Physiology textbooks teach us that the kidney controls water, ion and pH homeostasis, excretes metabolites and xenobiotics, but is also pivotal as an endocrine organ. Well-known for decades is the kidney's role in producing renin to control blood pressure and erythropoietin (EPO) to regulate red blood cell production. While renin secretion directly depends on the action of blood pressure sensors in the afferent arteriole of the kidney glomerulus, the sensing for EPO synthesis is somehow indirect: interstitial cells in the kidney register a decrease in venous oxygen pressure, as in anaemia, or arterial oxygen pressure, as in high altitude. Thus, it is not that erythrocytes are ‘counted’ but that decreased oxygen (i.e. hypoxia) is sensed. Control of EPO synthesis by hypoxia induction of its gene recently attracted even more attention when the Nobel Prize of 2019 was awarded for the discovery how cells sense oxygen (Fandrey et al. 2019). Understanding the fundamental process of oxygen sensing was achieved through tracing back oxygen-dependent EPO production, which led to the discovery of hypoxia-inducible factors (HIF). The isoform HIF-2 is now established to control EPO expression. Oxygen-dependent regulation of HIF-2 depends on the activity of hydroxylases, with enzymatic activity dependent on the oxygen tension. Prolyl hydroxylases (PHD-1, PHD-2 and PHD-3) determine the abundance of HIF proteins while an asparagyl hydroxylase (FIH-1) controls the activity of HIF as a transcription factor. PHD-2 and PHD-3 are already therapeutically targeted. PHD inhibitors were approved for clinical use in Europe last autumn to increase EPO production in renal anaemia by increasing HIF-2 in the kidneys. One further question has puzzled investigators all these years: which cell makes EPO? In this issue of The Journal of Physiology, work by Broeker et al. (2022) adds a fascinating new perspective to the discussion. In general, it is accepted that renal interstitial cells, positive for platelet-derived growth factor receptor β (PDGFR-β), synthesize EPO. Recently, genetic tagging and isolation of renal EPO-producing cells (REP) has been successful (Imeri et al. 2019). Likewise, renin-expressing cells can be tagged in mice. When Broeker et al. (2022) specifically deleted the oxygen sensors PHD-2 and PHD-3 in renin producing (renin+) cells, these juxtaglomerular cells surprisingly started to express EPO. Coincidently, kidney renin mRNA and plasma renin concentration were lowered indicating that co-deletion of PHD-2 and PHD-3 with substantial HIF-2 accumulation in juxtaglomerular renin+ cells converted these cells into EPO producers. The assumption that HIF-2 accumulation was pivotal was corroborated in mice with renin+ cell-specific deletion of von-Hippel–Lindau protein, which usually degrades HIF-2. However, when Broeker et al. used one of the above-mentioned PHD inhibitors, the endocrine shift from renin to EPO production was not observed. This is reassuring with respect to therapeutic use of the inhibitors because short term inhibition does not appear to interfere with renin-dependent blood pressure regulation when therapy is aimed at increasing EPO to treat anaemia. Obviously, only long-term HIF-2 stabilization induces EPO expression in juxtaglomerular renin+ cells and this requires cellular reprogramming. After PHD-2 and PHD-3 co-deletion, Broeker et al. found not only renin but connexin 40 expression (typical for renin+ cells) reduced and next to EPO also CD73 (typical for EPO-producing interstitial cells) increased, which indicates a phenotypic shift of juxtaglomerular cells. Even more surprising was the fact that some of the interstitial cells, positive for EPO expression and PDGFR-β, were found positive for renin mRNA. However, because the interstitial cells were negative for renin protein – other than juxtaglomerular cells – the authors distinguish these two cell types. Still, Broeker et al. (2022) conclude that interstitial renin-expressing cells belong to a pool of potential EPO producers which can be recruited by PHD inhibition. These results are in good accordance with a recent publication by Miyachi et al. (2021) who detected renin and EPO co-expression in the kidneys of anaemic mice. Interestingly, the study by Broeker et al. sheds new light on potentially different roles of the oxygen sensors PHD-2 and PHD-3 within the cellular context: while PHD-2 deletion alone resulted in increased EPO production in interstitial renin-positive cells, the total deletion of PHD-2 and PHD-3 was required to induce EPO expression in juxtaglomerular renin-positive cells. It is still unresolved how the different oxygen sensors provide specific sensitivity to hypoxia in different cell types. Broeker et al. (2022) suggest that PHD-3 activity under physiological circumstances prevents EPO expression in juxtaglomerular cells and leaves these cells as important producers of renin. High oxygen tension in the vas afferens, where juxtaglomerular cells for renin expression usually exist, perfectly fits with this assumption, because PHD-3 activity will be high. Inhibitors of PHDs are not yet highly specific for the PHD isoforms. Therefore, it will be of considerable interest to closely follow the effects of chronic PHD inhibitor treatment. Given the widespread functions of HIF, it is so far surprising that PHD inhibitors approved to ameliorate renal anaemia appear to only affect EPO-producing cells in the kidney. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. None declared. Sole author. Open Access funding enabled and organized by Projekt DEAL. No funding was received for this work.