ISX is a retinoic acid-sensitive gatekeeper that controls intestinal beta,beta-carotene absorption and vitamin A production

Abstract

The uptake of dietary lipids from the small intestine is a complex process that depends on the activities of specific membrane receptors with yet unknown regulatory mechanisms. Using both mouse models and human cell lines, we show here that intestinal lipid absorption by the scavenger receptor class B type 1 (SR-BI) is subject to control by retinoid signaling. Retinoic acid via retinoic acid receptors induced expression of the intestinal transcription factor ISX. ISX then repressed the expression of SR-B1 and the carotenoid-15,15'-oxygenase Bcmo1. BCMO1 acts downstream of SR-BI and converts absorbed beta,beta-carotene to the retinoic acid precursor, retinaldehyde. Using BCMO1-knockout mice, we demonstrated increased intestinal SR-BI expression and systemic beta,beta-carotene accumulation. SR-BI-dependent accumulation of beta,beta-carotene was prevented by dietary retinoids that induced ISX expression. Thus, our study revealed a diet-responsive regulatory network that controls beta,beta-carotene absorption and vitamin A production by negative feedback regulation. The role of SR-BI in the intestinal absorption of other dietary lipids, including cholesterol, fatty acids, and tocopherols, implicates retinoid signaling in the regulation of lipid absorption more generally and has clinical implications for diseases associated with dyslipidemia.

Figure 1.

RA induces ISX expression through RARs in CaCo-2 cells. CaCo-2 cells were seeded in DMEM and 10% FBS. After allowing the cells to adhere for 24 h, cells were treated with 1 μM RA or 1 μM 4-[(E)-2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthalenyl)-1-propenyl]benzoic acid (TTNPB) (a specific RAR agonist) or pretreated with 1 μM cycloheximide (CHX) and then 1 μM RA as indicated. After 4 h, total RNA was extracted and reverse transcribed, and qRT-PCR was performed with a specific probe set for ISX. mRNA levels were normalized for 18S rRNA expression. A) ISX mRNA expression in CaCo-2 cells treated with either RA or TTNPB. B) ISX mRNA levels in CaCo-2 cells pretreated with CHX and then with RA. Results are presented as fold induction vs. untreated control cells (n=3/condition). C) CaCo-2 cells treated with 1 μM RA or without (vehicle control) for 12 h were subjected to indirect immunostaining using the ISX primary antibody and Alexa Fluor-488 secondary antibody as indicated. ISX expression is detected as green fluorescence. Nuclei were also concurrently stained with DAPI, which was included in the mounting medium. Approximately 100 cells/experiment were counted; representative images from 3 independent experiments are shown. Images were acquired at ×40.

Figure 2.

RARs directly bind to the ISX promoter in the ChIP assay. A) Schematic representation of the retinoic acid response element (RARE) in the human ISX promoter region, along with PCR primer pair locations. B) ChIP assays in CaCo-2 cells were performed with the anti-RAR (M-454) antibody specific for human RARs, while IgG was used as a control in the IP reaction. To show binding of RARs to the putative RARE in the ISX promoter region, we designed two primer pairs for its detection in precipitated DNA fractions as indicated. C) Overexpression of ISX in human hepatocyte cells HepG2 decreased SR-BI protein levels. Human hepatocyte HepG2 cells were transiently transfected with either WT ISX or empty vector (V) using Lipofectamine 2000 as indicated. Untransfected parental (P) HepG2 cells were also included as a control. Cells were harvested at 48 h and 72 h post-transfection; total protein was isolated and then subjected to immunoblot analysis as indicated. RAN (Ras-related nuclear protein) was used as the protein loading control. D) Densiometric quantification of SR-BI protein levels as expressed by SR-BI to RAN levels (means ± sd, n=3 independent experiments).

Figure 3.

RA induces ISX expression in vitamin A-deficient Lrat^−/− mice. Eight-week-old Lrat^−/− mice were maintained on a diet lacking vitamin A. After 10 d on this diet, mice were orally gavaged with either RA (0.5 mg/animal) or vehicle control as indicated (n=3/condition). After 24 h, the gavage was repeated. After an additional 24 h, animals were sacrificed, and their small intestines were removed. Total RNA was then extracted; expression of relevant genes was quantified as indicated. A–C) mRNA expression of ISX (A) and its downstream targets SR-BI (B) and BCMO1 (C) was determined using gene specific probe sets (ABI) and qRT-PCR. Values are means ± sd from 2 independent experiments carried out in triplicate. *P ≤ 0.001. Gray bars indicate vehicle-gavaged control mice; solid black bars indicate RA-gavaged mice. D) Total protein from RA or vehicle-treated Lrat^−/− mice (n=3/condition) was obtained from intestinal tissue (duodenum and jejunum), subjected to immunoblot analysis, and probed with the respective antibodies as indicated. E) Densiometric quantification of ISX, SR-BI, and BCMO1 protein levels (means ± sd, n=3). F) Expression of SR-BI in the liver of RA gavaged or control (vehicle-treated) Lrat^−/− mice (n=3/condition) was also estimated by immunoblot analysis as indicated. RAN (Ras-related nuclear protein) was used as the protein loading control.

Figure 4.

Retinoids control intestinal β,β-carotene absorption levels. A) Immunoblot analysis for RBP4 in liver protein extracts from control and BCMO1^−/− mice. B, C) Levels of total vitamin A (all-trans-retinol and retinyl esters) (B) and β,β-carotene in livers of Bcmo1-knockout and WT mice (C). D) β,β-carotene levels in plasma of Bcmo1-knockout and WT mice. Gray bars indicate WT; solid black bars indicate Bcmo1-knockout mice. Values are means ± sd; n = 6 animals/genotype and supplementation group. Group 1: VAD diet with no supplementation; group 2: VAD diet with 150 μg/g β,β-carotene; group 3: VAS diet with 150 μg/g β,β-carotene and a weekly oral dose of 300 IU vitamin A.

Figure 5.

β,β-Carotene induces ISX expression in a BCMO1-dependent manner. Relative intestinal mRNA levels of ISX (A), SR-BI (B), and BCMO1 as determined by qRT-PCR (C). Gray bars indicate WT; solid black bars indicate Bcmo1-knockout mice. Values are means ± sd; n = 3 animals/genotype and supplementation group. Group 1: VAD diet with no supplementation; group 2: VAD diet with 150 μg/g β,β-carotene; group 3: VAS diet with 150 μg/g β,β-carotene and a weekly oral dose of 300 IU vitamin A. n.d., not detectable. *P ≤ 0.001.

Figure 6.

Crosstalk between RAR and ISX signaling controls lipid absorption. A) VAD: SR-BI and BCMO1 expression are increased significantly throughout the small intestine. Enhanced SR-BI activity facilitates the absorption of various lipids, including cholesterol, fatty acids, xanthophylls, tocopherols, and β,β-carotene . B) VAS: retinoids derived either from β,β-carotene conversion or preformed dietary retinoids promote the production of RA. RA binds to RARs, inducing ISX expression. Induction of ISX then leads to the repression of intestinal expression of SR-BI and BCMO1.