Vimentin regulates mitochondrial ROS production and inflammatory responses of neutrophils

Abstract

The intermediate filament vimentin is present in immune cells and is implicated in proinflammatory immune responses. Whether and how it supports antimicrobial activities of neutrophils are not well established. Here, we developed an immortalized neutrophil model to examine the requirement of vimentin. We demonstrate that vimentin restricts the production of proinflammatory cytokines and reactive oxygen species (ROS), but enhances phagocytosis and swarming. We observe that vimentin is dispensable for neutrophil extracellular trap (NET) formation, degranulation, and inflammasome activation. Moreover, gene expression analysis demonstrated that the presence of vimentin was associated with changes in expression of multiple genes required for mitochondrial function and ROS overproduction. Treatment of wild-type cells with rotenone, an inhibitor for complex I of the electron transport chain, increases the ROS levels. Likewise, treatment with mitoTEMPO, a SOD mimetic, rescues the ROS production in cells lacking vimentin. Together, these data show vimentin regulates neutrophil antimicrobial functions and alters ROS levels through regulation of mitochondrial activity.

Keywords: ROS; antimicrobial; inflammatory responses; mitochondria; neutrophils; vimentin.

Figure 1

HoxB8-transformed wild-type and Vim^-/- GMP differentiated into neutrophils. (A) Representative image of western blot showing vimentin expression. (B) Quantification of the western blots represented in panel A (C) Ly6G and CD11b expression for neutrophil progenitor cells and differentiated neutrophils. (D) Quantification of the percentage of Ly6G positivity. (E) Representative images of the morphology of the nuclei of Vim^+/+ and Vim^-/- GMP and neutrophils. Scale bar: 5 μm. (F) Quantification of percentage nuclei that are lobular. (B, D, F) Dots are independent experiments using 5 Vim^+/+ and 6 Vim^-/- cell lines. Data are mean ± SEM. *p < 0.05; ns, not significant; paired t-test (B) or one-way ANOVA with Sidak’s post hoc test (D, F).

Figure 2

Functional analysis of neutrophils with and without vimentin. (A) Representative images of Vim^+/+ and Vim^-/- neutrophils swarming to C. albicans. Cyan, neutrophils; magenta, C. albicans. Scale bar: 10 μm. (B) Quantification of neutrophil swarming efficiency by intensity. Swarming of Vim^+/+ neutrophils is 100%. (C) Representative images of phagocytosis of E. coli by Vim^+/+ and Vim^-/- neutrophils. blue, DNA; cyan, all E. coli; magenta, extracellular E. coli. Scale bar: 5 μm. (D) Quantification of number of intracellular bacteria per cell in Vim^+/+ and Vim^-/- neutrophils. Extracellular bacteria were defined as bacteria that stain with an anti-E. coli antibody, and intracellular bacteria are the GFP positive bacteria that do not stain with the anti-E. coli antibody. For each experiment, the average number of intracellular bacteria per cell was determined in at least 75 cells for Vim^+/+ and Vim^-/- neutrophils that contained bacteria. Dots are the mean intracellular bacteria per cell. (E, F) Quantification of TNFα (E) or IL-6 (F) produced by Vim^+/+ and Vim^-/- neutrophils in response to LPS stimulation at 6 hours. (G) Quantification of IL-1β produced by Vim^+/+ and Vim^-/- neutrophils in response to treatment with LPS alone or LPS and nigericin at 2 hours. (H) Representative western blot for pro-caspase-1, p12, and p10 after LPS and nigericin (to induce inflammasome activation) stimulation of Vim^+/+ and Vim^-/- neutrophils. (I, J) Quantification of pro-caspase-1 (I) and p12/p10 levels (J). (B, D, E–G, I, J) Data mean ± SEM. Dots are independent experiments. ns, not significant; *p <0.05; **p<0.01; ***p<0.001 by two-way ANOVA with Sidak’s post hoc test (B, E–G) and paired t-test (D–J).

Figure 3

Vimentin regulates mitochondrial ROS production. (A, B) ROS production from neutrophils was measured by cytochrome c reactivity in unstimulated cells (A) or PMA-treated cells (B). Blue, Vim^+/+ ; gray, Vim^-/- neutrophils. (C) Ingenuity Pathway Analysis showing multiple genes in electron transport chain complexes are down-regulated in Vim^-/- neutrophils. Pink diamond, downregulated genes in Vim^-/- as compared to Vim^+/+ neutrophils. (D) Vimentin reduces mitochondria and mitochondrial-derived ROS. Representative images. Dotted lines are cells boundaries. Cyan, mitotracker; magenta, mitosox. Scale bar: 5 μm. (E) Mitosox intensity per cell. (F) Mitotracker intensity per cell. (G) Inhibitor of complex I of the electron transport chain, rotenone, increases ROS production. (H) Quantification of total superoxide production of (G). (I) Addition of mitoTEMPO, a SOD mimetic, reduces ROS production. Dots are independent experiments, and data are mean ± SEM. ns, not significant; *p <0.05; **p<0.01, ***p-value<0.001. Significance was determined by paired t-test (E, F) or one-way ANOVA with Sidak’s post hoc test (H).

Figure 4

Schematic showing vimentin function in neutrophils. Vimentin supports mitochondrial function. In the absence of vimentin, mitochondria become dysregulated and produce more ROS. Vimentin loss also leads to decreased phagocytosis and swarming and increased inflammasome activation and proinflammatory cytokine production.