Role of Nox4 and p67phox subunit of Nox2 in ROS production in response to increased tubular flow in the mTAL of Dahl salt-sensitive rats

Abstract

Nox4 and Nox2 are the most abundant NADPH oxidases (Nox) in the kidney and have been shown to contribute to hypertension, renal oxidative stress, and injury in Dahl salt-sensitive (SS) hypertensive rats. The present study focused on the role of Nox4 and p67phox/Nox2 in the generation of H₂O₂ and O₂^·− in the renal medullary thick ascending limb of Henle (mTAL) of SS rats in response to increasing luminal flow (from 5 to 20 nl/min). Nox4 and p67phox/Nox2 genes were found to be expressed in the mTAL of SS rats. Responses of SS rats were compared with those of SS rats with knockout of Nox4 (SS^Nox4−/−) or functional mutation of p67phox (SS^{p67phox−/−}). Nox4 was the dominant source of increased intracellular H₂O₂ production in response to increased luminal flow as determined using the fluorescent dye peroxyfluor 6-AM (PF6-AM). The rate of mitochondrial H₂O₂ production [as determined by mitochondria peroxy yellow 1 (mitoPY1)] was also significantly reduced in SS^Nox4−/− compared with SS rats, but not in SS^{p67phox−/−} rats. In contrast, intracellular superoxide (O₂^·−) production (the ratio of ethidium to dihydroethidium) in the mTAL of SS^Nox4−/− rats was nearly identical to that of SS rats in response to luminal flow, indicating that Nox4 made no measurable contribution. mTAL O₂^·− production was reduced in SS^{p67phox−/−} compared with SS rats at the lower luminal flow of 5 nl/min and progressively increased when perfusion was changed to 20 nl/min. We conclude that increased mTAL luminal flow results in increases in intracellular and mitochondrial H₂O₂, which are dependent on the presence of Nox4, and that p67phox/Nox2 accounts solely for increases in O₂^·− production.

increased oxidative stress in the renal medulla induced by administration of H₂O₂ or the superoxide dismutase (SOD) inhibitor diethyldithiocarbamic acid (DETC) has been shown to result in salt-sensitive hypertension in Sprague-Dawley (SD) rats (27, 28, 44). Elevated levels of reactive oxygen species (ROS) have been observed in the renal medulla of Dahl salt-sensitive (SS) rats, and chronic medullary interstitial infusion of the antioxidants apocynin and/or catalase significantly reduced salt-sensitive hypertension in SS rats (45).

The sources and localization of ROS production within the kidney and the functional relevance of the various oxidative species have been of great interest but remain poorly understood. We have found that elevated ROS production within the renal outer medulla of SS rats can be largely attributed to the multicomponent NADPH oxidases (Nox) (45) that catalyze the reduction of molecular oxygen to superoxide (O₂^·−) and H₂O₂. Among the seven different members of the Nox family of enzymes, Nox2 (also named gp91phox) and Nox4 are expressed in the kidneys of rats (3, 5, 18, 21, 22, 30). Our data and findings of others indicate that Nox1 is expressed at very low levels in the medullary thick ascending limb of Henle (mTAL), and Nox5 has not been found in rodents (15, 21, 30). The p67phox cytosolic subunit [necessary for Nox2 enzyme activity (20)] and Nox4 have been found to be important determinants of salt sensitivity in SS rats based on observations that functional mutation of p67phox (Ncf2 gene) in SS (SS^{p67phox−/−}) (14) rats and knockout of Nox4 in SS (SS^Nox4−/−) rats (9) reduced salt-induced hypertension and renal injury of SS rats by 40–50%.

We previously characterized blood pressure salt sensitivity and renal function in SS^Nox4−/− and SS^{p67phox−/−} rats compared with SS rats (9, 14). SS^Nox4−/− and SS^{p67phox−/−} rats exhibited significant reductions (∼50%) in salt-induced hypertension, albumin excretion, glomerular injury, and tubular necrosis. A dynamic analysis of sequential changes in glomerular filtration rate (GFR) and in medullary blood flow during development of hypertension in unanesthetized SS^{p67phox−/−} rats compared with SS (wild-type) rats showed that SS^{p67phox−/−} rats were protected from the pattern of early reductions of medullary blood flow and proteinuria during the 1st wk of a high-salt diet and from the reductions of GFR observed after 2 wk of a 4% NaCl diet in SS rats (13).

We have been particularly interested in the mTAL, a segment of the nephron that reabsorbs ∼25% of the filtered Na⁺ and, thus, is important in NaCl and water homeostasis. The mTAL of SS rats has been found to transport and reabsorb excess amounts of NaCl compared with the salt-resistant Dahl R rats (24, 39) and produces excess ROS compared with salt-resistant control strains of rats (33). We and others have found that increased tubular flow and luminal Na⁺ content stimulate ROS production in the mTAL of normotensive SD rats (8, 17). The present study was designed to determine the extent to which Nox4 and p67phox contribute to total intracellular O₂^·− and H₂O₂ production and to mitochondrial H₂O₂ production in the mTAL of SS rats in the face of increased luminal flow and Na⁺ delivery. Selective fluorescent dyes were used to determine changes within whole cell O₂^·− and H₂O₂ and mitochondrial H₂O₂ of microperfused mTAL, as we previously described (32, 36). Our findings indicate that Nox4 contributes to mTAL cellular and mitochondrial H₂O₂ production and is not involved in intracellular O₂^·− production in response to increased luminal flow in SS rats. The p67phox subunit contributes to a lesser extent to the increased cellular H₂O₂ and is primarily responsible for basal levels of O₂^·− production.

MATERIALS AND METHODS

Experimental animals.

Male Dahl SS rats [SS rat; Medical College of Wisconsin (MCW), Milwaukee, WI], SS^Nox4−/− rats, and SS^{p67phox−/−} rats were obtained at weaning from colonies developed and maintained at the MCW under controlled environmental conditions. Parents and offspring were fed a purified AIN-76A rodent diet (Dyets, Bethlehem, PA) containing 0.4% NaCl until the experimental period, during which they were fed the 4.0% NaCl (high-salt) diet (Dyets); water was provided ad libitum. Rats were studied at 7–8 wk of age and 180–250 g body weight. All experimental protocols were approved by the MCW Institutional Animal Care and Use Committee. Generation of the knockout strains using zinc finger nuclease technology has been previously described for the SS^{p67phox−/−} (14) and SS^Nox4−/− (9) rats. For the RNA studies, rats were maintained on the 0.4% NaCl diet from weaning or switched to the 4.0% NaCl diet for 7 days prior to collection of tissue. For the microscopy studies, rats were maintained on the 4.0% NaCl diet for 3 days prior to tubule isolation.

Modified enzymatic isolation of nephron segments (mTALs, glomeruli, proximal tubules, and outer medullary collecting ducts) for RNA analysis.

Rats were anesthetized with pentobarbital sodium (50 mg/kg), and the kidneys were cleared of blood by retrograde infusion with 10 ml of cold saline solution. An equal volume of a collagenase digestion solution [Hanks' balanced salt solution (HBSS) containing 20 mM HEPES (HBSS-H), pH 7.4, with collagenase type 2 (200 U/ml)] was infused at 4 ml/min, and the kidneys were removed, decapsulated, and maintained at 4°C. The outer medulla and cortex tissues were cut into small pieces and incubated in separate tubes with 1–2 ml of collagenase digestion solution at 37°C for 20 min in a hybridization oven, allowing for continuous rotation of the contents of the tubes to facilitate digestion. The digested tissue formed a gradient, with the heaviest and most undigested tissue settling on the bottom of the tube and the middle supernatant containing the nephron components already separated. The heavier, undigested tissue remaining in the original tube was further digested by addition of an equal volume of collagenase solution and incubation at 37°C. The first supernatant after the initial 20-min digestion of the cortex sample was used to isolate glomeruli. The second and third supernatants after 30 and 40 min of digestion of the cortical sample were used to isolate proximal tubules (PTs), which were mixed together. The first, second, and third supernatants (20, 30, and 40 min of digestion) of the outer medulla tissue were used to collect medullary thick ascending limb (mTAL) and mixed together. The fourth and fifth digestions of the outer medullary sample (60 and 70 min of incubation at 37°C) were used to collect the outer medullary collecting ducts (OMCDs) and mixed together. The supernatant from each digestion was removed and centrifuged at 100 g for 5 min at 4°C, and the pellet was washed twice with 1% BSA-HBSS-H by centrifugation. The final pellet was resuspended in 1% BSA-HBSS-H, and this suspension was placed in a small petri dish and visualized using a dissecting stereomicroscope (model M3Z, Leica) for manual separation and segregation of the nephron elements. OMCDs, mTALs, PTs, and glomeruli were isolated from the cortex and the outer medulla using this method and transferred to individual tubes that were then frozen in liquid nitrogen and stored at −80°C until RNA extraction. Photomicrographs of the four components of the nephron that were isolated using this digestion and isolation method are shown in Fig. 1A.

Fig. 1.

A: photomicrographs at ×4 and ×10 magnification show nephron components from kidneys of salt-sensitive (SS) rats following digestion and manual isolation prior to RNA extraction. mTAL, medullary thick ascending limb; PT, proximal tubule; OMCD, outer medullary collecting duct. B: summary of marker gene expression of the nephron components from RNA-seq analysis. Slc12a1 gene [encodes NKCC2 (Na⁺-K⁺-Cl⁻ cotransporter)] was used as a marker of mTAL (16, 26, 47). Nphs2 (podocin), which is almost exclusively expressed in podocytes of fetal and mature kidney glomeruli, was used as a marker of glomeruli (6). Aqp1 (aquaporin-1 water channel protein) was used as a marker of PT (46). Aqp2 (aquaporin-2 collecting duct water channel protein) was used as a marker for the OMCD (10). Cspg4 [a chondroitin sulfate proteoglycan 4, also known as neuron-glial antigen 2 (NG2)] was used as a pericyte marker (vasa recta) (37). All values are expressed as fragments per kilobase of transcript per million mapped reads.

RNA sequencing.

TRIzol was used to extract RNA from each of the nephron segment samples, and the quality of the extracted RNA was determined by a bioanalyzer (model 2100, Agilent). Six rats with SS genetic background were used to obtain six RNA-seq libraries using ∼100 ng of total RNA pooled from two samples of each segment using an Ilumina TruSeq Stranded mRNA LT kit. Three libraries were prepared from rats fed the 0.4% NaCl diet and three from rats fed the 4.0% NaCl diet for 7 days. The libraries underwent cluster generation using the TruSeq PE Cluster Kit v3-cBot-HS and 100 cycles of paired-end sequencing using the TruSeq SBS Kit v3-HS. These libraries were multiplexed on one lane of a 300-Gb flow cell and sequenced using an Illumina HiSeq2000 sequencer, as previously described (19, 23). The same amount of RNA for RNA sequencing was used for all four segments (100 ng of total RNA) to enable a relative comparison of these segments. Purity of nephron segments is shown in Fig. 1B. The expression of segment markers was determined from RNA-seq data in the four isolated components of the nephron from rats with SS genomic background fed the 0.4% NaCl diet and showed that contamination of tubules with other segments was low (Fig. 1B).

Reagents and solutions.

HBSS (catalog no. 14025) was obtained from Gibco; HEPES (catalog no. H4034), BSA (catalog no. A9647), Tween 20 (catalog no. P7949), DETC (catalog no. 228680), and menadione (catalog no. M5750) from Sigma; collagenase type 2 (catalog no. LS004174) from Worthington Biochemical; Cell-Tak from BD Biosciences (Bedford, MA); and dihydroethidium (DHE; catalog no. D11347) from Molecular Probes. Mitochondria peroxy yellow 1 (mitoPY1) and peroxyfluor 6 (PF6)-AM were generously provided by Dr. Bryan Dickinson and Dr. Christopher Chang.

Nonenzymatic isolation of rat renal mTAL for fluorescence microscopy.

Rats were switched from the 0.4% to the 4.0% NaCl diet for 3 days before collection of the kidneys for the microscopy studies. On the day of study, the kidneys were flushed with 10 ml of chilled (4°C) HBSS-H (pH 7.4) and then hemisected to expose the inner strip of the outer medulla, which was removed, and the mTAL was isolated by microdissection at 4°C using the Leica M3Z stereomicroscope, as previously reported (36). The isolated mTAL tissue strip was placed on a glass coverslip coated with the tissue adhesive Cell-Tak in HBSS-H for fluorescence imaging. Several coverslips were prepared from each rat.

Microperfusion of mTAL.

Glass pipettes (Narishige) were pulled to an internal diameter of 10–15 μm, the tips were beveled and smoothed, and the pipettes were secured on a micromanipulator (World Precision Instruments, Sarasota, FL) mounted on the microscope stage. The micropipette tip was inserted into the open lumen of the mTAL, and the tubule was perfused with the HBSS-H solution with the flow rate controlled using a nano pump (model A1400, World Precision Instruments).

Fluorescence microscopy for determination of H₂O₂ responses to change in luminal flow.

As we described previously (36), intracellular and mitochondrial H₂O₂ levels in mTALs were determined by incubation of tubules with 5 μM PF6-AM (for intracellular H₂O₂) (11) or 5 μM mitoPY1 (for mitochondrial H₂O₂) (12) in HBSS-H for 30 min at 37°C. The mTALs were washed three times with HBSS-H to remove excess dye, and the coverslips were placed on a heated imaging chamber (Warner Instruments) maintained at 37°C. Several coverslips could be prepared from each kidney to permit different treatments of mTALs from the same rat, including nonperfused (no flow) time controls in each case. The number given for each experimental measurement corresponds to a separate rat in all cases. PF6-AM and mitoPY1 fluorescence images were obtained using a Nikon TE-2000U inverted microscope equipped with a 60/1.1 water immersion objective lens and a high-resolution digital camera (Photometrics Cascade 512B, Roper Scientific, Tucson, AZ) (1). Excitation was provided by a 175-W xenon arc lamp (model DG-4, Sutter Instrument, Novato, CA) at alternating wavelengths, and emission was controlled using an optical filter changer (Lambda 10-3, Sutter Instrument). PF6-AM and mitoPY1 were excited at 480/40 nm, and emission signals from 510/30 and 530/60 nm, respectively, were acquired every 1 min at the 5 nl/min flow rate and every 3 min at the 20 nl/min flow rate. PF6-AM and mitoPY1 signals were normalized by subtraction of the first time point value from each data point. A positive control stimulus was applied at the end of the experiment by administration of H₂O₂ (100 μM) to confirm that the tubule was viable and responsive and could be included in the analysis.

Fluorescence microscopy for determination of O₂^·− responses to changes in flow.

Tissue strips containing mTAL were loaded with DHE (50 mmol/l) in HBSS-H for 15 min at room temperature. The mTALs were washed three times for 5 min each with HBSS-H to remove excess dye, and the coverslips were placed on a heated imaging chamber maintained at 37°C. Fluorescence intensity of each image was quantified over an area of ∼10–15 mTAL cells using MetaFluor imaging software (Universal Imaging, Downingtown, PA). Fluorescence from regions of interest was measured every 10 s for 150 s at flow rates of 5 and 20 nl/min. The rate of O₂^·− production was determined as a change in ratio of the ethidium (Eth) fluorescent signal to the DHE fluorescent signal (Eth/DHE) in mTALs perfused at flow rates of 5–20 nl/min and nonperfused mTALs (no flow). We have found that using Eth/DHE to determine O₂^·− production in mTAL reduces measurement artifacts associated with cell volume changes and dye bleaching (31). A 445/40-nm and a 605/55-nm band-pass emission filter were used to collect DHE (380/40X–445/40E) and Eth (480/40X–605/55E), signals, respectively (35). Background Eth and DHE fluorescence signals were subtracted from the average intensity at all the regions of interest containing mTAL epithelial cells. We normalized the data by subtracting Eth/DHE value at the first time point from each data point. At the end of the experiment, the SOD inhibitor DETC (1 mM) and menadione (500 mM), a stimulant of O₂^·− production, were administered as a positive control for dye loading and cell viability. Regression analysis of the fluorescence ratios was performed, and differences in slopes (P_interaction) were evaluated using two-way repeated-measures (RM) ANOVA (150 s of recording at 5 and 20 nl/min, total 15 measurements for each mTAL flow rate).

Detection of total O₂^·− and NADPH oxidase activity by 2-hydroxyethidium fluorescence.

Outer medullary homogenate protein (100 μg) isolated from SS and SS^Nox4−/− rats was incubated with DHE (20 mmol/l), with salmon testes DNA (0.5 mg/ml), and with or without 100 mmol/l diphenylene iodonium (Sigma), an inhibitor of NADPH oxidase. Oxyethidium fluorescence was measured at an excitation of 485 nm and an emission of 570 nm on a SpectraFluor microplate reader (Tecan) every 5 min for 35 min at 37°C (45). The maximal increase in fluorescence within 35 min was used as an index of total O₂^·− level. The portion of oxyethidium fluorescence after diphenylene iodonium inhibition was used as an index of NADPH oxidase activity.

Statistical analysis.

Values are means ± SE. Two-way RM ANOVA followed by a Holm-Sidak test (2-way RM ANOVA, SigmaPlot 12.0) was used to assess differences between the strains and between perfused and nonperfused mTALs within each strain. P_interaction (obtained from 2-way RM ANOVA) was used to evaluate the differences in slopes. For NADPH activity measurements, significance was evaluated using a paired t-test. P < 0.05 was considered significant.

RESULTS

Nox4, p67phox, Nox1, Nox2, and Nox3 RNA expression in tubular segments and glomeruli of SS rats.

The average expression levels of Nox4, p67phox, Nox2, Nox1, and Nox3 of the isolated tubular segments and glomeruli of rats with SS genomic background fed the 0.4% NaCl diet or the 4.0% NaCl diet for 7 days are summarized in Table 1. It is evident from this RNA-seq analysis that Nox4 and Nox2 (Cybb) genes are expressed in the mTALs, OMCDs, and glomeruli of these rats and at quantitatively similar levels. Nox4 mRNA was expressed at ∼65-fold greater levels in the PTs than the mTALs of these rats, the relevance of which remains to be determined. The p67phox (Ncf2) mRNA was expressed in all tubular segments but at lower levels of expression than Nox2 (Cybb). Nox1 mRNA was expressed in the mTALs at levels only one-sixth of Nox2, but we cannot exclude Nox1 as a potential ROS generator. Nox3 was not detectable in any of the nephron segments or the glomeruli.

Table 1.

Gene expression of Nox1, Cybb (Nox2), Nox3, Nox4, and Ncf2 (p67phox) in isolated components of the nephron from rats with SS genomic background fed 0.4% and 4% NaCl for 7 days

	Nephron Segments
Gene Name	mTAL	Glo	PT	OMCD
0.4 % NaCl
Nox4	3.6 ± 0.6	3.0 ± 1.7	210.4 ± 75.5	1.1 ± 0.3
Ncf2 (p67phox)	0.2 ± 0.0	0.8 ± 0.1	0.2 ± 0.1	0.3 ± 0.1
Cybb (Nox2)	3.5 ± 0.3	9.9 ± 0.9	2.6 ± 0.3	3.3 ± 0.4
Nox1	0.7 ± 0.1	0.5 ± 0.0	0.6 ± 0.1	1.1 ± 0.1
Nox3	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0
4 % NaCl for 7 days
Nox4	3.0 ± 0.1	3.1 ± 1.2	155.9 ± 52.6	2.0 ± 0.4
Ncf2 (p67phox)	0.5 ± 0.1	0.8 ± 0.1	0.4 ± 0.1	0.3 ± 0.1
Cybb (Nox2)	5.6 ± 0.6	11.1 ± 0.2	6.1 ± 1.0	4.4 ± 0.8
Nox1	0.6 ± 0.0	0.4 ± 0.1	0.7 ± 0.1	1.2 ± 0.1
Nox3	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0	0.0 ± 0.0

Values (means ± SE) are expressed as FPKM (fragments per kilobase of transcript per million mapped reads). Gene expression was determined by RNA-seq in isolated components of the nephron from rats with salt-sensitive (SS) genomic background. Glo, glomerulus; mTAL, medullary thick ascending limb of Henle; OMCD, outer medullary collecting duct; PT, proximal tubule.

Comparison of total intracellular H₂O₂ production in mTALs isolated from SS, SS^Nox4−/−, and SS^{p67phox−/−} rats in response to increased tubular flow.

Intracellular mTAL H₂O₂ responses (PF6-AM fluorescence) to increased luminal flow (from 5 to 20 nl/min) were compared in SS, SS^Nox4−/−, and SS^{p67phox−/−} rats fed the 4.0% NaCl diet for 3 days. Nonperfused mTALs of each strain were used as time controls. Figure 2A shows that mTALs of SS rats (n = 12) responded with significant elevation of intracellular H₂O₂ as luminal flow was increased compared with nonperfused mTALs (slopes were statistically different, P_interaction < 0.01). In contrast, increases in intracellular H₂O₂ were absent in mTALs isolated from mutant SS^Nox4−/− rats (n = 8) and did not differ significantly from the nonperfused control mTALs (Fig. 2B). The mTALs of SS^{p67phox−/−} rats (n = 8) exhibited reduced H₂O₂ responses compared with SS rats but not to the extent seen in SS^Nox4−/− mTALs and also were not significantly different from control nonperfused mTALs (Fig. 2C). These differences are more apparent in Fig. 2D, which summarizes the mTAL H₂O₂ responses to luminal flow in the three rat strains. Statistical comparisons of mean data at each time point at 5 and 20 nl/min between SS and SS^{p67phox−/−} mTALs show that H₂O₂ production was reduced in the absence of p67phox (SS^{p67phox−/−} rats) after 20 min of 20 nl/min flow, despite no significant differences in slopes between the two strains (P_interaction = 0.2), indicating that Nox2 probably does contribute to the overall cellular increases in H₂O_2. The H₂O₂ production of mTALs from SS^Nox4−/− rats was significantly reduced compared with SS and SS^{p67phox−/−} rats whether the mean data at each time point or slopes (P_interaction < 0.01) were compared. This total elimination of the H₂O₂ response to flow in mTALs isolated from SS^Nox4−/− rats indicates that Nox4 was the major source of H₂O₂ production.

Fig. 2.

A–C: intracellular H₂O₂ responses [change in peroxyfluor 6 (PF6)-AM (PF6-AM) intensity] to an increase in mTAL luminal flow from 5 to 20 nl/min in SS (n = 12), SS^Nox4−/− (n = 8), and SS^{p67phox−/−} (n = 8) rats fed 4.0% NaCl diet for 3 days prior to study (■, ●, and ▲). □, ○, and △, Fluorescence changes of nonperfused mTALs. *Significant differences between slopes of perfused and nonperfused mTALs within each strain [2-way repeated-measures (RM) ANOVA, P_interaction < 0.05]. D: summary of responses to increased luminal flow in mTALs of SS (■), SS^Nox4−/− (●), and SS^{p67phox−/−} (▲) rats. †Significant differences between slopes of SS and SS^Nox4−/− rats; #significant differences between slopes of SS^Nox4−/− and SS^{p67phox−/−} rats (2-way RM ANOVA, P_interaction < 0.05).

mTAL luminal flow induced mitochondrial H₂O₂ production in SS, SS^Nox4−/−, and SS^{p67phox−/−} rats.

The mitochondria H₂O₂-specific fluorescent dye mitoPY1 was applied to measure responses to increased luminal flow in mTALs of rats fed the 4.0% NaCl diet for 3 days prior to the study. As summarized in Fig. 3, A and C, mitoPY1 fluorescence intensity significantly increased in the mTALs of SS and SS^{p67phox−/−} rats in response to increased luminal perfusion compared with the nonperfused control mTALs (slope differences, P_interaction < 0.05). However, intracellular H₂O₂ production did not differ significantly between perfused and nonperfused mTALs isolated from mutant SS^Nox4−/− rats (n = 8; Fig. 3B). Mitochondrial responses of mTALs isolated from SS, SS^Nox4−/−, and SS^{p67phox−/−} rats relative to flow are summarized in Fig. 3D. At a luminal perfusion of 5 nl/min, the slopes were not statistically different between the rat strains. When luminal flow was increased to 20 nl/min, the rate of increase in mitochondrial H₂O₂ production was reduced in SS^Nox4−/− compared with SS rats (P_interaction = 0.015), which was apparent only after 20 min. While the mitochondrial H₂O₂ production tended to be greater in SS^{p67phox−/−} than SS rats, these differences were not statistically significant in terms of slopes or mean data at each time point (Fig. 3D).

Fig. 3.

A–C: mitochondrial H₂O₂ responses [change in mitochondria peroxy yellow 1 (mitoPY1) intensity] to increased mTAL luminal flow from 5 to 20 nl/min in SS (n = 10), SS^Nox4−/− (n = 6), and SS^{p67phox−/−} (n = 7) rats fed 4.0% NaCl diet for 3 days prior to study (■, ●, and ▲). □, ○, and △, Fluorescence changes of nonperfused mTALs. *Significant differences between slopes of perfused and nonperfused mTALs within each strain (2-way RM ANOVA, P_interaction < 0.05). D: summary of responses to increased luminal flow in mTALs of SS (■), SS^Nox4−/− (●), and SS^{p67phox−/−} (▲) rats. †Significant differences between slopes of SS and SS^Nox4−/− rats; #significant differences between slopes of SS^Nox4−/− and SS^{p67phox−/−} rats (2-way RM ANOVA, P_interaction < 0.05).

Intracellular O₂^·− production in mTALs isolated from SS, SS^Nox4−/−, and SS^{p67phox−/−} rats in response to increased tubular flow.

Eth/DHE fluorescence responses were used to evaluate intracellular changes of O₂^·− in mTALs. The rate of O₂^·− production (slopes, P_interaction) was significantly different between mTALs perfused at luminal flow of 5 and 20 nl/min and nonperfused mTALs of each strain (Fig. 4, A–C). Only mTALs of SS^{p67phox−/−} rats did not show significant differences in slopes between mTALs perfused at 5 nl/min and nonperfused mTALs (Fig. 4C). Figure 4D shows that O₂^·− production (Eth/DHE) in mTALs of SS rats increased continuously throughout the 5 nl/min perfusion period and that although O₂^·− production continued to increase when luminal perfusion was changed to 20 nl/min, the rate of O₂^·− production (slope) was not further increased. Importantly, this response was nearly identical in mTALs of SS^Nox4−/− and SS rats (Fig. 4D), indicating that Nox4 did not contribute to the mTAL O₂^·− production. Consistent with these observations, NADPH oxidase activity determined from homogenates of outer medullary tissue of SS^Nox4−/− rats fed the 4.0% NaCl diet for 21 days showed no significant differences in total enzyme activity compared with SS rats (n = 5 for each strain). The rate of O₂^·− production was significantly lower in mTALs of SS^{p67phox−/−} than SS and SS^Nox4−/− rats (slope differences, P_interaction < 0.001; Fig. 4D) at the lower rates of luminal perfusion (5 nl/min). However, the response of SS^{p67phox−/−} rats to 20 nl/min flow was similar to that of SS and SS^Nox4−/− rats (slopes were not significantly different; Fig. 4D). We conclude that Nox4 makes no measurable contribution to the tubular flow-induced increase in O₂^·− production, while p67phox/Nox2 contributes significantly to the basal production of O₂^·−, at low (5 nl/min), but not high (20 nl/min), rates of luminal flow.

Fig. 4.

Intracellular superoxide (O₂^·−) production [ethidium-to-dihydroethidium (Eth/DHE) ratio] in response to increase in mTAL luminal flow from 5 to 20 nl/min in SS (n = 9), SS^Nox4−/− (n = 7), and SS^{p67phox−/−} (n = 8) rats fed 4.0% NaCl diet for 3 days prior to study (■, ●, and ▲). □, ○, and △, nonperfused mTALs. *Significant differences between slopes of perfused and nonperfused mTALs within each strain (2-way RM ANOVA, P_interaction < 0.05). D: summary of responses to increased luminal flow in mTALs of SS (■), SS^Nox4−/− (●), and SS^{p67phox−/−} (▲) rats. &Significant differences between slopes of SS and SS^{p67phox−/−} rats at 5 nl/min flow; #significant differences between slopes of SS^Nox4−/− and SS^{p67phox−/−} rats at 5 nl/min flow (2-way RM ANOVA, P_interaction < 0.05).

DISCUSSION

The present studies examined the contribution of Nox4 and the p67phox cytosolic subunit of Nox2 to ROS production in tubular epithelial cells of the mTAL in response to increased luminal flow and Na⁺ delivery in SS rats. Excess Na⁺ reabsorption and elevated ROS production in the mTAL and renal outer medulla are associated with blood pressure salt sensitivity of the SS rat (9, 14, 27, 28). In this study we confirmed that Nox4 and Nox2 are the most abundantly expressed Nox isoforms in the rat kidney (Table 1), as previously reported (4, 21, 30), and importantly we determined their functional contributions to ROS production in the mTAL. Development of the SS^Nox4−/− and SS^{p67phox−/−} rat models has enabled us to overcome the limitations of nonselective inhibitors and the limited in vivo efficiency of mRNA antisense oligonucleotides and allowed us to begin exploring these relationships in the salt-sensitive model of hypertension.

Nox4 and p67phox/Nox2 contributions to mTAL cellular and mitochondrial H₂O₂ elevations in response to increased tubular flow.

Our results indicate that Nox4 is the dominant source of intracellular production of H₂O₂ in response to physiological elevations of luminal flow (from 5 to 20 nl/min). This was apparent given the complete elimination of the H₂O₂ responses to increased luminal flow in mTALs of SS^Nox4−/− rats (Fig. 2D). Although SS^{p67phox−/−} rats tended to exhibit a reduced cellular H₂O₂ production compared with SS rats, the slope differences were not statistically significant. The contribution of Nox4 and p67phox/Nox2 to mitochondrial H₂O₂ production in the SS rat has not been studied. The rate of mitochondrial H₂O₂ production (slope) was significantly lower in mTALs of SS^Nox4−/− than SS rats, as became apparent after 20 min of luminal perfusion at 20 nl/min (Fig. 3D). This finding was in contrast to mTALs of SS^{p67phox−/−} rats, in which the rate of mitochondrial H₂O₂ production tended to be even greater than in SS rats, although it was not significantly different.

Localization of Nox4 has been elusive given the limitations of immunohistochemical methods and lack of specificity of Nox4 antibodies, but studies have suggested that Nox4 may be present in mitochondria and contribute to ROS formation (2, 7, 25). The results of the present study could be consistent with the presence of Nox4 in the mitochondria, which was stimulated by increased mTAL luminal flow or Na⁺ delivery to produce H₂O₂. However, we were previously unable to detect the presence of Nox1, Nox2, or Nox4 in a proteomic analysis of mitochondria isolated from the mTAL nephron segment of SS rats (48). We therefore speculate that the reductions in the rate of mitochondrial H₂O₂ production in SS^nox4−/− rats were in part due to 1) electron transport chain activity of complexes I and III perhaps due to reduced mTAL Na⁺ flux (36) and 2) reduced H₂O₂ production of membrane Nox4 and, thereby, less diffusion of H₂O₂ to the mitochondria. The kinetics of such responses could also explain why ∼20 min were required to detect the reduction in the rate of increase in mitochondrial H₂O₂ in SS^nox4−/− rats.

We observed previously (33) that apocynin pretreatment of mTALs isolated from the SD rat had no effect on mitochondrial H₂O₂ responses (mitoPY1 fluorescence) to increased tubular flow (from 5 to 20 nl/min). Since apocynin inhibits the translocation of cytosolic subunits of Nox1 and Nox2 required for their activation (38, 42), which are not required for Nox4, these data indicate that H₂O₂ produced from Nox1 or/and Nox2 in the mTAL did not feed-forward to stimulate mitochondrial H₂O₂ production (36). However, since Nox4 is constitutively active and its activation does not require cytosolic proteins, it is possible that H₂O₂ produced from membrane Nox4 could diffuse into the mitochondria (feed-forward) and enhance the measured rate of rise of mitochondrial H₂O₂ production in SS rats.

Independent of mitochondrial responses, it is curious that the knockout of Nox4 (SS^Nox4−/−) completely eliminated any increases of cytosolic H₂O₂ in response to increased tubular flow. Nox2 remains present in the mTALs of SS^Nox4−/− rats, although at reduced expression levels, as we have reported (9), but one could expect continued O₂^·− and H₂O₂ production. An absence of any increase in cellular H₂O₂ with increased luminal flow was not expected and suggests some type of interaction between Nox4 and Nox2. It is possible that an additive response from the knockout of Nox4 together with reduction of p67phox/Nox2 expression in SS^Nox4−/− rats pushed cytoplasmic H₂O₂ below detection limits of the PF6-AM fluorescence probe.

We previously showed in isolated mTALs of SD rats an increase in both cytoplasmic H₂O₂ (PF6-AM fluorescence) and mitochondrial H₂O₂ (mitoPY1 fluorescence) when tubular flow was increased from 5 to 20 nl/min (36). Cytoplasmic H₂O₂ responses were significantly reduced by pretreatment with the electron transport chain inhibitors rotenone and antimycin A and by inhibition of mTAL Na⁺ transport with furosemide or ouabain. It was evident from these studies that mitochondrial H₂O₂ production was determined by mTAL Na⁺ transport and that mitochondrial H₂O₂ production made an important contribution to the overall increases in cellular H₂O₂ in response to luminal flow. It is likely that similar events contributed to the overall increases in intracellular H₂O₂ in the SS rats.

Nox4 and p67phox/Nox2 contributions to mTAL cellular O₂^·− responses to increased tubular flow.

In contrast to the observed H₂O₂ responses, mTAL production of O₂^·− was not measurably affected by the absence of Nox4, as seen by the nearly identical mTAL Eth/DHE responses of the SS and SS^Nox4−/− rats (Fig. 4D). This finding is consistent with evidence that Nox4 produces H₂O₂, rather than O₂^·−, which is a unique feature of Nox4 and is thought to be a consequence of a highly conserved histidine residue in the E-loop of Nox4 that promotes the rapid dismutation of O₂^·− within the protein itself (41, 43).

As might be anticipated, SS^{p67phox−/−} rats exhibited significantly lower rates of mTAL O₂^·− production than SS and SS^Nox4−/− rats. Yet, this difference was observed only at the low luminal perfusion rate of 5 nl/min, and as luminal perfusion was increased to 20 nl/min, the rate of change in O₂^·− production (slope, P_interaction) in SS^{p67phox−/−} rats was similar to that in SS and SS^Nox4−/− rats. We believe that these observations are likely explained by the high concentrations of Na⁺ (149 mM) in the luminal perfusate used in our studies. We have found in SD rats that at lower mTAL luminal Na⁺ concentrations of 60 mM, increases in luminal flow from 5 to 20 nl/min resulted in a sharp rise of intracellular O₂^·− production, whereas tubules perfused at 149 mM Na⁺ exhibited a progressive rise of O₂^·− production, even at a luminal perfusion rate of 5 nl/min (1). Importantly, the rate of rise of O₂^·− production (Eth/DHE slopes) was not further influenced by increases in luminal perfusion to 20 nl/min, indicating that the Na⁺ transport stimulus was already driving the O₂^·− production to levels that could not be further enhanced by luminal flow. Since the present studies perfused the mTAL at luminal Na⁺ concentrations of 149 mM and the O₂^·− responses (Eth/DHE) were very similar to those observed in SD rats (1) and the O₂^·− responses of the SS^{p67phox−/−} rats mimicked those of SD rats perfused at lower Na⁺ concentration (60 mM), these data suggest reduced rates of Na⁺ transport in the mTALs of SS^{p67phox−/−} rats, which reduced the basal level of O₂^·− production. In support of this idea are observations of reduced blood pressure salt sensitivity and reduced levels of O₂^·− production in the outer medullary tissue and reduced outer medullary renal injury in SS^{p67phox−/−} rats (14, 40). We recognize that the present observations are not consistent with observations by others who concluded that Nox2 was not involved in luminal flow-mediated O₂^·− production (Eth/DHE) in Nox2^−/− mice and found that transduction of mTALs isolated from SD rats with a Nox4 siRNA adenovirus inhibited flow-stimulated O₂^·− production (21). While it is difficult to reconcile these differences, many laboratories have found that H₂O₂, and not O₂^·−, is the prominent product of Nox4 (5, 29, 34, 43), and we believe that the SS^Nox4−/− rat model system has meaningfully addressed the contributions of Nox4.

Conclusion.

We conclude that Nox4 is responsible for the increases in intracellular and, possibly, mitochondrial H₂O₂ observed with physiological increases in mTAL luminal flow but does not contribute to the increases in intracellular O₂^·−. In contrast, p67phox/Nox2 contributes little to the elevations of H₂O₂ and is responsible only for flow-mediated increases in intracellular O₂^·−.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants HL-116264 (A. W. Cowley, Jr.), HL-82798 (A. W. Cowley, Jr.), and HL-122662 (A. W. Cowley, Jr.).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

N.Z. and C.Y. performed the experiments; N.Z. and C.Y. analyzed the data; N.Z., C.Y., and A.W.C. interpreted the results of the experiments; N.Z. and C.Y. prepared the figures; N.Z. and A.W.C. drafted the manuscript; N.Z., C.Y., and A.W.C. edited and revised the manuscript; N.Z., C.Y., and A.W.C. approved the final version of the manuscript; A.W.C. developed the concept and designed the research.