NADPH oxidase-2 mediates zinc deficiency-induced oxidative stress and kidney damage

Abstract

Zn²⁺ deficiency (ZnD) is comorbid with chronic kidney disease and worsens kidney complications. Oxidative stress is implicated in the detrimental effects of ZnD. However, the sources of oxidative stress continue to be identified. Since NADPH oxidases (Nox) are the primary enzymes that contribute to renal reactive oxygen species generation, this study's objective was to determine the role of these enzymes in ZnD-induced oxidative stress. We hypothesized that ZnD promotes NADPH oxidase upregulation, resulting in oxidative stress and kidney damage. To test this hypothesis, wild-type mice were pair-fed a ZnD or Zn²⁺-adequate diet. To further investigate the effects of Zn²⁺ bioavailability on NADPH oxidase regulation, mouse tubular epithelial cells were exposed to the Zn²⁺ chelator N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN) or vehicle followed by Zn²⁺ supplementation. We found that ZnD diet-fed mice develop microalbuminuria, electrolyte imbalance, and whole kidney hypertrophy. These markers of kidney damage are accompanied by elevated Nox2 expression and H₂O₂ levels. In mouse tubular epithelial cells, TPEN-induced ZnD stimulates H₂O₂ generation. In this in vitro model of ZnD, enhanced H₂O₂ generation is prevented by NADPH oxidase inhibition with diphenyleneiodonium. Specifically, TPEN promotes Nox2 expression and activation, which are reversed when intracellular Zn²⁺ levels are restored following Zn²⁺ supplementation. Finally, Nox2 knockdown by siRNA prevents TPEN-induced H₂O₂ generation and cellular hypertrophy in vitro. Together, these findings reveal that Nox2 is a Zn²⁺-regulated enzyme that mediates ZnD-induced oxidative stress and kidney hypertrophy. Understanding the specific mechanisms by which ZnD contributes to kidney damage may have an important impact on the treatment of chronic kidney disease.

chronic kidney disease (CKD), regardless of etiology, is accompanied by Zn²⁺ deficiency (ZnD) (13, 16, 17, 34). In CKD, ZnD is attributed to decreased dietary intake, reduced intestinal Zn²⁺ absorption, and increased renal excretion (35). Since Zn²⁺ is an essential cofactor that influences the expression and activity of numerous enzymes, transcription factors, and regulatory proteins, disruption of Zn²⁺ homeostasis has negative consequences on epidermal, gastrointestinal, central nervous, immune, skeletal, and reproductive systems (43). ZnD has also been demonstrated to contribute to kidney damage (32, 51). Experimental findings show that ZnD exacerbates diabetic kidney damage, as evident by enhanced oxidative damage, fibrosis, and renal dysfunction (3, 32, 52–54). Moreover, the beneficial effects of adequate Zn²⁺ are underscored by Zn²⁺ supplementation studies showing attenuation of the progression of kidney disease (3, 32, 51).

Oxidative stress, resulting from sustained reactive oxygen species (ROS) levels, is implicated in the detrimental effects associated with ZnD (11, 28, 47, 58, 60). There is evidence that Zn²⁺ exerts antioxidative effects (11, 12, 24, 41). Specifically, Zn²⁺ associates with sulfhydryl groups, thereby protecting proteins from oxidation (7). Additionally, Zn²⁺ is an essential cofactor for antioxidant proteins, such as Cu,Zn superoxide dismutase (SOD) and metallothionein (7). Although it is well known that antioxidative mechanisms contribute to ZnD-induced oxidative stress (7, 18, 26), the specific prooxidative pathways continue to be defined.

In the kidney, NADPH oxidases (Nox) are the primary sources of oxidative stress (20, 55, 59). Physiologically, these enzymes are critical to signaling pathways involved in gene expression, apoptosis, differentiation, migration, and proliferation (9). Nox enzymes comprise several subunits that include catalytic, membrane-bound isoforms such as Nox1-5 or dual oxidase 1/2 and p22^phox, as well as regulatory, cytosolic components such as p47^phox, p67^phox, NoxA1, NoxO1, Rac1, and polymerase-δ-interacting protein 2 (Poldip2) (5). Various stimuli, including hyperglycemia, oxidatively modified lipoproteins, and advanced glycation end products, promote subunit assembly into a functional complex. Once assembled, the Nox enzyme complex generates O₂^·− by transfer of an electron from NADPH to molecular oxygen (6). Unlike other Nox-containing NADPH oxidases, Nox4 is constitutively active and primarily produces H₂O₂ (33, 36, 39). Although Nox4 does not require assembly of cytosolic subunits to be functional (19), its activity is enhanced by association with Poldip2 (33, 39).

Although Nox2 and Nox4 have been implicated in kidney disease (46, 57), the role of these Nox enzymes in ZnD-induced oxidative stress and renal damage has not been investigated. Therefore, the objectives of this study were to determine if Nox enzymes are 1) regulated by Zn²⁺ and 2) sources of ZnD-induced renal oxidative stress. In this study, in vivo and in vitro models of ZnD were used to demonstrate that Nox2 and Nox4 are Zn²⁺-regulated enzymes and that Nox2 mediates ZnD-induced oxidative stress and kidney damage. These novel findings identify Nox2-containing NADPH oxidase as a specific mechanism by which ZnD contributes to oxidative stress and subsequent kidney damage.

MATERIALS AND METHODS

Zn²⁺ Deficiency Models

In vivo.

An established protocol (4, 29) was used to induce ZnD in 8- to 12-wk-old male and female wild-type (WT) mice on a mixed genetic background. Mice received a ZnD diet (1 ppm; Harlan Teklad, Madison, WI) for 6 wk. Control mice received a pair-matched Zn²⁺-adequate (ZnA) diet (50 ppm; Harlan Teklad). To maintain a Zn²⁺-free environment, deionized drinking H₂O was provided in Zn²⁺-free containers and cages were changed daily. After the experimental period, urine and kidneys were collected for analysis. This animal use protocol was reviewed and approved by the Institutional Animal Care and Use Committee at the Atlanta Veterans Affairs Medical Center.

In vitro.

Mouse tubular epithelial cells (mTEC) derived from the distal convoluted tubular segment of nephrons (30, 44) were grown at 37°C in 5% CO₂ in DMEM supplemented with 5% fetal bovine serum and 1% penicillin-streptomycin. At 75% confluence, the culture medium was changed to serum-free Opti-MEM. To reduce intracellular Zn²⁺ levels, mTEC were cultured in 1 nM N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN)- or vehicle (DMSO)-containing medium for 24 h. To replete intracellular Zn²⁺, TPEN- or vehicle-exposed mTEC were then cultured in 1 nM Zn²⁺-supplemented medium for 24 h. To determine the role of Nox enzymes in ROS generation, select monolayers were pretreated with 10 nM diphenyleneiodonium (DPI), a general Nox enzyme inhibitor, or vehicle before and during TPEN exposure. To identify the specific Nox isoforms involved in ROS generation, mTEC were transfected with 100 nM scrambled oligonucleotides or oligonucleotides targeting Nox2 (catalog no. sc-35504, Santa Cruz Biotechnology) or Nox4 (catalog no. sc-41587, Santa Cruz Biotechnology). To determine the role of SOD in TPEN-induced ROS generation, mTEC were treated with 5 μM LCS-1 (catalog no. 41931-13-9, Sigma), a SOD1 inhibitor, with or without TPEN.

Zn²⁺ Assessment

FluoZin3.

To assess changes in intracellular Zn²⁺ levels, confluent cell monolayers were loaded with 5 μg/ml FluoZin3 (Thermo Fisher Scientific), a fluorescent Zn²⁺ probe, for 1 h at 37°C in Krebs-Ringer phosphate glucose (KRPG) buffer (in mM: 145 NaCl, 5.7 KH₂PO₄, 4.86 KCl, 0.54 CaCl₂, 1.22 MgSO₄, and 5.5 glucose, pH 7.35). After the cells were washed with 1× phosphate-buffered saline (PBS), intracellular FluoZin3 fluorescence was monitored by confocal microscopy (Olympus) at 494-nm excitation and 516-nm emission.

Zip10 mRNA expression.

To confirm changes in Zn²⁺ bioavailability, mRNA expression of Zn²⁺ influx transporter ZRT/IRT-like protein 10 (Zip10), a Zn²⁺-regulated Zn²⁺ transporter upregulated in ZnD, was measured. Briefly, total RNA was isolated from cells with TRIzol according to the manufacturer's protocol (Invitrogen). cDNA was generated and amplified using One-Step SYBR Green (Bio-Rad). All data were normalized to the 9s mRNA content of the same sample. Zip10 mRNA expression was calculated using the cycle threshold (ΔΔC_t) method.

Kidney Damage Assessment

Proteinuria.

To assess kidney damage, total urinary protein was measured using a bicinchoninic acid (BCA) protein assay (Pierce). Protein concentrations were calculated by extrapolation of values from a protein standard curve. Urinary protein was normalized to 24-h urine volume.

Microalbuminuria.

To confirm kidney damage, urinary albumin was measured. Briefly, albumin and creatinine were measured in spot urine samples by ELISA (EXOCELL). Microalbuminuria was determined by calculation of the albumin-to-creatinine ratios.

Hypertrophy.

To assess renal changes, whole kidney hypertrophy was determined from total kidney weight-to-body weight ratios. To assess cellular hypertrophy in vitro, protein-to-DNA ratios were determined. Briefly, cells were incubated with 1 μM Hoechst in KRPG buffer for 30 min at 37°C in darkness to detect DNA content. Cells were washed with KRPG buffer, and Hoechst fluorescence was measured at 350-nm excitation and 460-nm emission. Finally, cells were lysed, and the protein content was quantified using a BCA protein assay. Protein concentration for each sample was normalized by respective DNA content as determined by Hoechst fluorescence intensity.

Total urinary electrolytes.

To assess electrolyte balance, total urinary electrolytes were calculated using an electrolyte analyzer (EasyLyte, Medica). Briefly, Na⁺, K⁺, and Cl⁻ concentrations were measured in 24-h urine samples. Urinary electrolytes were normalized to 24-h urine volume.

ROS Measurement

Amplex Red.

H₂O₂ was measured using horseradish peroxidase-catalyzed oxidation of the nonfluorescent molecule N-acetyl-3,7-dihydroxyphenoxazine to the highly fluorescent molecule resorufin (Amplex Red Assay, Invitrogen). Kidney slices and cells were incubated in KRPG buffer containing 100 μl/ml Amplex Red and 0.2 U/ml horseradish peroxidase for 2 h at 37°C in darkness. After collection of supernatants, resorufin fluorescence was measured at 540-nm excitation and 590-nm emission. Sample fluorescence was compared with that generated by a H₂O₂ standard curve to calculate H₂O₂ concentrations. Renal H₂O₂ concentrations were normalized to respective kidney slice weight. For cellular H₂O₂ concentrations, mTEC were lysed in RIPA lysis buffer. Total cellular proteins were collected and quantified by a BCA protein assay. Cellular H₂O₂ concentrations were normalized to protein concentrations.

DCF.

O₂^·− was detected using the fluorescent probe 2′,7′-dihydrodichlorofluorescein diacetate (DCF-DA; Invitrogen). Confluent mTEC monolayers were loaded with 25 μg/ml DCF-DA for 1 h at 37°C in KRPG buffer in darkness. For quantification, cells were washed in 1× PBS, and then DCF fluorescence intensity was measured on a microplate reader (Omega, BMG LABTECH, Cary, NC) at 488-nm excitation and 520-nm emission. Thereafter, cells were lysed with RIPA lysis buffer, and total cellular proteins were quantified by a BCA protein assay. DCF fluorescence intensities were normalized to respective protein concentrations.

NADPH Oxidase Expression

Quantitative RT-PCR.

To measure Nox mRNA levels, quantitative RT-PCR was performed. Total RNA was isolated from cells with TRIzol according to the manufacturer's protocol (Invitrogen). cDNA was generated and amplified using One-Step SYBR Green (Bio-Rad). Validated Nox2 (GeneCopoeia no. MQP026893) and Nox4 (GeneCopoeia no. MQP031270) primers were used. All data were normalized to the 9s mRNA content of the same sample. Nox2 and Nox4 mRNA levels were calculated using the ΔΔC_t method.

Western blotting.

To measure Nox protein levels, Western blot analysis was performed. Briefly, cells were lysed using RIPA lysis buffer. Protein (50 μg) was separated by 7.5% SDS-PAGE and then transferred onto a PVDF membrane. The membrane was incubated in 1% bovine serum albumin in PBS and then immunoblotted with primary antibodies (1:500 dilution) specific for Nox2 (catalog no. ab129068, Abcam), Nox4 (catalog no. ab133303, Abcam), or β-actin or GAPDH (Cell Signaling Technology, Danvers, MA). After incubation in rabbit secondary antibody (1:5,000 dilution), immunoreactive bands were detected using the Syngene imaging system. Densitometry analyses were performed using GeneTools analysis software. Densitometric values for Nox4 high- and low-molecular-weight bands were combined for analysis. Nox densitometric values were normalized to respective loading control (GAPDH or β-actin).

NADPH Oxidase Activation

Coimmunoprecipitation.

To determine the activation status of NADPH oxidases, Nox2-p67^phox and Nox4-Poldip2 associations were examined by a coimmunoprecipitation assay performed according to the manufacturer's instructions (catalog no. 26149, Thermo Fisher Scientific). Briefly, cells were lysed in RIPA lysis buffer. Collected proteins were quantified by a BCA protein assay. Then 50 μg of protein were incubated with protein A beads conjugated with p67^phox (catalog no. 07-002, Millipore) or Poldip2 (catalog no. ab181841, Abcam). Bound Nox2 and Nox4 proteins were detected by Western blot analysis. Densitometry analyses of immunoreactive bands were performed using GeneTools analysis software.

Statistical Analysis

For all experiments, graphing and statistical analyses were performed using GraphPad software (Prism, San Diego, CA). For all experiments comparing only two groups, statistical analysis was performed by Student's t-test. When more than two groups were analyzed, two-way ANOVA was followed by Bonferroni's post hoc analysis. Data are expressed as means ± SE.

RESULTS

Dietary Zn²⁺ Restriction Promotes Kidney Damage and Upregulates Nox2 in Mice

To investigate renal consequences of reduced Zn²⁺ bioavailability, WT mice were pair-fed a ZnD or ZnA diet. Consistent with ZnD-induced growth retardation, body weights are significantly reduced in mice fed the ZnD diet compared with those fed the ZnA diet (Table 1). To access glomerular function, urinary proteins (proteinuria) were measured. After 6 wk of dietary Zn²⁺ restriction, total urinary proteins are significantly increased (Fig. 1A). Additionally, urinary albumin levels (microalbuminuria) are elevated in mice fed the ZnD diet (Fig. 1B). To access tubular function, total urinary electrolytes were measured. Na⁺, K⁺, and Cl⁻ are reduced in urine samples from mice fed the ZnD diet compared with those fed the ZnA diet (Fig. 1C). Consistent with changes in function, kidneys of ZnD mice undergo hypertrophy (Table 1), as assessed by kidney weight-to-body weight ratios. Collectively, glomerular and tubular dysfunction indicates that ZnD promotes kidney damage.

Table 1.

Kidney and body weights of mice fed a ZnA or ZnD diet

	ZnA (n = 5)	ZnD (n = 6)
Body wt, g	26.6 ± 1.652	22.49 ± 1.678*
Kidney wt, mg	294.2 ± 31.73	301.7 ± 27.01
Kidney wt/body wt, mg/g	10.65 ± 0.443	14.07 ± 0.889*

Values are means ± SE. ZnA, Zn²⁺-adequate (50 ppm); ZnD, Zn²⁺-deficient (1 ppm).

^*P < 0.05 vs. ZnA.

Fig. 1.

Dietary Zn²⁺ restriction promotes kidney damage in mice. To examine the effect of reduced Zn²⁺ bioavailability on the kidney, wild-type (WT) mice were pair-fed a Zn²⁺-deficient (ZnD) or Zn²⁺-adequate (ZnA) diet. To assess kidney damage, proteinuria (A), microalbuminuria (B), and total urinary electrolytes (C) were examined. Values are means ± SE of 5–6 mice per group. *P < 0.05 vs. ZnA.

To examine effects of reduced Zn²⁺ bioavailability on renal ROS generation, H₂O₂ levels were measured in WT mice fed a ZnD or ZnA diet. Renal (Fig. 2A) and urinary (Fig. 2B) H₂O₂ levels are significantly increased in mice fed the ZnD diet. To examine alterations in NADPH oxidases, Nox isoform expression was assessed. Elevated ROS generation (Fig. 2) is accompanied by increased Nox2 mRNA (Fig. 3A) and protein (Fig. 3, B and C) expression. Although Nox4 mRNA levels do not change (Fig. 3A), Nox4 protein expression (Fig. 3, B and D) is significantly reduced in ZnD diet-fed mice. These results show differential regulation of Nox isoform expression. Furthermore, these findings reveal that ZnD induces oxidative stress and Nox2 upregulation in kidneys.

Fig. 2.

Dietary Zn²⁺ restriction stimulates renal reactive oxygen species (ROS) generation in mice. To examine effects of reduced Zn²⁺ bioavailability on renal ROS generation, WT mice were pair-fed a ZnD or ZnA diet. To examine changes in ROS generation, renal (A) and urinary (B) H₂O₂ levels were measured. Values are means ± SE of 5–6 mice per group. *P < 0.05 vs. ZnA.

Fig. 3.

Dietary Zn²⁺ restriction stimulates renal NADPH oxidase (Nox)-2 upregulation in mice. To examine effects of reduced Zn²⁺ bioavailability on renal NADPH oxidase expression, WT mice were pair-fed a ZnD or ZnA diet. Alterations in Nox2 and Nox4 mRNA (A) and protein (B–D) expression were examined. Representative blots are shown. Values are means ± SE of 5–6 mice per group. *P < 0.05 vs. ZnA. ns, Not significant.

Intracellular Zn²⁺ Modulates ROS Generation and Nox2 Regulation in mTEC

To manipulate intracellular Zn²⁺ bioavailability, mTEC were treated with the Zn²⁺ chelator TPEN followed by Zn²⁺ supplementation. TPEN reduces intracellular Zn²⁺ levels compared with vehicle-treated cells (Fig. 4A); furthermore, Zn²⁺ supplementation restores intracellular Zn²⁺ to control levels. Consistent with these findings, intracellular Zn²⁺ reduction by TPEN promotes mRNA upregulation of Zip10 (Fig. 4B). Additionally, restoration of intracellular Zn²⁺ levels by Zn²⁺ supplementation reduces Zip10 mRNA to control levels.

Fig. 4.

Intracellular Zn²⁺ modulates ROS generation in mouse tubular epithelial cells (mTEC). To examine ROS generation, mTEC were exposed to the intracellular Zn²⁺ chelator N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN) followed by Zn²⁺ supplementation. Select monolayers were treated with TPEN in the presence or absence of the SOD1 inhibitor LCS-1. To assess Zn²⁺ bioavailability, intracellular Zn²⁺ (A) and Zn²⁺ influx transporter ZRT/IRT-like protein 10 (Zip10, B) mRNA levels were examined. Changes in ROS generation were assessed by examining H₂O₂ (C) and O₂^·− (D) levels. Values are means ± SE of 3 independent experiments performed in replicates. Representative confocal images are shown. *P < 0.05 vs. control vehicle. #P < 0.05 vs. control TPEN.

To investigate a possible role of Zn²⁺ in ROS generation, H₂O₂ and O₂^·− levels were measured in mTEC exposed to TPEN followed by Zn²⁺ supplementation. TPEN-induced ZnD stimulates H₂O₂ generation compared with vehicle-treated cells (Fig. 4C). Zn²⁺ repletion reverses TPEN-induced H₂O₂ generation to control levels. Contrary to expectations, O₂^·− is reduced in TPEN-treated mTEC, as assessed by two independent methods, DCF (Fig. 4D) and dihydroethidium (data not shown). Moreover, basal O₂^·− levels are reduced with Zn²⁺ supplementation compared with vehicle controls and remain low in TPEN-treated cells. However, O₂^·− generation is significantly increased in TPEN-treated cells exposed to the SOD-1 inhibitor LCS-1. Collectively, these results demonstrate that Zn²⁺ plays a role in ROS generation. Specifically, ZnD promotes O₂^·− generation, which is subsequently converted to H₂O₂ via SOD-1.

To investigate a possible role of Nox enzymes in ZnD-induced ROS generation, mTEC were treated with DPI, a general NADPH oxidase inhibitor, before and during TPEN exposure. ROS generation was examined by measurement of H₂O₂ and O₂^·− levels. Basal H₂O₂ levels are unaffected by DPI (Fig. 5A). However, TPEN-induced H₂O₂ generation is significantly blunted with DPI treatment. Consistent with results shown in Fig. 4D, O₂^·− levels are reduced in TPEN-treated cells. Additionally, basal O₂^·− generation is reduced in DPI-treated mTEC (Fig. 5B).

Fig. 5.

NADPH oxidases mediate ZnD-induced oxidative stress in mTEC. To investigate the role of NADPH oxidases in ZnD-induced ROS generation, mTEC were exposed to TPEN in the presence or absence of the NADPH oxidase inhibitor diphenyleneiodonium (DPI). To assess ROS generation, H₂O₂ (A) and O₂^·− (B) levels were examined. Values are means ± SE of 3 independent experiments performed in replicates. *P < 0.05 vs. control vehicle. #P < 0.05 vs. control TPEN.

Since these data support a role for Nox enzymes in ZnD-induced oxidative stress, the specific Nox isoforms involved were examined in mTEC exposed to TPEN followed by Zn²⁺ supplementation. Nox2 and, to a lesser extent, Nox4 mRNA expressions are upregulated in TPEN-exposed mTEC compared with vehicle-treated cells (Fig. 6A). Nox2 mRNA expression is significantly reduced with intracellular Zn²⁺ repletion, while Nox4 mRNA is unaffected. Moreover, Western blot analysis shows that Nox2 protein expression is increased, but Nox4 protein expression is significantly reduced with TPEN treatment (Fig. 6, B and C). In Zn²⁺-supplemented cells, elevated Nox2 protein expression is attenuated while Nox4 expression remains reduced. These results demonstrate that Nox2 and Nox4 expression is regulated by intracellular Zn²⁺.

Fig. 6.

Intracellular Zn²⁺ modulates Nox subunit expression in mTEC. To examine effects of Zn²⁺ bioavailability on NADPH oxidases, Nox subunit expression was examined in mTEC exposed to TPEN followed by Zn²⁺ supplementation. To assess changes in NADPH oxidases, Nox subunit mRNA (A) and protein (B) expression was examined. Values are means ± SE of 3 independent experiments performed in replicates. C: representative blots. *P < 0.05 vs. vehicle. #P < 0.05 vs. TPEN.

To investigate the role of Zn²⁺ in Nox enzyme activation, subunit association was examined in mTEC exposed to TPEN followed by Zn²⁺ supplementation. Nox2-p67^phox association is increased in TPEN-induced Zn²⁺-depleted cells compared with vehicle-treated cells (Fig. 7A), as assessed by a coimmunoprecipitation assay. Furthermore, Zn²⁺ repletion reverses enhanced Nox2-p67^phox association. Nox4 and Poldip2 are associated basally (Fig. 7B). However, this association was reduced in TPEN-exposed cells. The Nox4-Poldip2 association is not reversed with Zn²⁺ repletion. These results confirm that Nox2 and Nox4 activation are also differentially regulated by intracellular Zn²⁺.

Fig. 7.

Intracellular Zn²⁺ modulates Nox activation in mTEC. To examine effects of Zn²⁺ bioavailability on NADPH oxidases, Nox enzyme activation was examined in mTEC exposed to TPEN followed by Zn²⁺ supplementation. To identify the active Nox isoform, Nox2 coimmunoprecipitation (co-IP) with p67^phox (A) and Nox4 coimmunoprecipitation with polymerase-δ-interacting protein 2 (Poldip2, B) were assessed. Values are means ± SE of 3 independent experiments. Representative blots are shown. *P < 0.05 vs. vehicle. #P < 0.05 vs. TPEN.

Nox2-Containing NADPH Oxidase Mediates ZnD-Induced Oxidative Stress and Cellular Hypertrophy in mTEC

To identify the specific Nox isoform involved in ZnD-induced oxidative stress, mTEC were transfected with Nox2- or Nox4-targeting oligonucleotides to knock down Nox expression. In cells transfected with scrambled oligonucleotides, TPEN significantly increases H₂O₂ generation (Fig. 8A). However, TPEN-induced H₂O₂ generation is prevented in cells transfected with Nox2-specific siRNA (si-Nox2), but not in cells transfected with Nox4-specific siRNA (si-Nox4) (Fig. 8A).

Fig. 8.

Nox2-containing NADPH oxidase mediates ZnD-induced oxidative stress and kidney cellular hypertrophy in mTEC. To identify the Nox isoform involved in ZnD-induced renal effects, mTEC were transfected with Nox-specific siRNA (si-Nox) or scrambled oligonucleotides prior to TPEN exposure. A: ROS generation was examined by measuring H₂O₂ levels. B: cellular hypertrophy was assessed by calculating protein-to-DNA ratios. Values are means ± SE of 3 independent experiments performed in replicates. *P < 0.05 vs. vehicle.

Finally, to investigate the role of Nox isoforms in ZnD-induced renal hypertrophy, si-Nox-transfected cells were exposed to TPEN. TPEN induces cellular hypertrophy in cells transfected with scrambled oligonucleotides, as assessed by protein-to-DNA ratios (Fig. 8B). However, cellular hypertrophy is prevented by knockdown of Nox2, but not Nox4. Together, these findings demonstrate that Nox2-containing NADPH oxidase mediates ZnD-induced oxidative stress and subsequent cellular hypertrophy.

DISCUSSION

ZnD has been shown to exacerbate kidney damage, in part, by promoting oxidative stress (11, 28, 47, 58, 60). However, the contribution of prooxidant mechanisms continues to be defined. Since NADPH oxidases are a primary source of ROS generation in the kidney (20, 55, 59), this study tested the hypothesis that reduced Zn²⁺ bioavailability induces NADPH oxidase upregulation, thereby promoting oxidative stress and subsequent kidney damage. Using in vivo and in vitro models of ZnD, our novel findings demonstrate that Nox2-containing NADPH oxidase is a Zn²⁺-regulated enzyme that plays a critical role in ZnD-induced oxidative stress and kidney damage (Fig. 9).

Fig. 9.

Proposed schema. Nox-2 is a Zn²⁺-regulated enzyme that mediates ZnD-induced oxidative stress and subsequent kidney damage. [Zn²⁺]_i, intracellular Zn²⁺ concentration.

Disrupted Zn²⁺ homeostasis contributes to oxidative stress. Studies show that Zn²⁺ toxicity, due to elevated intracellular Zn²⁺ levels, promotes ROS generation (25). Specifically, increased ROS levels were observed in renal and neuronal cells exposed to high Zn²⁺ concentrations (25). Additionally, studies demonstrate that ZnD also stimulates ROS generation (28, 50) and implicate antioxidative mechanisms (11, 12, 24, 41). Recent findings indicate that prooxidative pathways are activated in response to ZnD. Specifically, this study (Fig. 4) and others (2, 40) indicate a role for NADPH oxidases in ZnD-induced oxidative stress. This study is the first to demonstrate that Nox2-derived ROS is modulated by intracellular Zn²⁺ bioavailability. Our in vitro findings demonstrate that Nox inhibition via DPI prevents ZnD-induced ROS generation (Fig. 5). Furthermore, our study further expands the current knowledge by demonstrating that Nox2-containing NADPH oxidase mediates ZnD-induced oxidative stress. Specifically, knockdown of Nox2 via siRNA inhibits H₂O₂ generation (Fig. 8A), indicating that Nox2 is the catalytic Nox isoform responsible for ROS generation. Additionally, this study identifies H₂O₂ as the predominant ROS that mediates ZnD-induced oxidative stress. Our results show that H₂O₂, but not O₂^·−, levels are sustained with ZnD (Fig. 4). These finds are unexpected, since Nox2, which produces O₂^·−, is upregulated, while Nox4, which primarily generates H₂O₂, is downregulated (Figs. 6 and 7). An explanation for this discrepancy is that Nox2-generated O₂^·− is rapidly converted to H₂O₂ by SOD. Our findings support this explanation by showing elevated O₂^·− levels in the presence of the SOD-1 inhibitor LCS-1 (Fig. 4), suggesting that SOD is activated with ZnD and mediates the dismutation of O₂^·− to H₂O₂. This is consistent with studies showing increased SOD activity in response to ZnD (56). Moreover, signaling events that occur as a result of O₂^·− are mediated by its more stable dismutation product H₂O₂ (1, 9)_.

NADPH oxidases are Zn²⁺-regulated enzymes (25, 37). In cardiomyocytes, ischemia-reperfusion-induced injury is attributed, in part, to ZnD (25). In this ZnD model, Nox2 expression and p47^phox phosphorylation are enhanced. These changes are prevented with Zn²⁺ supplementation (25), suggesting that reduced Zn²⁺ levels induce NADPH oxidase upregulation. In models of Zn²⁺ toxicity, increased p67^phox activation is observed (37). Specifically, Matsunaga et al. demonstrated p67^phox membrane translocation in response to toxic Zn²⁺ levels in renal proximal tubular cells (37), indicating that increased Zn²⁺ levels also stimulate NADPH oxidase upregulation. Together, these studies suggest that NADPH oxidases are modulated by Zn²⁺ bioavailability. However, the specific isoforms affected were unknown. The current study extends these findings by directly demonstrating that Nox2- and Nox4-containing NADPH oxidases are modulated by intracellular Zn²⁺ bioavailability. In vivo findings show that ZnD is accompanied by enhanced Nox2 expression and ROS generation (Figs. 2 and 3). To further investigate the role of Zn²⁺ bioavailability in Nox regulation, mTEC were treated with the intracellular Zn²⁺ chelator TPEN. Consistent with in vivo results, in vitro findings reveal that intracellular Zn²⁺ depletion promotes Nox2 expression, activation, and activity, which are reversed with Zn²⁺ repletion (Figs. 4, 6, and 7). Contrary to Nox2, Nox4 is downregulated with ZnD. Furthermore, Zn²⁺ repletion does not restore Nox4 to control levels. It can be assumed that Nox4 downregulation is a possible compensatory mechanism to reduce elevated intracellular ROS levels. Although differential Nox regulation has been previously shown (10, 21, 38), additional studies are needed to investigate the distinct mechanisms involved during ZnD. Collectively, these findings reveal that Nox2-containing NADPH oxidase is upregulated in response to reduced Zn²⁺ bioavailability while Nox4-containing NADPH oxidase is downregulated. This study is the first to directly demonstrate that NADPH oxidases are regulated by intracellular Zn²⁺ bioavailability.

Microalbuminuria is an early derangement that occurs in kidney damage (5). In diabetic mice with ZnD, microalbuminuria is worsened compared with diabetic mice with normal serum Zn²⁺ levels (32). Our study extends these findings by directly demonstrating that ZnD contributes to kidney damage. In vivo findings reveal that ZnD alone promotes proteinuria and microalbuminuria (Fig. 1). Furthermore, these derangements are accompanied by kidney hypertrophy (Table 1), which is also an early event in kidney damage and predicts eventual loss of kidney function (14). These findings are consistent with the renal dysfunction observed in mice with ZnD (3, 52–54). It is well established that NADPH oxidases play a critical role in kidney damage (6, 20, 23, 45, 49, 55, 59). Nox2-containing NADPH oxidase plays a role in cyclosporine-induced kidney damage (14, 15), while Nox4-containing NADPH oxidase is involved in diabetes-induced kidney damage (20, 23, 45). However, it is unknown whether Nox2- or Nox4-containing NADPH oxidase is responsible for ZnD-induced kidney damage. Previously, we showed that Nox2 contributes to high glucose-induced cellular hypertrophy (57). Consistent with this finding, the current study shows that Nox2 is also involved in ZnD-induced cellular hypertrophy. In vivo results show that whole kidney hypertrophy is accompanied by Nox2 upregulation (Table 1, Fig. 3). In vitro findings reveal that Nox2 knockdown prevents cellular hypertrophy (Fig. 8B). However, this was not observed with Nox4 knockdown. This is consistent with Nox2 mediation of ZnD-induced oxidative stress (Fig. 8A). Collectively, our findings suggest that Nox2-containing NADPH oxidase plays a role in ZnD-induced kidney damage. However, in vivo studies are needed to further investigate ZnD-induced kidney damage and the underlying mechanisms.

Clinical and experimental studies reveal that CKD, regardless of etiology, is accompanied by ZnD (8, 13, 16, 17, 22, 27, 34). The importance of Zn²⁺ monitoring in CKD patients is underscored by Zn²⁺ supplementation studies demonstrating reversal or attenuation of complications (31, 42, 48). This study identifies an underlying mechanism. We reveal that Nox2-derived oxidative stress is a mediator of ZnD-induced kidney damage. Furthermore, our findings show that Zn²⁺ supplementation is sufficient to reverse ZnD-induced oxidative damage and may represent a possible intervention in CKD.

GRANTS

This work was supported by National Institutes of Health Grants R25 DK-101390 (M. S. Li), F31 HL-114386 (S. E. Adesina), R15 GM-113120 (J. L. Gooch), R01 DK-085097 (R. S. Hoover), and T32 DK-007656 and R01 DK-085097-05S1 (C. R. Williams).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

M.S.L., S.E.A., C.L.E., J.L.G., and C.R.W. performed the experiments; M.S.L., S.E.A., J.L.G., R.S.H., and C.R.W. analyzed the data; M.S.L., S.E.A., C.L.E., J.L.G., R.S.H., and C.R.W. interpreted the results of the experiments; M.S.L. and C.R.W. prepared the figures; M.S.L. and C.R.W. drafted the manuscript; M.S.L., S.E.A., and C.R.W. edited and revised the manuscript; M.S.L., S.E.A., C.L.E., J.L.G., R.S.H., and C.R.W. approved the final version of the manuscript.