Roxadustat alleviates the inflammatory status in patients receiving maintenance hemodialysis with erythropoiesis-stimulating agent resistance by increasing the short-chain fatty acids producing gut bacteria

Anemia is a common complication of chronic kidney disease (CKD). Renal anemia is estimated to affect 2 to 4 million of the approximately 20 million people in the United States with CKD [1], and > 90% of the 500,000 patients who undergo dialysis in China, leading to high hospitalization rates and a high incidence of cardiovascular events [2]. Renal anemia is a multifactorial process caused by relative erythropoietin (EPO) deficiency, uremia-induced inhibitors of erythropoiesis, shortened erythrocyte survival, and disordered iron homeostasis [3]. Resistance to erythropoiesis-stimulating agents (ESAs) is mainly associated with abnormal iron metabolism and inflammatory anemia [4‐6]. There has recently been great progress in the development of drugs for the treatment of renal anemia, including hypoxia-inducible factor-prolyl hydroxylase inhibitors (HIF-PHIs), which have been shown to treat renal anemia safely and effectively, via multiple mechanisms [7]. The representative HIF-PHI, Roxadustat (FG-4592, FibroGen, Inc. San Francisco, CA, USA), has been widely used in clinical applications. Roxadustat induces hypoxia in cells, thus stimulating endogenous EPO synthesis, improving iron metabolism, and reducing inflammation [8]. HIF transcription factors comprise two subunits: an oxygen-sensitive α-subunit (HIF-1α, HIF-2α, or HIF-3α) and a constitutively expressed β-subunit [9]. HIF-1α is a master transcription factor related to cell proliferation and survival, iron metabolism, angiogenesis and glucose metabolism [10]. Notably, HIF-1α has been shown to maintain intestinal homeostasis by not only regulating the integrity of the intestinal epithelial barrier, but also improving the survival of intestinal microorganisms [11], while HIF-2α is involved in the regulation of intestinal iron transporters mediated by the intestinal microbiota. Intestinal microorganisms and their metabolites may inhibit HIF-2α under conditions of iron deficiency [12]. As the master intestinal transcriptional regulator of apical and basolateral iron transporters, HIF-2α is also sensitive to cellular iron and oxygen levels, which can affect transcriptional activation of the iron-absorption machinery to maintain systemic iron levels [13‐16].

The human gut microbiota is a complex ecosystem and a key component of gastrointestinal homeostasis. Many studies have shown that the gut microbiota in patients with CKD differs compared with that in healthy people [17‐19]. Changes in the gut microbiota and host reactions are related to the progression of CKD, as well as the increased risk of cardiovascular disease, uremic toxins, and inflammation [20, 21]. In 2011, Ritz proposed an “intestinal-renal syndrome” and derived the theory of an “intestinal-renal axis”, based on extensive studies [22]. This concluded that there was a bidirectional relationship between the kidney and intestine, in which functional impairment of one side could affect the normal function of the other side through various mechanisms, with important roles for the gut microbiota and its metabolites.

The mechanism of HIF-PHIs in the treatment of renal anemia has not yet been fully elucidated. We therefore conducted a prospective cohort study to explore the effect of roxadustat on the gut microbiota in hemodialysis (HD) patients with renal anemia and ESA resistance. The aim of this study was to determine if roxadustat could improve renal anemia in patients with ESA resistance by altering the intestinal microbiota, thereby correcting inflammation and iron metabolism disorders.

A total of 30 patients were included in this study to compare the changes in the gut microbiota spectrum after 3 months of roxadustat treatment (Fig. 1). The baseline characteristics of the participants are shown in Table 1. All participants were of Han ethnicity, born in northeastern China, and had a similar dietary pattern which did not alter during the 3 months. No adverse effects were observed during the 3 months of the participants’ medication. Changes in Hb levels during the 3 months are shown in Fig. 2. The Hb level in patient 7 was an outlier, but the results were confirmed by repeated analysis. Baseline Hb at enrollment was (85.93 ± 9.85) g/L, and this increased significantly to (94.97 ± 12.14) g/L after 2 weeks’ of roxadustat treatment (P < 0.001). Hb levels continued to increase gradually to (100.80 ± 11.97) g/L, (106.57 ± 13.06) g/L, and (109.20 ± 14.78) g/L after the 1st, 2nd, and 3rd months, respectively (P < 0.001). Ferritin, TIBC, UIBC, and hepcidin decreased significantly after the 1st and 3rd months (P < 0.05), while sTfR increased significantly (P < 0.05). Changes in iron-metabolism indices are shown in Table 2. Levels of inflammatory factors decreased significantly after the 1st and 3rd months (P < 0.001), as shown in Table 3.

EPO levels increased significantly (F = 1220.882, P < 0.001) from (6.76 ± 1.69) U/L at baseline to (9.33 ± 1.62) U/L after 1 month and (10.13 ± 1.56) U/L after 3 months.

Serum SCFA levels also increased significantly (F = 23.577, P < 0.001) from (53.21 ± 13.72) g/L at baseline to (60.71 ± 12.36) g/L after 1 month and (69.73 ± 11.67) g/L after 3 months.

The current study showed that Hb and EPO levels were increased after the administration of roxadustat in patients with ESA resistance undergoing maintenance HD. Roxadustat regulated the levels of inflammatory factors, increased iron utilization, and altered the diversity and abundance of the gut microbiota, especially in terms of SCFA-producing bacteria, thus alleviating ESA resistance.

In line with previous studies, roxadustat can effectively treat patients with renal anemia in end-stage renal disease (ESRD) [28, 29]. Numerous clinical trials have shown the advantage of roxadustat compared with epoetin alfa [30]. The mechanism of roxadustat involves stimulating the manufacture of endogenous EPO, improving iron utilization [31] (indicated by suppression of hepatic hepcidin and upregulation of transport genes related to iron metabolism, such as DCYTB, DMT1, and TF), and reducing inflammation. Roxadustat is known to adjust the inflammatory microenvironment and ameliorate disorders of iron metabolism (the two main characteristics of renal anemia) in CKD patients, especially those with ESA resistance [3, 32‐35]. Notably, our results showed a dramatic change in the gut microbiota after the administration of roxadustat, particularly characterized by an increase in SCFA-producing bacteria. Meanwhile, our study showed the abundance of Alistipes shahii was negatively correlated with IL-6 and TNF-α levels. Alistipes shahii, known as SCFAs-producing bacteria, can significantly increase acetate and propionate so that relieve the bowel inflammation [36]. These results thus suggest that roxadustat may alleviate ESA resistance by regulating the gut microbiota.

CKD is considered to be a chronic inflammatory condition [37, 38] and to be closely linked to the gut microbiota [39, 40]. There is growing acceptance that the gut microbiota and its products affect the pathogenesis and development of CKD. Several factors, such as toxin accumulation [41‐43], inflammatory status, drugs (corticosteroids, immunosuppressive agents, antibiotics), and dietary restrictions may affect the gut microbiota [44]. Recent studies found differences in the composition and function of the gut microbiota between dialysis patients with ESRD and healthy controls [45‐47]. Jiang et al. found that butyrate-producing microbiota, such as Roseburia spp. and Faecalibacterium prausnitzii, were significantly decreased in patients with any stage of CKD (n = 65) compared with healthy controls (n = 20) in a study in China. Wong et al. compared patients with ESRD (n = 24) and healthy controls (n = 12) in the United States, and showed that SCFAs played a dominant role in the gut microbiota related to uremic toxins in ESRD patients. Nineteen microbial families were increased, including 12 with urease-producing enzymes, five with uricase, and four with indole and p-cresol-producing enzymes, while four microbial families, including two with butyrate-producing enzymes, were decreased. Zhao et al. conducted a systematic review [48] of gut microbiota in patients with CKD, which showed significant increases in bacterial families possessing urease, uricase, and indole and p-cresol forming enzymes, and decreases in families possessing butyrate-forming enzymes among ESRD patients [49]. Previous studies comparing the intestinal flora between ESRD patients and healthy controls showed decreases in bacteria producing SCFAs, especially butyrate, in ESRD patients [48, 50]. Notably, however, the current study indicated that SCFA-producing bacteria increased after the administration of roxadustat. We thus speculated that the inflammatory and toxic microenvironment in patients with ESRD may be closely linked to the gut microbiota, which may play a dominant role in regulating ESA resistance.

Our results showed that SCFA-producing bacteria and serum SCFAs were increased and inflammatory factors were decreased after the administration of roxadustat. As an HIF-PHI, roxadustat can activate the HIF/PHD oxygen-sensing pathway, thus inducing the expression of HIF-1α and HIF-2α. HIF-1α has an anti-inflammatory effect while HIF-2α affects iron utilization.

Recent studies have highlighted the prominent role of HIF-1α in maintaining intestinal homeostasis, not only by regulating the integrity of the intestinal epithelial barrier but also improving the survival of intestinal microorganisms [11]. The intestinal lumen in mammals is characterized by a hypoxic environment, which can induce the expression of several HIF-1 target genes in intestinal epithelial cells. Thus HIF-1α up-regulates genes involved in intestinal barrier function, such as MUC2, ITF, CLDN1, and other tight-junction proteins, and down-regulates nuclear factor (NF)-κB signal transduction and the production of anti-inflammatory cytokines to exert an immunosuppressive effect [51]. HIF-1α induced the development of SCFA-producing bacteria, which are key to maintaining intestinal homeostasis. First, SCFAs especially butyrate, provides the energy framework of colon cells [52, 53], also plays a crucial role in maintaining intestinal epithelial integrity, which is related to its ability to produce mucin and tight-junction proteins [54‐56]. Second, SCFAs, as a major mode of communication between the microbiota and host cells, have demonstrated several immunomodulatory effects [57]. SCFAs have anti-inflammatory effects by reducing the secretion of proinflammatory cytokines and chemokines, and possibly reducing local macrophage infiltration. SCFAs also exert anti-inflammatory effects through inhibition of NF-κB activation in the host immune cells by binding to G-protein-coupled receptors (GPRs) 43 and 41 [58‐60]. In addition, SCFAs epigenetically regulates gene expression by inhibiting histone deacetylases, which can induce the differentiation of colonic Treg cells [61], and can also improve the host response to inflammation [62, 63]. These functions may also be conducive to the survival of the intestinal flora and may promote the interaction between the host and gut microorganisms. Meanwhile, the abundance of E. coli increased after Roxadustat treatment. In present, the function of E. coli of chronic kidney disease is still inexplicit. There is now substantial research showing that the E. coli is more abundant in patients with ESRD [17, 19, 64] and it may produce protein-bound uraemic toxins (PBUTs), such as indoxyl sulfate (IS) and p-cresyl sulfate (PCS) [65]. In vitro, the level of E. coli was elevated after the application of rhuEPO. Thus, the upregulation of E. coli after roxadustat treatment maybe a result of the promotion of EPO [66]. The function of E. coli in ESRD patients is an unsettled question and need further study.

HIF-1 was first identified as a transcription factor regulating EPO in human hepatoma cells, while in vivo hypoxic stimulation of EPO and erythropoiesis is primarily mediated by HIF-2 [67‐69]. Intestinal iron absorption has recently been shown to be mainly regulated by HIF-2α, which is an oxygen- and iron-regulated transcription factor that directly targets the three key intestinal iron transporters: divalent metal transporter 1 (DMT1), duodenal cytochrome b, and ferroportin (FPN) [15]. In the hypoxic environment of the duodenum, HIF-2α upregulates the expression of apical (DMT1) and basolateral (FPN) enterocyte iron transporters, thus improving iron absorption [70]. The importance of hypoxia signaling and the essential role of HIF-2α in iron homeostasis have been highlighted in several genetic and pathophysiologic mouse models [12]. These models showed that the gut microbiota regulates systemic iron homeostasis in the host in two ways: by repressing intestinal iron-absorption pathways via the inhibition of basal HIF-2α function, and by promoting cellular iron storage via the induction of FTN expression. The host iron-sensing mechanism is intimately connected to and regulated by the gut microbiome. The present study showed that hepcidin, ferritin, TIBC, and UIBC all decreased after the administration of roxadustat, indicating improved iron utilization. This could partly explain the reduced demand for iron in patients with renal anemia treated with roxadustat compared with ESAs. Hepcidin is a key regulator of iron absorption [71]. The commensal microbiota could regulate hepcidin expression during intestinal inflammation via STAT3 activation and modulation of erythropoietic activity [72]. Roxadustat could thus improve anemia by affecting iron metabolism, and especially by reducing hepcidin levels [73]. We, therefore, considered that roxadustat might correct ESA resistance via activation of HIF-2α to improve iron utilization.

Our study had several strengths. First, to the best of our knowledge, this was the first study to explore the effects of roxadustat on the gut microbiota in patients with renal anemia and ESA resistance undergoing maintenance HD. The results thus provide a basis for further studies to explore the mechanism of roxadustat for the treatment of renal anemia. Second, this was a before–after control study, which thus eliminated the potential effects of individual differences and other confounding factors on the results, whilst ensuring the accuracy of the results and reducing the sample size, with high statistical efficiency.

The study also had some limitations. First, the limited data meant that the study may have been influenced by some unmeasured confounders, such as exercise, diet, and sleep information. These factors could also affect the gut microbiota [74‐76], with potential implications for health. Second, although our results indicated that roxadustat significantly altered the gut microbiota, the species accumulation box plot suggested that the sample size was relatively limited. Further studies with larger sample sizes are therefore required. Third, the study enrolled patients from a single dialysis center, thus limiting the generalizability of the results to the overall HD population. Further studies are, therefore, needed to verify the current results.

Roxadustat alleviated renal anemia in patients with ESA resistance by decreasing inflammatory factors and hepcidin levels and improving iron utilization. These effects were at least partly mediated by improved diversity and abundance of SCFA-producing gut bacteria, probably via activation of HIF.