Kruppel-like factor 6 and miR-223 signaling axis regulates macrophage-mediated inflammation

Abstract

Macrophage-mediated inflammation is an explicitly robust biologic response that plays a critical role in maintaining tissue homeostasis by eliminating deleterious agents. These tissue macrophages tailor appropriate responses to external cues by altering inflammatory gene expression. Therefore, transcription factors and regulators that modulate inflammatory gene expression play an essential role in shaping the macrophage inflammatory response. Here, we identify that Kruppel-like factor (KLF)6 promotes inflammation by restraining microRNA-223 (miR-223) expression in macrophages. We uncovered that pro- and anti-inflammatory agents oppositely regulate KLF6 and miR-223 expression in macrophages. Using complementary gain- and loss-of-function studies, we observed that overexpression of KLF6 attenuates and deficiency of KLF6 elevates miR-223 expression in macrophages. Furthermore, heightened miR-223 expression in KLF6-deficient macrophages significantly attenuates inducible proinflammatory gene expression. Concordantly, myeloid-Klf6 deficiency significantly curbs diet-induced adipose tissue inflammation, obesity, glucose intolerance, and insulin resistance. At the molecular level, KLF6 directly represses miR-223 expression by occupying its promoter region. More importantly, genetic inhibition of miR-223-3P in KLF6-deficient macrophages completely reversed attenuated proinflammatory gene expression in macrophages. Collectively, our studies reveal that KLF6 promotes proinflammatory gene expression and functions by repressing miR-223 expression in macrophages.—Kim, G.-D., Ng, H. P., Patel, N., Mahabeleshwar, G. H. Kruppel-like factor 6 and miR-223 signaling axis regulates macrophage-mediated inflammation.

The innate immune response is essential for combating infection or injury and has served as a major force that has shaped human evolution (1). The monocyte-derived macrophages are the major innate immune cells recruited to the site of infection or injury that governs the tissue repair or regeneration through matrix remodeling, recruitment of fibroblasts, elimination of dead cells, and development of new blood vessels. However, macrophages that are recruited because of sterile inflammation contribute to chronic inflammatory disease pathogenesis (2). These classically activated macrophages are characterized by the production of high levels of TNF, IFN-γ, IL-1β, and consequently promote a robust proinflammatory milieu (3). In contrast, the alternatively activated tissue-resident macrophages are less sensitive to proinflammatory stimuli and are a significant source of anti-inflammatory cytokines. Recent clinical, pathologic, and experimental studies support an important role for the macrophages in a broad spectrum of chronic inflammatory conditions, such as obesity and insulin resistance (4, 5). Chronic inflammation is now recognized as a key step in the pathogenesis of obesity-induced insulin resistance and type 2 diabetes mellitus (T2D) (6, 7). Despite the acknowledged importance of inflammatory macrophages in the pathogenesis of sterile inflammation, the molecular events that govern inflammatory gene expression and tissue inflammation remain incompletely understood.

Studies over the past decade revealed that microRNAs (miRNAs) function as fine-tune regulators of inflammatory gene expression and immune cell functions (8). miRNAs are a group of highly conserved small (∼22 nt) noncoding RNAs that bind to 3′-UTRs, which results in protein translation cession as well as rapid mRNA degradation. Multiple studies have established that miRNA-223 (miR-223) is highly conserved across mammalian species (9) and is preferentially expressed in the hematopoietic cells (10–12). Deficiency of miR-223 is known to enhance proinflammatory macrophage activation and pathogenic functions (13–15). Recent human studies have uncovered that obesity and T2D status significantly altered miR-223 expression (16–19). More importantly, macrophage miR-223 deficiency significantly elevated high-fat diet (HFD)-induced metabolic tissue inflammation, obesity, and T2D (14, 15). In addition, myeloid-miR-223 is known to protect against complications of obesity and T2D, such as atherosclerosis (20). Despite the acknowledged importance of miR-223 in metabolic inflammation, the transcriptional events that govern miR-223 expression have not been investigated. Our studies uncovered that Kruppel-like factor (KLF)6 represses miR-223 levels and elevates target proinflammatory gene expression in macrophages.

KLF6 is a unique member of the zinc-finger family of transcription factors that is evolved from the Kruppel-like factor luna (LUNA) gene (21). Based on the genomic structure, intron/exon numbers, sequence differences, distinct DNA-binding domains, the position of nuclear localization signals and unique DNA-binding consensus sequences distinguished KLF6 from other members of the KLF family (21, 22). Studies from the Mahabeleshwar laboratory have demonstrated that KLF6 is predominantly expressed in myeloid cells and promotes proinflammatory gene expression while suppressing anti-inflammatory gene expression in macrophages (23, 24). Our previous molecular studies have provided evidence that KLF6 cooperates with NF-κB to promote proinflammatory gene expression while inhibiting peroxisome proliferator-activated receptor γ or signal transducer and activator of transcription 3 function to attenuate anti-inflammatory gene expression in macrophages (23–25). Furthermore, we previously demonstrated that KLF6 contributes to myeloid cell plasticity in human and experimental models of inflammatory diseases (25). Recent genome-wide association studies have reported that single nucleotide polymorphisms in and around the human KLF6 gene is associated with altered body mass index and T2D (26, 27). In this context, whether myeloid-KLF6 regulates diet-induced obesity and T2D by modulating anti-inflammatory miRNAs expression has not been investigated. In this study, we provide evidence that KLF6 represses miR-223 expression to enhance proinflammatory gene expression in macrophages. KLF6 deficiency elevated and overexpression of KLF6 attenuated miR-223 expression in primary macrophages. Our in vivo studies discovered that KLF6 deficiency significantly attenuated HFD-induced adipose tissue inflammation in vivo. More importantly, myeloid-KLF6–deficient mice are protected from HFD-induced obesity, insulin resistance, and glucose intolerance. Our gene expression analysis studies unveiled that KLF6 deficiency significantly attenuated miR-223 gene targets following proinflammatory cytokine exposure. Interestingly, inhibition of miR-223 completely reversed attenuated miR-223 target gene expression in KLF6-deficient macrophages. Based on our observations, we propose that KLF6 promotes proinflammatory gene expression in part by repressing miR-223 expression in macrophages.

MATERIALS AND METHODS

Materials

LPS was obtained from InvivoGen (San Diego, CA, USA). Recombinant mouse IFN-γ protein (485-MI-100) and recombinant macrophage colony-stimulating factor (M-CSF; 416-ML), and Quantikine ELISA kits for mouse monocyte chemotactic protein-1 (MCP1; MJE00B), TNF (DY410), IL-6 (DY406-05), and IL-10 (M1000B) were obtained from R&D Systems (Minneapolis, MN, USA). Thioglycollate broth was obtained from MilliporeSigma (108190; Burlington, MA, USA). Rat anti-mouse F4/80 antibody was obtained from Accurate Chemical and Scientific (AIADG31240; Jericho, NY, USA). Rabbit anti-mouse CD301 antibody was purchased from Abcam (ab64693; Cambridge, MA, USA). Hematoxylin (7211), eosin (71304), phycoerythrin-Texas Red–labeled anti-ITGAM (CD11b) antibody (RM2817), biotin-conjugated goat anti-rabbit IgG (65-6140), and Lipofectamine (L3000-008) transfection reagents were obtained from Thermo Fisher Scientific (Waltham, MA, USA). Insulin was obtained from Eli Lilly and Co. (Indianapolis, IN, USA). The neutral buffered formalin solution was purchased from MilliporeSigma (HT5012-1CS). Mouse miRNA hairpin inhibitors for mmu-miR-223-3p and mmu-miR-223-5p were obtained from GE Healthcare Dharmacon (Lafayette, CO, USA). Control and Klf6-specific small interfering RNA (siRNA) were obtained from Qiagen (Hilden, Germany). RAW264.7 cell line was obtained from the American Type Culture Collection (Manassas, VA, USA). All the tissue culture supplies were obtained from Corning Life Sciences (Tewksbury, MA, USA). All other chemicals and reagents used were of analytical grade and were obtained from commercial sources.

Cell culture

RAW264.7, J774 macrophage cell lines were cultured in DMEM supplemented with 10% fetal bovine serum, 100 U/ml penicillin, 10 μg/ml streptomycin, and 2 mM glutamine in a humidified incubator (5% CO₂ and 37°C). Mouse thioglycollate-elicited peritoneal macrophages (PMs) were obtained by inducing peritonitis with 3% thioglycollate broth in 8–12-wk-old mice. The adherent macrophage cell population was examined for any contaminating cells by fluorescence-activated cell sorting analysis as described in our previous studies (23, 28). These cells were utilized for the indicated experiments in complete DMEM. Bone marrow–derived macrophages (BMDMs) were generated by ex vivo differentiation of bone marrow cells. Briefly, bone marrow cells from 8–12-wk-old wild-type, Lyz2^cre, and Klf6^fl/fl:Lyz2^cre mice were harvested from the femur and tibia. These bone marrow cells were cultured in cell culture medium supplemented with recombinant mouse M-CSF for 7 d. These BMDMs were collected and utilized for the indicated experiments. Human monocytes were isolated from unfractionated peripheral blood mononuclear cells by negative selection using magnetic beads and then were cultured with recombinant human M-CSF ex vivo for 7 d to generate primary macrophages. All the studies involving human samples were approved by the Case Western Reserve University Institutional Review Board.

Experimental mouse models

All of the animal procedures were approved by the Institutional Animal Care and Use Committee at Case Western Reserve University and conformed to guidelines established by the American Association for Accreditation of Laboratory Animal Care. The investigators performing animal experiments were blinded to mouse genotypes by noncontinuous ear tag numbering. All mice were bred and maintained under pathogen-free conditions, fed standard laboratory chow (Envigo, Huntingdon, United Kingdom), and kept on a 12-h light/dark cycle. The mouse line expressing lysozyme-M promoter-driven Cre recombinase (Lyz2^cre) was obtained from The Jackson Laboratory (Bar Harbor, ME, USA). Myeloid KLF6–specific null mice were generated in our laboratory as previously described (23). In brief, Lyz2^cre mice were crossed with Klf6 floxed (KLF6^fl/fl) mice to generate a mouse line harboring Klf6 floxed allele and Lyz2^cre allele. These mouse lines with 2 Klf6 floxed and Lyz2^cre alleles were used as the KLF6 myeloid–specific null group (on a C57BL/6 background). Mice with only 2 Lyz2^cre alleles were used as the control. Primary macrophages from Klf6^fl/fl:Lyz2^cre mice displayed ∼85–90% reduction in KLF6 mRNA and protein expression, as assessed by quantitative PCR (qPCR) and Western blot analyses (23). Both Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice were fed an HFD (D12492; Research Diets, New Brunswick, NJ, USA) for the indicated time periods to induce obesity and insulin resistance. Mice were maintained in individually ventilated cages that were environmentally controlled with a 12-h light/dark cycle and monitored weekly. Oral glucose tolerance tests were performed (1 g/kg body weight) following an overnight unfed period. Blood glucose levels were monitored at 0-, 15-, 30-, 60-, and 120-min time points using Accu-Chek glucose readers (Roche, Basel, Switzerland). The insulin tolerance test was performed on mice that were unfed for 5 h by injecting insulin (1 U/kg body weight). Blood glucose levels were monitored at 0, 30, 60, 90, and 120 min time points. After 20 wk of HFD, Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice were assessed for body fat distribution using a 7T small animal MRI scanner. Plasma cytokine levels in Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice fed on chow or HFD for 20 wk were quantified for MCP1, TNF, IL-6, and IL-10 as per the manufacturer’s instructions.

Immunohistochemistry

Perigonadal visceral adipose tissues were obtained from mice on the day of experiment harvest, fixed in 10% buffered formalin, embedded in paraffin, and stained with hematoxylin and eosin. To quantify the size and number of the adipocytes, hematoxylin and eosin–stained adipose tissue images were analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA). For determining the macrophage infiltration in adipose tissue, paraffin sections of adipose tissues were deparaffinized in xylene and rehydrated in graded ethanol series. Samples were subjected to antigen retrieval steps with antigen unmasking solution (H-3300; Vector Laboratories, Burlingame, CA, USA). Samples were treated with 0.3% H₂O₂ for 30 min at room temperature and the nonspecific binding was blocked with blocking buffer. Samples were incubated with F4/80 or CD301 antibody overnight at 4°C and were subsequently incubated with biotin-conjugated goat anti-rabbit IgG for 30 min at room temperature. Samples were incubated in avidin-biotin complex reagent (PK-4000; Vector Laboratories), and the immunostaining was visualized using a 3,3'diaminobenzidine reagent (SK-4100; Vector Laboratories). Images were acquired utilizing a microscope and macrophage number, and area were quantified by ImageJ software.

RNA extraction, reverse transcription, and real-time qPCR assay

Total RNA was isolated from indicated cell types after completion of the specified treatment using the High Pure RNA Isolation Kit (11828665001; Roche). One microgram of total RNA is used for reverse transcription using M-MuLV reverse transcriptase (M0253L; New England Biolabs, Ipswich, MA, USA) in the presence of random hexamers and oligo-dT primers (N8080127, 18418020; Thermo Fisher Scientific). The resulting cDNA was subjected to real-time qPCR using gene-specific primers on a Step One Plus Real-Time PCR System (Thermo Fisher Scientific). A list of primers utilized in this study are provided in Table 1.

TABLE 1.

List of primers used

	Primer sequence, 5′–3′
Target gene	Forward	Reverse
mmu-miR-223	GCCATCTGCAGTGTCACG	ATAGGCATGAGCCACACTTG
hsa-miR-223	TCACTTCCCCACAGAAGCTC	CTGGCAGCTCATTCGTCATA
Acvr2a	CCCTCCTGTACTTGTTCCTACTCA	GCAATGGCTTCAACCCTAGT
F3	CCGAGCAATGGAAGAGTTTC	CGCTTGCACAGAGATATGGA
Gpc1	GTGCGGAGAGTGTCATTGG	TCCACAGGCCTGGATGAC
Icam1	CCCACGCTACCTCTGCTC	GATGGATACCTGAGCATCACC
Igf1r	GGTCTCTGAGGCCAGAAGTG	GGTAGGCCATGCCATCTG
Itgb1	CGTGGTTGCCGGAATTGTTC	ACCAGCTTTACGTCCATAGTTTG
Klf6	TCCCACTTGAAAGCACATCA	ACTTCTTGCAAAACGCCACT
Mip2	CCAGCCACACTTCAGCCTA	CAGTTCACTGGCCACAACAG
Mmp9	CTGGACAGCCAGACACTAAAG	CTCGCGGCAAGTCTTCAGAG
Nlrp3	CCCTTGGAGACACAGGACTC	GGTGAGGCTGCAGTTGTCTA
Nox1	CCTGATTCCTGTGTGTCGAAA	TTGGCTTCTTCTGTAGCGTTC
Rassf4	AATCTAGTGCACGCGACCTC	TTCAACTCTCGTCCTCAGACC
Sgms2	CACACTGTCGTGCTCACACTT	CGCTACGAGAATGCAGATGA
Vcam1	ACGTCAGAACAACCGAATCC	GTGGTGCTGTGACAATGACC
Il-1a	GCACCTTACACCTACCAGAGT	AAACTTCTGCCTGACGAGCTT
Il-6	TAGTCCTTCCTACCCCAATTTCC	TTGGTCCTTAGCCACTCCTTC
Tnf	CCCTCACACTCAGATCATCTTCT	GCTACGACGTGGGCTACAG
Mcp1	TTAAAAACCTGGATCGGAACCAA	GCATTAGCTTCAGATTTACGGGT
Nos2	GTTCTCAGCCCAACAATACAAGA	GTGGACGGGTCGATGTCAC
Il-12	TGGTTTGCCATCGTTTTGCTG	ACAGGTGAGGTTCACTGTTTCT
Il-10	GCTCTTACTGACTGGCATGAG	CGCAGCTCTAGGAGCATGTG
Arg1	TTTTAGGGTTACGGCCGGTG	CCTCGAGGCTGTCCTTTTGA
Mrc1	GGACGAGCAGGTGCAGTT	CAACACATCCCGCCTTTC
Chil3l3	CAGGTCTGGCAATTCTTCTGAA	GTCTTGCTCATGTGTGTAAGTGA
Retnla	CCCTCCACTGTAACGAAGACTC	CACACCCAGTAGCAGTCATCC
Myd88	AGGACAAACGCCGGAACTTTT	GCCGATAGTCTGTCTGTTCTAGT
Traf6	AAAGCGAGAGATTCTTTCCCTG	ACTGGGGACAATTCACTAGAGC
Irak4	CATACGCAACCTTAATGTGGGG	GGAACTGATTGTATCTGTCGTCG
Tab1	TCCAACCGCAGCTACTCTG	CCCGTACAGGAAGCAGTTATTTT
Tab2	CATGACCTGCGACAAAAATTCC	TGATTGCGTAGACCAGAAATTCC
Rela	AGGCTTCTGGGCCTTATGTG	TGCTTCTCTCGCCAGGAATAC
Relb	CCGTACCTGGTCATCACAGAG	CAGTCTCGAAGCTCGATGGC
Cul1	CTCAGTTTGTTGGCTTGGAGT	TGGAGAATCGGTAATCTTCCCA
Chuk	GTCAGGACCGTGTTCTCAAGG	GCTTCTTTGATGTTACTGAGGGC
Ikbkb	CTGAAGATCGCCTGTAGCAAA	TCCATCTGTAACCAGCTCCAG
Ikbkg	AAGCACCCCTGGAAGAACC	CCTGCTCTGAAGGCAGATGTA
Nfkb1	ATGGCAGACGATGATCCCTAC	TGTTGACAGTGGTATTTCTGGTG
Ccr2	ATCCACGGCATACTATCAACATC	CAAGGCTCACCATCATCGTAG
36B4	GCTCCAAGCAGATGCAGCA	CCGGATGTGAGGCAGCAG

Transient transfection, luciferase assay, and chromatin immunoprecipitation

RAW264.7 cells were transiently transfected with control siRNA, siKlf6, pCI-neo (empty vector control), and pCI-neo-Klf6 (overexpressing Klf6) with Lipofectamine transfection reagents as specified by the manufacturer’s instructions. Similarly, BMDMs from Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice were transfected with control or miRNA hairpin inhibitors using Lipofectamine transfection reagents. These transfected cells were stimulated with either LPS (100 ng/ml), IFN-γ (10 ng/ml), or PBS (control) and also subjected to indicated analyses.

A wild-type 1.8 kb miR-223 promoter (−4220 to −2404) upstream of the 5′ end of the miR-223 coding sequence was cloned into pGL3 basic vector. The KLF6 binding site (−4156 to −4144) on the miR-223 promoter was distorted by substitution mutation. RAW264.7 cells were transfected with wild-type or KLF6 binding site mutant miR-223 promoter reporter plasmid using the Lipofectamine transfection reagent according to the manufacturer’s instructions. These cells were unstimulated or stimulated with 100 ng/ml LPS for 18 h. Luciferase reporter activity was measured and normalized according to the manufacturer’s instructions. Results are presented as relative luciferase activity over the control group.

Chromatin immunoprecipitation (ChIP) analyses were performed using the EZ-Magna ChIP G Kit (17-409; MilliporeSigma) according to the manufacturer’s instruction. Briefly, wild-type mouse BMDMs were stimulated with 100 ng/ml LPS for 6 h and ChIPs were performed using anti-KLF6 antibody. Primer pairs (forward: 5′-TCAAATGCTATCCCCTTTCC-3′ reverse: 5′-CCAGGGCTAGGAAGTAGGAG-3′) flanking the KLF6-binding site on mouse miR-223 promoter region (−4207 to −4101 upstream of miR-223 coding sequence) were utilized for amplification by real-time quantitative RT-PCR. ChIP performed using isotype IgG was used as a negative control. DNA levels were first normalized to the internal control region in the first intron of the mouse Actb gene (forward: 5′-CGTATTAGGTCCATCTTGAGAGTAC-3′, reverse: 5′-GCCATTGAGGCGTGATCGTAGC-3′). Relative enrichment was calculated by dividing the normalized levels of ChIP DNA to that of input DNA at the corresponding locus.

Statistical analysis

All data, unless indicated, are presented as the means ± sd. The statistical significance of differences between 2 experimental groups were analyzed by a paired or unpaired Student’s t test. Statistical significance of differences between 2 or more in vivo experimental groups were analyzed by using 2-way ANOVA.

RESULTS

Macrophage miR-223 expression is modulated by inflammatory stimuli

Previous studies have demonstrated that alteration in miR-223 levels are associated with human inflammatory disorders and disease outcome in experimental animal models (13–16, 20). Furthermore, miR-223 is an intergenic miRNA whose expression is fine-tuned by a number of transcription factors (29). Therefore, we examined the expression pattern of miR-223 in diverse tissue types. Our analyses indicate that miR-223 is most abundantly expressed in mouse BMDMs compared with any other tissue types (Fig. 1A). Furthermore, evaluation of murine macrophage cell lines and primary cells confirmed that miR-223 is abundantly expressed in PMs and BMDMs of wild-type mice compared with macrophage cell lines (Fig. 1B). Next, we sought to identify the stable form of the mature miR-223 strand in human and murine primary macrophages. Our analyses establish that miR-223-3P is a stable strand in human primary macrophages derived from peripheral blood mononuclear cells (Fig. 1C) as well as mouse BMDMs (Fig. 1D). Next, we examined whether pro- or anti-inflammatory stimuli alter miR-223 expression in primary macrophages. Our studies demonstrate that wild-type BMDMs exposure to anti-inflammatory cytokines (IL-4 or IL-10) or pharmacological agents (dexamethasone or rosiglitazone) significantly elevated miR-223 expression (Fig. 1E, F). Conversely, exposure of wild-type BMDMs to proinflammatory agents, such as LPS or IFN-γ, significantly attenuated miR-223 expression (Fig. 1G, H). More importantly, differentiation of monocytes into macrophages significantly attenuated miR-223 while elevating KLF6 expression in human (Fig. 1I) and murine (Fig. 1J) systems. Taken together, our analyses establish that miR-223 is most abundantly expressed in macrophages and that expression of miR-223 are influenced by inflammatory stimuli and myeloid cell maturation status.

Figure 1.

The miR-223 expression is modulated by inflammatory stimuli. A) miR-223 expression was evaluated in mouse testis, liver, heart, kidney, adipose tissue, lung, spleen, and BMDMs by qPCR analysis, and relative fold changes are indicated (n = 5). B) Mouse macrophage cell lines (RAW264.7 and J774), PMs, and BMDMs were analyzed for miR-223 expression by qPCR analyses (n = 4). C, D) Human monocyte-derived macrophages (C) and wild-type murine BMDMs (D) were analyzed for miR-223-3P and miR-223-5P expression by qPCR analyses (n = 4). E, F) Wild-type murine BMDMs were separately stimulated with IL-4 (20 ng/ml) and IL-10 (20 ng/ml) (E), or dexamethasone (1 μM) and rosiglitazone (5 μM) (F) for 12 h, and miR-223 expression was evaluated by qPCR analysis (n = 4). G, H) Wild-type murine BMDMs were separately stimulated with LPS (100 ng/ml) or IFN-γ (10 ng/ml) for indicated time, and miR-223 expression was evaluated by qPCR analysis (n = 4). I, J) Human and murine monocytes and monocyte-derived macrophages were analyzed for miR-223 and KLF6 mRNA expression by qPCR analysis (n = 4). Data were analyzed by 2-way ANOVA (A) or Student’s t test (B–J). All values are reported as means ± sd. *P < 0.05, **P < 0.01, ***P < 0.001.

KLF6 represses miR-223 expression in macrophages

Our studies uncovered that proinflammatory agents attenuated and anti-inflammatory stimuli elevated miR-223 expression in macrophages. In addition, miR-223 and KLF6 expression are inversely regulated during monocyte to macrophage differentiation (Fig. 1I, J). Our previous studies have demonstrated that proinflammatory agents elevate and anti-inflammatory stimuli attenuate KLF6 expression in human and murine macrophages (23). Therefore, we hypothesized that KLF6 may negatively regulate miR-223 expression in macrophages. To test this hypothesis, KLF6 expression in RAW264.7 cells were altered by Klf6-specific siRNA or Klf6-overexpressing plasmid. Our analyses show that attenuation of KLF6 elevated (Fig. 2A) and overexpression of KLF6 diminished (Fig. 2B) miR-223 expression in macrophages. Next, we examined whether these observations are reproducible in primary macrophages. As shown in Fig. 2C, deficiency of KLF6 significantly elevated the basal level of miR-223 expression in Klf6^fl/fl:Lyz2^cre mouse BMDMs. More importantly, increased miR-223 expression in Klf6^fl/fl:Lyz2^cre mouse BMDMs resulted in heightened miR-223-3P strands abundance under resting condition (Fig. 2D). Previous studies have demonstrated that miR-223 is highly conserved across the species and that primary transcript for miR-223 gene is located on X-chromosome in both humans and rodents (30). Therefore, we examined whether expression of miR-223 and Klf6 are altered in primary macrophages derived from the male and female host. As denoted in Fig. 2E, F, miR-223 or Klf6 expression levels were not significantly altered between male and female mouse primary macrophages. Our previous studies have demonstrated that KLF6 levels are elevated under chronic inflammatory disease conditions (25). Therefore, we probed whether inflammatory milieu established by an HFD altered Klf6 and miR-223 expression in splenic macrophages. Our analyses denote that 8 wk of HFD significantly elevated Klf6 expression while attenuating miR-223 expression in splenic macrophages compared with mice fed on a chow diet (Fig. 2G). However, deficiency of KLF6 completely restrained attenuation of miR-223 expression in splenic macrophages following HFD in Klf6^fl/fl:Lyz2^cre mice (Fig. 2H). Next, we examined whether KLF6 directly represses miR-223 expression by binding to its promoter. Accordingly, BMDMs from wild-type mice were stimulated with LPS, and ChIP was performed using anti-KLF6 antibody. Interestingly, our KLF6-ChIP analysis depicts that LPS stimulation of wild-type BMDMs significantly induced KLF6 occupancy on the miR-223 promoter region (−4207 to −4101) compared with the vehicle control group (Fig. 2J). To validate the functionality of our findings, we constructed the miR-223 promoter-driven luciferase reporter system. Our analyses indicate that LPS exposure significantly attenuated miR-223 promoter-driven luciferase activity in macrophages (Fig. 2K). More importantly, a mutation in KLF6 binding site (−4156 to −4144) on miR-223 promoter robustly elevated luciferase reporter activity and greatly protected LPS-exerted attenuation of miR-223 promoter activity in macrophages (Fig. 2K). Collectively, our studies demonstrate that KLF6 directly represses miR-223 expression by enriching on its promoter region.

Figure 2.

KLF6 repress miR-223 expression in macrophages. A, B) RAW 264.7 cells were transfected with either Klf6 siRNA (A) or pCI-neo-Klf6 (B) plasmid and miR-223 expression was analyzed by qPCR analysis (n = 5). C, D) Total RNA extracts from Lyz2^cre and Klf6^fl/fl:Lyz2^cre mouse BMDMs were analyzed for miR-223 (C) and miR-223-3P (D) expression by qPCR analysis (n = 9). E, F) Total RNA samples from male and female wild-type mouse PMs and BMDMs were analyzed for miR-223 (E) and Klf6 (F) expression by qPCR analysis (n = 5). G) Total RNA samples derived from spleen macrophages following chow or HFD were analyzed for miR-223 and Klf6 expression by qPCR analysis (n = 5). H) Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice were fed on chow or HFD for 8 wk. Total RNA samples derived from spleen macrophages were analyzed for miR-223 expression by qPCR analysis (n = 5). I) Schematic representation of miR-223 promoter. J) Wild-type mouse BMDMs were stimulated with 100 ng/ml LPS for 4 h. The ChIP analysis was performed on the miR-223 promoter (−4207 to −4101) utilizing anti-KLF6 antibody or isotype-specific IgG (n = 3). K) RAW264.7 cells were transfected with wild-type or KLF6-binding mutant miR-223 promoter-driven luciferase reporter plasmid. These cells were stimulated with 100 ng/ml LPS for 18 h, and cell lysates were analyzed for luciferase reporter activity. N.S., not significant. Data were analyzed by Student’s t test. All values are reported as means ± sd. *P < 0.05, ***P < 0.001.

Myeloid-KLF6 deficiency curtails HFD-induced adipose tissue inflammation

Previous studies have reported that loss of miR-223 accelerated obesity-associated adipose tissue inflammation and HFD-induced insulin resistance (14, 15). Our analyses demonstrated that KLF6 represses miR-223 expression in macrophages. Therefore, we examined whether myeloid-specific deficiency of KLF6 altered HFD-induced adipose tissue inflammation and adipocyte morphology in vivo. Accordingly, perigonadal adipose tissues from chow or 20 wk of HFD-fed Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice were subjected to detailed histochemical and immunologic evaluations. Our analyses show that myeloid-KLF6 deficiency did not alter macrophage recruitment to adipose tissues on a chow diet (Fig. 3A–C). However, HFD feeding dramatically elevated macrophage recruitment to adipose tissues in Lyz2^cre mice, and these effects were significantly attenuated in Klf6^fl/fl:Lyz2^cre mice (Fig. 3A–C). Prior studies have highlighted the importance of CCR2/MCP1 axis in the recruitment of monocyte-derived macrophages to adipose tissues that contribute to diet-induced metabolic dysfunction (31, 32). Therefore, we examined whether myeloid-KLF6 deficiency alters adipose tissue Ccr2 levels because of differential recruitment of monocyte-derived macrophages following chow or HFD. Consistent with adipose tissue F4/80 positive macrophage abundance (Fig. 3A–C), our analyses revealed that myeloid-KLF6 deficiency did not significantly alter adipose tissue Ccr2 levels on a chow diet (Fig. 3D). As anticipated, HFD feeding significantly elevated Ccr2 level in Lyz2^cre mouse adipose tissues (Fig. 3D). Interestingly, myeloid-KLF6 deficiency significantly attenuated the HFD-induced elevation of Ccr2 levels in adipose tissues (Fig. 3D). Taken together (Fig. 3A–D), our analyses reveal that myeloid-KLF6 deficiency significantly attenuates monocyte-derived macrophage recruitment to adipose tissues following HFD feeding. Recent studies have highlighted the importance of resident tissue macrophages in health and diseases (33, 34). Therefore, we examined whether myeloid-KLF6 deficiency alters adipose tissue-resident macrophage population on chow or HFD. Our analyses indicated that myeloid-KLF6 deficiency modestly elevated resident macrophage population on chow diet (Fig. 3E–G). As anticipated (35), HFD significantly elevated resident macrophage population in Lyz2^cre mouse adipose tissues (Fig. 3E–G). However, myeloid-KLF6 deficiency did not significantly alter the abundance of resident macrophages following HFD (Fig. 3E–G). Next, we evaluated whether myeloid-KLF6 deficiency altered morphometrics of adipocytes on chow or HFD. Our studies indicate that Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice fed a chow diet contained an approximately similar number of adipocytes (Fig. 3H, I). However, Klf6^fl/fl:Lyz2^cre mice on a chow diet exhibit relatively smaller adipocytes compared with Lyz2^cre mice (Fig. 3H, J). As anticipated, Lyz2^cre mice fed an HFD exhibited significantly enlarged adipocytes and hence reduced adipocyte number for a given area (Fig. 3H–J). Interestingly, this increase in adipocyte size and reduction in adipocyte number for a given area were not observed in Klf6^fl/fl:Lyz2^cre mice fed on HFD (Fig. 3H–J). Concurrently, we examined whether myeloid-KLF6 deficiency altered macrophage-specific classic pro- or anti-inflammatory gene expression in white adipose tissue following HFD. Our analyses show that myeloid-KLF6 deficiency significantly attenuated proinflammatory gene transcript levels (Il1α, Il6, Tnf, Mcp1, Nos2, Il12 and Vcam1) while elevating anti-inflammatory (Il10, Arg1, Chil3l3 and Retnla) gene expression in adipose tissue (Fig. 3K). Furthermore, analyses of Lyz2^cre and Klf6^fl/fl:Lyz2^cre mouse plasma cytokines on chow and HFD demonstrate that myeloid-KLF6 deficiency significantly attenuated HFD-induced proinflammatory cytokines (MCP1, TNF, and IL-6) levels while modestly elevating anti-inflammatory cytokine (IL-10) level (Fig. 3L–O). Taken together, our studies establish that myeloid-KLF6 deficiency is protective against HFD-induced adipose tissue inflammation and adverse morphometric changes of adipocytes.

Figure 3.

Myeloid-KLF6 deficiency curtails HFD-induced adipose tissue inflammation. A–C) Perigonadal adipose tissue sections of Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice fed on chow (CW) or HFD for 20 wk were stained for macrophages by anti-F4/80 antibody (A). Area (B) and number (C) of F4/80 positive cells were quantified using ImageJ software (n = 8). D) Total RNA extracts from CW- or HFD-fed mouse adipose tissue was analyzed for expression of Ccr2 by quantitative RT-PCR analysis (n = 5). E–G) Perigonadal adipose tissue sections of Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice fed CW or HFD for 20 wk were stained for resident macrophages by anti-CD301 antibody (E). Area (F) and number (G) of CD301-positive cells were quantified using ImageJ software (n = 8). H–J) Perigonadal adipose tissue sections of Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice fed chow or HFD for 20 wk were stained for hematoxylin and eosin (H) to quantify adipocyte number (I) and size (J) (n = 8). K) Total RNA extracts from HFD-fed Lyz2^cre and Klf6^fl/fl:Lyz2^cre mouse adipose tissue were analyzed for expression of Il-1α, Il-6, Tnf, Mcp1, Nos2, Il-12, Vcam1, Il-10, Arg1, Mrc1, Chil3l3, and Retnla by quantitative RT-PCR analysis (n = 5). L–O) Plasma level of MCP1 (L), TNF (M), IL-6 (N), and IL-10 (O) were analyzed by Quantikine ELISA Kit (n = 5). N.S., not significant. Data were analyzed by 2-way ANOVA. All values are reported as means ± sd. Scale bars, 100 μm. *P < 0.05, **P < 0.01, ***P < 0.001.

Myeloid-KLF6 deficiency attenuates HFD-induced obesity and insulin resistance

Previous studies have demonstrated that macrophage-mediated adipose tissue inflammation accelerates the development of metabolic syndrome (4–7). Our studies established that myeloid-KLF6 deficiency is protective against HFD-induced adipose tissue inflammation (Fig. 3). Therefore, we examined whether the myeloid-specific deficiency of KLF6 would alter the development of diet-induced obesity and metabolic syndrome in vivo. Accordingly, cohorts of Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice were fed chow or an HFD, and body weight gain was recorded. As shown in Fig. 4A, Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice fed a chow diet did not exhibit any significant difference in body weight gain over a period of 17 wk. As anticipated, Lyz2^cre mice exhibited rapid body weight gain on the HFD (Fig. 4B). Surprisingly, Klf6^fl/fl:Lyz2^cre mice were highly protected from HFD-induced body weight gain and maintained significantly lower body weight compared with Lyz2^cre mice (Fig. 4B). Consistent with these observations, MRI analyses indicate that Klf6^fl/fl:Lyz2^cre mice fed on HFD exhibit significantly lower adipose tissue volume compared with Lyz2^cre mice fed on HFD (Fig. 4C, D). To explore the role of myeloid-KLF6 in obesity-induced insulin resistance, we performed insulin and glucose tolerance tests on HFD-fed Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice. Analysis of plasma glucose concentration following insulin administration uncovered that Klf6^fl/fl:Lyz2^cre mice are significantly more sensitive to insulin challenge compared with Lyz2^cre mice (Fig. 4E). Consistent with these observations, oral glucose tolerance testing revealed an enhanced ability of Klf6^fl/fl:Lyz2^cre mice to clear glucose compared with Lyz2^cre mice (Fig. 4F). Collectively, these observations illustrate that myeloid-Klf6 deficiency significantly attenuates HFD-induced obesity and insulin resistance in vivo.

Figure 4.

A, B) Myeloid-KLF6 deficiency attenuates HFD-induced obesity and T2D. A, B) Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice were fed on chow (A) or HFD (B), and body weight gains were recorded (n = 10). Data were analyzed by ANOVA. C, D) Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice that were fed an HFD for 20 wk were subjected to MRI (C) to assess the distribution and quantification of adipose tissue volume (D) (n = 8). E, F) Data were analyzed by Student’s t test. Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice were subjected to insulin tolerance test (E) or oral glucose tolerance test (F) after 16 wk of HFD (n = 8). N.S., not significant. Data were analyzed by 2-way ANOVA (A, B, E, F). All values are reported as means ± sd. *P < 0.05, **P < 0.01, ***P < 0.001.

miR-223 targets are attenuated in KLF6-deficient macrophages

Next, we sought to identify the putative miR-223-3P inflammatory target genes attenuated in KLF6-deficient macrophages. Utilizing publicly available miRNA target prediction databases (36–38), we have identified a set of miR-223 inflammatory gene targets (Fig. 5A). Our analyses depict that miR-223-3P binding sites on 3′-UTR of inflammatory targets genes are highly conserved in humans and mice (Fig. 5A). Next, we assessed whether deficiency of KLF6 altered miR-223-3P target gene expression in macrophages. Accordingly, BMDMs from Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice were stimulated with IFN-γ or LPS, and expression of miR-223-3P proinflammatory gene targets were evaluated by qPCR analyses. As shown in Fig. 5B, IFN-γ stimulation robustly elevated Vcam1, Icam1, Gpc1, F3, Nox1, Rassf4, and Acvr2a expression in Lyz2^cre mouse BMDMs. Compellingly, IFN-γ–induced expression of these proinflammatory gene targets were significantly diminished in Klf6^fl/fl:Lyz2^cre mouse BMDMs (Fig. 5B). Concordant with these observations, LPS stimulation strongly induced Mmp9, Mip2, Gpc1, Sgms2, Nlrp3, Itgb1, and Igf1r expression in Lyz2^cre mouse BMDMs (Fig. 5C). Intriguingly, LPS-induced expression of these proinflammatory gene targets were significantly dampened in Klf6^fl/fl:Lyz2^cre mouse BMDMs (Fig. 5C). Recent studies have demonstrated that miR-223 suppresses canonical NF-κB signaling in a zebrafish model (39). Therefore, we examined whether upstream regulators of NF-κB were altered in KLF6-deficient macrophages (Fig. 5D). Our analyses unveiled that Irak4, Rela, and Nfkb1 mRNA expression were significantly attenuated in Klf6^fl/fl:Lyz2^cre mouse BMDMs. Taken together, these results demonstrate that KLF6 deficiency resulted in significant attenuation of inducible expression of miR-223-3P proinflammatory targets as well as positive regulators of NF-κB signaling in macrophages.

Figure 5.

Deficiency of KLF6 attenuates miR-223 target gene expression in macrophages. A) Conserved miR-223-3P target sites in the 3′ UTRs of human and mouse genes associated with vascular inflammation. B, C) Lyz2^cre and Klf6^fl/fl:Lyz2^cre mouse BMDMs were stimulated with 10 ng/ml IFN-γ (B) or 100 ng/ml LPS (C) for 6 h. Total RNA samples from these experiments were evaluated for expression of indicated genes by qPCR analysis (n = 4). D) Total RNA extracts from Lyz2^cre and Klf6^fl/fl:Lyz2^cre mouse BMDMs were analyzed for expression of Myd88, Traf6, Irak4, Tab1, Tab2, Rela, Relb, Cul1, Chuk, Ikbkb, Ikbkg, and Nfkb1 by quantitative RT-PCR analysis (n = 4). The 36B4 gene was used as a housekeeping gene for qPCR analyses. N.S., not significant. Data were analyzed by Student’s t test. All values are reported as means ± sd. *P < 0.05, **P < 0.01, ***P < 0.001.

Blockade of miR-223 reverse attenuated proinflammatory target gene expression in KLF6-deficient macrophages

Our studies so far had demonstrated that deficiency of KLF6 resulted in elevated miR-223-3P expression and attenuated miR-223-3P proinflammatory target gene expression in macrophages. Therefore, we examined whether diminished levels of proinflammatory gene expression in KLF6-deficient macrophages are miR-223-3P dependent. We employed genetic approaches to block miR-223-3P function in primary macrophages. Accordingly, BMDMs from Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice were transfected with control or mmu-miR-223-3p–specific hairpin inhibitor. These cells were stimulated with LPS or IFN-γ, and miR-223-3P inflammatory target gene expression was evaluated by qPCR analysis (Fig. 6A–C). As anticipated, LPS or IFN-γ–induced proinflammatory and upstream positive regulators of NF-κB activation was attenuated in Klf6^fl/fl:Lyz2^cre mouse BMDMs (Fig. 6A–C). Intriguingly, blockade of functional miR-223-3P in KLF6-deficient macrophages completely reverse attenuated Itgb1, Mip2, Igf1r, Gpc1, Sgms2, Mmp9, Rela, Ikbkb, and Nfkb1 expression following LPS stimulation (Fig. 6A, B). Concordant with these observations, inhibition of miR-223-3P expression in KLF6-deficient macrophages completely rescued diminished Icam1, F3, and Vcam1 expression following IFN-γ stimulation (Fig. 6C). Taken together, our results illustrate that attenuation of NF-κB signaling and proinflammatory gene expression in KLF6-deficient macrophages are, in part, dependent on elevated miR-223-3P expression.

Figure 6.

Blockade of miR-223 reverse attenuated proinflammatory target gene expression in Klf6-deficient macrophages. Lyz2^cre and Klf6^fl/fl:Lyz2^cre mouse BMDMs were transfected with control or mmu-miR-223-3p–specific hairpin inhibitors, and the cells were stimulated with 100 ng/ml of LPS or 10 ng/ml IFN-γ for 6 h. Total RNA from these experiments was analyzed for Itgb1, Mip2, Igf1r, Gpc1, Sgms2, Mmp9 (A); Rela, Ikbkb, Nfkb1 (B); and Icam1, F3 and Vcam1 (C) by qPCR (n = 4). The 36B4 gene was used as a housekeeping gene for qPCR analyses. N.S., not significant. Data were analyzed by 2-way ANOVA. All values are reported as means ± sd. *P < 0.05, **P < 0.01, ***P < 0.001.

DISCUSSION

Our studies are the first to describe that KLF6 elevates macrophage-mediated inflammation by directly repressing anti-inflammatory miR-223 expression in macrophages. Genetic blockade of miR-223-3P completely reversed attenuated proinflammatory gene expression in KLF6-deficient macrophages. The key findings of our study are as follows: 1) miR-223 is most abundantly expressed in primary macrophages and expression of miR-223 is modulated by inflammatory stimuli, 2) KLF6 deficiency elevates basal level of miR-223 expression in macrophages, 3) KLF6 directly represses miR-223 expression by enriching on its promoter region, 4) myeloid-KLF6 deficiency attenuates HFD-induced monocyte-derived macrophage-mediated adipose tissue inflammation, 5) myeloid-KLF6 deficiency did not alter HFD-induced resident macrophage population expansion in adipose tissue, 6) myeloid-KLF6 deficiency curtails HFD-induced obesity and insulin resistance in vivo, 7) inducible proinflammatory gene targets of miR-223 are attenuated in KLF6-deficient macrophages, 8) heightened miR-223 expression attenuated positive regulators of NF-κB in KLF6-deficient macrophages, and 9) blockade of miR-223 reverse attenuated proinflammatory target gene expression in KLF6-deficient macrophages. Collectively, these findings establish that KLF6 boosts inflammation by suppressing miR-223 expression in macrophages (Fig. 7).

Figure 7.

KLF6 promotes macrophage-mediated inflammation by repressing miR-223 expression.

The transcriptional regulation of inducible inflammatory gene expression involves coordinated regulation of multiple transcription factors and cofactors. Studies over the last decade have uncovered miRNAs as important regulators of inflammatory gene expression in human health and diseases (8). miRNAs will add an additional layer of regulation in fine-tuning inflammatory gene expression in myeloid and nonmyeloid immune cells. Recent studies have revealed that miR-223 is a critical regulator of myeloid cell differentiation, and its dysregulation is correlated with human inflammatory disease conditions (12, 16, 30). Our previous studies have demonstrated that KLF6 elevates inflammatory gene expression in myeloid cells under ex vivo and in vivo settings (23–25). In this context, our current studies identified a novel mechanism by which KLF6 elevates proinflammatory gene expression by directly repressing miR-223 expression in inflammatory macrophages. Prior studies have demonstrated that miR-223 expression in myeloid cells is regulated by C/EBPα, GATA-1, NFI-A, and PU.1 transcription factors (9, 30). Therefore, it is feasible that KLF6 may interact with one of these key myeloid cell lineage–defining transcription factors to modulate miR-223 expression in macrophages. Past studies have demonstrated that miR-223 is required for granulocytic differentiation utilizing NB4 cell system (30). Interestingly, miR-223–deficient mice are viable and did not exhibit any gross abnormalities hematopoietic system except granulocyte functions (12). Indeed, hemizygous miR-223^−/y mice exhibited neutrophilia in advanced age (12). In this context, myeloid-KLF6–deficient mice did not exhibit any major changes in neutrophil or nonmyeloid immune cell compartments (23). It is important to note that differentiation of monocyte into macrophages attenuated miR-223 expression while elevating KLF6 levels in macrophages. This helps to establish an inflammatory milieu in macrophage-mediated tissue inflammation. In this context, past studies have discovered that miR-223 is instrumental in alternative macrophage activation and reduction in obesity-associated adipose tissue inflammation (14, 15). Concordant with these observations, our studies revealed that heightened miR-223 levels in KLF6-deficient macrophages dampened proinflammatory gene expression. Furthermore, myeloid-KLF6–deficient mice exhibited diminished adipose tissue inflammation by monocyte-derived macrophages and abrogated diet-induced obesity as well and insulin resistance. Our in vivo analyses revealed that myeloid-KLF6 deficiency attenuated monocyte-derived macrophage levels in adipose tissues following HFD. Indeed, our recent past studies have uncovered that myeloid-KLF6 deficiency results in a modest elevation of circulating monocyte levels (23). This further highlights the importance of myeloid-KLF6 signaling in monocyte-derived macrophage recruitment to the site of inflammation. Importantly, myeloid-KLF6 deficiency did not alter the abundance of adipose tissue-resident macrophage population following HFD. Recent studies have shown that miR-223 may modulate macrophage inflammatory response by repressing NLRP3 inflammasome activity (40). In this scenario, our studies uncovered that elevated miR-223 in KLF6-deficient macrophages attenuated Nlrp3 expression following IFN-γ exposure. More importantly, recent studies uncovered that myeloid-derived miR-223 repress intestinal inflammation by suppressing Nlrp3 expression in innate immune cells (13). In this direction, our previous studies have demonstrated that myeloid-KLF6 deficiency is protective against the pathogenesis of intestinal inflammation in vivo (25). Therefore, it is conceivable that some of these host protection observed in myeloid-KLF6 deficient mice against these inflammatory disease conditions may be due to elevated levels of miR-223 in the myeloid compartment. Collectively, these observations establish that miR-223 utilize a multipronged approach to squash inflammatory gene expression in macrophages.

Monocyte-derived macrophages are a critical component of the innate immune system that plays a crucial role in human health and disease (41). Our previous studies identified KLF6 as most abundantly expressed in myeloid cells and essential for classic macrophage activation (23). Studies by our group and others have shown that KLF6 contributes to a variety of human inflammatory disease conditions (25, 42). Studies by Qi et al. (43) implicated role for KLF6 in the early onset of diabetes utilizing rat model system. More recently, human genome-wide association studies have revealed single nucleotide polymorphisms near the KLF6 gene is associated with obesity and T2D (27). In this direction, our studies uncovered that blockade of myeloid-KLF6 is protective against HFD-induced obesity and insulin resistance. Previous studies have uncovered that loss of myeloid-miR-223 enhances diet-induced obesity and T2D utilizing mouse models (14, 15). In addition, altered miR-223 levels in humans are associated with the advancement of obesity and T2D (16–18). Our current studies have established that KLF6 directly represses miR-223 expression to advance macrophage proinflammatory gene expression and attendant functions. This novel KLF6-miR-223 signaling axis represents a molecular switch in regulating proinflammatory macrophage activation and systemic metabolic dysfunction. Our current studies have identified several inflammatory and metabolic gene targets that are regulated by this novel KLF6-miR-223 signaling axis macrophage. Little is known regarding the molecular mechanisms by which macrophages may lose their quiescence in response to chronic metabolic or inflammatory insults. In this direction, our studies illustrate that many of the stimuli that enhance KLF6 expression attenuate miR-223 levels in macrophages. In addition, deficiency of KLF6 elevated the basal level of miR-223 expression in macrophages. Therefore, this novel KLF6-miR-223 signaling axis serves as a nodal regulator that balances the expression of pro- and anti-inflammatory gene expression in macrophages. Our previous studies have demonstrated that KLF6 facilitates macrophage proinflammatory gene expression by elevating NF-κB functions (23). In this direction, our current studies uncovered that KLF6 deficiency attenuated expression of key positive regulators of NF-κB signaling that was due to heightened miR-223 expression. In summary, our in vitro, ex vivo, and in vivo observations presented here highlight the importance of KLF6-miR-223 signaling axis in macrophage inflammatory gene expression and tissue inflammation. This study provides the initial evidence that KLF6 promotes proinflammatory cytokine and chemokine levels by directly suppressing miR-223 expression in macrophages. Furthermore, inhibition of miR-223 completely reversed attenuated inducible proinflammatory gene expression in KLF6-deficient macrophages. Collectively, our studies discovered a novel myeloid KLF6-miR-223 signaling pathway that regulates inflammatory gene expression and progression of inflammation. Given the importance of KLF6-miR-223 signaling axis in regulation of broad inflammatory disease conditions, the results of our studies will bear implications for numerous pathologic processes, including, but not limited to, inflammatory bowel disease, atherosclerosis, thrombosis, aneurysm, and peripheral artery diseases.

Glossary

BMDM

bone marrow–derived macrophage

ChIP

chromatin immunoprecipitation

HFD

high-fat diet

KLF

Kruppel-like factor

M-CSF

macrophage colony-stimulating factor

MCP1

monocyte chemotactic protein-1

miR-223

miRNA-223

miRNA

microRNA

PM

peritoneal macrophage

qPCR

quantitative PCR

siRNA

small interfering RNA

T2D

type 2 diabetes mellitus

AUTHOR CONTRIBUTIONS

G.-D. Kim, H.-P. Ng, N. Patel, and G. H. Mahabeleshwar performed experiments and analyzed and interpreted the data; and G. H. Mahabeleshwar conceived of and designed the study as well as wrote and edited the manuscript that was approved by all authors.