Dnajb11-Kidney Disease Develops from Reduced Polycystin-1 Dosage but not Unfolded Protein Response in Mice

Abstract

Significance statement: Heterozygous DNAJB11 mutation carriers manifest with small cystic kidneys and renal failure in adulthood. Recessive cases with prenatal cystic kidney dysplasia were recently described. Our in vitro and mouse model studies investigate the proposed disease mechanism as an overlap of autosomal-dominant polycystic kidney disease and autosomal-dominant tubulointerstitial kidney disease pathogenesis. We find that DNAJB11 loss impairs cleavage and maturation of the autosomal-dominant polycystic kidney disease protein polycystin-1 (PC1) and results in dosage-dependent cyst formation in mice. We find that Dnajb11 loss does not activate the unfolded protein response, drawing a fundamental contrast with the pathogenesis of autosomal-dominant tubulointerstitial kidney disease. We instead propose that fibrosis in DNAJB11 -kidney disease may represent an exaggerated response to polycystin-dependent cysts.

Background: Patients with heterozygous inactivating mutations in DNAJB11 manifest with cystic but not enlarged kidneys and renal failure in adulthood. Pathogenesis is proposed to resemble an overlap of autosomal-dominant polycystic kidney disease (ADPKD) and autosomal-dominant tubulointerstitial kidney disease (ADTKD), but this phenotype has never been modeled in vivo . DNAJB11 encodes an Hsp40 cochaperone in the endoplasmic reticulum: the site of maturation of the ADPKD polycystin-1 (PC1) protein and of unfolded protein response (UPR) activation in ADTKD. We hypothesized that investigation of DNAJB11 would shed light on mechanisms for both diseases.

Methods: We used germline and conditional alleles to model Dnajb11 -kidney disease in mice. In complementary experiments, we generated two novel Dnajb11-/- cell lines that allow assessment of PC1 C-terminal fragment and its ratio to the immature full-length protein.

Results: Dnajb11 loss results in a profound defect in PC1 cleavage but with no effect on other cystoproteins assayed. Dnajb11-/- mice are live-born at below the expected Mendelian ratio and die at a weaning age with cystic kidneys. Conditional loss of Dnajb11 in renal tubular epithelium results in PC1 dosage-dependent kidney cysts, thus defining a shared mechanism with ADPKD. Dnajb11 mouse models show no evidence of UPR activation or cyst-independent fibrosis, which is a fundamental distinction from typical ADTKD pathogenesis.

Conclusions: DNAJB11 -kidney disease is on the spectrum of ADPKD phenotypes with a PC1-dependent pathomechanism. The absence of UPR across multiple models suggests that alternative mechanisms, which may be cyst-dependent, explain the renal failure in the absence of kidney enlargement.

Graphical abstract

Figure 1.

Dnajb11 loss results in abnormal biogenesis of PC1. (A) Immunoblots of cell lysate with anti-HA, anti-Dnajb11 (left panel), and anti-LRR PC1 N-terminal antibody (7e12) (right panel) show that Dnajb11^−/− cells have significantly reduced PC1-CTF as seen with HA epitope tag and increased PC1 full length—representing a profound alternation in cleavage ratio—and nearly absent mature N-terminal fragment (PC1-NTR). (B) Immunoblot of native PC1 using E8 antibody to PC1-CTF in IMCD3 cells of the indicated genotypes corroborate altered cleavage ratio resulting from loss of Dnajb11. (C) Quantification of PC1-CTF/PC1-FL ratio in biological triplicates of IMCD3 cells on the basis of blot shown in Supplemental Figure 1D. (D) Immunoblots showing unchanged expression of other evaluated cystoproteins in Dnajb11^−/− cells. (E) Agarose gel showing Xbp1 RT-PCR products where WT and Dnajb11^−/− cells have almost entirely unspliced (inactive) Xbp1u and no difference in activation to spliced Xbp1s on treatment with tunicamycin. (F) Immunoblots showing no difference in steady-state expression of UPR sensors/effectors Atf4, Xbp1s, Atf6, or BiP nor in response to tunicamycin treatment between WT and Dnajb11^−/− cells. BiP, binding immunoglobulin protein; IMCD3, inner medullary collecting duct; RT-PCR, reverse transcriptase PCR; UPR, unfolded protein response; WT, wild-type.

Figure 2.

Dnajb11^−/− mice have premature lethality and atypical polycystic kidneys phenotype. (A) Ratio of genotypes of live-born pups to crosses of Dnajb11^+/− breeders on the basis of samples collected from first 16 litters for which samples for all live births were available. Embryo counts are based on embryos obtained at E14–E18. (B) Embryonic kidney histology at approximately E16.5. (C) Body weight and photograph at P18. (D) Representative whole-kidney histologic scan, hematoxylin and eosin–stained kidney sections, and Pico Sirius–stained kidney sections mice of each genotype at P18. (E) Immunofluorescent staining of Dnajb11^−/− kidney sections showing that cysts arise from proximal (Megalin+) and distal nephron (DBA+) segments. (F) Higher magnification of a cyst with atypical epithelium as was seen in a subset of cysts in all Dnajb11^−/− mice. (G) Immunoblot of native PC1 in primary cells from Dnajb11^+/− and Dnajb11^−/− mice at age P21 using E8 antibody to PC1-CTF. (H) BUN and creatinine levels from mice serum. (I) Representative hematoxylin and eosin–stained liver and pancreas sections. (J) qPCR showing genes with significantly increased expression in Dnajb11^−/− kidneys. (K) Immunofluorescent staining showing areas of macrophages between cysts (left panel) and with patchy distribution particularly in the medulla on a low magnification scan (right panel). *indicates p<0.05, **indicates p<0.01 DBA, dolichos biflorus agglutinin; PC1, polycystin-1; PC1-CTF, PC1 C-terminal fragment; qPCR, quantitative PCR.

Figure 3.

Conditional distal nephron epithelial inactivation of Dnajb11 results in a polycystin dosage-dependent polycystic kidney phenotype. (A) Representative whole-kidney histologic scans of indicated genotypes at age 9 weeks. (B) BUN and creatinine levels from mouse serum (9 weeks) show no significant difference in kidney function. (C) Hematoxylin and eosin–stained kidney sections. Genotypes are as indicated by color blocks defined in (A). (D) KW:BW of indicated genotypes at 9 weeks. (E) qPCR of genes characterized in cystic and fibrotic kidney disease models. *indicates p<0.05, **indicates p<0.01 KW:BW, kidney weight-to-body weight ratio; qPCR, quantitative PCR.

Figure 4.

Characterization of interstitial fibrosis and assessment of UPR in Dnajb11 cystic kidneys. (A) Pico Sirius stain of kidney sections with indicated genotypes at 9 weeks. (B) Immunofluorescent staining of Dnajb11^fl/fl; Pkd1^+/−; Ksp-Cre kidney showing the pericystic distribution of macrophages as indicated by F4/80 staining. (C and D) Immunoblots of BiP expression on whole-kidney lysate (9 weeks) from the indicated genotypes. T: WT mouse injected with tunicamycin 6 hours before sacrifice. (E) qPCR for UPR effectors shows no activation of UPR in the 9 weeks Ksp model. Xbp1s primers specifically detect the active isoform. S: To demonstrate sensitivity to detect changes in Ksp-cre model whole-kidney cDNA, 7–9 weeks Sec63fl/fl; Ksp-Cre samples were compared in parallel. The tunicamycin-treated positive control was tested in parallel but not included in statistical analysis. (F) qPCR for UPR effectors in germline model at P18. Genotypes are defined by color blocks in (E). (G) Representative kidney sections with Pico Sirius stain for Dnajb11^fl/fl; Pkd1^+/−; Ksp-Cre (9 weeks) and Pkd2^WS25/− mice (13 weeks) with similar cyst area. (H) Cystic index for the indicated genotypes and ages. Statistical analysis was carried out independently to compare three genotypes at 9 weeks, and the two time points for the cystic mouse genotype. (I) Pico Sirius stain of Dnajb11^fl/fl; Pkd1^+/−; Ksp-Cre kidneys at earlier (3 weeks) and aged (25 weeks) time points show a significant increase in fibrosis with age. (J) Immunofluorescent staining of α-smooth muscle actin in 25 weeks Dnajb11^fl/fl; Pkd1^+/−; Ksp-Cre kidneys colabeled with proximal and distal nephron segment markers LTL and DBA. *indicates p<0.05, **indicates p<0.01, ***indicates p<0.001, ****indicates p<0.0001 BiP, binding immunoglobulin protein; DBA, dolichos biflorus agglutinin; LTL, lotus tetragonolobus lectin; qPCR, quantitative PCR; UPR, unfolded protein response; WT, wild-type.