HIF-1$\alpha$ was first discovered by Semenza et al. [8, 9], who in their study identified a transcription factor that binds to the enhancer region of the erythropoietin gene in response to hypoxic conditions. They named this factor HIF-1 and later discovered that it was composed of HIF-1$\alpha$ and HIF-1$\beta$ [10]. HIF-2$\alpha$ was discovered by Tian et al. [11]. By cloning and characterization, they found that HIF-2$\alpha$ shares 48% sequence identity with HIF-1$\alpha$ and dimerizes with HIF-1$\beta$ to activate the transcription of the erythropoietin gene [11]. Although both HIF-1$\alpha$ and HIF-2$\alpha$ bind to the hypoxia-responsive element of the erythropoietin gene, HIF-2$\alpha$ primarily regulates erythropoietin transcription in adults owing to differences in cell specificity and target genes between HIF-1$\alpha$ and HIF-2$\alpha$ [12, 13].

Several research groups contributed to the discovery of HIF degeneration through the von Hippel-Lindau tumor suppressor protein (pVHL). Kaelin Jr.’s group discovered that various hypoxia-inducible mRNAs were insensitive to oxygen and were overexpressed in renal cancer cell lines lacking functional pVHL [14]. Concurrently, Salceda and Caro [15] reported that HIF-1$\alpha$ undergoes rapid ubiquitination and subsequent proteasomal degradation under normoxic conditions. Ratcliffe’s group demonstrated that cells lacking pVHL did not target HIF-1$\alpha$ for oxygen-dependent proteolysis [16]. Kaelin’s group [17] revealed the mechanism of ubiquitination of HIF-1$\alpha$ by pVHL. pVHL-elongin B/C-CUL2 complex binds directly to the oxygen-dependent degradation domain of HIF-1$\alpha$, resulting in its polyubiquitination [17]. They further reported that pVHL binds to HIF when a conserved proline residue in HIF is hydroxylated, which requires molecular oxygen and ferrous ion [18]. Studies by Kaelin’s group and Ratcliffe’s group finally revealed that hypoxia sensing and pVHL-dependent degradation of HIFs are primarily regulated by HIF-PH [19, 20].

William G. Kaelin Jr., Peter J. Ratcliffe, and Greg L. Semenza were awarded the Albert Lasker Basic Medical Research Award in 2016 and were subsequently awarded the Nobel Prize in 2019 for their contributions to the discovery of HIF and pVHL-mediated oxygen-sensing regulation of HIF [21, 22]. The first orally active HIF-PH inhibitor, roxadustat, was developed by FibroGen [23, 24] and was approved in China in 2018 for the treatment of renal anemia in patients with dialysis-dependent CKD [25]. There are five HIF-PH inhibitors currently approved for the treatment of renal anemia in patients with CKD: roxadustat, daprodustat, vadadustat, enarodustat, and molidustat [26, 27, 28, 29].

Both ESAs and HIF-PH inhibitors are currently used for the treatment of renal anemia. ESAs are synthesized from human endogenous erythropoietin and possess additional glycosylation or polyethylene glycol residue to extend their half-life, allowing them to remain in the body and exert a hematopoietic effect for several days [30, 31, 32]. Consequently, ESAs are administered to patients every 2 to 4 weeks. Although their half-life is extended, they are degraded in approximately 1 week, which may cause a wide variation in hemoglobin levels between pre- and post-injections. HIF expression is induced under hypoxic conditions and further induces the transcription of hypoxia-inducible genes, including erythropoietin and iron metabolism-related genes. Once hypoxic conditions are corrected to normoxic conditions, HIFs are hydroxylated by HIF-PHs and promptly degraded by the ubiquitin-proteasome pathway. HIF-PH inhibitors stabilize HIFs regardless of the oxygen level by inhibiting the hydroxylation of HIF-PHs, which maintains the transcription of hypoxia-inducible genes. In contrast to ESAs, they are administered to patients every day or thrice a week, which induces stable erythropoietin production at physiological levels. Given that HIF-PH inhibitors promote iron use by regulating iron metabolism-related genes such as hepcidin, they improve renal anemia even in patients with chronic inflammation [32]. They are better for patients because they are oral medications.

Several studies have demonstrated the protection of renal tubules under AKI by HIF-PH inhibitors. An overview of the studies introduced in this section is shown in Table 1 (Ref. [33, 34, 35, 36, 37, 38]). Ito et al. [33] demonstrated that enarodustat offers a protective effect against AKI using a rat model of renal ischemia–reperfusion injury (IRI). Pretreatment with enarodustat reduced plasma creatinine 48 h after ischemia reperfusion, dependent on glycogenesis. Performing an in vitro experiment using proximal tubular-derived cell lines, they demonstrated that enarodustat induced the expression of HIF-1$\alpha$ via inhibition of prolyl hydroxylase domain-containing protein (PHD) 2, one of HIF-PHs, leading to an increase in glycogenesis, suppression of reactive oxygen species (ROS), and improvement of cell viability in cells under the deprivation of glucose and oxygen.

Yang et al. [34] reported the involvement of absent in melanoma 2 (AIM2) activation and 5’-Nucreotidase Ecto (CD73) expression in the renoprotective effect of roxadustat using an IRI mouse model. They confirmed that the expressions of the AIM2 inflammasome complex and CD73 were induced in the proximal tubular cells of the kidneys of patients with AKI and post-transplant kidneys. In IRI mice, treatment with roxadustat enhanced the expression of HIF-1$\alpha$ and CD73 in the proximal tubular cells and suppressed the expression of AIM2 complex, resulting in the improvement of renal function and tissue injury. Co-administration with a CD73 inhibitor abolished the effect of roxadustat on AIM2 complex expression and improved renal function and tissue injury.

Both the renoprotective and cardioprotective effects of HIF-PH inhibitors have been suggested. Deguchi et al. [39] reported the cardioprotective effects of a HIF-PH inhibitor. Pretreatment with roxadustat markedly reduced infarct size and suppressed plasma creatinine kinase activity in the mouse cardio-IRI model. The protective effect was accompanied by an increase in HIF-1$\alpha$ expression, suggesting that HIF-1$\alpha$ plays a key role in the protective effects of HIF-PH inhibitors against ischemic injury.

Both HIF-1$\alpha$ and HIF-2$\alpha$ could be involved in the improvement of AKI. We recently demonstrated that pretreatment with roxadustat improved renal function after renal IRI in Sprague–Dawley (SD) rats, which was accompanied by the production of erythropoietin in renal tubular cells [35]. The expression of nephron-specific genes after IRI was higher in rats pretreated with roxadustat than in rats pretreated with hypoxia, suggesting the protective effect of roxadustat on renal tubules. Pretreatment with roxadustat significantly maintained the increase in plasma erythropoietin levels and erythropoietin expression in the kidney during IRI, whereas the induction of erythropoietin by hypoxic preconditioning was transient before IRI. Immunohistochemical analysis revealed that roxadustat induced the expression of erythropoietin in the proximal and distal tubular cells but not in the interstitial cells of the kidney with IRI. It has been demonstrated that small amounts of HIF-2$\alpha$ are expressed in renal tubular cells and that its expression can be induced [40, 41]. Therefore, we hypothesize that pretreatment with HIF-PH inhibitor contributes to prolonged renoprotective effects on AKI by inducing HIF-2$\alpha$ and erythropoietin expressions in renal tubules.

The involvement of hypoxia-inducible factor prolyl hydroxylase 2 (HIF-PH2; also called PHD2) in the renoprotective effects of endothelial cells has been suggested. Tamoxifen-induced endothelial cell-specific deficits in HIF-PH2 in adult mice exerted protective effects against AKI caused by bilateral renal IRI, in which the infiltration of macrophages and expression of inflammation-related genes were significantly inhibited in HIF-PH2 knockout (KO) mice [36]. The renoprotective effect was attributed to the activation of HIF-1$\alpha$ alone. Furthermore, the study implied that a humoral factor, such as vascular cell adhesion molecule-1 and interferon regulatory factor 1, generated by the activation of HIF-1$\alpha$ in extrarenal endothelial cells contributed to the renoprotective effects. The same study reported that in endothelial HIF-2$\alpha$ or HIF-1$\alpha$ KO mice, the loss of HIF-2$\alpha$ alone exacerbated AKI induced by IRI [37]. The loss of HIF-PH2 probably changes both activation of HIF-1$\alpha$ and HIF-2$\alpha$, which may alter the effects of HIF-1$\alpha$ and HIF-2$\alpha$ on renal injury in both studies. Together, these studies suggest that HIF-1$\alpha$ and HIF-2$\alpha$ act cooperatively rather than antagonistically to protect against AKI.

Collectively, the renoprotective effects of HIF-PH inhibitor in AKI are likely exerted in proximal tubular cells via activation of HIF-1$\alpha$, which induces glycogenesis, suppresses inflammation, and enhances erythropoietin production. HIF-PH inhibitor inhibits HIF-PH2 in renal tubular and extrarenal endothelial cells in AKI, leading to the protective effects on AKI through the activation of HIF-1$\alpha$ and HIF-2$\alpha$. The activations of HIF-1$\alpha$ and HIF-2$\alpha$ in extrarenal endothelial cells suppress the secretion of inflammation-inducible humoral factors, possibly contributing to the renoprotective effect against AKI.

Kapitsinou et al. [38] reported that the renoprotective effect of an HIF-PH inhibitor was observed only when the kidney was pretreated with an HIF-PH inhibitor. Treatment after ischemic reperfusion did not exert the renoprotective effect on the kidneys. Treatment with GSK1002083A, a structural analog of 2-oxoglutarate, at 48 and 6 h before 20 min of bilateral renal pedicle clamping significantly suppressed the increase in blood urea nitrogen levels. IRI-induced fibrosis and inflammation were significantly improved following treatment with an HIF-PH inhibitor 21 days after IRI [38]. Therefore, HIF-PH inhibitors may be useful for the prevention of AKI due to ischemia, sepsis, and post-transplant kidney injury. In particular, in case of kidney transplantation, the time when renal ischemia occurs is clear. We believe that the use of HIF-PH inhibitors for pretreatment of transplanted kidney before the transplantation is practical. A search of ClinicalTrials.gov (https://clinicaltrials.gov) revealed that there appear to be no clinical studies examining the protective effect of HIF-PH inhibitors on AKI. The clinical studies are warranted to clarify the benefit of HIF-PH inhibitors for AKI.

Treatment with HIF-PH inhibitors induces the expression of both HIF-1$\alpha$ and HIF-2$\alpha$, and opposing roles of these two factors in chronic renal fibrosis have been suggested. An overview of the studies introduced in this section is shown in Table 2 (Ref. [42, 43, 44, 45, 46, 47]).

Higgins et al. [42] found that progression of renal fibrosis was HIF-1$\alpha$ dependent. They demonstrated that unilateral ureteral obstruction (UUO) caused hypoxia in renal tissue, which induced epithelial–mesenchymal transition (EMT) accompanied by the expression of HIF1$\alpha$, leading to renal fibrosis. An in vitro study revealed that the epithelial tubular cells derived from proximal tubular cell-specific HIF-1$\alpha$ KO mice did not exhibit EMT, consistent with the in vivo results that renal fibrosis and macrophage infiltration induced by UUO were mitigated in the HIF-1$\alpha$ KO mice. This suggests that HIF-1$\alpha$ in renal tubular cells could be a therapeutic target for renal fibrosis, although the study has not explored the role of HIF-2$\alpha$, whose expression was induced by UUO as well as HIF-1$\alpha$.

Kobayashi et al. [43] examined the effect of global deletion or activation of HIF-1$\alpha$ and HIF-2$\alpha$ on renal fibrosis using transgenic mice. In conditional global deletion of HIFs in mice, renal fibrosis was exacerbated by UUO compared to control mice with UUO. In contrast, UUO-induced renal fibrosis was mitigated in mice with global activation of HIFs. Renal fibrosis exacerbated by the deletion of HIFs was dependent on macrophage infiltration into the kidneys. They further demonstrated that myeloid-specific deletion and activation of HIFs enhanced and attenuated the infiltration of macrophages in the kidney, respectively, although they did not affect renal fibrosis caused by UUO. This study suggested that HIF-1$\alpha$ and HIF-2$\alpha$ play key roles in renoprotective effect through the regulation of infiltration of macrophages and fibrosis in the kidney. This finding suggests that the activation of HIF-2$\alpha$, rather than the suppression of HIF-1$\alpha$, contributes to the renoprotective effect [43].

The expression of HIF-2$\alpha$ is localized mainly to glomerular mesangial cells, vascular smooth muscle cells, endothelial cells, and interstitial fibroblast cells in the kidney. A small amount of HIF-2$\alpha$ is expressed in glomerular and tubular epithelial cells [40]. The HIF-2$\alpha$ expression in renal tubular cells was inducible with aldosterone and vasopressin, accompanied by erythropoietin expression [41]. Renoprotective role of HIF-2$\alpha$ has been suggested by a study using transgenic mice, in which the expression of HIF-2$\alpha$ was conditionally enhanced in renal epithelial cells [44]. The study examined the effect of renal tubular cell-specific induction of HIF-2$\alpha$ at early-stage or at late-stage kidney injury caused by a 0.2% adenine diet. The induction of HIF-2$\alpha$ at early-stage kidney injury aggravated renal fibrosis by increasing the expression of type I collagen and fibronectin. In contrast, the induction of HIF-2$\alpha$ at late-stage kidney injury restored reduced renal vasculature and improved renal tissue hypoxia, resulting in the improvement of renal fibrosis. The study further demonstrated that the expression of HIF-2$\alpha$ was increased in microdissected renal tubulointerstitium of patients with early-stage immunoglobulin A (IgA) nephropathy, while the expression of HIF-2$\alpha$ was decreased in the subjects with progressed IgA nephropathy [44]. In vitro studies showed that the induction of HIF-2$\alpha$ in isolated renal tubular cells inhibited transforming growth factor-$\beta$1 (TGF-$\beta$1)-induced expressions of fibronectin and $\alpha$-smooth muscle actin, confirming the renoprotective role of HIF-2$\alpha$ in renal tubular cells [48]. These findings suggest a possible therapeutic target of HIF-2$\alpha$ for the treatment of late-stage patients with CKD. We demonstrated that treatment of rats with roxadustat stimulated erythropoietin production not only in interstitial cells but also in proximal tubular cells in the kidney, suggesting that HIF-PH inhibitors induce HIF-2$\alpha$ in proximal tubules [49]. These findings suggest that HIF-2$\alpha$ in renal tubular cells, as well as in interstitial cells, is inducible with HIF-PH inhibitor and contributes to the prevention of CKD progression.

Schley et al. [45] reported the renoprotective effect of an HIF-PH inhibitor on tubulointerstitial nephritis caused by an adenine-containing diet in mice. In the study, the effects of isoquinoline derivative 2-(1-chloro-4-hydroxyisoquinoline-3-carboxamido) acetate (ICA), an inhibitor for PHD enzymes, and roxadustat were examined. Both ICA and roxadustat inhibited the progression of renal tubular damage and interstitial fibrosis induced by adenine, accompanied by a reduced infiltration of mononuclear phagocytes. Notably, the renoprotective effects of ICA were independent of HIF-1$\alpha$ and HIF-2$\alpha$ in renal tubular and myeloid cells. This study suggests the broad effects of HIF-PH inhibitors, independent of HIFs in kidney epithelial cells.

The roles of HIF-PH2 in chronic kidney injury have been examined. Zhao et al. [46] demonstrated that a deficit of HIF-PH2 in endothelial cells inhibited angiotensin II-induced renal fibrosis owing to decreased expression of angiotensin II receptor type 1 with subsequent suppression of ROS production and TGF-$\beta$1 expression. In contrast, another study demonstrated that a congenital deficit of HIF-PH2 in endothelial cells caused renal fibrosis (over 15 months) in endothelial HIF-PH2 KO mice [50]. In the study, a deficit of HIF-PH2 in endothelial cells increased the expression of HIF-1$\alpha$ and HIF-2$\alpha$ in the kidney, which subsequently induced Notch3 and TGF-$\beta$1, resulting in remodeling of arterioles and fibrosis in the kidney. These results suggest that the inhibition of endothelial HIF-PH2 exerts renoprotective effects on CKD through the activation of HIF-1$\alpha$ and HIF-2$\alpha$ in the kidney, while HIF-PH2 is important for normal development of the kidney.

Previous studies suggest that the timing of HIF-PH inhibitor administration is important for exerting the renoprotective effects in CKD. Yu et al. [47, 51] examined the effect of L-mimosine, a HIF-PH inhibitor, on kidney injury in rats subjected to subtotal nephrectomy. L-mimosine was administered during early, advanced, or end-stage kidney damage. The study demonstrated that the administration of L-mimosine at an advanced stage significantly improved renal function, renal fibrosis, and anemia, exacerbating them at the early stage and not having any effects at the end stage. The study further demonstrated that the renoprotective effect of L-mimosine at the advanced stage of kidney injury was accompanied by significantly increased expression of HIF-2$\alpha$ and erythropoietin. However, HIF-1$\alpha$ expression was induced only at the early stage, with HIF-2$\alpha$ and erythropoietin expressions occurring slightly at the end stage of kidney injury under the L-mimosine administration [47]. In the subtotal nephrectomy model, the expressions of HIF-1$\alpha$ and HIF-2$\alpha$ and their target genes were increased in the early stage, and returned to basal level in the advanced and end stages [51]. The same study suggested mechanisms of renoprotective effect by HIF-PH inhibitors, which involve miRNAs. They demonstrated that the expression of miR-29c decreased in subtotal nephrectomized kidneys, whereas treatment with L-mimosine restored its expression. miR-29c targets extracellular matrix genes and reduces their expression. In vitro study revealed that the expression of miR-29c was increased by cobalt chloride (CoCl₂) and decreased by knockdown of HIF-1$\alpha$ or HIF-2$\alpha$ [52].

These findings suggest the possible use of HIF-PH inhibitors in the treatment of patients with progressive CKD. Treatment with HIF-PH inhibitors probably maintains the expression of HIF-2$\alpha$ alone in progressed CKD, leading to renoprotective effect against renal fibrosis. The broad effects of HIF-PH inhibitors (except on HIFs) could also be involved in renoprotective effect. The appearance of renal anemia, which indicates erythropoietin deficiency due to lack of HIF-2$\alpha$ and advanced progression of CKD, would represent an appropriate timing to start HIF-PH inhibitor, not only for the treatment of anemia but also for the treatment of renal fibrosis. Administration of HIF-PH inhibitors restores HIF-2$\alpha$ expression in renal tissue and contributes to renoprotective effect. In any case, treatment with HIF-PH inhibitor in the early stage of CKD is not appropriate as it may induce increased erythropoiesis and subsequent risk of polycythemia, as well as inappropriate induction of HIF-1$\alpha$ (Fig. 1, Ref. [42, 44, 47, 51]).

Fig. 1.

Schematic illustration of hemoglobin levels, HIF-1$\alpha$ and HIF-2$\alpha$ expression, and the effects of HIF-PH inhibitors during CKD progression. Previous studies demonstrated that the expressions of HIF-1$\alpha$ and HIF-2$\alpha$ in the kidney were increased in the early stage, and returned to basal level in the advanced and end stages of CKD [51]. Activation of HIF-1$\alpha$ in renal tubules accelerated renal fibrosis by inducing EMT [42]. The activation of HIF-2$\alpha$ in renal tubules at early-stage kidney injury aggravated renal fibrosis by increasing the expression of type I collagen and fibronectin. In contrast, the induction of HIF-2$\alpha$ at late-stage kidney injury improved renal fibrosis [44]. Treatment with HIF-PH inhibitor at an advanced stage significantly improved renal function, renal fibrosis, and anemia, exacerbating them at the early stage and not having any effects at the end stage [47]. Treatment with HIF-PH inhibitors probably maintains the expression of HIF-2$\alpha$ alone in progressed CKD, leading to renoprotective effect against renal fibrosis. The broad effects of HIF-PH inhibitors (except on HIFs) could also be involved in renoprotective effect. As description of graphs, vertical axis indicates levels of HIF-1$\alpha$ and HIF-2$\alpha$ expression and hemoglobin. Horizontal axis indicates years lapsed. Hemoglobin level is shown with red line. Expressions of HIF-1$\alpha$ and HIF-2$\alpha$ are shown in cumulative graph with orange and blue, respectively.

The pathogenesis of diabetic kidney disease is unique and induced by vascular damage due to hyperglycemia. Once proteinuria occurs, disease progression is rapid and unavoidable. The establishment of a treatment method before the progression to renal fibrosis is necessary. An overview of the studies introduced in this section is shown in Table 3 (Ref. [53, 54, 55, 56, 57, 58]). Nordquist et al. [53] examined the effects of CoCl₂ on kidney injury in rats with streptozotocin-induced diabetes. In diabetic rats, oxygen consumption was significantly increased, accompanied by excessive glomerular filtration rate (GFR), proteinuria, and tubulointerstitial injury. Treatment with CoCl₂ reduced oxygen consumption and improved excessive GFR, proteinuria, and tubulointerstitial injury, suggesting the protective effect of HIFs on diabetic kidney injury. The beneficial effects of CoCl₂ were related to reduced oxidative stress [53]. In proximal tubule-specific conditional HIF-1$\alpha$ KO mice, diabetes induced by streptozotocin exacerbated proteinuria and tubular injury due to increased ROS accumulation and apoptosis in the kidney. An in vitro study demonstrated that HIF-1$\alpha$ inhibited mitochondrial fragmentation, ROS production, mitochondrial membrane potential loss, and apoptosis under hypoxic ambiance by activating heme oxygenase-1, a target gene of HIF-1$\alpha$ [54]. Systemic HIF-1$\alpha$-deficient mice exhibited exacerbation of streptozotocin-induced diabetic nephropathy [55, 56]. Notably, the above studies were conducted on streptozotocin-induced diabetic nephropathy, which could be considered as subacute kidney injury rather than chronic kidney injury caused by diabetes.

The opposite effects of HIF-1$\alpha$ on diabetic kidney injury have been suggested. Nayak et al. [57] reported the renoprotective effect of HIF-1 inhibitor on diabetic nephropathy. OVE26 mice, a type 1 diabetic animal model, were treated with 3-(5^′-hydroxymethyl-2^′-furyl)-1-benzyl indazole (YC-1), an HIF-1$\alpha$ inhibitor. The expression of HIF-1$\alpha$ was increased in the kidney of OVE26 mice. Treatment with YC-1 reduced glomerular hypertrophy, mesangial matrix expansion, extracellular matrix accumulation, and urinary albumin excretion, as well as NADPH oxidase 4 protein expression and NADPH-dependent reactive oxygen species production in OVE26 mice, while their blood glucose levels remained unchanged. The study suggests the renoprotective effect of YC-1 through the inhibition of HIF-1$\alpha$-induced glucose transporter 1 (GLUT-1) expression. It should be noted that YC-1 induces cGMP production in a nitric oxide-independent manner, promoting vasodilation and antiplatelet effect, as well as HIF-1$\alpha$ inhibition.

Sugahara et al. [58] demonstrated the renoprotective effect of an HIF-PH inhibitor in a diabetic kidney disease model. Diabetic black and tan brachyury (BTBR) ob/ob mice were treated with enarodustat or its vehicle between 4 and 22 weeks of age. BTBR ob/ob mice exhibited albuminuria with foot process effacement, glomerular hypertrophy, mesangial expansion, and interstitial fibrosis. Treatment with enarodustat improved albuminuria and foot process effacement, but did not alter glomerular hypertrophy and interstitial fibrosis. Treatment with enarodustat decreased the expression of C-C motif chemokine ligand 2/monocyte chemoattractant protein 1 (CCL2/MCP1) in the glomeruli of BTBR ob/ob mice, resulting in the suppression of macrophage infiltration. An in vitro study revealed that enarodustat suppressed palmitate-induced CCL2/MCP-1 production via HIF-1$\alpha$. Treatment with enarodustat improved hyperglycemia and insulin sensitivity in BTBR ob/ob mice. It also decreased cholesterol levels and increased adiponectin levels. The study concluded that HIF-PH inhibitors may prevent metabolic disorders and have potential renoprotective effects [58].

Generally, few studies have examined the effects of HIF-PH inhibitors in diabetic kidney disease models. Although they may be effective in diabetic kidney disease, activation of HIF-1$\alpha$ may be harmful depending on the stage of disease progression.

Given that HIF-PH inhibitors are involved in the regulation of various pathways, the effects of combining them with other drugs on CKD should be considered. Sung et al. [59] examined the combination therapy of roxadustat and dapagliflozin, a sodium glucose co-transporter 2 (SGLT2) inhibitor, for cardiorenal syndrome-induced damage in rats. SD rats were subjected to remnant kidney-induced chronic kidney injury, followed by acute myocardial infarction. The rats were treated with either roxadustat or dapagliflozin, or both, for a month. Combination therapy significantly improved cardiorenal syndrome—both cardiac and renal functions improved by suppressing tissue fibrosis. The study suggested that the combination therapy enhanced HIF-1$\alpha$ production by activating the Nrf2/ARE pathway and stabilizing HIF-1$\alpha$, although the involvement of HIF-2$\alpha$ in the cardiorenal protection has not been examined [59]. A previous study suggests the involvement of HIF-1$\alpha$ suppression in the renoprotective effect of SGLT2 inhibitors for diabetic kidney injury [60]. Treatment of diabetic db/db mice with luseogliflozin decreased the expression of cortical tubular HIF-1$\alpha$, accompanied by attenuation of tubular injury. An in vitro study demonstrated that the treatment with luseogliflozin reduced oxygen consumption under hypoxic conditions, resulting in suppression of HIF-1$\alpha$ expression in human renal proximal tubular epithelial cells [60].

The involvement of Zinc (Zn) deficiency in CKD progression and the renoprotective effects of Zn supplementation have been suggested [61]. Zhang et al. [62] reported that treatment with Zn inhibited tubular EMT and attenuated renal tubulointerstitial fibrosis by downregulation of HIF-1$\alpha$ in the kidneys of diabetic streptozotocin-treated mice, which was accompanied by improved hyperglycemia and proteinuria. In vitro experiments showed that the supplementation of Zn suppressed the expression of HIF-1$\alpha$ protein and phosphoinositide 3-kinase signaling-related proteins induced by high glucose and hypoxia [62].

The Rho-kinase inhibitor, fasudil, used to treat cerebral vasospasm after subarachnoid hemorrhage, is expected to be approved for other diseases such as pulmonary hypertension and heart failure [63]. Matoba et al. [64] demonstrated that fasudil improved kidney injury in db/db mice through the inhibition of HIF-1$\alpha$ expression by increased expression of HIF-PH2.

HIFs are involved in many signaling pathways, such as mitochondrial and glycolytic metabolism, erythropoiesis, ferroptosis, and inflammation [65]. Some of these pathways are thought to be common to other clinically approved drugs. Although the use of HIF-PH inhibitors in combination therapy is expected to have a renoprotective effect, caution is required as they may also inhibit the effects of combined medication.

Clinical trials have examined the safety and efficacy of HIF-PH inhibitors. The effect of vadadustat on renal anemia and major adverse cardiovascular events (MACE) was examined in non-dialysis patients with CKD who either received or did not receive darbepoetin alfa [66]. Treatment with vadadustat improved renal anemia comparably to darbepoetin alfa, demonstrating the non-inferiority of vadadustat to darbepoetin. However, MACE was more in patients with vadadustat than in patients with darbepoetin alfa (22.0% vs. 19.9%; hazard ratio, 1.17). The increased risk of MACE in patients treated with vadadustat was largely due to an excess of nonfatal myocardial infarctions and a higher incidence of death from non-cardiovascular causes. The effects of daprodustat on renal anemia and MACE were examined in non-dialysis patients with stage 3–5 CKD [67]. The control group comprised non-dialysis patients with stage 3–5 CKD who received an injection of darbepoetin alfa. The effect of daprodustat on Hb elevation was not significantly different from that of darbepoetin, confirming the non-inferiority of daprodustat to darbepoetin alfa. The occurrence of MACE in patients treated with daprodustat was not different from that in patients treated with darbepoetin, suggesting the safety of daprodustat in non-dialysis patients with CKD. These two large-scale clinical studies revealed controversial results regarding the safety of HIF-PH inhibitors. The incidence of MACE in different HIF-PH inhibitors including vadadustat and daprodustat is shown in Table 4 (Ref. [66, 67, 68, 69, 70]) for reference.

Systematic reviews of meta-analyses examining the efficacy and safety of HIF-PH inhibitors for the treatment of anemia in patients with CKD showed no notable disparities in safety outcomes between the HIF-PH inhibitor and ESA or placebo groups [71, 72]. These agents improved anemia as much as ESA by enhancing iron metabolism, decreasing hepcidin levels, and improving iron transport. Potential adverse events (AEs) associated with HIF-PH inhibitors have been suggested, including the development of malignant tumors, diabetic retinopathy, age-related macular degeneration, and progression of polycystic kidney diseases [73, 74]. The onset or progression of these diseases has been suggested to be induced by the activation of HIF-1$\alpha$ and the subsequent induction of VEGF. Although there is no clinical evidence that shows the correlation between HIF-PH inhibitors and potential AEs currently, the Japanese Society of Nephrology and the Asian Pacific Society of Nephrology have issued recommendation papers for the proper use of these agents and regular check-ups for low-risk patients [75]. Long-term careful observation of patients treated with HIF-PH inhibitors is required.

Two clinical trials described above reported that CKD progression was not delayed in patients treated with HIF-PH inhibitors compared to patients treated with darbepoetin alfa. The effect of the HIF-PH inhibitor on CKD progression has not been examined in comparison to a placebo. A few clinical studies have suggested a renoprotective effect of ESA on CKD progression; however, these studies were small-scale. Recently, we performed a retrospective study to examine the renoprotective effects of daprodustat [76]. The study included patients with stage 3–5 CKD (mean eGFR, 26.0 mL/min/1.73 m² $\pm$14.0) and revealed that daprodustat significantly attenuated the decline in eGFR compared to the period before the treatment (Fig. 2, Ref. [76]). The decline in eGFR was halted in approximately 90% of patients, although the sample size was small. The effect of daprodustat was not observed in patients who had already progressed to stage 5 CKD, suggesting a CKD stage-dependent effect of HIF-PH inhibitors, as implied by in vivo studies [46, 47]. The clinical trials described above seem to have included a relatively large number of patients with stage 5 disease. Another clinical study reported a beneficial effect of HIF-PH inhibitors on heart failure-induced CKD progression. The study included 69 patients with a median eGFR of 29.1 mL/min/1.73 m² (interquartile range: 19.0–35.1 mL/min/1.73 m²) at treatment initiation [77]. Treatment with HIF-PH inhibitors for 6 months resulted in an improvement in hemoglobin levels and eGFR, comparable to those observed 6 months before treatment. This study suggests that an improvement in heart failure results in an improvement in CKD. Our study examined long-term (about 15 months) effects of HIF-PH inhibitors on the progression of CKD. Given that the study was retrospective and included only a small number of patients, prospective studies including a large number of patients with moderately advanced CKD are required.

Fig. 2.

Renoprotective effect of daprodustat in patients with CKD. (A) The changes in the eGFR slope in three patients before and after (pre and post, respectively) the administration of daprodustat. The figures are shown as representative results from a retrospective study to examine renoprotective effect of daprodustat. (B) The changes in the slope of the decline of eGFR (mL/min/1.73 m²/month) by daprodustat. Daprodustat significantly reduced the decline of eGFR slope. Box and bar show mean $\pm$ SD. * p 0.05. The original figures were published in Ref. [76].

Due to their pharmacological effects, HIF-PH inhibitors have been developed as drugs to improve anemia. Clinical studies have been conducted on renal anemia, leading to their current practical application for renal anemia. It appears that renoprotective effect of HIF-PH inhibitors was not anticipated. A search on ClinicalTrials.gov revealed that currently, very few clinical studies examining the renoprotective effect of HIF-PH inhibitors as a primary endpoint have registered. It is expected that the large-scale clinical trials will be conducted with renoprotective effects as a primary endpoint.

In the last decade, SGLT2 and nonsteroidal mineralocorticoid receptor (MR) inhibitors have proven their renoprotective effects and have been used as promising new drugs for the treatment of patients with CKD. Although SGLT2 inhibitors are effective in early-stage patients with CKD with proteinuria, their renoprotective effect is not much expected in patients with advanced CKD, in whom proximal tubular function has already been abolished. MR inhibitors have been shown to suppress proteinuria and delay CKD progression; however, an initial dip in eGFR and hyperkalemia limits their use in patients with advanced CKD. Collectively, HIF-PH inhibitors may represent an unmet therapeutic need for delaying the progression of moderately advanced CKD.