Driven by the diabetes and metabolic syndrome epidemics, chronic kidney disease (CKD) has emerged as a major health and socioeconomic burden worldwide (1–4). Management of CKD remains a challenge for the nephrologist, evidenced by the increasing need for renal replacement therapy. Many patients reach end-stage renal disease (ESRD) having become refractory to regimen that correct disease etiology. Hemodialysis, the mainstay for management of ESRD, carries risks of cardiovascular and cerebrovascular accidents and infections (5), and is a bridge to organ transplantation. Given the relative paucity in the number of donor kidneys available for transplant, there is urgent need for novel therapeutics that target and improve the renal microarchitecture, offering the potential for tissue repair and restoration of function.

Autologous cell therapy employs an individual’s cells, which typically are cultured and expanded outside the body and then introduced into the donor (6). Advantages of such an approach include reduced risks for alloimmune events, bioincompatibility, and disease transmission (6). Autologous selected renal cell (SRC) therapy involves use of biopsy-sourced kidney cells that are expanded in culture, submitted to gradient separation followed by implantation of a select fraction into the donor’s kidney (7–13). Phenotype-labeling studies suggest that SRCs are composed primarily of renal epithelial cells, including podocytes (7–13), a finding consistent with the source of SRCs, viz. the renal cortex. Numerous reports indicate that the tubular epithelium has the capacity to regenerate, repair, and re-epithelialize in response to insult (14–16). Recent evidence suggests that podocytes harbor a distinct mechanism for repopulation as well (17, 18). Consistent with these observations, non-clinical data indicate that administration of SRCs to diseased kidneys stimulates a reparative response and improvement in renal filtration (8–10, 12). Evaluation of glomerular filtration rate in a subset of diabetic kidney disease (DKD) patients at increased risk for ESRD suggests that randomization to SRCs (REACT™) is associated with stabilization of renal function (13).

Transcriptomic analyses indicate overexpression of certain mRNA by SRCs (10). These mRNA, differentially expressed vs. the source biopsy, include cadherin 1/cdh1, cubulin/cubn, nephrin/nphs1, erythropoietin/epo, kinase domain region/kdr [vascular endothelial growth factor (VEGF) receptor gene], hairy and enhancer of split-1/hes1, and platelet endothelial cell adhesion molecule/pecam1 (10). In the present study, we coupled knowledgebase data with empirical results from in vitro and in vivo assays to elucidate a potential mechanism of action underlying reparative and restorative activity of SRCs.

Using a targeted transcriptomic approach, we previously reported that SRCs differentially express cubn, cdh1, pecam1, hes1, epo, kdr, and nphs1 (10). We queried Humanbase (19) to determine whether these mRNA form interactomes within the kidney. Each of these mRNA is indeed expressed by the kidney (Figure 1A), suggesting that it is not a product of the SRC preparation process. Predictive analytics coupled with empirical data indicate that while these mRNA form interactomes within the renal compartments (Figures 1B–E) strength of their interaction confidence profiles for kidney < tubules < glomeruli < podocytes (Figure 1F). These data suggest that cross-talk between these nodes is likely compartmentalized, occurring primarily within the tubular and glomerular compartments. The Human Protein Atlas (20) houses antibody-imaging based distribution profiles for gene products. Immunohistochemical mapping data indicate that staining for CDH1 antibody (Figure 2A) is punctate and prominent in the collecting ducts and distal tubules. Staining for the EPO (Figure 2B) and CUBN (Figure 2C) antibodies is localized to the tubules, whereas PECAM, NPHS1 and KDR antibodies exhibit punctate glomerular staining (Figures 2D–F, respectively). These gene product distribution profiles, consistent with the SRC interactome profiles, also point to compartmentalization of SRC activity within the tubules and glomeruli.

Next, we sought to identify functional mechanisms governing the tissue-reparative activity of SRCs. Seeding the SRC mRNA into Humanbase and gene ontology (GO) biological process (BP) (21) indicates that these nodes participate in a suite of kidney development activities, including epithelial cell differentiation, renal vasculature development, and glomerular and nephron development (Table 1). Not surprisingly, query of the Human Fetal Kidney Atlas (22) indicates that these mRNA are also expressed by cells involved in embryologic kidney development, including nephron and podocyte progenitor cells, s- and comma-shaped body cells, pre-tubular aggregate cells and early proximal tubule cells (Figure 3). These data suggest that SRCs harbor nephrogenic potential, which we next evaluated in proof-of-concept in vitro and in vivo assays. Placed in culture, SRCs formed organoids that express the tubular phenotypic marker chemokine receptor 4 [CXCR4, (23)] (Figures 4A, B), and when cultured in the presence of hydrogel, these SRC-derived organoids formed tubules (Figure 4C). Interventional effects of SRCs were evaluated in the rat model of sub-total nephrectomy (Nx). Five days following SRC administration into the renal cortex of Nx rats, glomerular development, including comma- and S-shaped body cell formation (Figures 5A, B, respectively) and capillary loop stage (Figure 5C), advancing to maturing stages of glomerular morphogenesis was observed (Figures 5D, E). Treatment with SRCs did not affect mean arterial pressure (MAP; 97 ± 3 mmHg in control; 119.8 ± 8 mmHg in Nx; 106 ± 9 mmHg in Nx + SRC; p > 0.05). Compared to the control cohort, rats submitted to Nx exhibited an increase in blood urea nitrogen (BUN) and serum creatinine (SCr) (Figures 5F, G, respectively). Randomization to SRCs was associated with improved renal filtration evidenced by reduced BUN and SCr levels (Figures 5F, G, respectively). Masson’s trichrome and periodic acid Schiff (PAS)-stained sections were evaluated in each group. Minimal histological changes were observed in kidneys from the control cohort (Figure 5). The remnant kidney in the Nx cohort was characterized by glomerular and tubulointerstitial injury (Figures 5H, I). Glomerular pathology included severe focal segmental glomerulosclerosis, glomerular atrophy, adhesions of sclerotic segment to Bowman’s capsule, and shrinkage of capillary tufts, accompanied by tubular dilatation and tubular casts. Dilated tubules with an accumulation of proteinaceous casts were prominent, as were tubular atrophy and tubulointerstitial fibrosis. By contrast, SRC treatment was associated with reduced glomerular changes, consisting predominantly of compensatory glomerular hypertrophy characterized by enlargement of glomeruli without appreciable injury, reduced tubular dilatation and protein accumulation, and markedly reduced tubulointerstitial fibrosis (Figures 5H, I).

SRCs, a renal epithelial cell-enriched platform, are being advanced as autologous cell-based therapy for the treatment of CKD (13). In the present study, we report that the gene products of cdh1, cubn, nphs1, epo, kdr, hes1, and pecam1, mRNA differentially expressed by SRCs (10), reside within the tubular and glomerular compartments. Interactomes formed by these nodes also appear localized within the tubules and glomeruli and are involved in a suite of kidney developmental activities. Culturing of SRCs results in formation of organoids, and in the presence of a scaffold, these SRC organoids assemble into tubules. Implanted into the diseased kidney, SRCs are associated with nephron development, and improvement in renal function and tissue microarchitecture. These data suggest that SRCs harbor nephrogenic potential, which may underlie its reparative and restorative effects (Figure 6).

SRCs (REACT™) represent autologous cell-based therapy—a platform that has been deployed to bioengineer skin substitutes, aid wound healing, counteract chronic inflammation, treat burns and pressure ulcers, and improve postoperative healing (6). Derived from a corticomedullary biopsy, SRCs are prepared via a multistage process (U.S. Patent 8,318,484), including enzymatic dissociation of renal tissue followed by buoyancy separation of the cells. Two cellular fractions are selected and combined to form rat SRCs, which comprise primarily renal epithelial cells, including podocytes (7–12). Extensive non-clinical evidence corroborated by emerging clinical data support the notion that orthotopic implantation of SRCs are reparative and restorative (8–10, 12, 13). In the ZSF1 rat model of DKD (12), intervention with SRCs improved survival and mitigated loss of renal structure and function. Emerging Phase II clinical data from a subset of diabetic patients at increased risk for ESRD suggests that randomization to SRCs are associated with with stabilization of renal filtration (18). To delineate the mechanism of action underlying SRC activity in the present study coupled data from search engines, including Humanbase (19), GO BP (21), and tissue atlases (20, 22), with results from in vitro and in vivo assays. HumanBase and GO BP learn biological associations from extensive data collections while modeling tissue- and BP-specific gene interactions by leveraging experimentally verified tissue expression and gene function. Tissue atlases, repositories of protein expression profiles at single cell and subcellular resolution, marry spatial biology with function. Query of these knowledgebases were associated with three independent findings united by a consistent and repetitive theme. Genes expressed by SRCs viz. cdh1, cubn, nphs1, epo, kdr, hes1, and pecam1, appear to form interactomes within the tubules and glomeruli. Second, SRC gene products are localized to these renal compartments. Third, SRC genes participate in a suite of kidney developmental activities. Epithelial cell differentiation, renal vasculature development, and glomerular and nephron development are outputs of these genes or of networks formed by these genes. That SRCs recapitulate events associated with embryologic kidney development is supported by the observation that renal progenitor cells also express these genes.

The hallmark finding of this study is proof-of-concept for the nephrogenic potential of SRCs. In culture, SRCs forms organoids characterized by the expression of a tubular marker CXCR4, a receptor involved in epithelial patterning, tubulogenesis and renal morphogenesis (23). Indeed, inactivation of this receptor in embryonic kidney explants results in impaired ureteric bud branching and mesenchymal tubulogenesis, and renal malformation (23). Consistent with the role of this receptor, SRC organoids, in the presence of a scaffolding material like hydrogel, assemble into tubules. Indeed, tubule formation in the presence of a scaffold involves budding and branching morphogenesis (24). The primary role of the scaffolding material is to act as an extracellular matrix (ECM) and overcome variations in differentiation capacity inside growing organoids, allowing them to survive, grow and proliferate (24). The ECM typically has tiny pores that allow the passage of nutrients and gasses to give the cells the environment they need to thrive. The ECM also provide essential cues to cells rendering them critical for the establishment of physiologically relevant 3D tissue cultures (24). Several important observations emerge from these data. First, SRC derived from the kidneys of obese ZSF1 rats, a model of progressive DKD (12), retains the capacity for organoid formation and tubular assembly. Second, SRC organoids appear committed to renal development, unlike induced pluripotent stem cells that can mature into a non-renal cell (25). These observations are clinically relevant from the perspective of use in a DKD patient population and test article safety (25). Of significance, administration of SRCs was associated with beneficial effects in the rodent Nx model of kidney disease. Randomization to SRCs was associated with improvement in BUN and SCr, clinically reported measures of renal filtration. Accompanying these physiologic findings was a preservation of renal microarchitecture with treatment benefits observed across both the tubulointerstitial and glomerular compartments. Since there were no observed hemodynamic changes associated with SRC therapy, these treatment effects were likely directly related to mitigation of renal injury. Furthermore, interventional benefit was observed in female rats atop the protection afforded by endogenous estrogen (26), a significant finding in terms of the demographics likely to benefit from this therapy. Administration of SRCs into the kidney was accompanied by nephrogenesis characterized by hallmark features including comma- and S-shaped bodies and maturing glomeruli (27, 28). Sourced from the renal cortex, the SRC manufacturing process involves cell expansion, buoyancy separation, and selection of tubular epithelial cells and podocytes (7, 10). Numerous studies have documented the innate capacity of the tubular epithelium to regenerate following insult (14–16). Indeed, renal epithelial regeneration is thought to be mediated by the actions of scattered tubular cells/renal progenitor cells and/or via dedifferentiation of certain tubules (14, 15). Recent studies indicate that the podocyte compartment is dynamic and harbors a mechanism for podocyte replenishment (16, 17). These cell populations with reparative capacity while presumably endogenous to every kidney may lack critical mass in CKD and/or find themselves in an ECM-rich environment not amenable to their activation. The cell expansion and selection process associated with its manufacture makes SRCs a conduit/platform carrying a critical and enriched mass of renal epithelial cells, including podocytes to restitute kidney microarchitecture and function. Indeed, this is borne out by the differential expression of mRNA with nephrogenic potential.

SRCs (REACT™) are being evaluated in a Phase 3 clinical trial in subjects at increased risk (CKD 3b/4) for ESRD. The present study suggests that SRCs are a standalone platform harboring nephrogenic potential, which may form the basis, at least in part, for its reparative and restorative activity.

The original contributions presented in this study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

This animal study was reviewed and approved by Charles River Labs.

PN, AB, and ER performed the studies and analyzed the data. TB and DJ oversaw the project. PN wrote the manuscript. AB, ER, TB, and DJ revised the manuscript. All authors contributed to the article and approved the submitted version.