Specific pathogen-free, C57BL/6J mice were sourced from KBT Oriental Co., Ltd. (Tosu, Japan). All animal experimental procedures were performed in accordance with the protocols and guidelines for animal experiments that were approved by the institutional animal care and use committees of Kagoshima University (Permit Numbers: MD19057). All animals were housed with standard chow¹ and tap water under standard conditions, in a 24°C temperature-controlled environment with a 12-h light/12-h dark cycle. Mice were monitored daily, and were euthanized by cervical dislocation under anesthesia if they showed distress during experiments. All efforts were made to minimize animal suffering.

Bone marrow-derived macrophages (BMDM) were obtained from the femurs and tibias of sacrificed mice (8–10 weeks of age) by flushing with Dulbecco's Modified Eagle Medium (DMEM; Thermo Fisher Scientific, MA, USA) containing 10% fetal bovine serum (FBS; Thermo Fisher Scientific, MA, USA) and 1% Penicillin-Streptomycin (P/S; Thermo Fisher Scientific, MA, USA). Cells were filtered, washed, re-suspended, and cultured in differentiation medium, containing DMEM with 10% FCS, 1% P/S, and 25 ng/mL macrophage colony-stimulating factor (M-CSF; R & D systems, MN, USA). Cells were cultured in differentiation medium for 7 days, following which, M-CSF was depleted. These cells were defined as M0 macrophages (> 95% of population F4/80-positive). For macrophage polarization, cells were seeded in 12-well plates (1 × 10⁶ cells/well), and were treated for 48 h either with 20 ng/mL IFN-γ (PeproTech, NJ, USA) and 10 ng/mL LPS (Sigma-Aldrich, MO, USA) in media (DMEM with 10% FCS, 1% P/S) to induce an M1 phenotype or with 20 ng/mL IL-4 (PeproTech, NJ, USA) to induce an M2 phenotype. Cells were treated for 24 h with recombinant human HGF (E3112; EA Pharma Co., Ltd, Tokyo, Japan) 24 h after polarization to an M1 or M2 phenotype.

After macrophages were polarized to an M1 phenotype, they were cultured in media with or without HGF treatment for 24 h. Cells were also treated for 1 or 24 h either with 20 μM of the PI3K inhibitor LY294002 (Selleck chemicals, Houston, USA) or the treatment vehicle dimethyl sulfoxide (DMSO, Nacalai Tesque Inc., Kyoto, Japan), before being harvested. This inhibitor dose was previously reported to be effective (21).

Total RNA was extracted from cells using the TRIzol reagent (Thermo Fisher Scientific, MA, USA). RNA purity was confirmed by spectrophotometry, and A₂₆₀/A₂₈₀ ratios ranged from 1.9 to 2.1. First-strand cDNA was synthesized from 500 ng of the total RNA using a PrimeScript RT Master Mix (Takara Bio Inc., Shiga, Japan). Real-time PCR was performed using TB Green Premix Ex Taq II (Takara Bio Inc., Shiga, Japan) and the ABI Prism 7700 sequence detection system (Applied Biosystems, Foster City, CA, USA). Data were collected and analyzed using the Step One Plus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). After calculation of the cycle threshold (Ct) value for each sample, relative levels were calculated and normalized to β-actin. PCR conditions were as follows: initial holding period at 95°C for 30 s, followed by 40 cycles of a 2-step program consisting of denaturation at 95°C for 5 s and annealing, and polymerization at 60°C for 34 s. All reactions were performed in duplicate. The following mouse primers were used (f, forward; r, reverse; Takara Bio Inc. Shiga, Japan): β-actin, iNOS, CD86, TNF-α, IL-6, Arg-1, CD206, IL-10, TGF-β1, c-Met, Slc2a1, hexokinase, pyruvate kinase, Slc1a5, glutaminase, glutamate dehydrogenase, CD36, fatty acid synthase, and acyl-CoA synthase (summarized in Supplemental Table 1A). Gene expression values were calculated using the comparative ΔΔCt method and the Ct values were normalized to those of β-actin and are expressed relative to mean level in M0 macrophages not treated with HGF, which is arbitrarily defined as 1.

Cells were lysed in RIPA buffer (Sigma Aldrich, MO, USA) with both a protease inhibitor (Nacalai Tesque Inc., Kyoto, Japan) and a phosphatase inhibitor (Nacalai Tesque Inc., Kyoto, Japan). The same amount of cell lysate sample (10 μL) was used and separated using 4–20% SDS polyacrylamide gel electrophoresis before being electrophoretically transferred to a polyvinylidene fluoride membrane (Bio-Rad Laboratories, Inc., CA, USA). The membranes were blocked for 1 h at room temperature using the SuperBlock™ (PBS) Blocking Buffer (Thermo Fisher Scientific, MA, USA) and were probed by incubating them overnight at 4°C with respective primary antibodies diluted in Solution 1 of Can Get Signal Immunoreaction Enhancer Solution (TOYOBO Co., Ltd., Osaka, Japan). The membranes were then incubated with horseradish peroxidase (HRP)-conjugated anti-mouse antibodies (Cell Signaling Technology, MA, USA). The protein bands were visualized with an enhanced chemiluminescence system (GE Healthcare, IL, USA) and a Fusion Solo S6 (Vilber Lourmat, France). Western blot bands showing phosphorylation of intracellular signaling were quantified by densitometry with Image J software (National Institutes of Health, MD, USA). All antibodies used in this study are listed in Supplemental Table 1B.

The concentration of TNF-α, IL-6, IL-10, and TGF-β1 in cell-free supernatants was measured using Mouse Quantikine TNF-α, IL-6, IL-10, and TGF-β1 ELISA Kits, respectively (R & D systems, MN, USA).

All data are expressed as the mean with a scatterplot showing individual data points. Statistical analysis was performed using either an unpaired Student's t-test or a one-way ANOVA with appropriate post-hoc tests (Graph Pad Prism version 8.4.0). P < 0.05 were considered statistically significant. In vitro experiments were independently performed two or more times.

We first investigated if M0, M1, and M2 macrophages expressed the c-Met. M0 macrophages were induced by treating bone marrow cells with M-CSF for 7 days. Macrophages were then polarized to either an M1 phenotype, by treating with IFN-γ and LPS, or an M2 phenotype via treatment with IL-4. M1 macrophages significantly upregulated mRNA expression of the M1 markers iNOS, CD86, TNF-α, and IL-6. mRNA expression of the M2 markers Arg-1 and CD206 was significantly upregulated in M2 macrophages, but mRNA expression of IL-10 was significantly upregulated in M1 macrophages (Figure 1A). Further, though the spontaneous secretion of TNF-α, IL-6, and IL-10 was significantly increased in the cell culture supernatant of M1 macrophages, TGF-β1 secretion was significantly increased in M2 macrophages (Figure 1B). c-Met expression was significantly upregulated in M1 macrophages. We also found that HGF treatment induced the phosphorylation of c-Met, especially in M1 macrophages (Figure 1C).

We next investigated if HGF-MET signaling was involved in determination of the macrophage phenotype. We had previously shown that the mRNA expression of iNOS, CD86, TNF-α, and IL-6 was upregulated in M1 macrophages and that Arg-1 and CD206 mRNA expression was upregulated in M2 macrophages. For this study, we focused on the phenotypic changes in macrophages with or without HGF treatment. There was no significant difference observed between M0 and M2 macrophages regardless of HGF treatment. However, for M1 macrophages, we observed significant downregulation of the mRNA expression of M1 markers, including iNOS, TNF-α, and IL-6, and the upregulation of Arg-1 with HGF treatment compared to levels in untreated cells (Figure 2A). Further, though HGF treatment significantly increased the secretion of IL-10 and TGF-β1 in the cell culture supernatant of M1 macrophages, no difference was observed for TNF-α and IL-6. TGF-β1 secretion was also significantly increased in both M1 and M2 macrophages that were treated with HGF (Figure 2B).

To identify mechanisms underlying changes in the M1 macrophage phenotype, we examined PI3K activity. We found that HGF treatment induced the phosphorylation of Akt, GSK3β, CREB, and C/EBPβ in M1 macrophages compared to that in untreated cells. We also observed the inactivation of NF-κB (Figure 3A and Supplemental Figure 1A). A PI3K inhibitor, LY294002, was used to verify that the PI3K pathway was crucial for changes in the M1 phenotype. Combinatorial treatment of M1 macrophages with HGF and LY294002 inhibited the phosphorylation of Akt, GSK3β, CREB, and C/EBPβ and enhanced the inactivation of NF-κB compared to that in cells treated with HGF and DMSO (Figure 3B and Supplemental Figures 1B,C). Differences between M1 macrophages that were treated with DMSO and HGF and the cells treated only with DMSO were similar to that previously observed (Figure 2A), with the exception of iNOS and IL-6 mRNA expression (Figure 3C). It is known that DMSO treatment on its own also downregulates the expression of iNOS and IL-6 (22). Therefore, there was a trend toward a decreased expression of iNOS and IL-6, but with no significant difference. In M1 macrophages treated with HGF and LY294002, Arg-1 expression was downregulated, iNOS expression was significantly upregulated, and no change was observed in TNF-α and IL-6 compared to those with HGF and DMSO treatment (Figure 3C). Combinatorial treatment of M1 macrophages with HGF and LY294002 also significantly decreased the secretion of IL-10 but not TGF-β1 when compared to that in cells treated with HGF and DMSO (Figure 3D).

Finally, we studied changes in M1 macrophage intracellular metabolism when treated with HGF to determine if there was a shift toward an M2-like phenotype. We evaluated the mRNA expression of transporters and enzymes involved in glucose, amino acid, and fatty acid metabolism in M1 macrophages. The data for M0 and M2 macrophages was shown as a reference. With HGF treatment, we found that pyruvate kinase, which is involved in glucose metabolism, was significantly downregulated, and Scl1a5, which is involved in amino acid metabolism, was significantly upregulated. With regards to fatty acid metabolism, M1 macrophages treated with HGF significantly upregulated CD36 expression and significantly downregulated fatty acid synthase and acyl-CoA synthase-1 (Figure 4).