New biomarkers of inflammation associated with haemodialysis

Abstract

Background: The first year of haemodialysis (HD) carries the highest risk of mortality, which to a large extent is attributed to the aggravation of inflammation. However, traditional markers such as C-reactive protein and interleukin-6 show only minor changes during the first year, suggesting that there are other factors involved. The present study evaluates the effect of HD on microinflammation and oxidative stress of uremic patients.

Methods: We conducted a prospective observational longitudinal study including 30 incident HD patients. Blood samples were collected at baseline and 6 and 12 months. Pro-inflammatory monocytes were quantified using flow cytometry. Proteomic analysis (Olink) was performed on serum. Concentrations of indoxyl sulphate (IS), growth differentiation factor 15 (GDF-15), oxidative status and circulating microRNA (miRNA) expression were also determined.

Results: A new population of activated monocytes was identified that progressively increased at 1 year of HD. In addition, an increase in the serum concentration of up to 29 inflammation-related proteins was detected, including interleukins, chemokines, tumour necrosis factor family molecules, cell activation molecules and apoptosis-related proteins. Conversely, leukaemia inhibitory factor receptor was downregulated. The concentration of IS was positively correlated with GDF-15 levels. Furthermore, patients exhibited decreased expression of miRNA-126-3p, -130a-3p, -146a-5p, 223-3p, -let7a-5p and -let7b-5p.

Conclusion: This study highlights the impact of HD on inflammation and oxidative stress, manifested by an increase in activated monocytes and inflammatory markers. The observed subclinical inflammation associated to HD treatment may help in understanding the mechanisms of cardiovascular damage in patients on HD.

Keywords: cardiovascular disease; haemodialysis; inflammation; monocytes; uraemic toxins.

Figure 1:

Gating strategy for monocyte subset determination. (A) CD14^high/SSC-A (side scatter area) was used to identify monocyte cells among other leucocytes. (B) Doublets were removed by gating on SSC-A versus SSC-H (side scatter height). (C) Forward and side scatter area (FSC-A/SSC-A) identified typical and atypical monocytes. (D, E) Expression of surface marker CD16 was analysed in gated cells.

Figure 2:

Monocyte subsets in blood from incident HD patients for 12 months of follow-up. (A) Representative FACS plots depicting monocyte subsets. The x-axis represents the forward scatter area (FSC-A), with higher signal indicating an increase in cell size. (B) Percentage of atypical monocytes and (C) typical monocytes at baseline, 6 months and 12 months following the initiation of HD treatment. Percentage of pro-inflammatory CD16⁺ cells within (D) atypical monocytes and (E) typical monocytes. (F) CD16 and (G) CD14 expression levels on typical and atypical monocytes quantified by changes in the MFI. Data are presented as median ± SEM. P-values <.05 were considered on the borderline of statistical significance.

Figure 3:

Inflammatory proteins identified in serum samples from HD patients. Volcano plot of 92 inflammation-related proteins at (A) 6 months and (B) 12 months compared with baseline in HD patients. The red points indicate significantly upregulated proteins, while blue points are significantly downregulated proteins. (C) Venn diagram shows differentially expressed proteins identified at 6 months, at 12 months and those common at both 6 and 12 months (overlapping region).

Figure 4:

Functional classification of the altered levels of the inflammatory proteome in HD patients by STRING platform. Protein–protein interaction networks of the differentially expressed proteins in serum of HD patients.

Figure 5:

Determination of uraemic toxins, inflammation and oxidative stress markers. Serum concentrations of (A) IS (μg/mL) and (B) GDF-15 (pg/mL) in healthy subjects and HD patients. (C) Pearson correlation coefficient test of the relationship between serum levels of IS and levels of GDF-15. (D) Total antioxidant capacity levels in micromolar copper reducing equivalents (CRE) in healthy subjects and HD patients. (E) MDA (μM) levels in HD patients and healthy subjects. The results are presented as the mean ± SEM. P-values <.05 were considered on the borderline of statistical significance.

Figure 6:

Expression levels of circulating miRNAs in HD patients. Relative miRNA expression levels of (A) miR-126-3p, (B) miR-130a-3p, (C) miR-146a-5p, (D) miR-223-3p, (E) miR-let7a-5p and (F) miR-let7b-5p in serum from patients undergoing HD. Samples were normalized to exogenous control cel-miR-39-3p. The results are presented as ΔCt values. P-values <.05 were considered on the borderline of statistical significance.