Podocyte-specific soluble epoxide hydrolase deficiency in mice attenuates acute kidney injury

Abstract

Podocytes play an important role in maintaining glomerular function, and podocyte injury is a significant component in the pathogenesis of proteinuria. Soluble epoxide hydrolase (sEH) is a cytosolic enzyme whose genetic deficiency and pharmacological inhibition have beneficial effects on renal function, but its role in podocytes remains unexplored. The objective of this study was to investigate the contribution of sEH in podocytes to lipopolysaccharide (LPS)-induced kidney injury. We report increased sEH transcript and protein expression in murine podocytes upon LPS challenge. To determine the function of sEH in podocytes in vivo we generated podocyte-specific sEH-deficient (pod-sEHKO) mice. Following LPS challenge, podocyte sEH-deficient mice exhibited lower kidney injury, proteinuria, and blood urea nitrogen concentrations than controls suggestive of preserved renal function. Also, renal mRNA and serum concentrations of inflammatory cytokines IL-6, IL-1β, and TNFα were significantly lower in LPS-treated pod-sEHKO than control mice. Moreover, podocyte sEH deficiency was associated with decreased LPS-induced NF-κB and MAPK activation and attenuated endoplasmic reticulum stress. Furthermore, the protective effects of podocyte sEH deficiency in vivo were recapitulated in E11 murine podocytes treated with a selective sEH pharmacological inhibitor. Altogether, these findings identify sEH in podocytes as a contributor to signaling events in acute renal injury and suggest that sEH inhibition may be of therapeutic value in proteinuria.

Enzymes: Soluble epoxide hydrolase: EC 3.3.2.10.

Keywords: knockout mice; podocyte; proteinuria; soluble epoxide hydrolase.

Figure 1. LPS treatment increases sEH expression in podocytes

A) Immunoblots of sEH, nephrin, and tubulin in total kidney lysates of control (saline-treated) and LPS-treated C57BL/6J wild-type male mice. Representative immunoblots are shown, and each lane represents an animal. Bar graphs represent protein (left panel) and mRNA (right panel) in kidney lysates from control (saline; n=6) and LPS-treated (LPS; n=9) mice and presented as means ± SEM. B) Lysates of podocytes isolated from control and LPS-treated C57BL/6J wild-type male mice were immunoblotted for sEH, nephrin, and tubulin. Representative immunoblots are shown. Bar graphs represent protein expression (left panel) and mRNA (right panel) in podocytes and presented as means + SEM. In A and B *p<0.05, **p<0.01 indicate a significant difference between saline- and LPS-treated mice. C) sEH, nephrin and tubulin protein expression in differentiated murine E11 podocytes without (PBS) and with LPS treatment for the indicated times. Bar graph represents sEH and nephrin expression and presented as means ± SEM from three independent experiments. *p<0.05, **p<0.01 indicate a significant difference between PBS (24 hours) and LPS-treated groups at the shown time versus PBS (0 hour).

Figure 2. Efficient and specific deletion of sEH in podocytes

A) sEH genomic locus and targeting; two loxP sites were designed in an intronic region of the sEH gene. B) Confirmation of sEH floxed and Cre mice by PCR. C) Genomic DNA from tails was amplified by PCR; primers were designed to distinguish the alleles with and without loxP insertions (left), and Cre (right). D) Immunoblots of sEH expression in podocytes, epididymal fat, liver and muscle of control (Ctrl) and pod-sEHKO (KO) mice. Representative immunoblots are shown. E) Ephx2 expression in podocytes of control (Ctrl; n=6) and pod-sEHKO (KO; n=6) mice. **p<0.01 indicates a significant difference between Ctrl and KO. F) Co-immunostaining of sEH (red) and nephrin (green) in kidney paraffin sections of Ctrl and KO mice. Scale bar: 200μm.

Figure 3. Podocyte sEH deficiency attenuates LPS-induced proteinuria

Body (A) and kidney (B) weights, and kidney to body weight ratio (C) of control (Ctrl) and pod-sEHKO (KO) mice without (saline) and with LPS treatment. Urinary proteins (D), albumin to creatinine ratio (E), BUN (F), serum albumin (G) and creatinine (H) from control and pod-sEHKO mice without (saline) and with LPS treatment (n=9 per group). *p<0.05, **p<0.01 indicate a significant difference between mice without and with LPS treatment, and †p<0.05, ††p<0.01 indicate a significant difference between pod-sEHKO and control mice. Albumin to creatinine ratio (I) and BUN (J) from control and pod-sEHKO female mice (n=5–6/group) after anti-GBM serum injection. *p<0.05 indicates a significant difference between pod-sEHKO and control mice. K) PAS-stained kidney sections of Ctrl and pod-sEHKO mice without and with LPS treatment. Arrows indicate dilated Bowman space and asterisks indicate dilated tubules. Scale bar: 200μm. Images in the right panel are a magnification of the boxed regions. L) Immunostaining for nephrin (green), and DAPI (blue) in kidney sections of control Ctrl and pod-sEHKO mice without and with LPS treatment. Scale bar: 50μm.

Figure 4. Decreased LPS-induced inflammatory response in podocyte sEH-deficient mice

Renal mRNA (A) and serum (B) levels of IL-1β, IL-6 and TNFα in saline- and LPS-treated control (n= 6) and pod-sEHKO (n= 6) mice. Serum IL-6 concentration was very low in both control and pod-sEHKO mice before LPS challenge. C) Kidney lysates from control and pod-sEHKO mice without (saline) and with LPS treatment were immunoblotted for pIKKα, pIκBα, pNF-κBp65, their respective unphosphorylated proteins, and Tubulin as a loading control. Representative immunoblots are shown. Bar graphs represent pIKKα/IKKα, pIΚBα/IΚBα, pNF-κBp65/NF-κBp65 and NF-κBp50/Tubulin as means ± SEM. D) Kidney lysates from control and pod-sEHKO were immunoblotted for pJNK and pp38 and their respective unphosphorylated proteins. Representative immunoblots are shown. Bar graphs represent pJNK/JNK and pp38/p38 as means ± SEM. *p<0.05, **p<0.01 indicate a significant difference between mice without (saline) and with LPS treatment, and †p<0.05, ††p<0.01 indicate a significant difference between pod-sEHKO and control mice.

Figure 5. Decreased LPS-induced ER stress in podocyte sEH-deficient mice

Immunoblots of key proteins in ER stress signaling in kidney lysates from saline- and LPS-treated control and pod-sEHKO mice (n= 6/group). Representative immunoblots are shown. Bar graphs of pPERK/PERK, peIF2α/eIF2α, pIRE1α/IRE1α, sXBP1/Tubulin, and cleaved caspase3/Tubulin are presented as means ± SEM. *p<0.05, **p<0.01 indicate a significant difference between mice without (saline) and with LPS treatment, and ††p<0.01 indicates a significant difference between pod-sEHKO and control mice.

Figure 6. Pharmacological inhibition of sEH attenuates LPS-induced inflammatory and ER stress signaling in immortalized podocytes

Immunoblots of key proteins in inflammation, MAPK, and ER stress signaling in differentiated E11 podocytes without (− LPS) and with (+ LPS) treatment for 24h and without (−) and with (+) sEH pharmacological inhibition (sEHI). Representative immunoblots are shown. Bar graphs of pIKKα/IKKα, pIΚBα/IΚBα, pNF-κBp65/NF-κB, NF-κBp50/Tubulin, pJNK/JNK, pp38/p38, pPERK/PERK, peIF2α/eIF2α, pIRE1α/IRE1α, sXBP1/Tubulin, and cleaved Caspase3/Tubulin are presented as means ± SEM from three independent experiments. **p<0.01 indicates a significant difference in cells without and with LPS treatment. †p<0.05, ††p<0.01 indicate a significant difference between non-treated and sEHI-treated cells.