Transcriptional regulation of Acsl1 by CHREBP and NF-kappa B in macrophages during hyperglycemia and inflammation

Abstract

Acyl-CoA synthetase 1 (ACSL1) is an enzyme that converts fatty acids to acyl-CoA-derivatives for lipid catabolism and lipid synthesis in general and can provide substrates for the production of mediators of inflammation in monocytes and macrophages. Acsl1 expression is increased by hyperglycemia and inflammatory stimuli in monocytes and macrophages, and promotes the pro-atherosclerotic effects of diabetes in mice. Yet, surprisingly little is known about the mechanisms underlying Acsl1 transcriptional regulation. Here we demonstrate that the glucose-sensing transcription factor, Carbohydrate Response Element Binding Protein (CHREBP), is a regulator of the expression of Acsl1 mRNA by high glucose in mouse bone marrow-derived macrophages (BMDMs). In addition, we show that inflammatory stimulation of BMDMs with lipopolysaccharide (LPS) increases Acsl1 mRNA via the transcription factor, NF-kappa B. LPS treatment also increases ACSL1 protein abundance and localization to membranes where it can exert its activity. Using an Acsl1 reporter gene containing the promoter and an upstream regulatory region, which has multiple conserved CHREBP and NF-kappa B (p65/RELA) binding sites, we found increased Acsl1 promoter activity upon CHREBP and p65/RELA expression. We also show that CHREBP and p65/RELA occupy the Acsl1 promoter in BMDMs. In primary human monocytes cultured in high glucose versus normal glucose, ACSL1 mRNA expression was elevated by high glucose and further enhanced by LPS treatment. Our findings demonstrate that CHREBP and NF-kappa B control Acsl1 expression under hyperglycemic and inflammatory conditions.

Fig 1. Acsl1 expression is upregulated in macrophages by HG.

A) BMDMs were differentiated under normal glucose (NG) and high glucose (HG), and Acsl1 mRNA absolute copy number was determined by quantitative real-time PCR (qPCR) using a standard curve generated with known quantities of mouse Acsl1 cDNA. B) Western blot of total cell lysates from BMDMs cultured in NG and HG using antibodies against ACSL1 and β-actin as (loading control). C) The ratio of ACSL1/β-actin band intensities was quantified using Image studio 5.2 (n = 3). D) Nascent Acsl1 RNA expression in BMDMs under NG and HG conditions was determined by qPCR. Nascent Acsl1 expression was quantified relative to cyclophilin A and shown as fold change. The data presented are the means ± standard error (n = 3); the p-value was calculated using the student’s t-test. (*p < 0.05).

Fig 2. CHREBP contributes to transcriptional upregulation of Acsl1 in HG.

A) CHREBP expression plasmid or vector only (VO) were co-transfected with Acsl1-GLuc reporter in HEK293 cells cultured under HG conditions. Luciferase assay was performed 48 hours post-transfection and shown as relative luciferase units (RLU). Expression of transfected Myc-tagged CHREBP is shown by western blot using a c-Myc antibody, with tubulin as a loading control. B) Wild type or Chrebp^-/- BMDMs were differentiated under HG conditions. Acsl1 mRNA absolute copy number was quantified by qPCR. The data presented are means ± standard error (n = 3); the p-value was calculated using student’s t-test. (*p < 0.05).

Fig 3. LPS induction of Acsl1 expression via NF-kappa B.

A) BMDMs were differentiated under NG and HG conditions. Cells were stimulated with LPS (10ng/mL) for 24 hours. Total RNA was isolated, and Acsl1 mRNA was determined by qPCR relative to cyclophilin A1 and shown as fold change. NG treated sample (M0) was set to 1. B) BMDMs were differentiated in HG. Cells were pretreated for 4 hours with NF-kappa B inhibitor CAPE (5 μM) and then treated with LPS for 16 hours. RNA was isolated, and Acsl1 mRNA copy number was determined by qPCR. The data presented are means ± standard error (n = 3); the p-value was calculated using one-way ANOVA (p < 0.05; **p < 0.01; and ***p < 0.001).

Fig 4. ACSL1 protein abundance and membrane localization increase under inflammatory conditions.

BMDMs were differentiated under NG conditions. A) Total cell lysates were prepared with (+) and without (-) LPS treatment (10 ng/ml for 24 hours). ACSL1 protein expression was determined by western blot with an anti-ACSL1 antibody. An anti-alpha tubulin antibody was used as a loading control. B) Bands were quantified using the Image studio 5.2. C) Cytoplasmic and membrane proteins were isolated to determine the localization of ACSL1 protein by western blotting as a function of LPS treatment. Alpha-tubulin (cytoplasmic protein) and Na+/K+ ATPase (membrane-associated protein) were used to confirm the fidelity of fractionation. D) Images were quantitated as in B with cytoplasmic ACSL1 protein normalized to tubulin and membrane ACSL1 protein normalized to Na+/K+ ATPase. The data presented are means ± standard error (n = 3); the p-value was calculated using student’s t-test. (*p < 0.05). E) Mitochondria were isolated from BMDMs, and ACSL1 abundance was determined by western blot with and without LPS treatment. The non-specific (ns) band serves as a loading control.

Fig 5. CHREBP and NF-kappa B increase Acsl1 transcriptional activity.

HEK293 cells cultured in HG conditions were transfected with Acsl1-GLuc reporter (250ng; columns 1–7) and individually with p65/RELA (150ng; column 2) or CHREBP (150ng; column 3). Cells were transfected with Acsl1-GLuc reporter and a fixed amount of p65/RELA (150ng; columns 4–7), along with increasing amounts of CHREBP (100ng, 150ng, 200ng, 250ng; columns 4–7, respectively). Total DNA was adjusted to 750ng with vector only. Luciferase assay was performed 48 hours post-transfection and shown as relative luciferase units (RLU). The data presented are means ± standard errors (n = 3); the p-value was calculated using one-way ANOVA. *p < 0.05. B) Western blot of lysates from panel A with antibodies against the Myc-tag on CHREBP, FLAG-tag on p65/RELA, and tubulin as a loading control. The p65/RELA blot for lanes 1–2 was run on a different gel than lanes 3–7 and denoted by a black line between lanes 2 and 3. Exposure times were the same.

Fig 6. CHREBP and p65/RELA occupy the mouse Acsl1 gene.

A) Putative binding sites of CHREBP and p65/RELA on mouse Acsl1. Arrows denote the region amplified by qPCR. B) Chromatin was immunoprecipitated from BMDMs differentiated in HG using an antibody against CHREBP and isotype-matched IgG control. C) Chromatin was immunoprecipitated from BMDMs differentiated in HG and stimulated with LPS (10ng/mL) for 1 hour using an antibody against p65/RELA along with isotype-matched IgG control. Percent of precipitated DNA compared to total input DNA is shown. The data are means with an error bar representing the spread of the mean from two independent experiments.

Fig 7. Model for HG and inflammation-induced Acsl1 expression by CHREBP and NF-kappa B in macrophages.

A) We propose that in high glucose, CHREBP is depressed, and active nuclear CHREBP promotes the expression of Acsl1. B) Under inflammatory conditions that activate the NF-kappa B via dismissal of the inhibitory protein I kappa B (IkB) that prevents NF-kappa B nuclear transport, Acsl1 transcription is induced. C) Acsl1 expression can be further increased by high glucose and inflammatory stimuli via CHREBP and NF-kappa B.