Cockayne syndrome mice reflect human kidney disease and are defective in de novo NAD biosynthesis

Abstract

Cockayne Syndrome (CS) is a premature aging disorder caused by mutations in the CSA and CSB genes involved in DNA metabolism and other cellular processes. CS patients display many features including premature aging, neurodegeneration, and kidney abnormalities. Nicotinamide dinucleotide (NAD⁺) deprivation has been observed in CS patient-derived cells. NAD⁺ has essential roles in regulating cellular health, stress responses, and renal homeostasis. While kidney dysfunction is a common feature in CS patients, its molecular pathogenesis is not understood. Here, we report that severe kidney pathology is present in CS A and B mice. We find that the NAD⁺ biosynthetic pathways are impaired in kidneys from these mice. Using human renal tubular epithelial cells, we show that CSA/B downregulation causes persistent activation of the ATF3 transcription factor on the quinolinate phosphoribosyl transferase gene locus, a rate-limiting enzyme in de novo NAD⁺ biosynthesis in the kidney, causing impaired transcription and deficient NAD⁺ homeostasis.

Fig. 1. Cockayne syndrome (CS) mice models exhibits kidney abnormalities.

A Body weight measurement of CS mice compared to wild type. For CSA (aged 47–68 weeks), 5 females and 4 males were used as controls (WT) and 5 females, and 6 males were used as homozygous knockout experimental group (HO CSA). For CSB (aged 45–68 weeks), 3 males and females were used for both control (WT), and the experimental groups (HO CSB). B Kidneys from each group were isolated and weights were normalized to the total body weight. Scale bar, 5 mm. C Representative image of CS mice kidneys, and (D) with arrows indicating fluid filled-blisters on the surface. E Transverse section of WT and CS kidneys (aged 45–68 weeks). Lengths of the renal cortex (C) and medulla (M) were measured and cortical to medullary ratio were compared. Scale bar, 1 mm. Data are presented as mean ± SD (n ≥ 3). A two‐way ANOVA Sidak test was used for the comparisons between males and females. p-values indicated (Blue bar, WT; Red bar, HO CSA/CSB).

Fig. 2. Histopathological and Morphometric analysis of CS mice.

A Histology findings of different lesions in male CS mice compared to wild type (aged 17–21 weeks) and examples include (1) Cytomegaly; (2) Karyomegaly; (3) Hyaline casts; (4) Karyomegaly; (5) Tubular regeneration; (6) Tubular regeneration, Karyomegaly, and epithelial degeneration; (7) Karyomegaly and epithelial degeneration. Scale bar, 20 μm. B Histology findings of different lesions in CSA and CSB mice compared to wild type (aged 45–68 weeks) and examples indicated include (1) hyaline glomerulopathy; (2) cortical scarring; (3) tubular nuclear enlargement consistent with tubular injury and occasional apoptotic cells; (4) mitotic figures; (5) foci of plasma cell rich inflammation; (6) hyaline glomerulopathy; (7) tubular cytoplasmic vacuolation; (8) focal tubular epithelial karyomegaly; (9) bland hyaline casts. Scale bar, 60 µm. C Representative image (left) and quantification (right) of PSR staining in CS mice. Scale bar, 100 µm. D Mean glomerular area quantification. E Whole-slide H&E images with glomeruli color-coded according to their corresponding histogram bin by digital histology (refer to “Materials and Methods” for details). Tubular feature quantifications (F) Median Tubular area, (G) Median tubule eccentricity, and (H) Median Tubule Perimeter by digital histology. I p57 immunohistochemistry (IHC) in kidney sections of CSA and CSB mice. Representative IHC image (Top) and quantification (bottom) of p57 staining in CS mice is shown. Scale bar, 60 µm. J Representative and quantification of TUNEL staining in CS mice kidney sections. Scale bar, 60 µm. Data are presented as mean ± SD (n ≥ 3). A two‐way ANOVA Sidak test was used for the comparisons between males and females. p values indicated (Blue bar, WT; Red bar, HO CSA/CSB).

Fig. 3. Kidney dysfunction in CS models.

A–G Serum isolated from the whole blood samples and (H) urines collected from both male and female mice (aged 45–68 weeks) for both genotypes: WT and CS (CSA/B) mice and analyzed for the parameters indicated. I–N Kidney toxicity markers in serum and urine were determined by an immunoassay panel (refer to “Method” section for details). Data are presented as mean ± SD (n ≥ 3), An unpaired t test was used for comparing CS with WT. A two‐way ANOVA Sidak test was used for the comparisons between males and females. p values indicated (Blue bar, WT; Red bar, HO CSA/CSB).

Fig. 4. Impaired tryptophan metabolism in CS kidneys.

Kidneys from CSA (4 females and 3 males for controls, 4 females and 5 males for HO CSA) and CSB (3 of each females and males for both controls and HO CSB) mice sets were isolated and RNA was purified. After RNA sequencing, differentially expressed gene sets (DEGs) were processed. A Significantly downregulated gene ontology (GO) set (p ≤ 0.05) analysis from DEGs. Gene set enrichment analysis (GSEA) plot and heatmap for the ‘Nicotinate and Nicotinamide metabolism’ panel from DEGs of CSA (B) and CSB (C) mice kidneys.

Fig. 5. CS kidneys have increased expression of toxicity markers.

A mRNA expression heatmap of the kidney injury markers (fold change ≥1.5-fold) in CS mice compared to WT mice. B qPCR was performed with the purified RNA from CS mice kidneys for kidney injury markers (Lcn2, Havcr1, and Timp1). Actin normalized expression values are shown. C Havcr1 immunohistochemistry (IHC) was performed with CS kidney tissue sections. A representative IHC image is shown (top), with quantitative analysis (bottom). Scale bar, 20 μm. Data are presented as mean ± SD (n ≥ 3). An unpaired t test was used for comparing CS with WT. A two-way ANOVA Sidak test was used for male vs female comparisons. p values indicated (Blue bar, WT; Red bar, HO CSA/CSB).

Fig. 6. mQPRT expression is downregulated in CS mice kidneys.

A Schematic diagram of intracellular NAD⁺ biosynthesis pathway in the kidney. The source of intracellular NAD⁺ comes from three pathways, Preiss-Handler (green), de novo (blue), and salvage pathways (yellow). B Total NAD⁺ quantification from the kidneys of CS and WT mice (n = 3). C mQPRT (mouse isoform) mRNA expression data from CSA and CSB mice kidney DEGs (D) mQPRT qPCR analysis with RNAs purified from WT/CS mice kidneys. Actin normalized expression values are shown. E mQPRT expression in kidney sections of CSA (left) and CSB (right) mice. Representative images for mQPRT IHC for CS mice for both sexes are shown. Scale bar, 20 μm. Data are presented as mean ± SD (n ≥ 3). An unpaired t test was used for comparing CS with WT. A two-way ANOVA Sidak test was used for male vs female comparisons. p values indicated (Blue bar, WT; Red bar, HO CSA/CSB).

Fig. 7. CSA/B availability determines QPRT expression.

A CS mice kidney tissues were subjected to western blot for probing indicated antibodies. B Actin normalized mATF and mQPRT blot quantification data are shown (Blue bar, WT; Red bar, HO CSA/CSB). C HK-2 cells were transfected with two sets of CSA/B siRNAs. Cells were harvested 72 h post-transfection and subjected to qPCR. GAPDH-normalized expression values and (D) western blot analysis with indicated antibodies are shown. E Actin normalized blot quantification of CSA, CSB, hQPRT (human isoform) and hATF3 are shown. F siRNA-transfected HK2 cells were harvested and subjected to total protein-normalized NAD⁺ and NAD⁺/NADH quantification analyses. Data are presented as mean ± SD (n ≥ 3). An unpaired t test was used for comparing CS with WT and HK-2 Knockdown vs SCR. A two-way ANOVA Sidak test used for male vs female comparisons. A Mann–Whitney test used for ATF3 quantification. p values indicated.

Fig. 8. Direct interaction of ATF3 on the QPRT gene locus.

A Schematic representation of hQPRT gene with potential hATF3 binding sites (marked with arrows). B Anti-ATF3/IgG chromatin immunoprecipitation (ChIP) was performed with HK-2 cells. Ten primer sets for respective potential hATF3 binding sites were investigated. C CSA/B siRNAs were transfected to HK-2 cells. 72 h post-transfection, hATF3 ChIP-qPCR was performed on hQPRT genomic locus. Data are presented as mean ± SD (n = 3). A paired t test used for analyses. p values indicated.

Fig. 9. NAD supplementation mitigates kidney injury and apoptosis in CSA/CSB knockdown HK-2 cells.

HK-2 cells transfected with two sets of CSA/B siRNAs and treated with 1 mM NR for 24 h. Cells were harvested after NR treatment and subjected to qPCR and TUNEL staining. mRNA expression for (A) kidney injury markers LCN2, CLU, and HAVCR1 and (B) NMNATs 1,2,3 were analyzed. ACTIN normalized expression values are shown. C TUNEL staining was performed on cells with/without NR treatment. A representative image (left) is shown with quantitative analysis (right). Scale bar, 100 µm. An unpaired t test was used to compare HK-2 Knockdown vs SCR with/without NR treatment.

Fig. 10. Potential mechanism of renal dysfunction in CS mice.

In healthy animals, renal tubular cells utilize high-demand NAD⁺ supplies from Preiss-Handler and de novo synthetic pathways by metabolizing nicotinamide (NA) and tryptophan (TRP), respectively. In the case of CS protein disruption (CS), increased ATF3 expression suppresses QPRT transcription, leading to an impaired de novo pathway. Moreover, decreased nicotinate phosphoribosyltransferase (NAPRT) expression may also lead to the Preiss-Handler pathway blockage. Impairment of both Preiss-Handler and de novo NAD⁺ synthetic pathways result in overall intracellular NAD⁺ deprivation, which may subsequently lead to kidney damage.