Kidney regeneration from pluripotent stem cells is receiving a lot of attention because limited treatments are currently available for chronic kidney disease (CKD). It has been shown that uremic state in CKD is toxic to somatic stem/progenitor cells, such as endothelial progenitor and mesenchymal stem cells, affecting their differentiation and angiogenic potential. Recent studies reported that specific abnormalities caused by the non-inherited disease are often retained in induced pluripotent stem cell (iPSC)-derived products obtained from patients. Thus, it is indispensable to first assess whether iPSCs derived from patients with CKD due to non-inherited disease (CKD-iPSCs) have the ability to generate kidneys. In this study, we generated iPSCs from patients undergoing haemodialysis due to diabetes nephropathy and glomerulonephritis (HD-iPSCs) as representatives of CKD-iPSCs or from healthy controls (HC-iPSCs). HD-iPSCs differentiated into nephron progenitor cells (NPCs) with similar efficiency to HC-iPSCs. Additionally, HD-iPSC-derived NPCs expressed comparable levels of NPC markers and differentiated into vascularised glomeruli upon transplantation into mice, as HC-iPSC-derived NPCs. Our results indicate the potential of HD-iPSCs as a feasible cell source for kidney regeneration. This is the first study paving the way for CKD patient-stem cell-derived kidney regeneration, emphasising the potential of CKD-iPSCs.
Chronic kidney disease (CKD) is a major problem worldwide and the number of patients with CKD continues to rise1,2. The replacement of kidney function in patients with end-stage renal disease requires dialysis or kidney transplantation. Although kidney transplantation can improve the quality of life and prolong the life expectancy of patients with CKD3, the insufficient number of donor organs make this a suboptimal solution in the treatment of severe renal diseases4. Hence, kidney regeneration by induced pluripotent stem cells (iPSCs) is expected to be particularly helpful.
Kidneys arise from metanephros, which develop via the reciprocal interaction between the metanephric mesenchyme, containing nephron progenitor and stromal progenitor cells and the ureteric bud (UB)5. Recently, kidney regeneration from pluripotent stem cells (PSCs) has made remarkable progress and several studies have reported the successful differentiation of PSCs into nephron progenitor cells (NPCs) and UB in vitro, by recapitulating the development of the metanephric kidney6,7,8,9,10,11. However, because kidney development is a very complex process, we need to combine NPCs, stromal progenitor cells and UB spatiotemporally at an appropriate ratio for kidney generation. More importantly, stromal progenitor cells have not yet been generated from PSCs and it will still take time to generate whole kidneys from PSCs in vitro.
To generate whole functional kidneys, we have employed the ‘organogenic niche method’, which uses heterozygous embryos as an organ factory12,13,14. In this method, progenitor cells are applied at the area of nephrogenesis, where they differentiate into kidneys by borrowing the developing programs of the growing xeno-embryos. Notably, we have recently generated transgenic embryos conditionally expressing the diphtheria toxin receptor and treated them with diphtheria toxin to ablate native NPCs from the nephrogenic area. Then, we applied allogenic (mouse) and xenogenic (rat) exogenous NPCs to the nephrogenic area and succeeded in replacing native NPC with the exogenous NPCs in isolated mouse metanephros15. In this system, the native niche supports the differentiation of exogenous NPCs into neo-nephrons, while exogenous NPCs support the proliferation of native UB. Importantly, the distal neo-nephrons connected with the native UB. In such way, we have succeeded in providing a suitable niche for exogenous NPCs, while keeping the original structure of the metanephros.
iPSCs generated from patients are a promising tool among the tailor-made therapeutic approaches because they supply patient-derived cells/tissues/organs. The regeneration of kidneys derived from iPSCs of patients with CKD will help to circumvent current problems such as organ shortages4, immune rejection and life-long immunosuppression16. We plan to use NPCs differentiated from iPSCs derived from patients with CKD that is not due to a genetic cause (CKD-iPSCs) as a cell source for the strategy described above, after optimizing it for human cells15, to regenerate kidneys from CKD-iPSCs. Because non-inherited kidney diseases are the major cause of CKDs, such as diabetic nephropathy and glomerulonephritis, a considerable proportion of patients with CKD would be able to benefit from this procedure. Although the cell products derived from inherited kidney disease-specific iPSCs demonstrate a disease specific phenotype17, nothing has been reported on the biological properties of cell products derived from patients with CKD due to non-inherited causes. Recently, several studies have shown that non-inherited disease-specific abnormalities are often retained in iPSC-derived products obtained from patients18,19,20. CKD leads to accumulation of organic compounds (uremic toxins) in the bloodstream; these cause many toxic effects, including reduced proliferation capacity, differentiation abnormalities,21,22 and angiogenic function22,23,24,25,26,27 in stem and progenitor cells and premature aging in stem27 and peripheral blood mononuclear cells (PBMCs)28.
Hence, to generate CKD-iPSC-derived kidneys, it is indispensable to first assess whether the CKD-iPSCs and CKD-iPSC-derived products retain the adverse CKD-specific abnormalities, especially the differentiation ability and angiogenic function, which are crucial for kidney regeneration. Therefore, in this study, we generated iPSCs from haemodialysis (HD) patients with CKD that was not due to a genetic cause (HD-iPSCs) and used them as representatives of CKD-iPSCs; additionally, we also generated iPSCs from age- and sex-matched healthy controls (HC; HC-iPSCs). Then, we compared the characteristics of HD-iPSCs, HD-iPSC-derived NPCs and HD-iPSC-derived nephrons to those from HC-iPSCs. In such way, we assessed the potential of HD-iPSCs as cell source for kidney regeneration.
Generation of iPSCs from patients on haemodialysis and healthy controls
Three patients with end-stage renal disease attending our institution for HD were enrolled in the study. All HD patients had received standard dialysis therapy for renal insufficiency. Of the three patients, one had chronic glomerulonephritis, one had diabetic nephropathy and one had rapidly progressive glomerulonephritis. HD-iPSC lines were generated from PBMCs of these patients. Similar to HD-iPSC lines, HC-iPSC lines were generated from PBMCs of two healthy controls. iPSC donor characteristics are presented in Table 1. Although the iPSC generation efficiency varied among the individuals, we succeeded in generating four HD-iPSC lines, namely HD-1, HD-2, HD-3 and HD-4 and three HC-iPSC lines, namely HC-1, HC-2 and HC-3. All iPSC lines used in this study showed the typical colony morphology of iPSCs29 (Fig. 1a), maintained a normal karyotype (Fig. 1b) and met the authenticity criteria of iPSCs: (1) expression of stem cell markers (Fig. 1c); (2) expression of human embryonic stem cell (hESC)-specific surface antigens, including stage-specific embryonic antigen-4 (SSEA-4), tumour-related antigen (TRA)-1-60 and TRA-1-81 (Fig. 1d); and (3) ability to differentiate into the three germ layers through teratoma formation (Fig. 1e). In HD-2 and HD-3, the replication-defective and persistent Sendai virus (SeVdp) vector was completely removed by RNA interference (Fig. 1f). Additionally, we also used 201B7 cells obtained from the RIKEN BioResource Centre as HC-iPSCs, namely HC-4, for matching patient age and sex against HD-iPSC lines. Flow cytometric analysis using rBC2LCN, which is a novel detection reagent for iPSCs30, showed that almost all the cells were rBC2LCN positive and there was no significant difference between the two groups (HC and HD group, n = 4 in each, Fig. 1g,h, Supplementary Figure S1). We cultured the iPSCs on feeders at first and then transitioned them to feeder-free conditions for differentiation. All iPSC lines reached 80–90% confluence under feeder-free conditions in 6-well plates after approximately 7–8 days. We found no significant differences in doubling time between the two groups (HC and HD group, n = 4 in each, Fig. 1i).
HD-iPSCs differentiated into NPCs as efficiently as HC-iPSCs
The four HD-iPSC lines and the four HC-iPSC lines were induced into nephron progenitor spheres (NPSs) including NPCs, using an embryoid body-mediated differentiation protocol6. The size of the spheres derived from the HD-iPSC lines increased over time (Fig. 2a). Quantitative reverse transcription polymerase chain reaction (qRT-PCR) showed that the spheres expressed specific markers such as Brachyury (T), Odd-skipped related transcription factor 1 (OSR1), Wilms tumour 1 (WT1), Paired box 2 (PAX2) and SIX homeobox 2 (SIX2) at the expected time points following the development of metanephric nephron progenitors6, similar to the HC-iPSC-derived spheres (Fig. 2b). On day 14, the NPSs, derived from both iPSC lines, expressed the NPC markers WT1, PAX2 and SIX2 (Fig. 2c).
We performed the following experiments to examine the NPC differentiation ability of HC- and HD-iPSC lines: (1) NPC marker expression analysis by qRT-PCR; (2) NPC marker protein expression analysis by immunostaining; and (3) NPC surface marker analysis by flow cytometry. For qRT-PCR, 6 NPSs were combined to represent one group and three groups for each iPSC line were analysed. We found no significant differences in the expression of the NPC markers WT1, PAX2 and SIX2, between HC and HD groups (n = 12 in each group, Fig. 2d). For immunostaining, 12 NPSs were combined for each iPSC line. Again, we found no significant differences in the percentage of cells expressing NPC markers (Fig. 2e,f). For flow cytometry and cell sorting, 48 NPSs were combined for each iPSC line. We plotted forward versus side scatter of dissociated NPSs; DAPI positive dead cells are shown in blue (Fig. 2g). To eliminate cell debris and dead cells from the analysis, we placed the gate shown in red in the figure. Kaku et al.31 have shown that NPCs constitute an integrin subunit alpha 8 (ITGA8)+/platelet derived growth factor receptor alpha (PDGFRA)− population. Therefore, we examined the ITGA8+/PDGFRA− population in the gated cells and found no significant difference between HC- and HD-iPSC-derived NPSs (n = 4 in each group, Fig. 2h,i, Supplementary Figure S2).
Isolated HC- and HD-iPSC-derived NPCs showed comparable levels of the nephron progenitor markers
To examine the iPSC-derived NPC characteristics between the HC and HD groups, we sorted NPCs from NPSs using the immunomagnetic separation method32. This system allows the separation of cells with lower damage and in shorter time when compared with conventional cell sorting methods. We isolated ITGA8+/PDGFRA− cells in two steps. First, we performed a PDGFRA negative selection and then an ITGA8 positive selection (Supplementary Figure S3). In agreement with previous studies6,31, our results showed that the size of the PDGFRA+ population obtained using the embryoid body-mediated differentiation protocol was very limited. Next, we further separated, in this population, ITGA8 negative or positive cells. As a result, we successfully enriched both the ITGA8−/PDGFRA− (from 76.2 ± 3.3% to 91.6 ± 3.8% in HC-iPSC-derived NPSs, p = 0.02 and from 74.2 ± 5.1% to 91.4 ± 2.6% in HD-iPSC-derived NPSs, p = 0.02, Fig. 3a,b, Supplementary Figure S2) and ITGA8+/PDGFRA− (from 23.4 ± 3.2% to 84.3 ± 2.8% in HC-iPSC-derived NPSs, p < 0.0001 and from 25.6 ± 5.2% to 87.2 ± 1.9% in HD-iPSC-derived NPSs, p < 0.0001, Fig. 3c,d, Supplementary Figure S1) fractions. Then, we performed gene expression analyses of glial-cell derived neurotrophic factor (GDNF), which is one of the NPC markers and is a key signal of the interaction between NPCs and UB5, in addition to the analyses of WT1, PAX2, SIX2. The expression levels of NPC markers in the ITGA8+/PDGFRA− population were higher than in the ITGA8−/PDGFRA− population (ITGA8+/PDGFRA− population; n = 8, ITGA8−/PDGFRA− population; n = 4, Fig. 3e). We found no significant difference in the expression level of nephron progenitor markers in the ITGA8+/PDGFRA− population between HC and HD groups (n = 4 in each group, Fig. 3f). Isolated ITGA8+/PDGFRA− and ITGA8−/PDGFRA− cells derived from HD-2 were cultured in 96-well plates for one day to form aggregates. Then, we co-cultured these aggregates with mouse embryonic spinal cords for nine days. ITGA8+/PDGFRA− aggregates showed tubulogenesis (Fig. 3g), while ITGA8−/PDGFRA− aggregates did not (Fig. 3h), as indicated in a previous study that used the conventional cell sorting method31. These data indicate that we could purify iPSC-derived NPCs using the immunomagnetic separation method and that the immunomagnets did not affect the differentiation of isolated iPSC-derived NPCs.
HD-iPSC-derived NPCs showed possibility to differentiate into nephrons similar to the HC-iPSC-derived NPCs
Next, we examined whether HD-iPSC-derived NPCs could differentiate into nephrons similar to HC-iPSC-derived NPCs. We co-cultured NPSs including NPCs with mouse embryonic spinal cords for nine days. Although the differentiation efficiency varied among NPSs, most NPSs underwent robust tubulogenesis (Fig. 4a). We selected three well-differentiated spheres from each iPSC line, separated the well-differentiated parts, named iPSC-derived nephrons and used them for further analysis (Fig. 4b). We found no significant difference in the percentage of iPSC-derived nephrons per sphere between the HC and HD groups (n = 12 in each group, Fig. 4c). Reverse transcription-PCR (RT-PCR) showed that marker genes were expressed in multiple segments of the HD-2-derived nephrons, including podocytes and proximal and distal tubules, as in the HC-4-derived nephrons (Fig. 4d). To eliminate the possibility that iPSC-derived nephrons were contaminated with mouse spinal cord tissue, we performed additional RT-PCR assays, using mouse spinal cord (Sp) as a negative control (Fig. 4d). Next, to quantify the efficiency of nephron formation between HC and HD groups, we performed qRT-PCR for representative nephron markers: NPHS1 and NPHS2 as terminally differentiated podocyte-specific markers; low density lipoprotein-related protein 2 (LRP2) as proximal tubule marker; solute carrier family 12 member 1 (SLC12A1) as loop of Henle marker; polycystin 1 (PKD1) and polycystin 2 (PKD2) as distal tubule markers. In addition, we examined the expression of vascular endothelial growth factor A (VEGFA), which is secreted from developing podocytes and plays a critical role in nephrogenesis and growth of glomerular capillaries33. We obtained comparable transcriptional profiles in HC- and HD-iPSC-derived nephrons (n = 12 in each group, Fig. 4e). Histological analysis of periodic acid-Schiff (PAS) stained HD-iPSC-derived nephrons showed the presence of numerous glomeruli (Fig. 4f,g). Transmission electron microscopy revealed cell clusters that had primary foot processes (Fig. 4h). Immunostaining showed that they expressed the typical podocyte marker, WT1, in their nuclei, whereas Nephrin and Podocin were expressed on the basal sides of the podocytes (Fig. 4i–m), indicating mature podocyte formation. Some tubules expressed markers specific for proximal tubules, such as Jagged1, Lotus tetragonolobus lectin (LTL) and Megalin (Fig. 4n–p). Other tubules had macula densa-like structures (Fig. 4g) and expressed PAX2 and E-cadherin, indicative of distal tubule formation (Fig. 4q).
HD-iPSC-derived glomeruli showed possibility to attract blood vessels similar to HC-iPSC-derived glomeruli
Finally, we examined the angiogenic function of HC- and HD-iPSC-derived glomeruli in vivo using the cluster of differentiation 31 (CD31)/nephrin assay described by Sharmin and colleagues34. The authors transplanted iPSC-derived spheres, which had been co-cultured with spinal cords for one day, to initiate tubulogenesis beneath the kidney capsules of immunodeficient mice. Because the spheres might differ in their differentiation ability, it is important, before transplantation, to identify the spheres that will likely differentiate well in vivo. The longer spheres and spinal cords are co-cultured, the easier it is to identify the spheres that will differentiate well because of their morphological changes. Meanwhile, upon extensive co-culture, the spheres miss the time window for angiogenesis. In kidney development, the infiltration of angioblasts starts at the S-shaped body stage35. Resembled S-shaped bodies were observed four days after initiation of co-culture for differentiation6,34. Hence, we decided to co-culture NPSs derived from each iPSC line with spinal cord for three days instead of one. Then, we transplanted HC- and HD-iPSC-derived tissues beneath the kidney capsules of immunodeficient mice (Fig. 5a,b). Sharmin et al.34 found that blood vessels integrated into iPSC-derived glomeruli ten days after transplantation (11 days after initiation of co-culture). Hence, we collected iPSC-derived tissues nine days after transplantation (12 days after initiation of co-culture) and found that the size of the transplanted HD-iPSC-derived tissues had increased and small vessels had integrated from mice kidneys, similar to the transplanted HC-iPSC-derived tissues (Fig. 5c,d). Haematoxylin/eosin (HE) staining of HD-iPSC-derived tissue showed more mature glomeruli, which had open lumens in the glomerular capillary loops and proximal tubules in the transplanted tissues than in the HD-iPSC-derived nephrons cultured in vitro (Fig. 5e,f). Electron microscopy of HD-iPSC-derived tissue showed that glomeruli containing red blood cells had the typical three-layer structure of the glomerular capillary wall: fenestrated endothelial cells with diaphragms, glomerular basement membrane (BM) and podocytes (Fig. 5g–i). Slit diaphragm-like structures, located above the BM, were formed among podocytes. Moreover, the lumens of the proximal tubules were lined with brush borders (Fig. 5j). Immunohistochemistry of HD-iPSC-derived tissue showed that CD31-positive endothelial cells were integrated just beneath the Nephrin-positive slit diaphragm-like structures (Fig. 5k,l). The apical regions of the Cadherin6-positive proximal tubules were Megalin-positive (Fig. 5m). We used the number of CD31-positive glomeruli in the Nephrin-positive glomeruli to assess the angiogenic function in vivo (CD31/Nephrin assay). The percentage of CD31-positive glomeruli was over 90% in both HC- and HD-iPSC-derived glomeruli (n = 31 and 32 in the HC and HD group, respectively) and there was no significant difference between the two groups (Fig. 5n).
In the present study, we described the generation and characterisation of iPSC lines representing age- and sex-matched patients on haemodialysis, as well as control individuals. HD-iPSCs could differentiate into NPCs as efficiently as HC-iPSCs. HD-iPSC-derived NPCs and HC-iPSC-derived NPCs showed comparable levels of NPC marker expression. Moreover, HD-iPSC-derived NPCs showed possibility to become mature and vascularised nephrons in vivo, similar to HC-iPSC-derived NPCs. These findings suggest that HD-iPSCs possess sufficient nephron differentiation ability and HD-iPSC-derived nephrons can become functional in vivo. Therefore, our results suggest that HD-iPSCs are a useful cell source for kidney regeneration.
Several groups have succeeded in differentiating iPSCs into kidney organoids, which are 3D multicellular tissues mimicking6,7,8. To achieve kidney regeneration, we plan to use the ‘organogenic niche method’, which requires iPSC-derived NPCs as cell source for transplantation. Although Morizane et al.7 generated NPCs from iPSCs more efficiently than Taguchi et al.6, the NPCs generated with the protocol used by Taguchi et al.10 could interact with UB and glomeruli differentiated from the NPCs attract blood vessels in vivo34: both functions are indispensable in the generation of functional kidneys composed of iPSC-derived nephrons. Therefore, in this study, we used the protocol developed by Taguchi and colleagues.
Although the induction efficiency of iPSCs varied among different individuals, all iPSC lines that met the authenticity criteria for iPSCs showed almost comparable characteristics including proliferation rate (Fig. 1h,i and Supplementary Figure S1) and could differentiate into NPCs and nephrons regardless of HC or HD group upon appropriate stimulation (Figs 2d,f,i and 4e). In this study, we used 7 iPSC lines that were generated from PBMCs and 1 iPSC line that was generated from fibroblast. Although it was difficult to statistically compare the differentiation ability between iPSCs derived from PBMCs and from fibroblast, our result indicated the usefulness of HD-iPSCs derived from PBMCs in kidney regeneration. This is a very important finding for the clinical application of CKD patient-derived iPSCs because PBMCs can easily be isolated from patients, without invasive clinical procedures.
To generate kidneys from CKD-iPSC-derived NPCs, it is indispensable to identify the characteristics of HD-iPSC-derived NPCs, used as representatives of CKD-iPSC-derived NPCs, by comparing them with HC-iPSC-derived NPCs. To isolate iPSC-derived NPCs from differentiated spheres, we used an immunomagnetic separation method against NPC surface markers (ITGA8 and PDGFRA). As a result, we purified iPSC-derived NPCs (ITGA8+/PDGFRA−) efficiently and in a short time (Fig. 3d,e, Supplementary Figure S2) and showed that isolated HD-iPSC-derived NPCs expressed comparable levels of NPC markers as the HC-iPSC-derived NPCs (Fig. 3f). In kidney development, NPCs that receive signals from UB differentiate into nephrons, while UBs grow and ramify in response to signals from NPCs and connect to the distal end of nephrons36,37. GDNF, secreted from NPCs, is one of the NPC markers and a key molecule for reciprocal interaction. We showed that HD-iPSC-derived NPCs and HC-iPSC-derived NPCs expressed comparable levels of GDNF. This observation is important when generating kidneys that are composed of HD-iPSC-derived nephrons because insufficient interaction would cause retardation in UB branching, decreased kidney growth and low nephron number38.
Finally, we examined the angiogenic function of the iPSC-derived glomeruli. Although there are protocols for generating iPSC-derived glomeruli containing iPSC-derived endothelial cells8,9, glomeruli generated using the protocol we followed have been shown not to have internal vascular endothelial cells in vitro34. Nevertheless, iPSC-derived glomeruli still need to attract external endothelial cells, regardless of containing or not containing internal endothelial cells in vivo, that will be constituent elements of the glomerular tuft to become functional nephrons as the preexisting internal endothelial cells are lost in in vitro conditions39. In such conditions, angiogenesis, induced by the VEGFA secreted from the developing podocytes, is crucial for glomerular development and vascularisation33,40,41. We showed that the level of VEGFA expressed by HD-iPSC-derived nephrons, including developing podocytes, was comparable to that by HC-iPSC-derived nephrons in vitro (Fig. 4e). The result that HD-iPSC-derived glomeruli showed possibility to attract CD31-positive endothelial cells as the HC-iPSC-derived glomeruli in vivo without any supporting growth factors was consistent with the in vitro experiment (Fig. 5n). Moreover, similar to transplanted kidney organoids generated by other protocols39,42, transplanted HD-iPSC-derived glomerular maturation was observed with BM, fenestrated endothelium and mature podocytes in glomerular tufts. HD-iPSC-derived proximal tubular epithelial maturation was also observed with the development of brush border on the apical side. Our results suggested that HD-iPSC-derived nephrons had possibility to become functional nephrons in vivo, similar to the HC-iPSC-derived nephrons.
In summary, our data imply that CKD does not affect the characteristics and differentiation ability of HD-iPSCs and HD-iPSC-derived NPCs and angiogenic function of HD-iPSC-derived glomeruli. Therefore, HD-iPSCs are a plausible cell source for kidney regeneration. In addition, we found an easy and efficient method for isolating NPCs from differentiated spheres in this study. This method enables us to eliminate undifferentiated or unwanted cell population that might cause adverse effects, such as tumorigenesis, without affecting the differentiation of isolated NPCs into nephrons. In the future, we plan to use isolated HD-iPSC-derived NPCs as a cell source for the ‘organogenic niche method’. In addition, by combing this method with our previously developed urine drainage system43, we plan to drain urine from neo-kidneys composed of HD-iPSC-derived neo-nephrons.
However, several limitations of this study should be acknowledged. First, when we generated HD-iPSCs from PBMCs, we did not assess whether the parental PBMCs, from which iPSCs were generated, were actually affected by CKD. The longer patients suffer from CKD, more PBMCs are known to be affected by CKD44. Because we generated iPSCs from patients with long-standing CKD, we considered the PBMCs derived from these to be affected by CKD. Second, we could not analyse the function of iPSC-derived nephrons over a longer period because the mouse embryonic spinal cords that we used as inducers of differentiation became degenerative in a short time and the kidney capsule in which we transplanted the iPSC-derived spheres was not a suitable place for kidney development. As CKD is known to cause premature aging of PBMCs28, we need to assess if the characteristics of HD-iPSC-derived nephrons will be different from those of HC-iPSC-derived nephrons on long-term maintenance. Further studies are necessary to overcome this limitation; a possibility could be to provide a suitable environment, which enables long-term maintenance of iPSC-derived nephrons, mimicking the in vivo nephrogenic area, such as the transplantation of iPSC-derived NPCs into metanephros in which native NPCs are eliminated.
To our knowledge, this is the first report examining the potential of HD-iPSCs for kidney regeneration and their capability to generate nephrons and vascularised glomeruli from HD-iPSCs. Our data suggest that HD-iPSCs are a suitable source for kidney regeneration and indicate the usefulness of CKD-iPSCs. This study paves the way for patient-stem cell-derived kidney regeneration. Indeed, the generation of CKD patient-stem cell-derived kidneys would overcome not only organ shortages but also immune rejection issues and the life-long immunosuppression related to kidney transplantation.
Generation of iPSCs
Two HD-iPSC lines, HD-1 and HD-4 and three HC-iPSC lines, HC-1, HC-2 and HC-3, were generated from human PBMCs using episomal vectors, according to the protocol of the Centre for iPS Cell Research and Application (Kyoto, Japan). Two iPSC lines, HD-2 and HD-3, were generated from human PBMCs using the SeVdp vector expressing the four reprogramming factors OCT3/4, SOX2, cMYC and KLF4 and removing the vector genome by RNA interference, as previously described45,46.
Nephron progenitor cells/nephron induction from iPSCs
The eight iPSC lines were induced to NPSs including NPCs by a previously established method6. To differentiate nephrons, NPSs were co-cultured with mouse embryonic spinal cords taken from E12.5 embryos at the air-fluid interface of a polycarbonate filter (0.8 mm; Corning Inc., Corning, NY, USA) for nine days. The induction was performed at least two times for each iPSC line, independently. The percentage of the well-differentiated part per spheres was calculated using ImageJ software (National Institute of Health [NIH], Bethesda, MD, USA).
Flow cytometric analyses
rBC2LCN-FITC (180-02991; Wako), biotinylated anti-ITGA8 (BAF4076; R&D Systems, Minneapolis, MN, USA), allophycocyanin-conjugated streptavidin (405207; BioLegend, San Diego, CA, USA) and phycoerythrin-conjugated anti-PDGFRA (323506; BioLegend) were used for cell staining. Data were acquired and analysed using the MACSQuantify analysis software (Miltenyi Biotec, Bergisch Gladbach, Germany).
Cell sorting using the immunomagnetic cell separation system
PDGFRA negative selection was performed using Anti-PE MicroBeads UltraPure (130-105-639; Miltenyi Biotec), LS Columns (130-042-041; Miltenyi Biotec) and a MidiMACS Separator (130-042-302; Miltenyi Biotec). Then, ITGA8 positive selection was performed using Anti-Biotin MicroBeads UltraPure (130-105-637; Miltenyi Biotec), MS Columns (130-0042-201; Miltenyi Biotec) and MiniMACS Separator (130-042-102; Miltenyi Biotec). To assess the differentiation ability of the sorted cells, 3 × 105 ITGA8−/PDGFRA- and ITGA8+/PDGFRA- cells were aggregated in differentiation medium with 1 µM CHIR and 5 ng/mL human FGF9 overnight at 37 °C and then co-cultured with mouse embryonic spinal cords for nine days.
Transplantation of iPSC-derived tissues into immunodeficient mice
We modified the transplantation method previously described34. iPSC-derived NPSs were cultured with mouse embryonic spinal cords on polycarbonate filters for three days to initiate tubulogenesis. Before transplantation, spheres with spinal cords were cut into half and then the tissues were transplanted beneath the kidney capsules using small dispensing spoons under three types of mixed anaesthetic agents: medetomidine (Zenoaq, Fukushima, Japan), butorphanol (Meiji Seika, Tokyo, Japan) and midazolam (Astellas Pharma Inc., Tokyo, Japan). NOD/Shi-scid, IL-2RγKO Jic mice (male, 6 months old, In-Vivo Science Inc., Tokyo, Japan) were used as the host animals. Nine days after transplantation, the transplanted tissues were collected.
Data are expressed as the mean ± SEM. Means were compared using the unpaired two-tailed Student’s t-test. All analyses were performed using the PRISM7 software (GraphPad Software, La Jolla, CA, USA). P < 0.05 was considered as statistically significant.
All animal experiments were approved by the Experimental Animal Committee of The Jikei University, Japan (Permit Number: 2016-001) and conducted in conformity with the NIH Guide for the Care and Use of Laboratory Animals. All experiments using human iPSCs were performed in accordance with institutional guidelines and approved by the Ethics Committee of The Jikei University School of Medicine (Permit Number: 25-015). All donors provided written informed consent for collection of samples and subsequent analyses.
Data Availability Statement
The datasets generated during the current study are available from the corresponding author on reasonable request.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
We would like to thank Dr. Mahito Nakanishi, Research Centre for Stem Cell Engineering, National Institute of Advanced Industrial Science and Technology, for providing SeVdp and Dr. Toshiaki Tachibana, Division of Molecular Cell Biology, Core Research Facilities for Basic Science, The Jikei University School of Medicine, for electron microscopy analysis. Additionally, we would like to thank Editage (www.editage.com) for English language editing and publication support. This research was supported by the Japan Agency for Medical Research and Development (AMED, Grant No. 17ek0310006h0002); The Kidney Foundation, Japan (Grant No. JKFB15-1); and The Japan Society for the Promotion of Science (JSPS-KAKENHI, Grant No. 16H03175).