Pharmacological inhibition of MyD88 homodimerization counteracts renal ischemia reperfusion-induced progressive renal injury in vivo and in vitro

The activation of innate immunity via myeloid differentiation factor 88 (MyD88) contributes to ischemia reperfusion (I/R) induced acute kidney injury (AKI) and chronic kidney injury. However, since there have not yet been any effective therapy, the exact pharmacological role of MyD88 in the prevention and treatment of renal ischemia reperfusion injury (IRI) is not known. We designed a small molecular compound, TJ-M2010-2, which inhibited MyD88 homodimerization. We used an established unilateral I/R mouse model. All mice undergoing 80 min ischemia through uninephrectomy died within five days without intervention. However, treatment with TJ-M2010-2 alone significantly improved the survival rate to 58.3%. Co-treatment of TJ-M2010-2 with the CD154 antagonist increased survival rates up to 100%. Twenty-eight days post-I/R of 60 min ischemia without nephrectomy, TJ-M2010-2 markedly attenuated renal interstitial and inhibited TGF-β1-induced epithelial-mesenchymal transition (EMT) of renal tubular epithelial cells. Furthermore, TJ-M2010-2 remarkably inhibited TLR/MyD88 signaling in vivo and in vitro. In conclusion, our findings highlight the promising clinical potential of MyD88 inhibitor in preventing and treating acute or chronic renal I/R injuries, and the therapeutic functionality of dual-system inhibition strategy in IRI-induced AKI. Moreover, MyD88 inhibition ameliorates renal I/R injury-induced tubular interstitial fibrosis by suppressing EMT.

Scientific RepoRts | 6:26954 | DOI: 10.1038/srep26954 interaction is a key step in I/R pathogenesis 10 . Some DAMPs, for example, high-mobility group box 1 11 and heat shock protein 12 , are released from intracellular compartments and play essential roles during renal I/R. Thus, TLR activation is required for downstream signaling events. TLRs are expressed in various cell types, including dendritic cells (DCs) 13 , B cells 14 and kidney tubular epithelial cells 15 . Myeloid differentiation factor 88 (MyD88) is widely used by all TLRs except TLR3 and activates the mitogen-activated protein kinases (MAPKs) and the translocation of nuclear factor-κ B (NF-κ B) from cytosol to nucleus to induce various inflammatory cytokines, for example, interleukin-1β (IL-1β ), interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α ) 16 . Thus, TLR/MyD88 signaling is a key mediator of renal IRI induction. Furthermore, MyD88 is a potential target for preventing and treating renal IRI.
We designed and synthesized a series of small compounds, 2-aminothiazole derivatives, named TJ-M2010 (WIPO Patent Application Number: PCT/CN2012/070811), based on the MyD88 Toll/interleukin-1 receptor (TIR) domain. In this study, we focused on the specific effects of compound TJ-M2010-2 in renal IRI. TJ-M2010-2 alone exhibited powerful protective effects against IRI induced acute kidney injury (AKI) and chronic kidney injury (CKI). The dual system inhibition strategy by using our MyD88 inhibitor (TJ-M2010-2) against the innate immune system and CD154 antagonist (MR1) against the adaptive immune system showed even more powerful protection. Moreover, because epithelial-mesenchymal transition (EMT) plays a role in renal interstitial fibrosis, we suggest that MyD88 inhibition attenuates renal fibrosis by blocking EMT.

TJ-M2010-2-MyD88 TIR domain interactions.
Molecule docking results demonstrated that TJ-M2010-2 localized to the 3D region constructed by box 1, box 3 and α A of the MyD88 TIR domain without the participation of box 2 (Box 1, a portion of the β A strand; box 2, the second part of the BB loop; box 3, the first part of the α E helix 17 ). TJ-M2010-2 interacted with the residues of α A, α E, β C, β D, DD loop and EE loop but not with residues in the BB loop (Fig. 1a). As shown in Fig. 1b, there were 15 interacting sites that TJ-M2010-2 acted with MyD88 TIR domain. The benzene ring of the TJ-M2010-2 benzyl acted on I179, Y187, M178, and I182 of α A, W286 and L193 of α E, V220 of β C (piperazidine ring), I165 and A163 of β A, and L191 of β B (benzene ring) by hydrophobic interactions. The aminothiazole benzene ring of TJ-M2010-2 interacted with F285, L289, and R288 of α E. The TJ-M2010-2 aminothiazole ring acted on I255 of β D and C274 of the DD loop (Y187, I182 and L193 are not shown because they overlap other structures).
Although the BB loop is essential for MyD88 dimerization, our results showed that TJ-M2010-2 interacted with I179 (Poc site residue adjacent to BB loop) through hydrophobic interactions, disrupting its interactions of I179 with L199 and T202, which play a key role in sustaining BB loop stability (Fig. 1c). Therefore, TJ-M2010-2 indirectly targets the BB loop. A much more mobile BB loop made the TIR:TIR domain association unstable 18 .
TJ-M2010-2 alone or with MR1 improves survival rate, prevents renal dysfunction and attenuates pathologic damage after IRI. To investigate the therapeutic effects of TJ-M2010-2 on IRI induced acute and chronic injury, we used different IRI mouse models. As shown in Fig. 2a, all mice from IRI group died within five days without intervention. Strikingly, TJ-M2010-2 treatment alone resulted in a significantly higher survival rate of 58.3%. Interestingly, the dual system inhibitor (TJ-M2010-2 plus MR1) increased the survival rate to as high as 100%, as all of the mice survived 28 days after surgery. Furthermore, serum creatinine (Cr) and blood urea nitrogen (BUN) decreased to normal levels by seven days post-IRI after TJ-M2010-2 or dual system inhibitor treatment (Fig. 2b,c). However, there was no significant difference in serum Cr and BUN levels between the IRI and TJ-M2010-2 groups on day 28 (see Supplementary Fig. S1). We observed severe acute tubular damage in the cortex and medulla of kidneys from the IRI group on day 1, primarily seen as tubular necrosis and cast formation (Fig. 2d). Though the degree of damage was reduced in the MR1 group, there was significantly less damage in the TJ-M2010-2 and TM groups (Fig. 2e). The above results demonstrate that TJ-M2010-2 robustly protects against renal IRI.

TJ-M2010-2 or dual system inhibition ameliorates inflammatory responses after IRI.
To evaluate the protective effects of TJ-M2010-2 or dual system inhibition on IRI-induced inflammatory responses, we measured NF-κ B nuclear translocation, activator protein-1 (AP-1) and cyclooxygenase 2 (COX2) expression, reactive oxygen species (ROS) and cytokine production and myeloperoxidase (MPO) activity. DAMPs released during renal IRI initiate TLR/MyD88 signaling, which leads to NF-κ B nuclear translocation 8 . To test whether TJ-M2010-2 inhibited IRI-induced NF-κ B activity, we performed an electrophoretic mobility shift assay (EMSA) assay to detect nuclear NF-κ B binding to an NF-κ B 3′ biotinylated probe (AGTTGAGGGGACTTTCCCAGGC) and found that IRI markedly increased NF-κ B activation. TJ-M2010-2 or dual system inhibition treatment remarkably inhibited NF-κ B activity (Fig. 3a,b). We next evaluated the inflammatory cytokines AP-1 and COX2. TJ-M2010-2 alone or in combination with MR1 significantly decreased AP-1 and COX2 levels (Fig. 3c,d). To explore how cytokine levels change after IRI, we measured the in vivo serum levels of IL-1β , IL-6, TNF-α and interleukin-10 (IL-10). As shown in Fig. 3e-h, TJ-M2010-2 treatment or dual system inhibition remarkably suppressed IL-1β , IL-6 and TNF-α levels compared to the IRI group. Furthermore, IL-10 levels significantly increased in the TJ-M2010-2 and TM groups. Reperfusion is associated with ROS formation, which is responsible for kidney injury 19,20 . Thus, we examined ROS levels in renal tissues. As shown in Fig. 3i, TJ-M2010-2 treatment or Scientific RepoRts | 6:26954 | DOI: 10.1038/srep26954 dual system inhibition significantly decreased ROS production. To evaluate neutrophil infiltration, we measured MPO activity in kidneys one day after IRI. As shown in Fig. 3j,k, TJ-M2010-2 treatment or dual system inhibition significantly reduced renal MPO activity. These results suggest that TJ-M2010-2 shows strong anti-inflammatory effects after IRI.

TJ-M2010-2 or dual system inhibition attenuates IRI-induced apoptosis.
To explore the anti-apoptotic effect of TJ-M2010-2, we performed a terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay, analyzed Caspase-3 and B cell lymphoma-2 (Bcl-2) expression in kidney tissue and determined whether TJ-M2010-2 protected proximal tubular cells against ischemia-induced cell death. IRI induced apoptosis in large areas all over the kidney, while TJ-M2010-2 or dual system inhibition limited apoptosis to small and scattered areas mainly localized to the junction of the cortex and medulla after IRI (Fig. 4a,b). As shown in Fig. 4c-e, TJ-M2010-2 treatment or dual system inhibition markedly decreased Caspase-3 protein and mRNA levels and Fas and FasL mRNA levels. In contrast, Bcl-2 protein levels increased. Furthermore, TJ-M2010-2 treatment reduced human kidney 2 (HK-2) cell apoptosis, with significant protection at 80 μ M (Fig. 4f,g). These results demonstrate that TJ-M2010-2 has a strong effect on relieving IRI-induced apoptosis by inhibiting Caspase-3 expression, promoting increased Bcl-2 and strongly protecting tubular cells against IRI.

TJ-M2010-2 inhibits DC maturation in vitro and decreases DC-mediated T-cell proliferation.
DCs play a critical role in immune response initiation after IRI 21,22 . To determine whether TJ-M2010-2 affected DCs maturation and subsequent T-cell proliferation, we tested the inhibitory effects of TJ-M2010-2 on in vitro lipopolysaccharides (LPS)-induced DC maturation and T-cell proliferation in mixed lymphocyte reaction (MLR) system. LPS treatment increased DC expression of CD80, CD86 and MHC-II. However, TJ-M2010-2 inhibited DC expression in a dose-dependent manner (Fig. 5a), and 40 μ M TJ-M2010-2 significantly inhibited DC maturation (Fig. 5b). CD4 + and CD8 + T-cell proliferation significantly increased in co-culture with LPS-stimulated DCs and dose-dependently decreased in co-culture with LPS-stimulated DCs pretreated with TJ-M2010-2 ( Fig. 5c,d). Furthermore, the concentration of TJ-M2010-2 used in this study did not directly influence the cell viability of DCs and lymphocytes (see Supplementary Fig. S2). These results demonstrate that TJ-M2010-2 inhibits DC maturation and effectively inhibits DC-induced proliferation of CD4 + and CD8 + T-cells.

TJ-M2010-2 attenuates renal fibrosis after IRI and prevents EMT in vitro.
To evaluate the protective effects of TJ-M2010-2 on renal fibrosis and its mechanism 28 days after surgery, we examined renal interstitial fibrosis, collagen deposition, serum transforming growth factor-β 1 (TGF-β 1 ) and EMT in HK-2 cells. As shown in Fig. 6a-c, the TJ-M2010-2 group showed less tubular atrophy and inflammatory cell infiltration compared to the IRI group. TJ-M2010-2 attenuated fibronectin expression, which is a fibroblast marker. Furthermore, Masson's trichrome staining and collagen IV staining indicated that renal interstitial fibrosis and collagen deposition were significantly decreased upon TJ-M2010-2 treatment compared to the IRI group. We also examined the level of α -smooth muscle actin (α -SMA), a myofibroblast activation marker, which was down-regulated upon TJ-M2010-2 administration. We measured TGF-β 1 serum levels, as it is known to play a . Mice were exposed to IRI for 80 min with uninephrectomy. (a) After the left renal pedicle was clamped for 80 min, mouse survival status was monitored for 30 days. The survival rate was 100% in Sham; 0% in IRI; 15.4% in MR1; 58.3% in TJ-M2010-2; 100% in TM (* p < 0.01 versus Sham; # p < 0.01 versus IRI). (b,c) Six mice were sacrificed for each group. Blood samples were collected on day 1, day 3 and day 7 after reperfusion to measure serum creatinine (Cr) and blood urea nitrogen (BUN) levels (* p < 0.01 versus Sham; # p < 0.01 versus IRI). Results are expressed as mean ± s.d. (d) Renal tissues were collected one day after IRI and stained with Hematoxylin & Eosin (three mice were sacrificed for each group). Original magnification × 200 over five fields. Histological sections of renal tissues are shown. Bar = 800 μ m in all panels. (e) Histogram of tubular necrosis scores (* p < 0.01 versus Sham; # p < 0.01 versus IRI). Results are expressed as mean ± s.d. vital role in renal interstitial fibrosis. TJ-M2010-2 treatment significantly attenuated TGF-β 1 levels compared to the IRI group (Fig. 6d). Moreover, the MyD88 KO group exhibited the best protective effects among the Sham, IRI and TJ-M2010-2 groups (Fig. 6a-d). We next evaluated the EMT process in HK-2 cells. As shown in Fig. 6e, normal HK-2 cells showed a cobblestone morphology, and TGF-β 1 treatment altered the morphology to an elongated shape with filopodial extensions. TJ-M2010-2 administration restored the normal cobblestone morphology. Furthermore, we measured E-cadherin expression and the mesenchymal cell markers vimentin and α -SMA. As  shown in Fig. 6f,g, TGF-β 1 administration significantly decreased E-cadherin expression and increased vimentin and α -SMA expression. However, TJ-M2010-2 incubation reversed the effects of TGF-β 1. Therefore, TJ-M2010-2 ameliorates renal fibrosis by decreasing TGF-β 1 levels and inhibiting EMT.

Body weight and kidney weight measurements 28 days after surgery.
To evaluate the degree of renal atrophy 28 days after IRI, we measured the body weight and kidney weight among different groups. As shown in Table 1, the ratio of left kidney weight to body weight and the ratio of left kidney weight to right kidney weight were significantly greater in the TJ-M2010-2 group compared to the IRI group, while the ratio in the Results are expressed as mean ± s.d. (c) Mice were exposed to IRI for 80 min with uninephrectomy (three mice were sacrificed for each group). Total proteins were extracted from kidney one day after IRI. TLR4, MyD88, p-IRAK4, TRAF6, p-p38, p-JNK and p-ERK protein levels were analyzed by Western Blot (one of three independent experiments). (d) Total proteins were extracted from HK-2 cells exposed to H/R injury. TLR4, MyD88, p-IRAK4, TRAF6, p-p38, p-JNK and p-ERK protein levels were analyzed by Western Blot (one of three independent experiments). (e) Nuclear proteins were extracted from HK-2 cells exposed to H/R injury. EMSA assay was used to detect NF-κ B activity (one of three independent experiments). (f) Densitometric analysis of the NF-κ B band in EMSA. (*p < 0.0001 versus Control; ※ p < 0.01 versus H/R; # p < 0.0001 versus H/R). Results are expressed as mean ± s.d. (g) Three mice were sacrificed for each group. Renal tissues from short-term observation (80 min with uninephrectomy, day 1) and long-term observation (60 min without nephrectomy, day 28) were stained with MyD88 (red) and nucleus (blue).

Discussion
Although the innate immune system initiates the immune responses and plays a critical role in IRI, there is not currently a drug available that targets the innate immune system or that prevents renal IRI. We designed and synthesized TJ-M2010-2 to inhibit innate immunity, and we investigated its anti-IRI effects and underlying mechanisms in a renal I/R mouse model. TJ-M2010-2 prolonged the survival rate, protected renal function and ameliorated the inflammatory responses and apoptosis in the short-term by inhibiting the TLR/MyD88 signaling pathway. TJ-M2010-2 also showed a long-term effect by attenuating renal fibrosis via EMT inhibition.
TLR/MyD88 signaling pathway plays a dominant role in mediating kidney damage induced by I/R 23 . During renal IRI, TLR expression, the indication of innate immune response activation, dramatically increases. Several TLRs have been widely studied in renal IRI, including TLR2 and TLR4 [24][25][26][27] . Leemans et al. found that TLR2 played a crucial role in the induction of inflammatory injury in renal I/R 25 . Li et al. showed that the inhibition of TLR4/MyD88 signaling protected mice against ischemia induced acute kidney injury 26 . All TLRs, except TLR3, need MyD88 as their adaptors 8 . Most TLR ligands bind to individual receptors to promote MyD88 homodimerization and then MyD88 recruits IRAK4, IRAK1, IRAK2, TRAF6 to induce inflammatory responses by activating NF-κ B and MAPKs 8,16 . In addition, several receptors and adapters in the TLR/MyD88 signaling pathway of innate immunity contain a TIR domain, which contains several highly conserved residues. Some of these proteins include TLRs, MyD88, TIR adaptor protein (TIRAP), TIR-domain-containing adapter protein inducing interferon-β (TRIF) and TRIF-related adapter molecule (TRAM) 28 . Therefore, we synthesized TJ-M2010-2  based on the MyD88 TIR domain structure. As we have shown, TJ-M2010-2 acts on α A, α E, β C, β D, DD loop, EE loop and the Poc site residue I179, which alters MyD88 configuration, electron cloud distribution and stability to influence TIR:TIR domain interactions. Our results show that TJ-M2010-2 blocks TLR/MyD88 signaling by affecting MyD88 homodimerization. Further studies to determine whether TJ-M2010-2 also influences MyD88 heterodimerization with TLRs and MAL, especially with TLR2 and TLR4, are required. In I/R, DCs participate in the early phase of IRI and play a central role in mediating renal I/R damage 21,29 by binding kidney endothelium infiltrating the kidneys 30 , and priming the adaptive immune response. The immune response is primarily controlled by TLR/MyD88 signaling in DCs 31 . After TLR engagement in DCs by DAMPs, MyD88 homodimerization leads to downstream signal transduction and subsequent NF-κ B nuclear translocation 32,33 . Thus, the TLR/MyD88/NF-κ B signaling in DCs plays a critical role in the induction of renal IRI and the blockade of that signaling by TJ-M2010-2 effectively protects against I/R induced AKI. Meanwhile, TJ-M2010-2 demonstrated strong inhibitory effects on TLR/MyD88 signaling in HK-2 cells that had undergone H/R injury and mouse renal tissues exposed to I/R injury. We also observed robust and multiple protective effects against renal IRI, for instance, decreased T cell proliferation and ROS production, the rescue of tubular apoptosis in vivo and the protection of HK-2 cells against apoptosis in vitro, and the decrease of pro-inflammatory cytokines, including IL-1β , IL-6 and TNF-α and the increase of anti-inflammatory cytokine IL10 accompanied with decreased neutrophil infiltration.
Although TLR/MyD88 signaling plays a dominant role in initiating the immune response during renal IRI, there are still other signaling pathways that participate in activating the adaptive immune response. Therefore, it is not adequate to target only innate immunity and we must also block adaptive immunity to achieve maximum protection against AKI. Thus, we designed a dual system inhibition protocol against the innate and adaptive immune systems simultaneously by applying TJ-M2010-2 with the CD154 antagonist MR1. Some studies have shown that CD154 plays an important role in IRI in the kidney 34 , heart 35 and liver 36,37 . In our study, the MR1 treatment group exhibited higher survival rates and better renal function compared to the mice in the Sham group, though the protective effects of MR1 were not as strong as those of TJ-M2010-2. To our surprise, 100% of the mice in the TM group survived and had the best renal function, as assessed by Cr and BUN levels. These results verify that targeting both immune systems simultaneously is important when designing new and better therapeutic strategies to prevent AKI.
Renal fibrosis is the final result of CKI in IRI 38 . EMT is an important pathological pathway to tubulointersitial fibrosis development, and TGF-β 1 is the critical profibrotic factor in this process 39 . In this study, TGF-β 1 initiated EMT in vitro, and TJ-M2010-2 treatment blocked EMT. Although some studies have reported that MyD88 inhibition relieved renal fibrosis, the exact mechanism is not clear. We found that the anti-fibrotic effect of TJ-M2010-2 was associated with the amelioration of TGF-β 1 -induced EMT. Therefore, MyD88 plays a critical role in EMT leading to renal fibrosis. MyD88 is a potential novel clinical target to reduce chronic kidney injury.
In summary, our results demonstrate the powerful protection of TJ-M2010-2 against acute and chronic renal IRI by inhibiting the TLR/MyD88 signaling pathway. The dual system inhibition strategy provides new insight into the approaches for treating and preventing AKI. The prevention of renal fibrosis through MyD88 inhibition results from the reduction of TGF-β 1 -induced EMT. Induction of renal IRI model. Renal IRI model was performed as the method reported with some modifications 40 . Briefly, 80 min ischemic time with uninephrectomy was used for short-term observation (less than 7 days) and 60 min without nephrectomy was used for long-term observation (28 days). Day 0 was the day of the operation. The mice were divided into six groups, as follows: Reagents and administration. TJ-M2010-2 was synthesized at the Academy of Pharmacy, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China (WIPO Patent, application number: PCT/CN2012/070811). Nuclear magnetic resonance and high performance liquid chromatography were used to analyze the structure and purity of TJ-M2010-2, respectively. The molecular structure and chemical synthesis of TJ-M2010-2 were shown (see Supplementary Fig. S3). Mice were intraperitoneally injected with 50 mg/kg TJ-M2010-2 dissolved in 0.5% carboxymethyl cellulose with or without MR1 (Bio X Cell, West Lebanon, NH, USA) (200 μ g/day) on day -2, -1, 0 for short-term administration. Furthermore, TJ-M2010-2 was given to mice from day -2 to day 7 for long-term administration.

Methods
Scientific RepoRts | 6:26954 | DOI: 10.1038/srep26954 Measurement of blood and renal parameters. Blood samples were collected from the inferior vena cava and centrifuged (7500 g, 10 min) for serum collection on day 1, day 3, day 7 and day 28. Serum Cr and BUN concentrations were measured by the clinical laboratory of Tongji Hospital (Wuhan, China). Serum IL1β , IL6, TNF-α and IL10 of day 1 and TGF-β 1 of day 28 concentrations were measured by ELISA kits (eBioscience, San Diego, CA, USA) according to the manufacturer's instructions. ROS of day 1 were measured by ROS assay kit (Beyotime Institute of Biotechnology) according to the manufacturer's instructions. Body weight and kidney weight were measured on day 28. Plasmids and cell culture. Flag-MyD88, HA-MyD88 and Flag-tagged control pcDNA 3.1-(Flag-MyD88, HA-MyD88, Flag-con) were generated by Shanghai GeneChem. Co. Ltd. (Shanghai, China). The Flag-con lacked the MyD88 sequence (NM_002468). Human embryonic kidney 293 T (HEK293T) cells (Cell Bank of the Committee of Type Culture Collection of the Chinese Academy of Sciences, Shanghai, China) were cultured in Dulbecco's modified Eagle medium supplemented with 10% fetal bovine serum (both from Gibco, Carlsbad, CA, USA) in a 37 °C humidified atmosphere of 5% CO 2 . HK-2 cells (China Center for Type Culture Collection, Wuhan, China) were cultured in RPMI medium 1640 supplemented with 10% fetal bovine serum (Gibco) in a 37 °C humidified 5% CO 2 atmosphere.

Generation and intervention of bone marrow-derived dendritic cells (BMDCs).
BMDCs were obtained as the method described with some modifications 43,44 . On incubation day 3 and 6, fresh medium containing granulocyte-macrophage colony-stimulating factor (GM-CSF) and IL-4 was added to the plate. MLR. The MLR was performed as the method previously described 44 . After four days reaction, the cells were collected and stained with PE-conjugated anti-CD4 and APC-conjugated anti-CD8 (eBioscience). Cell proliferation was analyzed by FCM and the proliferation index (PI).
Histology, immunohistochemistry and immunofluorescence. Left kidneys were dissected from mice 1 day and 28 day after I/R and fixed in 10% formalin. Formalin-fixed kidneys were embedded in paraffin, and 4 μ m sections were stained with Hematoxylin and Eosin (H&E). Tubular necrosis was semi-quantitatively analyzed according to previously reported methodology 47 . Renal MPO, fibronectin, collagen IV and α -SMA was evaluated by immunohistochemistry as the method previously described 40 . Masson's trichrome staining was used to evaluate renal fibrosis. Twenty-four hours after I/R, cell apoptosis in paraffin sections was analyzed by TUNEL assay with the in Situ Cell Death Detection kit, POD (Roche, Basel, Switzerland) according to the manufacturer's instructions. Histopathology and immunohistochemistry scoring was performed by pathologists blinded to the treatment groups. Frozen (day 1) and formalin-fixed (day 28) kidneys were used to perform immunofluorescence analysis with a MyD88 antibody (Abcam) at 4 °C overnight. A Cy3-conjugated secondary antibody was then added at 37 °C for 50 min. Cell nuclei were stained with DAPI.
Cell hypoxia/reoxygenation (H/R) injury. HK-2 cells were cultured in a 6-well plate at 7.0 × 10 5 cells/ml for 24 h. The cells in the H/R group were induced to become transiently ischemic by culturing them in a 37 °C humidified atmosphere with no oxygen and 5% CO 2 for 5 h, and then cultured under normal conditions for 1 h. The cells in the TJ-M2010-2 group were pre-treated with TJ-M2010-2 (20 μ M or 40 μ M) for 24 h before H/R injury. In vitro EMT reversal assay in HK-2 cells. HK-2 cells were seeded in a 6-well plate at 1.0 × 10 5 cells/ml for 24 h to approximately 60% confluence. The medium was then replaced with serum-free medium. TJ-M2010-2 was added (0 μ M, 20 μ M) to the culture for 30 min, followed by addition of TGF-β 1 (4 ng/ml) for 72 h. Morphological changes were assessed by phase contrast microscopy and E-cadherin, vimentin and α -SMA expression were evaluated by Western Blot.

In vitro apoptosis quantification in HK-2 cells. For in vitro
In vitro MyD88 overexpression assay in HK-2 cells. HK-2 cells were transfected with HA-MyD88 in the presence or absence of TJ-M2010-2 (40 μ M) for 24 h before H/R injury. MyD88 and p-IRAK4 protein levels and NF-κ B nuclear translocation were measured.
Real-time PCR. Total RNA was extracted from renal tissues using Trizol reagent (Invitrogen). Reverse transcription was performed using RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific). Real-time PCR was performed using SYBR ® Green Realtime PCR Master Mix (TOYOBO, Japan) with ABI PRISM ® 7700. The primer sequences are listed as Supplementary Table S1 online. The results were analyzed using the comparative CT method, and actin was used as the housekeeping gene.
EMSA. Nuclear proteins were extracted from the left kidney (harvest on day 1) and HK-2 cells with a Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime Institute for Biotechnology). EMSA was performed with the LightShift Chemiluminescent EMSA Kit (Thermo Scientific, Waltham, MA, USA) according to the manufacturer's protocol.
Statistical Analysis. Data are presented as mean ± standard deviation (s.d.). Survival rates were compared with a log-rank test, and different groups were compared with a Student's t-test or one-way ANOVA, as appropriate. Statistical analysis was performed on Graphpad Prism Software (GraphPad Software, Inc. La Jolla, CA, USA). P values < 0.05 were considered statistically significant.