Circulating and Urinary CCL20 in Human Kidney Disease

Abstract

CC motif chemokine ligand 20 (CCL20), a chemokine involved in immune cell migration through its receptor CCR6, has been implicated in kidney inflammation in crescentic glomerulonephritis and acute kidney injury. However, clinical information for other kidney diseases is scarce. We have analysed CCL20 levels in plasma and urine from patients with diabetic kidney disease (DKD, n = 98) and autosomal dominant polycystic kidney disease (ADPKD, n = 85) treated according to the guidelines and studied their association with baseline characteristics and long-term (median follow-up 4.9 and 7.1 years, respectively) clinical outcomes. Single-cell kidney transcriptomics were mined to identify CCL20-expressing cells. Plasma CCL20 was higher in DKD and ADPKD than in a reference group: median 12.8 (3.5-33.2), 6.0 (1.2-19.2), and 0.0 (0.0-9.0) pg/mL, respectively. Urinary CCL20 was quantifiable in 48% of patients with DKD but not in the reference group. Transcriptomics data support a local kidney source of CCL20. In DKD, plasma CCL20 was higher in early compared to advanced CKD. Urinary CCL20 was higher in patients with A2 albuminuria than in those with other albuminuria categories. In ADPKD, higher plasma and urinary CCL20 levels tended to be associated with lower eGFR, higher albuminuria, and larger kidneys. However, no significant association was found between CCL20 levels and progression to kidney failure or death. In conclusion, CCL20 is increased in biological fluids and locally produced in CKD. While this may point to a potential role in risk stratification, further studies are necessary.

Keywords: CC motif chemokine ligand 20 (CCL20); autosomal dominant polycystic kidney disease; chronic kidney disease; clinical outcomes; diabetic kidney disease; observational study.

Figure A1

Flow chart of the study design to assess CCL20 in human kidney disease. CCL20 expression levels were analysed by ELISA in plasma and urine samples from CKD patients with diabetic kidney disease (DKD) or autosomal dominant polycystic kidney disease (ADPKD). Participants in each cohort were categorised according to eGFR and UACR criteria defined by KDIGO. ADPKD patients were further stratified using the Mayo Clinic total kidney volume (TKV) classification. In parallel, the origin and expression of CCL20 and its receptor CCR6 in DKD and ADPKD were explored using publicly available transcriptomic resources, Kidney Interactive Transcriptome (KIT) and Nephroseq, to assess their transcriptional expression in individual kidney cell types and whole/microdissected kidney tissue, respectively.

Figure A2

KDIGO risk categories of CKD according to eGFR (G) and albuminuria (A) categories [63]. Green: low risk (if no other markers of kidney disease, no CKD); yellow: moderately increased risk; orange: high risk; red: very high risk.

Figure A3

Study flow diagram illustrating the biobanking and follow-up periods for participants in the ADPKD and DKD cohorts. Coloured bars indicate the duration of sample collection (biobanking), while arrows represent ongoing clinical follow-up.

Figure 1

Plasma CCL20 detection and levels in the reference group (Ref) and DKD patients measured by enzyme-linked immunosorbent assay (ELISA): (A) percentage of participants with detectable plasma CCL20, * p < 0.05; (B) plasma CCL20 protein levels. Data are presented as boxplots showing median, interquartile range (IQR), and range, ** p < 0.01; (C) percentage of detectable plasma CCL20 among DKD patients across different eGFR categories, * p < 0.05; (D) plasma CCL20 protein levels in DKD patients across different eGFR categories. Data are presented as boxplots showing the median, IQR, and range, * p < 0.05, ** p < 0.01; (E) percentage of detectable plasma CCL20 among DKD patients across different UACR categories, ** p < 0.01; (F) plasma CCL20 protein levels in DKD patients across different UACR categories. Data are presented as boxplots showing median, IQR, and range, * p < 0.05. Inset charts show a magnified view of the lower range of values (B,D,F).

Figure 2

Urine CCL20 detection and levels in the reference group (Ref) and DKD patients detected by ELISA: (A) percentage of participants with detectable CCL20 in the urine, * p < 0.05; (B) urine CCL20 protein levels. Data are presented as boxplots showing the median, interquartile range (IQR), and range. Not significant (ns), p = 0.0534; (C) percentage of detectable urine CCL20 in DKD patients across different eGFR categories. (D) Urine CCL20 protein levels measured by ELISA in DKD patients across different eGFR categories. Data are presented as boxplots showing the median, IQR, and range, * p < 0.05; (E) percentage of detectable urine CCL20 among DKD patients across different UACR categories, * p < 0.05; (F) urine CCL20 protein levels measured by ELISA in DKD patients across different UACR categories. Data are presented as boxplots showing the median, IQR, and range, ** p < 0.01, *** p < 0.001.

Figure 3

Relationship between plasma and urinary CCL20 levels in DKD patients. A linear regression analysis showed no significant association.

Figure 4

Plasma CCL20 detection and levels in the reference group (Ref) and ADPKD patients assessed by ELISA: (A) percentage of participants with detectable plasma CCL20, * p < 0.05; (B) plasma CCL20 protein levels. Data are presented as boxplots showing the median, interquartile range (IQR), and range, * p < 0.05; (C) percentage of detectable plasma CCL20 among ADPKD patients across different eGFR categories. (D) Plasma CCL20 protein levels in ADPKD patients across different eGFR categories. Data are presented as boxplots showing the median, IQR, and range, * p < 0.05, ** p < 0.01; (E) percentage of detectable plasma CCL20 among ADPKD patients across different UACR categories; (F) plasma CCL20 protein levels in ADPKD patients across different UACR categories. Data are presented as boxplots showing the median, IQR, and range, * p < 0.05; (G) percentage of detectable plasma CCL20 among ADPKD patients across different total kidney volume (TKV) categories; (H) plasma CCL20 protein levels in ADPKD patients across different TKV categories. Data are presented as boxplots showing the median, IQR, and range. Inset charts show a magnified view of the lower range of values (B,D,F).

Figure 5

Kaplan–Meier survival curves for freedom from KRT or death. (A) Kaplan–Meier survival curves in the combined cohort of DKD and ADPKD patients, stratified by plasma CCL20 tertiles. ADPKD patients already on dialysis at baseline (time 0) were excluded from this analysis. No significant differences were observed among survival curves. Log-rank test: χ² = 0.85, gl = 2, p = 0.65. (B) Kaplan–Meier survival curves in DKD patients comparing those with detectable versus undetectable urinary CCL20 levels. No significant differences were observed among survival curves. Log-rank test: χ² = 1.77, df = 1, p = 0.18. The hazard ratio (HR) was 2.00 (95% CI: 0.77–5.19).

Figure 6

CCL20 expression in kidney cell populations from patients with DKD (A) and ADPKD (B). Data visualised using the Kidney Interactive Transcriptomics (KIT) platform, developed by the Humphreys Lab (https://humphreyslab.com/SingleCell/) [20], based on single-nucleus RNA sequencing (snRNA-seq) datasets from references [21] and [22], respectively. The dot plots (left) indicate average expression (colour scale) and proportion of expressing cells (dot size) across kidney cell populations. Uniform manifold approximation and projection (UMAPs) (right) illustrate the spatial distribution of CCL20-positive cells in snRNA-seq datasets. Red boxes in the dot plots and UMAPs indicate the cell populations with CCL20 expression. Cont: control; PT: proximal tubule; PTVCAM1: proximal tubular cells expressing VCAM; PEC: parietal epithelial cells; ATL: ascending thin limb of the loop of Henle; TAL1: thick ascending limb 1; TAL2: thick ascending limb 2; DCT1: distal convoluted tubule 1; DCT2: distal convoluted tubule 2; PC: principal cells; ICA: intercalated cells A; ICB: intercalated cells B; PODO: podocytes; ENDO: endothelial cells; MES: mesangial cells; FIB: fibroblasts; LEUK: leukocytes; FR-PTC: failed-repair proximal tubule cells; TAL: thick ascending limb; DCT: distal convoluted tubule; CNT_PC: connecting tubule/principal cells.

Figure 7

CCR6 expression in kidney cell populations from patients with DKD (A) and ADPKD (B). Data visualised using the KIT platform, developed by the Humphreys Lab (https://humphreyslab.com/SingleCell/) [20] based on snRNA-seq datasets from Wilson et al. (2019) and Muto et al. (2022), respectively [21,22]. The dot plots (left) indicate average expression (colour scale) and proportion of expressing cells (dot size) across kidney cell populations. UMAPs (right) illustrate the spatial distribution of CCL20-positive cells in single-nucleus RNA-seq datasets. Red boxes in the dot plots and UMAPs indicate the cell populations with CCR6 expression.