Erythropoietin gene-enhanced marrow mesenchymal stromal cells decrease cisplatin-induced kidney injury and improve survival of allogeneic mice

Abstract

Bone marrow-derived mesenchymal stromal cells (MSCs) are promising for regenerative medicine applications, such as for renoprotection and repair in acute kidney injury (AKI). Erythropoietin (Epo) can also exert cytoprotective effects on various tissues including the kidney. We hypothesized that MSCs gene-enhanced to secrete Epo may produce a significant beneficial effect in AKI. Mouse Epo-secreting MSCs were generated, tested in vitro, and then implanted by intraperitoneal injection in allogeneic mice previously administered cisplatin to induce AKI. Epo-MSCs significantly improved survival of implanted mice as compared to controls (67% survival versus 33% with Vehicle only). Also, Epo-MSCs led to significantly better kidney function as shown by lower levels of blood urea nitrogen (72 ± 9.5 mg/dl versus 131 ± 9.20 mg/dl) and creatinine (74 ± 17 µmol/l versus 148±19.4 µmol/l). Recipient mice also showed significantly decreased amylase and alanine aminotransferase blood concentrations. Kidney sections revealed significantly less apoptotic cells and more proliferating cells. Furthermore, PCR revealed the presence of implanted cells in recipient kidneys, with Epo-MSCs leading to significantly increased expression of Epo and of phosphorylated-Akt (Ser473) (P-Akt) in these kidneys. In conclusion, our study demonstrates that Epo gene-enhanced MSCs exert significant tissue protective effects in allogeneic mice with AKI, and supports the potential use of gene-enhanced cells as universal donors in acute injury.

Figure 1

Characterization of mesenchymal stromal cells (MSCs) and Epo-MSCs. Flow cytometry analysis was conducted (a) on tissue-culture expanded primary MSCs and (b) Epo gene-enhanced MSCs to determine the expression of cell surface antigens CD31, CD34, CD44, CD45, CD73, CD105, H-2Kb (Kb) and I-Ab, as indicated in the Materials and Methods. The darker line represents the specific antibody, and the lighter line the isotype control. (c) Also, a portion of MSCs and Epo-MSCs were tested in vitro by exposure in culture to specific media inducing the osteogenic differentiation (center) or adipogenic differentiation (lower) of both cell populations as described in the Materials and Methods, and as revealed following Alizarin Red S (100X) or Oil Red O (400X) staining, respectively. These cells were photographed under bright-field microscopy with an Axiovert 25 microscope.

Figure 2

Effect of conditioned media (CM) from mesenchymal stromal cells (MSCs) and Epo-MSCs on kidney cell survival in vitro. MM55.K mouse kidney epithelial cells were exposed to CM from MSCs, CM from Epo-MSCs, or complete media, with or without cisplatin at 7.5 µg/ml for 42 hours as described in the Materials and Methods. (a) All kidney cells were then recovered, stained, and analyzed by flow cytometry for annexin V and propidium iodide (PI) expression. (b) Experiment was performed three separate times and average of live kidney cells (negative for both annexin V and PI) calculated. Mean ± SEM.

Figure 3

Effect of mesenchymal stromal cells (MSCs) and Epo-MSCs conditioned media (CM) on protein expression in kidney cells. MM55.K mouse kidney epithelial cells were exposed, as described in Materials and Methods, to cisplatin at 7.5 µg/ml in the presence of CM from MSCs, CM from Epo-MSCs, or complete media for 42 hours. All kidney cells were then recovered, lysed, and used for Western blot analysis to assess expression of (a) Cleaved Caspase-3, (b) Cleaved Caspase-7, (c) Bcl-2, and (d) Bcl-xl, with β-tubulin levels used as internal control. The experiment was conducted three separate times and mean ± SEM values for fold change shown under representative blot for each protein.

Figure 4

Effect of conditioned media (CM) from mesenchymal stromal cells (MSCs) and from Epo-MSCs on P-Akt expression in kidney cells. MM55.K mouse kidney epithelial cells were exposed to cisplatin, with/without MSCs CM or Epo-MSCs CM (a) at increasing concentrations or (b) at highest concentrations with/without a neutralizing anti-IGF-1 antibody (IGF1 Ab), and/or anti-VEGF antibody (VEGF Ab), or anti-Epo antibody (Epo Ab), as detailed in the Materials and Methods. Cells were collected 42 hours later and lysates used for western blot analysis of phosphorylated Akt (Ser473) (P-Akt) expression, as well as of loading controls GAPDH or total Akt expression. (b) The bar graph represents the mean ± SEM of seven independent experiments.

Figure 5

Effect of mesenchymal stromal cells (MSCs) and Epo-MSCs on blood urea nitrogen (BUN) and survival of cisplatin-treated mice. (a) Mice given cisplatin were injected intraperitoneally with allogeneic MSCs (gray box, n = 30) or Epo-MSCs (black box, n = 30), or with Vehicle only (white box, n = 30), and BUN levels determined from blood collected at day 4 (median seen as horizontal bar within box in boxplot with whiskers from minimum to maximum), (b) and at three time points for mice kept long-term (mean ± SEM, n = 18 on day 4 for each of the three groups). (c) Survival of these mice over time was also assessed (n = 18). Normal mice not given cisplatin were used as controls.

Figure 6

Blood creatinine, amylase and alanine aminotransferase (ALT) levels in mice. Mice implanted by intraperitoneal injection with allogeneic mesenchymal stromal cells (MSCs) (gray columns) or Epo-MSCs (black columns), or with Vehicle only (white columns) 1 day following cisplatin administration, were killed 4 days later, and blood collected analyzed for (a) creatinine, (b) amylase and (c) ALT levels (n = 9–15, mean ± SEM). Normal mice not given cisplatin were utilized as controls (striped columns).

Figure 7

Immunohistochemical analysis of mouse kidneys. Mice implanted intraperitoneally with allogeneic mesenchymal stromal cells (MSCs), Epo-MSCs, or Vehicle only, 1 day following cisplatin administration, were killed 4 days later and kidneys removed, sectioned and stained for detection of Caspase-3 and Ki-67. (a) Caspase-3 (top panel) or Ki-67 (lower panel) expressing cells are indicated by black arrows in representative images. (b) The percentages of Caspase-3 positive cells and (c) Ki-67 positive cells per mm² kidney area were also evaluated as described in the Materials and Methods (n = 4–6 mice/group, mean ± SEM). Normal mice not given cisplatin were used as controls.

Figure 8

PCR and western blot analysis of kidneys. (a) Genomic DNA isolated from kidneys of cisplatin-treated mice implanted with mesenchymal stromal cells (MSCs), Epo-MSCs, or Vehicle only, were used to amplify a 444-base pair fragment of mouse Y chromosome and a 363-base pair fragment of internal control β-globin, and findings from individual mice shown. The same mouse kidneys were also used for isolation of total protein used for (b) western blot detection of Epo or (c) of phosphorylated-Akt (Ser473) (P-Akt), and (b,c) of GAPDH internal control. Results from individual mice are shown in upper portions of panels b and c, whereas lower portions reveal levels (means ± SEM (n = 6)) of Epo or P-Akt expression relative to GAPDH.