The new 4-O-methylhonokiol analog GS12021 inhibits inflammation and macrophage chemotaxis: role of AMP-activated protein kinase α activation

Abstract

Preventing pathologic tissue inflammation is key to treating obesity-induced insulin resistance and type 2 diabetes. Previously, we synthesized a series of methylhonokiol analogs and reported that compounds with a carbamate structure had inhibitory function against cyclooxygenase-2 in a cell-free enzyme assay. However, whether these compounds could inhibit the expression of inflammatory genes in macrophages has not been investigated. Here, we found that a new 4-O-methylhonokiol analog, 3',5-diallyl-4'-methoxy-[1,1'-biphenyl]-2-yl morpholine-4-carboxylate (GS12021) inhibited LPS- or TNFα-stimulated inflammation in macrophages and adipocytes, respectively. LPS-induced phosphorylation of nuclear factor-kappa B (NF-κB)/p65 was significantly decreased, whereas NF-κB luciferase activities were slightly inhibited, by GS12021 treatment in RAW 264.7 cells. Either mitogen-activated protein kinase phosphorylation or AP-1 luciferase activity was not altered by GS12021. GS12021 increased the phosphorylation of AMP-activated protein kinase (AMPK) α and the expression of sirtuin (SIRT) 1. Inhibition of mRNA expression of inflammatory genes by GS12021 was abolished in AMPKα1-knockdown cells, but not in SIRT1 knockout cells, demonstrating that GS12021 exerts anti-inflammatory effects through AMPKα activation. The transwell migration assay results showed that GS12021 treatment of macrophages prevented the cell migration promoted by incubation with conditioned medium obtained from adipocytes. GS12021 suppression of p65 phosphorylation and macrophage chemotaxis were preserved in AMPKα1-knockdown cells, indicating AMPK is not required for these functions of GS12021. Identification of this novel methylhonokiol analog could enable studies of the structure-activity relationship of this class of compounds and further evaluation of its in vivo potential for the treatment of insulin-resistant states and other chronic inflammatory diseases.

Fig 1. Chemical structures of honokiol, 4-O-methylhonokiol, and their derivatives.

Fig 2. Effects of treatment with a honokiol analog on cell viability, expression of iNOS and COX-2, and production of NO in RAW 264.7 cells.

(A) RAW 264.7 cells were incubated with each compound (20 μM or 40 μM) in the presence of serum for 24 h. Cell viability was determined by the MTT assay. (B)-(C) Cells were pretreated with 20 μM of compounds 1 h before LPS treatment (10 ng/mL) for 24 h. The protein level was measured by western blotting. The experiments were repeated at least three times and representative blots (B) and quantification data (C) are shown. β-tubulin was used as a loading control. (D) Nitrite content. Nitrite levels were measured by spectrophotometric means with Griess reagent using supernatant media obtained from the cells in (B). HK, honokiol; GS, GS12021; c1–c9, arbitrary names for the synthetic honokiol analogs. Data are provided as mean ± SEM values. *P < 0.05, **P < 0.01 versus CON; ^# P < 0.05, ^## P < 0.01 versus LPS alone treatment.

Fig 3. Anti-inflammatory effect of GS12021 in RAW 264.7 cells.

(A) Concentration-dependent effect of GS12021 on expression of iNOS and COX-2. Representative western blot images and quantification data are shown. Cells were pretreated with the indicated concentrations of GS12021 1 h before LPS treatment (10 ng/mL) for 24 h. (B) Nitrite production in cells from (A). (C) mRNA expression of inflammatory genes/cytokines. Cells were treated with agents for 6 h and RT-qPCR was performed. (D) Cytokine levels in supernatant media from cells treated with LPS ± GS12021 for 6 h (TNFα) or 24 h (IL-6 and MCP-1). All the experiments were repeated at least three times and the data are provided as mean ± SEM values. ** P < 0.01 versus CON; ^# P < 0.05, ^## P <0.01 versus LPS treatment alone.

Fig 4. Macrophage chemotaxis was markedly inhibited by GS12021 treatment.

Macrophage migration assays using DMEM (migration media) or conditioned medium (CM) collected from adipocytes were conducted. RAW 264.7 cells were incubated with DMSO or GS12021 at 20 μM for 3 h and detached cells used for the migration assay in the presence of DMEM or CM in the lower well. Cells placed in transwells were incubated for 3 h for migration, and migrated cells found on the lower parts of transwells were counted after staining with crystal violet. The experiments were repeated three times and the representative microscope images for migrated cells are shown below at ×100 magnification. **P < 0.01 versus DMEM; ^## P < 0.01 versus CM.

Fig 5. Effect of GS12021 treatment on LPS-stimulated cell signaling in RAW 264.7 cells.

Immunoblot analyses of phosphorylated IKK, IκBα and phosphorylated- and total-p65. The experiments were repeated at least three times, and the representative images and the densitometry results for phosphorylated p65 are shown in (A). Phosphorylations of JNK, ERK, p38 MAPK, Akt, c-Jun and S6 protein (B). Cells were pretreated with vehicle or GS12021 (20 μM for 1 h) and stimulated with LPS for the time periods indicated. (C) NF-κB reporter assay (N = 4). (D) AP-1 reporter assay (N = 4). RAW 264.7 cells were transfected with NF-κB or AP-1 luciferase reporter plasmid and, after 24 h, cells were incubated with GS12021 (20 μM or 40 μM) for a further 24 h. **P < 0.01 versus CON; ^## P < 0.05 versus LPS alone.

Fig 6. Effect of GS12021 treatment on AMPKα activation and SIRT1 expression in RAW 264.7 cells.

(A) Levels of phosphorylated AMPKα and total AMPKα were measured by western blotting. Cells were pretreated with vehicle or GS12021 (20 μM for 1 h) and stimulated with LPS for the time periods indicated. Quantification results are shown in the right panel. *P < 0.05 versus CON; ^# P < 0.05, ^## P <0.01 versus LPS alone. (B) Cells were treated with GS12021 (20 μM) for the indicated time periods and western blotting was performed with specific antibodies against p-ACC, p-AMPKα, and SIRT1. β-tubulin was used as a loading control. (C) Concentration-dependent effect of GS12021 on AMPK activation and SIRT1 expression. Cells were treated with different concentrations of GS12021 for 12 h, and western blotting analyses were performed.

Fig 7. Anti-inflammatory effect, but not anti-chemotaxis effect, of GS12021 was attenuated in AMPKα1 knockdown cells.

RAW 264.7 cells were transfected with siRNA of control (CON) or AMPKα1 and incubated for 24 h. Cells were subsequently treated with LPS or GS12021+LPS and mRNA or protein levels were determined. (A) AMPKα expression after siRNA transfection. (B) mRNA levels of iNOS, TNFα, and MCP-1/CCL2 in AMPKα1 knockdown macrophages. **P < 0.01 versus vehicle; ^# P < 0.05, ^## P < 0.01 versus LPS alone. (C)-(D) The effect of AMPKα1 knockdown on the GS12021 suppression of p-p65 and the adipocyte CM-mediated macrophage chemotaxis. Representative microphotographs of cell migration are shown above the quantification results (N = 3). AU means arbitrary units. **P < 0.01 versus vehicle.

Fig 8. Anti-inflammatory effect of GS12021 was preserved in SIRT1 knockout (KO) macrophages.

(A) SIRT1 expression in peritoneal macrophages obtained from wild-type (WT) or myeloid-specific SIRT1 KO mice. Peritoneal macrophages were isolated from WT or SIRT1 KO mice as described in the Materials and Methods section. (B) mRNA levels of inflammatory genes measured by RT-qPCR in macrophages from WT or SIRT1 KO mice. Cells were pretreated with GS12021 for 1 h, and then treated with LPS (10 ng/mL) for 6 h. ** P < 0.01 versus CON, # P<0.05 ## P<0.01 versus LPS alone.

Fig 9. The effect of GS12021 treatment on the adipocyte inflammation.

Fully differentiated 3T3-L1 adipocytes were treated with TNFα (10 ng/ml) for 6 h in the presence or absence of GS12021 (20 μM). mRNA levels of iNOS, MCP-1/CCL2, IL-1β and IL-6 were measured by RT-qPCR (N = 3). **P < 0.01 versus CON; ^# P < 0.05, ^## P < 0.01 versus TNFα alone.

Fig 10. Graphical abstract shows that AMPK is required for GS12021 inhibition of inflammation by but not for its inhibition of chemotaxis in macrophages.