Regulation of Monocyte Adhesion and Migration by Nox4

Abstract

We showed that metabolic disorders promote thiol oxidative stress in monocytes, priming monocytes for accelerated chemokine-induced recruitment, and accumulation at sites of vascular injury and the progression of atherosclerosis. The aim of this study was to identify both the source of reactive oxygen species (ROS) responsible for thiol oxidation in primed and dysfunctional monocytes and the molecular mechanisms through which ROS accelerate the migration and recruitment of monocyte-derived macrophages. We found that Nox4, a recently identified NADPH oxidase in monocytes and macrophages, localized to focal adhesions and the actin cytoskeleton, and associated with phospho-FAK, paxillin, and actin, implicating Nox4 in the regulation of monocyte adhesion and migration. We also identified Nox4 as a new, metabolic stress-inducible source of ROS that controls actin S-glutathionylation and turnover in monocytes and macrophages, providing a novel mechanistic link between Nox4-derived H2O2 and monocyte adhesion and migration. Actin associated with Nox4 was S-glutathionylated, and Nox4 association with actin was enhanced in metabolically-stressed monocytes. Metabolic stress induced Nox4 and accelerated monocyte adhesion and chemotaxis in a Nox4-dependent mechanism. In conclusion, our data suggest that monocytic Nox4 is a central regulator of actin dynamics, and induction of Nox4 is the rate-limiting step in metabolic stress-induced monocyte priming and dysfunction associated with accelerated atherosclerosis and the progression of atherosclerotic plaques.

Figure 1. Nox4 localizes to focal adhesions and the F-actin cytoskeleton in mature human monocyte-derived macrophages.

HMDM were fixed, permeabilized and stained as described in “Materials and Methods.” Nuclei were also stained by DAPI (blue). A–C: Nox4 co-localizes with paxillin. HMDM were labeled with antibodies directed against Nox4 (red) and paxillin (green). Colocalization is indicated in yellow (Pearson coefficient  = 0.79). D: Magnification of filed highlighted in panel C with white box in C showing the red (Nox4) and green signals (paxillin) and the overlay. E–G: Nox4 co-localizes with activated FAK. HMDM were labeled with antibodies directed against Nox4 (red) and FAK-Y397Pi (green). Colocalization is indicated in yellow (Pearson coefficient  = 0.75). H–J: Nox4 co-localizes with F-actin. HMDM were labeled with antibodies directed against Nox4 (red) and stained with Alexa Fluor 488-labeled phalloidin (green). Colocalization is indicated in yellow (Pearson coefficient  = 0.84). Scale bar  = 10 µm.

Figure 2. Nox4 co-immunoprecipitates with components of focal adhesions.

Lysates of HMDM were treated with either rabbit monoclonal antibodies directed against Nox4 (IP) or rabbit serum (Control), and antibodies were precipitated with pre-washed Protein G Sepharose as described under “Materials and Methods.” Immunoprecipitates were separated by SDS-PAGE and blots were probed for p22^phox (A), known components of focal adhesions (B) and proteins presumably unrelated to focal adhesions (C). Representative images of at least 3 independent experiments are shown. WCL: whole cell lysate.

Figure 3. Nox4 knockdown prevents increased monocyte adhesion induced by metabolic stress.

(A) THP-1 monocytes were either treated for 24 h with control medium (5 mM glucose, 2% FBS, open bars) or metabolically primed in control medium supplemented with 100 μg/ml native LDL plus 20 mM D-glucose (final concentration 25 mM, MS, closed bars). Cells were then allowed to adhere to fibronectin-coated wells either in the presence or absence of MCP-1 (2 nM) as described in “Materials and Methods.” *: P<0.05 vs. “Control – no MCP-1”; **: P<0.05 vs. “Control + no MCP-1,” n = 4. (B) THP-1 monocytes were transfected for 24 h with either scrambled (siControl) or Nox4-targetting siRNA (siNox4), and subsequently metabolically primed for 24 h where indicated (solid bars). Monocyte adhesion in response to MCP-1 (2 nM) was measured as described in “Materials and Methods”. Insert: Representative Western blot showing suppression of Nox4 protein levels by siRNA. *: P<0.05 vs. Control. **: P<0.05 vs. MS (siControl), n = 4.

Figure 4. Overexpression of Nox4 mimics metabolic priming and promotes monocyte adhesion.

THP-1 monocytes were infected with adenoviruses (Ad, MOI = 50) expressing Nox4 under the control of a doxycycline (Dox)-sensitive promoter. Nox4 expression was induced by adding Dox (1 ug/ml) to the culture medium. Monocyte adhesion in response to MCP-1 (2 nM) was measured as described in “Materials and Methods”. Insert: Representative Western blot showing adenovirus-mediated overexpression of Nox4. *: P<0.05 vs. “Ad”; n = 4.

Figure 5. Nox4 association with actin and S-glutathionylation of Nox4-associated actin are increased metabolically stressed monocytes.

THP-1 monocytes were either treated for 24 h with control medium (5 mM glucose, 2% FBS, Control, open bars) or metabolically primed in control medium supplemented with 100 μg/ml native LDL plus 20 mM D-glucose (final concentration 25 mM, MS, closed bars). Lysates of control cells and metabolically primed THP-1 monocytes were prepared and subjected to immunoprecipitations using rabbit monoclonal antibodies directed against Nox4 (+) or rabbit serum as a control (−). The lysates were pre-cleared by protein G sepharose and co-immunoprecipitation experiments were performed as detailed in “Materials and Methods.” When the co-IP experiments were repeated with protein G beads only (no serum), no actin was observed in the pull down. To assess the association of Nox4 with actin, Western blots were probed with antibodies directed against Nox4 and actin (A+B). To determine whether actin associated with Nox4 was S-glutathionylated, protein separation was conducted under non-reducing conditions, and Western blots were probed with antibodies directed against Nox4 and glutathione (GSH), the latter to identify S-glutathionylated actin (Actin-SSG; C+D). Panels B and D show the densitometric analysis of gels from three independent experiments. *:P<0.05 vs. Control; n = 3.