Angiotensin II Reduces Transport-Dependent Oxygen Consumption but Increases Transport-Independent Oxygen Consumption in Immortalized Mouse Proximal Tubular Cells

Abstract

Oxidative stress is closely associated with renal dysfunction following diabetes and hypertension. Angiotensin II (Ang II) can activate the NADPH- oxidase, increasing oxidative stress that is thought to blunt proximal tubular electrolyte transport and thereby oxygen consumption (QO₂). We investigated the effect of Ang II on QO₂ in immortalized mouse proximal tubular cells over-expressing the NADPH oxidase subunit p22^phox; a model of increased oxidative stress. Cultured cells were exposed to either Ang II or H₂O₂ for 48 h. QO₂ was determined during baseline (113 mmol/l NaCl; transport-dependent QO₂) and during sodium-free conditions (transport-independent QO₂). Ang II reduced transport-dependent QO₂ in wild-types, but not in p22^phox which also displayed increased QO₂ at baseline. Transport-independent QO₂ was increased in p22^phox and Ang II had no additional effect, whereas it increased QO₂ in wild-type. Addition of H₂O₂ reduced transport- dependent QO₂ in wild-types, but not in p22^phox. Transport-independent QO₂ was unaffected by H₂O₂. The similar effects of Ang II and H₂O₂ to reduce transport- dependent QO₂ suggest a direct regulatory role of oxidative stress. In accordance, the transport-dependent QO₂ was reduced in p22^phox already during baseline. The effects of Ang II on transport-independent QO₂ was not replicated by H₂O₂, indicating direct regulation via Ang II-receptors independently of oxidative stress. However, the Ang II effect was absent in p22^phox, suggesting that oxidative stress also modulates normal Ang II signaling. In conclusion, Ang II affects both transport-dependent and transport-independent QO₂ in proximal tubular cells and may be an important pathway modulating renal QO₂.

1 Introduction

Development of nephropathy due to diabetes and hypertension is strongly connected with concurrent development of renal hypoxia [1]. Renal hypoxia may develop with increased metabolic demand, renal anemia, or when renal peritubular capillary blood flow is decreased due to glomerular injury or vasoactive substances constricting the arterioles. Angiotensin II (Ang II), a vasoactive substance known to be increased in both hypertensive and diabetic kidneys [2, 3], induces oxidative stress and constriction of afferent as well as efferent arteriole [4]. Both these effects reduce renal blood flow and therefore reduce oxygen delivery, contributing to renal hypoxia. The metabolic demand by the kidney is largely determined by electrolyte reabsorption, accounting for 80 % of total kidney oxygen consumption (QO₂) [5]. Interestingly, the energy-demanding Na⁺/K⁺-ATPase-activity is increased in diabetes [6], increasing QO₂ and possibly limiting renal oxygenation. Diabetes-induced increased QO₂ and development of renal hypoxia are closely linked to increased levels of oxidative stress, as demonstrated by the prevention of renal hypoxia using antioxidants [7]. Interestingly, Ang II activates the NADPH oxidase via the Ang II receptor subtype 1 (AT1-R), resulting in increased superoxide formation [8] and several studies have suggested a role for NADPH-oxidase in the development of both hypertension and nephropathy [9, 10].

In summary, renal QO₂ is affected by mitochondrial activity, electrolyte transport and cellular QO₂, all processes that are altered by oxidative stress. However, the detailed pathways of Ang II-mediated effects on proximal tubular QO₂ and their relation to increased oxidative stress are presently unknown. Therefore, the present study separated the effects of Ang II on electrolyte transport-dependent QO₂ from those on transport-independent QO₂ in immortalized wild-type and p22^phox overexpressing mouse proximal tubular cells. The latter is a model of increased oxidative stress and was utilized to separate the effects of AT1-R signaling per se from those of Ang II-induced oxidative stress.

2 Material and Methods

Immortalized proximal tubular cells with and without a stable over-expression of the NADPH oxidase subunit p22^phox were maintained at 37°C and 5 % CO₂ in Dulbecco’s Modified Eagle Medium/F12 medium containing 5 % fetal bovine serum. At 50 % confluency, cell splitting was routinely performed using 2.5 % trypsin. Both cell types were grown with or without Ang II (10⁻⁷ mol/l, Sigma-Aldrich, St Louis, MO, USA replaced every 12 h) and H₂O₂ (2.5 × 10⁻⁵ mol/l, Sigma-Aldrich, St Louis, MO, USA replaced every 12 h) for 48 h. Before measurements, cells were placed in suspension with 2.5 % trypsin followed by triple rinsing by slow centrifugation (100 × g, 4 min, 4°C) in ice-cold respiration buffer (in mmol/l: glucose 23.2; NaCl 113; KCl 4.0; NaHCO₃ 27.2; KH₂PO₄ 1.0; MgCl₂ 1.2; CaCl₂ 1.0; HEPES 10.0; Ca²⁺-lactate 0.5; glutamine 2.0 and streptomycin 50 U/ml; osmolality 298 ± 2 mOsm checked with a freezing point osmometer, Fiske laboratories; pH 7.4) with or without Na⁺. In absence of NaCl and NaHCO₃, osmolality was kept constant with 280.4 mOsm mannitol. The cell-suspension was kept on ice until QO₂ was measured as described previously [11]. Briefly, a custom-made 1.1 ml gas-tight Plexiglas chamber thermostatically controlled at 37°C with continuous stirring from an air driven magnetic stirrer was used to determine QO₂ on free-floating cells in respiration buffer with or without Na⁺ to evaluate transport-dependent QO₂ and transport-independent QO₂ respectively. QO₂ was determined as rate of oxygen disappearance as measured by a modified Unisense-500 O₂-sensor (Unisense A/S, Aarhus, Denmark) calibrated with the air-equilibrated buffer solution as 228 μmol/l O₂ and Na₂S₂O₅-saturated buffer as zero, and normalized to protein concentration. Statistical analyses were performed using GraphPad Prism software (GraphPad Software Inc., San Diego, CA, USA). Multiple comparisons between groups were performed using analysis of variance (ANOVA) followed by Sidak multiple comparisons test. All data are presented as mean ± standard error of the mean (SEM) and p < 0.05 was considered statistically significant.

3 Results

Cells overexpressing p22^phox had reduced transport-dependent QO₂ compared to wild-type. Ang II reduced transport-dependent QO₂ in wild-type but not in p22^phox. During baseline conditions, transport-independent QO₂ was increased in p22^phox compared to wild-type. Ang II displayed only a tendency to decrease transport- dependent oxygen consumption in p22^phox whereas it increased transport- independent QO₂ in wild-type (Fig. 21.1). Addition of H₂O₂ reduced transport-dependent QO₂ in wild-type but only tended to reduce QO₂ in p22^phox. However, transport-independent QO₂ was unaffected by H₂O₂ in both cell types (Fig. 21.2).

Fig. 21.1.

Transport-dependent (left) and transport-independent oxygen consumption (QO₂, right) in immortalized wild-type mouse proximal tubular cells and corresponding cells overexpressing the NADPH oxidase subunit p22^phox and the effect of 48 h exposure to angiotensin II. Asterisk denotes p < 0.05 and data are presented as mean ± SEM

Fig. 21.2.

Transport-dependent (left) and transport-independent oxygen consumption (QO₂, right) in immortalized wild-type mouse proximal tubular cells and corresponding cells overexpressing the NADPH oxidase subunit p22^phox and the effect of 48 h exposure to H₂O₂. Asterisk denotes p < 0.05 and data are presented as mean ± SEM

4 Discussion

The present study demonstrates a role for Ang II in regulating transport-dependent as well as transport-independent QO₂ in mouse proximal tubular cells. It is likely that Ang-II decreases transport-dependent QO₂ by inducing oxidative stress, a hypothesis strengthened by the fact that the effect is mimicked by elevating oxidative stress via H₂O₂. Furthermore, the response to Ang II is reduced in cells with increased levels of oxidative stress during baseline, i.e. the p22^phox, further highlighting oxidative stress as a crucial component in the mechanism of Ang II.

However, as the effects on transport-independent QO₂ is not mimicked by H₂O₂ other mechanisms than oxidative stress may be involved.

It is known that Ang II increases superoxide formation by activating NADPH oxidase via AT-1R [8, 12]. Indeed, inhibition of AT-1R by olmesartan reduces oxidative stress independently of its blood pressure-lowering effect [13]. Several studies have suggested a role for NADPH oxidase in the development of hypertension and kidney disease [10, 14]. Oxidative stress is indeed increased in human hypertension [15] as well as in several hypertensive animal models [16]. AT₁-Receptor blockade with candesartan has been reported equally effective as the superoxide dismutase mimetic tempol in normalizing renal pO₂, implying that oxidative stress is directly involved [17]. Furthermore, inhibition of Ang II with angiotensin- converting enzyme (ACE) inhibitors and AT1-R blockers lowers renal QO₂ [18]. Ang II acting on AT₁-Receptors has been implied for the development in intrarenal hypoxia during hypertension, since 2 weeks of treatment with the AT1-R blocker candesartan to SHR rats normalize renal pO₂ [19]. Finally, Ang II-dependent hypertension in rats increases transport-dependent QO₂ in the thick ascending limb [20].

Cells with increased levels of oxidative stress displayed increased transport- independent QO₂. This may be due to mitochondrial uncoupling by uncoupling protein (UCP)-2, a phenomenon known to be induced during conditions of increased oxidative stress, resulting in increased QO₂ unrelated to ATP production [11, 21]. Importantly, diabetes-induced mitochondria uncoupling via UCP-2 is prevented by antioxidant treatment [22]. In the present study, addition of H₂O₂ to wildtype cells decreased transport-dependent oxygen consumption and did not affect transport- independent oxygen consumption, arguing against the presence of mitochondria uncoupling in these cells. However, it has been shown that mitochondria uncoupling is regulated by oxidative stress originating from the matrix side of the electron transport chain [23] and addition of H₂O₂ may therefore not be a strong enough signal to increase mitochondria uncoupling in cells with normal levels of oxidative stress. However, in cells with increased oxidative stress an increase in transport- independent oxygen consumption was evident, suggesting increased mitochondria uncoupling. Interestingly, the specific effect of Ang II on increasing transport- independent QO₂ may indeed be due to increased mitochondrial uncoupling. In a study by Doughan et al., addition of Ang II to cultured bovine aortic endothelial cells increased mitochondrial superoxide production and mitochondrial uncoupling [24]. Recently, Ang II receptors have been localized to the mitochondrial inner membrane [25]. In kidneys from SHR rats, oxidative stress was increased and the mitochondria displayed increased H₂O₂ generation, decreased membrane potential and increased UCP-2 expression [26]. Importantly, the effect of Ang II on transport- independent QO₂ was not replicated by increased oxidative stress per se, implying a crucial and specific role for AT1-R signaling. It is tempting to speculate that AT1-R located in the mitochondria are involved in these specific effects of Ang II. However, there was no effect of Ang II on transport-independent QO₂ in cells with increased levels of oxidative stress, demonstrating that oxidative stress may still have a regulatory role in mediating the observed effect on transport-independent QO₂ in wild-type cells. Reduced AT1-R expression or blunted receptor response in conditions with increased oxidative stress cannot be excluded.

In conclusion, the present study demonstrates that Ang II can affect both transport- dependent and transport-independent QO₂ in mouse proximal tubular cells and may be an important pathway modulating renal QO₂.