RIPK3: A New Player in Renal Fibrosis

Chronic kidney disease (CKD) is the end result of a plethora of renal insults, including repeated episodes of acute or toxic kidney injury, glomerular, or diabetic kidney disease. It affects a large number of the population worldwide, resulting in significant personal morbidity and mortality and economic cost to the community. Hence it is appropriate to focus on treatment strategies that interrupt the development of kidney fibrosis, the end result of all forms of CKD, in addition to upstream factors that may be specific to certain diseases. However, the current clinical approach to prevent or manage renal fibrosis remains unsatisfactory. The rising importance of receptor-interacting serine/threonine-protein kinase (RIPK) 3 in the inflammatory response and TGF-β1 signaling is increasingly recognized. We discuss here the biological functions of RIPK3 and its role in the development of renal fibrosis.


INTRODUCTION
Chronic kidney disease (CKD) is defined as a loss of glomerular filtration and or proteinuria, persisting for at least 3 months or structural abnormalities in the kidney. In the majority of cases, CKD eventually leads to end-stage kidney disease (ESKD) requiring renal replacement therapy or death will ensue. CKD affects a large proportion of the population and considerably more than is widely appreciated by the general public. In 2016-2017, 1.8 million hospitalizations in Australia were associated with CKD, which accounts for 16% of all hospital admissions in Australia. Of those hospitalizations, 80% were for regular dialysis (AIHW, 2019). Having CKD increases the length of stay, cost, and complications of non-CKD related hospital admissions. In the United States, the overall prevalence of CKD in the general adult population was 14.8% in 2011(United States Renal Data System, 2018. Regardless of the cause of the initial renal injury, progressive renal fibrosis is common to all forms of CKD, characterized pathologically by extracellular matrix (ECM) accumulation, myofibroblast activation, and inflammatory cell infiltration (Lee and Kalluri, 2010;Carew et al., 2012).
To date, inhibition of the renin-angiotensin-aldosterone system (RAAS) is the crucial strategy utilized to slow deterioration of renal functional decline. However, this influences intrarenal and extrarenal hemodynamics, and only secondarily reduces the development of renal fibrosis. It is primarily beneficial in patients with proteinuric renal disease, and at best, it delays the time to ESKD, leading to renal replacement therapy or death, by a factor of months. More recently, sodiumglucose linked transport inhibitors have been shown to reduce the development of end-stage kidney disease in patients with diabetic kidney disease (Ingelfinger and Rosen, 2019), and a recent trial in both diabetic and non-diabetic CKD was prematurely terminated in light of positive results in favor of the SGLT2 inhibitor (ClinicalTrials.gov, 2020). However, a treatment gap remains, and novel therapies directed toward reducing the ultimate fibrotic response in the kidney are urgently needed to arrest the progression of CKD and improve the outcome of patients.
TRANSFORMING GROWTH FACTOR BETA-1 (TGF-β1) TGF-β is the prototype of a family of secreted polypeptide growth factors. Three isoforms of TGF-β have been identified in mammals, including TGF-β1, TGF-β2 and TGF-β3 (Yu et al., 2003). All TGF-βs are synthesized as homodimeric proproteins together with the latency-associated peptide (LAP), which binds to the TGF-β homodimer to promote the formation of the latent TGF-β binding protein (LTBP) (Robertson et al., 2015). The synthesized complex, consisting of TGF-β dimer, LAP dimer, and LTBP, remains inactive and stored in the ECM (Hinz, 2015). LTBP serves as a localizer to interact with the ECM (Annes et al., 2003). LAP inhibits TGF-β activity by preventing TGF-β binding to its receptors (Annes et al., 2003;Hinz, 2015). This mechanism controls free and, therefore, active TGFβ tissue levels. To cleave the TGF-β complex into the active component and release active TGF-β, one or more of a wide range of proteases, including plasmin, matrix metalloproteinase (MMP) 2, and MMP9, thrombospondin, integrins, and the cationic independent mannose 6 phosphate receptor, are needed (Annes et al., 2003).
It is well accepted that overexpression of active TGF-β1 induces a fibrotic response in multiple organs, including the kidney (Sanderson et al., 1995). TGF-β1 is a well-characterized key mediator in the pathogenesis of tubulointerstitial fibrosis, due to its direct and indirect effect on various cells types (Roberts, 1998;Wang et al., 2005;Bottinger, 2007). The direct action of TGF-β1 includes the transition of cells to a fibroblastic phenotype and synthesis of profibrotic proteins, such as collagens and fibronectin (Border et al., 1990;Haberstroh et al., 1993;Wilson et al., 1993). TGF-β1 also facilitates an indirect fibrotic response, via accelerating apoptosis of resident healthy cells and promoting resident and infiltrating cells to increase ECM deposition (Lebrin et al., 2005;Das et al., 2014;Mack and Yanagita, 2015). Inhibiting TGF-β1 in animal models of kidney disease attenuates fibroblast activation and ECM accumulation (Moon et al., 2006;Russo et al., 2007;Murphy et al., 2012;McGaraughty et al., 2017).
The central signal transduction in response to TGF-β1 is mediated by two specific receptors, TGF-β type II receptor (TGFβRII) and the TGF-β type I receptor (TGFβRI). TGF-β1 firstly binds with TGFβRII in an active form (homodimers) and recruits the low-affinity receptor (TGFβRI) by the ligandbound high-affinity receptor (TGFβRII) (Groppe et al., 2008). The activation of TGFβRI initiates receptor signaling (Xu et al., 2012a) and phosphorylates the substrates, the Smad proteins. Specifically, TGF-β1 signaling stimulates receptorregulated Smad (R-Smad) phosphorylation. This is followed by translocation of R-Smads and the common mediator Smad (Co-Smad) complexes in the nucleus to regulate gene transcription (Moustakas et al., 2001). By contrast, inhibitory Smads (I-Smads) antagonize the activity of the R-Smads by preventing phosphorylation of R-Smads (Hill, 1999).
Relative to Smad3, the function of Smad2 in renal fibrosis is not fully elucidated. Because of the unavailability of Smad2 knock out (KO) mice, conditional kidney tubular epithelial cells Smad2 KO mice were generated by crossing the Smad2 floxed mouse with the kidney-specific promoter (Cadherin 16)-driven Cre transgenic mouse (Shao et al., 2002). Unexpectedly, deletion of Smad2 in tubular cells significantly enhances fibrosis, with an associated elevated Smad3 signaling in the UUO mouse model (Meng et al., 2010). Similarly, Smad2-/-fibroblasts have an increased fibrotic response (Meng et al., 2010). Additional evidence has shown that Smad3, but not Smad2, mediates fibrotic process (Wang et al., 2006;Yang et al., 2009Yang et al., , 2010Chung et al., 2010;Zhou et al., 2010). Hence Smad2 and Smad3 may have distinct roles in mediating the fibrosis upon exposure to TGF-β1.

I-SMADS
Smad 6 and Smad7 are inhibitory mediators in the TGF-β1 signaling pathway. They provide a negative feedback loop through multiple mechanisms, including competing with R-Smads in activating the receptors by associating directly with TGFβRI (Hanyu et al., 2001;Nakayama et al., 2001), indirectly affecting the activity of TGFβRI by cooperation with BMPs (Murakami et al., 2003;Yan et al., 2009), interference in the formation of R-Smad/Co-Smad complex (Hata et al., 1998;Yan et al., 2016) and abolishing transcription in the nucleus (Lin et al., 2003;Zhang et al., 2007).
The deletion of Smad7 accelerates fibrogenesis in a number of mouse models, including UUO , diabetic (Chen et al., 2011), and hypertensive nephropathy (Liu et al., 2013). However, the importance of Smad6 in renal fibrogenesis is unclear.
TGF-β1 increases phosphorylation of tyrosine residues on TGFRs (I and II) and recruits ERK through Ras, Raf, and their downstream MAPK cascades. Specifically, ERK regulates target gene transcription through its downstream transcription factors in conjunction with Smads to control epithelial-mesenchymal transition (EMT) (Lee et al., 2007). ERK also regulates the activity of R-Smads, including Smad1, Smad2, and Smad3 (Kretzschmar et al., 1997(Kretzschmar et al., , 1999Funaba et al., 2002;Matsuura et al., 2005). Moreover, ERK is involved in the autoinduction of TGF-β1 via distinct transcriptional and translational mechanisms in tubular epithelial cells (Zhang et al., 2006). These studies suggest a dominant role of ERK in the non-Smad mediated transduction of TGF-β1.

RECEPTOR-INTERACTING SERINE/THREONINE-PROTEIN KINASE (RIPK) 3
The RIP kinase family contains seven members, each of which possesses a homologous kinase domain. To date, the functions of RIPK4-7 are poorly understood . RIPK2 is a mediator of mucosal immunity. Extensive studies have clarified the importance and physiological roles of RIPK1 and RIPK3 in inflammation and cell death (Christofferson et al., 2014;Newton, 2015).
To date, several phosphorylation sites of RIPK3 have been identified. The human Ser227 site (Thr231/Ser232 for mouse RIPK3) and Ser199 site (Ser204 in mouse) are particularly crucial for the activation of its downstream substrate in the necroptosis pathway, mixed-lineage kinase domain-like (MLKL) (He et al., 2009;Sun et al., 2012;Chen et al., 2013;McQuade et al., 2013).
In contrast to the obligatory involvement of RIPK3, RIPK1 is not always required to cause necroptosis. The M45-mutant strain of murine cytomegalovirus (MCMV) infection causes RIPK3 activation in the absence of activation of RIPK1 (Upton et al., 2010(Upton et al., , 2012. There is also evidence that the RHIM-containing protein (ICP) 6 protein of herpes simplex virus 1 recruits RIPK3 directly and independent of RIPK1 (Wang et al., 2014b). In addition, RIPK1 may have dual influences on cell death by both promoting necroptosis and protecting cells from death under certain conditions (Filliol et al., 2017).

RIPK3 IN KIDNEY FIBROSIS
To date, few studies have investigated the role of RIPK3 in kidney fibrosis, and most of them have not dissected the role of RIPK3 from necroptosis. The majority of these studies used acute injury models where there is known increased necroptosis.
Necrostatin-1 reduces interstitial fibrosis in a mouse model of UUO (Xiao et al., 2017) by inhibiting necroptosis, associated with lower protein and mRNA expression of RIPK, RIPK3, and MLKL and TGF-β1. In parallel, collagen accumulation and fibroblast activation (Xiao et al., 2017) were reduced. This study showed the integral relationship between necroptosis and TGF-β1 activation leading to renal fibrosis.

RIPK3 IN APOPTOSIS AND INFLAMMATION
Under certain conditions, RIPK3 also serves as a pro-apoptosis adaptor, which recruits RIPK1 and Fas-associated protein with death domain (FADD) to activate caspase 8 and thus apoptosis. This effect relies on the involvement of caspase 8 when RIPK3 is inactive, or MLKL is absent (Mandal et al., 2014;Newton et al., 2014). RIPK3 deficient animals develop normally, whereas mice expressing catalytically inactive RIPK3 D161N die around embryonic day 11.5 from increased RIPK1-and caspase-8dependent apoptotic injury (Newton et al., 2014). Similar effects are observed in a study using compounds that selectively inhibit RIPK3 interaction with caspase 8 independent of pro-necrotic kinase activity and MLKL (Mandal et al., 2014).
Recent studies also identify that RIPK3 is an essential mediator in NOD-, LRR-and pyrin domain-containing protein (NLRP) 3 inflammasome formation (Wang et al., 2014a;Lawlor et al., 2015;Chen et al., 2018;Guo et al., 2019). In LPS-treated mouse bone marrow-derived dendritic cells, activation of the NLRP3 inflammasome was initiated by necroptosis (Kang et al., 2014). In a podocyte cell line, the RIPK3 specific inhibitor GSK'872 blocked both the necroptosis pathway and the NLRP3 inflammasome activation (Guo et al., 2019). These indicate the RIPK3 mediated NLRP3 inflammasome can be dependent of the necroptosis. Specifically, RIPK3-MLKL triggers NLRP3 activation when the activation of caspase 8 is reduced (Lawlor et al., 2015). In this setting, RIPK3 activity is required. RIPK3 can also promote NLRP3 inflammasome independent of the MLKL and RIPK3 kinase activity when caspase 8 is active (Lawlor et al., 2015). Collectively, RIPK3 mediated NLRP3 activation can be in both a necroptosis-independent and -dependent manner, depending on the levels of caspase-8 activity.

IMPLICATIONS FOR ANTI-FIBROTIC THERAPY
TGF-β1-specific, humanized, neutralizing monoclonal antibody added to RAAS inhibitors failed to slow the progression of diabetic nephropathy (Voelker et al., 2017). Therefore, targeting the full spectrum of downstream TGF-β1 signaling to prevent renal disease is unlikely to be fruitful, and the development of blockers of more targeted downstream pathways, such as the RIPK3/necroptotic pathway may be more beneficial.
To date, several small-molecule compounds Fauster et al., 2015;Martens et al., 2017Martens et al., , 2020Park et al., 2018;Pan et al., 2019;Hart et al., 2020) have been developed to inhibit RIPK3 activity, providing an impressive toolbox for the investigation of the role of RIPK3 in diverse tissues. These inhibitors of RIPK3 can be divided into three types: ATP mimetic inhibitors targeting the active ATP-binding site in the kinases located between two catalytic domain lobes (type I), targeting the inactive states (type II), and unclassified inhibitors (Muller et al., 2015;Martens et al., 2020; Table 1).
GSK'872 is the most widely used cell-permeable inhibitor of the RIPK3-selective kinase, with >1,000-fold selectivity over a vast majority of more than 300 other kinases (Kaiser et al., 2013). GSK'872 binds the kinase domain and inhibits kinase activity with high specificity, targeting a broad range of pronecrotic stimuli (Mandal et al., 2014) and has been used to specifically inhibit RIPK3 (Lu et al., 2017;Chen et al., 2018;Imamura et al., 2018).
The serine/threonine kinase B-Raf V600E inhibitor dabrafenib is the only type I RIPK3 inhibitor approved for clinical use (Rheault et al., 2013;Li et al., 2014;Sugaya et al., 2019). Previous studies have reported that dabrafenib is a selective RIPK3 inhibitor in various models, including human hepatocytes , mouse models of acetaminophen-caused liver injury , and ischemic brain injury (Cruz et al., 2018). In addition, dabrafenib is a well-known inhibitor of B-Raf, which suppresses the downstream Ras/Raf/ERK/MAPK pathway (Spagnolo et al., 2014), which has been approved for clinical use for the treatment of non-small cell lung cancer expressing B-Raf V600E mutations and in melanoma (Odogwu et al., 2018). Inhibition of Raf kinase has found to attenuate renal fibrosis (Xu et al., 2012b;Chen et al., 2019).
Collectively, inhibition of RIPK3 is a promising anti-fibrotic strategy. RIPK3 facilitates necrosome and necroptosis. RIPK3 stimulates downstream activation of TGF-β1 cascades and regulates TGF-β1 transcription through NLRP3 inflammasome activation. Given inhibitors of RIPK3 are already approved for use in patients with non-small cell lung cancer and melanoma, an accelerated route to market in patients with CKD should be available if early phase clinical studies prove positive.

AUTHOR CONTRIBUTIONS
YS wrote the manuscript. YS, CP, XC, and CH provided the critical discussion of the manuscript. YS and CP revised the manuscript. Figure 1 was created with BioRender.com.