Cold Shock Domain Protein DbpA Orchestrates Tubular Cell Damage and Interstitial Fibrosis in Inflammatory Kidney Disease

Abstract

DNA-binding protein A (DbpA) belongs to the Y-box family of cold shock domain proteins that exert transcriptional and translational activities in the cell via their ability to bind and regulate mRNA. To investigate the role of DbpA in kidney disease, we utilized the murine unilateral ureter obstruction (UUO) model, which recapitulates many features of obstructive nephropathy seen in humans. We observed that DbpA protein expression is induced within the renal interstitium following disease induction. Compared with wild-type animals, obstructed kidneys from Ybx3-deficient mice are protected from tissue injury, with a significant reduction in the number of infiltrating immune cells as well as in extracellular matrix deposition. RNAseq data from UUO kidneys show that Ybx3 is expressed by activated fibroblasts, which reside within the renal interstitium. Our data support a role for DbpA in orchestrating renal fibrosis and suggest that strategies targeting DbpA may be a therapeutic option to slow disease progression.

Keywords: cold shock proteins; kidney fibrosis; kidney injury.

Figure 1

Summary of the YBX3 gene expression data of microdissected glomerular (Glom.) and tubulointerstitial (Tubulo.) compartments from patients with different kidney diseases. Data are presented as a heatmap of the log fold change. Red-marked genes mean upregulated and blue-marked genes mean downregulated in comparison to the control (living donor). Asterisks indicate a q-value < 5% and are therefore to be regarded as significantly regulated. Diabetic nephropathy (DN; Glom: n = 14, Tub: n = 18); hypertensive nephropathy (HTN; Glom: n = 15, Tub: n = 21); minimal change disease (MCD; Glom: n = 14, Tub: n = 15); rapidly progressive glomerulonephritis (RPGN; Glom: n = 23, Tub: n = 21); controls (living donors (LD); Glom: n = 41. Tub: n = 42).

Figure 2

Deletion of exons 2–5 within the Ybx3 gene leads to the formation of an alternative translation/splice product. (A) Genotyping of tail biopsies showed the presence of the expected PCR products at ~400 bp and ~600 bp for wild-type and knockout mice, respectively. (B) DbpA protein expression was assessed via Western blotting of testis tissue using an anti-MSY4 antibody recognizing the C-terminal domain of DbpA (residues 249 to 263) [33]. (C) PCR amplification of the whole Ybx3 coding sequence detected modified mRNA products from the knockout mice (KO_a and KO_b). Sequence analyses identified two protein products in knockout mice. One was previously reported as CSDA CRA_a (denoted KO_a), and the second is a fusion of the Ybx3 N-term (exon 1) with parts of CSDA CRA_a (denoted KO_b). (D) The cartoon depicts the relationship between the exons and coding sequences (filled boxes). The cold shock domain (CSD) is indicated as well as the N- and C-terminus. The DbpA isoforms differ by the presence or absence of the alternative domain (AD) encoded in exon 6. The 3′ untranslated region (UTR) encoded in exon 10 is indicated by a line. Alternatively translated exons are indicated with stripped bars. (E) The KO-a and KO-b fragments were subcloned into the pEGFP vector and transiently expressed in HEK293 cells. Western blot analysis detected both KO_a (37 kDa) and KO_b (40 kDa and 48 kDa) proteins in transfected HEK293 cells. KO_b could be detected with an anti-DbpA antibody directed against epitopes within the protein N-terminus. The lower MW band may result from alternative splicing.

Figure 3

Ybx3 knockout mice develop normally. (A) The body weight of wild-type (black) and Ybx3-deficient mice (white) was comparable at 3 and 6 months of age. (B) The weight of the kidneys at 3 and 6 months of age is shown after normalization to the body weight. (C) Photographs of whole kidneys from wild-type and Ybx3-deficient animals showed no morphological differences. (D) Kidney function (i.e., plasma creatinine levels) was similar in wild-type and knockout animals. (E) Periodic acid–Schiff (PAS) staining of kidney tissue slices revealed no structural differences between wild-type and Ybx3 knockout mice; scale bars represent 200 µm and 50 µm. The number (F) and size (G) of the glomeruli per visual field were comparable between wild-type and knockout animals. (H) Blood counts revealed no alterations in the cell number between wild-type and knockout animals.

Figure 4

The DbpA protein is induced following ureter ligation. To investigate the role of DbpA in tubulointerstitial nephritis, we utilized the murine unilateral ureteral obstruction (UUO) model. (A) Western blot analysis of contralateral or ligated kidney lysates revealed massively upregulated DbpA expression following UUO in wild-type animals on day 14. PDGFR-β, E-cadherin, and β-catenin served as markers for successful disease induction, while GAPDH served as a loading control. Apparent molecular weights are indicated in kDa. (B) Immunohistochemistry using the N-term DbpA antibody showed profound induction of DbpA protein expression in the interstitium of the kidney in UUO compared to the unaffected contralateral kidney tissue. Staining of the blood vessels served as a positive control in healthy kidneys. Scale bars correspond to 200 μm and 50 μm as indicated. (C) Single-cell RNA transcriptomics data were analyzed using the KIT software (http://humphreyslab.com/SingleCell/, accessed on 22 April 2022) [42,43]. Expression data for Ybx3 in healthy mouse kidneys before and after 14 days of UUO are shown. Abbreviations: Pod—podicyte; MC—mesangial cell; EC—endothelial cell; PT—proximal tubule; LH(DL)—loop of Henle descending loop; LH(AL)—loop of Henle ascending loop; DCT—distal convoluted tubule; CNT—connecting tubule; CD-PC—collecting duct-principal cell; IC-A—intercalated cell type A; IC-B—intercalated cell type B; MØ—macrophage; Dediff. PT—dedifferentiated PT; Prolif. PT—proliferating PT; Act. Fib.—activated fibroblast; JGA—juxtaglomerular apparatus; DL+tAL—descending loop + thin ascending loop. (D) Automated multidimensional fluorescence microscopy of diseased (UUO) kidney cryosections from wild-type and Ybx3-deficient mice. The architecture of the kidney was visualized using collagen IV staining (white) to show the extracellular matrix; vimentin staining (yellow) depicts the glomeruli, and propidium iodide (PI, gray) staining depicts the nuclei. Tissue-resident dendritic cells were visualized using CD11c (cyan) and infiltrating T cells CD3e (red). A scale bar of 30 μm is indicated. (E) Immunohistochemistry of collagen type I served as a marker of renal fibrosis. Scale bars corresponding to 200 μm and 100 μm are indicated. Quantification is provided for wild-type and knockout animals (*** p < 0.001).

Figure 5

UUO-dependent tubular cell damage is not altered in Ybx3-deficient animals; however, immune cell infiltration is reduced. (A) Periodic acid–Schiff (PAS) staining of obstructed kidneys on days 5 and 14 showed no observable difference in the loss of brush border membranes or tubular dilation. In Ybx3-deficient mice, tubular cell death was determined by counting the number of cells per tubule. Scale bars represent 200 μm and 100 μm as indicated. Quantification is provided for wild-type and knockout animals. (B) Flow cytometry analyses of infiltrating immune cells on days 5 and 14 are shown for CD45+ leukocytes, myeloid cells (CD45+/CD11b+), macrophages (CD11b+/CD11c+/CD64+), NK cells (CD11b+/CD11c+/MHCII-), Ly6G+ neutrophils, and T cells (CD45+/CD3+) from both wild-type and knockout animals (* p < 0.05, ** p < 0.01, *** p < 0.001).

Figure 6

UUO-dependent renal tubulointerstitial fibrosis is reduced in Ybx3-deficient mice. Representative images of obstructed kidneys on day 14 in wild-type and Ybx3 knockout mice stained for (A) tenascin-C and (B) α-smooth muscle actin (αSMA) are shown. Scale bars represent 100 μm and 50 µm. Quantification was performed by assessing the positively stained cortical area (%) in wild-type and knockout animals. Data obtained via computer-based morphometric analysis on day 14 after UUO induction in obstructed and ‘healthy’ contralateral kidneys showed decreased α-smooth muscle actin expression in fibrotic material in the Ybx3 knockout compared with wild-type animals. Mean values are shown with red bars (* p < 0.05, *** p < 0.001).

Figure 7

Collagen I and III synthesis is reduced in Ybx3 knockout animals compared to wild-type animals. (A) Total collagen levels were estimated by determining the hydroxyproline content of the tissue for wild-type and knockout animals. (B) Sirius red staining of healthy and obstructed kidneys on day 14 from wild-type and Ybx3 knockout mice (scale bars, 100 μm). (C) Quantification was performed by assessing the positively stained cortical area (%) for wild-type and knockout animals. The diagram shows data obtained via computer-based morphometric analysis on day 14 after UUO induction. Sirius red staining for collagen deposits showed less fibrotic material in knockout animals compared to the wild type. (D) Further analysis of the type I/type III collagen expression ratio showed a significant decrease in the type I/type III ratio in the Ybx3 knockout group. (E) TaqMan analysis of the relative Col1a1 and Col1a3 transcripts in healthy contralateral and obstructed kidneys. The analysis showed a significant reduction in collagen transcripts in the absence of Ybx3. Data represent means ± SDs. Quantification is provided for wild-type and knockout animals (* p < 0.05, ** p < 0.01, *** p < 0.001).