KLF13 promotes SLE pathogenesis by modifying chromatin accessibility of key proinflammatory cytokine genes

Abstract

Although significant progress has been achieved in elucidating the genetic architecture of systemic lupus erythematosus (SLE), identifying genes underlying the pathogenesis has been challenging. The NZM2410-derived lupus susceptibility Sle3 locus is associated with T cell hyperactivity and activated myeloid cells. However, candidate genes associated with these phenotypes have not been identified. Here, we narrow the Sle3 locus to a smaller genomic segment (Sle3k) and show that mice carrying Sle3k and Sle1 loci developed lupus nephritis. We identify Klf13 as the primary candidate gene that is associated with genome-wide transcription changes resulting in higher levels of proinflammatory cytokines, enhanced T cell activation, and hyperresponsiveness of myeloid cells. Correspondingly, Klf13 ^-/- mice display repression of genes involved in mediating immune activation, including key proinflammatory cytokines/chemokines in T cells and dysregulation in cytokine signaling pathways in myeloid cells in response to toll receptor ligands. Klf13 upregulation is associated with increased production of RANTES, a key chemokine in lupus nephritis, in activated T cells and the kidneys of lupus-prone mice. In sum, our findings reveal Klf13 as a key gene in the Sle3 interval in mediating lupus pathogenesis that may have implications in the rational design of new therapies for SLE.

Fig. 1. Congenic dissection of Sle3 interval.

a Diagrammatic representation of B6 congenic mouse interval harboring NZM2410- derived lupus susceptibility locus Sle3t flanked by D7mit157 and D7Mit233 on Chromosome 7 and all the Sle3t-derived truncated congenic intervals produced in the study for fine-mapping and gene identification. b Flow cytometry analysis showing CD40 expression BMDMs [unstimulated (U) or stimulated with LPS] derived from B6 or Sle3t strains. N = 3 mice per group. BMDMs cultured from B6 and Sle3t mice were unstimulated (U) and stimulated with 10 ng/ml LPS for 6 h, and supernatants were analyzed by ELISA for TNFα (c) and IL-6 (d). N = 3 mice per group. Values below the detection levels in ustimulated samples were assigned a value of 1. Mean fluorescence intensity (MFI) of CD40, MHCII, and CD80 on BMDMs [unstimulated (U) or stimulated with LPS] derived from B6 and different truncated Sle3 recombinant strains is shown in e–h. N = 3 mice per group. i BMDMs cultured from B6 and truncated Sle3 recombinant mice were unstimulated (U) and stimulated with 10 ng/ml LPS for 6 h, and supernatants were analyzed by ELISA for TNFα and IL-6. N = 3–5 mice per group. Each data point represents a single mouse in the figures. Student’s t-test was performed for statistical analysis. Results are shown as means ± SEM. Statistical significance is represented as *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 2. Phenotypic characterization of Sle3 congenic mice.

Assessment of serum levels of anti-nuclear autoantibodies (IgG) by Elisa in 4-month-old females (a) and 6–7-month old females (b). In panel a, B6 n = 8; Sle1 n = 8; Sle3t n = 7; Sle3a n = 8; Sle3b n = 6; Sle1Sle3t n = 8; Sle1.Sle3a n = 8; Sle1.Sle3b n = 8. In panel B, B6 n = 8; Sle1 n = 8; Sle3t n = 7; Sle3a n = 7; Sle3b n = 6; Sle1Sle3t n = 7; Sle1.Sle3a n = 7; Sle1.Sle3b n = 7. c Anti-GBM disease was induced in 6–12-week-old B6, and Sle3a strains and nephritis score was calculated at day 14 (n = 5 mice per group). d Twenty-four-hour urine samples collected from 2–3-month-old female mice on days 0 and 14 after the anti-GBM challenge were assessed for proteinuria (n = 5 mice per group). All mice used in this study are females. Each data point represents a single mouse in the figures. Results are shown as means ± SEM. Statistical significance is represented as *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 3. Phenotypic characterization of Sle1Sle3k congenic mice.

a Measurement of splenic weight in Sle1Sle3k (n = 16) mice versus B6 (n = 6) and Sle1 (n = 6) mice. b Measurement of proteinuria in Sle1Sle3k (n = 13) mice versus B6 (n = 5) and Sle1 (n = 5). c Serum levels of BUN in Sle1Sle3k (n = 14) mice versus B6 (n = 10) and Sle1 (n = 9). d Representative H & E slides showing the development of severe glomerulonephritis (GN) in kidneys of Sle1.Sle3k (n = 13) mice compared to B6 (n = 5) or Sle1 (n = 5) mice at 9–12 months of age. Arrows indicate hyper-cellularity observed in glomeruli. Sacle bars, 100 µm. e GN scores graded in a blinded fashion are shown in Sle1.Sle3k mice (n = 13) compared to B6 (n = 5) or Sle1 (n = 5) mice at 9–12 months of age. f Serum levels of anti-nuclear autoantibodies (IgG) in Sle1.Sle3k mice (n = 14) compared to B6 (n = 3) or Sle1 (n = 4) were assessed by Elisa. g Heatmap depicting increased serum autoantibody titers in Sle1Sle3k mice compared to B6 and Sle1 mice against a panel of 96 autoantigens spotted on a custom array. Mice per group (n = 5). h Representative images of immunohistochemistry for CCL5 and CD3 of paraffin-fixed kidney sections of 9–12-month-old mice. Quantification of tubules stained for CD3 +ve cells per field (i) and RANTES +ve tubules per field (j) in kidney sections is shown. N = 4-5 per strain. Scale bars, 100 µm. All mice used were 9–12-month-old females for all three strains. Each data point represents a single mouse in the figures. Results are shown as means ± SEM. Statistical significance is represented as *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 4. Identification of Klf13 as a lupus susceptibility gene.

a RPKM gene expression values of NZW allele versus B6 allele of Klf13 gene in splenic CD4 T cells assessed by RNA-seq (n = 1). b Real-time PCR analysis of KLF13 mRNA levels measured in splenic CD4 T cells isolated from B6 versus Sle3k (n = 3 mice per group). c Western blot showing KLF13 protein expression in splenic CD4 T cells isolated from B6 versus Sle3k mice at 72 h and 96 h post-stimulation with anti-CD3/anti-CD28. GAPDH and KLF13 protein run at the same size and hence the bands shown for 72 h post timepoint are from two separate gels. d Splenic CD4 T cells were stimulated with plate-bound anti-CD3/anti-CD28 for 96 h, and levels of RANTES was measured by ELISA (n = 3 per group). e Whole genome sequencing of NZB, NZW, and NZM2410 mouse strains depicting the mapping of Sle3k congenic interval on Chr7. All the variants in three mouse strains compared to the B6 genome are colored in purple. f All SNPs in the NZW allele of the Klf13 gene were identified by genome sequencing. g Comparison of ATAC-seq data from CD4 T cells and BMDMs in B6 (Gray Panel) versus Sle3k (Blue Panel) mice. Bedgraph panels for the Klf13 gene are shown with peak locations relative to the transcription start site (TSS). The top Black panels are the ATAC-seq Bed files of mouse Encode data obtained from the UCSC genome browser. Yellow bars indicate regulatory elements identified in Ensemble regulatory build. h Bedgraph panels generated from CHIP Atlas data for the Klf13 gene shown for multiple immune cells including CD4 T cells and the position of upstream variant 7:63939829 within the epigenetic mark overlapping with the peak of regulatory regions and several transcription factors binding sites as indicated by black arrow in the Top panel. The position of the intronic variant 7: 63919320 is shown nearby to the peak regulatory signal in the intron is shown in bottom panel indicated by black arrow. Results are shown as means ± SEM. Statistical significance is represented as *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 5. Regulation of T cell activation and production of proinflammatory cytokines mediated by KLF13.

a Splenic CD4 T cells were stimulated with plate-bound anti-CD3/anti-CD28 for 48 h and levels of IL2 and IFNγ in the cell culture supernatants were determined by ELISA (n = 6 per group). b Heatmap of differentially regulated selected genes in splenic CD4 T cells isolated from B6 versus Sle3k mice as determined by RNA-seq analysis at 6 h post-stimulation with anti-CD3/anti-CD28. c Gene expression values for IL-2 and IFNγ are shown as determined by RNA-seq analysis of CD4 T cells isolated from B6 versus Sle3k mice. d Splenic CD4 T cells were stimulated with plate-bound anti-CD3/anti-CD28 for different time points (hour), and levels of IL2, IFNγ, and CCL5 in the cell culture supernatants were determined by ELISA in Klf13^–/– versus WT mice (n = 4). e RNA-seq panels showing the read depth of KLF13 mRNA level confirming the knockdown of the Klf13 gene in Klf13^–/– mice compared to WT mice. RPKM gene expression values for IFNγ and IL-2 are shown as determined by RNA-seq analysis of CD4 T cells isolated from Klf13^–/– versus WT mice. f Heatmap of differentially regulated genes in splenic CD4 T cells isolated from Klf13^–/– mice versus WT mice as determined by RNA-seq analysis at 18 h post-stimulation with anti-CD3/anti-CD28. N = 2 experiments. g PCA analysis of RNA seq data of unstimulated or CD4 T cells stimulated with anti-CD3/anti-CD28 for 6 h and 18 h isolated from Klf13^–/– versus WT mice. h Bar graph showing number of significantly variable genes between Klf13^–/– versus WT mice with FDR < 0.05 and LOG10 fold change greater than one. i Volcano plot comparing RNA seq data in Klf13^–/– versus WT mice in unstimulated and stimulated CD4 T cells at 6 h and 18 h post-stimulation (anit-CD3 and anti-CD28) by plotting the fold change on the x-axis and -log10 of the p-value on the y-axis. j ATAC-seq was performed on CD4 T cells unstimulated or stimulated with anti-CD3/anti-CD28 for 3 h isolated from Klf13^–/– versus WT mice. N = 2. Bedgraph panels for IL2 (0 h) and IFNγ (3 h) are shown with peak locations relative to the transcription start site (TSS). k Klf13, Ccl5, and Ifih1 gene expression in the various cell subsets in published single cell datasets of kidney biopsies from Lupus Nephritis (LN) patients. Cell cluster annotations are as follows: CE0- Epithelial cells ; CD0- dividing cell cluster including T cells and NK cells; CM0, CM1 and CM4 are monocyte/macrophage clusters; CM2- Macrophages; CM3- CD1C+ dendritic cells (DCs); CT1 and CT5b are NK cell clusters; CT0a, CT3b, CT0b, CT2, CT3a and CT5a, and CT4 are T cell clusters; CB2a and CB0 are B cell clusters; CB2b- plasmacytoid DCs (pDCs); CB1- plasma cells/plasmablasts; CT6- CD4 + T cells with high levels of ISGs; CB3- B cells with high levels of ISGs. Results are shown as means ± SEM. Statistical significance is represented as *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 6. KLF13 modulates toll receptor signaling in myeloid cells.

a BMDMs cultured from B6 and Sle3k mice were unstimulated and either stimulated with 1 mg/ml R848 or 10 ng/ml LPS for 6 h, and supernatants were analyzed by ELISA for CCL2 and TNFα, respectively (n = 6). b Heatmap of differentially regulated genes in BMDCs derived from Sle3k versus B6 mice unstimulated or stimulated with 1 mg/ml R848 for 6 h as analyzed by RNA-seq analysis. N = 1 per timepoint. c BMDMs cultured from WT versus Klf13^–/– mice were unstimulated or stimulated with 1 mg/ml R848 or with 10 ng/ml LPS for 6 h, and supernatants were analyzed by ELISA for CCL2 (n = 6) and TNFα (n = 4) respectively. d Global gene expression profiling of BMDMs derived from wild type, and Klf13^–/– mice either unstimulated or stimulated with 1 mg/ml R848 or 10 ng/ml LPS or 10 mg/ml ODN1826 for 6 h. N = 1 per timepoint. Heatmap represents differentially expressed genes between WT. and Klf13^–/– mice determined by RNA-seq analysis. e Global gene expression profiling of BMDCs derived from wild type and Klf13^–/– mice either unstimulated or stimulated with 1 mg/ml R848 or with 10 ng/ml LPS for 6 h. Heat maps represent differentially expressed genes between WT. and Klf13^–/– mice determined by RNA-seq analysis at 6 h in response to LPS. N = 2 per time-point. f Total number of genes significantly upregulated or downregulated between wild type and Klf13^–/– mice in each treatment group (FDR < 0.05). g PCA analysis of RNA seq data of BMDCs unstimulated or stimulated with LPS or R848 for 6 h isolated from Klf13^–/– versus WT mice. ATAC-seq was performed on WT versus Klf13^–/– derived BMDMs (h) and BMDCs (i) treated with 1 mg/ml R848 for 1 h or 3 h (N = 2 per time point). Bedgraph panels for CCL2, IL-12β, IL-1β, IL-6 and TNF-α are shown with peak locations relative to the transcription start site (TSS) highlighted by boxes and arrows. BMDMs, Bone-marrow derived macrophages; BMDCs, Bone-marrow derived dendritic cells. Results are shown as means ± SEM. Statistical significance is represented as *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.