Transplantation of Human Hematopoietic Stem Cells into Ischemic and Growing Kidneys Suggests a Role in Vasculogenesis but Not Tubulogenesis



Transplantation of murine bone marrow-derived stem cells has been reported recently to promote regeneration of the injured kidney. We investigated the potential of human adult CD34+ progenitor cells to undergo renal differentiation once xenotransplanted into ischemic and developing kidneys. Immunostaining with human-specific antibodies for tubular cells (broad-spectrum cytokeratin), endothelial cells (CD31, PECAM), stromal cells (vimentin), and hematopoietic cells (pan-leukocyte CD45) demonstrated that although kidney ischemia enhanced engraftment of human cells, they were mostly hematopoietic cells (CD45+) residing in the interstitial spaces. Few other engrafted cells demonstrated an endothelial phenotype (human CD31+in morphologically appearing peritubular capillaries), but no evidence of tubular or stromal cells of human origin was found. Upregulation of SDF1 and HIF1 transcript levels in the ischemic kidneys might explain the diffuse engraftment of CD45+cells following injury. Similarly, when embryonic kidneys rudiments were co-transplanted with human CD34+cells in mice, we found both human CD45+and CD31+cells in the periphery of the developing renal grafts, whereas parenchymal elements failed to stain. In addition, human CD34+cells had no effect on kidney growth and differentiation. This first demonstration of human CD34+stem cell transplantation into injured and developing kidneys indicates that these cells do not readily acquire a tubular phenotype and are restricted mainly to hematopoietic and, to a lesser extent, to endothelial lineages. Efforts should be made to identify additional stem cell sources applicable for kidney growth and regeneration.


Regenerative medicine is focused on the development of cells, tissues, and organs for the purpose of restoring function through transplantation [1]. The general thought that replacement, repair, and restoration of function is best accomplished by cells and tissues that can perform the appropriate physiologic/metabolic duties better than any mechanical device could also be applicable to the kidney. In that regard, the use of stem cells as starting material offers new and powerful strategies for future tissue development and engineering.

Perhaps the most characterized adult stem cell is one of those residing in the adult bone marrow, that is, the hematopoietic stem cell (HSC), which gives rise to all blood cell types [2]. In addition, mesenchymal stem cells are multipotent cells that can be isolated from adult bone marrow and be induced in vitro and in vivo to differentiate into a variety of mesenchymal tissues [3]. Recently, it has been suggested that bone marrow-derived stem cells can cross boundaries and give rise to a broader array of differentiated cell types, that is, turning blood into liver, brain, pancreas, skin, intestine, and eventually kidney [49]. Indeed, several studies have shown that murine hematopoietic [10, 11] and mesenchymal [12] stem cells adopt the phenotype of renal tubular cells during acute tubular injury, whereas bone marrow-derived progenitor cells can give rise to glomerular endothelial and mesangial cells during glomerular injury [1315]. Moreover, human CD34+-enriched cell population (containing mostly HSCs, but also other hematopoietic progenitors, endothelial precursors, and potentially mesenchymal-like progenitor cells) have been shown to differentiate into hepatocytes, intestinal cells, microglia, myotubes, and cardiomyocytes following xenotransplantation in the severe combined immunodeficient (SCID) mouse model [1618]. Nevertheless, recent reports have shown little evidence for adult stem cell plasticity, bringing into question the clinical relevance of stem cell plasticity [1922]. This difficulty in reproducing HSC plasticity and the fact that the fate of adult human stem cells has not yet been evaluated in the injured kidney prompted us to test whether human CD34+ cells can participate in kidney regeneration. Moreover, we have recently established grafts of human and pig developing kidneys in immunodeficient mice [2325]. This model affords the opportunity to examine the contribution of stem cell preparations to growth and development of the kidneys in vivo [26].

Here, we xenotransplanted human peripheral blood CD34+ progenitor cells in ischemic and growing kidneys and demonstrated in both instances that although these cells can engraft and participate in renal neovascularization (possibly revealing functional hemangioblast activity), they mostly remain of hematopoietic lineage and fail to adopt a renal tubular phenotype.

Materials and Methods


Animals were maintained under conditions approved by the Institutional Animal Care and Use Committee at the Weizmann Institute. Immune-deficient nonobese diabetic (NOD)/SCID mice (Weizmann Institute Animal Breeding Center, Rehovot, Israel) were used at the age of 8–10 weeks as hosts for the transplantation studies. All mice were kept in small cages (up to five animals in each cage), fed sterile food, and given acidulated water containing ciprofloxacin.

Porcine Embryonic Kidneys

Pig embryos were obtained from the Lahav Institute of Animal Research (Kibbutz Lahav, Israel). The study protocol was approved by ethics committees both in Kibbutz Lahav and at the Weizmann Institute. Pregnant sows were operated on under general anesthesia, and E28 embryos were extracted. Warm ischemia time was <10 minutes, and the embryos were transferred in cold phosphate-buffered saline (PBS). Pig kidney precursors for transplantation were extracted under a light microscope and were kept under sterile conditions at 4°C in RPMI medium 1640 (Biological Industries, Bett HaEmek, Israel, before transplantation. Cold ischemia time until transplantation was <2 hours.

Human Peripheral Blood CD34+ Cells

Granulocyte colony-stimulating factor-mobilized peripheral blood cells from healthy adult donors for clinical transplantation were obtained after informed consent was given and were used in accordance with procedures approved by the human ethics committee of the Weizmann Institute. The samples were separated on Ficoll-Paque (Pharmacia Biotech, Uppsala, Sweden). CD34+CD133+ cells were enriched using the magnetic-activated cell sorting (MACS) cell isolation kit and the auto-MACS magnetic cell sorter (Miltenyi Biotec, Bergisch Gladbach, Germany, according to the manufacturer's instructions, obtaining purity of approximately 97% and 84% for CD34+ and CD133+ cells, respectively. Purified cells were used freshly.

Ischemia/Reflow Experiments and Cell Transplantation

Mice were anesthetized with 100 mg/kg ketamine and 10 mg/kg xylazine injected intraperitoneally, and a flank incision was made. For unilateral ischemia/reflow (I/R), the left renal pedicle was clamped for 40 minutes using a vascular clamp (Fine Science Tools Inc., Foster City, CA). The abdomen was covered with gauze moistened in PBS, and the mice were maintained at 37°C using a warming pad. After 40 minutes, the clamp was removed, and reperfusion was confirmed visually. To determine the extent of acute injury, control mice were sacrificed 24 hours after I/R, and kidneys were collected and processed for histology using H&E and sirius red staining. A volume of 50 μl of cell suspension containing 4 million CD34+ cells was injected directly into the left kidney through the renal pelvis with a 32-gauge needle immediately after removal of the vascular clamp. The needle was advanced into the renal parenchyma, and the cell suspension was slowly injected. For controls, 1) mice were subjected to an identical protocol of I/R but received PBS instead of human cells, or 2) mice received the same amounts of human cells without bearing ischemic injury.

Cotransplantation of Embryonic Kidney Tissue and Human CD34+ Cells

Implantation of pig embryonic tissue and CD34+ cells was performed under general anesthesia (2.5% 2,2,2-tribromoethanol, 97% in PBS, 10 ml/kg i.p.). Host kidney was exposed through a left lateral incision. A 1.5-mm incision was made at the caudal end of the kidney capsule, and a fragment of donor tissue (1–2 mm in diameter) was grafted concomitant with 50 μl of cell suspension containing 4 million human CD34+ cells. Control mice received fetal grafts but no CD34+ cells.

In Situ Detection of Human Cells

For immunohistochemical labeling the following non-cross-reactive antibodies were used: monoclonal mouse anti-human cytokeratin clones MNF116 (broad-spectrum cytokeratin) and BA17 (cytokeratin19), monoclonal anti-human vimentin (clones V9 and SP20), monoclonal mouse anti-human CD31 (PECAM) (clone JC/70A), monoclonal mouse anti-human CD45 (clone 2B11 + PD7/26), monoclonal mouse anti-human CD68 (clone PG-M1), and polyclonal rabbit anti-human CD3 (SP20 was purchased from Lab Vision, Fremont, CA,; all others from DAKO, Glostrup, Denmark, Table 1 depicts the specificity of the anti-human monoclonal antibodies (mAbs) that we used for detection of human cell engraftment in animal models. The antibodies (Abs) have been previously shown to be non-cross-reactive with murine (all antibodies listed) and pig (BA17, SP20, JC/70A, 2B11 + PD7/26) cells in chimeric SCID mice transplanted with human and porcine tissues [27]. Four-micrometer paraffin sections were xylen-deparaffinized and rehydrated. Endogenous peroxidase was blocked with 0.3% H2O2 in 70% methanol for 10 minutes. Antigen retrieval procedures were performed according to the manufacturer's instructions. After blocking, both paraffin sections and 6-μm cryosections were incubated with specific first antibody for 60 minutes. Detection of antibody binding was performed using the following secondary reagents: DAKO peroxidase envision system for detection of mouse and rabbit antibodies, Histofine simple stain MAX PO for rat antibodies, biotinylated anti-goat antibody (Sigma-Aldrich, St. Louis, (followed by extra avidin peroxidase reagent) for goat. In all cases, diaminobenzidine was used as chromogen.

Polymerase Chain Reaction for Human Alu Sequences

Genomic DNA was extracted with the DNeasy tissue kit (Qiagen, Hilden, Germany, from ischemic mouse kidneys and developing grafts. The polymerase chain reaction (PCR) primers were positioned in the most conserved areas of human Alu sequences and produced a PCR product of 224 bp [28]. For PCR, the following primers for Alu sequences were used: Alusense, 5′-ACG CCT GTA ATC CCA GCA CTT-3′; Alu-antisense, 5′-TCG CCC AGG CTG GAG TGC A-3′. PCR was carried out under the following conditions: 95°C for 10 minutes; 25 cycles of 95°C for 30 seconds, 58°C for 45 seconds, and 72°C for 45 seconds; and 72°C for 10 minutes. The PCR product was electrophoresed on a 2% agarose gel and stained with ethidium bromide (10 ng/ml).

Real-Time PCR

cDNA was synthesized using Omniscript reverse transcriptase (Qiagen) on total RNA. Real-time PCR of human and mouse samples was performed using an ABI7900HT sequence detection system (PerkinElmer Life Sciences, Boston,; Applied BioSystems, Foster City, CA, in the presence of SYBR-Green (SYBR Green PCR kit; Qiagen). This fluorochrome incorporates stoichiometrically into the amplification product, providing real-time quantification of double-stranded DNA PCR product. Primers were designed to amplify an 80–120-bp fragment with 50°–65°C annealing temperature. The following primers were used: mouse SDF1α'3a sense, 5′-ATGA ACGCCAAG-GTCGTGGTC-3′; antisense, 5′-GGTCTGTTGTGC TTACTT-GTTT-3′; and mouse HIF1′: sense, 5′-TCAGAGGAAGC-GAAAAATGGA-3′; antisense, 5′-CAGTCACCTGGTTGCT-GCAA-3′. For standard curve determination, we used a pool of all the samples, serially diluted in four log2 steps and run in parallel to the samples. The total volume of each reaction was 20 μl, containing 300 nM forward and 300 nM reverse primer and 125 ng of cDNA. Appropriate negative controls were run for each reaction. All of the reactions were performed in triplicate. Optimization of the real-time PCR was performed according to the manufacturer's instructions. For each analysis, transcription of the gene of interest was compared with transcription of the housekeeping gene β-actin, whose level of expression was not changed significantly according to the microarray data (data not shown) and which was amplified in parallel.


Morphological Changes in the Ischemic Kidney

We initially examined the gross appearance and histopathology of kidneys subjected to I/R injury (Fig. 1). Twenty-four hours after transient ischemia, there were gross changes (shape, size, and color) (Fig. 1A, 1B), as well as morphological changes characteristic of ischemic damage, that is, tubular cell swelling and the disappearance of nuclei in both cortex and medulla. Changes were especially evident when sirius red staining was applied (Fig. 1C, 1D).

Engraftment and Differentiation of Human CD34+ Cells in Ischemic Kidneys

The goal of these experiments was to assess the ability of CD34+ cells to engraft and differentiate in the ischemic adult kidney. Because human CD45 represents a universal marker for hematopoietic differentiation, it is useful for monitoring CD34+ cell engraftment. Nevertheless, we initially determined whether human CD45 detection correlates with a highly sensitive poly-merase chain reaction that detects human-specific Alu sequence (Alu PCR). For this, 4 × 106 CD34+ cells were injected through the renal pelvis into the renal parenchyma of intact immunodeficient mice (n = 8). In all instances where human CD45+ cells were detected at 4 weeks after injection (four of eight mice), Alu PCR was positive and vice versa, confirming that human CD45 is a reliable marker for CD34+ cell engraftment (Fig. 2). Interestingly, in the intact kidneys, CD45+ cells were mostly maintained in foci at the site of injection (Fig. 3A, 3B). Furthermore, human CD31+ cells were detected in only one of these intact kidneys. We then injected 4 × 106 human CD34+ cells into the renal parenchyma of mice immediately following the ischemic period (n = 12). At 4 weeks after injection, we identified human CD45-expressing cells in kidneys of 8 of 12 mice. In these animals, we observed widespread engraftment, demonstrated by CD45-expressing cells that were either dispersed throughout the kidney interstitium and perivascular spaces (Fig. 3C) as individual cells (Fig. 3D–3F) or as clusters of cells (Fig. 3G, 3H), both in close proximity to tubules or glomeruli but clearly distinguishable from the parenchymal structures. Thus, analysis of regions distant to the injection site in ischemic versus intact kidneys exhibited 33.2 ± 6.5 and 4.6 ± 2.4 human CD45+ cells per high-power field (HPF) (×40) (p < .001). To further delineate the nature of the CD45+ cell infiltrate, we immunostained sections for CD68 (macrophages), CD3 (T cells), and CD20 (B cells). We observed engraftment of CD68+ and CD3+ cells as individual cells (CD68) (Fig. 3I, 3J) and cell clusters (CD3) (Fig. 3K, 3J). We did not identify CD45+ cells in 4 of 12 animals, probably as a result of the technical difficulties associated with injecting into the renal parenchyma or due to death of cells early after infusion.

To determine the ability of the human HSCs to support angiogenesis, we identified human PECAM (CD31), a marker of sprouting endothelial cells, in the ischemic kidneys. We observed human CD31+ cells in 5 of the 8 kidneys expressing CD45. Although this marker is not entirely specific and positive cells resembling leukocyte subsets were found (Fig. 4A), most of these cells were detected along the renal microvasculature in morphologically appearing peritubular capillaries, especially in proximity to the area of injection (Fig. 4B–4H) (glomerular capillaries were uniformly negative). There, counts of immunoreactive cells (three consecutive high power fields [×100] per kidney in five kidneys) demonstrated an average of 12.3 ± 5.3 CD31+ cells of human origin per HPF [27]. In contrast, in all the kidney sections used in this study, we did not detect tubular or stromal/fibrocytic cells reacting with the MNF116 or V9 antibodies, respectively (data not shown).

To show that human cells expressing these tubular and stromal markers can indeed be detected following differentiation of stem cells, we stained kidneys of NOD/SCID mice that were directly injected with pluripotent human embryonic stem cells and developed intrarenal tertaoma (positive control staining). Both human MNF116+ epithelial and V9+ stromal cells were clearly identified in these sections (Fig. 5A, 5B).

Control kidney sections obtained from mice subjected to I/R and PBS injection (n = 4) were uniformly negative for all human antibodies. Thus, ischemic injury promotes engraftment of CD34+ cells and neovascularization by human cells.

HIF1 and SDF1 Upregulation in Ischemic Kidneys

Following the observation that human CD34+ cells show higher rates of engraftment, as well as a diffuse engraftment pattern in ischemic kidneys, we determined transcript levels of both HIF1 and SDF1, previously shown to promote migration of progenitor cells including human CD34+ cells [16, 29]. Real-time PCR of ischemic kidneys, kidneys contralateral to ischemia, and sham-operated intact kidneys was performed at consecutive time points after injury (Fig. 6). SDF1 and HIF1 mRNA levels were mostly significantly elevated in ischemic compared to sham kidneys up to 2 (SDF1) and 4 weeks (HIF1) after injury (for HIF1, the 1 week time point was insignificant). In some time points, SDF1 and HIF1 were also significantly induced in contralateral compared to sham kidneys (Fig. 6). Interestingly, differences in SDF1 and HIF1 levels between ischemic and contralateral kidneys were not significant, indicating remote effects of kidney injury. Thus, rapid and prolonged induction of SDF1 and HIF1 mRNA might underlie the enhanced engraftment of human CD34+ cells in the ischemic kidneys.

Engraftment and Differentiation of Human CD34+ Cells in Growing Kidneys

To investigate whether human CD34+ cells have the capacity to participate in kidney growth and development, we transplanted 4 × 106 cells into the renal subcapsular space concomitant with implantation of E28 pig kidney precursors (n = 6). Although developing pig kidney grafts established in immunodeficient mice without CD34+ cells administration were negative for all human markers (n = 5) 4 weeks after grafting, four out of six of those co-transplanted with human CD34+ cells showed positive staining for human CD45. Interestingly, human CD45+ cells were mostly concentrated in the periphery of the developing grafts as cell clusters (Fig. 7A, 7B) or scattered cells (Fig. 7C, 7D), whereas few individual cells engrafted in the transplant in proximity to immature tubules and glomeruli (Fig. 7E, 7F). Similarly, in three out of four of these grafts, we found a small number of human CD31+ cells appearing as elongated cells (possibly angioblasts) in the periphery of the graft (Fig. 7G) and also in peripheral vessels (Fig. 7H–7J). Although rare (1–2 cells or cell clusters per HPF showed human CD31 expression in multiple fields), these cells were not present in control specimens (data not shown). We failed to detect human CD31 in intragraft peritubular or glomerular vasculature. Furthermore, all of the developing grafts analyzed were negative for human CK19, whereas Vsp20 stained only CD31-expressing cells, indicating lack of differentiation into tubular or stromal/fibrocytic components of the developing kidney. Here, too, immunostaining of teratoma induced by injection of human embryonic stem cells into the kidney subcapsular space demonstrated CK19+ and Vsp20+ cells in tubular structures and stromal cells, indicating epithelial and stromal differentiation (Fig. 5C, 5D). In addition, the detection of human CD31 in vascular structures (Fig. 5E, 5F), reaffirmed the use of this marker for endothelial differentiation. Finally, comparison of chimeric grafts (developing kidney and CD34+ cells) with those grown without human CD34+ cells showed no differences in both growth and differentiation into glomeruli and tubules (Table 2).


The data presented here collectively suggest that transplantation of human CD34+ cells into the injured or developing kidney does not result in de novo tubulogenesis. No cells expressing broad human epithelial markers were observed after injection of adult stem cells into diseased/growing kidneys, indicating that transplanted cells had not differentiated into renal tubular cells. Rather, even 4 weeks after infusion into ischemic and developing kidneys, implanted CD34+ cells remained mainly consistent with hematopoietic cells expressing the common leukocyte marker CD45. This finding is consistent with recent data showing mostly hematopoietic fates for HSCs in the infracted myocard [21, 22], all supporting the restricted developmental plasticity of HSCs.

Importantly, the presence of cells with distinct mature morphology expressing human epithelial (MNF116, CK19), stromal (V9, Vsp20), and endothelial markers (CD31) following differentiation of hESC in vivo (control experiment), as well as the finding of renal vasculature expressing human CD31 after transplantation of CD34+ cells suggests that differentiation/transdifferentiation of the injected human cells can be detected using the methods employed here. Similarly, hepatic differentiation was demonstrated within the liver of NOD/SCID mice transplanted with human CD34+ stem cells using a human albumin-specific mAb [16]. In addition, Belicchi et al. [30] used human-specific mAbs for neuronal markers to show evidence for glial differentiation after the injection of human skin-derived stem cells into the ventricular space of adult SCID brain.

We observed improved engraftment of human cells in the injured kidneys, a process likely to be dependent on local signals induced by ischemia, including SDF1 and HIF1, which in turn affect CD34+ cell migration and function [16, 29]. Even though the human CD34+ cells were highly purified (97%), we cannot exclude the possibility that cell contaminants delivered in the initial preparation represent, at least in part, engrafted cells. Indeed, human CD3+ cells detected in the ischemic kidneys are unlikely to have differentiated from CD34+ progenitor cells and might have been directly transferred. Recently, it has been shown that single HSCs generate skeletal muscle through myeloid intermediates and that HSC-derived hepatocytes are primarily derived from mature myelomonocytic cells [31, 32]. Thus, human CD68+ cells detected in the ischemic kidney, which have the capacity to promote tissue regeneration, are more likely to have been derived from CD34+ progenitor cells.

Our data are the first to show that human CD34+CD133+ cells differentiate to form intrarenal blood vessels in the presence of ischemic kidney injury. These capillaries could only have derived from the progenitor fraction delivered and not from contaminating mature cells. Nevertheless, this contribution to kidney endothelium is at low levels. In the embryo, the mesodermal precursor cell, the hemangioblast, gives rise to blood and blood vessels [33]. Similarly, during adult life, it seems that HSCs also exhibit bipotential hemangioblast activity, serving as a source for circulating endothelial progenitor cells, especially when enriched for the CD133 marker [34]. Indeed, previous reports have demonstrated the vasculogenic potential of murine and human HSCs and CD34+ cells in other model systems, such as heart infarct or retinal neovascularization [3538]. In the kidney, Rookmaaker et al. induced murine experimental glomerulonephritis and showed that whole murine bone marrow-derived cells (not fractioned into stem cell subtypes) participate in endothelial repair in injured glomeruli [13]. We extend these findings by showing that in the bone marrow a purified population of CD34+CD133+ cells, is endowed with vasculogenic potential during kidney injury. In contrast with the findings of Rookmaaker et al. [13], we detected human CD31 exclusively in peritubular and not glomerular vasculature. Because glomerular damage is only minor in the ischemia/reperfusion model, such differences are possibly related to the different types of kidney injury induced.

Interestingly, the “growing kidney” model, even though employed with different species (human/pig), showed similar findings of restricted hematopoietic and, to a lesser extent, vascular differentiation after human CD34+ cell administration. Nevertheless, this vasculogenic potential was more prominent following ischemic kidney injury than that observed in the developing kidney, where human cells, especially those positive for CD31, failed to integrate with the bulky developing mass and were observed in the periphery of the grafts, surrounding the growing kidneys. In addition, CD34+ cells had no effect on graft growth and differentiation. Organogenesis of complex tissues, such as the kidney, requires a coordinated sequential transformation process, with individual stages involving time-dependent expression of cell-cell, cell-matrix, and cell-signal interactions in three dimensions. Precursor tissues, such as the E28 embryonic pig kidney, are composed of functionally diverse inherent stem/progenitor cell types (epithelial, stromal, and endothelial) that are organized in spatially complex arrangements. We have shown that at this developmental stage, the theme of temporal-spatial patterning of progenitor cell interactions is programmed in the kidney precursor, leading to its growth and differentiation after transplantation [26, 39]. Accordingly, exogenous stem cell preparations, such as CD34+ progenitor cells, show little contribution to graft development.

Our results are in contrast with the work of Kale et al. and Lin et al. [10, 11], who reported extensive tubular regeneration after transplantation of Lin c-kit+ mouse HSCs into mice subjected to renal ischemia. Several explanations related to differences in the protocols for stem cell isolation, transplantation, and detection might account for this discrepancy. First, we employed human rather than mouse cells. Differences in progenitor components in these stem cell preparations might contribute to the different outcomes. Thus, cells in blood or bone marrow that do not have the CD34+CD133+ cell phenotype might be indeed capable of renal tubular differentiation. Second, whereas mouse stem cells were transplanted systemically prior to injury to establish chimerism, we injected our preparation directly into the kidney after injury, so as to mimic the clinical setting more closely. It is therefore probable that either priming or differentiation in bone marrow is a prerequisite for human CD34+ cells to undergo renal tubular differentiation after they recirculate to the kidney. Third, different assays were used to detect tubular differentiation. The studies by Kale et al. and Lin et al. [10, 11] relied on X-gal staining to track cell fate and to monitor cell differentiation after HSC transplantation. This approach requires the establishment of a signal threshold, above which cells are designated as positive for β-galactosidase. Establishing such a threshold in renal tubules that express high levels of endogenous β-galactosidase might be rather difficult and staining might be therefore hard to interpret. In contrast, the creation of human/mouse chimeras in the ischemia model afforded us the opportunity to use highly specific human Abs that do not cross react with mouse tissue and therefore do not produce background staining and preclude contamination by mouse cells.

The data presented here did not address the potential beneficial effects of stem cell injection on kidney function after ischemic injury or when attempting to grow kidney rudiments in vivo, regardless of whether cells undergo overt tubular differentiation. Fraidenraich et al. [40] have recently demonstrated an embryonic stem cell-based rescue of a lethal cardiac defect in vivo via defined secreted factors. In addition, as recently shown for adult mouse skeletal muscle stem cells [41], human HSCs may exert functional benefits via engraftment into renal microvasculature and angiogenesis, which in turn can salvage tubular cells. In any event, the failure of HSCs to contribute to the formation of new tubules in the present study underscores the need for identifying additional stem cell sources applicable for kidney regeneration.


The authors indicate no potential conflicts of interest.

Table Table 1.. Primary monoclonal antibodies used to detect species-specific cellular antigens
original image
Table Table 2.. Graft growth and differentiation following subcapsular transplantation of E28 pig kidney precursors ± HSC
original image
Figure Figure 1..

Ischemic kidney injury. Twenty-four hours after transient ischemia, macroscopic (ischemia [A] and control [B]) and microscopic (ischemia [C] and control [D]) changes were noted. For histology, sirius red staining was performed. Magnification ×40.

Figure Figure 2..

Polymerase chain reaction for the detection of human-specific Alu sequences after injection of human hematopoietic stem cells into kidneys of nonobese diabetic/severe combined immunodeficient mice. There is a single discrete band (210 bp) in four of eight kidneys, indicating positive amplification for human Alu. These kidneys were all engrafted with human CD45+ cells, whereas kidneys negative for Alu lacked human CD45+ cells.

Figure Figure 3..

Transplantation of human CD34+ into adult kidneys. Immunostaining for human CD45 (A–H), CD68 (I, J), and CD3 (K, L). (A, B): Intact kidneys of nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice transplanted with human CD34+CD133+ cells. Maintenance of human CD45+ cells, primarily at the injection site, is shown at low magnification (×4) (A) and higher magnification (×40) (B). (C–H): Ischemically injured kidneys of NOD/SCID mice transplanted with human CD34+CD133+ cells. (C): Human CD45+ hematopoietic cells diffusely engraft 1 month after transplantation (magnification ×4). Individual cell engraftment of human CD45+ hematopoietic cells in renal interstitium is shown at a mangification of ×40 (D, E). (F): Larger image (original magnification ×40) showing maintenance in the perivascular space. (G, H): Engraftment of clusters of human CD45+ hematopoietic cells (×40). In all instances, cells are in close proximity to renal tubules and glomeruli but can be clearly distinguished from both. (I, J): Individual human CD68+ cells engraft in the interstitial spaces (I, magnification ×40), appear as elongated cells (I, inset), and are in close relation to vessels (J, magnification × 100). (K, L): Human CD3+ cells appear in clusters (I, magnification × 100) as round cells (K, original magnification ×100, arrows).

Figure Figure 4..

Ischemically injured kidneys of nonobese diabetic/severe combined immunodeficient mice transplanted with human CD34+CD133+ cells. Immunostaining for human CD31. Positive staining was detected in individual cells possibly of hematopoietic origin (A, magnification ×40) and in morphologically appearing peritubular capillaries (B, original magnification ×100, arrowheads). (C–F): Three different fields of human CD31-expressing vessels (C, magnification ×100; D, a corresponding larger image, original magnification ×100; E, magnification ×100; F, a corresponding larger image, original magnification ×100). (D, F): Arrowheads within lumens denote positive staining. (G, H): Continuous vascular structures near injection site (G, magnification ×100; H, a larger image, original magnification ×100); in both G and H, lumen is less apparent.

Figure Figure 5..

In vivo differentiation of human embryonic stem cells (clone H19) into morphologically distinct cells expressing epithelial and stromal markers. Detection of human MNF116+(A) and V9+(B) cells in teratoma growing from the renal pelvis after direct injection of hES cells into adult kidneys of nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice. Detection of human CK19+(C) Vsp20+(D) and CD31+(E, F) cells in teratoma induced by injection of hES cells into the renal subcapsular space of NOD/SCID mice. E, magnification × 10; all other images, original magnification ×40.

Figure Figure 6..

SDF1 and HIF1 gene expression following ischemia/reperfusion injury to murine kidneys. Ischemic kidneys (IR, black bars), kidney contralateral to ischemia (CL, gray bars), and sham-operated intact (S, white bars) were analyzed by real-time polymerase chain reaction at consecutive time points after ischemia. Three mice were analyzed for each time point. Data were calculated as average ± SD. *, p < .05 compared to sham.

Figure Figure 7..

E28 pig embryonic kidney co-transplanted with human CD34+CD133+ cells under the kidney capsule of nonobese diabetic/severe combined immunodeficient mice. (A–F): Immunostaining for human CD45. (E–H): Immunostaining for CD31. Human CD45+ cells are present in the periphery of the developing graft as cell clusters (A and B, magnification ×10 and ×40, respectively) or individual cells (C and D, magnification ×40). Few human CD45+ cells can be traced inside the graft in proximity to immature tubules (E) and glomerulus (F) (original magnifications ×40). Human CD31+ cells engraft in the periphery of the growing transplant as elongated cells (G) and into different small-size capillaries (H–J) (all original magnifications ×40). Abbreviations: g, graft; m, mouse kidney; p, periphery of the graft (appearing as a loose connective tissue).