Oxygen is an essential molecule for the sustenance of aerobic organisms. Reactive oxygen species (ROS), including superoxide radicals (O₂ ^•−), hydrogen peroxide (H₂O₂), hydroxyl radicals (^•OH), and singlet oxygen (¹O₂), are constantly released when oxygen is consumed to produce adenosine triphosphate (ATP). ROS are harmful to cells and tissues, so specific biological systems remove and detoxify them. A proper balance between ROS production and removal is required to maintain normal biological conditions. Oxidative stress is defined as a ROS level increase and the damage it causes. It is caused by an imbalance between ROS production and removal and is associated with a variety of diseases, such as cancers, cardiovascular diseases, neurological diseases, and infertility [1–5]. The concept of oxidative stress and its components are well known, and recent successes with ROS-oriented therapeutic approaches have attracted the attention of both basic and clinical researchers. Persistent oxidative stress is known to play a crucial role in the pathogenesis of acute kidney injury (AKI) and to be involved in the AKI-to-chronic kidney disease (CKD) transition [6,7].

Because oxygen has an oxidative effect, hyperoxia causes oxidative stress. In clinical settings, oxygen therapy must only be administered to patients for whom an oxygen supplement is safe because hyperoxia often causes lung injury [8]. Hypoxia, paradoxically, also induces oxidative stress through increased ROS production, superoxide radicals (O₂ ^•−) in particular, by oxygen-dependent enzymes such as cytochrome c oxidase, nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, and uncoupled endothelial nitric oxide synthase [9]. Given that the kidney is the organ most susceptible to hypoxia and that renal hypoxia usually occurs during kidney disease progression, as shown below, hypoxia, and not hyperoxia, must play an important role in the aggravation of oxidative stress in CKD [10–13]. This review examines recent evidence for the pathogenic role of chronic hypoxia and oxidative stress in renal diseases and describes promising therapeutic approaches against oxidative stress.

Although renal blood flow accounts for as much as 20% of the cardiac output, renal tissue oxygen tensions are comparatively low due to oxygen shunt diffusion between the arterial and venous vessels, which run parallel to each other in the kidney [10]. For example, the partial pressure of oxygen was reported to be higher in the renal vein than at any site in the efferent arterioles or the cortex, which implies precapillary oxygen shunting from the arteries to the veins inside the kidney [14]. The existence of a decreasing oxygen gradient with respect to the depth from the renal surface (i.e., 10–20 mmHg in the renal medulla vs. 50 mmHg in the renal cortex), also supports the presence of oxygen shunt diffusion because renal arterial flow spreads from the renal pedicle to the renal surface [15]. This oxygen shunt makes the kidney the organ most susceptible to systemic hypoxia; microvascular oxygen pressure in the kidney starts to decrease with a smaller hematocrit drop than required for other organs, even in normal kidneys [16]. In CKD, tubulointerstitial hypoxia is caused by multifactorial mechanisms: loss of peritubular capillaries, decreased oxygen diffusion from the peritubular capillaries to tubular and interstitial cells as a result of fibrosis, stagnation of peritubular capillary blood flow induced by sclerosis of ‘parent’ glomeruli, decreased peritubular capillary blood flow as a result of an imbalance of the vasoactive substances, increased metabolic demands of tubular cells, inappropriate energy usage as a result of uncoupled mitochondrial respiration induced by oxidative stress, and decreased oxygen delivery as a result of anemia [11]. Therefore, chronic tubulointerstitial hypoxia occurs in parallel with CKD progression, and it has been assumed that chronic tubulointerstitial hypoxia is a common final step in the development of end-stage renal disease (ESRD) from a variety of causes [11].

Investigating tubulointerstitial hypoxia requires an ability to measure the oxygen status in the kidney. Several methods have been developed to evaluate that oxygen status, and each of them has advantages and disadvantages (Table 1). In animal experiments, the use of microelectrodes has been the gold standard for quantitative oxygen tension measurement in solid organs. The oxygen tension of the cortex and medulla of a normal kidney, measured by microelectrodes, was reported to be approximately 50 mmHg and 30 mmHg, respectively [17]. This method can be applied to models of kidney diseases such as ischemia/reperfusion injury and polycystic kidney disease [18,19]. It also has disadvantages: it is limited to extracellular regions, it is difficult to use in clinical applications because of its invasiveness, and the oxygen tension and renal flow results might be affected by renal decapsulation [20]. Therefore, other hypoxia detection methods are currently under development: blood oxygen level–dependent magnetic resonance imaging (BOLD-MRI), nitroimidazole probes, and phosphorescence lifetime measurement [17].

BOLD-MRI measures the deoxyhemoglobin concentration in veins and is a non-invasive technique. Though it has been tested in clinical trials, studies using BOLD-MRI in CKD patients to examine the relationship between renal prognosis and renal hypoxia have shown conflicting results [21,22]. Recently, a more precise renal BOLD-MRI value, the twelve-layer concentric objects method, was developed [23] to expand the potential to use renal hypoxia detected by BOLD-MRI as a progression marker for CKD [24]. However, only venous oxygen status can be assessed by BOLD-MRI. Although intravenous oxygen status is important, oxygen status in other compartments, such as intraarterial or intracellular oxygen status, are important as well. Nitroimidazole probes are frequently used as intracellular hypoxic markers. Pimonidazole, a 2-nitroimidazole hypoxic marker, is reductively activated in an oxygen-dependent manner and is covalently bound to thiol-containing proteins in hypoxic cells. Intracellular hypoxia (below 10 mmHg of oxygen partial pressure) can be detected immunohistochemically by pimonidazole–protein adducts. Several positron emission tomography tracers with nitroimidazoles have been studied for clinical applications [17,25,26]. Another promising method is phosphorescence lifetime measurement, which can visualize intravital oxygen status and quantify intracellular oxygen concentrations, which are otherwise difficult to do. The phosphorescent dye, which is based on an iridium (III) complex, (benzothienylpyridine) 2Ir (acetylacetone) with a cationic dimethylamino group, enabled the measurement of intracellular oxygen tension [27]. This technique was applied to mouse kidneys and successfully measured the intracellular partial pressures of oxygen in tubular cells [28]. In that study, ischemia/reperfusion models showed that oxygen pressure dropped at the ischemia phase and subsequently recovered at the reperfusion phase [28]. Afterward, the phosphorescence lifetime imaging technique was established and showed the presence of an oxygen gradient in the surface of the cortex, even in normal mouse kidneys [29]. This method could also be expanded to assess oxygen levels in other compartments using different phosphorescent probes. The biggest limitation of this technique is its need for special instruments. Precise understanding and complementary use of these different methods are crucial for the accurate assessment of renal oxygenation.

According to the International Society of Nephrology, 10% of the world’s population is affected by CKD, and millions die from it every year [30]. As indicated by its severe effects on public health, renal hypoxia, a common prognostic factor in a variety of kidney diseases, is certainly an important challenge.

CKD-associated pathologic mechanisms are varied: changes in signaling pathways, mitochondrial dysfunction and oxidative stress, changes in autophagy, chronic inflammation, vascular dysfunction, and epigenetic regulation [31]. An accumulation of basic and clinical research has established that hypoxia-induced renal fibrosis is a common pathological hallmark of CKD. CKD aggravates tubulointerstitial chronic hypoxia through the multifactorial causes listed above [11]. Renal hypoxia, in turn, accelerates CKD progression. That is, renal hypoxia is both the cause and consequence of kidney disease [32]. Its vicious cycle contributes to CKD progression.

Renal hypoxia is associated not only with CKD but also with the AKI-to-CKD continuum. Although renal function can recover, AKI often results in the development of CKD. Observational studies report that 20% to 50% of AKI survivors develop CKD. From a pathological perspective, incomplete or maladaptive repair after AKI results in kidney fibrosis, which develops and progresses to CKD. Recent studies show that tubulointerstitial hypoxia is a key player in the pathophysiology of the AKI-to-CKD transition and CKD [33]. Capillary rarefaction after AKI episodes induces renal hypoxia, which can affect tubular epithelial cells, fibroblasts, and inflammatory cells, resulting in tubulointerstitial fibrosis. Damaged tubular epithelial cells that fail to re-differentiate might supply a decreased amount of vascular endothelial growth factor (VEGF) and contribute to capillary rarefaction, thereby aggravating hypoxia and forming a vicious cycle [33,34]. Other studies have demonstrated that renal hypoxia induces diverse epigenetic changes such as DNA methylation, histone modification, and chromosome conformational changes, and those epigenetic changes are stored in cells as “hypoxic memory,” which can induce the AKI-to-CKD transition as a long-term effect [7,35]. Among elderly people, AKI frequently occurs in CKD patients. AKI on top of CKD has poor renal prognosis because CKD-associated pathological changes can increase renal sensitivity or susceptibility to AKI and suppress kidney repair or recovery from AKI in CKD patients [31].

Oxidative stress caused by an imbalance between ROS production and scavenging occurs through a dysfunction of mitochondrial respiration. Mitochondrial dysfunction can be caused by renal damage that stems from multiple factors, including aging, diabetes mellitus, and inflammation.

Oxidative stress is enhanced in CKD patients, especially those with diabetic kidney disease (DKD). It is well-known that chronic hyperglycemia induces ROS production in diabetes [36]. ROS-mediated renal inflammation and renal fibrosis contribute to the pathology of DKD through multiple signaling pathways that involve transforming growth factor-β, monocyte chemoattractant protein-1, connective tissue growth factor, tumor necrosis factor-α, interleukin (IL) 1, IL-6, IL-18, and cell adhesion molecules [12].

Several markers related to oxidative stress, such as 8-hydroxy-2′-deoxyguanosine (8-OHdG), malondial-dehyde (MDA), and advanced glycation end products (AGEs), have been used to predict renal prognosis. 8-OHdG is a product of oxidative DNA damage, and 8-OHdG excreted in urine has been widely used as a non-invasive systemic oxidative stress marker in both animal experiments and clinical research. Recent studies have demonstrated that urinary 8-OHdG reflects an increased systemic level of oxidative DNA damage in DKD patients and the severity of DKD [37,38]. A high serum concentration of MDA, a lipid peroxidation marker, is found in uremic patients. An increased plasma level of MDA could be an early prognostic indicator of graft dysfunction after kidney transplantation [39]. AGEs are formed during glycation and oxidative stress, and therefore, an increase in oxidative stress results in increased AGEs, which explain the accumulation of AGEs in CKD patients who are not diabetic. Therefore, AGEs are uremic toxins. AGE deposition in skin, reflecting systemic AGE accumulation, can be noninvasively measured by skin autofluorescence, and is predicted to become a novel risk marker for CKD [40–42]. The utility of these oxidative stress–induced prognostic biomarkers corroborates the importance of oxidative stress in CKD patients. The best predictor for disease prognosis has not yet been identified. 8-OHdG differs substantially among individuals because it is affected by smoking, exercise, and dietary habits, and MDA results depend on the detection method used [43]. Although the effectiveness of skin autofluorescence in measuring AGEs has been shown in many previous reports on renal and cardiovascular diseases, AGEs do not reflect only oxidative stress [44].

Chronic renal hypoxia is certainly present in CKD. However, living organisms are often exposed to hypoxia for a variety of reasons, such as ischemia, hypoxemia, and anemia. Therefore, a defense mechanism against hypoxia called the hypoxia-inducible factor (HIF) pathway evolved. In hypoxic conditions, HIF accumulates due to decreased HIF hydroxylation by prolyl hydroxylase (PHD), a rate-limiting enzyme in HIF degradation that requires an oxygen molecule as a substrate. HIF activation promotes the expression of downstream genes such as glucose transporter 1 (GLUT1), erythropoietin (EPO), and VEGF, which counteract hypoxia by increasing glucose uptake, erythropoiesis, and angiogenesis, respectively [11]. Another function of HIF activation is oxidative stress reduction by the inhibition of ROS production and the enhancement of detoxification. An example of inhibiting ROS production is mitochondrial NDUFA4L2 (NADH dehydrogenase [ubiquinone] 1 alpha subcomplex, 4-like 2) expression, which is induced by hypoxia. NDUFA4L2 expression is upregulated by HIF, which attenuates mitochondrial oxygen consumption by inhibiting the activity of complex I (NADH: ubiquinone oxidoreductase) in the electron transport chain to limit intracellular ROS production under low-oxygen conditions [45]. Another study showed that HIF optimized the efficiency of respiration by altering the composition of a cytochrome c oxidase subunit, resulting in decreased ROS and increased ATP production [46]. Heme oxygenase 1 (HO-1) and superoxide dismutase 1 (SOD-1) are well-known ROS-detoxifying enzymes regulated by HIF [47–49].

A master regulator of defense responses to oxidative stress is the nuclear factor erythroid 2-related factor 2 (Nrf2)–Kelch-like ECH-associated protein 1 (KEAP1) interaction. In normal conditions, KEAP1 binds to Nrf2, resulting in proteasomal degradation of Nrf2. Under oxidative stress, the molecular structure of KEAP1 changes until it loses the ability to bind Nrf2. The resulting accumulation of Nrf2 leads to Nrf2 translocation to the nucleus, which promotes the expression of a variety of cytoprotective genes related to redox and detoxification [50]. On the other hand, p38 mitogen-activated protein kinase (MAPK) and c-Jun amino terminal kinase (JNK) signaling are upregulated under oxidative stress, and they are associated with cell death and inflammation [50,51].

The effectiveness of activating the HIF pathway or Nrf2 and inhibiting apoptosis signal-regulating kinase 1 (ASK-1), an upstream signaling kinase of p38 MAPK and JNK, in the treatment of renal diseases is now under active investigation to enable future clinical applications of antioxidative treatments, as described in detail below.

The significance of hypoxia and oxidative stress in CKD has been described above. There is an intricate link between renal hypoxia and oxidative stress.

Oxidative stress is enhanced in CKD, especially in DKD [36]. Increased oxidative stress causes increased kidney oxygen consumption, which results in kidney tissue hypoxia. An underlying mechanism is the accumulation of uremic toxin. Indoxyl sulfate (IS), a representative uremic toxin, causes oxidative stress, which results in increased oxygen consumption and hypoxia [52]. Other uremic toxins, such as phenyl sulfate and ρ-cresyl sulfate, enhance tubular cell susceptibility to oxidative stress by depleting the glutathione level. Hyperuricemia is another mechanism that causes increased oxidative stress in CKD. Long-term hyperuricemia results in increased renal oxidative stress and mitochondrial dysfunction [53]. These findings indicate that oxidative stress induced by uremic toxins in CKD aggravates renal hypoxia.

Renal hypoxia, in turn, magnifies renal oxidative stress. Oxidative stress caused by excess ROS production generally leads to renal inflammation and fibrosis via diverse signaling pathways [12]. It is certain that hypoxia and hyperoxia both result in mitochondrial generation of ROS in various organs, including the kidney. To explain the paradoxical increase in ROS production during hypoxia, an interesting experiment demonstrated that ROS generated by complex III of the electron transport chain stabilized HIF, which implies that mitochondria have a potential oxygen sensing mechanism at complex III. The discovery of the molecules that act as oxygen sensors will provide a therapeutic strategy for oxidative stress [54]. A study that used dinitrophenol, a mitochondrial uncoupler that increases oxygen consumption, found that kidney tissue hypoxia, per se, without confounding hyperglycemia or oxidative stress, could be sufficient to initiate the development of nephropathy [55].

Hypoxia and oxidative stress are intricately linked, and both aggravate the progression of CKD. Therefore, treatments against hypoxia and oxidative stress are needed to dissolve this vicious cycle.

As explained in this review, oxidative stress and chronic hypoxia are intricately linked in CKD, and the discovery of therapeutic approaches against increased oxidative stress is in progress (Fig. 1, Table 2).

It has been well established that increased oxidative stress correlates with a variety of diseases in multiple organs. In the kidney, oxidative stress is upregulated in CKD. Given that the kidney is uniquely susceptible to hypoxia, renal hypoxia (not hyperoxia) is an important source of oxidative stress. Renal hypoxia increases oxidative stress, which in turn exacerbates renal hypoxia. The vicious cycle between oxidative stress and tissue hypoxia plays an important part in CKD progression. Therefore, novel therapeutic approaches against oxidative stress are needed for the treatment of renal diseases. Several promising agents, such Nrf2 activators and HIF stabilizers, have shown beneficial results in clinical trials (Table 2). Future studies are needed to accumulate more evidence of the renoprotective effects and safety of those agents in clinical applications.