Activation of AMPK enhances neutrophil chemotaxis and bacterial killing

Abstract

An inability of neutrophils to eliminate invading microorganisms is frequently associated with severe infection and may contribute to the high mortality rates associated with sepsis. In the present studies, we examined whether metformin and other 5' adenosine monophosphate-activated protein kinase (AMPK) activators affect neutrophil motility, phagocytosis and bacterial killing. We found that activation of AMPK enhanced neutrophil chemotaxis in vitro and in vivo, and also counteracted the inhibition of chemotaxis induced by exposure of neutrophils to lipopolysaccharide (LPS). In contrast, small interfering RNA (siRNA)-mediated knockdown of AMPKα1 or blockade of AMPK activation through treatment of neutrophils with the AMPK inhibitor compound C diminished neutrophil chemotaxis. In addition to their effects on chemotaxis, treatment of neutrophils with metformin or aminoimidazole carboxamide ribonucleotide (AICAR) improved phagocytosis and bacterial killing, including more efficient eradication of bacteria in a mouse model of peritonitis-induced sepsis. Immunocytochemistry showed that, in contrast to LPS, metformin or AICAR induced robust actin polymerization and distinct formation of neutrophil leading edges. Although LPS diminished AMPK phosphorylation, metformin or AICAR was able to partially decrease the effects of LPS/toll-like receptor 4 (TLR4) engagement on downstream signaling events, particularly LPS-induced IκBα degradation. The IκB kinase (IKK) inhibitor PS-1145 diminished IκBα degradation and also prevented LPS-induced inhibition of chemotaxis. These results suggest that AMPK activation with clinically approved agents, such as metformin, may facilitate bacterial eradication in sepsis and other inflammatory conditions associated with inhibition of neutrophil activation and chemotaxis.

Figure 1

Effects of metformin or AMPK inhibitor compound C on neutrophil chemotaxis. (A, B, C) Representative images (A) and quantitative data (B, C) show neutrophil migration after dose- and time-dependent stimulation with W-peptide. In (B), transmigration assay was performed by inclusion of W-peptide (0, 12.5, 25 or 50 nmol/L) for 60 min, whereas, in (C), migration was determined after inclusion of 50 nmol/L W-peptide for the indicated time period. Means ± SD (n = 3), ***P < 0.001, compared with untreated neutrophils. (D) Representative Western Blots and quantitative data show the amount of phosphorylated and total AMPK in neutrophils treated with W-peptide (50 nmol/L) for 0, 20, 40 or 60 min. Average (mean ± SEM) of optical bend density was obtained from three independent experiments (*P < 0.05, compared with untreated). (E, F) HL-60 cells were treated with IL-8 for 60 min followed by chemotaxis assay (E) and Western Blot analysis (F) for phospho and total AMPK. Means ± SD (n = 3), *P < 0.05, **P < 0.01, ***P < 0.001, compared with untreated cells.

Figure 2

Effects of AMPK activation or inhibition on neutrophil chemotaxis. Neutrophils were pretreated with compound C (0, 1, 2, 3 or 10 μmol/L) for 60 min, metformin (500 μmol/L) or berberine (10 μmol/L) for 2.5 h and then chemotaxis was measured after inclusion of cells in transmigration chambers and exposure to W-peptide for 60 min. (A) Representative images show amount of control (untreated) or compound C–treated (30 μmol/L) neutrophils that migrated into the lower reservoir. Panel (B) shows dose-dependent inhibition by compound C, whereas (C) metformin exposure or berberine exposure enhanced neutrophil chemotaxis. Means ± SD (n = 3), *P < 0.05, compared with control. Lower panels B or C show Western blots of phospho AMPK, total AMPK and actin obtained from neutrophils treated with compound C (10 μmol/L) for 60 min, metformin (500 μmol/L) for 2.5 h, or berberine (10 μmol/L) for 2.5 h. (D) Chemotaxis assays were performed in control (untreated) cells and cells treated with control (scrambled siRNA) or specific siRNA to AMPKα1 subunit. (E) Representative Western blots show amounts of AMPKα1 and actin before and after treatment with scrambled or siRNA to AMPKα1. ctr, Control; met, metformin; ber, berberine; scr., scrambled.

Figure 3

Exposure to LPS diminished neutrophil chemotaxis and AMPK phosphorylation. (A) Representative images show neutrophil transmigration before and after treatment with LPS. Neutrophils were treated with LPS (0 or 1 μg/mL) for 90 min. The cells were then washed and chemotaxis examined over a 60 min period. In (B), neutrophils were incubated with LPS for 90 min followed by measurement of neutrophil chemotaxis. (C, D) Representative Western blots and quantitative data show the amounts of phosphorylated and total AMPK obtained from neutrophils treated with LPS (0–1,000 ng/mL) for 90 min (C) or after exposure to LPS (300 ng/mL) for the indicated time periods (D). Means ± SEM (n = 3), *P < 0.05. (E, F) LPS suppresses activation of AMPK. Representative Western Blots and quantitative data show amounts of phospho and total AMPK, and actin. Neutrophils were treated with AICAR (0 or 250 μmol/L) for 60 min and then cultured with LPS (0 or 300 ng/mL) for an additional 60 min. Cells also were treated with LPS (300 ng/mL) for 60 min followed by inclusion of AICAR (250 μmol/L) in the cultures for an additional 60 min. Means ± SD, n = 3, *P < 0.05, compared with neutrophils treated with LPS alone. ctr, Control; AIC, AICAR.

Figure 4

Metformin diminished LPS-induced inhibition of neutrophil chemotaxis. (A) Neutrophils were treated with metformin (met; 0 or 500 μmol/L) for 90 min and then LPS (0 or 300 ng/mL) for 60 min, or cells were first treated with LPS (300 ng/mL) for 60 min followed by inclusion of metformin (500 μmol/L) in the cultures for an additional 90 min. Neutrophil chemotaxis was then examined. Means ± SD, n = 3, *P < 0.05, **P < 0.01 compared with neutrophils treated with LPS alone. (B) Representative images show direction and distance (length of arrows) passed by the neutrophils a pretreated with metformin, LPS, compound C or combination of metformin and LPS. Panel (C) shows neutrophil velocity. Mean ± SEM, n ≥ 20, *P < 0.05, **P < 0.01 compared with control, ^‡P < 0.001 compared with metformin and ^§P < 0.001 compared with LPS alone. Large arrow indicates direction to W-peptide. (D, E). Metformin stimulates, whereas compound C diminishes neutrophil chemotaxis in vivo. Panel (D) shows time line administration of metformin (125 mg/kg; IP), compound C (3 mg/kg; IP), PBS (200 μL; IP) or W-peptide (0.43 mg/kg; IP) followed by acquisition of peritoneal neutrophils. Panel (E) shows amount of peritoneal neutrophils obtained from mice treated as indicated in (D). Mean ± SEM (n ≥ 3). *P < 0.05, compared control or mice treated with compound C. ctr, Control; met or Met., metformin; W-pep, W-peptide; com. C, compound C.

Figure 5

Metformin increases bacteria uptake and killing in vitro and in vivo. (A, B). Neutrophils (10⁶ cells/mL) were pretreated with or without metformin (500 μmol/L) for 2.5 h and then incubated with fluorescently tagged E. coli (10⁷/mL) (A) or S. aureus (10⁷/mL) (B) for additional 20 min. Neutrophil-dependent uptake of bacteria was determined using flow cytometry. Means ± SEM (n = 3), *P < 0.05 or **P < 0.01. (C, D) Neutrophils (2 × 10⁶ cells/mL) were pretreated with metformin (0 or 500 μmol/L) for 2.5 h, AICAR (0 or 250 μmol/L) for 2 h, LPS (1 μg/mL), compound C (10 μmol/L) or rapamycin (10 nmol/L) for 60 min. Next, neutrophils were incubated with E. coli (2 × 10⁷/mL) for 90 min. Images (C) show agar plates with colonies formed by viable E. coli that were obtained after incubating bacteria with neutrophils. Panel (D) shows the number of viable bacteria recovered from killing assay. Means ± SEM (n = 4), *P < 0.05 compared with control. (E, F, G) Mice were subjected to the administration of metformin (125 mg/kg; IP) for 12 h, compound C (3 mg/kg; IP), or vehicle (0.9% saline) for 2 h followed by IP injection of E. coli (2 × 10⁸) for an additional 6 h. (E) Representative images showed the number of E. coli colonies formed from viable bacteria that were recovered from peritoneal lavages. (F, G) Average of viable bacteria (F) and neutrophils (G) obtained from peritoneal lavages (mean ± SEM, n ≥ 5). *P < 0.05, **P < 0.01. RFU, relative fluorescence unit; ctr, control; met, metformin; CFU, colony-forming unit; com. C, compound C.

Figure 6

AMPK activation induces neutrophil cytoskeletal rearrangement and leading edge formation. Neutrophils were pretreated with AICAR (0 or 250 μmol/L) for 90 min or metformin (0 or 500 μmol/L) for 2.5 h followed by inclusion of LPS (0 or 1 μg/mL) for additional 90 min. (A,B) Representative images show actin (green) and nuclei (blue) staining (A), whereas (B) shows magnified region of interest. (C) Average of actin fluorescence is shown. Mean ± SD actin fluorescence intensity (n < 3 ~ 5), *P < 0.05.

Figure 7

Effects of metformin, AICAR and IKK inhibitor on LPS-dependent activation of TLR4 or mTOR signaling pathways. Neutrophils were cultured with AICAR (0 or 500 μmol/L) for 2.5 h, metformin (0 or 500 μmol/L) for 2.5 h, IKK inhibitor PS-1145 (0 or 10 μmol/L) for 60 min or rapamycin (0 or 30 nmol/L) for 30 min. Next, neutrophils were cultured with LPS (0 or 300 ng/mL) for an additional 60 min. (A, C, F, G) Representative Western blots show amount of IκBα, total and phosphorylated ribosomal protein S6 (rpS6) (Ser^240/244), 4E-binding protein 1 (4E-BP1), and actin. Panels (B, D, F and H) show average of Western blots optical bend densitometry. Means ± SEM (n = 3); (B, D) *P < 0.05, **P < 0.01 compared with LPS only; (F) **P < 0.01 compared with control (untreated); (H) **P < 0.01 compared with control (untreated), LPS or rapamycin- and LPS-treated cells. Panel (E) shows the ability of IKK inhibitor PS-1145 to prevent LPS-mediated inhibition of neutrophil chemotaxis.