 2C11, 1A2, and 2E1 activities were down-regulated, but CYP2D2 was not. P450 down-regulation during DSS-induced colitis was prevented by treatment with polymyxin B (CYP3A2, 2E1) or metronidazole (CYP3A2, 2C11, 2E1) (Masubuchi and Horie, 2004), indicating that endotoxins of commensal bacteria are probably involved in some of the effects. Portal blood endotoxin levels were elevated in these rats, but liver weight and serum alanine aminotransferase enzymes were normal, suggesting little liver damage. However, it is unclear from these studies whether these changes in P450 activity reflected changes in gene expression.
We hypothesized that the LPS model may not predict how P450 enzymes will be regulated in other models of inflammation, such as a live infection. Therefore, we investigated hepatic and renal P450 expression during C. rodentium bacterial infection in wild-type mice and mice deficient in functional TLR4, a critical component of LPS signaling. We found that C. rodentium infection produced effects on hepatic P450 expression that were more enzyme-selective than the effects produced by LPS injection, and were largely independent of TLR4 activation.
Materials and Methods
Bacteria. A wild-type strain of C. rodentium (#51116) was obtained from The American Type Culture Collection (Manassas, VA). Before infection, C. rodentium was grown in Luria broth without shaking overnight, and harvested by centrifugation. Bacteria were resuspended in phosphate-buffered saline, and concentration was determined by retrospective plating on MacConkey agar, on which C. rodentium forms small pink colonies with white rims.
Chemicals, Animals, and Treatments. Unless otherwise specified, all reagents and chemicals were obtained from Sigma-Aldrich (St. Louis, MO). Female wild-type (C3H/HeOuJ; HeOu) and TLR4 mutant (C3H/HeJ; HeJ) mice were obtained from Jackson Laboratory (Bar Harbor, ME). These mice derive from the same parent strain and are genetically very similar, except that C3H/HeJ mice incurred a spontaneous inactivating mutation in the TLR4 gene approximately 40 years ago (http://jaxmice.jax.org). The mice were acclimatized to the animal facility for 1 week, and mice were 4 to 5 weeks of age at the time of infection. Mice were housed in a Biosafety Level-2 facility to prevent transmission of infection to other mouse colonies. Two groups of mice were infected with 2.0 × 108 colony-forming units of C. rodentium in saline by gavage. Control groups for each genotype were administered saline by gavage the day after the infected group, and were pair-fed; i.e., each day they received the amount of food eaten by the infected group on the prior day. Mice were pair-fed to control for the possible role of reduced food intake in P450 regulation during infection. Mice were sacrificed 6 days after administration of saline or bacteria. Livers and kidneys were collected, rinsed in cold 1.15% potassium chloride, and stored at –80°C until RNA or microsome preparation. Four or six mice were in each group (n = 4, HeJ infected group; n = 6, HeOu pair-fed, HeOu infected, HeJ pair-fed). The Institutional Animal Care and Use Committee of Emory University approved all procedures.
Preparation of Total RNA. Total liver and kidney RNAs were prepared using RNA-Bee isolation reagent according to the manufacturer's instructions (Tel-Test, Friendswood TX). Total RNA concentration was determined spectrophotometrically by measuring absorbance at 260 nm, and RNA purity and integrity were confirmed by formaldehyde-agarose gel electrophoresis followed by visualization with ethidium bromide.
Microsome Preparation. Liver microsomes were prepared by differential centrifugation and stored at –80°C (Haugen and Coon, 1976). Microsomal protein concentrations were determined by the method of Lowry et al. (1951) using bovine serum albumin as the standard.
cDNA Synthesis. Purified total RNA was reverse-transcribed using the SuperScript First-Strand Synthesis System for the reverse transcriptase-polymerase chain reaction (RT-PCR) kit (Invitrogen, Carlsbad, CA), according to the manufacturer's protocol.
Primer Sequences. Primers for mouse P450s (excluding CYP4Fs), cytokines, acute phase proteins, and glyceraldehyde dehydrogenase (GAPDH) were designed using the Primer Select software program (DNASTAR, Inc., Madison, WI). To exclude cross-reactivity with other mouse P450 sequences, as well as other enzymes, all primers were submitted to the National Center for Biotechnological Information for nucleotide comparison by the basic local alignment search tool (BLASTn; Altschul et al., 1990). Oligonucleotides with a high degree of similarity (>80%) to other mouse P450 mRNA transcripts were eliminated from further consideration. Primers were custom-synthesized on a 50-nmol scale by MWG Biotech, Inc. (High Point. NC), and obtained desalted and lyophilized. Primers were diluted to 100 μM in deionized water and stored at –80°C. For mouse P450 4F genes, specific sequences for PCR primers and dual-labeled fluorescent probes were designed at the region of highest sequence specificity or intron/exon boundaries using Primer Express software (Applied Biosystems, Foster City, CA) and custom synthesized by Integrated DNA Technologies, Inc. (Coralville, IA). The isoform specificity of P450 primers and probes was confirmed through multiple alignments with the other members of the CYP4F subfamily as well as homology-searched to ensure that there was no cross reactivity with other genes using an National Center for Biotechnological Information BLAST search. All primer sequences have been published previously (Pan et al., 2000; Overbergh et al., 2003; Richardson and Morgan, 2005), except for the CYP4F primer and probe sequences (L. Antonovic, T. A. Richardson, E. T. Morgan, and H. W. Strobel, manuscript in preparation).
Quantitative Real-Time RT-PCR. With the exception of the CYP4Fs, real-time RT-PCR was performed using the ABI PRISM 7000 Sequence Detection System and SyBr Green Master Mix reagent (Applied Biosystems, Bedford, MA) to determine expression of mRNAs of interest in mouse liver and kidney, as described previously by Richardson and Morgan (2005). Briefly, reactions were performed in a total volume of 25 μl using SyBr Green Master Mix reagent (Applied Biosystems); 2 μl of cDNA/sample was used as template for the reaction, with 10 μM forward and reverse primers. P450 and GAPDH cDNA amplification was performed in duplicate wells using the same sample. Thermal cycling conditions included 2 min at 50°C and 10 min at 95°C, followed by 40 cycles of 95°C for 15 s and 1 min at the appropriate annealing temperature. This technique allows identification of the threshold cycle (Ct) value when PCR product is first detectable, using fluorescence emission. To normalize the amount of total mRNA present in each reaction, levels of the housekeeping gene GAPDH were monitored in parallel samples. Results are expressed as relative levels of P450 mRNA, as defined by the Ct method described by Livak and Schmittgen (2001). Control samples were chosen to represent 1× expression of the gene. The amount of P450 mRNA in treated samples was calculated relative to the control P450 sample. All primer sets yielded a single PCR product of expected size by agarose gel electrophoresis. Specificity was routinely monitored by checking product melting curves (dissociation curves) in each reaction well.
For analysis of mouse P450 4Fs, quadruplicate aliquots of total RNA (200 ng) for each sample were reverse-transcribed, including a blank [without the reverse transcriptase (RT) enzyme] to account for amplification of contaminating genomic DNA, at 50°C for 30 min, followed by 72°C for 5 min. Ten-microliter RT sample reactions consisted of 4 μl of RNA, 1× SSII buffer, 300 nM reverse primer, 500 μM deoxynucleoside-5′-triphosphates, and 10 U/10 μl SuperScript II (Invitrogen). After this, the QPCR reaction was carried out by adding 40 μl of QPCR mix (containing 1× PCR buffer, 300 nM forward primer, 300 nM reverse primer, 4 mM MgCl2, 2.5 U/50 μl Taq polymerase, and 100 nM fluorogenic probe) to each RT sample. Amplification was performed using an ABI Prism 7700 (Applied Biosystems) at 95°C for 1 min, followed by 40 cycles of 95°C for 15 s, and 60°C for 1 min. During the QPCR reaction, an increase in fluorescence generated by the increase in cleaved reporter dye is detected, normalized for its reporter signal fluorescence, and plotted by the instrument versus cycle number. Standard curves for individual CYP4F isoforms were generated by plotting Ct versus the log of their amplicon amount (custom made by Integrated DNA Technologies, Inc.) in the range of 200 ng to -2 pg, and were used to calculate the relative amount of a particular CYP4F mRNA in the samples. Calculations for Ct and standard curves were performed using the instrument software (Heid et al., 1996). Data were analyzed and the absolute amount of each CYP4F mRNA in samples was determined by normalizing the values for copy number of the gene of interest to the copy number values of m-cyclophilin, used as an internal standard. For comparison to the other P450s, CYP4F data are presented in relative amounts of RNA, rather than absolute RNA amounts.
Western Immunoblotting. P450 protein levels in mouse hepatic microsomes were measured by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blotting, as described previously by Richardson and Morgan (2005). Antibodies to rat CYP3A2, 4A1, and 2E1 were generously provided by Dr. James Halpert (University of Texas Medical Branch, Galveston, TX), Dr. Gordon Gibson (University of Surrey, Guildford, UK), and Dr. Magnus Ingelman-Sundberg (Karolinska Institute, Stockholm, Sweden), respectively. Polyclonal antibodies to rat CYP3A2, 4A1, and 2E1 proteins were diluted 1:5000, whereas 2C11 antibody was diluted 1:20,000. Secondary antibodies were as follows: goat anti-rabbit, 3A, 2C, and 2E; rabbit anti-sheep, 4A; dilution for each was 1:2500, with the exception of 2C, which was 1:10,000. All assays were performed within a linear range, and the intensity of stained bands was measured by laser densitometry.
Statistical Analysis. Control and experimental groups were compared by Student's t test.
Results
C. rodentium-infected mice exhibited overt signs of infection, with diarrhea and body weight loss. Body weights of all mice were stable for the first 4 days of the experiment. The body weights of infected HeJ mice and their pair-fed controls declined from 19 g on day 4 to 17 g on day 7, with no significant difference between the groups (data not shown). The body weights of infected HeOu mice declined similarly in this period, but the body weights of pair-fed HeOu mice remained stable. C3H/HeJ and HeOu mice are highly susceptible to infection, requiring 4 days for bacterial colonization, and death occurring between 6 and 10 days postinfection (Vallance et al., 2003). Two of the six mice in the HeJ infected group died during the exposure time, with a resultant n = 4 for the HeJ infected group. All mice in the other groups (n = 6) survived the entire exposure period.
Effect ofC. rodentiumInfection on Hepatic P450 mRNA and Protein Expression. Bacterial infection significantly down-regulated the hepatic expression of CYP2C29, 3A11, 4A10, 4A14, 4F14, and 4F15 mRNAs, but not CYP1A2, 2E1, 3A13, and 4F13 mRNAs in HeOu mice (Figs. 1 and 2). Of the P450 isoforms studied, CYP4A10 and 4A14 mRNAs were the most affected by C. rodentium infection (reduced to less than 4% of control) (Fig. 2). CYP3A11 mRNA was down-regulated to 16% of control, whereas CYP2C29, 4F14, and 4F15 mRNAs exhibited intermediate reductions to 50%, 55%, and 48% of control, respectively. In contrast to the other P450 isoforms, hepatic CYP2A5 and CYP4F16 mRNAs were induced to 180% and 188% of control levels, respectively, in infected HeOu mice, and CYP4F18 mRNA levels were induced to 600% of control (Figs. 1 and 2).
The pattern of P450 regulation after infection in HeJ mice was similar to that in HeOu mice for most mRNAs (CYP2C29, 3A11, 3A13, 4A10, 4A14, 4F14, 4F18). In contrast to the effects in HeOu mice, hepatic CYP1A2 mRNA expression in HeJ mice was decreased to 24% of control, and CYP2A5 and 4F16 mRNAs were unaffected (Figs. 1 and 2).
C. rodentium infection decreased P450 protein expression (2C, 3A, and 4A) in both HeOu and HeJ mice (Fig. 3), which was consistent with mRNA results. Again, CYP4A proteins were the most significantly affected by C. rodentium infection, decreasing to 6% and 18% of control in HeOu and HeJ mice, respectively (Fig. 3).
Effect ofC. rodentiumInfection on Renal P450 mRNA Expression. Bacterial infection had no significant effect on most of the renal P450 isoforms studied in HeOu mice (CYP1A2, 2A5, 2C29, 2E1, 3A11, and 3A13) (Fig. 4). CYP4A10 and 4A14 mRNAs were significantly down-regulated by C. rodentium infection in kidneys of HeOu mice, with reductions to 43% and 34% of control, respectively (Fig. 5), whereas a tendency toward up-regulation of CYP4F18 mRNA in HeOu mice missed significance (p = 0.07). In contrast, renal P450 mRNAs in HeJ mice tended to increase after C. rodentium infection, with significant increases for CYP2A5 (500% of control), 3A13 (740% of control), 4A10 (170% of control), 4A14 (270% of control), and 4F18 (270% of control) mRNAs (Figs. 4 and 5).
Effect ofC. rodentiumInfection on Hepatic and Renal Nuclear Receptor mRNA Expression. Hepatic peroxisome proliferator-activated receptor α (PPARα) mRNA levels tended to decrease in both HeOu and HeJ mice after C. rodentium infection, but this was only statistically significant in the HeJ strain (50% of control) (Fig. 6). In contrast, renal PPARα mRNA levels were unaffected in either strain. Hepatic expression of retinoid X receptor α (RXRα) mRNA, the heterodimer partner of PPARα, tended to increase in both HeOu and HeJ mice (225% and 128% of control), although this parameter was significant only for HeOu mice (Fig. 6). Renal RXRα mRNA levels tended to increase in HeOu and HeJ mice (300% and 230%), although both effects narrowly missed significance (p = 0.067 for HeOu mice and p = 0.056 for HeJ mice).
Effect ofC. rodentiumInfection on Hepatic Cytokine and Acute Phase Protein mRNA Expression. As expected, C. rodentium infection increased hepatic mRNA expression of proinflammatory cytokines and acute phase proteins (Fig. 7). Hepatic mRNA levels of IL-1β (450% of control), IL-6 (193% of control), and TNFα (650% of control) were significantly increased in infected HeOu mice. Similar to HeOu mice, IL-1β and IL-6 tended to increase in HeJ mice, but only TNFα mRNA was significantly increased (200% of control). Hepatic α1-acid glycoprotein was induced in infected HeOu (690%) and HeJ (400%) mice, whereas α-fibrinogen and angiotensinogen were unaffected (Fig. 7).
Discussion
Injection of LPS is a well characterized inflammation model, and treatment of rodents or hepatocytes with LPS suppresses multiple P450 mRNAs and proteins and induces acute phase proteins (Morgan, 2001; Renton, 2004). We hypothesized that the specificity of P450 regulation may be different in live infections than in the LPS model, because other bacterial components may be involved and because the profile, time course, and sources of cytokines affecting the hepatocyte are likely to be dissimilar. To our knowledge, this is the first report investigating hepatic and extrahepatic P450 regulation using C. rodentium as a model of infection. Consistent with our hypothesis, we found that the pattern of hepatic P450 regulation by C. rodentium infection was dramatically different from and more enzyme-specific than that produced by LPS. The relatively specific targeting of CYP4A and CYP4Fproteins, which have important roles in metabolism of arachidonic acid and other eicosanoids, supports the concept that P450 regulation in infection may be important in modulating the inflammatory and vascular responses via these metabolites (Morgan, 2001). Furthermore, hepatic P450 mRNA expression was similarly down-regulated in both wild-type (HeOu) and TLR4-mutant (HeJ) mice, indicating that hepatic P450 down-regulation during infection is largely independent of TLR4-LPS signaling. Patterns of renal P450 regulation were different in wild-type mice and TLR4-mutant mice, also, and may indicate TLR4 involvement in regulation of renal P450 isoforms.
Hepatic P450 expression during C. rodentium infection is more specifically regulated than in the LPS model of inflammation. Intraperitoneal injection of LPS down-regulates the expression of several mouse hepatic mRNAs, with significant reductions in CYP1A2, 2A5, 2C29, 2E1, 3A11, 4A10, and 4A14 mRNAs at 16 h (Richardson and Morgan, 2005), and 4F15 and 4F16 mRNAs at 24 h (Cui et al., 2001). The effect of C. rodentium infection on hepatic P450 mRNAs exhibits both similarities to and striking differences from LPS administration. CYP2C29, 3A11, 4A10, 4A14, and 4F15 mRNAs are likewise down-regulated both during LPS exposure (Cui et al., 2001; Richardson and Morgan, 2005) and in infected HeOu mice. CYP1A2, 2E1, 2A5, and 4F16 mRNAs, each down-regulated by LPS, are either unchanged or induced by C. rodentium infection. Notably, hepatic CYP3A13 mRNA expression is unaffected by LPS administration or infection (Fig. 2; Richardson and Morgan, 2005).
Both C. rodentium infection and LPS administration suppress CYP4A10 and 4A14 mRNAs, but CYP4A mRNAs are more affected by bacterial infection as compared with LPS exposure. Infection nearly abolished expression of both CYP4A10 and 4A14 mRNAs (<4% of control), whereas a 16-h LPS exposure decreased CYP4A10 and CYP4A14 to about 20 to 30% of control (Richardson and Morgan, 2005). Down-regulation of CYP4A proteins (Fig. 3) during infection was consistent with results for CYP4A mRNAs. Bacterial infection seems to have a far more dramatic effect on CYP4A mRNAs and proteins compared with other P450 subfamilies, These data indicate that exposure to a single bacterial component, LPS, cannot predict the response to a live infection, in which several bacterial components may be a factor.
Most of the hepatic P450 mRNAs studied during infection were significantly down-regulated in both HeOu and HeJ mice, indicating that the effects are TLR4-independent. C. rodentium infection induced hepatic CYP2A5 mRNA to 180% of control in HeOu mice (Fig. 1), but had no effect in HeJ mice, indicating that this up-regulation is dependent on TLR4. Down-regulation of CYP1A2 in HeJ mice could indicate that maintenance of CYP1A2 mRNA levels in HeOu mice is TLR4-dependent, also. CYP4F15 was significantly decreased to 48% of control in HeOu mice, and this effect was not apparent in HeJ mice, indicating that the decrease of 4F15 mRNA in HeOu mice is TLR4-dependent (Figs. 1 and 2).
Renal P450s during infection also differed significantly from their regulation in the LPS model. Unlike hepatic CYP4A mRNAs, which are down-regulated during inflammation, renal CYP4A mRNAs are induced in both rat and mouse after LPS treatment (Sewer et al., 1997; Barclay et al., 1999; Mitchell et al., 2001). Mouse renal CYP4F15 and 4F16 mRNAs are induced and unaffected, respectively, by LPS (Cui et al., 2001). Infection had no significant effect on most renal P450 isoforms in HeOu mice, except for CYP4A10 and 4A14 mRNAs, which were significantly down-regulated. Renal CYP4F15 mRNA is induced after LPS treatment, but remained unchanged during infection, as does CYP4F16 (Fig. 5).
In contrast to HeOu mice, several renal P450 mRNAs in HeJ mice were significantly increased after C. rodentium infection. This may be a compensatory effect for down-regulation of hepatic P450s, as has been suggested for Fasciola hepatica-infected rats with decreased hepatic P450 activities, which also have increased renal P450 activities (Biro-Sauveur et al., 1995). It is unclear why this increase in renal P450s is only seen in HeJ mice, and not in HeOu mice.
C. rodentium infection reduced expression of hepatic CYP4A10 and 4A14 mRNAs to less than 4% of control and renal expression to 30 to 40% of control levels. Because CYP4A regulation is mediated by PPARα, it could be surmised that down-regulation of CYP4A mRNA after infection could be due to down-regulation of PPARα mRNA and RXRα. Hepatic PPARα mRNA was decreased to 50% of control in both HeOu and HeJ mice. Hepatic RXRα levels were significantly increased in HeOu mice and tended to increase in HeJ mice (Fig. 6), unlike the LPS-mediated decrease in RXRα levels (Beigneux et al., 2000). It seems unlikely that the striking down-regulation of hepatic CYP4A10 and 4A14 mRNAs after infection (<4% of control) is due to this modest down-regulation of PPARα mRNA. Surprisingly, renal mRNA levels of PPARα and RXRα both tended to increase after infection. The mechanism for CYP4A mRNA down-regulation in this model requires further study.
Expression of CYP2A mRNAs and proteins is inducible by inflammation, microbial and parasitic infection, and cancer (Montero et al., 1999; Su and Ding, 2004). C. rodentium infection induced hepatic CYP2A5 mRNA to 180% of control in HeOu mice (Fig. 1), but had no effect in HeJ mice, indicating that this effect may be dependent on TLR4. Although the mechanism of CYP2A5 induction remains unknown, hepatic inflammation could be a precursor to induction (Su and Ding, 2004). It has also been suggested that CYP2A5 induction by Helicobacter hepaticus is mediated via increased production of reactive oxygen species (Su and Ding, 2004).
With the exception of CYP3A11, 4A10, and 4F15, basal hepatic P450 mRNAs tended to be higher in HeJ mice compared with HeOu mice (Figs. 1 and 2), suggesting that TLR4 may somehow function in regulation of P450 isoforms in the healthy animal. Alternatively, the deficiency of this gene in mouse development could affect P450 expression. We are unaware of any significant genetic difference between the strains, other than TLR4 deficiency, that could explain the difference. Most renal P450 mRNAs in HeJ mice were similar to or lower than that in HeOu mice (except CYP1A2 and 2E1). Thus, the influence of TLR4 on basal P450 regulation is both tissue- and isoform-specific.
Down-regulation of multiple P450s during LPS-induced inflammation can be mimicked by in vivo and in vitro treatment with proinflammatory cytokines such as IL-1β, IL-6, and TNFα (Morgan, 2001). C. rodentium infection increased hepatic cytokine mRNAs and α1-acid glycoprotein in both HeOu and HeJ mice (Fig. 7), indicating activation of Kupffer cells. However, the lack of effect on fibrinogen and angiotensinogen suggests that the level of hepatic inflammation produced during infection is mild. Despite this, P450 enzymes are sensitive to the resulting hepatic inflammation, particularly the CYP4A isoforms. The mechanism for the selective down-regulation of hepatic P450s during infection and experimental colitis must be investigated further, although cytokine-mediated effects are a likely possibility.
Signaling mechanisms other than TLR4 exist, including signaling through other toll-like receptors such as TLR2 (Yang et al., 1998, 1999), or signaling through the release of bacterial lipoprotein (LP) (Zhang et al., 1997). Human TLR2-transfected HEK 293 cells respond to LPS, require LPS-binding protein and CD14, and activate nuclear factor κB (Yang et al., 1998, 1999). Bacterial LP induces TNFα and IL-6 production in vivo in LPS-responsive and -nonresponsive mice, and acts synergistically with LPS to induce lethal shock and cytokine production (Zhang et al., 1997). Bacterial culture supernatants containing LP induce IL-6 production in macrophages obtained from LPS-nonresponsive mice (Zhang et al., 1998). Bacterial LP activates TLR2 (Aliprantis et al., 2000), and rodent hepatocytes express TLR2 mRNA that can be up-regulated by LPS and regulated, in part, by IL-1 and TNFα (Liu et al., 2000; Matsumura et al., 2000). Additional studies are needed to investigate other signaling mechanisms for C. rodentium infection.
In summary, our hypothesis that the specificity of P450 regulation may be different, in live infections, from that in the LPS model was supported by the experimental data. Hepatic inflammation induced by C. rodentium infection is mainly TLR4-independent because hepatic P450 mRNA expression was similarly down-regulated and cytokine mRNAs were similarly induced, in both wild-type and TLR4-mutant mice. In addition, renal P450 expression patterns were different in wild-type mice and TLR4-mutant mice, and TLR4 may be involved in their regulation. The results suggest that the hepatic P450 down-regulation in this model may be cytokine-mediated.
Acknowledgments
We thank Kimberly L. Pierce for excellent technical assistance. We also thank Dr. Gary W. Miller, Emory University Center for Neurodegenerative Disease and the Department of Environmental and Occupational Health, for the use of equipment.
Footnotes
-
The National Institutes of Health Grants GM46897 and DK072372 (E.T.M.), NS44174 (H.W.S.), AR002157 (M.S.), and AI056067-01 (D.K.) provided funding for this study. Portions of this work were presented at the 2005 Experimental Biology Meeting, San Diego, California, April 2–6, 2005.
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.105.007393.
-
ABBREVIATIONS: P450, cytochrome P450; TLR, toll-like receptor; IL, interleukin; TNFα, tumor necrosis factor α; LPS, lipopolysaccharide; PPARα, peroxisome proliferator-activated receptor α; EPEC, enteropathogenic Escherichia coli; DSS, dextran sulfate sodium; IBD, inflammatory bowel disease; RT-PCR, reverse transcription-polymerase chain reaction; GAPDH, glyceraldehyde phosphate dehydrogenase; HeOu, C3H/HeOuJ; HeJ, C3H/HeJ; Ct, threshold cycle; QPCR, quantitative PCR; RXRα, retinoid X receptor α; LP, lipoprotein.
- Received September 15, 2005.
- Accepted December 2, 2005.
- The American Society for Pharmacology and Experimental Therapeutics