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

Abstract

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.

Figure 1. TJ-M2010-2 interacts with MyD88 TIR domain.

(a) The crystal structure of most TIR domains contains a five-stranded β-sheet (a–e) surrounded by five α-helices (a–e). TJ-M2010-2 embedded into the MyD88 TIR domain and interacting with the αA, αE, βC, βD, DD loop and EE loop. The MyD88 TIR domain crystal structure based on the Protein Data Bank (PDB ID: 4DOM). The non-bond interaction score was −747.325 kcal/mol. (b) TJ-M2010-2 acted on the indicated MyD88 TIR domain amino acid residues. The red arrow indicates the Poc residue site I179. (c) The dominating interacting site I179 is magnified and marked by the white rectangle with dotted lines. Gray: carbon atom; blue: nitrogen atom; red: oxygen atom; white: hydrogen atom; yellow: sulfur atom. (d) Co-immunoprecipitation assay (Co-IP) of MyD88. HEK293T cells were co-transfected with HA-MyD88/Flag-MyD88 or HA-MyD88/Flag-con. Seven hours after transfection, TJ-M2010-2 was added to the medium. Total protein was extracted 48 hours after transfection. Proteins were incubated with anti-HA antibody and Protein A + G Agarose, and Flag-MyD88 bound to HA-MyD88 was detected by an anti-Flag antibody. As the concentration of TJ-M2010-2 increased (0 μM, 10 μM, 40 μM), the binding capacity between the Flag-MyD88 and HA-MyD88 proteins decreased (Co-IP, lane 2, 3, 4) (one of three independent experiments). (e) Densitometric analysis of Co-IP assays. The inhibition of dimerization of Flag- and HA- MyD88 proteins was dose-dependent, as 91% inhibited at 10 μM TJ-M2010-2 and 99% inhibited at 40 μM compared to 0 μM. The density of each HA-MyD88 lane was divided by that of Flag-MyD88 (*p < 0.01 versus Control; ^#p < 0.01 versus 0 μM). Results are expressed as mean ± s.d. Control: lane 1 in Co-IP; 0 μM: lane 2 in Co-IP; 10 μM: lane 3 in Co-IP; 40 μM: lane 4 in Co-IP.

Figure 2. TJ-M2010-2 alone or in combination with MR1 prolongs survival, protects the renal function, attenuates pathologic damage of mice subjected to renal ischemia reperfusion injury (IRI).

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 μm 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.

Figure 3. TJ-M2010-2 alone or with MR1 attenuates inflammatory responses after IRI.

Mice were exposed to IRI for 80 min with uninephrectomy. (a) Nuclear proteins were extracted from kidney tissues one day after renal IRI and incubated with an NF-κB probe for 25 min (three mice were sacrificed for each group). EMSA assay was used to detect NF-κB activity (one of three independent experiments). (b) Densitometric analysis of the NF-κB band in EMSA. (^#p < 0.01 versus IRI). Results are expressed as mean ± s.d. Blank: no protein was added. (c) Total proteins were extracted from renal tissues one day after IRI. AP-1 and COX2 expression were measured by Western Blot. (one of three independent experiments). (d) Densitometric analysis of Western Blot results. The density of each β-actin lane was divided by that of AP-1 and COX2 (*p < 0.0001 versus Sham; ^#p < 0.0001 versus IRI). Results are expressed as mean ± s.d. (e–h) Serum samples were collected on day 1 after IRI (three of six mice sacrificed for the measurement of renal function for each group). IL1β, IL6, TNF-α and IL10 serum levels were quantified by ELISA. (*p < 0.01 versus Sham; ^#p < 0.01 versus IRI). Results are expressed as mean ± s.d. (i) Kidney tissues were obtained one day after IRI and homogenized to measure ROS formation. (*p < 0.01 versus Sham; ^#p < 0.01 versus IRI). Results are expressed as mean ± s.d. (j) Renal tissues were collected one day after IRI and stained by immunohistochemistry to detect MPO-positive cells (the same mice that were sacrificed for the assessment of pathologic damage were used for each group). Original magnification × 400 over five fields. MPO staining for neutrophil infiltration (brown: MPO-positive cells). Bar = 400 μm in all panels. (k) Semi-quantitative analysis of MPO positive cells. (*p < 0.01 versus Sham; ^#p < 0.01 versus IRI). Results are expressed as mean ± s.d.

Figure 4. TJ-M2010-2 alone or with MR1 decreases IRI-induced apoptosis.

Mice were exposed to IRI for 80 min with uninephrectomy. (a) Renal tissues were dissected one day after IRI and the amount of renal apoptosis was analyzed by TUNEL assay (the same mice that were sacrificed for the assessment of pathologic damage were used for each group). Original magnification × 200 over five fields. Bar = 800 μm in all panels. (b) Semi-quantitative analysis of apoptotic area. The level of apoptosis was expressed as the percent TUNEL-positive area from total area. (*p < 0.01 versus Sham; ^#p < 0.01 versus IRI). Results are expressed as mean ± s.d. (c) Total proteins were extracted from the kidney one day after reperfusion. Caspase-3 and Bcl-2 protein levels were analyzed by Western Blot. (one of three independent experiments). (d) Densitometric analysis of Western Blot results. The density of each lane of β-actin was divided by that of Caspase-3 and Bcl-2 (*p < 0.01 versus Sham; ^#p < 0.01 versus IRI). Results are expressed as mean ± s.d. (e) RNA was extracted from kidneys one day after IRI. Caspase-3, Fas and FasL mRNA levels were measured by Real-time PCR. (*p < 0.0001 versus Sham; ^#p < 0.0001 versus IRI; ^※p < 0.001 versus IRI). Results are expressed as mean ± s.d. (f) HK-2 cells were pre-treated with TJ-M2010-2 (20 μM, 40 μM, 80 μM) and subjected to transient ischemia followed by re-oxygen. HK-2 cells were then stained with annexin V and PI. (one of three independent experiments). (g) Quantitative analysis of the results of FCM. (*p < 0.01 versus Control; ^#p < 0.05 versus 40 μM; ^†p < 0.01 versus 20 μM). Results are expressed as mean ± s.d.

Figure 5. TJ-M2010-2 interferes with DC maturation and decreases T cell proliferation.

(a) Bone marrow cells from BALB/c mice were cultured with GM-CSF and IL-4 to induce the production of BMDCs. Seven days later, DCs were incubated with TJ-M2010-2 for one hour and then stimulated with LPS for 48 h. CD80, CD86 and MHC-II levels were measured by FCM. TJ-M2010-2 inhibited CD80, CD86 and MHC-II levels dose-dependently (one of three independent experiments). (b) Quantitative analysis of the results above. (*P < 0.01 versus Control; ^※p < 0.05 versus LPS; ^#p < 0.01 versus LPS; †p < 0.01 versus 10 μM). Results are expressed as mean ± s.d. Blank: Unstained DCs; Control: DCs stained with CD80, MHC-II, CD86 and CD11c antibodies in the absence of intervention. (c) Lymphocytes as responder cells were obtained from C57BL/6 mice and stained with CFSE. BMDCs as stimulator cells were derived from BALB/c mice and treated with LPS with or without TJ-M2010-2. The lymphocytes were co-cultured with BMDCs at a 10:1 ratio for four days, and the proliferation of CD4⁺ and CD8⁺ T cells was measured by FCM. The proliferation index (PI) was regarded as the parameter (one of three independent experiments). PI is the total number of divisions divided by the number of cells that went into division. (d) Quantitative analysis of the results from MLR. (^*p < 0.01 versus Control; ^※p < 0.05 versus LPS; ^#p < 0.01 versus LPS; ^†p < 0.01 versus 10 μM). Results are expressed as mean ± s.d. Control: CD4⁺/CD8⁺ T-cells stained with CFSE and co-cultured with DCs in the absence of intervention.

Figure 6. TJ-M2010-2 attenuates IRI-induced renal fibrosis and EMT.

Mice were exposed to IRI for 60 min without nephrectomy. (a) Left renal tissues were collected 28 days after IRI and stained with Hematoxylin & Eosin, Masson’s trichrome staining, fibronectin, collagen IV and α-SMA (six mice were sacrificed for each group). Original magnification × 400 over five fields. Bar = 400 μm in all panels. (b) Semi-quantitative analysis of the fibrotic area using Masson’s trichrome staining. (*p < 0.0001 versus Sham; ^#p < 0.0001 versus IRI). Results are expressed as mean ± s.d. (c) Semi-quantitative analysis of positive areas for fibronectin, collagen IV and α-SMA. (^*p < 0.0001 versus Sham; ^†p < 0.05 versus IRI; ^※p < 0.01 versus IRI; ^#p < 0.0001 versus IRI). Results are expressed as mean ± s.d. (d) Serum samples were collected 28 days after IRI (three of six mice sacrificed for renal function measurement for each group). TGF-β₁ levels were quantified by ELISA. (*p < 0.0001 versus Sham; ^※p < 0.01 versus; ^#p < 0.0001 versus IRI). Results are expressed as mean ± s.d. (e) Normal HK-2 cells are shown (left, Control group). HK-2 cells were treated with TGF-β₁ (4 ng/ml) for 72 h (middle, TGF-β₁ group). HK-2 cells were pre-treated with TJ-M2010-2 for 30 min and then TGF-β₁ was given as previously described (right). One of three independent experiments is shown. Original magnification × 100. (f) Total protein was extracted after TJ-M2010-2 and TGF-β₁ treatment. E-cadherin, vimentin and α-SMA protein levels were analyzed by Western blot. (one of three independent experiments). (g) Densitometric analysis of Western Blot results. The density of β-actin in each lane was divided by that of E-cadherin, vimentin or α-SMA (^*p < 0.0001 versus Control group; ^#p < 0.0001 versus TGF-β₁ group). Results are expressed as mean ± s.d.

Figure 7. TJ-M2010-2 down-regulates the TLR/MyD88 signaling pathway.

(a) HK-2 cells were co-transfected with HA-MyD88/Flag-MyD88 or HA-MyD88/Flag-con. Co-IP assays were performed the same as with HEK293T cells (one of three independent experiments). (b) Densitometric analysis of Co-IP assays. The density of each HA-MyD88 lane was divided by that of Flag-MyD88 (^*p < 0.0001 versus Control; ^※p < 0.01 versus 0 μM; ^#p < 0.0001 versus 0 μM). 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-κB activity (one of three independent experiments). (f) Densitometric analysis of the NF-κB 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).

Figure 8. MyD88 DNA reverses the protective effect of TJ-M2010-2.

HK-2 cells were transfected with or without MyD88 DNA followed by H/R injury. (a) Total proteins were extracted from HK-2 cells. MyD88 and p-IRAK4 protein levels were analyzed by Western Blot (one of three independent experiments). (b) Densitometric analysis of Western Blot results. (^*p < 0.0001). Results are expressed as mean ± s.d. (c) Nuclear proteins were extracted from HK-2 cells. An EMSA assay was used to detect NF-κB activity (one of three independent experiments). (d) Densitometric analysis of the NF-κB band in EMSA. (^*p < 0.05). Results are expressed as mean ± s.d.