Fatty acid binding protein 5 (FABP5) is mainly involved in the uptake, transport, and metabolism of fatty acid in the cytoplasm, and its role in immune cells has been recognized in recent years. However, the role of FABP5 in macrophage inflammation and its underlying mechanisms were not fully addressed. In our study, the acute liver injury and sepsis mouse models were induced by i.p. injection of LPS and cecal contents, respectively. Oleic acid (0.6 g/kg) was injected four times by intragastric administration every week, and this lasted for 1 wk before the LPS or cecal content challenge. We found that myeloid-specific deletion of FABP5 mitigated LPS-induced acute liver injury with reduced mortality of mice, histological liver damage, alanine aminotransferase, and proinflammatory factor levels. Metabolic analysis showed that FABP5 deletion increased the intracellular unsaturated fatty acids, especially oleic acid, in LPS-induced macrophages. The addition of oleic acid also decreased LPS-stimulated macrophage inflammation in vitro and reduced acute liver injury in LPS-induced or cecal content–induced sepsis mice. RNA-sequencing and molecular mechanism studies showed that FABP5 deletion or oleic acid supplementation increased the AMP/ATP ratio and AMP-activated protein kinase (AMPK) activation and inhibited the NF-κB pathway during the inflammatory response to LPS stimulation of macrophages. Inhibiting AMPK activation or expression by chemical or genetic approaches significantly rescued the decreased NF-κB signaling pathway and inflammatory response in LPS-treated FABP5-knockout macrophages. Our present study indicated that inhibiting FABP5 or supplementation of oleic acid might be used for the treatment of sepsis-caused acute liver injury.

Sepsis is characterized by the life-threatening dysfunction of many organs, including the liver (1, 2). Bacterial LPS, also known as endotoxin, can cause acute liver injury and sepsis and is often used to induce the endotoxin shock mouse model (3). TLR-mediated activation of the MyD88/NF-κB signaling pathway in macrophages produces large amounts of TNF-α, IL-6, and IL-1β inflammatory cytokines that cause inflammatory storms in the liver, leading to sepsis-associated acute liver injury (4, 5). The central role of fatty acids (FAs) is to provide energy sources and signals for metabolism and gene transcriptional regulation networks, and they closely participate in cell proliferation, differentiation, and survival pathways. FAs, classified into saturated and unsaturated FAs, are the significant components of neutral fats, phospholipids, and glycolipids (6). Unsaturated FAs include monounsaturated FAs (MUFAs) and polyunsaturated FAs (PUFAs). MUFAs mainly consist of n-9 oleic acid (C18:1) (7). Although n-3 PUFAs are widely described as anti-inflammatory fats, oleic acid has also been reported in recent years to play an anti-inflammatory role in immune cells and to reduce TNF-α–induced endothelial reactive oxygen species production (810). The entry of FAs into cells is mainly controlled by transporters such as CD36, FA binding proteins (FABPs), and the FA transporter family on the plasma membrane (11). As a member of the FABP family, FABP5 is mainly involved in the uptake, transport, and metabolism of FAs in the cytoplasm, and its role in immune cells has been reported in recent years (11). FABP5 deletion in macrophages inhibits early atherosclerosis in mice by reducing the expression of cyclooxygenase 2 (COX2), and IL-6 (12). FABP5-knockout (FABP5KO) mice display higher mRNA levels of the anti-inflammatory cytokines IL-10, arginase 1, YM-1, and Fizz-1 in the liver than wild-type (WT) mice after LPS stimulation, which promotes more anti-inflammatory macrophages (13). However, the molecular mechanism by which FABP5 regulates the inflammatory response of macrophages has not been fully understood. In the present study, we found that FABP5 deletion caused the accumulation of intracellular unsaturated FAs, especially oleic acid, in macrophages and decreased the secretion of TNF-α, IL-6, and IL-1β in the LPS-stimulated macrophages by regulating the AMP-activated protein kinase (AMPK)/NF-κB signaling pathway. Similarly, additional administration of oleic acid to WT mice significantly reduced LPS- or cecal content–induced acute liver injury and mortality, which provides evidence for the protective effect of oleic acid in acute liver injury caused by sepsis.

C57BL/6 mice (6–8 wk old) were purchased from Beijing University Experimental Animal Center (Beijing, China) or Sipeifu Biotechnology Co. (Beijing, China). FABP5KO mice, FABP5flox/flox, and LysMCre mice were obtained from Dr. Lianfeng Zhang, Key Laboratory of Human Diseases Comparative Medicine, Ministry of Health, Institute of Laboratory Animal Science, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. To obtain myeloid specificity of FABP5KO mice, FABP5flox/flox mice were bred with transgenic mice harboring Cre-recombinase driven by a myeloid-specific lysozyme M promoter to generate the following genotypes: FABP5flox/flox (control) and LysMCre-FABP5 (Lyzs-FABP5KO). The primers used for genotyping are shown in Table I. Mice were fed and bred under a specific pathogen-free condition and allowed unrestricted access to food and water in the animal facility of the Institute of Zoology, Chinese Academy of Sciences. All the experiments were approved by the animal ethics committee of the Institute of Zoology.

Table I.

Primers used for genotyping and real-time PCR

GenePrimer sequence (5′–3′)Use
TNF-α  RT-PCR 
 Forward CTTCTGTCTACTGAACTTCGGG  
 Reverse CAGGCTTGTCACTCGAATTTTG  
Il-6  RT-PCR 
 Forward CAAAGCCAGAGTCCTTCAGAG  
 Reverse GTCCTTAGCCACTCCTTCTG  
Il-1β  RT-PCR 
 Forward ACGGACCCCAAAAGATGAAG  
 Reverse TTCTCCACAGCCACAATGAG  
Vacm1  RT-PCR 
 Forward GCAAAGGACACTGGAAAAGAG  
 Reverse TCAAAGGGATACACATTAGGGAC  
Cxcl2  RT-PCR 
 Forward GAAGTCATAGCCACTCTCAAGG  
 Reverse CTTCCGTTGAGGGACAGC  
Ptgs2  RT-PCR 
 Forward CTCACGAAGGAACTCAGCAC  
 Reverse GGATTGGAACAGCAAGGATTTG  
Gadd45b  RT-PCR 
 Forward CCTGGTCACGAACTGTCATAC  
 Reverse GTTGCTTTAGATGTTTGGAGTGG  
Traf1  RT-PCR 
 Forward GAAATCATGTGCCCCTTTGC  
 Reverse TTCCACTCCTTTAAGACCGC  
Ccl4  RT-PCR 
 Forward TGACCAAAAGAGGCAGACAG  
 Reverse GTGAGAAGCATCAGGGCTG  
HPRT  RT-PCR 
 Forward AGTACAGCCCCAAAATGGTTAAG  
 Reverse CTTAGGCTTTGTATTTGGCTTTTC  
Lyz  Genotyping 
 Lyz-common CTTGGGCTGCCAGAATTTCTC  
 Lyz-mutant CCCAGAAATGCCAGATTACG  
 Lyz-wt TTACAGTCGGCCAGGCTGAC  
FABP5  Genotyping 
 FABP5-S CATTCAATTCCTATAGCGCCAAG  
 FABP5-A CTGCAGAAACCACGCCCTAC  
FABP5-flox  Genotyping 
 FABP5-f-P1 CCCTGGCTCCTCATCCTTGT  
 FABP5-f-P2 GCGGGCACTCCATCCTAACT  
 FABP5-f-P3 CAGACGAGTCGCAGCTTGCAGGAG  
 FABP5-f-P4 GGTGGTGTCGGGCAGCCAAATA  
GenePrimer sequence (5′–3′)Use
TNF-α  RT-PCR 
 Forward CTTCTGTCTACTGAACTTCGGG  
 Reverse CAGGCTTGTCACTCGAATTTTG  
Il-6  RT-PCR 
 Forward CAAAGCCAGAGTCCTTCAGAG  
 Reverse GTCCTTAGCCACTCCTTCTG  
Il-1β  RT-PCR 
 Forward ACGGACCCCAAAAGATGAAG  
 Reverse TTCTCCACAGCCACAATGAG  
Vacm1  RT-PCR 
 Forward GCAAAGGACACTGGAAAAGAG  
 Reverse TCAAAGGGATACACATTAGGGAC  
Cxcl2  RT-PCR 
 Forward GAAGTCATAGCCACTCTCAAGG  
 Reverse CTTCCGTTGAGGGACAGC  
Ptgs2  RT-PCR 
 Forward CTCACGAAGGAACTCAGCAC  
 Reverse GGATTGGAACAGCAAGGATTTG  
Gadd45b  RT-PCR 
 Forward CCTGGTCACGAACTGTCATAC  
 Reverse GTTGCTTTAGATGTTTGGAGTGG  
Traf1  RT-PCR 
 Forward GAAATCATGTGCCCCTTTGC  
 Reverse TTCCACTCCTTTAAGACCGC  
Ccl4  RT-PCR 
 Forward TGACCAAAAGAGGCAGACAG  
 Reverse GTGAGAAGCATCAGGGCTG  
HPRT  RT-PCR 
 Forward AGTACAGCCCCAAAATGGTTAAG  
 Reverse CTTAGGCTTTGTATTTGGCTTTTC  
Lyz  Genotyping 
 Lyz-common CTTGGGCTGCCAGAATTTCTC  
 Lyz-mutant CCCAGAAATGCCAGATTACG  
 Lyz-wt TTACAGTCGGCCAGGCTGAC  
FABP5  Genotyping 
 FABP5-S CATTCAATTCCTATAGCGCCAAG  
 FABP5-A CTGCAGAAACCACGCCCTAC  
FABP5-flox  Genotyping 
 FABP5-f-P1 CCCTGGCTCCTCATCCTTGT  
 FABP5-f-P2 GCGGGCACTCCATCCTAACT  
 FABP5-f-P3 CAGACGAGTCGCAGCTTGCAGGAG  
 FABP5-f-P4 GGTGGTGTCGGGCAGCCAAATA  

Bone marrow cells were stimulated by 50 ng/ml M-CSF (315-02, PeproTech) for 7 d to differentiate into bone marrow–derived macrophages (BMDMs) (14). Primary peritoneal macrophages (PEMs) were obtained from the peritoneal exudates of mice that were pretreated with 3% thioglycolate (LA4590, Solarbio) i.p. for 5 d. The peritoneal exudate cells were collected by cold PBS and adjusted to 1 × 106 cells/ml in DMEM cultured at 37°C and 5% CO2 in a cell incubator (15). Two hours later, using PBS to wash out the supernatant cells, the remaining adherent cells contained more than 90% macrophages.

BMDMs or PEMs isolated from WT or FABP5KO mice were treated with LPS (100 ng/ml, Escherichia coli 0111:B4; Sigma-Aldrich, St. Louis, MO) and GolgiPlug (100 ng/ml, 555029, BD Biosciences) to induce proinflammatory macrophages. Proinflammatory macrophages were harvested after 6 h. Compound C (Com C; 1 μM, S7840, Selleckchem), doxorubicin (0.5 μM, HY-15142, MedChemExpress), or 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR; 250 μM, HY-13417, MedChemExpress) were used to pretreat BMDMs or PEMs for 1 h before administering LPS. Oleic acid (100 μM, O1383, Sigma-Aldrich) was used to pretreat PEMs for 24 h before administering LPS. After different treatment, BMDMs and PEMs were collected using cold PBS (15, 16). For surface staining, BMDMs and PEMs were stained with anti-mouse CD11b (15-0112-83, Thermo Fisher Scientific), anti-mouse F4/80 (123110, BioLegend), for 30 min at 4°C. The cells were fixed using a buffer containing paraformaldehyde and then permeabilized by washing two times with 1× BD Perm/Wash buffer. For intracellular cytokine staining, the fixed/permeabilized cells were resuspended in 50 µl BD Perm/Wash buffer containing a predetermined optimal concentration of a fluorochrome-conjugated anti-mouse TNF-α Ab (12-7321-82, Thermo Fisher Scientific) at 4°C for 30 min in the dark (17). After being washed two times, cells were immediately analyzed by flow cytometry.

Total protein of cultured PEMs or BMDMs was extracted using radioimmunoprecipitation assay buffer (MultiSciences, Hangzhou, China) with protease and phosphatase inhibitor combinations (Roche, Basel, Switzerland) (18). The lysates were mixed with 5× SDS-PAGE loading buffer and boiled at 100°C for 5 min. SDS-PAGE was used to separate protein samples, which were subsequently transferred to polyvinylidene difluoride membranes (EMD Millipore) (19). After blocking in TBS plus 5% nonfat dry milk for 1 h at room temperature, the membranes were incubated with Abs against p-AMPKα (2531, Cell Signaling Technology), AMPKα (5831, Cell Signaling Technology), p-P65 (3033, Cell Signaling Technology), P65 (6596, Cell Signaling Technology), phosphorylated IκBα (9246, Cell Signaling Technology), IκBα (ab32518, Abcam), phosphorylated inhibitor of NF-κB kinase-α (p-IKKα; 2697, Cell Signaling Technology), IKKα (ab178870, Abcam), and β-actin (A5441, Sigma-Aldrich). After an overnight incubation at 4°C, secondary Abs were conjugated with peroxidase and then detected through chemiluminescence (EMD Millipore).

Total RNA was extracted with TRIzol (Ambion/Life Technologies, Carlsbad, CA), and reverse transcription was performed with avian myeloblastosis virus reverse transcriptase (2621, TaKaRa) according to the manufacturer’s instructions (20). Quantitative real-time PCR was performed using Power SYBR PCR Master Mix (RR420A, TaKaRa) in 96-well plates. Plates were read with CFX96 (Bio-Rad Laboratories). All mRNA expression levels were analyzed using the comparative cycle threshold method and normalized to the housekeeping gene hypoxanthine phosphoribosyltransferase. Table I lists the primers used in the amplification.

HPLC was used to measure intracellular AMP or ATP. Cells were washed with PBS, quenched with 6% ice-cold HClO4, and flash-frozen in liquid nitrogen after removing the culture media swiftly. Cells were taken, frozen with liquid nitrogen, and thawed for 20 min on ice. After vortexing, samples were centrifuged for 10 min at 4°C at 10,000 × g. The supernatant was collected and neutralized with 2 M KHCO3 to pH 7.4 after another 10-min centrifugation at 10,000 × g at 4°C. As previously mentioned (21), nucleotides in the supernatant were separated using a reverse-phase Discovery C18 column (Sigma-Aldrich) with a Hewlett-Packard series 1100 HPLC system (Agilent, Santa Clara, CA). At a flow rate of 0.7 ml/min, a phosphate buffer containing tetrabutylammonium sulfate and methanol was used as the mobile phase. The nucleotide separation was completed in 20 min after using gradient elution. AMP or ATP levels were normalized by cell number.

BMDMs of WT and FABP5KO mice were cultured and obtained from M-CSF–induced bone marrow cells for 7 d. Furthermore, BMDMs were induced with or without LPS stimulation for 6 h. Then macrophages were collected and frozen in liquid nitrogen immediately. Next, 1 × 107 macrophages were used for testing. FA was detected by the Novogene Technology Company using gas chromatography/mass spectrometry (6890N/5975B; Agilent).

Mouse TNF-α and IL-6 levels in sera or PEM culture supernatant were detected by ELISA kits of mouse TNF-α (430904, BioLegend) and mouse IL-6 (431304, BioLegend). All these kits were used according to the manufacturer’s instructions.

Mice were i.p. injected with 10 mg/kg of LPS (5). Then, mouse survival was recorded as lasting 4 d. Blood was collected from the submaxillary veins of mice at 0, 2, 6, and 12 h after LPS treatment. The liver was photographed for comparative analysis. Subsequently, a small part of the liver was put into TRIzol for preservation, and a small part of the liver was put into 4% paraformaldehyde.

Mouse cecal contents were diluted with 0.9% NaCl solution (20 mg/ml). Mice were i.p. injected with 1 ml diluted and filtered cecal contents. Mouse survival was observed for at least 4 d. Blood samples were collected from the submaxillary veins of mice at 0 and 6 h after the mouse cecal content treatment. In addition, a small part of the liver was put into 4% paraformaldehyde.

Oleic acid (0.6 g/kg) was injected four times by intragastric administration every week, and this lasted for 1 wk before LPS challenge. The control group received the same amount of saline as the experimental group by intragastric administration. Similarly, all these mice were i.p. challenged by LPS, and the induction method used was as described above. Blood samples were collected from the submaxillary veins of these mice for FA detection as described above.

BMDMs were transfected with 20 nM control siRNA or siRNA for AMPKα1 and AMPKα2 (Table II) using the Lipofectamine 3000 Transfection Kit (catalog no. L3000008, Thermo Fisher Scientific) according to its transfection protocol. siRNA 1 represented the combination of siAMPKα1-1 and siAMPKα2-1, siRNA 2 represented the combination of siAMPKα1-2 and siAMPKα2-2, and siRNA negative represents a meaningless RNA sequence, all of which were purchased from Tsingke Biotechnology. After 6 h of incubation at 37°C with 5% CO2, the culture medium was replaced with fresh DMEM. After cells were cultured for an additional 30 h, these cells were stimulated with LPS (100 ng/ml) for 30 min or 6 h and harvested for RT-PCR, Western blotting, and flow cytometry as described above.

Table II.

The sequence of siRNA used to knock down AMPKα

NameSequence (5′–3′)
SiAMPKα1-1 sense GUUCAACCAUGAUCGAUGATT 
SiAMPKα1-1 antisense UCAUCGAUCAUGGUUGAACTT 
SiAMPKα1-2 sense GGCAGAAGUUUGUAGAGCATT 
SiAMPKα1-2 antisense UGCUCUACAAACUUCUGCCTT 
SiAMPKα2-1 sense GCAUACCAUCUUCGAGUAATT 
SiAMPKα2-1 antisense UUACUCGAAGAUGGUAUGCTT 
SiAMPKα2-2 sense GGUGAAUUUCUACGAACUATT 
SiAMPKα2-2 antisense UAGUUCGUAGAAAUUCACCTT 
NameSequence (5′–3′)
SiAMPKα1-1 sense GUUCAACCAUGAUCGAUGATT 
SiAMPKα1-1 antisense UCAUCGAUCAUGGUUGAACTT 
SiAMPKα1-2 sense GGCAGAAGUUUGUAGAGCATT 
SiAMPKα1-2 antisense UGCUCUACAAACUUCUGCCTT 
SiAMPKα2-1 sense GCAUACCAUCUUCGAGUAATT 
SiAMPKα2-1 antisense UUACUCGAAGAUGGUAUGCTT 
SiAMPKα2-2 sense GGUGAAUUUCUACGAACUATT 
SiAMPKα2-2 antisense UAGUUCGUAGAAAUUCACCTT 

Serum alanine aminotransferase measurements were analyzed using the ALT/GPT kit (Nanjing Jiancheng). The kit was used according to the manufacturer’s instructions.

BMDMs were cultured and obtained as previously described. Furthermore, BMDMs in WT and FABP5KO mice were induced with or without LPS stimulation for 6 h. Total RNA was extracted using TRIzol reagent (15596018, Thermo Fisher Scientific) (22). The NovaSeq 6000 high-throughput sequencing platform (Illumina) was used for sequencing. The raw data can be low-quality reads (Q < 20) and were assessed by FastQC, and the adaptor sequence was filtered by Trimgalore. We used the mapping software HISAT2 to map the reads to the mm10 reference genome and StringTie to construct transcripts independently for each cell (23). The DEGseq package was used to identify the differentially expressed genes (24). We set q < 0.05 and |log2(foldchange)| > 0 as a significant difference.

We performed pathway enrichment analysis on all differentially expressed genes using the DAVID Bioinformatics Resources 6.8 online search tool (https://david.ncifcrf.gov/) and the Kobasonline search tool (http://kobas.cbi.pku.edu.cn/) for gene ontology functional annotation and KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analysis (25, 26). The protein interaction network was obtained by STRING (https://stringdb.org/), and the metabolic regulation network was obtained by MetScape and visualized by Cytoscape (27, 28).

The results are presented as mean ± SD. An unpaired Student t test was used to test differences between two groups. GraphPad Prism version 7 was used to examine all data. The legends of the figures provide statistical values.

We attempted to investigate the role of FABP5 on macrophage inflammation. First, we found that the expression of the FABP5 gene in mouse PEMs is significantly and gradually reduced with the prolongation of LPS stimulation time (p < 0.001) (Fig. 1A), indicating the potential roles of FABP5 in the inflammatory response to the LPS of macrophages. We then created FABP5KO mice and performed RNA-seq analysis for WT and FABP5KO mouse BMDMs with or without LPS stimulation to perform differential gene expression analysis. The heatmap analysis showed that FABP5 deficiency reduced the expression levels of chemokines and cytokines associated with LPS-induced inflammatory macrophages (M1) (Fig. 1B). To verify the RNA-seq analysis results, we stimulated peritoneal macrophages with LPS and obtained RNA samples at different stimulation times. Real-time PCR analysis showed that FABP5 deficiency significantly reduced the expression levels of inflammatory macrophage–related marker genes, such as TNF-α, IL-6, and IL-1β, in peritoneal macrophages within 12 h after LPS stimulation (p < 0.01 or p < 0.001) (Fig. 1C). Consistently, the secretion of TNF-α was significantly reduced in the culture medium of LPS-treated FABP5KO PEMs compared with those of WT PEMs, as determined by ELISA (p < 0.001) (Fig. 1D). The percentage of TNF-α+ cells in F4/80+CD11b+ macrophages in LPS-treated FABP5KO PEMs was significantly decreased compared with LPS-treated WT PEMs, as detected by flow cytometry (p < 0.01) (Fig. 1E). In addition to peritoneal macrophages, BMDMs induced by M-CSF in vitro also showed similarly decreased TNF-α, IL-6, and IL-1β expression, as shown in (Fig. 1F (p < 0.001). The results of ELISA and flow cytometric analysis also showed that FABP5 deletion significantly reduced TNF-α protein level in LPS-treated BMDMs (p < 0.05 or p < 0.001) (Supplemental Fig. 1). These results implied that FABP5 deficiency in macrophages remarkably decreased LPS-induced proinflammation in vitro.

FIGURE 1.

FABP5 deficiency decreases LPS-induced inflammation in macrophages in vitro. PEMs were obtained from WT and FABP5KO mice after 5 d of 3% Thioglycollate treatment. BMDMs were obtained from bone marrow cells induced with M-CSF for 7 d. (A) Real-time PCR showing expression of the FABP5 gene in WT PEMs after LPS stimulation for different time periods. (B) Heatmap represents the expression levels of cytokine and chemokine genes involved in proinflammatory macrophages (M1) in BMDMs between WT and FABP5KO mice with or without LPS stimulation for 6 h. (C) Gene expression of TNF-α, IL-6, and IL-1β in WT and FABP5KO PEMs with or without LPS stimulation for different time periods was measured by real-time PCR. (D) TNF-α levels in the supernatants of WT and FABP5KO PEMs with or without LPS stimulation for 6 h were measured by ELISA. (E) TNF-α expression in F4/80+CD11b+ cells in WT and FABP5KO PEMs with or without LPS stimulation for 6 h was measured by flow cytometry. (F) Real-time PCR showing expression of TNF-α, IL-6, and IL-1β in WT and FABP5KO BMDMs with or without LPS stimulation for 6 h. Data are representative of three independent experiments. Data are presented as mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, Student t test. See also Supplemental Fig. 1.

FIGURE 1.

FABP5 deficiency decreases LPS-induced inflammation in macrophages in vitro. PEMs were obtained from WT and FABP5KO mice after 5 d of 3% Thioglycollate treatment. BMDMs were obtained from bone marrow cells induced with M-CSF for 7 d. (A) Real-time PCR showing expression of the FABP5 gene in WT PEMs after LPS stimulation for different time periods. (B) Heatmap represents the expression levels of cytokine and chemokine genes involved in proinflammatory macrophages (M1) in BMDMs between WT and FABP5KO mice with or without LPS stimulation for 6 h. (C) Gene expression of TNF-α, IL-6, and IL-1β in WT and FABP5KO PEMs with or without LPS stimulation for different time periods was measured by real-time PCR. (D) TNF-α levels in the supernatants of WT and FABP5KO PEMs with or without LPS stimulation for 6 h were measured by ELISA. (E) TNF-α expression in F4/80+CD11b+ cells in WT and FABP5KO PEMs with or without LPS stimulation for 6 h was measured by flow cytometry. (F) Real-time PCR showing expression of TNF-α, IL-6, and IL-1β in WT and FABP5KO BMDMs with or without LPS stimulation for 6 h. Data are representative of three independent experiments. Data are presented as mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, Student t test. See also Supplemental Fig. 1.

Close modal

Proinflammatory macrophages play an important role in LPS-induced acute endotoxin shock (29). We therefore used WT mice and mice with a myeloid-specific deletion of FABP5 (Lyzs-FABP5KO) to construct LPS-induced acute liver injury models, as reported previously (5, 30). After WT and Lyzs-FABP5KO mice received a dose of LPS (10 mg/kg body weight), more than 50% of Lyzs-FABP5KO mice survived for at least 96 h, but all WT mice died within 36 h (p < 0.001) (Fig. 2A), indicating the resistance to endotoxic shock of Lyzs-FABP5KO mice. Pathological examination showed that the hemorrhage in livers of LPS-treated Lyzs-FABP5KO mice was significantly reduced compared with WT mice (Fig. 2B). H&E staining showed that the livers of LPS-treated Lyzs-FABP5KO mice had less infiltration of immune cells than the livers of WT mice (Fig. 2C). Clinical tests showed that the levels of alanine aminotransferase were significantly reduced in the serum of LPS-treated Lyzs-FABP5KO mice compared with WT mice (p < 0.001) (Fig. 2D). These results suggested that myeloid-specific deletion of FABP5 mitigated LPS-induced liver damage in mice.

FIGURE 2.

Myeloid-specific deletion of FABP5 alleviates acute liver injury in a sepsis mouse model. The sepsis model of WT and Lyzs-FABP5KO mice was induced by i.p. injection of LPS (10 mg/kg body weight) or cecal contents (20 mg/ml). Control means the mice without sepsis-inducing treatment. (A) Kaplan-Meier plots of WT and Lyzs-FABP5KO mouse survival after LPS injection are shown (n = 11). (B) Photographs of the liver in WT and Lyzs-FABP5KO mice are shown. Scale bar represents 1 cm. (C) H&E staining of liver tissues of WT and Lyzs-FABP5KO mice after LPS injection (scale bar represents 20 μm). These arrows represent infiltrating inflammatory cells. (D) The levels of alanine aminotransferase in the sera of WT and Lyzs-FABP5KO mice after LPS injection are shown. (E and F) The levels of TNF-α (E) and IL-6 (F) were measured in the sera of WT and Lyzs-FABP5KO mice after LPS injection by ELISA. (G) Gene expression of TNF-α, IL-6, and IL-1β in livers of WT and Lyzs-FABP5KO mice with or without LPS injection for different times measured by real-time PCR. (H) Kaplan-Meier plots of WT and Lyzs-FABP5KO mouse survival after cecal content injection are shown (n = 8). (I) H&E staining of liver tissues of WT and Lyzs-FABP5KO mice after cecal content injection (scale bar represents 20 μm). These arrows represent infiltrating inflammatory cells. (J and K) The levels of TNF-α (J) and IL-6 (K) in the sera of WT and Lyzs-FABP5KO mice were measured by ELISA. Experiments were done more than two times. Data are presented as mean ± SD (n = 6). *p < 0.05, **p < 0.01, ***p < 0.001, Student t test.

FIGURE 2.

Myeloid-specific deletion of FABP5 alleviates acute liver injury in a sepsis mouse model. The sepsis model of WT and Lyzs-FABP5KO mice was induced by i.p. injection of LPS (10 mg/kg body weight) or cecal contents (20 mg/ml). Control means the mice without sepsis-inducing treatment. (A) Kaplan-Meier plots of WT and Lyzs-FABP5KO mouse survival after LPS injection are shown (n = 11). (B) Photographs of the liver in WT and Lyzs-FABP5KO mice are shown. Scale bar represents 1 cm. (C) H&E staining of liver tissues of WT and Lyzs-FABP5KO mice after LPS injection (scale bar represents 20 μm). These arrows represent infiltrating inflammatory cells. (D) The levels of alanine aminotransferase in the sera of WT and Lyzs-FABP5KO mice after LPS injection are shown. (E and F) The levels of TNF-α (E) and IL-6 (F) were measured in the sera of WT and Lyzs-FABP5KO mice after LPS injection by ELISA. (G) Gene expression of TNF-α, IL-6, and IL-1β in livers of WT and Lyzs-FABP5KO mice with or without LPS injection for different times measured by real-time PCR. (H) Kaplan-Meier plots of WT and Lyzs-FABP5KO mouse survival after cecal content injection are shown (n = 8). (I) H&E staining of liver tissues of WT and Lyzs-FABP5KO mice after cecal content injection (scale bar represents 20 μm). These arrows represent infiltrating inflammatory cells. (J and K) The levels of TNF-α (J) and IL-6 (K) in the sera of WT and Lyzs-FABP5KO mice were measured by ELISA. Experiments were done more than two times. Data are presented as mean ± SD (n = 6). *p < 0.05, **p < 0.01, ***p < 0.001, Student t test.

Close modal

LPS can cause a storm of inflammatory cytokines in sepsis, with the liver being the major affected organ (31). Similar to the decreased liver pathological alteration in LPS-treated Lyzs-FABP5KO mice, TNF-α and IL-6 levels in the serum of Lyzs-FABP5KO mice were significantly lower than those of WT mice after LPS challenge, as detected by ELISA (p < 0.001) (Fig. 2E, 2F). In addition, we examined the expression of inflammatory genes in the liver tissue. Real-time PCR analysis showed that the expression of inflammatory genes, such as TNF-α, IL-6, and IL-1β, in the livers of Lyzs-FABP5KO mice was significantly lower than that of WT mice after LPS challenge at different time points (Fig. 2G). Thus, FABP5 deficiency, specifically in macrophages, alleviated LPS-induced acute inflammation and liver injury in mice.

In addition, we also used a cecal content–induced sepsis mouse model, which is more clinically relevant, to observe the effect of FABP5 deletion (32). Similarly, we found that ?60% of Lyzs-FABP5KO mice survived for at least 96 h, but all WT mice died within 30 h (p < 0.001) (Fig. 2H). H&E results of liver tissue showed that the inflammatory infiltration of the liver in Lyzs-FABP5KO mice was alleviated after i.p. injection of cecal contents compared with that in WT mice (Fig. 2I). The results of ELISA showed that the serum levels of TNF-α and IL-6 in Lyzs-FABP5KO mice were significantly lower than those in WT mice after cecal content stimulation (p < 0.001) (Fig. 2J, 2K). These results suggested that deletion of FABP5 in macrophages attenuated cecal content–induced acute liver injury and inflammatory response in mice.

Considering that the main role of FABP5 is to participate in FA uptake, transport, and metabolism within the cell cytoplasm (11), we examined the FA metabolism of LPS-stimulated WT and FABP5-deficient macrophages with MetScape metabolic analysis. The heatmap showed that FABP5 deletion caused changes in both glucose and lipid metabolism genes and that the main increased genes were related to lipid metabolism (Fig. 3A). Further analysis showed that FABP5KO macrophages significantly increased in glycolysis, gluconeogenesis, especially saturated FA β-oxidation, and glycerophospholipid metabolism compared with WT macrophages after LPS stimulation (Fig. 3B). Therefore, we hypothesized that FABP5 may affect macrophage polarization mainly through lipid metabolism. We thus detected free FAs in LPS-stimulated WT and FABP5KO macrophages by lipidomics. The findings revealed that long-chain unsaturated FAs, especially oleic acid and arachidonic acid, in LPS-stimulated FABP5KO macrophages were significantly higher than those of LPS-stimulated WT macrophages (p < 0.01) (Fig. 3C). However, we found that the absence of FABP5 did not affect long-chain saturated and medium-chain FAs (Supplemental Fig. 2). These data suggested that the absence of FABP5 mainly affects long-chain unsaturated FAs, especially oleic acid, which is the most abundant one.

FIGURE 3.

Oleic acid reduces LPS-induced proinflammation of macrophages. PEMs were obtained from WT and FABP5KO mice after 5 d of 3% Thioglycollate treatment. BMDMs were obtained from M-CSF–induced bone marrow cells for 7 d. (A) Heatmap showing the changes in glucose metabolism and lipid metabolism genes in WT and FABP5KO mouse macrophages with or without LPS treatment. (B) MetScape analysis of differentially expressed genes and metabolite data between WT and FABP5KO BMDMs after LPS stimulation for 6 h. Gene node color is green if the gene was significantly downregulated (log fold change <0.05), and gene color is red if the gene was significantly upregulated (log fold change >0.05). (C) Metabolomic profiling of long-chain unsaturated FA content between WT and FABP5KO BMDMs with LPS stimulation for 6 h. These samples were measured by gas chromatography/mass spectrometry. (D) PEMs of WT and FABP5KO mice were pretreated with 0, 0.1, 1, 10, 100, and 500 μM oleic acid for 24 h and followed by stimulation with LPS for 6 h. The control group was not stimulated by LPS and oleic acid. The mRNA expression of TNF-α, IL-6, and IL-1β was determined by real-time PCR. (E and F) TNF-α+ cells in F4/80+CD11b+ gates in WT PEMs were measured by flow cytometry. PEMs were incubated in medium containing a blank group and a 100 μM oleic acid group for 24 h, followed by stimulation with or without LPS for 6 h. Data are representative of three independent experiments. Data are presented as mean ± SD (n = 3). **p < 0.01, ***p < 0.001, Student t test. See also Supplemental Fig. 2.

FIGURE 3.

Oleic acid reduces LPS-induced proinflammation of macrophages. PEMs were obtained from WT and FABP5KO mice after 5 d of 3% Thioglycollate treatment. BMDMs were obtained from M-CSF–induced bone marrow cells for 7 d. (A) Heatmap showing the changes in glucose metabolism and lipid metabolism genes in WT and FABP5KO mouse macrophages with or without LPS treatment. (B) MetScape analysis of differentially expressed genes and metabolite data between WT and FABP5KO BMDMs after LPS stimulation for 6 h. Gene node color is green if the gene was significantly downregulated (log fold change <0.05), and gene color is red if the gene was significantly upregulated (log fold change >0.05). (C) Metabolomic profiling of long-chain unsaturated FA content between WT and FABP5KO BMDMs with LPS stimulation for 6 h. These samples were measured by gas chromatography/mass spectrometry. (D) PEMs of WT and FABP5KO mice were pretreated with 0, 0.1, 1, 10, 100, and 500 μM oleic acid for 24 h and followed by stimulation with LPS for 6 h. The control group was not stimulated by LPS and oleic acid. The mRNA expression of TNF-α, IL-6, and IL-1β was determined by real-time PCR. (E and F) TNF-α+ cells in F4/80+CD11b+ gates in WT PEMs were measured by flow cytometry. PEMs were incubated in medium containing a blank group and a 100 μM oleic acid group for 24 h, followed by stimulation with or without LPS for 6 h. Data are representative of three independent experiments. Data are presented as mean ± SD (n = 3). **p < 0.01, ***p < 0.001, Student t test. See also Supplemental Fig. 2.

Close modal

The activation of M1 macrophages is dependent on lipid metabolism (33). To test whether intracellular long-chain unsaturated FAs affect macrophage inflammation, we pretreated WT and FABP5KO macrophages with different doses of oleic acid and then stimulated them with LPS. Real-time PCR analysis revealed that the expression of inflammation-related genes, such as TNF-α, IL-6, and IL-1β, in WT macrophages reduced with increasing oleic acid doses (p < 0.01) (Fig. 3D). However, adding oleic acid, even at high concentrations, had no detectable effect on the inflammatory response to LPS in FABP5-deficient macrophages (p < 0.01) (Fig. 3D). Oleic acid significantly reduced the percentage of TNF-α+ cells in LPS-stimulated WT F4/80+CD11b+ macrophages (p < 0.001) (Fig. 3E, 3F). These results suggested that adding oleic acid could inhibit LPS-induced inflammation in the same way as the increased intracellular oleic acid caused by FABP5 depletion.

To test whether additional oleic acid can reduce the macrophage inflammatory response and its related disease in vivo, we performed intragastric administration of oleic acid to WT mice. We measured the serum oleic acid content by gas chromatography/mass spectrometry and found that the serum oleic acid content in oleic acid–treated mice was significantly higher than that in the normal saline-treated mice (p < 0.001) (Supplemental Fig. 3). This means that this method can indeed increase the amount of circulating oleic acid by additional application of oleic acid. Subsequently, LPS was injected i.p. to induce the acute liver injury model. Oleic acid treatment significantly increased the survival time and survival rate of mice after LPS injection compared with the control mice (p < 0.001) (Fig. 4A). Pathological tests showed that oleic acid treatment significantly reduced liver bleeding and inflammatory cell infiltration after LPS injection (Fig. 4B, 4C). Clinical tests revealed that oleic acid treatment significantly decreased the serum alanine aminotransferase levels after LPS injection compared with the normal saline group (p < 0.001) (Fig. 4D). In parallel with the less pathological alteration in oleic acid– and LPS-treated mice, oleic acid treatment significantly reduced the levels of TNF-α and IL-6 in the serum after LPS injection, as detected by ELISA (p < 0.001) (Fig. 4E, 4F). Real-time PCR analysis showed that the expression of inflammatory genes, such as TNF-α, IL-6, and IL-1β, in the liver tissue of oleic acid–treated mice was significantly lower than in the control mice (Fig. 4G). Similarly, we established a cecal content–induced sepsis model. Oleic acid treatment significantly increased the survival time and survival rate of mice after injection of cecal contents compared with the control mice (p < 0.001) (Fig. 4H). H&E staining of liver tissues showed that the infiltration of inflammatory cells was reduced in the oleic acid–treated mice after injection of cecal contents compared with the saline-treated mice (Fig. 4I). After injection of cecal contents, the levels of TNF-α and IL-6 in the serum of the oleic acid–treated mice were significantly lower than those in saline-treated mice (p < 0.01 and p < 0.001) (Fig. 4J, 4K). Thus, additional supplementary oleic acid could reduce acute liver inflammation and injury in mice.

FIGURE 4.

Additional intake of oleic acid alleviates acute liver injury caused by sepsis in mice. Intragastric administration of oleic acid or saline was performed, and i.p. injection of LPS or cecal content was used to induce the sepsis mouse model. (A) Kaplan-Meier plots of saline group and oleic acid group mouse survival after LPS injection are shown (n = 9). (B) Photographs of the livers of saline and oleic acid group mice are shown. Scale bar represents 1 cm. (C) H&E staining of liver tissues of saline and oleic acid group mice after LPS injection (scale bar represents 20 μm). These arrows represent infiltrating inflammatory cells. (D) The levels of alanine aminotransferase in the sera of saline and oleic acid group mice after LPS injection are shown. (E and F) The levels of TNF-α (E) and IL-6 (F) were measured in the sera of saline and oleic acid group mice after LPS injection by ELISA. (G) Gene expression of TNF-α, IL-6, and IL-1β in the livers of saline and oleic acid group mice with or without LPS injection for different times was measured by real-time PCR. (H) Kaplan-Meier plots of the survival of the saline- and oleic acid–treated mice after injection of cecal content are shown (n = 8). (I) H&E staining of liver tissues of saline and oleic acid group mice after injection of cecal content (scale bar represents 20 μm). These arrows represent infiltrating inflammatory cells. (J and K) The levels of TNF-α (J) and IL-6 (K) in the sera of saline- and oleic acid–treated mice after injection of cecal content were measured by ELISA. Experiments were done more than two times. Data are presented as mean ± SD (n = 6). *p < 0.05, **p < 0.01, ***p < 0.001, Student t test. See also Supplemental Fig. 3.

FIGURE 4.

Additional intake of oleic acid alleviates acute liver injury caused by sepsis in mice. Intragastric administration of oleic acid or saline was performed, and i.p. injection of LPS or cecal content was used to induce the sepsis mouse model. (A) Kaplan-Meier plots of saline group and oleic acid group mouse survival after LPS injection are shown (n = 9). (B) Photographs of the livers of saline and oleic acid group mice are shown. Scale bar represents 1 cm. (C) H&E staining of liver tissues of saline and oleic acid group mice after LPS injection (scale bar represents 20 μm). These arrows represent infiltrating inflammatory cells. (D) The levels of alanine aminotransferase in the sera of saline and oleic acid group mice after LPS injection are shown. (E and F) The levels of TNF-α (E) and IL-6 (F) were measured in the sera of saline and oleic acid group mice after LPS injection by ELISA. (G) Gene expression of TNF-α, IL-6, and IL-1β in the livers of saline and oleic acid group mice with or without LPS injection for different times was measured by real-time PCR. (H) Kaplan-Meier plots of the survival of the saline- and oleic acid–treated mice after injection of cecal content are shown (n = 8). (I) H&E staining of liver tissues of saline and oleic acid group mice after injection of cecal content (scale bar represents 20 μm). These arrows represent infiltrating inflammatory cells. (J and K) The levels of TNF-α (J) and IL-6 (K) in the sera of saline- and oleic acid–treated mice after injection of cecal content were measured by ELISA. Experiments were done more than two times. Data are presented as mean ± SD (n = 6). *p < 0.05, **p < 0.01, ***p < 0.001, Student t test. See also Supplemental Fig. 3.

Close modal

To understand the potential signaling pathways involved in the FABP5-mediated regulation of the M1 inflammatory response, we performed KEGG pathway enrichment analysis on the differentially expressed genes of LPS-induced WT and FABP5KO macrophages. The results showed that the Ag processing and presentation, TLR signaling pathway, NOD-like receptor signaling pathway, NF-κB signaling pathway, and TNF signaling pathway were significantly downregulated in LPS-stimulated FABP5-deficient macrophages compared with LPS-stimulated WT macrophages (Fig. 5A). Meanwhile, we found that most of the significantly decreased genes in the NF-κB pathway were canonical NF-κB–regulated genes (Fig. 5B). Activating the NF-κB signaling pathway is critical for the LPS-induced inflammatory response and secretion of inflammatory cytokines in macrophages (34, 35). To further verify the presence of a downregulated NF-κB signaling pathway in FABP5-deficient macrophages, we detected the essential downstream genes of the NF-κB pathway. Real-time PCR analysis revealed that critical downstream genes of NF-κB pathway expression, such as Vacm1, Cxcl2, Ptgs2, Gadd45b, Traf1, and Ccl4, in FABP5KO macrophages were significantly lower than in WT macrophages after LPS stimulation (p < 0.01) (Fig. 5C). IKKα appears to play a supporting role in activating the canonical NF-κB pathway (36), and the IKK complex then phosphorylates IκBα. Phosphorylated IκB family members undergo ubiquitylation and proteasomal degradation, resulting in the release and nuclear translocation of the canonical NF-κB family members, predominantly the NF-κB1 p50-RELA (also called p65) dimers (37, 38). Therefore, we detected NF-κB signaling pathway–related proteins and their active forms by Western blot analysis. The results showed that phosphorylation of IKKα, IκBα, and p65 proteins was significantly reduced in FABP5KO macrophages compared with WT macrophages after LPS stimulation (p < 0.05, p < 0.01, and p < 0.001, respectively) (Fig. 5D). In line with these results, oleic acid supplementation also reduced the phosphorylation expression of IKKα, IκBα, and p65 proteins in WT macrophages after LPS treatment (p < 0.05, p < 0.01, and p < 0.001, respectively) (Fig. 5E). These results suggested that FABP5 deletion or adding oleic acid inhibited the activation of the NF-κB pathway in LPS-treated macrophages Table I and II.

FIGURE 5.

FABP5 deficiency inhibits NF-κB activation during LPS-induced macrophage inflammation. (A) KEGG (Kyoto Encyclopedia of Genes and Genomes) pathways of the top five upregulated and downregulated pathways in FABP5KO versus WT inflammatory BMDMs. (B) The protein interaction network of canonical NF-κB–regulated genes. (C) Real-time PCR showing the expression of NF-κB pathway–related genes, including Vacm1, Cxcl2, Ptgs2, Gadd45b, Traf1, and Ccl4, between WT and FABP5KO BMDMs with or without LPS stimulation for 6 h. (D) Immunoblot analysis of key total proteins of the NF-κB pathway and their phosphorylation in WT and FABP5KO mice PEMs after LPS stimulation for different times (n = 3). (E) Immunoblot analysis of key total proteins of the NF-κB pathway and their phosphorylation in medium- and oleic acid–treated PEMs after LPS stimulation for different times (n = 3). Data are representative of three independent experiments. Data are presented as mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, Student t test.

FIGURE 5.

FABP5 deficiency inhibits NF-κB activation during LPS-induced macrophage inflammation. (A) KEGG (Kyoto Encyclopedia of Genes and Genomes) pathways of the top five upregulated and downregulated pathways in FABP5KO versus WT inflammatory BMDMs. (B) The protein interaction network of canonical NF-κB–regulated genes. (C) Real-time PCR showing the expression of NF-κB pathway–related genes, including Vacm1, Cxcl2, Ptgs2, Gadd45b, Traf1, and Ccl4, between WT and FABP5KO BMDMs with or without LPS stimulation for 6 h. (D) Immunoblot analysis of key total proteins of the NF-κB pathway and their phosphorylation in WT and FABP5KO mice PEMs after LPS stimulation for different times (n = 3). (E) Immunoblot analysis of key total proteins of the NF-κB pathway and their phosphorylation in medium- and oleic acid–treated PEMs after LPS stimulation for different times (n = 3). Data are representative of three independent experiments. Data are presented as mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, Student t test.

Close modal

AMPK activation significantly attenuates NF-κB signaling and inhibits mRNA and protein levels of proinflammatory cytokines, such as TNF-α and IL-6 (39, 40). AMPK is obviously activated in highly aggressive carcinoma cell lines treated by oleic acid (40, 41). Therefore, we hypothesized that FABP5 deletion impaired macrophage inflammation by activating the AMPK signaling pathway. First, we performed gene set enrichment analysis on genes that were relevant to AMPK positively regulated genes and found that the AMPK were significantly enriched in LPS-treated FABP5KO macrophages compared with LPS-treated WT macrophages (Fig. 6A). Consistent with the RNA-seq data, the phosphorylation of AMPKα in LPS-treated FABP5KO macrophages was enhanced compared with LPS-treated WT macrophages (p < 0.05 and p < 0.001) (Fig. 6B). Similarly, oleic acid pretreatment also increased the phosphorylation of AMPKα in LPS-induced macrophages (p < 0.05, p < 0.01, and p < 0.001) (Fig. 6C). The activation of AMPK was positively correlated with the AMP/ATP ratio in cells (42). The results of liquid chromatography and mass spectrometry showed that FABP5 deletion or oleic acid supplementation significantly increased the AMP/ATP ratio in macrophages with or without LPS stimulation (p < 0.001) (Fig. 6D, 6E).

FIGURE 6.

FABP5 deficiency impairs macrophage inflammation by activating the AMPK signaling pathway. (A) Gene set enrichment analysis of AMPK positively regulated genes in FABP5KO versus WT inflammatory BMDMs. (B) The expression of AMPKα and p-AMPKα proteins in WT and FABP5KO mouse PEMs after LPS stimulation at different time points was measured by immunoblot analysis (n = 3). (C) The expression of AMPKα and p-AMPKα proteins in medium group and oleic acid group PEMs after LPS stimulation for different times was measured by immunoblot analysis (n = 3). (D) The AMP/ATP ratio of WT and FABP5KO PEMs with or without LPS stimulation for 6 h was measured by HPLC. (E) The AMP/ATP ratio of medium group and oleic acid group PEMs with or without LPS stimulation for 6 h was measured by HPLC. (F) Flow cytometric analysis of TNF-α expression in F4/80+CD11b+ PEMs that were pretreated by Com C (1 μM) for 1 h and then stimulated with LPS for 6 h in WT and FABP5KO mice. (G) Immunoblot analysis of p65 and p-p65 protein in PEMs that were pretreated by Com C (1 μM) for 1 h and then stimulated with LPS for 6 h in WT and FABP5KO mice (n = 3). (H) Immunoblot analysis of p65 and p-p65 protein in PEMs that were pretreated by Com C (1 μM) for 1 h and then stimulated with LPS for 6 h in medium group and oleic acid group mice (n = 3). (I) Flow cytometric analysis of TNF-α expression in WT and FABP5KO F4/80+CD11b+ PEMs that were pretreated by AICAR (250 μM) for 1 h and then stimulated with LPS for 6 h. Data are representative of three independent experiments. Data are presented as mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, Student t test. See also Supplemental Fig. 4.

FIGURE 6.

FABP5 deficiency impairs macrophage inflammation by activating the AMPK signaling pathway. (A) Gene set enrichment analysis of AMPK positively regulated genes in FABP5KO versus WT inflammatory BMDMs. (B) The expression of AMPKα and p-AMPKα proteins in WT and FABP5KO mouse PEMs after LPS stimulation at different time points was measured by immunoblot analysis (n = 3). (C) The expression of AMPKα and p-AMPKα proteins in medium group and oleic acid group PEMs after LPS stimulation for different times was measured by immunoblot analysis (n = 3). (D) The AMP/ATP ratio of WT and FABP5KO PEMs with or without LPS stimulation for 6 h was measured by HPLC. (E) The AMP/ATP ratio of medium group and oleic acid group PEMs with or without LPS stimulation for 6 h was measured by HPLC. (F) Flow cytometric analysis of TNF-α expression in F4/80+CD11b+ PEMs that were pretreated by Com C (1 μM) for 1 h and then stimulated with LPS for 6 h in WT and FABP5KO mice. (G) Immunoblot analysis of p65 and p-p65 protein in PEMs that were pretreated by Com C (1 μM) for 1 h and then stimulated with LPS for 6 h in WT and FABP5KO mice (n = 3). (H) Immunoblot analysis of p65 and p-p65 protein in PEMs that were pretreated by Com C (1 μM) for 1 h and then stimulated with LPS for 6 h in medium group and oleic acid group mice (n = 3). (I) Flow cytometric analysis of TNF-α expression in WT and FABP5KO F4/80+CD11b+ PEMs that were pretreated by AICAR (250 μM) for 1 h and then stimulated with LPS for 6 h. Data are representative of three independent experiments. Data are presented as mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, Student t test. See also Supplemental Fig. 4.

Close modal

To determine whether the reduction of inflammatory macrophages caused by FABP5 deletion is through the AMPK signaling pathway, we pretreated macrophages with an AMPK-specific inhibitor, Com C, before LPS stimulation (43). Flow cytometric analysis showed that Com C pretreatment completely reversed the decreased percentage of TNF-α+ cells in FABP5KO F4/80+CD11b+ macrophages (p < 0.001) (Fig. 6F). Real-time PCR analysis showed that Com C pretreatment recovered the decreased expression levels of inflammatory cytokines, such as IL-6, IL-1β, and TNF-α, in FABP5-deficient macrophages (p < 0.001) (Supplemental Fig. 4A). Furthermore, the reduced phosphorylation of the p65 protein in macrophages caused by FABP5 deletion or oleic acid supplementation was also restored by adding Com C (p < 0.001) (Fig. 6G, 6H). In addition, we pretreated macrophages with another AMPK inhibitor, doxorubicin, before LPS stimulation. Similarly, we found that doxorubicin treatment also restored phenotypic differences of M1 caused by FABP5 depletion (p < 0.01 and p < 0.001) (Supplemental Fig. 4B, 4C). These results suggested that chemical inhibition of AMPK could reverse the reduction in inflammation caused by FABP5 depletion. In parallel with these results, we found that AMPK-specific activator AICAR pretreatment significantly reduced TNF-α+ cells in both WT and FABP5KO macrophages (p < 0.001) (Fig. 6I). Western blot analysis results showed that FABP5 deletion significantly increased the phosphorylation of AMPKα and decreased the phosphorylation of p65 (p < 0.001) (Supplemental Fig. 4D). To further confirm whether FABP5 acts through the AMPK pathway, we knocked down AMPK by siRNA. Western blot analysis results showed that siRNA 1 or siRNA 2 significantly knocked down AMPKα expression in both WT and FABP5KO macrophages (Fig. 7A), indicating the efficiency of siRNA to knock down AMPKα expression in macrophages. Importantly, siRNA 1 or siRNA 2 significantly rescued the decreased phosphorylation of p65 in LPS-stimulated FABP5KO macrophages, and there was no difference in phosphorylation of p65 between LPS-stimulated WT and FABP5KO groups after siRNA 1 or siRNA 2 treatment (Fig. 7A). Meanwhile, flow cytometry results showed that siRNA 1 or siRNA 2 significantly reversed the decreased proportion of TNF-α+ cells in LPS-stimulated FABP5KO macrophages (Fig. 7B, 7C). These results collectively suggested that the reduced inflammatory response to LPS in FABP5KO macrophages was likely mediated by the upregulated AMPK activation, which subsequently inhibits the NF-κB signaling pathway.

FIGURE 7.

Knockdown of AMPK expression by siRNA reverses the reduction of inflammatory response caused by FABP5-deleted macrophages. BMDMs were obtained from bone marrow cells induced by M-CSF for 7 d. (A) Immunoblot analysis of AMPKα, p65, and p-p65 protein in WT and FABP5KO macrophages under different siRNA treatments (n = 3). 0 represents siRNA negative (meaningless RNA sequence), 1 represents siRNA 1 (siAMPKα1-1 + siAMPKα2-1), and 2 represents siRNA 2 (siAMPKα1-2 + siAMPKα2-2). (B and C) Flow cytometric analysis of TNF-α expression (B) in WT and FABP5KO F4/80+CD11b+ BMDMs under different siRNA treatments. (C) Statistical histogram (n = 3). (D) Schematic diagram of the signaling pathway that FABP5 deletion regulates the inflammatory response in macrophages. Data are representative of three independent experiments. Data are presented as mean ± SD (n = 3). ***p < 0.001, Student t test.

FIGURE 7.

Knockdown of AMPK expression by siRNA reverses the reduction of inflammatory response caused by FABP5-deleted macrophages. BMDMs were obtained from bone marrow cells induced by M-CSF for 7 d. (A) Immunoblot analysis of AMPKα, p65, and p-p65 protein in WT and FABP5KO macrophages under different siRNA treatments (n = 3). 0 represents siRNA negative (meaningless RNA sequence), 1 represents siRNA 1 (siAMPKα1-1 + siAMPKα2-1), and 2 represents siRNA 2 (siAMPKα1-2 + siAMPKα2-2). (B and C) Flow cytometric analysis of TNF-α expression (B) in WT and FABP5KO F4/80+CD11b+ BMDMs under different siRNA treatments. (C) Statistical histogram (n = 3). (D) Schematic diagram of the signaling pathway that FABP5 deletion regulates the inflammatory response in macrophages. Data are representative of three independent experiments. Data are presented as mean ± SD (n = 3). ***p < 0.001, Student t test.

Close modal

Under different induction conditions, macrophages can differentiate into proinflammatory macrophages or anti-inflammatory macrophages (44). LPS-induced macrophages express the corresponding proinflammatory factors TNF-α, IL-1β, and IL-6 to remove pathogenic microorganisms and increase the physiological inflammatory response (5). Endotoxin shock and sepsis caused severe liver damage and increased mortality, in which inflammatory macrophages play an essential role in acute liver injury (45). Our study found that the expression of the FABP5 gene in macrophages gradually decreased with LPS stimulation. FABP5 deficiency reduced the expression of inflammatory genes such as TNF-α, IL-6, and IL-1β in both PEMs and BMDMs after LPS stimulation. Myeloid-specific deletion of FABP5 significantly reduced the levels of inflammatory cytokines such as TNF-α and IL-6 in the liver and sera, inflammatory cell infiltration in the liver, and mouse mortality in an LPS-induced endotoxin shock model or cecal content–induced sepsis mouse model. This is consistent with reports that FABP5 deletion can reduce COX2 and IL-6 expression and inhibit early atherosclerosis in mice (12). The continually decreased FABP5 expression induced by LPS and the essential role of FABP5 in the inflammatory response of M1 macrophages indicated that the late downregulated FABP5 expression might contribute to the turnoff of the inflammatory response in macrophages at the late stage of LPS stimulation.

As lipid chaperones, FABPs actively promote lipid transport to specific compartments within the cell, including endoplasmic reticulum, lipid droplets, mitochondria, peroxidase bodies, and the nucleus, and even signal out of the cell by an autocrine or paracrine process (11). FA metabolism analysis showed that unsaturated FAs, but not other FAs, increased in macrophages after FABP5 deletion, and the alteration of oleic acid was the highest among these changed FAs. This means that FABP5 deletion increases the intracellular oleic acid content, possibly because FABP5 deletion reduces the participation of oleic acid in the synthesis reaction and/or lipid transport, which needs to be clarified by a series of experimental studies in the future. To see whether the accumulation of intracellular oleic acid by FABP5 deletion was related to the reduction of macrophage inflammatory response, we increased the intracellular oleic acid content of WT and FABP5KO macrophages by exogenous oleic acid supplementation and found that the LPS-induced inflammatory response in WT macrophages was reduced by additional oleic acid, but no detectable effect was observed in FABP5KO macrophages. This indicates that when the intracellular oleic acid reaches a certain high content, additional increase of oleic acid may not have a further effect on the inflammatory response. Interestingly, excessive intake of oleic acid significantly reduced LPS- or cecal content–induced acute liver injury in WT mice, which is consistent with the previous report that oleic acid can alleviate rheumatoid arthritis (46). These results indicated that high levels of oleic acid in the diet might have a protective effect on the macrophage inflammation-mediated tissue damage, such as in the liver. It should be pointed out that although FABP5 deletion or the addition of oleic acid to increase the content of intracellular oleic acid significantly reduced the inflammatory response to LPS of macrophages, the inhibitory effect of FABP5 deletion and the additional oleic acid supplementation on the inflammatory response to LPS of macrophages was modest. The positive and negative cooperative effects of this pathway with other relevant pathways should be explored in the future.

It is well known that LPS acts on TLRs to activate the NF-κB signaling pathway in LPS-induced macrophages (4750). Our RNA-seq analysis showed that FABP5 deletion downregulated the classical NF-κB signaling pathway. The key downstream genes of the NF-κB pathway, such as Vacm1, Cxcl2, Ptgs2, Gadd45b, Traf1, and Ccl4, in FABP5KO macrophages were significantly decreased as determined by real-time PCR analysis. Our study also found that the increased content of intracellular oleic acid as a result of FABP5 deletion or oleic acid supplementation decreased the activation of key proteins in the classic NF-κB signaling pathway. These results suggested that FABP5 deficiency reduced the inflammatory response of macrophages by regulating the activation of the classic NF-κB signaling pathway. Many upstream signaling pathways can affect the classic NF-κB signaling pathway. Which pathway mediates the connection of FABP5 with the NF-κB pathway in macrophages? Our RNA-seq and biochemical studies showed that the deletion of FABP5 enhanced the AMPK pathway in LPS-treated macrophages. We treated macrophages with AICAR, an AMPK activator, and found that the percentage of TNF-α+ cells in LPS-stimulated WT and FABP5KO macrophages was significantly decreased, indicating that activation of the AMPK pathway inhibits M1 macrophage polarization. In the reverse case, when we treated macrophages with two AMPK inhibitors, Com C and doxorubicin, respectively, and knocked down AMPKα expression in macrophages by siRNA, we found that inhibition of the AMPK pathway could restore the classical NF-κB signaling pathway and rescued the reduced inflammatory cytokines TNF-α, IL-6, and IL-1β in LPS-treated FABP5KO macrophages. This observation is consistent with the previous studies showing that activation of AMPK by the increased intracellular AMP/ATP ratio inhibited the activity of the NF-κB signaling pathway (51).

In conclusion, the deletion of FABP5 in macrophages protected the liver from endotoxin damage by reducing proinflammatory macrophages in mice. Metabolic analysis showed that FABP5 deletion increased the content of intracellular free unsaturated FAs, especially oleic acid, in LPS-treated macrophages. Additional oleic acid intake reduced macrophage inflammation and acute liver injury by increasing intracellular oleic acid content. Mechanically, either FABP5 deletion or oleic acid addition increased the intracellular AMP/ATP ratio and AMPK activation and subsequently inhibited the classical NF-κB pathway, thereby reducing M1 macrophages and alleviating acute liver injury (Fig. 7D). Inhibiting FABP5 or supplementation of oleic acid might protect the liver from LPS-induced injury.

We appreciate Dr. Yang Zhao for her careful review of our manuscript and Ling Li for her excellent laboratory management.

This work was supported by grants from the National Natural Science Foundation for General and Key Programs (31930041 to Y.Z.), the National Key Research and Development Program of China (2017YFA0105002, 2017YFA0104401, 2017YFA0104402 to Y.Z.), and the Knowledge Innovation Program of the Chinese Academy of Sciences (XDA16030301 to Y.Z.).

flox/flox

Y.H. and Y.Z. conceived the study and experimental design. Y.H., D.W., Z.Z., E.A.B., S.L., and J.B. carried out the experiments; Y.H. and D.W. carried out most of the experiments, analyzed the data, and wrote the manuscript; E.A.B. carried out part of the experiments and revised the manuscript; Z.Z. analyzed the RNA-sequencing data; S.L. and H.X. assisted with the mouse genotyping, flow cytometry, and acute liver injury mouse model; J.B. assisted with the real-time PCR analysis and genotyped the genetically modified mice; L.Z. produced the FABP5-knockout mice and FABP5 mice and revised the manuscript. Y.Z. oversaw the project, designed the experiments, and revised the manuscript.

The RNA-sequencing datasets presented in this article have been submitted to the National Center for Biotechnology Information Sequence Read Archive database (http://www.ncbi.nlm.nih.gov/sra/) under accession number PRJNA786866 (SAMN23720060, SAMN23720061, SAMN23720063, SAMN23720064).

The online version of this article contains supplemental material.

Abbreviations used in this article:

     
  • AICAR

    5-aminoimidazole-4-carboxamide ribonucleotide

  •  
  • AMPK

    AMP-activated protein kinase

  •  
  • BMDM

    bone marrow–derived macrophage

  •  
  • Com C

    compound C

  •  
  • COX2

    cyclooxygenase 2

  •  
  • FA

    fatty acid

  •  
  • FABP5

    fatty acid binding protein 5

  •  
  • IKKα

    inhibitor of NF-κB kinase-α

  •  
  • KO

    knockout

  •  
  • MUFA

    monounsaturated fatty acid

  •  
  • PUFA

    polyunsaturated fatty acid

  •  
  • PEM

    peritoneal macrophage

  •  
  • RNA-seq

    RNA sequencing

  •  
  • WT

    wild type

1.
Strnad
P.
,
F.
Tacke
,
A.
Koch
,
C.
Trautwein
.
2017
.
Liver — guardian, modifier and target of sepsis.
Nat. Rev. Gastroenterol. Hepatol.
14
:
55
66
.
2.
Recknagel
P.
,
F. A.
Gonnert
,
M.
Westermann
,
S.
Lambeck
,
A.
Lupp
,
A.
Rudiger
,
A.
Dyson
,
J. E.
Carré
,
A.
Kortgen
,
C.
Krafft
, et al
2012
.
Liver dysfunction and phosphatidylinositol-3-kinase signalling in early sepsis: experimental studies in rodent models of peritonitis.
PLoS Med.
9
:
e1001338
.
3.
Ramaiah
S. K.
,
H.
Jaeschke
.
2007
.
Role of neutrophils in the pathogenesis of acute inflammatory liver injury.
Toxicol. Pathol.
35
:
757
766
.
4.
Osuchowski
M. F.
,
K.
Welch
,
J.
Siddiqui
,
D. G.
Remick
.
2006
.
Circulating cytokine/inhibitor profiles reshape the understanding of the SIRS/CARS continuum in sepsis and predict mortality.
J. Immunol.
177
:
1967
1974
.
5.
Zhu
L.
,
T.
Yang
,
L.
Li
,
L.
Sun
,
Y.
Hou
,
X.
Hu
,
L.
Zhang
,
H.
Tian
,
Q.
Zhao
,
J.
Peng
, et al
2014
.
TSC1 controls macrophage polarization to prevent inflammatory disease.
Nat. Commun.
5
:
4696
.
6.
Fahy
E.
,
S.
Subramaniam
,
R. C.
Murphy
,
M.
Nishijima
,
C. R. H.
Raetz
,
T.
Shimizu
,
F.
Spener
,
G.
van Meer
,
M. J. O.
Wakelam
,
E. A.
Dennis
.
2009
.
Update of the LIPID MAPS comprehensive classification system for lipids.
J. Lipid Res.
50
(
Suppl
):
S9
S14
.
7.
Schwingshackl
L.
,
G.
Hoffmann
.
2014
.
Monounsaturated fatty acids, olive oil and health status: a systematic review and meta-analysis of cohort studies.
Lipids Health Dis.
13
:
154
.
8.
Berbert
A. A.
,
C. R. M.
Kondo
,
C. L.
Almendra
,
T.
Matsuo
,
I.
Dichi
.
2005
.
Supplementation of fish oil and olive oil in patients with rheumatoid arthritis.
Nutrition
21
:
131
136
.
9.
Linos
A.
,
E.
Kaklamanis
,
A.
Kontomerkos
,
Y.
Koumantaki
,
S.
Gazi
,
G.
Vaiopoulos
,
G. C.
Tsokos
,
P.
Kaklamanis
.
1991
.
The effect of olive oil and fish consumption on rheumatoid arthritis – a case control study.
Scand. J. Rheumatol.
20
:
419
426
.
10.
Massaro
M.
,
R.
De Caterina
.
2002
.
Vasculoprotective effects of oleic acid: epidemiological background and direct vascular antiatherogenic properties.
Nutr. Metab. Cardiovasc. Dis.
12
:
42
51
.
11.
Furuhashi
M.
,
G. S.
Hotamisligil
.
2008
.
Fatty acid-binding proteins: role in metabolic diseases and potential as drug targets.
Nat. Rev. Drug Discov.
7
:
489
503
.
12.
Babaev
V. R.
,
R. P.
Runner
,
D.
Fan
,
L.
Ding
,
Y.
Zhang
,
H.
Tao
,
E.
Erbay
,
C. Z.
Görgün
,
S.
Fazio
,
G. S.
Hotamisligil
,
M. F.
Linton
.
2011
.
Macrophage Mal1 deficiency suppresses atherosclerosis in low-density lipoprotein receptor-null mice by activating peroxisome proliferator-activated receptor-γ-regulated genes.
Arterioscler. Thromb. Vasc. Biol.
31
:
1283
1290
.
13.
Moore
S. M.
,
V. V.
Holt
,
L. R.
Malpass
,
I. N.
Hines
,
M. D.
Wheeler
.
2015
.
Fatty acid-binding protein 5 limits the anti-inflammatory response in murine macrophages.
Mol. Immunol.
67
:
265
275
.
14.
Shi
L.
,
H.
Tian
,
P.
Wang
,
L.
Li
,
Z.
Zhang
,
J.
Zhang
,
Y.
Zhao
.
2021
.
Spaceflight and simulated microgravity suppresses macrophage development via altered RAS/ERK/NFκB and metabolic pathways.
Cell. Mol. Immunol.
18
:
1489
1502
.
15.
Chu
Z.
,
C.
Feng
,
C.
Sun
,
Y.
Xu
,
Y.
Zhao
.
2021
.
Primed macrophages gain long-term specific memory to reject allogeneic tissues in mice.
Cell. Mol. Immunol.
18
:
1079
1081
.
16.
Chu
Z.
,
C.
Sun
,
L.
Sun
,
C.
Feng
,
F.
Yang
,
Y.
Xu
,
Y.
Zhao
.
2020
.
Primed macrophages directly and specifically reject allografts.
Cell. Mol. Immunol.
17
:
237
246
.
17.
Hou
Y.
,
L.
Zhu
,
H.
Tian
,
H. X.
Sun
,
R.
Wang
,
L.
Zhang
,
Y.
Zhao
.
2018
.
IL-23-induced macrophage polarization and its pathological roles in mice with imiquimod-induced psoriasis.
Protein Cell
9
:
1027
1038
.
18.
Yang
T.
,
L.
Zhu
,
Y.
Zhai
,
Q.
Zhao
,
J.
Peng
,
H.
Zhang
,
Z.
Yang
,
L.
Zhang
,
W.
Ding
,
Y.
Zhao
.
2016
.
TSC1 controls IL-1β expression in macrophages via mTORC1-dependent C/EBPβ pathway.
Cell. Mol. Immunol.
13
:
640
650
.
19.
Chen
Y.
,
Y.
Li
,
H.
Guo
,
Z.
Zhang
,
J.
Zhang
,
X.
Dong
,
Y.
Liu
,
Y.
Zhuang
,
Y.
Zhao
.
2021
.
The effects of adoptively transferred IL-23/IL-18-polarized neutrophils on tumor and collagen-induced arthritis in mice.
J. Inflamm. Res.
14
:
4669
4686
.
20.
Liang
Z.
,
Q.
Zhang
,
X.
Dong
,
Z.
Zhang
,
H.
Wang
,
J.
Zhang
,
Y.
Zhao
.
2021
.
mTORC2 negatively controls the maturation process of medullary thymic epithelial cells by inhibiting the LTβR/RANK-NF-κB axis.
J. Cell. Physiol.
236
:
4725
4737
.
21.
Wang
F.
,
S.
Zhang
,
I.
Vuckovic
,
R.
Jeon
,
A.
Lerman
,
C. D.
Folmes
,
P. P.
Dzeja
,
J.
Herrmann
.
2018
.
Glycolytic stimulation is not a requirement for M2 macrophage differentiation.
Cell Metab.
28
:
463
475.e4
.
22.
Zhang
Q.
,
Z.
Liang
,
J.
Zhang
,
T.
Lei
,
X.
Dong
,
H.
Su
,
Y.
Chen
,
Z.
Zhang
,
L.
Tan
,
Y.
Zhao
.
2021
.
Sirt6 regulates the development of medullary thymic epithelial cells and contributes to the establishment of central immune tolerance.
Front. Cell Dev. Biol.
9
:
655552
.
23.
Pertea
M.
,
D.
Kim
,
G. M.
Pertea
,
J. T.
Leek
,
S. L.
Salzberg
.
2016
.
Transcript-level expression analysis of RNA-seq experiments with HISAT, StringTie and Ballgown.
Nat. Protoc.
11
:
1650
1667
.
24.
Wang
L.
,
Z.
Feng
,
X.
Wang
,
X.
Wang
,
X.
Zhang
.
2010
.
DEGseq: an R package for identifying differentially expressed genes from RNA-seq data.
Bioinformatics
26
:
136
138
.
25.
Huang
W.
,
B. T.
Sherman
,
R. A.
Lempicki
.
2009
.
Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources.
Nat. Protoc.
4
:
44
57
.
26.
Xie
C.
,
X.
Mao
,
J.
Huang
,
Y.
Ding
,
J.
Wu
,
S.
Dong
,
L.
Kong
,
G.
Gao
,
C. Y.
Li
,
L.
Wei
.
2011
.
KOBAS 2.0: a web server for annotation and identification of enriched pathways and diseases.
Nucleic Acids Res.
39
(
Suppl 2
):
W316
W322
.
27.
Karnovsky
A.
,
T.
Weymouth
,
T.
Hull
,
V. G.
Tarcea
,
G.
Scardoni
,
C.
Laudanna
,
M. A.
Sartor
,
K. A.
Stringer
,
H. V.
Jagadish
,
C.
Burant
, et al
2012
.
Metscape 2 bioinformatics tool for the analysis and visualization of metabolomics and gene expression data.
Bioinformatics
28
:
373
380
.
28.
Shannon
P.
,
A.
Markiel
,
O.
Ozier
,
N. S.
Baliga
,
J. T.
Wang
,
D.
Ramage
,
N.
Amin
,
B.
Schwikowski
,
T.
Ideker
.
2003
.
Cytoscape: a software environment for integrated models of biomolecular interaction networks.
Genome Res.
13
:
2498
2504
.
29.
Liu
R.
,
J.
Cui
,
Y.
Sun
,
W.
Xu
,
Z.
Wang
,
M.
Wu
,
H.
Dong
,
C.
Yang
,
S.
Hong
,
S.
Yin
,
H.
Wang
.
2021
.
Autophagy deficiency promotes M1 macrophage polarization to exacerbate acute liver injury via ATG5 repression during aging.
Cell Death Discov.
7
:
397
.
30.
Hsiao
H. B.
,
J. B.
Wu
,
H.
Lin
,
W. C.
Lin
.
2011
.
Kinsenoside isolated from Anoectochilus formosanus suppresses LPS-stimulated inflammatory reactions in macrophages and endotoxin shock in mice.
Shock
35
:
184
190
.
31.
Li
Q.
,
Y.
Tan
,
S.
Chen
,
X.
Xiao
,
M.
Zhang
,
Q.
Wu
,
M.
Dong
.
2021
.
Irisin alleviates LPS-induced liver injury and inflammation through inhibition of NLRP3 inflammasome and NF-κB signaling.
J. Recept. Signal Transduct. Res.
41
:
294
303
.
32.
Young
W. A.
,
E. A.
Fallon
,
D. S.
Heffernan
,
P. A.
Efron
,
W. G.
Cioffi
,
A.
Ayala
.
2017
.
Improved survival after induction of sepsis by cecal slurry in PD-1 knockout murine neonates.
Surgery
161
:
1387
1393
.
33.
Batista-Gonzalez
A.
,
R.
Vidal
,
A.
Criollo
,
L. J.
Carreño
.
2020
.
New insights on the role of lipid metabolism in the metabolic reprogramming of macrophages.
Front. Immunol.
10
:
2993
.
34.
Hah
Y. S.
,
Y. H.
Cheon
,
H. S.
Lim
,
H. Y.
Cho
,
B. H.
Park
,
S. O.
Ka
,
Y. R.
Lee
,
D. W.
Jeong
,
H. O.
Kim
,
M. K.
Han
,
S. I.
Lee
.
2014
.
Myeloid deletion of SIRT1 aggravates serum transfer arthritis in mice via nuclear factor-κB activation.
PLoS One
9
:
e87733
.
35.
Kono
Y.
,
S.
Kawakami
,
Y.
Higuchi
,
F.
Yamashita
,
M.
Hashida
.
2014
.
In vitro evaluation of inhibitory effect of nuclear factor-κB activity by small interfering RNA on pro-tumor characteristics of M2-like macrophages.
Biol. Pharm. Bull.
37
:
137
144
.
36.
Israël
A.
2010
.
The IKK complex, a central regulator of NF-κB activation.
Cold Spring Harb. Perspect. Biol.
2
:
a000158
37.
Yang
H. T.
,
S.
Papoutsopoulou
,
M.
Belich
,
C.
Brender
,
J.
Janzen
,
T.
Gantke
,
M.
Handley
,
S. C.
Ley
.
2012
.
Coordinate regulation of TPL-2 and NF-κB signaling in macrophages by NF-κB1 p105.
Mol. Cell. Biol.
32
:
3438
3451
.
38.
Sun
S. C.
2017
.
The non-canonical NF-κB pathway in immunity and inflammation.
Nat. Rev. Immunol.
17
:
545
558
.
39.
Chen
X.
,
X.
Li
,
W.
Zhang
,
J.
He
,
B.
Xu
,
B.
Lei
,
Z.
Wang
,
C.
Cates
,
T.
Rousselle
,
J.
Li
.
2018
.
Activation of AMPK inhibits inflammatory response during hypoxia and reoxygenation through modulating JNK-mediated NF-κB pathway.
Metabolism
83
:
256
270
.
40.
Xiang
H. C.
,
L. X.
Lin
,
X. F.
Hu
,
H.
Zhu
,
H. P.
Li
,
R. Y.
Zhang
,
L.
Hu
,
W. T.
Liu
,
Y. L.
Zhao
,
Y.
Shu
, et al
2019
.
AMPK activation attenuates inflammatory pain through inhibiting NF-κB activation and IL-1β expression.
J. Neuroinflammation
16
:
34
41.
Samanta
J.
,
A.
Mondal
,
S.
Saha
,
S.
Chakraborty
,
A.
Sengupta
.
2020
.
Oleic acid protects from arsenic-induced cardiac hypertrophy via AMPK/FoxO/NFATc3 pathway.
Cardiovasc. Toxicol.
20
:
261
280
.
42.
Lin
S. C.
,
D. G.
Hardie
.
2018
.
AMPK: sensing glucose as well as cellular energy status.
Cell Metab.
27
:
299
313
.
43.
Liu
X.
,
R. R.
Chhipa
,
I.
Nakano
,
B.
Dasgupta
.
2014
.
The AMPK inhibitor compound C is a potent AMPK-independent antiglioma agent.
Mol. Cancer Ther.
13
:
596
605
.
44.
Gordon
S.
2003
.
Alternative activation of macrophages.
Nat. Rev. Immunol.
3
:
23
35
.
45.
Possamai
L. A.
,
M. R.
Thursz
,
J. A.
Wendon
,
C. G.
Antoniades
.
2014
.
Modulation of monocyte/macrophage function: a therapeutic strategy in the treatment of acute liver failure.
J. Hepatol.
61
:
439
445
.
46.
Kremer
J. M.
,
D. A.
Lawrence
,
W.
Jubiz
,
R.
DiGiacomo
,
R.
Rynes
,
L. E.
Bartholomew
,
M.
Sherman
.
1990
.
Dietary fish oil and olive oil supplementation in patients with rheumatoid arthritis. Clinical and immunologic effects.
Arthritis Rheum.
33
:
810
820
.
47.
Ju
M.
,
B.
Liu
,
H.
He
,
Z.
Gu
,
Y.
Liu
,
Y.
Su
,
D.
Zhu
,
J.
Cang
,
Z.
Luo
.
2018
.
MicroRNA-27a alleviates LPS-induced acute lung injury in mice via inhibiting inflammation and apoptosis through modulating TLR4/MyD88/NF-κB pathway.
Cell Cycle
17
:
2001
2018
.
48.
Blood Editors
.
2013
.
Pandey MK, Sung B, Ahn KS, Kunnumakkara AB, Chaturvedi MM, Aggarwal BB. Gambogic acid, a novel ligand for transferrin receptor, potentiates TNF-induced apoptosis through modulation of the nuclear factor-κB signaling pathway. Blood. 2007;110(10):3517-3525.
Blood
121
:
3778
.
49.
Yuan
Q.
,
D.
Zhang
,
C.
Liu
,
C.
Zhang
,
D.
Yuan
.
2018
.
Chikusetsusaponin V inhibits LPS-activated inflammatory responses via SIRT1/NF-κB signaling pathway in RAW264.7 cells.
Inflammation
41
:
2149
2159
.
50.
Kauppinen
A.
,
T.
Suuronen
,
J.
Ojala
,
K.
Kaarniranta
,
A.
Salminen
.
2013
.
Antagonistic crosstalk between NF-κB and SIRT1 in the regulation of inflammation and metabolic disorders.
Cell. Signal.
25
:
1939
1948
.
51.
Chiang
C. F.
,
T. T.
Chao
,
Y. F.
Su
,
C. C.
Hsu
,
C. Y.
Chien
,
K. C.
Chiu
,
S. G.
Shiah
,
C. H.
Lee
,
S. Y.
Liu
,
Y. S.
Shieh
.
2017
.
Metformin-treated cancer cells modulate macrophage polarization through AMPK-NF-κB signaling.
Oncotarget
8
:
20706
20718
.

The authors declare no conflicts of interest.

Supplementary data