We read with interest and surprise the study of Zhao et al. on the therapeutic effects of the SGLT2 inhibitor empagliflozin in systemic lupus erythematosus (SLE) and lupus-like nephritis of MRLlpr mice.1 The concerns are the following: (1) SGLT2 is a sodium-glucose transporter mainly expressed in the proximal tubule of the kidney. SGLT2 inhibitors enhance sodium and glucose excretion into the urine, and also other mechanisms that exert beneficial effects on the cardiorenal system, glucose metabolism and haematopoiesis. In contrast, no direct effects on autoimmunity are to be expected. The fact that the authors report a suppression of all aspects of SLE and the related autoimmunity, that is, the profound suppression of total IgG and of double stranded DNA (dsDNA) autoantibodies produced by autoreactive immune cell clones in lymphoid organs and bone marrow, raises serious doubts, and no mechanism for this immunosuppressive effect is provided. (2) The authors seek to demonstrate SGLT2 protein expression in podocytes in human kidney biopsies and MRLlpr mouse kidneys but the lack of a tubular signal in figure 2 clearly demonstrates that the antibodies used do not detect SGLT2.1 Indeed, a prominent staining of SGLT2 in the brush border of the tubular cells and almost no expression in the glomeruli of human kidney biopsies with anti-neutrophil cytoplasm antibodies (ANCA) vasculitis or lupus nephritis2 is consistent with several single-cell RNA (scRNA) sequencing data sets of healthy and diseased human kidneys documenting no or low SGLT2 expression levels in podocytes. This contrasts the author’s finding of strong SGLT2 protein expression in a transgenic ‘podocyte’ cell line. (3) Podocytes in such healthy and diseased kidney scRNA sequencing data sets are also negative for NLRP3 transcripts, hence, all data and speculations presented on the involvement of the NLRP3 inflammasome are incompatible with known evidence. No appropriate experimental tools and controls were used. In particular, NLRP3 immunostaining (in figure 4G)1, once more, lacks a positive signal in resident or infiltrating mononuclear phagocytes identifying the presented signals as unspecific. In this sense, we recently disproved the presence of a functional NLRP3 inflammasome in primary human podocytes and mouse podocytes in vivo.3 In addition, we have performed similar studies with empagliflozin in the same mouse model and did not observe any of the reported findings (unpublished). While we believe that patients with lupus nephritis will benefit from SGLT2 inhibition in terms of the progression of chronic kidney disease and the related cardiovascular morbidity, the report of Zhao and coauthors seems to imply that SGLT2 inhibition would be a potent suppressor of systemic autoimmunity in SLE itself. The paper, as it stands, makes conclusions not supported by the provided data.

Data sharing not applicable as no data sets generated and/or analysed for this study. NA.

Not applicable.

Not applicable.