The kidney papilla is a stem cells niche

Stem Cell Reviews Copyright © 2006 Humana Press Inc. All rights of any nature whatsoever are reserved. ISSN 1535–1084/06/2:181–184/$30.00 (Online) 1558–6804

The Kidney Papilla is a Stem Cells Niche Qais Al-Awqati* and Juan A. Oliver Department of Medicine, College of Physicians and Surgeons of Columbia University, 630 W 168th St, New York, NY 10032 Abstract Stem cells are characterized by low cycle time, which has allowed us to identify such cells in the mature kidney. These putative stem cells are located mostly outside the renal tubule and are concentrated in the papilla of the kidney potentially under the urinary epithelium of the papilla. Clonal analysis of these cells shows that they can differentiate into epithelial, neuronal, and other uncharacterized cells. Induction of ischemic renal failure resulted in increased proliferation of these papillary cells. Injection of these cells under the renal capsule led to their incorporation into various tubule segments. It is likely that these stem cells sense a “damage” signal from the cortex resulting in proliferation followed by migration to the site of injury. Index Entries: Kidney; stem cells; acute renal failure; renal papilla.

Introduction

*Correspondence and reprint requests to: Qais Al-Awqati, Department of Medicine, College of Physicians & Surgeons of Columbia University, 630 W 168th St, New York, NY 10032. E-mail: [email protected]

The adult mammalian kidney develops from the lateral mesoderm, which eventually produces both the Wolffian duct and the metanephric mesenchyme. An outgrowth of the Wolffian duct, the ureteric bud invades the metanephric mesenchyme and induces it to convert to the epithelial and vascular structures of the kidney. Whereas organ specific stem cells are likely to exist in the embryonic kidney, their role is complicated by the stage at which one looks for them. The lateral mesoderm is known to generate the Wolffian duct as well as the metanephric mesenchyme. Hence, there might be stem cells in this region, which could generate all lineages of the kidney. At a later stage, pluripotent cells located in the metanephric mesenchyme generate the nephron from the glomerulus till the collecting tubule, while the ureteric bud generates the collecting system. Hence, one can state that the metanephric kidney is produced by at least two lineages one mesenchymal and the other epithelial derived from the ureteric bud. We demonstrated that the metanephric mesenchyme contains stem cells capable of producing all nephron segments from the glomerulus to the distal

181

tubule (1). But whether these stem cells can also differentiate into endothelial or smooth muscle cells remained to be established. To answer such questions, we recently generated an immortalized cell line from the metanephric mesenchyme, which in vitro was capable of producing endothelial and smooth muscle cells in addition to epithelial cells (2).

Characteristics of Organ-Specific Stem Cells and Their Niche The sine qua non of stem cells is pluripotency (or for organ specific cells, multipotency), i.e., they are able to generate several differentiated cell types. In addition, they must self-generate by asymmetric division. Stem cells have been found to have a very low cycling time (3), such that when animals are injected with a “pulse” of bromo-deoxy uridine, (BrdU) or tritiated thymidine followed by a prolonged “chase” in the absence of these reagents they retain the label, whereas cells that divide more frequently lose the label as a consequence of dilution induced by recurrent (label-free) DNA synthesis during cell replication. Rare cycling allows stem cells to conserve their proliferative potential thereby minimizing the

182 __________________________________________________________________________________________Al-Awqati and Oliver errors associated with DNAreplication. Stem cells of bone marrow, brain, intestine, liver, and skin have been found to be label-retaining cells. Most stem cells exist in specific compartments in the tissues, regions that shield them from toxic or harmful environments. Such niches include the “bulge” in the hair follicle (surrounded by pigmented cells that protect them from ultraviolet light) or the bone marrow, which protects them from circulating toxins, or under the ependyma of the cerebral ventricles. A stem cell niche is a restricted environment where presumably the factors that are needed to control growth and differentiation abound allowing stem cell protection, selfrenewing capacity, and differentiation. There is now increasing evidence that supporting cells in the niche provide stem cells with important regulatory molecules. One such factor was recently shown to suppress cell division perhaps accounting for the low cycling time (4). Recent studies have suggested that the bone marrow has a much lower O2 tension than peripheral blood (5); and that hypoxia might provide an important protective milieu for stem cell self regeneration (6,7). In addition, it is now well demonstrated that proliferation of hematopoietic stem cells requires a hypoxia-mediated pathway (8). Recent studies have shown that hypoxia can maintain the undifferentiated state owing to interaction between HIF-1a and Notch signaling effectors (9). Although hypoxia might appear to be counter-intuitive as a protective environment, it is likely that hypoxic milieus have low reactive oxygen species, a major toxic metabolite. It seems that all organs have special regions where the environment is quite different from that of the rest, allowing a region that might protect stem cells from whatever toxins the organ is exposed to. As will be seen below, we believe that the renal papilla is the stem cell niche in the kidney, a region that has the most special of all environments. Not only is it hyperosmotic but also is well known to be hypoxic (10).

Stem Cells in the Adult Kidney We started our search for stem cells of the adult kidney by looking for slowly cycling cells, which retained BrdU. Their abundance increased from cortex to papilla. Remarkably they seemed to be concentrated under the epithelial lining of the pelvis, a location reminiscent of the neural stem cells, which are concentrated under the ependyma of the ventricles (11). When the papillary label-retaining cells were isolated and cultured they grew as spheres, another characteristic of neural (neurospheres; 12) and embryonic stem cells (Embryoid bodies; 13). When individual cells were cultured in clonal assays they led to the appearance of differentiated cells with an epithelial or smooth muscle phenotype. These results satisfy most of the criteria of pluripotent stem cells.

Label Retaining Cells Traditionally this is tested by injecting animals with bromodeoxy uridine (BrdU) followed by a long period of “chase” during which cells that divide will dilute the BrdU label as a result of continuous synthesis of new (label-free) DNA. We injected newborn rats and mice with BrdU and examined them after a period of 2–4 mo of chase. The kidneys were then cut sagitally and stained for BrdU using fluorescein-coupled antibodies. The sections were also counter stained with

rhodamine-coupled antibodies to collagen IV to allow us to see the borders of the structures. We had remarkable results. There was hardly any BrdU label-retaining cells in the cortex and medulla, but the papilla was full of label retaining cells (see Fig. 1). Note the lower panel, which shows a section of the papilla at low magnification. We did find a rare label-retaining cell in the blood vessel of a glomerulus. It had been demonstrated that the juxta-glomerular apparatus contains some cells that are capable of generating mesangial cells (14). In the papilla some label-retaining cells reside in the tubules but many were outside the vessels or tubules. Often, they existed in clusters. These studies were performed in both rats and mice, but mice had a much higher number of renal papillary label-retaining cells.

Subepithelial Niche One remarkable characteristic is that these cells were highly concentrated under the epithelium lining the urinary surface of the papilla in the pelvis of the kidney. The tip of the papilla is shown in this composite photograph (see Fig. 2); note the presence of the cells under the surface of the uroepithelium, a finding similar to that of neural stem cells, which are located under the ependyma of the ventricles (15).

Nephro-Spheres To further characterize these cells we prepared single cell suspensions from isolated rat papillae and developed a simple cell fractionation scheme by differential centrifugation. A fraction was enriched in BrdU positive cells where more than 30% of the cells were BrdU positive and this fraction was used for further work. Remarkably, when these cells were cultured they formed spheres especially in the absence of growth factors and serum, which seemed to favor their growth as a monolayer. It is well known that embryonic, neural, and hematopoietic stem cells grow as spheres (16). Culture in fetal calf serum resulted in a dispersal of the sphere and expression of smooth muscle proteins. Following trypsinzation we cultured individual cells in wells and allowed them to develop. We found that a single cell was able to generate progeny that stained for ZO-1 (epithelial marker) and smooth muscle (smooth muscle actin) Right figure. In addition we found that many cells derived from clones expressed the neural marker nestin and exhibited a neuronal cell shape with a long process resembling an axon.

Acute Renal Failure We used an animal model of ischemic acute renal failure to investigate whether the BrdU cell indeed participate in the regeneration of the tubular epithelium. If these cells were to sense injury and start to divide, to replace the cells injured by ischemia then they will be expected to lose the label. Preliminary studies were performed on BrdU labeled rats “chased” for three months. Ischemia of the left renal artery was induced for 45 min and the animals were allowed to recover for 3 wk. Indeed, this was the case. The question of whether these cells disappeared because of apoptosis or cell division was approached by simultaneous staining with BrdU and Ki-67, a marker of cell division, and found that the papilla contained many cells that stained for both. In addition, when we tested the papilla for apoptosis we did not find any apoptotic cells there, whereas the

Stem Cell Reviews ♦ Volume 2, 2006

The Kidney Papilla is a Stem Cells Niche __________________________________________________________________________183

Fig. 2. Composite microphotograph of the tip of a renal papilla from a 4-mo-old rat that was administered a pulse of BrdU at 3–6 d of age (50-µm sections). Three weeks before harvesting the kidney, it was exposed to 45 min of ischemia by renal artery clamping. This maneuver leads to a marked reduction of the BrdU-retaining cells, making apparent that the remaining BrdU-positive cells are most prominent under the epithelia lining the urinary surface of the papilla, next to the urinary space (US).

Fig. 1. Composite microphotograph of the kidney of a 3-mo-old mouse that was administered a pulse of BrdU at 3–6 d of age. FITC, BrdU; rhodamine, collagen IV. The four panels shown are 50-µm sections from the regions of the kidney depicted in the schematic drawing of this organ: 1, outer cortex: 2, medulla; 3, junction between the medulla and papilla; and 4, tip of the papilla. Note that, except in the papilla, there are few BrdU-retaining cells in the kidney,and that these cells appear in the transition between the medulla and papilla.

cortico-medullary region of the kidney was full of apoptotic cells, as had been found by others. These studies demonstrate that ischemic damage induces the papillary label retaining cells to divide. The important question is that given the fact that the ischemic injury is largely cortical or cortico-medullary; how does the signal travel all the way to the papilla?

Migration of Stem Cells From the Papilla to the Cortex If the cells that divide in response to injury are destined to repair the tubules in the cortex; how are they going to do it? Will they travel to the cortex and replace the cells that are injured? If so, what is the pathway? The cells will need to migrate and cross at least one basement membrane to enter the tubule. Recently we found that the site of maximum proliferation is at the junction of the papilla with the medulla where most of the newly proliferating cells are present. Clearly,

future studies will have to identify the pathway from that site to the damaged tubules, is it a vascular pathway, capillary, or lymphatic? Or do the cells migrate through the extravascular space. How do they sense the damage in the cortex? Are there gradients of chemoattractant molecules released by injured cells? As usual with new developments in any field, more questions have been opened than answers were provided.

The role of the Renal Papilla as a Stem Cell Niche in Kidney Disease Although interest in the renal papilla has long centered on its critical role in osmotic regulation, our findings shed a new light on this part of the kidney. We believe that the response of the papilla to renal injury may well hold clues to several previously mysterious events in the kidney’s response to a number of diseases. For example, many patients with sickle cell anemia develop renal disease and approx 5% will need dialysis. Although papillary abnormalities including necrosis has long been recognized as a frequent complication of this disease, the renal disease they develop includes proteinuria and glomerulosclerosis, both diseases of the renal cortex. Analgesic nephropathy has also long been known to be a primarily papillary disease because the nephrotoxic drugs accumulate in the papilla by countercurrent multiplication. Yet, these patients all develop cortical tubulo-interstitial disease. Similarly, it has been suggested, based on clinical studies, that global renal damage results from other forms of papillary necrosis. Perhaps it is the loss of papillary renal stem cells that leads to the global renal damage in these diseases because of the lack of enough cells to repopulate damaged nephrons. Further, the remarkable capacity of the kidney to recover its function after prolonged ischemia could be owing to the papilla’s resistance to the ischemic damage. The administration of nephrotoxic compounds such as

Stem Cell Reviews ♦ Volume 2, 2006

184 __________________________________________________________________________________________Al-Awqati and Oliver mercuric chloride, which induces transient acute renal failure, is associated with marked morphological changes in the renal papilla but these changes do not suggest cellular injury but rather increased cellular activity.

References 1. Herzlinger D, Koseki C, Mikawa T, al-Awqati Q. Development 1992;114:565–572. 2. Oliver JA, Barasch J, Yang J, Herzlinger D, Al-Awqati Q. Am J Physiol Renal Physiol 2002;283:F799–F809. 3. Cotsarelis G, Cheng SZ, Dong G, Sun TT, Lavker RM. Cell 1989;57:201–209. 4. Arai F, Hirao A, Ohmura M, et al. Cell 2004;118(2):149–161. 5. Harrison JS, Rameshwar P, Chang V, Bandari P. Blood 2002; 99:394. 6. Cipolleschi MG, Dello Sbarba P, Olivotto M. Blood 1993;82: 2031–2037.

7. Danet GH, Pan Y, Luongo JL, Bonnet DA, Simon MC. J Clin Invest 2003;112:126–135. 8. Adelman DM, Maltepe E, Simon MC. Genes Dev 1999;13: 2478–2483. 9. Gustafsson MV, Zheng X, Pereira T, et al. Dev Cell 2005;9: 617–628. 10. Epstein FH. Kidney Int 1997;51:381–385. 11. Morshead CM, Reynolds BA, Craig CG, et al. Neuron 1994; 13:1071–1082. 12. Xian HQ, Gottlieb DI. Trends Neurosci 2001;24:685–686. 13. Gardner RL, Brook FA. Int J Dev Biol 1997;41:235–243. 14. Hugo C, Shankland SJ, Bowen-Pope DF, Couser WG, Johnson RJ. J Clin Invest 1997;100:786–794. 15. Morshead CM, Craig CG, van der Kooy D. Development 1998;125:2251–2261. 16. Layer PG, Robitzki A, Rothermel A, Willbold E. Trends Neurosci 2002;25:131–134.

Stem Cell Reviews ♦ Volume 2, 2006