Subcellular localization of Nox4 and regulation in diabetes

Abstract

Oxidative stress is implicated in human diseases. Some of the oxidative pathways are harbored in the mitochondria. NAD(P)H oxidases have been identified not only in phagocytic but also in somatic cells. Nox4 is the most ubiquitous of these oxidases and is a major source of reactive oxygen species (ROS) in many cell types and in kidney tissue of diabetic animals. We generated specific Nox4 antibodies, and found that Nox4 localizes to mitochondria. (i) Immunoblot analysis in cultured mesangial cells and kidney cortex revealed that Nox4 is present in crude mitochondria, in mitochondria-enriched heavy fractions, and in purified mitochondria; (ii) immunofluorescence confocal microscopy also revealed that Nox4 localizes with the mitochondrial marker Mitotracker; and (iii) the mitochondrial localization prediction program MitoProt indicated that the probability score for Nox4 is identical to mitochondrial protein cytochrome c oxidase subunit IV. We also show that in purified mitochondria, siRNA-mediated knockdown of Nox4 significantly reduces NADPH oxidase activity in pure mitochondria and blocks glucose-induced mitochondrial superoxide generation. In a rat model of diabetes, mitochondrial Nox4 expression is increased in kidney cortex. Our data provide evidence that a functional Nox4 is present and regulated in mitochondria, indicating the existence of a previously undescribed source of ROS in this organelle.

Fig. 1.

Subcellular localization of Nox4 in rat kidney cortex. (A) Nox4 protein or prohibitin was detected by Western blot analysis in crude subcellular fractions isolated from rat kidney cortex, Total (Tot), Mit, C, or Mb. (B) Immunoblotting of subcellular Mb fractions of rat kidney cortex with Nox4 antibody. Prohibitin was used as a marker for mitochondria. (C) Electron microscopy of freshly isolated pure mitochondria from kidney cortex. (Scale bar, 2 μm.) (D) Immunoblotting of pure Mit (Mit) and total cortical homogenate (Tot) with Nox4 antibody, prohibitin, Na⁺/K⁺-ATPase, calnexin, ARF, or LAMP-1. (E) Western blot analysis of homogenates from Tot and Percoll-purified Mit fractions increasing protein concentrations by using our Nox4 antibody.

Fig. 2.

Subcellular localization of Nox4 in cultured rat MCs. (A) Nox4 protein or Cyt c was detected by Western blot analysis in 1/5th vol of crude subcellular fractions, Tot, Mit, C, or Mb. (B) Mit were isolated by using a mitochondria purification kit (Pierce). Nox4 protein or Cyt c was detected by Western blot analysis. C, Cytosolic fraction. (C) Pure mitochondria were isolated by using differential centrifugation and Percoll gradient. Fractions were collected, and immunoblotting of each fraction was performed with Nox4 or prohibitin antibodies. Also shown is the immunoblotting of the total cell homogenate. (D) Increasing protein concentration of pure Mit fraction were immunoblotted with Nox4 antibody. (E) MC mitochondria were labeled with MTG, fixed, permeabilized, and stained by using our Nox4 antibody and a Cyanin-3-linked donkey anti-rabbit secondary antibody. Nuclei were counterstained by using the dye DAPI. (F) MC mitochondria were visualized with MTR and then stained with our Nox4 antibody by using a FITC-linked donkey anti-rabbit secondary antibody. Nuclei were counterstained with DAPI. (G) MC mitochondria were labeled with MTR and stained with a commercial rabbit anti-Nox4 antibody and Cyanin-3-linked secondary antibody. The color of the overlay was assigned as red and green respectively. (H) Endoplasmic reticulum (ER) was stained by using an antibody directed against the marker PDI labeled with Alexa Fluor 488 dye. (I) Golgi apparatus was stained by using Golgin-97 as marker and appropriate secondary antibody.

Fig. 3.

Nox activity in mitochondria. NADPH oxidase activity was assessed in Percoll gradient-purified mitochondria isolated from control MCs or cells treated with DPI (5 μM) either by measuring NADPH-dependent superoxide generation with the lucigenin-enhanced chemiluminescence method (A) or NADPH-dependent hydrogen peroxide production with Amplex Red reagent (B). **, P < 0.01 versus control cells. Values are the mean ± SE of three independent experiments.

Fig. 4.

Mitochondrial Nox4 is functional. (A) Nox4 mRNA expression was assessed by real-time PCR in MCs untransfected (UTr) or transfected with Nox4 siRNA (siNox4) or scrambled siRNA (scr). (B) Nox4 expression was analyzed by Western blot analysis in total cell lysates from UTr and siNox4- or scr-transfected cells by using our antibody. The histogram represents the ratio of the intensity of Nox4 bands quantified by densitometry factored by the densitometric measurement of GAPDH bands. The data are expressed as percentage of control where the ratio in the control was defined as 100%. Values are the means ± SE from three independent experiments. ##, P < 0.01 versus UTr cells. (C) Nox4 was visualized by immunofluorescence in UTr and siNox4- or scr-transfected cells with our antibody. Fluorescence intensity was semiquantified, and values are the means ± SE from three independent experiments. **, P < 0.01 versus UTr cells. (D) NADPH-dependent superoxide generation was measured in pure Mit fraction prepared from siNox4- or scr-transfected cells. Values are the means ± SE from three independent experiments. **, P < 0.01 versus scr-transfected cells. (E) Digitonin-permeabilized mitochondria were immunoprecipitated by using our Nox4 antibody, the commercial antibody [Nox4 (49)] or IgG control. NADPH-dependent superoxide generation was then measured in the immunoprecipitate. Values are the means ± SE from three independent experiments. **, P < 0.01 versus IgG.

Fig. 5.

Mitochondrial Nox4 is involved in glucose-induced ROS generation in MCs. (A) MCs were treated with HG for the indicated times. Equivalent amounts of cell lysates were analyzed by Western blot analysis for Nox4 expression by using our antibody. Actin was used as a loading control. The histogram represents the ratio of the intensity of Nox4 bands quantified by densitometry factored by the densitometric measurement of actin band. The data are expressed as percentage of control where the ratio in the control was defined as 100%. Values are the means ± SE from three independent experiments. *, P < 0.05 and **, P < 0.01 versus control. (B) NADPH-dependent ROS generation was measured from total and pure Mit fractions isolated from MCs after exposure to NG or HG for 4 h. (C) Representative images obtained by confocal fluorescence microscopy of MitoSOX Red fluorescence in UTr, scr-transfected, and siNox4-transfected MCs after exposure to NG (5 mM d-glucose) or HG (25 mM d-glucose) for 1 h. After fixation and permeabilization, it was cell stained with our Nox4 antibody and appropriate FITC-conjugated secondary antibody. Nuclei were counterstained with DAPI. Fluorescence intensity was semiquantified, and values are the means ± SE from three independent experiments. **, P < 0.01 versus NG.

Fig. 6.

Regulation of mitochondrial Nox4 in diabetes. (A) (Left) Immunoblots showing Nox4 protein expression in total cortex homogenate and crude Mit fractions from control (Con) and diabetic (DM) animals. (Right) Quantification of Nox4 expression by using our Nox4 antibody in the total and Mit fraction. Cyt c oxidase subunit VI (COX VI) was included as a mitochondrial marker. Each histogram represented the ratio of the intensity of Nox4 bands quantified by densitometry factored by the densitometric measurement of COX VI band. Values are the means ± SE from six animals in each group. **, P < 0.01 versus control rats. (B) Immunoblots showing Nox4 protein expression by using our antibody in total cortex homogenate and Mit fractions from Con and DM animals. Porin was included as a mitochondrial marker. Each histogram represented the ratio of the intensity of Nox4 bands quantified by densitometry factored by the densitometric measurement of porin band. Values are the means ± SE from four animals in each group. **, P < 0.01 versus control rats. (C) NADPH-dependent superoxide generation was measured in parallel from total and Percoll-purified Mit fractions isolated from the Con or DM rat kidney cortex. Values are the means ± SE from four animals in each group. **, P < 0.01 versus control rats.