Mesenchymal stem cell-derived microvesicles protect against acute tubular injury

Abstract

Administration of mesenchymal stem cells (MSCs) improves the recovery from acute kidney injury (AKI). The mechanism may involve paracrine factors promoting proliferation of surviving intrinsic epithelial cells, but these factors remain unknown. In the current study, we found that microvesicles derived from human bone marrow MSCs stimulated proliferation in vitro and conferred resistance of tubular epithelial cells to apoptosis. The biologic action of microvesicles required their CD44- and beta1-integrin-dependent incorporation into tubular cells. In vivo, microvesicles accelerated the morphologic and functional recovery of glycerol-induced AKI in SCID mice by inducing proliferation of tubular cells. The effect of microvesicles on the recovery of AKI was similar to the effect of human MSCs. RNase abolished the aforementioned effects of microvesicles in vitro and in vivo, suggesting RNA-dependent biologic effects. Microarray analysis and quantitative real time PCR of microvesicle-RNA extracts indicate that microvesicles shuttle a specific subset of cellular mRNA, such as mRNAs associated with the mesenchymal phenotype and with control of transcription, proliferation, and immunoregulation. These results suggest that microvesicles derived from MSCs may activate a proliferative program in surviving tubular cells after injury via a horizontal transfer of mRNA.

Figure 1.

Cytofluorimetric characterization of mesenchymal stem cell (MSC)-derived microvesicles (MVs). Representative FACS analyses of MVs (A) and MVs treated with RNase (B) showing the size (with 1-, 2- and −4-μm beads used as internal size standards) and the expression of CD44, CD29, α-4 integrin, α-5 integrin, CD73, α-6 integrin, and HLA-class I (thick lines) surface molecules. Dot lines indicate the isotypic controls. Ten different MV preparations were analyzed with similar results. In the CD44, CD29, α-4 integrin, α-5 integrin, and CD73 experiments, the Kolmogrov-Smirnov statistical analyses between relevant antibodies and the isotypic control was significant (P < 0.001). No significant expression of α-5-integrin and HLA class I was observed.

Figure 2.

Electron microscopy analyses of microvesicles (MVs). (A) Representative micrographs of scanning electron microscopy of purified MSC-derived MVs showing a spheroid shape (white line = 100 nm). Images were obtained by secondary electron at a working distance of 15 to 25 mm and an accelerating voltage of 20 and 30 kV. Digital acquisition and analysis were performed using the Jeol T300 system. (B–E) Representative micrographs of transmission electron microscopy obtained on purified MVs (B) and on the MSC monolayer cultured over night in the medium used for collection of MVs (see Concise Methods). C shows the release of MVs from the surface of a MSCs; D shows the presence of MVs within larger vesicles in the cell cytoplasm. E shows the extrusion of a MV from the surface of a MSC. Ultrathin sections, stained with led citrate were viewed by JEOL Jem 1010 electron microscope (black line = 500 nm).

Figure 3.

Incorporation of MVs in tubular epithelial cells (TECs). (A) Representative micrographs of internalization by TECs (30 min at 37 °C) of microvesicles (MVs) labeled with PKH26 preincubated or not with trypsin (0.5 mM) (Try); or with 100 μg/ml of sHA; or with 1 μg/ml blocking monoclonal antibody against CD44, CD29, and α4 integrin. Three experiments were performed with similar results. (B) Representative FACS analyses of internalization, after 30 min of incubation at 37 °C, by TECs of MVs labeled with PKH26 (black curves) preincubated or not with trypsin (Try) or with 100 μg/ml of sHA or with 1 μg/ml blocking monoclonal antibodies against CD44, CD29, and α4 integrin. Black curves indicate the internalization of untreated MVs. In the first panel, dot curve indicates the negative control (cells not incubated with MVs); red curve indicates the MVs treated with RNase and labeled with PKH26. In the other panels, dot curves indicate internalization of MVs after pretreatment with trypsin or incubation with blocking antibodies or sHA. Three experiments were performed with similar results.

Figure 4.

Proliferative and anti-apoptotic effects of mesenchymal stem cell (MSC)-derived microvesicles (MVs). (A) 10 μM BrdU was added to 4000 cells/well (TECs) into 96-well plates incubated for 48 h in DMEM deprived of FCS in the presence of vehicle alone or of different doses of MVs (black bars) or of RNase-treated MVs (white bars) or MVs pretreated with trypsin or with 100 μg/ml of sHA. EGF-induced (10 ng/ml) proliferation was also evaluated in TECs incubated or not with RNase-pretreated MVs (30 μg/ml). Results are expressed as mean ± SD of six different experiments. ANOVA with Newmann-Keuls multicomparison test was performed; *P < 0.05 MVs or EGF versus vehicle alone; §P < 0.05 MVs untreated versus MV treated with RNase, trypsin (Try-MV) or sHA. (B) Release of MSP and HGF by 1 × 10⁵ TECs incubated for 24 or 48 h with 30 μg/ml MVs compared with TECs incubated with vehicle alone (Ctrl). Results are expressed mean ± SD of six experiments, and t test was performed between MV treatment and control; *P < 0.05. (C) The percentage of apoptotic cells after 48-h serum deprivation (0% FCS, Ctrl) was evaluated by the TUNEL assay. TECs were incubated with vehicle alone, or with different doses of MVs (black bars) or of RNase-treated MVs (white bars), or MVs pretreated with trypsin or with 100 μg/ml of sHA. Results are expressed as mean ± SD of six different experiments. ANOVA of variance with Newmann-Keuls multicomparison test was performed; *P < 0.05 MVs versus vehicle alone; §P < 0.05 MVs treated versus MVs untreated. (D) The percentage of apoptotic cells after 48-h incubation with 100 ng/ml of vincristine or 5 μg/ml of cisplatinum was evaluated by the TUNEL assay. TECs were incubated with the apoptotic stimuli with or without 30 μg/ml MVs (black bars) or RNase-treated MVs (white bars) in the presence of 3% FCS (Ctrl = TECs incubated 48 h in the presence of 3% FCS only). Results are expressed as mean ± SD of six different experiments. Analyses of variance with Newmann-Keuls multicomparison test was performed; P < 0.05 MVs versus vehicle alone; §P < 0.05 MVs treated versus MV untreated.

Figure 5.

mRNA horizontal transfer and human protein expression in tubular epithelial cells (TECs) treated with mesenchymal stem cell (MSC)-derived microvesicles (MVs). (A) 1 × 10⁵ TECs cultured in the absence (TEC) or in the presence (TEC+MV) of 30 μg MVs for 1 and 3 h were analyzed by RT-PCR for specific human mRNA. Bands of PCR products specific for human POLR2E and SUMO-1 of the expected size (90 bp) were detected in a 4% agarose gel electrophoresis. As positive control the extract of human bone marrow-derived MSCs (BM-MSC) was used. The asterisk indicates the control without cDNA. B: Representative micrographs showing the expression of human POLR2E and SUMO-1 proteins by TECs, cultured in the absence or in the presence of 30 μg MVs for 24 and 48 h. After 24 h, POLR2E protein was detected in the cytoplasm and SUMO-1 in the cytoplasm and nuclei of TECs. After 48 h, both proteins were translocated to the nucleus. Nuclei were counterstained with Hoechst dye. Independent experiments using four different MV preparations were performed with similar results. Original magnification: ×630.

Figure 6.

Schematic representation of the protocol of glycerol induced acute kidney injury (AKI) and treatment with mesenchymal stem cells (MSCs) or MSC-derived microvesicles (MVs). Glycerol was injected intramuscularly at time 0; the arrow at day 3 indicate the administration of 75,000 MSCs; or 15 μg of MSC-derived MVs; or MSC-derived MVs treated with RNase, trypsin, or sHA; or fibroblast-derived MVs; or vehicle alone; the subsequent arrows indicate the time of sacrifice.

Figure 7.

Effects of intravenous injection of microvesicles (MVs) or mesenchymal stem cells (MSCs) into acute kidney injury (AKI) mice. Mice were given intramuscular injection of 8 ml/kg of 50% glycerol on day 0, followed by intravenous injection of MVs or RNase-treated MVs or MSCs or vehicle as control on day 3. (A and B) Creatinine and blood urea nitrogen values at the beginning of the experiments and on day 3, 5, 8, and 15 after glycerol administration. ANOVA with Dunnet's multicomparison test: *P < 0.05 MV- or MSC- treated AKI mice versus control AKI mice. (C) Representative micrographs of renal histology at day 5 and 8 after glycerol administration in control AKI mice, in AKI mice injected with 15 μg of MVs or RNase-MV or with 75,000 MSCs. Magnification: ×400.

Figure 8.

Ultrastructural changes of tubules in acute kidney injury (AKI) mice and effects of microvesicle (MV) injection. Representative electron micrographs of tubules of (A) normal control mouse, (B) AKI mouse at day 5 receiving saline, (C) AKI mouse at day 8 receiving saline, (D, E) MV-treated AKI mouse at day 5, and (F) MV-treated AKI mouse at day 8. Original magnification: A, D, F: ×3000; B, C: 2500; E: ×6000.

Figure 9.

Renal cell proliferation in acute kidney injury (AKI) mice untreated or treated with MSCs or MVs. (A) Quantification of BrdU-positive cells/high power field (hpf). BrdU was injected intraperitoneally for 2 successive days before mice being killed. (B) Quantification of PCNA-positive cells/hpf. All quantitative data were obtained from eight different mice for experimental conditions. ANOVA with Dunnet's multicomparison test: *P < 0.05 MSC- or MV-injected mice versus AKI control mice. (C) Representative micrographs of PCNA or BrdU uptake staining performed on sections of kidneys 5 and 8 d after glycerol treatment (2 and 5 d after vehicle or MSC or MV injection). Magnification: ×400. AKI, AKI untreated with MVs; AKI+MV, AKI treated with 15 μg MSC-derived MVs; AKI+RNase MV, AKI treated with 15 μg RNase inactivated MSC-derived MVs; AKI+MSC, AKI treated with 75,000 MSCs; AKI+MV sHA, AKI treated with 15 μg MSC-derived MVs preincubated with sHA; AKI+F-MV, AKI treated with 15 μg fibroblast-derived MVs.

Figure 10.

Distribution of microvesicles (MVs) after in vivo injection. (A) Quantification by spectrofluorimetric analyses of the amount of PKH26-labeled MVs, treated or not with trypsin, in different organs of acute kidney injury (AKI) and healthy mice (control+MV) as described in Concise Methods. The amount of MV-PKH26 is expressed in μg/g of dry tissue or μg/μl of plasma. (B–M): Representative confocal micrographs of frozen tissue sections of mice injected with PKH26-labeled MVs (red) and stained with vWF (B and K) or laminin (C–J, L, M) antibodies (green staining). MVs were detectable, after 1 h, within the endothelial cells of a renal vessel stained with anti-vWF antibody and within the lumen of injured tubules (B); after 3 h several tubular cells contained red MVs (C, D); at 6 h (E, F) the number of tubular cells containing red MVs was enhanced. MVs treated with trypsin were not detected in the kidney (G). In a normal control mouse, red MVs were not detected in tubular cells (H). In the lung, red MVs were not detected in the alveolar capillaries (J), whereas they were located in the endothelial cells of a large vessel (K). Liver accumulation of red MVs was detected both in normal controls (not shown) and in AKI mice (L). No signal was observed in specimens of AKI not injected with labeled MVs (I = kidney; M = liver). Nuclei were counterstained with Hoechst dye. B, C, E, G-M, original magnifications: ×400; D and F, original magnifications: ×630. G, glomeruli. Per each group (control+MV, AKI+MV, AKI+ trypsin treated MV) and each time points three animals were studied with similar results.

Figure 11.

Detection of human mRNA and human protein expression in kidneys of mice treated with human mesenchymal stem cell (MSC)-derived microvesicles (MVs). (A) Representative RT-PCR of acute kidney injury (AKI) mice untreated (AKI) or treated with 15 μg of MSC-derived MVs (AKI+MV) and sacrificed 1 and 3 h after MV injection. Bands of PCR products specific for human POLR2E of the expected size (90 bp) were detected in a 4% agarose gel electrophoresis in AKI treated with MVs but not in untreated AKI or in normal murine kidney (Ctrl). As positive control the extract of human bone marrow-derived MSCs (BM-MSC) was used. The asterisk indicates the control without cDNA. (B) Representative confocal micrographs showing the nuclear expression of human POLR2E and SUMO-1 proteins in kidney sections of AKI mice treated or not with MVs and sacrificed 48 h later. Nuclei were counterstained with Hoechst dye. Original magnification: ×400. Arrows indicate positive nuclei. The right panels of each show merge for the SUMO-1 or POLR2E staining and the nuclear staining with Hoechst. Eight animals per groups were examined with similar results.