SEARCH

SEARCH BY CITATION

Keywords:

  • Human mesenchymal stem cells;
  • Acute renal failure;
  • Tubular cells;
  • Kidney repair

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References

Transplantation of bone marrow mesenchymal stem cells (BM-MSC) or stromal cells from rodents has been identified as a strategy for renal repair in experimental models of acute kidney injury (AKI), a highly life-threatening clinical setting. The therapeutic potential of BM-MSC of human origin has not been reported so far. Here, we investigated whether human BM-MSC treatment could prevent AKI induced by cisplatin and prolong survival in an immunodeficient mouse model. Results showed that human BM-MSC infusion decreased proximal tubular epithelial cell injury and ameliorated the deficit in renal function, resulting in reduced recipient mortality. Infused BM-MSC became localized predominantly in peritubular areas and acted to reduce renal cell apoptosis and to increase proliferation. BM-MSC also induced protection against AKI-related peritubular capillary changes consisting of endothelial cell abnormalities, leukocyte infiltration, and low endothelial cell and lumen volume density as assessed by morphometric analysis. These findings indicate that human MSC of bone marrow origin hold potential to prolong survival in AKI and should be considered for testing in a clinical trial.

Disclosure of potential conflicts of interest is found at the end of this article.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References

Author contributions: M.M. and B.I.: conception and design, collection and/or assembly of data, data analysis and interpretation, manuscript writing; M.I.: conception and design, provision of study material or patients, data analysis and interpretation; M.M., M.I., and B.I. contributed equally to this work; D.C. and D.R.: data analysis and interpretation; M.A.: collection and/or assembly of data, data analysis and interpretation, manuscript writing; C.R. and C.Z.: collection and/or assembly of data, data analysis and interpretation; A.B.: administrative support, collection and/or assembly of data, data analysis and interpretation, final approval of manuscript; N.P.: collection and/or assembly of data; A. Rambaldi: provision of study material or patients; A. Remuzzi: data analysis and interpretation, collection and/or assembly of data; G.R.: conception and design, administrative support, data analysis and interpretation, manuscript writing, final approval of manuscript.

Acute kidney injury (AKI) [1] is a potentially devastating disorder in clinical medicine. The mortality associated with AKI has remained dismally high over the last 50 years [2]. Pharmacologic strategies have focused on targeting mechanisms and individual mediators thought to contribute to ischemic or toxic renal injury in experimental AKI models. However, translational research efforts have yielded disappointing results [3, [4], [5]6]. In the attempt to create innovative interventions for the cure of AKI, recent studies have tested stem cell-based technology, which provides the exciting prospect of a powerful treatment to repair acutely damaged organs by virtue of the unique stem cell tropism and proregenerative capacity [7, [8]9].

Mesenchymal stem cells (MSC) are a multipotent stem cell population that mostly resides in the bone marrow (BM) [10]. BM-MSC, or stromal cells, deserve attention for their proved plasticity, being capable of differentiating into cell types other than their tissue of origin [10, [11]12], and for their spectrum of growth factors, including vascular endothelial growth factor, hepatic growth factor, insulin-like growth factor-1 (IGF-1), and antiapoptotic cytokines that might be released to the place of engraftment [13, [14]15]. In rodents, BM-MSC treatment enhanced neoangiogenesis and tissue repair in ischemic skeletal muscle [16] and myocardium [17, 18] by local production of angiogenic factors [19]. Increasing evidence suggests that the therapeutic potential of MSC could be applied to AKI [14, 20, [21], [22]23]. A key target in AKI is the tubular epithelium undergoing detachment and apoptosis [24]. The infusion of murine MSC of BM origin protected cisplatin-treated syngenic mice from tubular injury and renal function deterioration by stimulating the proliferation of resident renal cells via local production of IGF-1 and to a lesser extent by differentiation in tubular epithelium [20, 25]. Studies in the ischemia/reperfusion model found that BM-MSC could be beneficial via paracrine production of antiapoptotic, mitogenic, and vasculotropic factors [14, 26]. Renal endothelial cell dysfunction leading to persistent vasoconstriction and reduction of blood flow has indeed been suggested to amplify the deleterious effects of tubular cell injury in AKI, thus representing an additional candidate target of therapy [2, 27, 28].

Preclinical and early clinical studies have shown effectiveness of human BM-MSC in genetic disorders, ischemic cardiomyopathy, and hematological diseases [29]. The potential of BM-MSC of human origin in AKI is awaiting investigation. Experimental studies have been conducted on models using BM-MSC of rodent origin, and the effect on mortality has never been reported. Here, we sought to assess whether BM-MSC of human origin may represent a therapeutic option in the clinical perspective to cure AKI. We used immunodeficient NOD-SCID mice with cisplatin-induced AKI, which in this strain caused animal death within 7 days after cisplatin injection. We evaluated whether (a) human BM-MSC could be protective against renal structural and functional injury, and (b) human BM-MSC treatment could reduce mortality in this AKI model.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References

Experimental Design

Female 2-month-old NOD-SCID mice (Charles River Italia s.p.a., Calco, Italy, http://www.criver.com) were used. Animal care and treatment were conducted in conformity with the institutional guidelines that are in compliance with national (Decreto legislativo no. 116, Gazzetta Ufficiale, suppl 40, 18 Febbraio 1992, Circolare no. 8, GU, 14 Luglio 1994) and international (EEC Council Directive 86/609, OJL 358, Dec 1987; NIH Guide for the Care and Use of Laboratory Animals, U.S. National Research Council, 1996) laws and policies. AKI was set up by subcutaneous injection of the nephrotoxic drug cisplatin (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) at a dose of 12.7 mg/kg [20] (n = 5 mice). Body weight, renal function, and survival were assessed. A preliminary experiment was done to characterize the extent of tubular cell injury in NOD-SCID cisplatin-treated mice sacrificed at 24 hours (n = 3), the time at which human BM-MSC were infused. Ultrastructural analysis revealed mitochondrial swelling, loss of brush border, and myelin figures in tubular epithelial cells (not shown). To test the effect of human BM-MSC injection, mice were divided into two groups and intravenously (i.v.) injected into tail vein, 24 hours after cisplatin, as follows: group 1, saline (n = 10); group 2, human BM-MSC (5 × 105 cells per 500 μl) (n = 9).

Mice were sacrificed at 4 days after cisplatin for determination of renal function, histology, apoptosis, and proliferation and morphometric analysis. For survival studies, seven cisplatin-treated mice given saline and six cisplatin-treated mice infused with human BM-MSC were used.

The effect of i.v. injection of human fibroblasts (5 × 105 cells per 500 μl) was assessed in mice 24 hours after cisplatin administration (n = 9), in comparison with mice given saline (n = 10). Three mice treated with human fibroblasts and four saline-treated mice were sacrificed at 4 days to assess renal function and histology. The remaining mice were followed during time for survival.

Renal function was assessed as blood urea nitrogen (BUN) by the Reflotron test (Roche Diagnostics, Indianapolis, http://www.roche-applied-science.com). Blood samples for BUN determination were collected at different time intervals after cisplatin injection. BUN levels exceeding 30 mg/dl were considered abnormal. Normal mice served as controls.

Human BM-MSC

Human BM aspirates were collected from adult subjects (30–40 years old) within the Hematology Division, Azienda Ospedaliera, Ospedali Riuniti di Bergamo [30]. Heparinized diagnostic samples were obtained during the clinical follow-up of patients previously treated by autologous or allogeneic bone marrow transplantation who were in complete confirmed remission, after informed consent [30]. Briefly, cells were separated by Ficoll-Hypaque gradient centrifugation (Lympholyte-H; Cedar Lane, Hornby, ON, Canada, http://www.cedarlanelabs.com) to yield mononuclear cells [30]. Human BM-MSC were plated at 2 × 105 cells per cm2 in Dulbecco's modified Eagle's medium low-glucose (Gibco, Carlsbad, CA, http://www.invitrogen.com) containing 0.1 mM gentamicin and 1,000 IU of heparin in the presence of 10% fetal bovine serum. Nonadherent cells were removed after 3–4 days, and fresh medium was added. At subconfluence, cells were recovered after trypsin-EDTA treatment and subsequently replated in the same medium [30]. Before in vivo experiments, cells were characterized for their capability to differentiate toward adipocytes and osteoblasts as previously reported [30]; their mesenchymal phenotype was also assessed by fluorescence-activated cell sorting (FACS) [30]. As control cells, human dermal fibroblasts (HuDe cell line; Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia Romagna, Brescia, Italy) were used at passage 10.

Renal Morphology

Renal Histology.

Kidney samples were fixed in Duboscq-Brazil. Paraffin sections (3-μm thickness) were stained with hematoxylin and eosin or with periodic acid-Schiff reagent. Luminal hyaline casts and cell loss (denudation of tubular basement membrane) were assessed in nonoverlapping fields (up to 28 for each section) using a ×40 objective (high-power field [HPF]). Numbers of casts and tubular profiles showing necrosis were recorded in a single-blind fashion.

Electron Microscopy.

Fragments of kidney tissue were fixed for 4 hours in 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4, and washed repeatedly in the same buffer. After postfixation in 1% OsO4, specimens were dehydrated through ascending grades of alcohol and embedded in Epon resin (Fluka–Sigma-Aldrich). Ultrathin sections were stained with uranyl acetate and lead citrate and examined using a Philips Morgagni electron microscope (Fei Company, Hillsboro, OR, http://www.fei.com).

Identification of PKH-26-Labeled Human BM-MSC

To study intrarenal localization and perform quantification of human BM-MSC, cells were labeled with PKH-26 [25] red fluorescence cell linker (Sigma-Aldrich) and then were infused in cisplatin-treated mice (n = 5). Labeling efficacy was >98%. Viability, evaluated by trypan blue exclusion, was >96%. After 4 days, mice were sacrificed, and kidney samples were fixed in 4% paraformaldehyde, infiltrated with 30% sucrose/phosphate-buffered saline (PBS), embedded in Tissue-Tek OCT Compound (Sakura Finetek, Torrance, CA, http://www.sakura-america.com), and fresh-frozen in liquid nitrogen. Eight-micrometer-thick sections were stored at −70°C. Sections were fixed in acetone (10 minutes) and incubated with fluorescein isothiocyanate (FITC)-labeled lectin wheat germ agglutinin (WGA; Vector Laboratories, Burlingame, CA, http://www.vectorlabs.com), which binds membrane glycoproteins and sialic acid and was used to better identify tubular structures. Nuclei were stained with 4′,6′-diamidino-2-phenylindole dihydrochloride hydrate (DAPI; Sigma-Aldrich). Twenty sections per mouse (n = 5 mice) were analyzed, and PKH-26-positive cells were counted. Data were expressed as number of PKH-26-positive cells per 105 renal cells.

To document the human origin of BM-MSC, kidney sections of mice injected with PKH-26-labeled cells were stained with mouse anti-human CD105 antibody (DakoCytomation, Glostrup, Denmark, http://www.dakocytomation.com) followed by anti-mouse Cy5 (1 hour at room temperature; Jackson Immunoresearch Laboratories, West Grove, PA, http://www.jacksonimmuno.com). Samples were counterstained with FITC-labeled lectin WGA, and nuclei were stained with DAPI.

Immunohistochemistry

Apoptosis.

Apoptosis was measured by a terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay (Roche Diagnostics) followed by rhodamine-labeled lectin lens culinaris agglutinin (Vector Laboratories) and DAPI. Slides were analyzed by confocal microscopy. Apoptotic nuclei and DAPI-positive cells per field were counted, and the results were expressed as TUNEL-positive cells per 103 cells.

Proliferation.

To determine the number of proliferating tubular cells, formalin-fixed paraffin sections were stained with a monoclonal anti-proliferating cell nuclear antigen (anti-PCNA) clone PC10 (1:250; Sigma-Aldrich) followed by biotinylated sheep anti-mouse IgG (1:50; Chemicon, Temecula, CA, http://www.chemicon.com). PCNA signal was developed by Vectastain ABC kit (Vector Laboratories) using 3,3′ diaminobenzidine as substrate. Nuclei were visualized by counterstaining with Harris hematoxylin. Evaluation was performed by counting PCNA-positive cells in tubular profiles in at least 15 HPF (×400) per section.

Length Density and Volume Density of Endothelial Cells and Lumen of Peritubular Capillaries

Kidney samples were fixed in 4% paraformaldehyde (4 hours), infiltrated with 30% sucrose/PBS (2 hours), embedded in OCT, and fresh-frozen in liquid nitrogen. Three-micrometer-thick sections were stained with rat anti-mouse endothelial cell antigen (MECA)-32, an endothelial cell marker (1 hour at room temperature; Developmental Studies Hybridoma Bank, Iowa City, IA, http://www.uiowa.edu/∼dshbwww) followed by goat anti-rat Cy3 (Jackson Immunoresearch Laboratories). Slides were counterstained with FITC-labeled lectin WGA (Vector Laboratories) and mounted with Moviol. To estimate the area density occupied by MECA-32 staining and by capillary lumen, 30 renal sections per mouse were digitalized using an inverted confocal laser microscopy (original magnification, ×630; LSM 510 meta; Carl Zeiss, Jena, Germany, http://www.zeiss.com). Each image (512 × 512 pixels) was digitally overlapped with an orthogonal grid composed of 676 points (ImageJ; NIH, Bethesda, MD). The volume density (Vv)—also defined as volume fraction—of endothelial cells was calculated as the ratio of the number of grid points hitting MECA-32 staining to the total number of grid points falling into tissue. The Vv of capillary lumen was calculated as the ratio of the number of points falling into lumen to the total number of grid points. The length density (Jv) of peritubular capillary was calculated using the formula Jv = 2 × Q/A, where Q is the number of transections of capillary axes across tissue section and A is the area of microscopic field (in μm2). From Jv and Vv of capillary lumen, the mean diameter of peritubular capillaries was calculated.

Statistical Analysis

The results are expressed as mean ± SE. Analysis of variance followed by Tukey's test for multiple comparisons was used to analyze renal function (BUN), apoptosis, proliferation, and morphometric parameters. Renal histology and survival data were analyzed by nonparametric Kruskal-Wallis test and the log-rank test, respectively. Statistical significance level was defined as p < .05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References

Human BM-MSC Phenotype

Human MSC, obtained from BM aspirates, were grown in culture as previously reported [30] and used at the fifth to the sixth passages. By FACS analysis, human BM-MSC expressed CD90, CD73, and CD105, whereas they did not exhibit expression of the hematopoietic markers CD45, CD34, CD14, and human leukocyte antigen-DR. Human BM-MSC had the capability to differentiate toward adipocytes and osteoblasts (not shown).

Human BM-MSC Infusion Preserves Tubular Epithelial Integrity and Renal Function in Mice with Cisplatin-Induced Acute Kidney Injury

To establish the renoprotective potential of BM-MSC of human origin in mice with AKI, we first set up an experimental model of acute renal failure induced by cisplatin in immunodeficient NOD-SCID mice. Renal function, assessed as BUN, and body weight were evaluated at different time intervals from cisplatin treatment. As shown in Table 1, the subcutaneous injection of cisplatin resulted in a significant increase in serum BUN at 2–3 days, which peaked at 4–5 days and stabilized at high values until animal death. In parallel, animals lost weight starting at 1 day from cisplatin treatment (Table 1). Mice almost invariably died within 5–7 days after receiving cisplatin (Table 1).

Table Table 1.. Time course of body weight, BUN, and survival in NOD-SCID mice injected with cisplatin
Thumbnail image of

Having established the AKI model in NOD-SCID mice, we tested the protective effect of human BM-MSC. For this purpose, mice were intravenously injected in the tail vein, 1 day after cisplatin treatment, with saline or with 5 × 105 human BM-MSC, a dose that did not induce respiratory complications. Since all mice with AKI given saline were still alive at 4 days, renal function and histology were investigated at this time point. Human BM-MSC infusion markedly protected mice from renal function impairment, as reflected by significantly lower BUN levels at 3 and 4 days with respect to mice given saline (Fig. 1A). By contrast, injection of cisplatin-treated mice with human fibroblasts failed to improve renal function (BUN: human fibroblasts, 121 ± 12 vs. saline, 135 ± 4.8 mg/dl) at 4 days.

thumbnail image

Figure Figure 1.. hBM-MSC are renoprotective and engraft the kidney in mice with cisplatin-induced acute kidney injury. (A): Renal function, measured as BUN, in mice treated with cisplatin that received saline (n = 10) or hBM-MSC (n = 9) 1 d after cisplatin. Data are mean ± SE. *, p < .01 versus saline; °, p < .01 versus d 0. (B): Histological evaluation of kidney samples taken from control mouse or cisplatin-treated mice receiving saline or hBM-MSC, 4 d after cisplatin. Original magnification, ×400. (C): Representative micrographs of kidney tissue from cisplatin-treated mouse injected with PKH-26-labeled hBM-MSC at 4 d. PKH-26-fluorescent BM-MSC (red, arrows) were localized in peritubular areas (left) and within tubular epithelium (right). Sections were stained with fluorescein isothiocyanate (FITC)-labeled lectin wheat germ agglutinin (WGA; green), and 4′,6′-diamidino-2-phenylindole dihydrochloride hydrate (DAPI) for nuclei (blue). Original magnification, ×630. (D): Coexpression of PKH-26 (red) and human antigen CD105 (white) by representative hBM-MSC (merged image, yellow). Sections were stained for FITC-labeled lectin WGA (green) and DAPI (blue). Original magnification, ×630. Abbreviations: BUN, blood urea nitrogen; d, days; hBM-MSC, human bone marrow mesenchymal stem cells.

Download figure to PowerPoint

By light microscopy, the kidneys of cisplatin-treated mice given saline showed AKI-associated tubular lesions at 4 days, consisting of loss of brush border, flattening and loss of the tubular epithelium, nuclear fragmentation, and hyaline casts (Fig. 1B; Table 2). No histologic glomerular changes were appreciable. Human BM-MSC treatment markedly attenuated tubular injury in cisplatin-treated NOD-SCID mice. In particular, kidneys of mice receiving human BM-MSC showed a significant reduction in the number of tubules affected by necrosis (Fig. 1B; Table 2). Mice injected with human fibroblasts had severe tubular alterations, comparable to those observed in mice given saline (human fibroblasts: casts, 2.5 ± 2.2; tubular necrosis, 1.8 ± 1.3; vs. saline: casts, 2.8 ± 1.4; tubular necrosis, 1.7 ± 0.9 n/HPF).

Table Table 2.. Effect of hBM-MSC on renal histology at 4 days
Thumbnail image of

We explored the capacity of human BM-MSC to engraft the kidney by evaluating the presence of PKH-26-labeled cells in the renal parenchyma of cisplatin-treated mice receiving BM-MSC infusion at day 4. Their frequency averaged 3.4 ± 0.76 per 105 renal cells. Labeled MSC were localized predominantly in peritubular areas and to lesser degrees within the tubular epithelium or the glomeruli (PKH-26 cells in each compartment per total PKH-26 cells: peritubular areas, 84% ± 0.12%; tubules, 8% ± 0.05%; and glomeruli, 8% ± 0.08%) (Fig. 1C). The human origin of PKH-26-labeled BM-MSC was confirmed in kidney sections costained for human CD105 antigen (Fig. 1D).

Human BM-MSC Treatment Limits Tubular Cell Apoptosis and Accelerates Tubular Cell Regeneration in Cisplatin-Treated Mice

In experimental models of cisplatin-induced AKI, apoptosis is partly responsible for tubular cell loss [24]. We investigated whether human BM-MSC could exert antiapoptotic activity. A significant increase in the number of TUNEL-positive cells was detected in renal sections of cisplatin-treated mice given saline at day 4 with respect to control mice (7.34 ± 2.0 vs. 1.7 ± 0.7 apoptotic cells per 103 cells) (Fig. 2A). TUNEL signal was clearly identified in the tubular epithelium. Human BM-MSC infusion significantly reduced apoptotic cells at 4 days (2.5 ± 0.8 apoptotic cells per 103 cells). Figure 2B shows representative images of kidneys taken from mice receiving saline or human BM-MSC.

thumbnail image

Figure Figure 2.. hBM-MSC reduce apoptosis and enhance kidney regeneration in cisplatin-treated mice with acute kidney injury. (A): Quantification of apoptotic cells, identified by terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay, in kidney sections of control mice (n = 3), cisplatin-treated mice receiving saline (n = 9) or hBM-MSC (n = 9) at 4 d. ○, p < .01 versus control; *, p < .01 versus saline. (B): Representative images of kidney sections labeled with TUNEL (green), rhodamine-labeled lectin lens culinaris agglutinin (red), and 4′,6′-diamidino-2-phenylindole dihydrochloride hydrate (blue). Original magnification, ×630. (C): Quantification of PCNA-positive tubular cells in control mice (n = 3), cisplatin-treated mice receiving saline (n = 8), or hBM-MSC (n = 8) at 4 d. *, p < .01 versus control and saline. (D): Representative images of PCNA staining (arrows) in kidney sections of control and cisplatin-treated mice given saline or hBM-MSC. Original magnification, ×400. Abbreviations: d, days; hBM-MSC, human bone marrow mesenchymal stem cells; HPF, high-power field; PCNA, proliferating cell nuclear antigen.

Download figure to PowerPoint

The effect of human BM-MSC treatment on tubular cell regeneration was explored by evaluating PCNA expression [14, 21, 31] in renal sections of controls, cisplatin-treated mice given saline or human BM-MSC at 4 days (Fig. 2C, 2D). Mice with AKI given saline exhibited a slight increase in the number of PCNA-positive cells compared with normal control mice (6.6 ± 0.7 vs. 2.7 ± 0.5 PCNA-positive cells per HPF). Higher numbers of PCNA-positive cells were detected in tubules upon human BM-MSC infusion (12.5 ± 1.1 PCNA-positive cells per HPF; p < .01 vs. controls and saline) (Fig. 2C, 2D).

Human BM-MSC Treatment Attenuates Peritubular Capillary Changes in Cisplatin-Treated Mice

The maintenance of the microvasculature could have importance for the prevention of acute tubular damage. By morphometric analysis of the peritubular capillary endothelium labeled for MECA-32, we observed a marked reduction (p < .01) in volume density of endothelial cells and capillary lumen in cisplatin-treated mice given saline at 4 days compared with normal NOD-SCID mice (Fig. 3A, 3B, 3D). Neither parameter was altered at day 1 after cisplatin injection (not shown). The length density of peritubular capillaries (Jv) was not altered in cisplatin-treated mice at 4 days given saline or human BM-MSC administration (Fig. 3C). The capillary diameter was significantly reduced by cisplatin treatment compared with controls (3.21 ± 0.24 μm vs. 5.94 ± 0.30 μm, respectively; p < .01). Human BM-MSC treatment significantly restored the volume density of endothelial cells and lumen (Fig. 3A, 3B, 3D) and the capillary diameter (averaging 4.56 ± 0.24 μm; p < .01 vs. cisplatin + saline).

thumbnail image

Figure Figure 3.. Effect of hBM-MSC on volume density of endothelial cells and lumen, and diameter of peritubular capillaries in mice with acute kidney injury. Renal peritubular capillary alterations were evaluated in controls (n = 3) and in cisplatin-treated mice given saline (n = 9) or hBM-MSC (n = 9) at 4 d, by morphometric analysis of digitalized images of renal tissues stained for mouse endothelial cell antigen-32, a marker specific for mouse vascular endothelial cells (red) and fluorescein isothiocyanate-labeled lectin wheat germ agglutinin (green). (A): Endothelial volume density. (B): Lumen volume density. (C): Length density of peritubular capillaries. ○, p < .01 versus control; *, p < .01 versus cisplatin + saline. (D): Representative images of renal tissue of control mouse and cisplatin-treated mice receiving saline or hBM-MSC, at 4 d. Selected areas (on the right) were digitally enlarged to show peritubular capillary morphology. Original magnification, ×630. Abbreviations: d, days; hBM-MSC, human bone marrow mesenchymal stem cells; Jv, length density; Vv, volume density.

Download figure to PowerPoint

We performed electron microscopy analysis to establish whether capillary endothelial cell injury may occur in this model. No ultrastructural changes were observed at day 1 after cisplatin injection. At 4 days, peritubular capillary endothelial cells showed focal abnormalities consisting of mild cytoplasmic swelling or retraction, in association with polymorphonuclear leukocytes in the capillary lumen or, less frequently, in contact with the underlying basement membrane (Fig. 4B, 4C). Cisplatin-treated mice receiving human BM-MSC showed no appreciable ultrastructural peritubular changes (Fig. 4D).

thumbnail image

Figure Figure 4.. Ultrastructural changes of peritubular capillaries in cisplatin-treated mice at 4 days and effects of human bone marrow mesenchymal stem cell (hBM-MSC) infusion. Shown are representative electron micrographs of tubulointerstitial areas of normal control mouse (A), cisplatin-treated mouse receiving saline (B,C), and cisplatin-treated mouse injected with BM-MSC (D). Arrows (C) indicate a site at which a PMN is in contact with the basement membrane of a peritubular capillary. Original magnifications, ×2,800 (A, D), ×5,600 (B), and ×14,000 (C). Abbreviations: PMN, polymorphonuclear cell; PTC, proximal tubular cell.

Download figure to PowerPoint

Human BM-MSC Treatment Ameliorates Mortality of Mice with Acute Kidney Injury

The high susceptibility of NOD-SCID mice to cisplatin toxicity allowed us to test whether BM-MSC treatment could affect survival, the key clinical outcome in AKI. Survival curves of NOD-SCID mice with AKI given saline or human BM-MSC are shown in Figure 5. Mice infused 1 day after cisplatin with human BM-MSC survived significantly longer than saline-treated mice (p < .02). At 7 days, 50% of mice injected with human BM-MSC were still alive, whereas all mice died in the saline group. At 14 days, the percentage of surviving mice given human BM-MSC was 34%. Cisplatin-treated mice injected with human fibroblasts had the same mortality as cisplatin-treated mice given saline. At 7 days, no animal in either group survived.

thumbnail image

Figure Figure 5.. hBM-MSC treatment prolongs survival in mice with acute kidney injury. All mice receiving saline died within 7 days. NOD-SCID mice given hBM-MSC survived significantly longer than saline-treated mice. *, p < .02 versus cisplatin + saline. Abbreviation: hBM-MSC, human bone marrow mesenchymal stem cells.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References

Experimental studies have shown homing of BM-MSC and beneficial effects of stem cell-based therapy on tissue structural repair in myocardial infarction [32, 33], neurologic diseases [34, 35], and acute kidney injury [14, 20, 21]. However, studies have disclosed an improvement of organ function [14, 20, 21, 23, 34, 36], leaving open the question of whether the treatment may affect clinical outcome. The present report is the first to document a positive effect of human stem cell treatment on survival. We used a model of acute kidney failure whereby animals die of uremia. The treatment with human BM-MSC was life-sparing via improvement of kidney function. Intravenous infusion of human BM-MSC protected mice from renal function impairment and consequent death primarily by exerting protective actions against tubular cell damage after cisplatin injection. Finding that human BM-MSC are protective strongly supports their potential clinical use as a way to ameliorate the outcome of AKI. Among the many different stem cells recently described, human BM-MSC offer a very useful tool in terms of accessibility, expansion potential, and genetic stability [29, 37, 38]. Moreover, human BM-MSC migrate to sites of injury [29, 37], produce high amounts of anti-inflammatory cytokines and growth factors [13, 15, 39], and possess immunomodulatory and tolerogenic properties [40, 41] that assign advantages over other stem cell types. Cultured BM-MSC have been infused in humans successfully for safety and early clinical testing for treatment of genetic disorders [42, 43], in ischemic cardiomyopathies [44], and in hematological pathologies such as graft-versus-host disease after hematopoietic cell transplantation [45].

Mechanisms underlying protective effects of MSC are beginning to be elucidated. Experiments in the ischemia reperfusion model in rats suggested paracrine pathways via growth factors, prosurvival molecules, and anti-inflammatory cytokines [14] of MSC origin. In fact, the administration of conditioned medium derived by murine BM-MSC to mice with cisplatin-induced AKI ameliorated renal function, enhanced tubular cell proliferation, and limited renal cell apoptosis [23]. Importantly, the regeneration of surviving tubular epithelial cells has recently been shown to represent the predominant mechanism of kidney repair after acute tubular injury [46]. In this respect, the release of IGF-1 [25] and probably other prosurvival growth factors [14] by BM-MSC stimulates tubular cell proliferation and exerts a powerful inhibitory action on caspase-dependent tubular epithelial cell apoptosis induced by cisplatin [24]. Here, in agreement with our previous data using murine BM-MSC [20, 25], finding PKH-26-fluorescent human BM-MSC in the proximity of tubular profiles and, less frequently, in the context of tubular epithelium in kidneys of NOD-SCID mice with AKI supports the concept that BM-MSC were protective by contributing locally to tubular regeneration, acting mainly via paracrine mechanisms to promote strong proliferative and antiapoptotic responses.

The relative paucity of BM-MSC in renal tissue in the face of remarkable renoprotective effects is consistent with previous findings in animals with AKI, showing that injected BM-MSC transiently localized at early time points, in peritubular areas, and disappeared after 24 hours in ischemic rat kidney [14]. In the glycerol-induced model of AKI, injected murine BM-MSC were recruited early via a CD44/hyaluronic acid-mediated mechanism largely in peritubular areas, also resulting in kidney recovery [47]. Besides the kidney, systemically infused rodent BM-MSC also localized in other organs, including lung, spleen, bone marrow, and liver [12, 14]. Engraftment of human BM-MSC intravenously injected in NOD-SCID mice was also detected, at a very low level, in lungs, bone marrow, heart, liver, and brain [48].

The main recognized direct target of cisplatin toxicity is the proximal tubular epithelial cell, which is endowed with highly cationic transporters for basolateral uptake of cisplatin from the adjacent peritubular capillary [49]. Cisplatin causes DNA damage, mitochondrial dysfunction, and reactive oxygen species formation followed by tubular necrosis [50]. Cisplatin activates the intrarenal expression of vasoactive mediators [51, 52] and proinflammatory factors, particularly tumor necrosis factor-α, which perturbs the peritubular endothelium [53], favoring the migration of inflammatory cells and leukocyte-mediated changes in vascular tone and perfusion [54, [55]56]. Here, peritubular endothelial ultrastructural changes and polymorphonuclear leukocyte infiltration in AKI mice developed later than the tubular epithelial damage, indicating that endothelial cell dysfunction amplifies deleterious effects of tubular cell damage, as suggested in ischemic AKI [27]. Another novel finding in the present study is that peritubular capillaries were significantly reduced in volume and diameter in cisplatin-treated mice at 4 days, and the BM-MSC treatment almost normalized peritubular capillary diameter and volume density of both endothelium and lumen. Our data, which show that human BM-MSC protected mice against tubular cell damage, peritubular capillary changes, and inflammatory cell infiltration, suggest that stem cell therapy by exerting epithelial cytoprotective action also reduces endothelial cell activation and ultimately preserves microvascular integrity. Notably, this beneficial effect is likely to translate into amelioration of hemodynamic changes. Thus, assuming a constant perfusion pressure among groups and dependence of flow resistance from blood viscosity and capillary radius [57], we calculated a >490% increase in resistance to blood flow and >80% decrease in perfusion of peritubular capillaries in cisplatin-treated mice, which were attenuated to 180% and 44%, respectively, upon human BM-MSC treatment. Tissue oxygenation would be improved in this setting. On the other hand, one cannot exclude direct effects of BM-MSC on endothelial cells. MSC-conditioned medium was recently reported to activate a phosphoinositide 3-kinase prosurvival pathway and to reduce hypoxia-induced apoptosis in cultured human aortic endothelial cells [13].

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References

In conclusion, we have shown that human BM-MSC infusion, by preserving the integrity of the tubular epithelium and of peritubular microvessels, prolongs survival in AKI mice. The lack of progress in the survival of patients with AKI is a major, long-standing issue in clinical medicine. BM-MSC represent an ideal cell population for future cell therapy and, in perspective, may hold potential for successful application in human AKI.

Disclosure of Potential Conflicts of Interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References

The authors indicate no potential conflicts of interest.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References

We thank Stefania Angioletti for valuable help in morphometric analysis and Fabio Sangalli for technical assistance with confocal microscope. We are also indebted to Dr. Annalisa Perna for statistical analysis. C.R. is a recipient of the fellowship “Fondazione Aiuti per la Ricerca sulle Malattie Rare, ARMR, Delegazione di Pisa.” The work was partially supported by the Italian Association for Cancer Research and the “Associazione Italiana contro le Leucemie, linfomi e mieloma (AIL) Bergamo-Sezione Paolo Belli.” The monoclonal antibody antiMECA-32, developed by Dr. Eugene C. Bucher, was obtained from the Developmental Studies Hybridoma Bank, developed under the auspices of the National Institute of Child Health and Human Development and maintained by the Department of Biological Sciences, University of Iowa, Iowa City, Iowa.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References