The human kidney contains up to 2 million epithelial nephrons responsible for blood filtration. Regenerating the kidney requires the induction of the more than 20 distinct cell types required for excretion and the regulation of pH, and electrolyte and fluid balance. We have previously described the simultaneous induction of progenitors for both collecting duct and nephrons via the directed differentiation of human pluripotent stem cells. Paradoxically, although both are of intermediate mesoderm in origin, collecting duct and nephrons have distinct temporospatial origins. Here we identify the developmental mechanism regulating the preferential induction of collecting duct versus kidney mesenchyme progenitors. Using this knowledge, we have generated kidney organoids that contain nephrons associated with a collecting duct network surrounded by renal interstitium and endothelial cells. Within these organoids, individual nephrons segment into distal and proximal tubules, early loops of Henle, and glomeruli containing podocytes elaborating foot processes and undergoing vascularization. When transcription profiles of kidney organoids were compared to human fetal tissues, they showed highest congruence with first trimester human kidney. Furthermore, the proximal tubules endocytose dextran and differentially apoptose in response to cisplatin, a nephrotoxicant. Such kidney organoids represent powerful models of the human organ for future applications, including nephrotoxicity screening, disease modelling and as a source of cells for therapy.

With the prevalence of end-stage renal disease rising 8% per annum globally, there is an urgent need for renal regenerative strategies. The kidney is a mesodermal organ that differentiates from the intermediate mesoderm (IM) through the formation of a ureteric bud (UB) and the interaction between this bud and the adjacent IM-derived metanephric mesenchyme (MM). The nephrons arise from a nephron progenitor population derived from the MM (ref. ). The IM itself is derived from the posterior primitive streak. Although the developmental origin of the kidney is well understood, nephron formation in the human kidney is completed before birth. Hence, there is no postnatal stem cell able to replace lost nephrons. In this study, we have successfully directed the differentiation of human embryonic stem cells (hESCs) through posterior primitive streak and IM under fully chemically defined monolayer culture conditions using growth factors used during normal embryogenesis. This differentiation protocol results in the synchronous induction of UB and MM that forms a self-organizing structure, including nephron formation, in vitro. Such hESC-derived components show broad renal potential ex vivo, illustrating the potential for pluripotent-stem-cell-based renal regeneration.

Mutations in SALL4, the human homolog of the Drosophila homeotic gene spalt (sal), cause the autosomal dominant disorder known as Okihiro syndrome. In this study, we show that a targeted null mutation in the mouse Sall4 gene leads to lethality during peri-implantation. Growth of the inner cell mass from the knockout blastocysts was reduced, and Sall4-null embryonic stem (ES) cells proliferated poorly with no aberrant differentiation. Furthermore, we demonstrated that anorectal and heart anomalies in Okihiro syndrome are caused by Sall4 haploinsufficiency and that Sall4/Sall1 heterozygotes exhibited an increased incidence of anorectal and heart anomalies, exencephaly and kidney agenesis. Sall4 and Sall1 formed heterodimers, and a truncated Sall1 caused mislocalization of Sall4 in the heterochromatin; thus, some symptoms of Townes-Brocks syndrome caused by SALL1 truncations could result from SALL4 inhibition.