By continuing to browse this site you agree to us using cookies as described in About Cookies
Wiley Online Library is migrating to a new platform powered by Atypon, the leading provider of scholarly publishing platforms. The new Wiley Online Library will be migrated over the weekend of March 17 & 18. You should not experience any issues or loss of access during this time. For more information, please visit our migration page: http://www.wileyactual.com/WOLMigration/
Author contributions: A.R.C.: conception and design, collection and/or assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript; X.-Y.Z.: conception and design, collection and/or assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript; J.D.K.: collection and/or assembly of data, data analysis and interpretation; K.L.J.: collection and/or assembly of data, data analysis and interpretation; S.C.T.: data analysis and interpretation, manuscript writing; J.P.G.: data analysis and interpretation, collection and/or assembly of data. A.L.: data analysis and interpretation, manuscript writing; L.O.L.: conception and design, collection and/or assembly of data, data analysis and interpretation, financial support, manuscript writing, final approval of manuscript. A.R.C. and X.-Y.Z. contributed equally to this article.
Disclosure of potential conflicts of interest is found at the end of this article.
First published online in STEM CELLS EXPRESS April 7, 2010.
Tissue injury triggers reparative processes that often involve endothelial progenitor cells (EPCs) recruitment. We hypothesized that atherosclerotic renal artery stenosis (ARAS) activates homing signals that would be detectable in both the kidney and EPCs, and attenuated on renal repair using selective cell-based therapy. Pigs were treated with intrarenal autologous EPC after 6 weeks of ARAS. Four weeks later, expression of homing-related signals in EPC and kidney, single kidney function, microvascular (MV) density, and morphology were compared with untreated ARAS and normal control pigs (n = 7 each). Compared with normal EPC, EPC from ARAS pigs showed increased stromal cell-derived factor (SDF)-1, angiopoietin-1, Tie-2, and c-kit expression, but downregulation of erythropoietin (EPO) and its receptor. The ARAS kidney released the c-kit-ligand stem cell factor, uric acid, and EPO, and upregulated integrin β2, suggesting activation of corresponding homing signaling. However, angiopoietin-1 and SDF-1/CXCR4 were not elevated. Administration of EPC into the stenotic kidney restored angiogenic activity, improved MV density, renal hemodynamics and function, decreased fibrosis and oxidative stress, and attenuated endogenous injury signals. The ARAS kidney releases specific homing signals corresponding to cognate receptors expressed by EPC. EPC show plasticity for organ-specific recruitment strategies, which are upregulated in early atherosclerosis. EPC are renoprotective as they attenuated renal dysfunction and damage in chronic ARAS, and consequently decreased the injury signals. Importantly, manipulation of homing signals may potentially allow therapeutic opportunities to increase endogenous EPC recruitment. STEM Cells 2010;28:1039–1047
The increasing incidence and prevalence of hypertension, diabetes, obesity, dyslipidemia, and metabolic syndrome is reflected in the increase in the number of patients with cardiovascular disease. Coexistence of these cardiovascular risk factors also amplifies the number of insults that the kidneys are constantly exposed to. By triggering injurious mechanisms such as inflammation, oxidative stress, fibrosis, and ischemia, renal function may progressively deteriorate and advance to irreversible chronic kidney disease.
One of the mechanisms by which cardiovascular risk factors may ultimately impair renal function is by inducing atherosclerosis, the predominant etiology of renal artery stenosis (RAS) , and thereby compromising blood supply. Even before obstruction by atherosclerotic plaques, the atherogenic process can harm renal tissue by promoting functional and structural deterioration of small vessels and the renal parenchyma [2, 3]. Atherosclerotic renovascular disease is on the rise as a cause of end-stage renal disease, specially in the older population . Simply restoring blood flow does not regularly restore function. There is a pressing need for improved strategies to arrest or reverse intrarenal injury in this disease.
Endogenous endothelial progenitor cells (EPC) are often mobilized to mediate neovascularization and endothelial replacement that contribute to healing ischemic tissues. The mobilization from bone marrow and subsequent homing of progenitor cells can be regulated by a variety of mediators such as stromal cell-derived factor (SDF)-1, stem cell factor (SCF), erythropoietin (EPO), or angiopoietins [5, 6], which are released by injured tissue to attract the cells and ensure their adherence. In turn, the cells express corresponding cognate receptors such as CXCR4, c-kit, erythropoietin receptors (EPOr), and Tie-2, respectively, which allow them to be recognized, recruited, and retained at the injured tissues.
However, the endogenous system may be overwhelmed or dysfunctional, and hence fail to repair the tissues. Therefore, exogenous delivery of EPC collected and expanded in vitro offers the potential for targeted treatment of conditions such as myocardial  and hind-limb ischemia , acute renal injury , and glomerulonephritis . We have recently shown the beneficial effects of intrarenal administration of autologous EPC in a porcine model of chronic nonatherosclerotic RAS . Conceivably, a decrease in tissue damage may resolve the injury signals and homing cues that it releases.
Specific signals that portend chronic ischemic injury and regulate the homing and adherence of endogenous circulating cells into the ischemic kidney, or the ability of successful renal repair to alleviate these signals, have not been elucidated. Therefore, the current study was designed to test the hypotheses that, first, renovascular disease activates homing signals detectable in both the ischemic kidney and EPC, and second, that these signals are attenuated on renal repair using selective intrarenal cell-based therapy. For this purpose, we used a pig model of experimental atherosclerotic renal artery stenosis (ARAS), which recapitulates many characteristics of early human atherosclerotic renovascular disease .
MATERIALS AND METHODS
All procedures were approved by Mayo Clinic Institutional Animal Care and Use Committee. Domestic pigs (35–40 kg) were fed with 2% cholesterol diet for 6 weeks, to induce pre-existing early atherosclerosis. The animals were then anesthetized with 0.5 g of intramuscular ketamine and xylazine, and maintained with a mixture of ketamine (0.2 mg/kg/minute) and xylazine (0.03 mg/kg/min), a local-irritant coil was implanted in the main renal artery to induce RAS, and the high-cholesterol diet continued. Six weeks after induction of RAS, animals were randomized into two groups: one was sham treated (ARAS, n = 7) and the other received an intrarenal infusion of autologous EPC (ARAS + EPC, n = 7). Additional seven pigs were used as normal controls. Four weeks later, renal hemodynamics and function were assessed in all pigs by multidetector computed tomography (MDCT), as previously described [11, 12]. Mean arterial pressure (MAP) was determined via a carotid artery catheter and the degree of stenosis by renal angiography. Blood samples were collected from a systemic and stenotic renal vein for measurement of creatinine, the EPC homing signal SCF, and the renal injury signal uric acid .
Three days after completion of in vivo studies, pigs were euthanized (sodium pentobarbital 100 mg/kg, Fort Dodge Laboratories, Fort Dodge, IA, http://www.fortdodge.eu). Kidneys were removed and lobes were either shock-frozen in liquid nitrogen and stored at −80°C, preserved in formalin, or prepared for micro-computed tomography (CT) to assess renal microvascular (MV) architecture. In vitro studies were then performed to assess renal inflammation, redox status, and expression of angiogenic and fibrogenic factors. EPC homing signals were examined in both the stenotic kidneys and isolated EPC using immunostaining or Western blotting for SDF-1, angiopoietin-1, EPO and their receptors CXCR4, Tie-2, and EPOr, as well as the SCF receptor c-Kit, and integrin β2, which mediates the adherence of leukocytes and EPCs to endothelial cell monolayers.
EPC Characterization, Preparation, and Delivery
EPC were cultured from mononuclear cells obtained from peripheral blood (100 ml), as described previously . Briefly, mononuclear cells were isolated from the blood, cultured for 1–3 weeks in endothelial media, and examined for the number of colony forming units (CFU), EPC function, phenotype, and progenitor markers . Cells were then labeled with a fluorescent membrane dye (CM-DiI) and kept in phosphate-buffered saline on ice until delivery (10 ml, 106 cells per milliliter) into the stenotic renal artery after 6 weeks of ARAS. Retention rate of the injected EPC in the kidney was calculated 4 weeks after administration, as described previously .
To evaluate homing signals, EPC from normal and ARAS pigs were stained using immunofluorescence with antibodies against SDF-1, CXCR-4, and angiopoietin-1 (all 1:50, Abcam, Cambridge, U.K., http://www.abcam.com), Tie-2 (1:50, Calbiochem, San Diego, CA, http://www.emdbiosciences.com), EPO and EPOr (1:50, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, http://www.scbt.com), c-kit (1:100, Abcam), and integrin β2 (1:50, Thermo Scientific, Rockford, IL, http://www.thermo.com). Surface receptor expression was confirmed by Western blotting, using the same antibodies used for immunofluorescence for CXCR-4, Tie-2, c-kit (all 1:500), and EPOr (1:200). Immunofluorescence and Western blotting were also repeated on 1-week older EPC (next passage).
A saline-filled cannula was ligated in a segmental artery perfusing the intact end of the stenotic kidney. Initially saline (containing heparin) was perfused under physiological perfusion pressure, and replaced with infusion (0.8 ml/minute) of an intravascular contrast agent (Microfil MV 122, Flow Tech, Carver, MA, http://www.flowtech-inc.com). Then, a lobe of the contrast-filled tissue was trimmed from the kidney, prepared, and scanned at 0.5° increments and cubic voxels of 18 μm on a side using micro-CT, as previously described [14–17].
Histology and Immunohistochemistry
Trichrome staining was performed in 5-μm paraffin midhilar renal cross-sections to assess fibrosis. Immunostaining in 5-μm frozen renal cross-sections examine the expression of integrin β2 (1:50, Thermo Scientific). Renal oxidative stress was assessed by the in situ production of superoxide anion detected using dihydroethidium (DHE) . To investigate the engraftment of EPC in endothelial or tubular structures, double immunofluorescence of the CM-DiI label and CD31 (1:50, AbD Serotec, Raleigh, NC) or cytokeratin (1:100, AbD Serotec), respectively, was performed on frozen slices, as described previously .
Standard blotting protocols were followed, as previously described , using specific antibodies against vascular endothelial growth factor (VEGF), angiopoietin-1, transforming growth factor (TGF)-β, matrix-metalloproteinase (MMP)-2, tissue-inhibitor of MMP (TIMP)-1, the NAD(P)H oxidase subunit p47phox, the VEGF receptor KDR (Santa Cruz, 1:200 for all), xanthine oxidase (1:10,000, nonreducing, no milk condition, Chemicon International, Temecula, CA, http://www.chemicon.com), SDF-1, angiopoietin-1 (1:1,000, Abcam), EPO, and EPOr (Santa Cruz, 1:200). Glyceraldehyde 3-phosphate dehydrogenase (1:5,000, Covance, Emeryville, CA, http://www.covance.com) served as loading control. Protein expression was determined in each kidney, and the intensities of the protein bands (one per animal) quantified and averaged in each group. For SCF, renal vein plasma proteins were enriched through albumin removal (ProMax, Polysciences, Warrington, PA, http://www.polysciences.com) following manufacture's instruction, and then standard Western blotting was performed with Ponceau-S staining as loading control .
Tissue attenuation was sampled in regions of interest selected in MDCT images in the aorta, renal cortex, and medulla. Time-attenuation curves were fitted with extended γ-variate curve fits  and analyzed for renal regional perfusion (ml/minute/g tissue), single kidney glomerular filtration rate (GFR), and renal blood flow (RBF), using previously validated methods [3, 12, 21-24].
Images were digitized for reconstruction of 3D volumes, and analyzed with the Analyze software package (Biomedical Imaging Resource, Mayo Clinic, Rochester, MN, http://www.mayo.edu.bir). The cortex was tomographically divided into inner, middle, and outer thirds , and the spatial density of cortical microvessels (diameters < 500 μm) calculated in each level [14, 15].
Midhilar 5-μm cross-sections of each kidney (one per animal) were examined using a computer-aided image analysis program (MetaMorph, Meta Imaging, Molecular Devices, Sunnyvale, CA). In each slide, trichrome staining or DHE fluorescence was semiautomatically quantified in 15–20 fields by the computer program, expressed as fraction of kidney surface area, and the results from all fields averaged [3, 17, 18, 23, 24]. For EPC immunofluorescence, six fields per sample of three pigs from each group were used positive and nuclei area was traced using MetaMorph, and expressed as ratio of positive/total nuclei.
Results are expressed as mean ± SEM. Comparisons within groups were performed using paired Student's t test, and among groups using analysis of variance, with the Bonferroni correction for multiple comparisons, followed by unpaired Student's t test. Statistical significance was accepted for p ≤ .05.
MAP and the angiographic degree of stenosis were similarly greater in ARAS and ARAS + EPC animals compared with normal controls (Table 1). ARAS pigs showed a significant increase in renal vein levels of uric acid compared with normal controls (1.78 ± 0.1 vs. 1.27 ± 0.1 mg/dl, p < .01), which was reduced after EPC treatment (1.5 ± 0.1 mg/dl, p = NS vs. normal, p = .05 vs. ARAS).
Table 1. Mean arterial pressure, degree of stenosis, and basal single kidney hemodynamics and function (mean ± SEM), in normal, ARAS, and ARAS pigs treated with intrarenal autologous EPCs (ARAS + EPC)
Characterization of the EPC
The number of CFU was similar in normal and ARAS pigs (61.2 ± 5.3 and 58.9 ± 4.1 CFU/cm2, respectively). As we have previously shown , both CD14 and CD133 were initially expressed in early EPC, but after 21 days the expression of KDR and CD34 increased as CD14 and CD133 diminished in late cells (data not shown), suggesting acquisition of endothelial characteristics.
EPC and Renal Homing and Angiogenic Signaling
EPC obtained from ARAS animals showed greater expression of SDF-1, angiopoietin-1, its receptor Tie-2, and c-kit than EPC obtained from normal pigs (Fig. 1A, 1B), indicating angiogenic and homing activity. However, CXCR-4 remained unchanged, and the expression of EPO and EPOr in ARAS EPC was blunted (Fig. 1A, 1B). Western blotting confirmed these trends of surface receptor expression (Fig. 1B, right). Repeated immunofluorescence and Western blotting showed a similar pattern in next passage EPC (data not shown), suggesting that this expression was sustained under culture conditions. In the kidney ARAS pigs showed a significant increase in renal vein SCF levels (Fig. 2, top) and Tie-2, VEGF, integrin β2, and EPO expression, while renal expression of angiopoietin-1 and EPOr decreased, and SDF-1 and CXCR-4 remained unchanged (Fig. 2, right).
Effects of EPC Delivery in the ARAS Kidney
An average of 12.9 ± 5.8% of the injected EPC was detected 4 weeks after delivery in the stenotic kidney. Some of the injected EPC coexpressed CD31 or cytokeratin, and appeared to incorporate into microvessels and tubules, respectively, suggesting acquisition of endothelial and tubular characteristics, as we have recently shown . EPC delivery into the ARAS kidney further elevated the renal expression of VEGF, improved angiopoietin-1, renal Tie-2, and integrin β2, and decreased renal vein plasma levels of SCF (Fig. 2, bottom).
Renal MV 3D Architecture and Function
MV density was diminished throughout the ARAS renal cortex (inner, middle, and outer). EPC selectively improved MV density in the outer cortex (Fig. 3), whereas the inner and middle cortical MV density remained attenuated. The increased MV density in ARAS + EPC might have contributed to improve (albeit not normalize) RBF and GFR, which had been decreased in ARAS (Table 1). This improvement was accompanied by downregulated renal expression of NAD(P)H oxidase (p47 subunit), xanthine oxidase, and decreased renal superoxide that were elevated in ARAS, indicating a decrease in oxidative stress by EPC (Fig. 4).
Renal Fibrosis and Morphology
The ARAS kidney showed renal glomerulosclerosis and tubulointerstitial and perivascular fibrosis, which was greater in the inner than outer cortex, and accompanied by increased expression of renal TGF-β and TIMP-1, and attenuated MMP-2 (Fig. 5). These deleterious changes were largely improved in ARAS + EPC, indicating an overall decrease in renal remodeling, mainly evident in the outer cortex.
The current study shows that experimental renovascular hypertension modulates systemic EPC homing signaling, and identifies candidate homing/retention signals used by the ischemic kidney to recruit and retain endogenous circulating EPC to stimulate its reparative processes. Autologous EPC expanded in culture expressed a variety of homing signal receptors, some of which corresponded to those expressed in the kidney and likely facilitated their retention and action. Compared with EPC isolated from normal animals, EPC from ARAS showed increased expression of SDF-1, angiopoietin-1, Tie-2, and c-kit, suggesting that early atherosclerosis modulated the homing capability of EPC. On the other hand, although the stenotic kidney activated some corresponding homing signaling, angiopoietin-1 and SDF-1/CXCR4, common mediators of cell homing, were either blunted or unchanged, implying incomplete heralding of an ischemic insult. Notably, delivery of autologous EPC restored renal angiogenic potency, improved cortical MV density, renal hemodynamics and function, decreased fibrosis of the stenotic kidney, and consequently decreased the endogenous injury signals.
Peripheral blood contains stem cell-like and monocytic-like endothelial cell progenitors, which have in vivo homing specificity to injured sites. These circulating cells have the potential to change phenotype when needed to promote vascular or parenchymal repair, and can stimulate neovascularization through paracrine secretion of growth factors, activate resident stem cells, and decrease tissue injury. A number of factors have been proposed to be released from damaged tissues, mobilize endogenous progenitor, and stem cells , and subsequently facilitate their engraftment in the host tissue. The interaction between tissues and circulating cells is mediated by corresponding molecules and cognate receptors released by or expressed in the injured tissue, and reciprocally the EPC.
Of the proposed key mediators of such signaling, SDF-1 and its CXCR4 receptor have been identified as a central signaling axis regulating migration and homing of progenitor cells to ischemic tissues [5, 25, 26]. A recent report suggests that an ischemic milieu, as characterizes the stenotic kidney, preconditions these cells and improves their migratory and homing capabilities  partly through stimulation of SDF-1/CXCR4 . We found that EPC isolated from ARAS pigs highly expressed SDF-1, while the expression of CXCR4 was similar to EPC from normal pigs. This pattern was relatively stable, as it was sustained during several passages in culture. However, the stenotic kidney did not increase the expression of either SDF-1 or CXCR4 compared with controls, implying a minor (if any) role of the SDF-1/CXCR4 axis in contributing to renal homing of EPC in our model of ARAS. Our observation is supported by recent evidence showing that stem cells can migrate and home into the kidney with acute ischemia/reperfusion injury independent of the SDF-1/CXCR4 axis .
On the other hand, a number of factors were upregulated in ARAS. Uric acid, which is overproduced by ischemic tissues and capable of mobilizing EPCs , as well as SCF, which improves the EPC homing ability  via its c-kit receptor (upregulated in EPC from ARAS), were both increased in the stenotic kidney vein. These were accompanied by increased renal expression of EPO [30–32], who attracts and directs circulating endogenous cell progenitors into ischemic tissues, and integrin β2, which promotes the adhesion of cell progenitors and augments neovascularization [30, 33]. Along the same lines, since angiopoietin-1 is essential to promote endothelial cell survival, vascular branching, and pericyte recruitment , the increased expression of angiopoietin-1/Tie-2 in EPC from ARAS supports their potential for cell homing and vascular proliferation . However, the coexpression of both the ligand and receptor in EPC is intriguing and of uncertain significance, yet likely does not diminish the homing potential and may prolong the survival of circulating EPCs . It is interesting that we observed a dissociation of the upregulated expression of SDF/CXCR4 and angiopoietin-1/Tie-2 in the EPC but not the kidney. This likely reflects the plasticity of EPC to respond to an array of signals released from different injury sites, whereas in the injured ARAS kidney oxidized lipids  or increased oxidative stress  can interfere with EPC function [39–41]. Indeed, the signals might be tissue- and disease-specific, and thus different in the ARAS kidney from those observed in other conditions, such as the ischemic myocardium  or hind-limb , or diabetes .
Progenitor cells exert their effects not only by lodging in a host tissue but also by influencing surrounding parenchymal cells in a paracrine and autocrine fashion. We have recently shown in RAS uncomplicated by atherosclerosis that intrarenally infused EPC successfully engraft and incorporate into vascular and tubular structures in the stenotic kidney, improving angiogenic signaling and MV proliferation and decreasing renal damage . Since atherosclerosis is the main etiology of RAS in humans, in the current study, we used a clinically relevant model that mimics early atherosclerotic renovascular disease. This study extends our previous observations [3, 19, 37], and demonstrates that renal dysfunction and injury in this model manifests in a decrease in cortical MV density and inadequate angiogenic signaling. Importantly, intrarenal delivery of EPC improved renal expression of most of the angiogenic, mobilizing, and homing factors, likely consequent to partial repair of the ARAS + EPC kidney and decreased need for progenitors. Progenitor cells are rich in vasculogenic factors like VEGF and its mediators, and tissue repair by EPC is partly mediated through increased delivery of VEGF and modulation of angiopoietins in the target tissue . Angiopoietin-1 is a potent homing  and angiogenic factor that interacts with VEGF  to not only promote neovascularization and maturation of new vessels [47, 48] but also to increase and extend circulating cell progenitors . Therefore, the increase in renal expression of the specific angiopoietin receptor Tie-2 may imply improvement in both homing signal for EPC and angiogenic signaling in the ischemic kidney.
We have recently shown that in non-ARAS EPC effectively increased MV proliferation throughout the renal cortex . Contrarily, generation of new vessels after EPC administration in ARAS was only evident in the small vessels in the outer renal cortex. This diminished effect may reflect preferential lodging of EPC in small vessels or sprouting of microvessels of smaller branching order, which are more abundant in the outer cortex . The increased vascularity in the outer cortex was accompanied by a decrease (relative also to the inner cortex) in renal fibrosis, implying that the inner cortex may be more vulnerable to insults, possibly due to its spatial and functional relationship with the medulla. It is also possible that marked fibrosis in the inner cortex at the time of EPC delivery impeded neovascularization in this region. Because the number of cells (CFU) in both RAS  and ARAS was similar to normal pigs, the greater renal functional and structural damage in ARAS [3, 19] or lower EPC MV adherence and retention may have reduced their efficacy. In fact, a failure to increase EPC numbers on injury may have further interfered with renal repair. Nevertheless, decreased cortical fibrosis and MV proliferation in the outer cortex of the ARAS + EPC kidneys was sufficient to improve in the decreased RBF and GFR that characterize ARAS . This functional recovery likely resulted also from decreased renal oxidative stress and thereby vasoconstriction due to improved function of new or repaired vessels. Furthermore, EPC also distinctly improved the blunted expression of MMP-2 in the ARAS kidney, without affecting the upregulated expression of its inhibitor TIMP-1, and decreased the expression of the profibrotic TGF-β, thereby facilitating matrix turnover. Overall, the increased blood supply and decreased noxious milieu resulted in attenuated renal fibrosis in ARAS + EPC.
The limitations of this study included the early stage of ARAS, because comorbid conditions, duration, and severity of ischemia may modulate regulation of different mediators. Additional factors may have also contributed to the mobilization, homing, and engraftment of EPC in the ARAS kidney. EPC expanded in culture may not be identical to the endogenous circulating cells, yet may partly represent their function. In addition, the contribution of signals to EPC mobilization and homing, although widely accepted, was not directly demonstrated in this study. Further studies are needed to determine how this expression correlates with the efficacy of EPC-based reparative processes, and whether inadequate renal repair achieved by endogenous EPC is related to incomplete homing signals and progenitor recruitment capacity in ARAS.
In summary, using a clinically relevant model of experimental ARAS, the current study suggests that the stenotic ARAS kidney expresses specific homing factors that correspond to cognate receptors expressed on EPC, suggesting organ-specific mechanisms for EPC recruitment. Furthermore, cultured autologous EPC from ARAS pigs reveal angiogenic capacity and express homing receptors that correspond to a wide array of signals that may permit their response to a variety of signals from injured organs. Indeed, intrarenal delivery of exogenous EPC partly reversed the functional deterioration and damage to the ARAS kidney, which in turn decreased the injury signal it activated. These findings sheds light into the mechanisms by which EPC exert their renoprotective effects and, importantly, show feasibility and potential for cell-based therapy to rescue the kidney in chronic atherosclerotic renovascular disease.
This work was supported by NIH (grant numbers: DK-73608, DK-77013, HL-77131, and PO1HL085307) and by American Heart Association (Scientist Development Grant: 0830100N). A.R.C. is currently affiliated with the Department of Physiology and Biophysics, University of Mississippi Medical Center, Jackson, MS.
Disclosure of Potential Conflicts of Interest
The authors indicate no potential conflicts of interest.