On the basis of a large amount of publicly available genetic data, this study explored the causal relationship between 731 immune cell phenotypes and CKD. To date, this is the first bidirectional MR to investigate genetic causality between multiple immune cell phenotypes and CKD. This study identified a strong contributory genetic causality of two immune cell phenotypes for CKD (PFDR < 0.05), as well as an association of CKD for an increase in five immune cell phenotypes (PFDR < 0.6).
In this study, CKD was observed to increase with HLA DR on DC (cDC panel) and HLA DR on CD33- HLA DR+ (Myeloid cell panel). cDC are exclusively responsible for presenting antigens to T cells, recruiting and activating effector T cells, and releasing inflammatory mediators, which contribute to tissue damage in CKD16. Cytokines produced by activated cDC or T cells, including IL-1, transforming growth factor-β (TGFβ), and TNF, also stimulate fibroblasts, which drive the progression of CKD17,18. Myeloid cells are characterised by common myeloid progenitors and all the cell types they give rise to, including neutrophils, eosinophils, basophils, dendritic cells, mast cells, and monocytes/macrophages, which constitute the major cellular components of innate immune events19. The HLA DRhi monocyte subtype, which is increased in patients with CKD compared with adults with normal renal function, generates higher amounts of TNF and IL-1β than other monocyte populations, and accelerates the progression of CKD and the development of complications through enhanced adhesion to the endothelium and enhanced response to specific inflammatory migratory signals20. Of note is that monocytes infiltrating tissues, which commonly transform into macrophages, mediate CKD-associated cardiac insufficiency and ultimately death from end-stage renal disease21,22.
Reverse MR studies have demonstrated that CKD induces an increase in Naive CD8 + T cells, CD45RA + CD8 + T cells, CD28 + CD45RA + CD8 + T cells, CD64 on CD14- CD16- and HLA DR on CD14- CD16 + monocytes. CD8 T cells were divided into naïve, effector, and memory cell subpopulations, with the memory cell population subdivided into two major subpopulations of central memory (CM) and effector memory (EM) T lymphocytes, and EMT lymphocytes further divided into CD45RA- and CD45RA + effector memory T lymphocytes, with CD28 distinguishing between two effector memory populations in early and late differentiation states23. The amount of CD8 T cells increases in the early stages of CKD and has been shown to peak on day five in a model of UUO-associated CKD24. CD8 T cells generate multiple chemokines to attract monocyte recruitment and proliferation, thereby amplifying the inflammatory response; IFN-γ and TNF-α secreted by CD8 T cells initiate relevant signaling pathways that induce macrophage polarization toward the M1 phenotype, resulting in the production of higher levels of pro-inflammatory cytokines; and IL-4 and IL-13 are secreted by CD8 T cells and Th2 cells, which subsequently induce a shift in the infiltrating M1 phenotypic shift to pro-fibrotic M2 macrophages involved in renal repair and regeneration25. Notably, when CKD progresses to end-stage renal disease, there is an increase in apoptosis and a decrease in the number of naïve and CM T-lymphocytes, a phenomenon that may reflect, in part, an impaired immune response26,27. In CKD, the proportion of CD28-T cells increases due to the progressive differentiation of memory T cells resulting from persistent chronic injury, which leads to further damage to renal tissue28. Human monocytes can be categorized into classical, intermediate and non-classical monocytes based on the surface expression of CD14 and CD16. Classical monocytes are the most common subtype and highly express CD14, which mediate antimicrobial defense and show high adhesion, migration, and phagocytosis29. Non-classical monocytes express CD16, which is strongly associated with vascular stiffness and cardiovascular event rates in advanced CKD30. Immature monocytes express HLA-DR surface markers, but CD13 and CD14 are not expressed until monocyte maturation. Increased antigenic HLA-DR expression reflects the mature activated state of monocytes, and decreased HLA-DR is usually associated with a decrease in the number of non-classical monocytes (pro-inflammatory monocytes), representing a state of immunosuppression31.
The two-sample MR study in this study was conducted with the published large GWAS cohorts, which included a CKD dataset with a total sample size of more than 400,000 and an immunologically characterized European population, making this a statistically analytical study with a large sample size and strong statistical efficacy. The results of this study are generally plausible due to the use of genetic IVs, significant mitigation of confounding factors, and causal inference employing multiple MR analyses. However, some limitations remain. Firstly, because the GWAS data employed were limited to European populations, although population heterogeneity was reduced, selective bias may have occurred, affecting the representativeness and generalisability of the results. Second, because GWAS data do not currently contain individual patient information, this study was unable to perform stratified analyses to obtain more precise results. Third, bi-directional MR analyses are very helpful in solving the causal network direction problem, but biological mechanisms must also be considered when interpreting MR results, and statistical effect values alone cannot provide this information. Fourth, while applying FDR adjustments for multiple testing, the use of a relatively loose threshold (1 × 10− 5) given the limited sample size and flexible criteria for assessing results may lead to more false positives. Overall, a randomised controlled trial in CKD would be the next step in this study to reduce the potential impact of confounding factors to obtain a higher level of evidence for causality.