MEK1/2 Inhibition Promotes Macrophage Reparative Properties

Abstract

Macrophages have important functional roles in regulating the timely promotion and resolution of inflammation. Although many of the intracellular signaling pathways involved in the proinflammatory responses of macrophages are well characterized, the components that regulate macrophage reparative properties are less well understood. We identified the MEK1/2 pathway as a key regulator of macrophage reparative properties. Pharmacological inhibition of the MEK1/2 pathway by a MEK1/2 inhibitor (MEKi) significantly increased expression of IL-4/IL-13 (M2)-responsive genes in murine bone marrow-derived and alveolar macrophages. Deletion of the MEK1 gene using LysM^Cre+/+Mek1^fl/fl macrophages as an alternate approach yielded similar results. MEKi enhanced STAT6 phosphorylation, and MEKi-induced changes in M2 polarization were dependent on STAT6. In addition, MEKi treatment significantly increased murine and human macrophage efferocytosis of apoptotic cells, independent of macrophage polarization and STAT6. These phenotypes were associated with increased gene and protein expression of Mertk, Tyro3, and Abca1, three proteins that promote macrophage efferocytosis. We also studied the effects of MEKi on in vivo macrophage efferocytosis and polarization. MEKi-treated mice had increased efferocytosis of apoptotic polymorphonuclear leukocytes instilled into the peritoneum. Furthermore, administration of MEKi after LPS-induced lung injury led to improved recovery of weight, fewer neutrophils in the alveolar compartment, and greater macrophage M2 polarization. Collectively, these results show that MEK1/2 inhibition is capable of promoting the reparative properties of murine and human macrophages. These studies suggest that the MEK1/2 pathway may be a therapeutic target to promote the resolution of inflammation via modulation of macrophage functions.

FIGURE 1

BMDM M2 gene expression is increased during IL-4/IL-13 polarization by the MEK1/2 inhibitor PD0325901. (A) One of three representative experiments showing constitutive expression of pERK1/2 in resting (M0) and IL-4/IL-13 treated conditions in murine BMDM. There was early reduction of pERK1/2 at 15-60 minutes post-stimulation with MEKi (PD0325901). (B) One of three representative experiments showing constitutive expression of pERK1/2 in resting (M0) and IL-4 treated human MDM. Western blots of protein lysates from 0 and 48 hours show decreased pERK1/2 after PD0325901 treatment. (C) M0 and IL-4/IL-13-treated BMDM exposed to carrier control or MEKi over 48 h were processed for qRT-PCR to determine relative quantification (RQ) of Retnla, Ym1, Ccl17, Tgfb1, and Arg1 mRNA normalized to time-matched M0 + carrier control samples. Treatment with MEKi led to a significant increase in IL-4/IL-13 dependent gene expression of Retnla, Ym1, Ccl17, Tgfb1, but not Arg1. Data from 3-4 biological replicates (mean±SEM). (D) At 48 h, IL-4/IL-13 treated cells exposed to carrier control or MEKi were processed for surface staining of CD71 and CD206. Change in mean fluorescent intensity (ΔMFI) was determined by subtracting the MFI of the isotype control samples from that of each antibody. MEKi treatment led to increased surface expression of both CD71 and CD206. Data mean ± SD are from triplicate samples from one representative experiment of two. (E) LPS-treated BMDM were treated with IL-4/IL-13 + carrier or PD0325901. At 48 h, MEKi treatment led to increased expression of Retnla, Ym1, and Tgfb1. Data are mean ± SD from 6 independent experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

FIGURE 2

Efficacy of MEK1/2 pathway inhibition results in differential enhancement of BMDM IL-4/IL-13 polarization. BMDM were stimulated with IL-4/IL-13 with the addition of media (control), DMSO (+ carrier), PD0325901, U0126, or PD98059 for 48 hours. Cells were processed for RNA and protein. (A) Relative quantification (RQ) of Retnla gene expression normalized to M0 showing the greatest effect of PD0325901 (mean ± SD of 4 biological replicates). (B) BMDM protein lysates were probed for ERK1/2 phosphorylation (pERK1/2), ERK1/2, Relmα, and β-actin and quantified using densitometry. (C) Relative quantification of the ratio of pERK1/2:ERK1/2 expressed as percent of IL-4/IL-13 + carrier controls showing the greatest effect of PD0325901. (D) Ratio of Relmα:β-Actin also demonstrating the greatest effect of PD0325901. (E) BMDM protein lysates from LysM^Cre+/+MEK1^fl/fl BMDM (LysM^Cre+/+) compared to LysM^Cre−/−MEK1^fl/fl controls (LysM^Cre−/−) demonstrating a reduction in MEK1 protein from LysM^Cre+/+ cells. Lane 1: M0 + Carrier, Lane 2: IL-4/IL-13 + Carrier, Lane 3: IL-4/IL-13 + MEKi. (F) Relative quantification of Retnla gene expression from IL-4/IL-13 treated LysM^Cre+/+ and LysM^Cre−/− BMDM normalized to M0 (mean ± SD of triplicate samples). Data are from one of two representative experiments. (G) BAL alveolar macrophages were IL-4/IL-13 polarized with the addition of either DMSO (+ carrier) or PD0325901 (+MEKi) ex vivo for 48 hours. qRT-PCR was used to measure the relative quantification (RQ) of Retnla, Ym1, Tgfb1, and Ccl17. Data are ± SD of the fold increase in PD0325901 compared to carrier. (n=7 biological replicates for each group collected from 2 independent experiments.) Statistical comparisons are versus carrier-treated samples or as indicated. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

FIGURE 3

MEKi increases BMDM STAT6 pathway activation during IL-4/IL-13 stimulation. (A) BMDM protein lysates collected at 1, 4, 24, and 48 hours after stimulation of M0 + Carrier and IL-4/IL-13 + Carrier or + MEKi. Blots were probed for pSTAT6, STAT6, SOCS1, SOCS3, and β-Actin one representative experiment of n=4-6. (B) Densitometry quantitation of the ratio of pSTAT6/STAT6, SOCS1/Actin, and SOCS3/Actin normalized to carrier-treated samples demonstrating increased pSTAT6 and reduced SOCS1 and SOCS3 proteins in MEKi treated samples. (C, D) BMDM from wild-type (WT) or Stat6^−/− Balb/c mice were stimulated with IL-4/IL-13 with the addition of DMSO (+ Carrier) or 0.5 μM PD0325901 (+MEKi). (C) At 48 h, Retnla, and Ym1 mRNA expression was measured and expressed as fold change (RQ) relative to respective M0 conditions. There was marked reduction in both Retnla and Ym1 in Stat6−/− compared to WT cells. Data are from 3-5 biological replicates and show the mean for each sample comparing matched carrier and inhibitor treated samples. (D) Protein lysates were collected at serial time points after stimulation. Relmα, STAT6 and β-Actin were detected by western blot. There was no detectable STAT6 or Relmα proteins in Stat6−/− cells compared to that of WT. Blots are from one representative experiment of three. (E) BMDM were stimulated with IL-4/ IL-13 with the addition of carrier or MEKi. RNA was collected over 48 h to determine the relative quantification (RQ) of PU.1 normalized to time-matched M0 conditions. MEKi treatment led to increases in PU.1 mRNA starting at 12 h (mean of 3-4 independent experiments). *P<0.05, **P <0.01.

FIGURE 4

MEKi increases macrophage efferocytosis of apoptotic cells. Murine BMDM (A) or alveolar macrophage (AM) (B) were stimulated with IL-4/IL-13 or BMDM cultured as M0 (C) with the addition of DMSO (+ Carrier) or 0.5 μM PD0325901 (+MEKi) for 48 hours. Unlabeled and CFSE-labeled, apoptotic human neutrophils (PMN) were added to BMDM cultures. Efferocytosis was quantified as the percentage of CSFE⁺ macrophages using FACS. In all MEKi treated samples, efferocytosis was significantly enhanced. Data are mean percent BMDM CFSE⁺ (A,C) (n=3) or (B) AM CFSE⁺ (n=6 each) from paired carrier and MEKi-treated samples. (D) M0 BMDM treated with either carrier or MEKi for 6 or 24 hours were incubated with Jurkat as AC target, and efferocytosis evaluated by flow cytometry. MEKi increased efferocytosis in M0 cells at 24 h but not at 6 h. Data are the mean ± SD from triplicate samples from one representative experiment of three. (E) Apoptotic PMNs were added to human MDM (M0 or IL-4 treated) with either carrier or MEKi. Efferocytosis was enhanced by MEKi across all conditions. Data are mean ± SD of the fold increase in the percentage of MDM CFSE⁺ from 4 different donors. (F,G) Representative contour plots from a single human donor showing CD14⁺ MDM after efferocytosis of either unlabeled PMN (+ Unl PMN) or CFSE-labeled PMN (+ CFSE PMN) from (F) resting (M0) or (G) IL-4 (M2) stimulated with either carrier or MEKi. *P<0.05, **P<0.01, ***P<0.001.

FIGURE 5

MEKi increases macrophage MertK, Tyro3, and Abca1 independent of polarization and STAT6. (A) BMDM from WT mice were left resting (M0) or polarized with IL-4 /IL-13 for 24 hours with the addition of either DMSO (+ Carrier) or 0.5 uM PD0325901 (+MEKi). (B) BMDM from Wild-type (WT) or Stat6^−/− mice or (C) murine alveolar macrophages were IL-4/IL-13 polarized for 48 hours with the addition of either vehicle or MEKi (PD0325901). RNA was collected to determine relative expression (RQ) of Mertk and Abca1 compared to M0 + carrier control. Mertk and Abca1 were increased in MEKi-treated M0 and M2 polarized WT and Stat6−/− BMDM, and in WT alveolar macrophages. Data are mean ± SD of 3-7 biological replicates. (D) Representative contour plots showing IL-4/IL-13 BMDMs polarized for 48 hours with the addition of either carrier or MEKi. Gates were set based on Mertk isotype control staining to evaluate the percent of macrophages that are Mertk⁺. (E,F) Quantitation of FACS data showing a MEKi-dependent increase in % BMDM Mertk⁺, ΔMFI of Mertk⁺, Abca1 and Tyro3. Data are ± SEM of 2-3 biological replicates. *P<0.05, **P <0.01, ***P <0.001.

FIGURE 6

MEKi promotes in vivo macrophage efferocytosis and M2 polarization. (A) Mice received either carrier or MEKi 24 h prior to intraperitoneal delivery of apoptotic neutrophils. Additional control mice did not receive PMNs. Mice were subjected to peritoneal lavage and cytospin preparations were made from recovered cells and stained with Diff-Quick. The percent of macrophages with ingested cells was quantified and data from three independent experiments are shown. MEKi treatment led to a significant increased in efferocytosis. (B-E) In a separate in vivo model of lung injury, mice received oropharyngeal delivery of LPS on day 0 and MEKi on days 1 and 3 post-LPS. Mice were monitored for weight change (B) and euthanized on days 2 and 4 for assessment of BAL cell counts and differential (C,D). On day 4, alveolar macrophages were isolated for assessment of M2 gene and protein expression (E). (B) Initial weight loss was similar between MEKi and carrier control groups, and MEKi treated mice had faster recovery of their weights starting at day 2. Day 2 (n=26-27/condition), Days 3-4 (n=16/condition). (C,D) On days 2 and 4, there were fewer BAL total cells due to reduced numbers of neutrophils in MEKi treated mice. Macrophage numbers were similar in both groups. (n=5-6 mice/group; 2-3 experimental replicates). (E) On day 4, alveolar macrophages from MEKi treated mice had greater expression of Ym1 and Ccl17 mRNA (normalized to carrier control) and greater expression of CD71 as measured by FACS. (n=10-12/group). *P<0.05, **P<0.01, ***P<0.001, ***P<0.0001.