Supplementary Material for: Integrative analysis by mendelian randomization and large-scale single-cell transcriptomics reveals causal links between B cell subtypes and diabetic kidney disease

Introduction: The growing incidence of diabetic kidney disease (DKD) and the difficulties in its management underscore the need for a more comprehensive understanding of its pathogenesis. Recent studies have emphasized the significant impact of circulating immunity on the development of diabetic microvascular complications including retinopathy and neuropathy, research on circulating immunity in DKD remains limited. Methods: This study utilized mendelian randomization (MR) analysis to explore the potential independent causal relationships between circulating immune cells and DKD pathogenesis. Additionally, a combination of single-cell disease relevance score (scDRS) and immune cell infiltration analysis was employed to map the circulating immunity landscape in DKD patients. Results: Ten immune traits, including 5 of B cells, 2 of T cells, 2 of granulocytes and one of monocytes, were defined to be associated with the pathogenesis of DKD. Notably, IgD-CD27- B cell Absolute Count [IVW: OR, 1.102 (1.023 to 1.189), p=0.011] and IgD-CD24- B cell Absolute Count [IVW: OR, 1.106 (1.030 to 1.188), p=0.005] were associated with promoting DKD pathogenesis, while CD24+CD27+ B cell %B cell [IVW: OR, 0.943 (0.898 to 0.989), p=0.016] demonstrated a protective effect against DKD onset. The presence of B cell activating factor receptor (BAFF-R) on CD20-CD38- B cell [IVW: OR, 0.946 (0.904 to 0.989), p=0.015] and BAFF-R on IgD-CD38+ B cell [IVW: OR, 0.902 (0.834 to 0.975), p=0.009] also indicated a potential role in preventing DKD. Single-cell disease relevance score (scDRS) analysis revealed that two main subsets of B cells, naïve B and memory B cells, had a higher proportion of DKD-related cells or a higher scDRS score of DKD phenotype, indicating their strong association with DKD. Furthermore, immune infiltrate deconvolution analysis showed a notable decrease in the circulating memory B cells and class-switched memory B cells in DKD patients compared to those of DM patients without DKD. Conclusion: Our study revealed the causal relations between circulating immunity and DKD susceptibility, particularly highlighted the potential roles of B cell subtypes in DKD development. Further studies addressing the related mechanisms would broaden the current understanding of DKD pathogenesis.