cMet agonistic antibody prevents acute kidney injury to chronic kidney disease transition by suppressing Smurf1 and activating Smad7

We aimed to investigate the role of cMet agonistic antibody (cMet Ab) in preventing kidney fibrosis during acute kidney injury (AKI) to chronic kidney disease (CKD) transition. Additionally, we explored the effect of cMet Ab on TGF-β1/Smad pathway during the pathogenesis of kidney fibrosis. A unilateral ischemia-reperfusion injury (UIRI) mouse model was established to induce AKI-to-CKD transition. Furthermore, we incubated human proximal tubular epithelial cells under hypoxic conditions as in vitro model of kidney fibrosis. We analyzed the soluble plasma cMet level in patients with AKI requiring dialysis. Patients who did not recover kidney function and progressed to CKD presented a higher increase in the cMet level. The kidneys of mice treated with cMet Ab showed fewer contractions and weighed more than the controls. The mice in the cMet Ab-treated group showed reduced fibrosis and significantly decreased expression of fibronectin and α-smooth muscle actin. cMet Ab treatment decreased inflammatory marker (MCP-1, TNF-α, and IL-1β) expression, reduced Smurf1 and Smad2/3 level, and increased Smad7 expressions. cMet Ab treatment increased cMet expression and reduced the hypoxia-induced increase in collagen-1 and ICAM-1 expression, thereby reducing apoptosis in the in vitro cell model. After cMet Ab treatment, hypoxia-induced expression of Smurf1, Smad2/3, and TGF-β1 was reduced, and suppressed Smad7 was activated. Down-regulation of Smurf1 resulted in suppression of hypoxia-induced fibronectin expression, whereas treatment with cMet Ab showed synergistic effects. cMet Ab can successfully prevent fibrosis response in UIRI models of kidney fibrosis by decreasing inflammatory response and inhibiting the TGF-β1/Smad pathway. comparison between groups including UIRI/vehicle vs. UIRI/cMet Ab groups. One-way ANOVA post-hoc Tukey tests were applied for cell culture related experiments. Detailed statistics are listed in the Supplementary Data. A value of P < 0.05 was considered as a threshold for statistical significance.


Clinical perspectives
-A higher increase in soluble plasma cMet level is a prognostic indicator for AKI-to-CKD transition, and the TGF-β1/Smad signaling pathway has been shown to be involved in hypoxia-induced AKI-to-CKD transition both in vivo in mice and in vitro in human primary tubular cell models.
-Treatment with a cMet agonistic antibody successfully modulated the TGF-β1/Smad signaling pathway (through Smad7 dependent-and independent-mechanisms) via the suppression of Smurf1 and activation of Smad7, thus reversing inflammatory responses and kidney fibrosis.
-The cMet agonistic antibody can prevent AKI-to-CKD transition by suppressing Smurf1 and activating Smad7.

Introduction
Acute kidney injury (AKI) is clinically important and may result in the progression of chronic kidney disease (CKD), thus, worsening the overall prognosis of the patients [1,2]. Evidence indicates that AKI is not just a benign or reversible condition, but can also induce CKD development [3][4][5]. Thus, if the pathophysiological changes associated with AKI are wellmanaged, CKD incidences caused by AKI can be reduced [6]. Kidney fibrosis is an important pathological feature of CKD and is involved in the process of AKI-to-CKD transition [7].
Nevertheless, effective treatment options against the development of kidney fibrosis remain cMet is a transmembrane tyrosine kinase receptor for hepatocyte growth factor (HGF) and is involved in cell survival, growth, and regeneration [9]. The activation of HGF/cMet axis improves kidney diseases by inhibiting oxidative stress, apoptosis, fibrosis, and inflammation [10,11]. In particular, cMet monoclonal antibody has been suggested as a potential therapeutic agent [12]. Recently, we confirmed the involvement of the HGF/cMet pathway in kidney fibrosis and showed that the cMet agonistic antibody (cMet Ab) reduces fibrosis and improves apoptosis in both glomerular endothelial cell model and unilateral ureteral obstruction model [13,14]. However, the inhibitory activity of cMet Ab on kidney fibrosis during AKI-to-CKD transition remains to be fully elucidated.
The aim of this study was to investigate whether the cMet Ab can prevent CKD development by inhibiting kidney fibrosis using a unilateral ischemia-reperfusion injury (UIRI) induced in vivo AKI-to-CKD transition model. We also explored the regulatory mechanisms associated with the TGF-β1/Smad pathway in an in vitro hypoxia-induced kidney cell model and investigated whether cMet Ab treatment can modulate the TGF-β1/Smad pathway and prevent kidney fibrosis.
Four-week AKI-to-CKD transition model was established by UIRI and sham surgery, as reported previously [16]. Briefly, the mice were anesthetized with intraperitoneal injection of Rompun™ (xylazine 10 mg/kg; Bayer korea Co., Ansan, Korea) mixed with Zoletil™ (zolazepam 30 mg/kg; Virbac Korea). During the procedure, the temperature of the animals was maintained at around 37 ℃ by placing them on heating pads. The left kidney of the mice was pulled out following a left flank incision, and the kidney pedicles were exposed to allow easy manipulation. Pedicles containing left renal artery were cross-clamped with proper devices (Roboz Surgical Instrument Co., Gaithersburg, MD) for 27 min. Left kidneys of sham operated groups were also exposed but renal artery clamping was not performed. cMet Ab (VM507), produced by the R&D Center for Innovative Medicines, Helixmith (Seoul, Korea) [14], was intravenously injected via tail vein 1 day before surgery and biweekly thereafter. The mice in the control group were administered with saline following the same schedule as that of treatment group. Mice were sacrificed at post-operative day 28 (4 weeks) followed by harvesting of left kidney. The harvested kidney was cut in transverse direction. The upper half of the kidney was dissected transversely into 3 tissue specimens from top to center and was used for mRNA extraction, frozen tissue specimen, and for estimating protein expression using western blot. Lower half was cut into 2 tissue specimens; upper section was used for preparing paraffin tissue block and lower section was used for estimating protein expression using western blot. Animal experiments were performed three times to evaluate statistical significance. Paraffin embedded kidney tissue sections (4 μm thick) were stained with Masson's trichrome and Sirius red (All from ScyTek, Logan, Utah, USA) to evaluate the extent of tissue fibrosis [16]. For each kidney section, at least 8 fields were randomly selected and photographed using light microscope (BX53F2; Olympus, Tokyo, Japan). The area of fibrosis and total tissue were measured at ×100 magnification using ImageJ 1.52d software (Wayne Rasband, National Institute of Health, USA).

Immunohistochemistry.
Paraffinized kidney tissue blocks were sliced to obtain 4 μm-thick sections, deparaffinized in xylene, and rehydrated in ethanol. Sliced specimens were heated in a microwave oven for 5 min repeatedly 3 times with 10% citrate buffer solution (pH 6.0) to retrieve the antigen.
Endogenous streptavidin activity was blocked with 3% hydrogen peroxide in methanol for 10 min at room temperature. Kidney sections were probed with primary antibodies against TGF-β1, Smad2/3, and Smad7 and incubated overnight at 4 °C. Polink HRP DAB detection kit was used to detect rabbit primary antibodies (GBI Labs, Bothell, WA, USA). The sections were counterstained with Mayer's hematoxylin (ScyTek Laboratories, Logan, UT, USA). The stained slides were evaluated at ×200 magnification and images were captured in at least 5 selected fields. The percentage of TGF-β1, Smad2/3, and Smad7 positive areas were measured by using ImageJ 1.52d software (Wayne Rasband, National Institute of Health, USA).

Western blot analysis.
Kidney tissues harvested 4 weeks after inducing UIRI and human proximal tubular epithelial cells (hPTECs) incubated for 72 hours were homogenized, followed by protein extraction using RIPA buffer containing complete protease inhibitors cocktail (Thermo fisher, Rockford, IL, USA). Protein concentrations were determined using the bicinchoninic acid (BCA) assay (Thermo scientific, Rockford, IL, USA) and equal amounts of protein extracts were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Separated proteins were transferred and immobilized onto the membranes (Millipore Corporation, Bedford, MA, USA). After blocking the non-specific proteins, the membranes were incubated with specific primary antibodies overnight at 4 ℃ to probe target proteins (Supplementary Table 2). Anti-rabbit IgG or anti-mouse IgG antibodies (All from Cell Signaling Technology, Danvers, MA, USA) were used as secondary antibodies. Protein bands were visualized on an Downloaded from http://portlandpress.com/clinsci/article-pdf/doi/10.1042/CS20210013/913136/cs-2021-0013.pdf by guest on 02 June 2021 enhanced chemiluminescence system (Advansta, CA, USA) and quantified using ImageJ 1.52d software (Wayne Rasband, National Institute of Health, USA).

Establishment of the in vitro model of AKI-to-CKD transition.
In this study, we established an in vitro model of hypoxia-induced AKI-to-CKD transition model using hPTECs [17,18]. The protocol for obtaining and processing human kidney specimen was reviewed and approved by the institutional review board of Seoul National University Hospital (IRB no. 1002-045-309). Human donors of kidney specimens provided written informed consent before nephrectomy. Human proximal tubule segments were isolated from the kidneys surgically removed from patients diagnosed with renal cell carcinoma. After dissecting the cortex, the unaffected specimens were minced and digested with Hank's balanced salt solution (HBSS) containing 3 mg/mL collagenase (Sigma-Aldrich, St. Louis, MO, USA) and incubated at 37 °C for 1 hour. The digested kidney cells were washed through a series of sieves (120, 70, and 40 μm in diameter) using phosphate buffer saline, followed by centrifugation at 500 g for 5 min. hPTECs were recovered from the pellet and incubated in DMEM/F12 for 4 hours. Tubules floating in the media were collected and cultured on collagencoated petri dishes (BD Biosciences, Franklin Lakes, NJ, USA) until the establishment of epithelial cell colonies. Cells after 2-3 passages were used in the current study.
One hour before the start of hypoxic treatment, the cells were treated with 0.25, 0.5, or 1 μg/mL cMet Ab or human IgG (R&D Systems, Wiesbaden, Germany). Furthermore, we established oxidative stress-induced cell fibrosis using H2O2. The hPTECs were treated with 0.5 mM H2O2,

Immunofluorescence staining.
Immunofluorescence staining was performed on hPTECs, which were seeded onto 4-chamber slides. After stimulation, the cells were fixed and permeabilized using 4% paraformaldehyde and 0.1% Triton X-100. The cells were then incubated with primary antibodies (Supplementary Table 2) overnight at 4 °C after blocking the slides with 2% bovine serum albumin (BSA). Aldrich, St. Louis, MO, USA). In addition, we performed immunofluorescence staining to detect the level of apoptosis in 4-week-old mouse tissues. The primary antibodies were not considered for negative controls during the staining procedure. The stained slides were scanned by confocal microscopy using a Leica TCS SP8 STED CW instrument (Leica, Wetzlar Germany) The mean values of signal intensity were expressed with the total intensity per region of interest or per field calculated by the MetaMorph image analysis software.

Flow cytometry analysis.
A single-cell suspension was prepared by filtering the homogenate using 40 μm pore cell

Relationship between plasma HGF and cMet concentrations and AKI-to-CKD transition.
A total of 66 AKI patients (among the 131 patients enrolled) survived after three months following the initiation of the CRRT. Of these, for 60 patients, 3-month serum creatinine and estimated GFR related data was available and were considered for further analysis. Thirty patients did not recover to attain normal kidney functions (CKD progressors whose estimated GFR decreased ≥15% from the baseline levels) including the 14 patients who remained on dialysis. CKD progressors showed statistically decreased eGFR at three months compared with that with the CKD non-progressor (25.9 ± 33.3 vs. 80.4 ± 35.2 ml/min/1.73 m 2 , respectively) despite the similar baseline clinical and demographical characteristics (Supplementary Table   5). CKD progressors had lower baseline log-transformed plasma cMet levels than nonprogressors ( Figure 1A, 6.40 ± 0.51 ln(pg/ml) vs. 6.69 ± 0.52 ln(pg/ml), P = 0.034). In addition, change fraction in day 2 plasma log-transformed cMet levels from day 0 were higher among the CKD progressors (100.5% ± 3.9% vs. 98.3% ± 4.2%, P = 0.048). The correlation between change of eGFR from baseline to 3 months and increase of soluble cMet from day 0 to day 2 was statistically significant (P = 0.034). Log-transformed plasma HGF levels were not different between groups at day 0, day 2, and day 7 (data not shown).

cMet agonistic antibody treatment ameliorates kidney fibrosis following AKI.
Body weights of mice were comparable at the time of UIRI induction and sacrifice. Figure   1B shows the representative gross morphology of both kidneys in each group. Compared to the right kidneys, left kidneys had contracted and decreased in size in the UIRI and vehicle treated groups. However, the left kidneys of the UIRI models treated with cMet Ab were relatively Downloaded from http://portlandpress.com/clinsci/article-pdf/doi/10.1042/CS20210013/913136/cs-2021-0013.pdf by guest on 02 June 2021 less contracted than those of vehicle-treated group. Furthermore, the weight of the left kidneys per body weight was significantly higher in the UIRI models treated with cMet Ab than in the vehicle-treated group (P = 0.004). The areas of interstitial fibrosis as observed by Masson's trichrome or Sirius red staining were increased in the UIRI group compared to those in the sham-operated group ( Figure 1C,S1A). The UIRI models treated with cMet Ab showed less degree of fibrosis, and areas of fibrosis were significantly lower in the cMet Ab-treated group (UIRI+cMet Ab vs. UIRI, 3.90% ± 1.74% vs. 20.24% ± 2.57%, P < 0.001 in Masson's trichrome staining and 5.54% ± 1.35% vs. 19.97% ± 4.19%, P = 0.005 in Sirius red staining, Figure 1D). The mRNA expression of α-smooth muscle actin (αSMA) and fibronectin increased in the kidney tissue of UIRI and vehicle-treated groups and decreased following treatment with cMet Ab ( Figure 1E). Furthermore, cMet Ab treatment downregulated the overexpression of COL1A1 and αSMA in the UIRI groups and upregulated the decreased expression of E-cadherin protein ( Figure 1F,G).

Changes in the expression of apoptosis and inflammation-related markers following treatment with cMet agonistic antibody.
We investigated the changes in apoptosis-related markers in the UIRI models treated with and without cMet Ab antibody. The tissue expression of phospho-p21 and cleaved caspase-3 increased in the UIRI group compared with that in the sham group ( Ab treatment. The overexpression of mRNAs coding inflammation-related markers including IL-1β, TNF-α, and MCP-1, and IL-1β in the UIRI and vehicle-treated groups was downregulated following cMet Ab treatment ( Figure 2E,F).

cMet activation controls TGF-β1/Smad pathway in kidney fibrosis.
We investigated whether the increased activity of TGF-β1 in the UIRI models induced AKIto-CKD transition and whether cMet Ab reduced the TGF-β1 activity in the kidney tissues.

Inhibition of Smurf1 and activation of Smad7 as a protective mechanism for kidney fibrosis following cMet agonistic antibody treatment.
Confocal microscopy showed changes in TGF-β1/Smad signaling in hPTECs subjected to hypoxia for 72 hours. Under hypoxic condition, cMet expression was downregulated whereas that of Smurf1 was upregulated when compared to normoxic conditions ( Figure 5A,S4A). The treatment with cMet Ab led to significant overexpression of cMet, whereas that of Smurf1 decreased the expression. In addition, suppressed Smad7 expression in hypoxia was reversed in response to decreased Smurf1 expression following cMet Ab treatment ( Figure 5B,S4B).
Hypoxia-induced Smurf1 activation and Smad2/3 overexpression were inhibited following treatment with cMet Ab (Figure 5C,S4C). The treatment of hPTECs subjected to hypoxia and treated with cMet Ab led to decreased mRNA expression of TGF-β1, Smad2, Smad3, and Smurf1, and increased expression of Smad7 ( Figure 5D). The expression pattern of these genes was further confirmed at the protein level by western blotting following cMet Ab treatment ( Figure 5E,F).   Figure S5). Furthermore, the TGF-β1 and Smad signaling pathways were found to be involved in hypoxia-induced AKI-to-CKD transition in in vivo and in vitro models. cMet Ab simultaneously inhibits Smurf1 and activates Smady7, and these modulations in Smurf1 and Smad7 independently suppress the TGF-β1 and Smad signaling pathway, thereby leading to the reversal of kidney fibrosis. The anti-fibrosis effects of cMet Ab were also proved in the hydrogen peroxide-challenged in vitro human primary tubular epithelial cell models ( Figure S6).
HGF is a pleiotropic protein that binds to its receptor cMet, activates downstream signaling pathways, and performs various biological functions including organ development, cellular proliferation, tissue regeneration, injury repair, and wound healing [19,20]. In cancer cells or malignant tumors, HGF/cMet activation is closely related to cancer progression or metastasis through abnormal cellular proliferation and angiogenesis [21]. HGF/cMet tissue expression levels were elevated, and elevated serum or plasma HGF/cMet levels were associated with Researchers have shown that IRI can result in kidney fibrosis; animal models of chronic IRI can be used to study AKI-to-CKD transition and CKD development as reported by us previously [45,46]. Different mechanisms have been suggested to be involved in the AKI-to-CKD transition including endothelial dysfunction, tubular epithelial injury, interstitial inflammation, and tissue fibrosis [47]. Kidney cell death including apoptosis and necrosis and inflammatory responses contribute to AKI following IRI [42]. Increased inflammatory responses associated with the presence of TGF-β1 in kidney tissue contribute to fibrosis and progression to CKD, even after functional recovery from AKI. In mouse models of cardiac arrest and cardiopulmonary resuscitation, tubular necrosis accompanied by cellular apoptosis and tissue inflammation occurs prior to recovery from AKI, and increased TGF-β1 and tissue fibrosis were observed in the CKD progression group [48]. Negative feedback mechanism of Smad signaling is regulated by E3 ubiquitin ligases including Smurfs. Smurfs bind to Smad7 and down-regulate TGF-β1 receptor via ubiquitinmedicated degradation pathway. In pathological conditions, Smurf/Smad7 complex formation is inhibited, resulting in activation of Smurfs, which down-regulates and degrades Smad7.
During the progression of kidney fibrosis, the negative feedback mechanism is impaired.

Data Availability
All data associated with this study are available in the main text or supplementary materials.

Competing Interests
The authors declare no conflict of interests.