http://www.kidney-international.org
mini review
& 2014 International Society of Nephrology
Nephron reconstitution from pluripotent stem cells Atsuhiro Taguchi1 and Ryuichi Nishinakamura1 1
Department of Kidney Development, Institute of Molecular Embryology and Genetics, Kumamoto University, Kumamoto, Japan
It has been a challenge in developmental biology and regenerative medicine to generate nephron progenitors that reconstitute the three-dimensional (3D) nephron structure in vitro. Many studies have tried to induce nephron progenitors from pluripotent stem cells by mimicking the developmental processes in vivo. However, the current developmental model does not precisely address the spatiotemporal origin of nephron progenitors, hampering our understanding of cell fate decisions in the kidney. Here, we present a revised model of early-stage kidney specification, suggesting distinct origins of the two major kidney components: the ureteric bud and metanephric mesenchyme. This model enables the induction of metanephric nephron progenitors from both mouse and human pluripotent stem cells. The induced cells self-organize in the presence of Wnt signaling and reconstitute 3D nephron structures including both nephric tubules with a clear lumina and glomeruli with podocytes. The engrafted kidney tissue develops vascularized glomeruli and nephric tubules, but it does not produce urine, suggesting the requirement for further maturation. Nevertheless, the generation of nephron components from human-induced pluripotent stem cells will be useful for future application in regenerative therapy and modeling of congenital kidney diseases in vitro. This review discusses the possibility of de novo organogenesis of a functional kidney from pluripotent stem cells and the future direction toward clinical applications. Kidney International advance online publication, 3 December 2014; doi:10.1038/ki.2014.358 KEYWORDS: development; intermediate mesoderm; kidney; origin; regenerative medicine; stem cell
Correspondence: Ryuichi Nishinakamura, Department of Kidney Development, Institute of Molecular Embryology and Genetics, Kumamoto University, 2-2-1 Honjo, Kumamoto 860-0811, Japan. E-mail: [email protected] Received 31 March 2014; revised 21 May 2014; accepted 22 May 2014 Kidney International
STRATEGIES FOR REGENERATIVE MEDICINE OF THE KIDNEY
Increasing number of patients are suffering from chronic kidney disease caused by diabetes, hypertension, glomerulonephritis, polycystic kidney diseases, and other disorders. However, no curative treatments have been established other than renal transplantation. Moreover, chronic kidney disease not only has a risk of mortality but also causes an economic burden. However, recent progress in developmental biology and stem cell biology is realizing the potential of regenerative medicine.1 The phrase ‘regenerative medicine’ implies several perspectives such as tissue repair and de novo organogenesis. In terms of tissue repair, one strategy involves the use of the endogenous adult stem cell population of the organ. If such cells can be expanded and then differentiated into the desired cell types, they would be a useful resource for cell therapy. However, the mammalian adult kidney is one of the typical organs that does not possess a robust regenerative capacity. In the research fields of Drosophila and zebrafish, the existence of stem cells has been shown in the adult kidneys.2,3 On the other hand, it is unclear whether the adult kidney in mammals contains such multipotent stem cells.4 Recently, intrinsic renal tubular epithelial cells, which can regenerate renal epithelium after acute kidney injury, have been identified by lineage trace experiments.5 Other studies have revealed that epithelial cells within Bowman’s capsule possess the capacity to reverse the loss of podocytes,6 and the dedifferentiation and proliferative capacity of adult podocytes.7 However, these adult cells possess limited potentials compared with the embryonic nephron progenitors, which are multipotent and give rise to all the epithelial nephron components.8,9 Another strategy for tissue repair is the in situ direct reprograming of cells into the desired cell types by introducing key transcription factors. For other organs, including the pancreas and heart, the core factors are well defined for efficient conversion to desired cell types.10,11 However, a similar strategy applied to the kidney has not generated fully competent renal cells.12 For de novo organogenesis, the availability of pluripotent stem cells is essential. Because there has been establishment and propagation of pluripotent stem cells, such as embryonic stem cells13 and induced pluripotent stem cells (iPSCs),14 the reconstitution of organs in vitro has become more realistic. Many attempts to induce defined cell types from such pluripotent stem cells have been made during the past few decades. The most common approach is recapitulation of the key signals required for lineage specification during 1
mini review
developmental processes in vivo.1 Indeed, various types of tissues, including all the three germ layer derivatives such as the heart, cartilage, liver, nerve, and retina, as well as germ cells, have been successfully generated by such strategies.1,15 However, there are no promising reports on reconstruction of three-dimensional (3D) kidney structures, partly because of the lack of knowledge concerning the developmental processes in vivo. THE NEWLY IDENTIFIED ORIGIN OF THE KIDNEY
Unlike other organs, the kidney develops three distinct primordia during embryogenesis, the pronephros, mesonephros, and metanephros, which develop from the anterior to the posterior direction. The mammalian adult kidney is derived from the embryonic metanephros located in the caudal end of the embryonic trunk, whereas the former two rudiments regress during gestation. The metanephros is composed of two major tissues, the metanephric mesenchyme (MM) and the ureteric bud (UB), the latter of which evaginates from the Wolffian duct. Although the UB gives rise to the collecting ducts and ureters, the MM contains the nephron progenitors that give rise to the majority of the kidney components including the glomeruli and nephric tubules.8,9 In chicken and axolotl, by taking advantage of the easy accessibility of their early-stage embryos, it has been clearly shown that the anteriorly located intermediate mesoderm (the so-called ‘pronephric anlagen’) migrates in the anteriorto-posterior direction and develops into the Wolffian duct.16 However, no report has shown direct evidence of the origin of the metanephric or mesonephric mesenchyme. In mammals, although the Pax2/8 þ /Lhx1 þ anterior region of the intermediate mesoderm is considered to be the corresponding cell population of the pronephric anlagen, no lineage tracing experiments have been performed to show that these cells are the origin of the Wolffian duct and UB.17 Conversely, one report showed that the cell population expressing the transcription factor Osr1, which includes the anterior intermediate mesoderm region, is the origin of the metanephros.18 However, this report did not precisely examine the preferential contribution of Osr1 þ cells from each early developmental stage toward MM or UB lineage cells in the metanephros. Nonetheless, it is widely accepted in the field of directed differentiation of the kidney in vitro that all urogenital tissues, including the MM and UB, originate from the common intermediate mesoderm (Figure 1).19–22 In contrast to these conventional models, we have recently identified the origin of the MM as a caudally located Brachyury (T) þ immature cell population in the postgastrulation stage embryo, which is distinct from that of the UB (Figure 1).23 T is a well-known gene that is initially expressed in the primitive streak, namely the precursors of all mesodermal tissues, during the gastrulation stage (embryonic day (E)6.5–7.5 in mice). Thus, for directed cell differentiation, the expression of T has been widely recognized as a transient state toward mesodermal lineage induction including both the MM and UB.19–22 However, even after 2
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
gastrulation (E8.5–12.5 in mice), T þ immature cells are maintained in the posterior end of the embryo.24 In addition, recent studies have shown that the posterior immature population, called ‘axial progenitor’, serves as the source of caudal body trunk.25,26 Our lineage trace experiment revealed that the MM—i.e., nephron progenitors—originates from the caudal T þ cell population at E8.5, whereas the precursors of the UB are excluded from this population, although it remains to be examined whether the caudal T þ population is really equivalent to the axial progenitor (Figure 1). An alternative explanation is that a subset of intermediate mesoderm cells is T þ. It is possible that the T þ population is heterogeneous, and only a subset of the population is destined for the kidney lineage. However, it may not be appropriate to call this cell population ‘intermediate mesoderm’, because no obvious anatomical boundary exists between paraxial, intermediate, and lateral plate mesoderm at the posterior end of the E8.5 embryo, where the T gene is expressed. Perhaps, the definition of ‘intermediate mesoderm’ has been too broad and ambiguous, leading to misunderstanding of kidney developmental processes. Intermediate mesoderm is defined as the anatomical region that lies between the paraxial and lateral plate mesoderm. However, it does not distinguish the anteroposterior positioning or the developmental stages (E8.5 or E9.5 in mice; Figure 1). For example, there is a clear distinction of the (T )/Osr1 þ /Pax2/8 þ / Lhx1 þ intermediate mesoderm at E8.5 and the Osr1 þ / Wt1 þ /(Pax2 )/(Six2 )/Hox11 þ intermediate mesoderm at E9.5.23 Thus, we refer to the former Osr1 þ UB precursors as ‘anterior intermediate mesoderm’. The latter, which we refer to as ‘posterior intermediate mesoderm’ in this review, is a transient population that is generated from T þ precursors at E8.5 and finally develops into the MM (containing nephron progenitors) as described below (Figures 1 and 2). In summary, the precursors of the UB transiently express T during gastrulation and give rise to the (T )/Osr1 þ / Pax2/8 þ /Lhx1 þ anterior intermediate mesoderm at the post-gastrulation stage. In contrast, the precursors of the MM are not derived from this anterior intermediate mesoderm but are maintained as a caudal T þ /Cdx2 þ /(Osr1 ) cell population until the post-gastrulation stage. Indeed, we have successfully induced metanephric nephron progenitors in vitro from the T þ and Osr1 cell population, but not the Osr1 þ cell population, in E8.5 embryos.23 INDUCTION OF NEPHRON PROGENITORS
On the basis of the conventional development model, the signals required for kidney specification have been predicted as the following processes: (1) mesoderm formation, (2) anterior (common) intermediate mesoderm formation, and (3) MM formation. Several reports have indicated the requirement of various growth factors for kidney lineage specification. A developmental biology approach revealed the requirement of activin, Bmp4, and retinoic acid signaling for the formation of the anterior intermediate mesoderm.27,28 Kidney International
mini review
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
Conventional model E3.5
E7.5 (gastrulation stage)
E8.5 (postgastrulation stage)
E9.5
Extend caudally
Blastocyst
Primitive streak
Oct3/4+ Nanog+
T+
E10.5
Wolffian duct
(Osr1–) Pax2+ Lhx1+
Intermediate mesoderm (nephrogenic cord)
Osr1+ Wt1+ Pax2+ (Six2–)
Ureteric bud
Intermediate mesoderm (pronephric anlagen) (T–) Osr1+ Pax2/8+ Lhx1+
Metanephric mesenchyme (metanephric nephron progenitor)
Osr1+ Wt1+ Pax2+ Six2+ Gdnf+ Hox11+
Metanephric mesenchyme (metanephric nephron progenitor)
Osr1+ Wt1+ Pax2+ Six2+ Gdnf+ Hox11+
Novel model E3.5
E7.5 (gastrulation stage)
E8.5 (postgastrulation stage)
E9.5
E10.5
Extend caudally
Blastocyst
Primitive streak
Oct3/4+ Nanog+
T+ Cdx2+
Anterior intermediate mesoderm (pronephric anlagen)
(T–) Osr1+ Pax2/8+ Lhx1+
Wolffian duct
(Osr1–) Pax2+ Lhx1+
Ureteric bud
Stay caudally
Posterior nascent mesoderm (axial progenitor)
T+ Cdx2+ (Osr1–)
Posterior intermediate mesoderm (nephrogenic cord)
Osr1+ Wt1+ (Pax2–) (Six2–) Hox11+
Figure 1 | Conventional and novel model for lineage segregation of the ureteric bud (UB) and the metanephric mesenchyme (MM). Conventional model shows the MM and UB originate from common intermediate mesoderm at E8.5. Novel model proposes the spatiotemporally distinct intermediate mesoderm gives rise to the MM and UB.
Another report showed the requirement of Fgf signals for development of the MM.29 Accordingly, in vitro-directed differentiation approaches using pluripotent stem cells have shown the efficacy of activin, Bmp, retinoic acid, and Wnt signaling to induce expression of anterior intermediate mesoderm marker genes.19,30,31 However, the anterior intermediate mesoderm is unlikely to be the precursor of the MM. Therefore, as discussed below, we need to reconsider the signals required for MM induction as follows: (1) nascent mesoderm formation, (2) posterior T þ precursor formation (posteriorization), (3) posterior intermediate mesoderm formation (kidney lineage specification), and (4) MM formation (maturation; Figure 2). Posteriorization
The precursors of the MM are maintained as a posterior T þ immature cell population. Thus, we need to introduce the concept of ‘axial progenitors’ to understand the posteriorization processes during kidney development. The Kidney International
importance of Wnt signaling has been shown for the formation and maintenance of caudal immature cell populations, thereby serving as the progenitor source for caudal body extension.24 Moreover, Bmp4 has a role in the induction of posterior Hox genes.32 Indeed, we have shown that a high concentration of a Wnt agonist together with Bmp4 is effective for the maintenance of T þ nascent mesoderm and the induction of posterior Hox gene expression. Interestingly, this combination of factors is also effective for the induction of nascent T þ /Cdx2 þ mesoderm from undifferentiated pluripotent stem cells (Figures 1 and 2a). In agreement with our results, two recent reports indicate the importance of early-phase Wnt signaling to confer ‘caudal identity’ to paraxial mesoderm and/or neuromesodermal progenitors.33,34 Taken together, these data suggest that the process of rostral and caudal body formation is distinct, even at the early stage of development (gastrulation stage). Thus, such information should be considered when aiming to differentiate pluripotent stem cells into a caudally located organ such as the kidney. 3
mini review
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
Nascent mesoderm induction Activin
Posteriorization
Bmp/ high Wnt
Bmp/ high Wnt
Lineage specification
Maturation
Activin/Bmp mid Wnt/RA
Fgf9/low Wnt a gfr
Oct3/4
Pd Osr1 Wt1 Hox11
T,Cdx2, Hox11
T,Cdx2
Activin Bmp/RA
RA ES/iPS
Nascent mesoderm
Posterior nascent mesoderm (axial progenitor)
Posterior intermediate mesoderm
a
gfr
Pd
Osr1 Wt1 Hox11 Pax2 Six2
a8
Itg
Metanephric mesenchyme
Figure 2 | In vitro reconstitution of 3D kidney structures. (a) Novel model of directed differentiation of pluripotent stem cells toward the metanephric mesenchyme. (b–e) Three-dimensional kidney structures from mouse ES cells and human iPS cells. Immunostaining of induced nephron progenitors for nephric tubule marker (E-cadherin; green) and glomerular markers (Nephrin; yellow, Wt1; red). (b, c) Lower magnification images of induced mouse (b) and human (c) kidney tissue. (d, e) Higher magnification images of induced mouse (d) and human (e) kidney tissue. Scale bars, 50 mm (b, c), 20 mm (d, e). ES cells, embryonic stem cells; iPS cells, induced pluripotent stem cells; RA, retinoic acid.
Kidney lineage specification
Next, we identified lineage specification factors that induce the T þ /Cdx2 þ /(Osr1 ) posterior nascent mesoderm into the Osr1 þ /Wt1 þ /Hox11 þ cell population corresponding to the posterior intermediate mesoderm at E9.5 in mice. This cocktail includes activin, Bmp4 and intermediate concentration of Wnt agonist, and retinoic acid.
factors, including activin, Bmp, and retinoic acid, in the maturation step induces nephron progenitors, indicating specific requirements for signaling at each stage (Figure 2). Importantly, our protocol induces metanephric nephron progenitors from mouse E8.5 T þ nascent mesoderm, mouse embryonic stem cells, and human iPSCs. These findings are the first evidence of conserved mechanisms for kidney specification in mice and humans.
Maturation
Finally, Fgf9 and a low concentration of Wnt agonist successfully generate the Osr1 þ /Wt1 þ /Pax2 þ /Six2 þ /Gdnf þ / Hox11 þ MM (nephron progenitors). Notably, in this step, administration of exogenous activin, Bmp4, and retinoic acid markedly inhibits the induction of metanephric nephron progenitors. In summary, the initial high concentration of Wnt agonist that is essential for posteriorization should be gradually attenuated, which allows the differentiation of the immature population similar to the conditions during axis formation in vivo.24 Moreover, removal of the lineage specification 4
RECONSTITUTION OF 3D KIDNEY STRUCTURES FROM NEPHRON PROGENITORS
During metanephros development, signals from the UB maintain and induce proliferation of nephron progenitor cells around the UB tips, as well as induce differentiation of nephron progenitors into epithelialized nephron components.28 These biphasic signals allow both expansion of the kidney and construction of the organized epithelialized nephron structures. Among these signals, the differentiation signal of Wnt9b secreted from the UB triggers mesenchymalto-epithelial transition and elevates the Wnt4 signal of Kidney International
mini review
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
the MM, resulting in well-organized nephron structures governed by intrinsic signals and local cell–cell interactions.28 Therefore, the nephron structure can be reconstituted from the MM by a substitutive source for the Wnt signaling, such as the embryonic spinal cord or Wnt4-expressing feeder cells. When the MM is cultured at the air–liquid interface with such Wnt signals, it reconstitutes 3D nephron components with nephric tubules and glomeruli.8,35 Accordingly, the cells induced by our novel method from mouse embryonic stem cells or human iPSCs readily reconstitute 3D nephron components including nephric tubules with proximal–distal segmentation, as well as glomeruli with podocytes (Figure 2). Recently, several groups have also reported the induction of kidney cells from human stem cells. Mae et al.19 showed efficient induction of an OSR1 þ cell population with a combination of activin, Bmp, and a Wnt agonist, which allows the derivation of various types of cells expressing urogenital tissue markers. However, the percentage of the kidney precursors that co-express MM markers and their capacity to generate nephron structures such as glomeruli and renal tubules are unclear. Xia et al.20 showed the induction of a UB progenitor-like population that is integrated preferentially into UB structures when reaggregated with mouse embryonic kidney tissues. However, it is unclear whether these induced UB cells possess the capacity to reconstitute the typical branched structure with the MM or without mouse UB cells. Lam et al. and Takasato et al.21,22 showed the induction of SIX2 þ MM-like cells by a PAX2/ LHX1 þ anterior intermediate mesoderm-like population. Takasato et al.21 also demonstrated the derivation of UB-like structures. However, our in vivo analysis showed that the anterior intermediate mesoderm is unlikely to be the precursor of the MM. Thus, it is unknown whether such protocols induce a genuine MM. Interestingly, such protocols partly share the same concept with ours: addition of a Wnt agonist for the initial phases, followed by a withdrawal of Wnt and the addition of Fgf. This aspect may be the reason why their protocols induce epithelialized structures to some extent but lack the generation of glomeruli. Thus, each MM and UB lineage should be induced by distinct conditions that are precisely optimized in terms of timing, concentration, and the combination of factors.
bility is the accessibility of the blood supply. Because the blood vessels integrate from host tissues, a certain period of time is required for perfusion of the grafted tissue, which may result in graft degeneration. The second possibility involves the non-physiological process of nephrogenesis. Unlike signals from the UB, the spinal cord lacks spatially distinct signaling that enables maintenance of the progenitor population, resulting in immediate differentiation and loss of nephron progenitors and limited the generation of the nephrons. The third possibility is the absence of stromal cells that have been recently identified as the third major population of the metanephric primordia.36 These cells are originally present in the MM and give rise to stromal cell types including interstitial cells, vascular smooth muscle cells, pericytes, and mesangial cells.5 Accumulating evidence shows the requirement of this population for proper development of the kidney, including UB branching and nephron differentiation.36–38 Interestingly, for liver reconstitution, concomitant culture with blood vessel cells and stromal cells generates a functional 3D liver-bud-like structure in vitro, indicating the importance of these supportive cell populations39 (Figure 3).
FUTURE DIRECTIONS FOR REGENERATIVE MEDICINE OF THE KIDNEY
There are similar problems for regenerative medicine of various organs, including methods to increase the size of 3D tissues and induce functional maturation, namely from embryonic to adult cell types. One of the key factors is the integration of well-organized blood vessels to allow circulation of blood flow. For this purpose, there has been development of various bioengineering methods.40,41 A recent study showed the possibility of using decellularized donor kidneys as a scaffold for bioengineered kidneys.42 This technique enables efficient reconstruction of both the endothelium and epithelial cells, allowing the partial function of urine production in vitro and in vivo. It may be possible to
‘Environmental ‘Proliferation’ factors?’ ‘Efficient ‘Bioengineering’ differentiation’ Reconstructed Pluripotent Kidney kidney stem cells precursors
CHALLENGES FOR RECONSTITUTION OF A FUNCTIONAL KIDNEY
Because the kidney produces urine by filtration of blood in the glomeruli, the integration of blood perfusion is essential for its function. Thus, we have co-transplanted mouse embryonic stem cell–derived nephron progenitors with an embryonic spinal cord beneath the kidney capsule of immunodeficient mice. Notably, similar to in vivo development, the podocytes in the induced glomeruli expressed blood vessel attractants, Vegfa and Ephrinb2, and we observed extensive integration of blood vessels into the glomeruli. However, they could not support the transplanted tissue long enough to produce urine, probably because of the following reasons. One possiKidney International
Ureteric bud Nephron progenitor ES iPS Stromal cell Endothelial cell
Figure 3 | Model for functional kidney reconstitution. Kidney precursors induced by individual conditions should be appropriately reconstructed to recreate the functional kidney organoid. 5
mini review
combine this technique with nephron progenitors induced from pluripotent stem cells. The efficacy of differentiation protocols and methods to expand induced progenitors in vitro are also essential aspects. Fgfs and Bmp7 are reported to be important for nephron progenitor maintenance and proliferation in vivo and in vitro. However, a robust condition has not been elucidated for long-term propagation of nephron progenitors.43 In addition, unknown factors produced by tissues surrounding kidneys might be necessary for further maturation of embryonic organs (Figure 3). In parallel with these directions, increasing attention has been paid to the applications of induced tissues for disease modeling in vitro. Using iPSCs established from patients, we would be able to recreate the diseased tissue in vitro. For example, our protocol may be applicable for reconstitution of congenital kidney anomalies, especially affecting the glomeruli. Although our current settings allow only limited maturation of induced kidney tissue, it would be beneficial to examine the early onset of such diseases, because it has been reported that 20–30% of all anomalies are identified in the prenatal period.44 However, improvement of the method to further mature induced kidney tissue is required for modeling adult diseases, which would also be useful for drug screening. CONCLUSIONS
Many obstacles remain to be overcome for reconstitution of a de novo functional kidney. However, the accumulating evidence shows the possibility that we can at least partially recapitulate in vivo developmental processes in vitro using pluripotent stem cells. Studies of in vitro reconstruction models will provide novel insights into the biological mechanisms in vivo. These mutually interactive analyses, together with bioengineering approaches, may achieve the production of a clinically usable bioartificial kidney in the near future. DISCLOSURE
All the authors declared no competing interests.
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
8.
9.
10.
11. 12.
13. 14.
15. 16.
17. 18.
19.
20.
21.
22.
23.
24. 25.
ACKNOWLEDGMENTS
26.
We thank all the members of the Nishinakamura laboratory for their contributions.
27.
REFERENCES
28.
1.
2. 3. 4.
5.
6. 7.
6
Murry CE, Keller G. Differentiation of embryonic stem cells to clinically relevant populations: lessons from embryonic development. Cell 2008; 132: 661–680. Diep CQ, Ma D, Deo RC et al. Identification of adult nephron progenitors capable of kidney regeneration in zebrafish. Nature 2011; 470: 95–100. Singh SR, Liu W, Hou SX. The adult Drosophila malpighian tubules are maintained by multipotent stem cells. Cell Stem Cell 2007; 1: 191–203. Cirio MC, de Groh ED, de Caestecker MP et al. Kidney regeneration: common themes from the embryo to the adult. Pediatr Nephrol 2014; 29: 553–564. Humphreys BD, Lin SL, Kobayashi A et al. Fate tracing reveals the pericyte and not epithelial origin of myofibroblasts in kidney fibrosis. Am J Pathol 2010; 176: 85–97. Lasagni L, Romagnani P. Glomerular epithelial stem cells: the good, the bad, and the ugly. J Am Soc Nephrol 2010; 21: 1612–1619. Shkreli M, Sarin KY, Pech MF et al. Reversible cell-cycle entry in adult kidney podocytes through regulated control of telomerase and Wnt signaling. Nat Med 2012; 18: 111–119.
29. 30.
31.
32.
33.
34.
Osafune K, Takasato M, Kispert A et al. Identification of multipotent progenitors in the embryonic mouse kidney by a novel colony-forming assay. Development 2006; 133: 151–161. Kobayashi A, Valerius MT, Mugford JW et al. Six2 defines and regulates a multipotent self-renewing nephron progenitor population throughout mammalian kidney development. Cell Stem Cell 2008; 3: 169–181. Ieda M, Fu JD, Delgado-Olguin P et al. Direct reprogramming of fibroblasts into functional cardiomyocytes by defined factors. Cell 2010; 142: 375–386. Zhou Q, Brown J, Kanarek A et al. In vivo reprogramming of adult pancreatic exocrine cells to beta-cells. Nature 2008; 455: 627–632. Hendry CE, Vanslambrouck JM, Ineson J et al. Direct transcriptional reprogramming of adult cells to embryonic nephron progenitors. J Am Soc Nephrol 2013; 24: 1424–1434. Evans MJ, Kaufman MH. Establishment in culture of pluripotential cells from mouse embryos. Nature 1981; 292: 154–156. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 2006; 126: 663–676. Williams LA, Davis-Dusenbery BN, Eggan KC. SnapShot: directed differentiation of pluripotent stem cells. Cell 2012; 149: 1174–1174, e1171. Drawbridge J, Meighan CM, Lumpkins R et al. Pronephric duct extension in amphibian embryos: migration and other mechanisms. Dev Dyn 2003; 226: 1–11. Bouchard M, Souabni A, Mandler M et al. Nephric lineage specification by Pax2 and Pax8. Genes Dev 2002; 16: 2958–2970. Mugford JW, Sipila P, McMahon JA et al. Osr1 expression demarcates a multi-potent population of intermediate mesoderm that undergoes progressive restriction to an Osr1-dependent nephron progenitor compartment within the mammalian kidney. Dev Biol 2008; 324: 88–98. Mae S, Shono A, Shiota F et al. Monitoring and robust induction of nephrogenic intermediate mesoderm from human pluripotent stem cells. Nat Commun 2013; 4: 1367. Xia Y, Nivet E, Sancho-Martinez I et al. Directed differentiation of human pluripotent cells to ureteric bud kidney progenitor-like cells. Nat Cell Biol 2013; 15: 1507–1515. Takasato M, Er PX, Becroft M et al. Directing human embryonic stem cell differentiation towards a renal lineage generates a self-organizing kidney. Nat Cell Biol 2014; 16: 118–126. Lam AQ, Freedman BS, Morizane R et al. Rapid and efficient differentiation of human pluripotent stem cells into intermediate mesoderm that forms tubules expressing kidney proximal tubular markers. J Am Soc Nephrol 2013; 25: 1211–1225. Taguchi A, Kaku Y, Ohmori T et al. Redefining the in vivo origin of metanephric nephron progenitors enables generation of complex kidney structures from pluripotent stem cells. Cell Stem Cell 2013; 14: 53–67. Wilson V, Olivera-Martinez I, Storey KG. Stem cells, signals and vertebrate body axis extension. Development 2009; 136: 1591–1604. Tzouanacou E, Wegener A, Wymeersch FJ et al. Redefining the progression of lineage segregations during mammalian embryogenesis by clonal analysis. Dev Cell 2009; 17: 365–376. Takemoto T, Uchikawa M, Yoshida M et al. Tbx6-dependent Sox2 regulation determines neural or mesodermal fate in axial stem cells. Nature 2011; 470: 394–398. Dressler GR. Advances in early kidney specification, development and patterning. Development 2009; 136: 3863–3874. Costantini F, Kopan R. Patterning a complex organ: branching morphogenesis and nephron segmentation in kidney development. Dev Cell 2010; 18: 698–712. Poladia DP, Kish K, Kutay B et al. Role of fibroblast growth factor receptors 1 and 2 in the metanephric mesenchyme. Dev Biol 2006; 291: 325–339. Kim D, Dressler GR. Nephrogenic factors promote differentiation of mouse embryonic stem cells into renal epithelia. J Am Soc Nephrol 2005; 16: 3527–3534. Moriya N, Uchiyama H, Asashima M. Induction of pronephric tubules by activin and retinoic acid in presumptive ectoderm of Xenopus laevis. Dev Growth Differ 1993; 35: 6. Lengerke C, Schmitt S, Bowman TV et al. BMP and Wnt specify hematopoietic fate by activation of the Cdx-Hox pathway. Cell Stem Cell 2008; 2: 72–82. Mendjan S, Mascetti VL, Ortmann D et al. NANOG and CDX2 pattern distinct subtypes of human mesoderm during exit from pluripotency. Cell Stem Cell 2014; 15: 310–325. Gouti M, Tsakiridis A, Wymeersch FJ et al. In vitro generation of neuromesodermal progenitors reveals distinct roles for wnt signalling in Kidney International
mini review
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
35.
36.
37.
38.
the specification of spinal cord and paraxial mesoderm identity. PLoS Biol 2014; 12: e1001937. Kispert A, Vainio S, McMahon AP. Wnt-4 is a mesenchymal signal for epithelial transformation of metanephric mesenchyme in the developing kidney. Development 1998; 125: 4225–4234. Li W, Hartwig S, Rosenblum ND. Developmental origins and functions of stromal cells in the normal and diseased mammalian kidney. Dev Dyn 2014; 243: 853–863. Mendelsohn C, Batourina E, Fung S et al. Stromal cells mediate retinoiddependent functions essential for renal development. Development 1999; 126: 1139–1148. Das A, Tanigawa S, Karner CM et al. Stromal-epithelial crosstalk regulates kidney progenitor cell differentiation. Nat Cell Biol 2013; 15: 1035–1044.
Kidney International
39.
40. 41. 42.
43. 44.
Takebe T, Sekine K, Enomura M et al. Vascularized and functional human liver from an iPSC-derived organ bud transplant. Nature 2013; 499: 481–484. Sackmann EK, Fulton AL, Beebe DJ. The present and future role of microfluidics in biomedical research. Nature 2014; 507: 181–189. Badylak SF, Weiss DJ, Caplan A et al. Engineered whole organs and complex tissues. Lancet 2012; 379: 943–952. Song JJ, Guyette JP, Gilpin SE et al. Regeneration and experimental orthotopic transplantation of a bioengineered kidney. Nat Med 2013; 19: 646–651. Barak H, Huh SH, Chen S et al. FGF9 and FGF20 maintain the stemness of nephron progenitors in mice and man. Dev Cell 2012; 22: 1191–1207. Schedl A. Renal abnormalities and their developmental origin. Nat Rev Genet 2007; 8: 791–802.
7
mini review
& 2014 International Society of Nephrology
Nephron reconstitution from pluripotent stem cells Atsuhiro Taguchi1 and Ryuichi Nishinakamura1 1
Department of Kidney Development, Institute of Molecular Embryology and Genetics, Kumamoto University, Kumamoto, Japan
It has been a challenge in developmental biology and regenerative medicine to generate nephron progenitors that reconstitute the three-dimensional (3D) nephron structure in vitro. Many studies have tried to induce nephron progenitors from pluripotent stem cells by mimicking the developmental processes in vivo. However, the current developmental model does not precisely address the spatiotemporal origin of nephron progenitors, hampering our understanding of cell fate decisions in the kidney. Here, we present a revised model of early-stage kidney specification, suggesting distinct origins of the two major kidney components: the ureteric bud and metanephric mesenchyme. This model enables the induction of metanephric nephron progenitors from both mouse and human pluripotent stem cells. The induced cells self-organize in the presence of Wnt signaling and reconstitute 3D nephron structures including both nephric tubules with a clear lumina and glomeruli with podocytes. The engrafted kidney tissue develops vascularized glomeruli and nephric tubules, but it does not produce urine, suggesting the requirement for further maturation. Nevertheless, the generation of nephron components from human-induced pluripotent stem cells will be useful for future application in regenerative therapy and modeling of congenital kidney diseases in vitro. This review discusses the possibility of de novo organogenesis of a functional kidney from pluripotent stem cells and the future direction toward clinical applications. Kidney International advance online publication, 3 December 2014; doi:10.1038/ki.2014.358 KEYWORDS: development; intermediate mesoderm; kidney; origin; regenerative medicine; stem cell
Correspondence: Ryuichi Nishinakamura, Department of Kidney Development, Institute of Molecular Embryology and Genetics, Kumamoto University, 2-2-1 Honjo, Kumamoto 860-0811, Japan. E-mail: [email protected] Received 31 March 2014; revised 21 May 2014; accepted 22 May 2014 Kidney International
STRATEGIES FOR REGENERATIVE MEDICINE OF THE KIDNEY
Increasing number of patients are suffering from chronic kidney disease caused by diabetes, hypertension, glomerulonephritis, polycystic kidney diseases, and other disorders. However, no curative treatments have been established other than renal transplantation. Moreover, chronic kidney disease not only has a risk of mortality but also causes an economic burden. However, recent progress in developmental biology and stem cell biology is realizing the potential of regenerative medicine.1 The phrase ‘regenerative medicine’ implies several perspectives such as tissue repair and de novo organogenesis. In terms of tissue repair, one strategy involves the use of the endogenous adult stem cell population of the organ. If such cells can be expanded and then differentiated into the desired cell types, they would be a useful resource for cell therapy. However, the mammalian adult kidney is one of the typical organs that does not possess a robust regenerative capacity. In the research fields of Drosophila and zebrafish, the existence of stem cells has been shown in the adult kidneys.2,3 On the other hand, it is unclear whether the adult kidney in mammals contains such multipotent stem cells.4 Recently, intrinsic renal tubular epithelial cells, which can regenerate renal epithelium after acute kidney injury, have been identified by lineage trace experiments.5 Other studies have revealed that epithelial cells within Bowman’s capsule possess the capacity to reverse the loss of podocytes,6 and the dedifferentiation and proliferative capacity of adult podocytes.7 However, these adult cells possess limited potentials compared with the embryonic nephron progenitors, which are multipotent and give rise to all the epithelial nephron components.8,9 Another strategy for tissue repair is the in situ direct reprograming of cells into the desired cell types by introducing key transcription factors. For other organs, including the pancreas and heart, the core factors are well defined for efficient conversion to desired cell types.10,11 However, a similar strategy applied to the kidney has not generated fully competent renal cells.12 For de novo organogenesis, the availability of pluripotent stem cells is essential. Because there has been establishment and propagation of pluripotent stem cells, such as embryonic stem cells13 and induced pluripotent stem cells (iPSCs),14 the reconstitution of organs in vitro has become more realistic. Many attempts to induce defined cell types from such pluripotent stem cells have been made during the past few decades. The most common approach is recapitulation of the key signals required for lineage specification during 1
mini review
developmental processes in vivo.1 Indeed, various types of tissues, including all the three germ layer derivatives such as the heart, cartilage, liver, nerve, and retina, as well as germ cells, have been successfully generated by such strategies.1,15 However, there are no promising reports on reconstruction of three-dimensional (3D) kidney structures, partly because of the lack of knowledge concerning the developmental processes in vivo. THE NEWLY IDENTIFIED ORIGIN OF THE KIDNEY
Unlike other organs, the kidney develops three distinct primordia during embryogenesis, the pronephros, mesonephros, and metanephros, which develop from the anterior to the posterior direction. The mammalian adult kidney is derived from the embryonic metanephros located in the caudal end of the embryonic trunk, whereas the former two rudiments regress during gestation. The metanephros is composed of two major tissues, the metanephric mesenchyme (MM) and the ureteric bud (UB), the latter of which evaginates from the Wolffian duct. Although the UB gives rise to the collecting ducts and ureters, the MM contains the nephron progenitors that give rise to the majority of the kidney components including the glomeruli and nephric tubules.8,9 In chicken and axolotl, by taking advantage of the easy accessibility of their early-stage embryos, it has been clearly shown that the anteriorly located intermediate mesoderm (the so-called ‘pronephric anlagen’) migrates in the anteriorto-posterior direction and develops into the Wolffian duct.16 However, no report has shown direct evidence of the origin of the metanephric or mesonephric mesenchyme. In mammals, although the Pax2/8 þ /Lhx1 þ anterior region of the intermediate mesoderm is considered to be the corresponding cell population of the pronephric anlagen, no lineage tracing experiments have been performed to show that these cells are the origin of the Wolffian duct and UB.17 Conversely, one report showed that the cell population expressing the transcription factor Osr1, which includes the anterior intermediate mesoderm region, is the origin of the metanephros.18 However, this report did not precisely examine the preferential contribution of Osr1 þ cells from each early developmental stage toward MM or UB lineage cells in the metanephros. Nonetheless, it is widely accepted in the field of directed differentiation of the kidney in vitro that all urogenital tissues, including the MM and UB, originate from the common intermediate mesoderm (Figure 1).19–22 In contrast to these conventional models, we have recently identified the origin of the MM as a caudally located Brachyury (T) þ immature cell population in the postgastrulation stage embryo, which is distinct from that of the UB (Figure 1).23 T is a well-known gene that is initially expressed in the primitive streak, namely the precursors of all mesodermal tissues, during the gastrulation stage (embryonic day (E)6.5–7.5 in mice). Thus, for directed cell differentiation, the expression of T has been widely recognized as a transient state toward mesodermal lineage induction including both the MM and UB.19–22 However, even after 2
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
gastrulation (E8.5–12.5 in mice), T þ immature cells are maintained in the posterior end of the embryo.24 In addition, recent studies have shown that the posterior immature population, called ‘axial progenitor’, serves as the source of caudal body trunk.25,26 Our lineage trace experiment revealed that the MM—i.e., nephron progenitors—originates from the caudal T þ cell population at E8.5, whereas the precursors of the UB are excluded from this population, although it remains to be examined whether the caudal T þ population is really equivalent to the axial progenitor (Figure 1). An alternative explanation is that a subset of intermediate mesoderm cells is T þ. It is possible that the T þ population is heterogeneous, and only a subset of the population is destined for the kidney lineage. However, it may not be appropriate to call this cell population ‘intermediate mesoderm’, because no obvious anatomical boundary exists between paraxial, intermediate, and lateral plate mesoderm at the posterior end of the E8.5 embryo, where the T gene is expressed. Perhaps, the definition of ‘intermediate mesoderm’ has been too broad and ambiguous, leading to misunderstanding of kidney developmental processes. Intermediate mesoderm is defined as the anatomical region that lies between the paraxial and lateral plate mesoderm. However, it does not distinguish the anteroposterior positioning or the developmental stages (E8.5 or E9.5 in mice; Figure 1). For example, there is a clear distinction of the (T )/Osr1 þ /Pax2/8 þ / Lhx1 þ intermediate mesoderm at E8.5 and the Osr1 þ / Wt1 þ /(Pax2 )/(Six2 )/Hox11 þ intermediate mesoderm at E9.5.23 Thus, we refer to the former Osr1 þ UB precursors as ‘anterior intermediate mesoderm’. The latter, which we refer to as ‘posterior intermediate mesoderm’ in this review, is a transient population that is generated from T þ precursors at E8.5 and finally develops into the MM (containing nephron progenitors) as described below (Figures 1 and 2). In summary, the precursors of the UB transiently express T during gastrulation and give rise to the (T )/Osr1 þ / Pax2/8 þ /Lhx1 þ anterior intermediate mesoderm at the post-gastrulation stage. In contrast, the precursors of the MM are not derived from this anterior intermediate mesoderm but are maintained as a caudal T þ /Cdx2 þ /(Osr1 ) cell population until the post-gastrulation stage. Indeed, we have successfully induced metanephric nephron progenitors in vitro from the T þ and Osr1 cell population, but not the Osr1 þ cell population, in E8.5 embryos.23 INDUCTION OF NEPHRON PROGENITORS
On the basis of the conventional development model, the signals required for kidney specification have been predicted as the following processes: (1) mesoderm formation, (2) anterior (common) intermediate mesoderm formation, and (3) MM formation. Several reports have indicated the requirement of various growth factors for kidney lineage specification. A developmental biology approach revealed the requirement of activin, Bmp4, and retinoic acid signaling for the formation of the anterior intermediate mesoderm.27,28 Kidney International
mini review
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
Conventional model E3.5
E7.5 (gastrulation stage)
E8.5 (postgastrulation stage)
E9.5
Extend caudally
Blastocyst
Primitive streak
Oct3/4+ Nanog+
T+
E10.5
Wolffian duct
(Osr1–) Pax2+ Lhx1+
Intermediate mesoderm (nephrogenic cord)
Osr1+ Wt1+ Pax2+ (Six2–)
Ureteric bud
Intermediate mesoderm (pronephric anlagen) (T–) Osr1+ Pax2/8+ Lhx1+
Metanephric mesenchyme (metanephric nephron progenitor)
Osr1+ Wt1+ Pax2+ Six2+ Gdnf+ Hox11+
Metanephric mesenchyme (metanephric nephron progenitor)
Osr1+ Wt1+ Pax2+ Six2+ Gdnf+ Hox11+
Novel model E3.5
E7.5 (gastrulation stage)
E8.5 (postgastrulation stage)
E9.5
E10.5
Extend caudally
Blastocyst
Primitive streak
Oct3/4+ Nanog+
T+ Cdx2+
Anterior intermediate mesoderm (pronephric anlagen)
(T–) Osr1+ Pax2/8+ Lhx1+
Wolffian duct
(Osr1–) Pax2+ Lhx1+
Ureteric bud
Stay caudally
Posterior nascent mesoderm (axial progenitor)
T+ Cdx2+ (Osr1–)
Posterior intermediate mesoderm (nephrogenic cord)
Osr1+ Wt1+ (Pax2–) (Six2–) Hox11+
Figure 1 | Conventional and novel model for lineage segregation of the ureteric bud (UB) and the metanephric mesenchyme (MM). Conventional model shows the MM and UB originate from common intermediate mesoderm at E8.5. Novel model proposes the spatiotemporally distinct intermediate mesoderm gives rise to the MM and UB.
Another report showed the requirement of Fgf signals for development of the MM.29 Accordingly, in vitro-directed differentiation approaches using pluripotent stem cells have shown the efficacy of activin, Bmp, retinoic acid, and Wnt signaling to induce expression of anterior intermediate mesoderm marker genes.19,30,31 However, the anterior intermediate mesoderm is unlikely to be the precursor of the MM. Therefore, as discussed below, we need to reconsider the signals required for MM induction as follows: (1) nascent mesoderm formation, (2) posterior T þ precursor formation (posteriorization), (3) posterior intermediate mesoderm formation (kidney lineage specification), and (4) MM formation (maturation; Figure 2). Posteriorization
The precursors of the MM are maintained as a posterior T þ immature cell population. Thus, we need to introduce the concept of ‘axial progenitors’ to understand the posteriorization processes during kidney development. The Kidney International
importance of Wnt signaling has been shown for the formation and maintenance of caudal immature cell populations, thereby serving as the progenitor source for caudal body extension.24 Moreover, Bmp4 has a role in the induction of posterior Hox genes.32 Indeed, we have shown that a high concentration of a Wnt agonist together with Bmp4 is effective for the maintenance of T þ nascent mesoderm and the induction of posterior Hox gene expression. Interestingly, this combination of factors is also effective for the induction of nascent T þ /Cdx2 þ mesoderm from undifferentiated pluripotent stem cells (Figures 1 and 2a). In agreement with our results, two recent reports indicate the importance of early-phase Wnt signaling to confer ‘caudal identity’ to paraxial mesoderm and/or neuromesodermal progenitors.33,34 Taken together, these data suggest that the process of rostral and caudal body formation is distinct, even at the early stage of development (gastrulation stage). Thus, such information should be considered when aiming to differentiate pluripotent stem cells into a caudally located organ such as the kidney. 3
mini review
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
Nascent mesoderm induction Activin
Posteriorization
Bmp/ high Wnt
Bmp/ high Wnt
Lineage specification
Maturation
Activin/Bmp mid Wnt/RA
Fgf9/low Wnt a gfr
Oct3/4
Pd Osr1 Wt1 Hox11
T,Cdx2, Hox11
T,Cdx2
Activin Bmp/RA
RA ES/iPS
Nascent mesoderm
Posterior nascent mesoderm (axial progenitor)
Posterior intermediate mesoderm
a
gfr
Pd
Osr1 Wt1 Hox11 Pax2 Six2
a8
Itg
Metanephric mesenchyme
Figure 2 | In vitro reconstitution of 3D kidney structures. (a) Novel model of directed differentiation of pluripotent stem cells toward the metanephric mesenchyme. (b–e) Three-dimensional kidney structures from mouse ES cells and human iPS cells. Immunostaining of induced nephron progenitors for nephric tubule marker (E-cadherin; green) and glomerular markers (Nephrin; yellow, Wt1; red). (b, c) Lower magnification images of induced mouse (b) and human (c) kidney tissue. (d, e) Higher magnification images of induced mouse (d) and human (e) kidney tissue. Scale bars, 50 mm (b, c), 20 mm (d, e). ES cells, embryonic stem cells; iPS cells, induced pluripotent stem cells; RA, retinoic acid.
Kidney lineage specification
Next, we identified lineage specification factors that induce the T þ /Cdx2 þ /(Osr1 ) posterior nascent mesoderm into the Osr1 þ /Wt1 þ /Hox11 þ cell population corresponding to the posterior intermediate mesoderm at E9.5 in mice. This cocktail includes activin, Bmp4 and intermediate concentration of Wnt agonist, and retinoic acid.
factors, including activin, Bmp, and retinoic acid, in the maturation step induces nephron progenitors, indicating specific requirements for signaling at each stage (Figure 2). Importantly, our protocol induces metanephric nephron progenitors from mouse E8.5 T þ nascent mesoderm, mouse embryonic stem cells, and human iPSCs. These findings are the first evidence of conserved mechanisms for kidney specification in mice and humans.
Maturation
Finally, Fgf9 and a low concentration of Wnt agonist successfully generate the Osr1 þ /Wt1 þ /Pax2 þ /Six2 þ /Gdnf þ / Hox11 þ MM (nephron progenitors). Notably, in this step, administration of exogenous activin, Bmp4, and retinoic acid markedly inhibits the induction of metanephric nephron progenitors. In summary, the initial high concentration of Wnt agonist that is essential for posteriorization should be gradually attenuated, which allows the differentiation of the immature population similar to the conditions during axis formation in vivo.24 Moreover, removal of the lineage specification 4
RECONSTITUTION OF 3D KIDNEY STRUCTURES FROM NEPHRON PROGENITORS
During metanephros development, signals from the UB maintain and induce proliferation of nephron progenitor cells around the UB tips, as well as induce differentiation of nephron progenitors into epithelialized nephron components.28 These biphasic signals allow both expansion of the kidney and construction of the organized epithelialized nephron structures. Among these signals, the differentiation signal of Wnt9b secreted from the UB triggers mesenchymalto-epithelial transition and elevates the Wnt4 signal of Kidney International
mini review
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
the MM, resulting in well-organized nephron structures governed by intrinsic signals and local cell–cell interactions.28 Therefore, the nephron structure can be reconstituted from the MM by a substitutive source for the Wnt signaling, such as the embryonic spinal cord or Wnt4-expressing feeder cells. When the MM is cultured at the air–liquid interface with such Wnt signals, it reconstitutes 3D nephron components with nephric tubules and glomeruli.8,35 Accordingly, the cells induced by our novel method from mouse embryonic stem cells or human iPSCs readily reconstitute 3D nephron components including nephric tubules with proximal–distal segmentation, as well as glomeruli with podocytes (Figure 2). Recently, several groups have also reported the induction of kidney cells from human stem cells. Mae et al.19 showed efficient induction of an OSR1 þ cell population with a combination of activin, Bmp, and a Wnt agonist, which allows the derivation of various types of cells expressing urogenital tissue markers. However, the percentage of the kidney precursors that co-express MM markers and their capacity to generate nephron structures such as glomeruli and renal tubules are unclear. Xia et al.20 showed the induction of a UB progenitor-like population that is integrated preferentially into UB structures when reaggregated with mouse embryonic kidney tissues. However, it is unclear whether these induced UB cells possess the capacity to reconstitute the typical branched structure with the MM or without mouse UB cells. Lam et al. and Takasato et al.21,22 showed the induction of SIX2 þ MM-like cells by a PAX2/ LHX1 þ anterior intermediate mesoderm-like population. Takasato et al.21 also demonstrated the derivation of UB-like structures. However, our in vivo analysis showed that the anterior intermediate mesoderm is unlikely to be the precursor of the MM. Thus, it is unknown whether such protocols induce a genuine MM. Interestingly, such protocols partly share the same concept with ours: addition of a Wnt agonist for the initial phases, followed by a withdrawal of Wnt and the addition of Fgf. This aspect may be the reason why their protocols induce epithelialized structures to some extent but lack the generation of glomeruli. Thus, each MM and UB lineage should be induced by distinct conditions that are precisely optimized in terms of timing, concentration, and the combination of factors.
bility is the accessibility of the blood supply. Because the blood vessels integrate from host tissues, a certain period of time is required for perfusion of the grafted tissue, which may result in graft degeneration. The second possibility involves the non-physiological process of nephrogenesis. Unlike signals from the UB, the spinal cord lacks spatially distinct signaling that enables maintenance of the progenitor population, resulting in immediate differentiation and loss of nephron progenitors and limited the generation of the nephrons. The third possibility is the absence of stromal cells that have been recently identified as the third major population of the metanephric primordia.36 These cells are originally present in the MM and give rise to stromal cell types including interstitial cells, vascular smooth muscle cells, pericytes, and mesangial cells.5 Accumulating evidence shows the requirement of this population for proper development of the kidney, including UB branching and nephron differentiation.36–38 Interestingly, for liver reconstitution, concomitant culture with blood vessel cells and stromal cells generates a functional 3D liver-bud-like structure in vitro, indicating the importance of these supportive cell populations39 (Figure 3).
FUTURE DIRECTIONS FOR REGENERATIVE MEDICINE OF THE KIDNEY
There are similar problems for regenerative medicine of various organs, including methods to increase the size of 3D tissues and induce functional maturation, namely from embryonic to adult cell types. One of the key factors is the integration of well-organized blood vessels to allow circulation of blood flow. For this purpose, there has been development of various bioengineering methods.40,41 A recent study showed the possibility of using decellularized donor kidneys as a scaffold for bioengineered kidneys.42 This technique enables efficient reconstruction of both the endothelium and epithelial cells, allowing the partial function of urine production in vitro and in vivo. It may be possible to
‘Environmental ‘Proliferation’ factors?’ ‘Efficient ‘Bioengineering’ differentiation’ Reconstructed Pluripotent Kidney kidney stem cells precursors
CHALLENGES FOR RECONSTITUTION OF A FUNCTIONAL KIDNEY
Because the kidney produces urine by filtration of blood in the glomeruli, the integration of blood perfusion is essential for its function. Thus, we have co-transplanted mouse embryonic stem cell–derived nephron progenitors with an embryonic spinal cord beneath the kidney capsule of immunodeficient mice. Notably, similar to in vivo development, the podocytes in the induced glomeruli expressed blood vessel attractants, Vegfa and Ephrinb2, and we observed extensive integration of blood vessels into the glomeruli. However, they could not support the transplanted tissue long enough to produce urine, probably because of the following reasons. One possiKidney International
Ureteric bud Nephron progenitor ES iPS Stromal cell Endothelial cell
Figure 3 | Model for functional kidney reconstitution. Kidney precursors induced by individual conditions should be appropriately reconstructed to recreate the functional kidney organoid. 5
mini review
combine this technique with nephron progenitors induced from pluripotent stem cells. The efficacy of differentiation protocols and methods to expand induced progenitors in vitro are also essential aspects. Fgfs and Bmp7 are reported to be important for nephron progenitor maintenance and proliferation in vivo and in vitro. However, a robust condition has not been elucidated for long-term propagation of nephron progenitors.43 In addition, unknown factors produced by tissues surrounding kidneys might be necessary for further maturation of embryonic organs (Figure 3). In parallel with these directions, increasing attention has been paid to the applications of induced tissues for disease modeling in vitro. Using iPSCs established from patients, we would be able to recreate the diseased tissue in vitro. For example, our protocol may be applicable for reconstitution of congenital kidney anomalies, especially affecting the glomeruli. Although our current settings allow only limited maturation of induced kidney tissue, it would be beneficial to examine the early onset of such diseases, because it has been reported that 20–30% of all anomalies are identified in the prenatal period.44 However, improvement of the method to further mature induced kidney tissue is required for modeling adult diseases, which would also be useful for drug screening. CONCLUSIONS
Many obstacles remain to be overcome for reconstitution of a de novo functional kidney. However, the accumulating evidence shows the possibility that we can at least partially recapitulate in vivo developmental processes in vitro using pluripotent stem cells. Studies of in vitro reconstruction models will provide novel insights into the biological mechanisms in vivo. These mutually interactive analyses, together with bioengineering approaches, may achieve the production of a clinically usable bioartificial kidney in the near future. DISCLOSURE
All the authors declared no competing interests.
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
8.
9.
10.
11. 12.
13. 14.
15. 16.
17. 18.
19.
20.
21.
22.
23.
24. 25.
ACKNOWLEDGMENTS
26.
We thank all the members of the Nishinakamura laboratory for their contributions.
27.
REFERENCES
28.
1.
2. 3. 4.
5.
6. 7.
6
Murry CE, Keller G. Differentiation of embryonic stem cells to clinically relevant populations: lessons from embryonic development. Cell 2008; 132: 661–680. Diep CQ, Ma D, Deo RC et al. Identification of adult nephron progenitors capable of kidney regeneration in zebrafish. Nature 2011; 470: 95–100. Singh SR, Liu W, Hou SX. The adult Drosophila malpighian tubules are maintained by multipotent stem cells. Cell Stem Cell 2007; 1: 191–203. Cirio MC, de Groh ED, de Caestecker MP et al. Kidney regeneration: common themes from the embryo to the adult. Pediatr Nephrol 2014; 29: 553–564. Humphreys BD, Lin SL, Kobayashi A et al. Fate tracing reveals the pericyte and not epithelial origin of myofibroblasts in kidney fibrosis. Am J Pathol 2010; 176: 85–97. Lasagni L, Romagnani P. Glomerular epithelial stem cells: the good, the bad, and the ugly. J Am Soc Nephrol 2010; 21: 1612–1619. Shkreli M, Sarin KY, Pech MF et al. Reversible cell-cycle entry in adult kidney podocytes through regulated control of telomerase and Wnt signaling. Nat Med 2012; 18: 111–119.
29. 30.
31.
32.
33.
34.
Osafune K, Takasato M, Kispert A et al. Identification of multipotent progenitors in the embryonic mouse kidney by a novel colony-forming assay. Development 2006; 133: 151–161. Kobayashi A, Valerius MT, Mugford JW et al. Six2 defines and regulates a multipotent self-renewing nephron progenitor population throughout mammalian kidney development. Cell Stem Cell 2008; 3: 169–181. Ieda M, Fu JD, Delgado-Olguin P et al. Direct reprogramming of fibroblasts into functional cardiomyocytes by defined factors. Cell 2010; 142: 375–386. Zhou Q, Brown J, Kanarek A et al. In vivo reprogramming of adult pancreatic exocrine cells to beta-cells. Nature 2008; 455: 627–632. Hendry CE, Vanslambrouck JM, Ineson J et al. Direct transcriptional reprogramming of adult cells to embryonic nephron progenitors. J Am Soc Nephrol 2013; 24: 1424–1434. Evans MJ, Kaufman MH. Establishment in culture of pluripotential cells from mouse embryos. Nature 1981; 292: 154–156. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 2006; 126: 663–676. Williams LA, Davis-Dusenbery BN, Eggan KC. SnapShot: directed differentiation of pluripotent stem cells. Cell 2012; 149: 1174–1174, e1171. Drawbridge J, Meighan CM, Lumpkins R et al. Pronephric duct extension in amphibian embryos: migration and other mechanisms. Dev Dyn 2003; 226: 1–11. Bouchard M, Souabni A, Mandler M et al. Nephric lineage specification by Pax2 and Pax8. Genes Dev 2002; 16: 2958–2970. Mugford JW, Sipila P, McMahon JA et al. Osr1 expression demarcates a multi-potent population of intermediate mesoderm that undergoes progressive restriction to an Osr1-dependent nephron progenitor compartment within the mammalian kidney. Dev Biol 2008; 324: 88–98. Mae S, Shono A, Shiota F et al. Monitoring and robust induction of nephrogenic intermediate mesoderm from human pluripotent stem cells. Nat Commun 2013; 4: 1367. Xia Y, Nivet E, Sancho-Martinez I et al. Directed differentiation of human pluripotent cells to ureteric bud kidney progenitor-like cells. Nat Cell Biol 2013; 15: 1507–1515. Takasato M, Er PX, Becroft M et al. Directing human embryonic stem cell differentiation towards a renal lineage generates a self-organizing kidney. Nat Cell Biol 2014; 16: 118–126. Lam AQ, Freedman BS, Morizane R et al. Rapid and efficient differentiation of human pluripotent stem cells into intermediate mesoderm that forms tubules expressing kidney proximal tubular markers. J Am Soc Nephrol 2013; 25: 1211–1225. Taguchi A, Kaku Y, Ohmori T et al. Redefining the in vivo origin of metanephric nephron progenitors enables generation of complex kidney structures from pluripotent stem cells. Cell Stem Cell 2013; 14: 53–67. Wilson V, Olivera-Martinez I, Storey KG. Stem cells, signals and vertebrate body axis extension. Development 2009; 136: 1591–1604. Tzouanacou E, Wegener A, Wymeersch FJ et al. Redefining the progression of lineage segregations during mammalian embryogenesis by clonal analysis. Dev Cell 2009; 17: 365–376. Takemoto T, Uchikawa M, Yoshida M et al. Tbx6-dependent Sox2 regulation determines neural or mesodermal fate in axial stem cells. Nature 2011; 470: 394–398. Dressler GR. Advances in early kidney specification, development and patterning. Development 2009; 136: 3863–3874. Costantini F, Kopan R. Patterning a complex organ: branching morphogenesis and nephron segmentation in kidney development. Dev Cell 2010; 18: 698–712. Poladia DP, Kish K, Kutay B et al. Role of fibroblast growth factor receptors 1 and 2 in the metanephric mesenchyme. Dev Biol 2006; 291: 325–339. Kim D, Dressler GR. Nephrogenic factors promote differentiation of mouse embryonic stem cells into renal epithelia. J Am Soc Nephrol 2005; 16: 3527–3534. Moriya N, Uchiyama H, Asashima M. Induction of pronephric tubules by activin and retinoic acid in presumptive ectoderm of Xenopus laevis. Dev Growth Differ 1993; 35: 6. Lengerke C, Schmitt S, Bowman TV et al. BMP and Wnt specify hematopoietic fate by activation of the Cdx-Hox pathway. Cell Stem Cell 2008; 2: 72–82. Mendjan S, Mascetti VL, Ortmann D et al. NANOG and CDX2 pattern distinct subtypes of human mesoderm during exit from pluripotency. Cell Stem Cell 2014; 15: 310–325. Gouti M, Tsakiridis A, Wymeersch FJ et al. In vitro generation of neuromesodermal progenitors reveals distinct roles for wnt signalling in Kidney International
mini review
A Taguchi and R Nishinakamura: De novo nephrogenesis in vitro
35.
36.
37.
38.
the specification of spinal cord and paraxial mesoderm identity. PLoS Biol 2014; 12: e1001937. Kispert A, Vainio S, McMahon AP. Wnt-4 is a mesenchymal signal for epithelial transformation of metanephric mesenchyme in the developing kidney. Development 1998; 125: 4225–4234. Li W, Hartwig S, Rosenblum ND. Developmental origins and functions of stromal cells in the normal and diseased mammalian kidney. Dev Dyn 2014; 243: 853–863. Mendelsohn C, Batourina E, Fung S et al. Stromal cells mediate retinoiddependent functions essential for renal development. Development 1999; 126: 1139–1148. Das A, Tanigawa S, Karner CM et al. Stromal-epithelial crosstalk regulates kidney progenitor cell differentiation. Nat Cell Biol 2013; 15: 1035–1044.
Kidney International
39.
40. 41. 42.
43. 44.
Takebe T, Sekine K, Enomura M et al. Vascularized and functional human liver from an iPSC-derived organ bud transplant. Nature 2013; 499: 481–484. Sackmann EK, Fulton AL, Beebe DJ. The present and future role of microfluidics in biomedical research. Nature 2014; 507: 181–189. Badylak SF, Weiss DJ, Caplan A et al. Engineered whole organs and complex tissues. Lancet 2012; 379: 943–952. Song JJ, Guyette JP, Gilpin SE et al. Regeneration and experimental orthotopic transplantation of a bioengineered kidney. Nat Med 2013; 19: 646–651. Barak H, Huh SH, Chen S et al. FGF9 and FGF20 maintain the stemness of nephron progenitors in mice and man. Dev Cell 2012; 22: 1191–1207. Schedl A. Renal abnormalities and their developmental origin. Nat Rev Genet 2007; 8: 791–802.
7
