L-carnitine and PPARα-agonist fenofibrate are involved in the regulation of Carnitine Acetyltransferase (CrAT) mRNA levels in murine liver cells

Abstract

Background: The carnitine acetyltransferase (CrAT) is a mitochondrial matrix protein that directly influences intramitochondrial acetyl-CoA pools. Murine CrAT is encoded by a single gene located in the opposite orientation head to head to the PPP2R4 gene, sharing a very condensed bi-directional promoter. Since decreased CrAT expression is correlated with metabolic inflexibility and subsequent pathological consequences, our aim was to reveal and define possible activators of CrAT transcription in the normal embryonic murine liver cell line BNL CL. 2 and via which nuclear factors based on key metabolites mainly regulate hepatic expression of CrAT. Here we describe a functional characterization of the CrAT promoter region under conditions of L-carnitine deficiency and supplementation as well as fenofibrate induction in cell culture cells.

Results: The murine CrAT promoter displays some characteristics of a housekeeping gene: it lacks a TATA-box, is very GC-rich and harbors two Sp1 binding sites. Analysis of the promoter activity of CrAT by luciferase assays uncovered a L-carnitine sensitive region within -342 bp of the transcription start. Electrophoretic mobility shift and supershift assays proved the sequence element (-228/-222) to be an L-carnitine sensitive RXRα binding site, which also showed sensitivity to application of anti-PPARα and anti-PPARbp antibodies. In addition we analysed this specific RXRα/PPARα site by Southwestern Blotting technique and could pin down three protein factors binding to this promoter element. By qPCR we could quantify the nutrigenomic effect of L-carnitine itself and fenofibrate.

Conclusions: Our results indicate a cooperative interplay of L-carnitine and PPARα in transcriptional regulation of murine CrAT, which is of nutrigenomical relevance. We created experimental proof that the muCrAT gene clearly is a PPARα target. Both L-carnitine and fenofibrate are inducers of CrAT transcripts, but the important hyperlipidemic drug fenofibrate being a more potent one, as a consequence of its pharmacological interaction.

Figure 1

Quantification of CrAT mRNA levels. (A) TIB-73 murine liver cells were cultivated in DMEM and 10%FCS in comparison to cells treated 24 h with dialyzed FCS. (B) TIB-73 murine liver cells were grown for 24 h with dialyzed FCS and supplemented afterwards for 4 hours with L-carnitine (40–120 μM). (C) Following 24 hours of treatment with dialyzed FCS TIB-73 cells were supplemented with 80 μM L-carnitine for 2–48 h. A second L-carnitine boost (80 μM) took place after 15 hours. Mean value for 0 h L-carnitine suplementation was designated as 1. Supplemented cultures were compared to non-supplemented control (DMEM + 10%diaFCS) in B and C. Values represent means ± SD (n = 3). Means without asterisk show no statistical significance (p > 0.05); (p-values of asterisk marked means are as followed: *p < 0.05, **p < 0.01, ***p < 0.001) (D) TIB-73 cells were grown in DMEM + 10%FCS for 24 hours and afterwards treated with fenofibrate (10–40 μM). Values represent means ± SD (n = 3). Supplemented cultures were compared to physiological control. ***p < 0.001.

Figure 2

PPARα Western blot from nuclear extracts of TIB-73 cells. (A) Cells were treated as described in Figure 1. Nuclear extracts were prepared after 4, 15 and 24 hours of L-carnitine supplementation; values are mean ± SD, n = 3, *p < 0.05 and ***p < 0.001 vs. DMEM + 10% dia.FCS (B) CrAT Western blot from whole cell extracts of TIB-73 cells. Cells were treated as described in Figure 1. values are mean ± SD, n = 3, *p < 0.05 vs. DMEM + 10 % dial.FCS.

Figure 3

Organisation of the murine CrAT promoter. (A) Presentation of the 5′ flanking sequence of the murine CrAT gene. Consensus binding sites are underlined. (B) Schematic structure of the murine CrAT promoter with putative binding sites. mCrAT-1 represents the construct for luciferase-assays ranging from −342 bp to +15 bp and mCrAT-2 from −763 to −328 bp. The artificial promoter construct mCRAT-1 contained the RXR-box, two Sp1 elements and one CACbP region in front of the luc-gene of the pGL2-basic luciferase reporter vector. The promoter construct mCRAT-2 contained three YY-1 sites, two GR1 elements and a HNFβ box in front of the reporter luc-gene.

Figure 4

Activity of different CrAT promoter constructs after supplementation with L-carnitine. TIB-73 cells cultured in medium containing 10% FCS to cause artificial L-carnitine deficiency were cotransfected either with mCrAT-1, mCrAT-2 or mCrAT-3 together with pCMV-ßgal and supplemented with increasing concentrations of L-carnitine for 4 hours. Luciferase activity was normalized for ß-gal activity. Data represent the mean ± SD, n = 4; the mean value for non-supplemented cultures is designated as 1. Comparison of mCrAT-2 and mCrAT-3 constructs at all 3 supplementation levels for 10, 40 and 80 μM L-Carnitine revealed ***p-values (p < 0.001) indicated with parentheses. A Kruskal-Wallis test for all three constructs resulted in slightly less significant p-values (p = 0.094 for 10 μM L-carnitine, p = 0.0002 for 40 and 80 μM L-carnitine).

Figure 5

Protein complexes binding to RXRα binding site at the CrAT promoter. (A) EMSA. Nuclear extracts from TIB-73 cells supplemented with increasing concentrations of L-carnitine were incubated with γ-³²P-labeled oligonucleotides representing the RXRα-binding site with anti-PPARα as indicated. (B) Histogramm of the denstitometrical scan of the EMSA presented in A. Values are mean ± SD, n = 3, ***p < 0.001 and *p < 0.05 vs. Corresponding supplementation levels are indicated with parentheses and asterisks on top to indicate the statistical significance of the p-values.

Figure 6

South-Western Blot of nuclear extracts from TIB-73. (A) cells cultivated in DMEM + 10 % FCS and DMEM + 10 % dialyzed FCS. Marker proteins adjacent to the blot indicate the size range 116 kD, 66 kD and 45 kD. (B) Graphical analysis of 3 distinct bands in kDa (145 ± 3, 70 ± 4, 51 ± 4), values are mean ± SD, n = 4. ***p < 0.001.