Madonna R. et al., page 1
Transplantation of mesenchymalcell improves peripheral limb ischemia in the diabetic rats
Short title: Mesenchymal cells and tissue ischemia in diabetes
Online Supplemental Material
METHODS
Animal Care – All studies were performed with the approval of the Institutional Ethics Committee for animal research. The investigations conformed to the Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals published by the U.S.A. National Institutes of Health.
Materials – All chemicals were purchased from Sigma-Aldrich (St. Louis, MO), unless otherwise specified. Type-I collagenase was obtained from Worthington Biochemical Corporation (Lakewood, NJ).
Isolation of rat ADSCs – Under general anesthesia with penthobarbital sodium (50 mg/kg intra-peritoneally), ADSCs were isolated from inguinal fat pads (1-2 g) from Sprague-Dawley rats. The adipose tissue was mechanically minced and digested with type I collagenase [300 U/mL in phosphate-buffered saline (PBS), 1% bovine serum albumin (BSA)] for 1 h. After adipocyte removal by centrifugation at 1200 r.p.m. for 5 min, the vascular stromal fraction was plated (at 1000 cells/cm2 density)in Dulbecco's Modified Eagle Medium: Nutrient Mixture F-12 (DMEM/F12) medium supplemented with penicillin (100 U/mL), streptomycin sulfate (100 μg/mL) and 10% fetal bovine serum (FBS). After 24 h, non-adherent cells were removed. When adherent cells reached subconfluence (P0), the attached ADSCs were reseeded in the same medium at a concentration of 2 x 103 cells/cm2.For injections, a pool of unpassaged (P0) or 1st (P1) and 2nd (P2) passage ADSCs was used (with a 1:1:1 ratio). Before implantation, ADSCs were incubated with 50 μmol/L bisbenzimide at 37 °C for 15 min. Cells were resuspended in PBS, and filtered through a 75 µm nylon filter before injection into the muscular tissue.
Characterization of rat ADSCs by flow cytometry – For flow cytometry analyses of endothelial progenitor cell (EPC) markers, a total of 1 106 ADSCs were incubated for 30 min at 4 °C with 10 μL of the following fluorescence-labeled monoclonal antibodies or their respective isotype control: (1) fluorescein isothyocyanate (FITC)-conjugated mouse anti-CD34 antibody (Miltenyi Biotec, Bergisch Gladbach, Germany); (2) phycoerythrin (PE)-conjugated mouse anti-CD133 antibody (Miltenyi); 3) Pan-CD45 peridinin-chlorophyll-protein (PerCP)-conjugated antibody (detecting all isoforms and glycoforms of CD45, from BD Biosciences, San Jose, CA, USA). For flow cytometry analyses of mesenchymal stem cell and monocyte/macrophage markers cells were incubated with the following antibodies: (1) purified anti-mouse CD105 (endoglin) monoclonal antibody (Becton, Dickinson, S. Diego, CA, USA), specifically cross-reacting with (cross-specific for) the analogous rat antigen; (2) purified anti-mouse CD44 monoclonal antibody (Cedarlane Laboratories, Burlington, ON, Canada), cross-specific for the analogous rat antigen; (3) FITC-conjugated anti-human CD29 (ImmunoTools, Friesoythe, Germany); (4) FITC-conjugated anti-rat CD73 (ImmunoTools); (5) FITC-conjugated anti-human CD29 (Immuno-Tools); (6) FITC-coniugated anti-rat CD106 (Santa Cruz Biotechnology).. Cells were analyzed by flow cytometry using a FACS Calibur flow cytometer (BD Biosciences). For each sample, 30.000 events were acquired and analyzed with the CellQuest software (BD Biosciences).
Multiple low-dose streptozotocin-induced hyperglycemia–Six months-old male Sprague Dawley rats with initial body weight of 350-400 g were allowed free access to diet and water. Diabetes was induced by daily tail vein injection of STZ, 50 mg/kg dissolvedin trisodium citrate buffer: 1 mL/kg of 0.01 M, pH 4.5 (Sigma) for 6 consecutive days, usinga protocol that reduces the acute effects of STZ[1]. Blood glucose levels, which are typically stable for 8 weeks (Kunjathoor, Wilson et al. 1996), were monitored weekly over the following 30 days using an Ascensia Elite XL one-touch blood glucometer (Bayer, Leverkusen, Germany). Animals were considered diabetic when their fasting blood glucose level was > 16 mmol/l. One week after STZ administration, rats with plasma glucose concentrations of >16 mmol/l were selected as the STZ-induced diabetic group. Except for a gradual weight loss, the health of the mice was not overtly affected by thetreatment throughout the experimental period.
Unilateral Hind Limb Ischemia – Eightmonths-oldmaleSprague-Dawley rats (350-400 g) were anesthetized with a mixture of oxygen and halothane (2.5%), sodium penthobarbital (50 mg/kg intraperitoneally (i.p.)) and sodium heparin (1000 U/kg). Adequate level of anesthesia was monitored by the lack of palpebral and toe pinch reflexes. After anesthesia, unilateral hind limb ischemia was induced. The common right iliac-femoral artery, including superficial and deep branches with all arterial branches between the ligation and the distal femoral artery, were ligated with 6.0 silk suture, using the contralateral limb as a control. The overlying skin was closed with 3.0 silk suture. After the surgical procedure, animals were monitored every 24 hours for 1 week, for the absence or presence of sign of pain, distress or discomfort. One day after femoral ligation, rats (n=6 for each group) were randomly treated by multiple intramuscular (i.m.) injections (1 mL of total volume, divided into 3 aliquots injected in the adductor and 3 others in the semimembranous muscles) in the ischemic leg with: (a) allogeneic ADSCs (106-107-108 cells/mL, P0 to P2, see above); or (b) supernates from a pool of unpassaged (P0) or 1st (P1) and 2nd (P2) passage ADSC cultures (conditioned media, CM, with a 1:1:1 ratio); or (c) phosphate buffered saline (PBS), as a non-cellular control; or (d) fresh Dulbecco’s Modified Eagles Medium (DMEM):F12/1% fetal calf serum (FCS) non-conditioned medium (NCM), as a second, non-cellular control; or (e) allogeneic fibroblasts (F, 107 cells/mL) isolated from the subcutaneous tissue, as a cellular control. As sham control, all groups of animals received multiple i.m. injections of PBS (3 injections in the adductor and 3 in the semimembranous muscles) in the non-ischemic leg. Animals received standard postoperative care and returned to the laboratory 4 weeks later for magnetic resonance (MR) angiography and1H-MR spectroscopy studies. Norats died during the experimentation. All data obtained from MRI and1H-MR spectroscopy were obtained by two independent radiologists (with minimal (<10%) intra- and interobserver variability), andwere managed blindly.
Cell Engraftment Rate Determination – After 1H-MR spectroscopy and IR imaging, evaluations of the number and distribution of ADSCs in the transplanted tissues at postoperative days 1 and 30, were performed by counting blue-fluorescent, 4',6-diamidino-2-phenylindole (DAPI)-positive cells. Tissue specimens from the adductor and semimembranous muscles of the ischemic and non-ischemic limbs were frozen in Optimal Cutting Temperature compound (OCT, Miles, Elkhart, IN, USA) immediately after the procedure and stored at –80 °C. Five μm sections were obtained from each specimen, then observed and photographed at 5x and 10x magnification under a fluorescence microscope. The number of engrafted cells was analyzed by counting DAPI-stained nuclei in an average of 20 fields in each section by two independent observers and photographed.
Magnetic Resonance sessions –All animals were underwent the MR protocol four weeks after the right iliac-femoral artery ligation. Each MR session was performed according with the methods previously described(Delli Pizzi, Madonna et al. 2012). Specifically, before each MR acquisition, rats were anesthetized bythe sameprotocol used during surgery. Measurements were performed with a 3 T scanner (Philips Medical System, Best, the Netherlands), equipped with a sense flex surface coil. The animals were placed in a supine position and their hind limbs were placed between the two coil rings. Axial, coronal and sagittal images of hind limbs were obtained by T1-weighted spin-echo sequence [Repetition time/echo time (RT/ET)=742/17 ms, slice thickness 2.5 mm, matrix size 156x162, 12 slices, and Field of view (FOV)=100x100 mm]. Short Time Inversion Recovery (STIR) sequences in axial orientation were acquired [RT/ET/inversion time (IT)=3393/30/160 ms, slice thickness 2.5 mm, matrix size 148x144, 18 slices, and FOV=80x130 mm]. MR angiography (MRA) was performed with time of flight (TOF) 2D sequence: TR/TE=18/3.5 ms, slice thickness 2 mm, matrix size 150x50, 60 slices, and FOV=150x50 mm. A single voxel (12x12x15 mm3) was located on the adductor and semimembranous muscles of each hind limb and a point-resolved spectroscopy (PRESS) sequence (RT/ET=2000/50 ms, 16-step phase-cycle, averaged for 192 scans and 2048 points with 2.000 Hz spectral width) with and without chemical shift selective suppression (CHESS) water suppression was acquired.
Magnetic Resonance Imaging data assessment – All MR imaging (MRI) data were analyzed on a Philips MR work station using the same protocol previously described(Delli Pizzi, Madonna et al. 2012). MRA images were visually checked for the presence ofa signal void due to flow discontinuation distal to the right common iliac-femoral artery. T1-weighted and Short Time Inversion Recovery (STIR) axial images were used under double blinding to measure the areas of the occluded (right) and non-occluded (left) hind limb and to identify the presence of signal hyperintensity related to oedema. On T1-W images, the boundaries of each occluded and non-occluded hind limb were manually drawn. In order to assess the size difference between occluded and non-occluded limbs, quantitative “swelling index” and STIR scoring were calculated for each animal [swelling index = (right limb area - left limb area)/(right limb area + left limb area); STIR scoring = (right limb area) / (left limb area)].Visual evaluation of STIR images was performed independently by two radiologists. The “edema-like” extension detected as signal hyperintensity in STIR images, was scored on a three-grade scale(Stramare, Beltrame et al. 2010): not detectable/light: grade 0), “edema-like” extension ≤50% of the whole limb: grade 1; and “edema-like” extension >50% of the whole limb: grade 2. In order to resolve discordant STIR scoring, consensus reading agreement between the two radiologists was acquired.
Magnetic Resonance Spectroscopy analysis – All spectra were analyzed by using theAdvanced Method for Accurate, Robust and Efficient Signal fitting (AMARES) algorithm within java Magnetic Resonance User Interface (jMRUI)(Vanhamme, van den Boogaart et al. 1997). Water suppressed spectra were filtered for removal of residual water by using the Hankel Lanczos Singular Value Decomposition (HLSVD) method(Cabanes, Confort-Gouny et al. 2001). Autophasing and baseline correction were applied. From each unsuppressed spectra, the area of the water peak was calculated by the same protocol to establish a reference signal for use as an internal standard(Torriani, Thomas et al. 2005). All non-water signals were removed from the unsuppressed free-induction decays by using the HLSVD method. Finally, for each rat, the total creatine (tCr)/water value measured in the right limb were normalized respect to the tCr/water value measured in the left (non-occluded) limb.
Tissue preparation and histological evaluation of capillary density – At postoperative day 30 and after in vivo analyses (MRI), rats were anaesthesized with a mixture of oxygen and halothane (2.5%), sodium penthobarbital (50 mg/kg i.p.) and sodium heparin (1000 U/kg). The chest of each animal was opened, the heart was exposed and perfusion was started through an apical cardiac puncture with a rinsing solution containing 1 x PBS for5 min, followed by fixation with 4% paraformaldehyde and 5% sucrose in PBS for 10 minutes. The adductor and semimembranous muscles were then taken, cut into two parts, one part of each muscle was snap-frozen on dry ice for protein collection and ELISA, and the other was embedded in optimal-cutting-temperature (OCT) compound, frozen on dry ice and stored at -70°C until sectioning. Five μm-thick sections in OCT were permealized, blocked for 30 min in PBS containing 1% bovine serum albumin, and incubated with a fluorescein-conjugated anti-von Willebrand factor (vWF) monoclonal antibody (Sigma, St. Louis, MO, USA) for 1 h at 4 °C. Non-immune IgG was used as the isotype control (Becton, Dickinson). Sections were washed, mounted in glycerol mounting medium (VectaShield, Vector Labs Burlingame, CA) and viewed under a fluorescence microscope. The number of vessels were counted. Five fields from the 3 different muscle samples of each animal were randomly selected for the vessel density analysis. Data are presented as vessel number/muscle area ratio.
Immunoblotting – Total proteins from ischemic and non-ischemic skeletal muscle tissues were isolated in ice-cold radioimmunoprecipitation assay (RIPA) buffer, separated under reducing conditions, and electroblotted to polyvinylidene fluoride (PVDF) membranes (Immobilon-P, Millipore, Bedford, MA, USA). Western analysis of α-sarcomeric actinin, cycloxygenase (COX)-2 and aquaporin (AQP)-1 was performed by incubating the membranes overnight at 4 °C with the following primary antibodies: (1) monoclonal rabbit anti-α-sarcomeric actinin antibody (Sigma); (2) monoclonal mouse anti-COX-2 (Santa Cruz); (3) monoclonal mouse anti-AQP1 (Santa Cruz). Blots were incubated with horseradish peroxidase-coupled secondary antibodies, washed and developed by using a SuperSignal West Pico Chemiluminescent Substrate Kit (Pierce, Rockford, IL, USA). The intensity of each immunoreactive protein band was quantified by densitometric analysis. To verify equal loading of proteins, membranes were stripped and re-probed with a monoclonal anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibody (Sigma).
Enzyme-Linked Immunosorbent Assay for rat HGF and rat VEGF– Supernatants were collected from ADSC cultures at passage 0-2 (conditioned media, CM); or from fibroblasts; or from the stromal vascular fraction at 48 h after final change of fresh medium (DMEM:F12/1% fetal calf serum (FCS)). Concentrationsof rat hepatocyte growth factor (HGF) or rat vascular endothelial growth factor (VEGF) in the cellular extracts isolated from ADSCs (passage 0-2) and fibroblasts, or in the supernatants, or in the frozen tissues (the adductor and semimembranous muscles from non-ischemic or ischemic limbs treated with PBS, CM, ADSCs and fibroblasts, limited to HGF), were determinedby enzyme-linked immunosorbent assays (rat HGF ELISA kit, B-Bridge International Inc., Cupertino CA, USA); quantikine rat VEGF Immunoassay, R&D System, Minneapolis, MN, USA) according to manufacturers’instructions. A standard curve was prepared from 7 HGF dilutions or 9 VEGF dilutions. The lower limits of sensitivity were 0.3 ng/mL for the HGF ELISA and 8.4 pg/mL for the VEGF ELISA. The overall inter-assay and intra-assay coefficients of variation were <10%. In each well, total cellular proteins were measured with a bicinchoninic acid (BCA) assay (Pierce, Rockford, IL, USA). The HGF and VEGF content was normalized to cell protein values.
Statistical Analyses– All results are presented as meansS.D. Differences between groups of treatments and differences in dose-responses at individual ADSC concentrations were analyzed with the Kruskal-Wallis test. When this test demonstrated a significant effect, post-hoc analysis was performed with the non-parametric Student-Newman-Keuls and Dunn tests. Pearson’s correlation investigated the potential relation between the T1-W and between 1H-MR spectroscopy and histological outcomes (sarcomeric α-actinin, capillary density), respectively. Spearman's correlation described the relation between Short Time Inversion Recovery(STIR) “edema-like” score and histological outcomes. All authors had full access to the data and take full responsibility for their integrity. All authors read and agreed to the final manuscript as written.
REFERENCES
Cabanes, E., S. Confort-Gouny, et al. (2001). "Optimization of residual water signal removal by HLSVD on simulated short echo time proton MR spectra of the human brain." J Magn Reson 150(2): 116-25.
Delli Pizzi, S., R. Madonna, et al. (2012). "MR angiography, MR imaging and proton MR spectroscopy in-vivo assessment of skeletal muscle ischemia in diabetic rats." PLoS One 7(9): e44752.
Kunjathoor, V. V., D. L. Wilson, et al. (1996). "Increased atherosclerosis in streptozotocin-induced diabetic mice." J Clin Invest 97(7): 1767-73.
Stramare, R., V. Beltrame, et al. (2010). "MRI in the assessment of muscular pathology: a comparison between limb-girdle muscular dystrophies, hyaline body myopathies and myotonic dystrophies." Radiol Med 115(4): 585-99.
Torriani, M., B. J. Thomas, et al. (2005). "Intramyocellular lipid quantification: repeatability with 1H MR spectroscopy." Radiology 236(2): 609-14.
Vanhamme, L., A. van den Boogaart, et al. (1997). "Improved method for accurate and efficient quantification of MRS data with use of prior knowledge." J Magn Reson 129(1): 35-43.
LEGEND TO FIGURES
Online Supplement Figure 1. Representative MR Angiography (MRA) from a rat 4 weeks after femoral artery ligation. MRA shows a signal void due to discontinuation of flow in the right common iliac-femoral artery.
Online Supplement Figure 12. Short Time Inversion Recovery (STIR) of rat hind limb ischemia treated with ADSCs, conditioned medium and controls.
Representative axial STIR images show an “edema-like” hyperintense signal (indicated by the white arrows) on the occluded (right) muscle in both non-diabetic (panel A) and diabetic (panel B) rats treated with multiple injections of phosphate buffered saline (PBS, 1 mL), fibroblasts (F, 107 cells/mL; panel C), non-conditioned medium (NCM, 1 mL; panel D), adipose tissue-derived mesenchymal stromal cells (ADSCs, 106-107-108 cells/mL; panels E-J) or ADSC conditioned medium (CM, 1 mL; panels L and M). Axial images show a hyperintense T2 signal on the right muscle related to ischemic damage of soft-tissue in occluded limbs. The signal appears more pronounced in the non-occluded and occluded limbs in diabetic rats treated with PBS, F or NCM (panels A-D).
Online Supplement Figure 3: AQP1 and COX-2 expression in rat hind limb ischemia treated with ADSCs, conditioned medium and controls.
Representative Western blot (upper panels) and densitometric quantification (data normalized for beta-actin, lower panels) of aquaporin (AQP)1 and cycloxygenase (COX)-2 from normal limbs in nondiabetic rats treated with multiple injections of phosphate buffered saline (PBS, 1 mL) or nonoccluded limbs in diabetic rats treated multiple injections of PBS (1 ml) or occluded limbs with femoral artery ligation in diabetic rats treated multiple injections of PBS (1 ml) (panel A and B), ADSCs (C3: 108 cells/ml), conditioned medium (CM, 1 ml), non-conditioned medium (NCM, 1 ml), fibroblasts (F, 107 cells/mL) (panel B). Values are mean±SD, with n=6 for each treatment group. §§ and §: P<0.01 and P<0.05, respectively, vs nondiabetic non-occluded hind limb; °° and °: P<0.01 and P<0.05, respectively, vs diabetic nonoccluded hind limb; **: P<0.05 vs diabetic occluded hind limb treated with PBS.