Lipopolysaccharide opposes the induction of CYP26A1 and CYP26B1 gene expression by retinoic acid in the rat liver in vivo

Abstract

Retinoic acid (RA), a principal metabolite of vitamin A (retinol), is an essential endogenous regulator of gene transcription and an important therapeutic agent. The catabolism of RA must be well regulated to maintain physiological concentrations of RA. The cytochrome P450 (CYP) gene family CYP26, which encodes RA-4-hydroxylase activity, is strongly implicated in the oxidation of RA. Inflammation alters the expression of numerous genes; however, whether inflammation affects CYP26 expression is not well understood. We investigated the regulation of CYP26A1 and CYP26B1 mRNA levels by RA and LPS in the rat liver, as the liver is centrally involved in retinoid metabolism and the acute-phase response to LPS. Both CYP26A1 and CYP26B1 mRNA were induced in <4 h by a single oral dose of all-trans-RA. RA-induced responses of both CYP26A1 and CYP26B1 were significantly attenuated in rats with LPS-induced inflammation whether LPS was given concurrently with RA or after the RA-induced increase in CYP26A1 and CYP26B1 mRNA levels. When RA and LPS were administered simultaneously (6-h study), LPS alone had little effect on either CYP26A1 or CP26B1 mRNA, but LPS reduced by 80% the RA-induced increase in CYP26A1 mRNA (P<0.02), with a similar trend for CYP26B1 mRNA. When LPS was administered 4 h after RA (16-h study), it abrogated the induction of CYP26A1 (P<0.02) and CYP26B1 (P<0.01). Overall, these results suggest that inflammation can potentially disrupt the balance of RA metabolism and maintenance of RA homeostasis, which may possibly affect the expression of other RA-regulated genes.

Fig. 1

Cytochrome P-450 (CYP)26A1 expression in the liver of rats treated simultaneously with retinoic acid (RA) and LPS. A: Northern blot analysis of CYP26A1 mRNA 6 h after treatment with RA, LPS, and the combination of RA with LPS. Each lane contains poly(A)⁺ RNA pooled equally from 3 rat livers. B: nuclear run-on assay of CYP26A1 gene transcription in the rat liver. Nuclei were isolated from pooled livers of vitamin A-depleted rats and rats treated with an RA analog, Am580, for 3 h (see MATERIALS AND METHODS) and used to synthesize nascent RNA transcripts in the presence of [³²P]UTP. Labeled RNA was extracted and hybridized to empty plasmid vector DNA (top left) or to a plasmid vector containing cDNA of CYP26A1 (top right). β-Actin (bottom left) and GAPDH (bottom right) were blotted onto the Nytran membrane. After a wash, membranes were exposed to X-ray film for 1.5 days for CYP26A1 and 20 h for reference genes. Signals for β-actin and GAPDH were essentially equal for both treatment conditions. C: liver fractionation and Northern blot analysis of CYP26A1 mRNA in rats treated as in B. RNA was prepared from the whole homogenate (Homog), homogenate after filtration (Sup), washed nuclear pellet (Nuclei), and cytosolic fractions (Cytosol) and analyzed as in A. D: quantitative PCR analysis of CYP26A1 transcripts measured in the liver of individual rats (n = 6 rats/group) 6 h after simultaneous treatment with RA (500 μg orally), LPS (50 μg/100 g body wt ip), or the combination of RA + LPS as described in MATERIALS AND METHODS. For comparison, data were normalized to the expression of 18S rRNA and the mean value for the control group for CYP26A1 was set to 1.00, after which the other groups were compared with this value. Bars show means ± SE; statistical results (by a least-squares means test) are displayed.

Fig. 2

CYP26B1 expression in the liver of rats treated simultaneously with RA and LPS. A: quantitative PCR analysis of CYP26B1 transcripts measured in the liver of individual rats (n = 6 rats/group) 6 h after simultaneous treatment with RA (500 μg orally), LPS (50 μg/100 g body wt ip), or the combination of RA + LPS as described in MATERIALS AND METHODS. Data were normalized and referenced to the control group for CYP26A1, set to 1.00 as in Fig. 1D, to compare the relative expression values of these two genes with each other. Bars show means ± SE; statistical results (by a least-squares means test) are displayed. B: dose-response curves showing linear increases in CYP26B1 mRNA with increasing doses of RA (liver RNA samples are from rats in Ref. 49). C: linear regression analysis of log¹⁰-transformed values for CYP26B1 mRNA in A and log¹⁰-transformed values for CYP26A1 mRNA from Fig. 1D.

Fig. 3

Plasma retinol, CYP26A1 and CYP26B1 transcripts, and retinoid metabolites in rats treated with LPS 4 h after the administration of RA. Rats were given 500 μg RA and then, 4 h later, LPS, and plasma and liver samples were collected 12 h later. A: changes in plasma retinol concentrations due to RA and RA + LPS. B: quantitative PCR analysis of CYP26A1. C: quantitative PCR analysis of CYP26A1 transcripts. Data were normalized to the control group and then expressed as means ± SE. D: aqueous-phase metabolites of RA formed after an intravenous administration of a test dose of [³H]RA given 90 min before the end of the experiment to a subset of rats (n = 3–4 rats/group) in the same experiment. Bars (means ± SE) show data for the liver (dark gray bars) and for the lungs, small intestine, spleen, kidney, and liver combined (total bars). The value of 0.061 represents a borderline significant difference between RA- and RA + LPS-treated rats for aqueous-phase metabolites in the liver.

Fig. 4

Poly-I:C (PIC), as well as LPS, abrogates the response of CYP26A1 after its induction by RA. Rats were treated with 500 μg RA or vehicle and then with LPS or PIC or vehicle 4 h later. A: CYP26A1 mRNA was detected by Northern blot analysis using poly(A)⁺-enriched RNA from pooled rat livers from each treatment group. B: loading control, showing the ethidium bromide-stained agarose gel prior to transfer to the Nytran membrane for Northern blot analysis. C: RT-PCR am-plification of rat liver RNA using [³³P]dATP incorporation during PCR (see MATERIALS AND METHODS). RNA from n = 3–4 individual rats/group was subjected to PCR analysis. Bars show means ± SE. The inset shows results from 1 or 2 rats/group visualized by X-ray film exposure of the polyacrylamide gel before the gel was cut and counted.