Enhanced Cardiomyogenic Lineage Differentiation of Adult Bone Marrow Derived Stem Cells grown on Cardiogel.
Sreejit P, R S Verma *
Electronic Supplementary Material - Materials and Methods
FACS analysis
Cell surface antigens on Bone marrow derived stem cells (BMSC) were analyzed by FACS as previously described (Sreejit, et al., 2012). Cells, after being washed with DPBS containing 5 mM EDTA, were trypsinized (0.05% trypsin) for 5 min, centrifuged (400g) for 5 min at room temperature, resuspended in culture medium, washed with DPBS, fixed in 4% paraformaldehyde (PFA; Himedia) in DPBS for 0.5 h at room temperature, washed again twice with DPBS, and blocked in 2% FCS in DPBS for 0.5 h at room temperature with agitation. Cells were washed again twice with DPBS, and ∼1×105 cells were then incubated with antibodies (raised in rat) against CD markers (CD 29, 34, 44, 45, 90, 105, 106) and Sca-1 (BD Biosciences) for 90 min at room temperature with intermittent shaking. After incubation, cells were washed twice with DPBS and incubated with fluorescein isothiocyanate (FITC)-conjugated anti-rat antibodies and isotype antibodies (BD Biosciences) for 60 min at room temperature in the dark with intermittent shaking. The antibodies used for immunophenotyping are described in Electronic Supplementary Material, Table S1. After two more DPBS washes, the cells were stored in 0.4% PFA at 4°C. FACS analysis was performed on a BD FACS Calibur flow cytometer (BD Biosciences) by using CellQuest software (BD Biosciences) with 10,000 events being recorded for each sample. Appropriate Isotype antibodies were also used as controls.
Real Time PCR (qTR-PCR)
Real-time PCR was carried out using the Quantitech® SYBR® Green PCR Kit (Qiagen, Germany) and various gene specific primers (Electronic Supplementary Material, Table S2). For normalization of various gene expression, β-Actin abundances were measured using the following primer pairs: β-Actin forward, 5’-TTC CTT CTT GGG TAT GGA AT-3’ and β-Actin reverse, 5′- GAG CAA TGA TCT TGA TCT TC -3′. The relative gene expression levels were determined by calculating the 2(−ΔΔCt) values (MacRae, et al., 2006).
1 μl of cDNA preparation per 10 μl reaction and Quantitech® SYBR® Green PCR Kit in an Eppendorf Mastercycler® ep realplex Thermal Cycler were used. Specific primers to amplify specific regions of the mouse genes were used. For normalization of the endogenous gene expression, a 206 bp β-Actin (as a house-keeping control) were amplified under identical conditions. The PCR program used for the real-time reaction was: 95°C for 15 min; 94°C for 15 sec, 54°C & 60°C for 20 sec, 72°C for 20 sec for 40 cycles as per the manufacturer’s instruction.
Induction of multilineage differentiation
Osteogenic differentiation
After primary culture in BMM, the BMSC were plated onto 12-well plates, with and without cardiogel pre-coating, at a density of 104 cells per well and incubated further. On attaining 70% confluency, the media were then replaced with osteogenic medium: BMM supplemented with 500 μM L-ascorbic acid (Sigma), 0.1 μM dexamethasone (Sigma) and 10 mM β-glycerophosphate (β-glycerophosphate disodium pentahydrate) (Sigma). The media were changed every 3 days (Kang, et al., 2005, Kretlow, et al., 2008, Peister, et al., 2004). Osteogenic differentiation was assessed by von Kossa, Fast violet staining, Alizarin Red S (ARS) staining and osteopontin expression using immunocytochemistry, 2 weeks after initial osteogenic induction and compared with control and adipogenesis induced cells grown on normal and cardiogel coated plates.
Adipogenic differentiation
Immortalized BMSCs were harvested and seeded in 12-well cell culture plates with and without cardiogel pre-coating. On attaining 70% confluency, the cells were treated with adipogenic differentiation medium, respectively. Adipogenic differentiation medium: BMM supplemented with 1 μM dexamethasone (Sigma), 100 μM indomethacin (Sigma) and 500 μM 1-methyl-3-isobutylxamthine (Sigma) (Kang, et al., 2005, Kretlow, et al., 2008, Peister, et al., 2004). 5 μg/mL bovine insulin (USV) was added in addition along with adipogenic differentiation medium for the initial induction. The media was changed every 4th d., with a single day of no treatment after the 2nd treatment. Adipogenic differentiation was assessed by Oil Red O (ORO) staining and Nile Red staining, 2 weeks after initial adipogenic induction and compared with control and osteogenesis induced cells grown on normal and cardiogel coated plates.
The differentiation experiments were repeated thrice in 12 well plates with different passage numbers.
Microscopy
The induced and control (uninduced) BMSCs were observed under bright field and phase contrast microscopy at various magnifications in separate microscopes (Nikon Ts 1000 and Nikon TiE). Photographs were recorded and analyzed further. Immunocytochemistry and Histochemical analysis data were also recorded and analyzed likewise in fluorescent microscope (Nikon TiE).
Immunocytochemistry
ICC was performed as described in the manuscript using primary and secondary antibodies outlined in Supplementary Table S1.
Histochemical analysis
Fast violet staining
To detect osteogenic differentiation following induction, fast violet staining was performed as previously reported to detect alkaline phosphatase (ALP) produced by osteoblasts with some modifications (Hoshiba, et al., 2009). ALP assay kit (Sigma) was used for the staining. After culturing for 14 days with medium changed every 3 days, cells were fixed in 4% paraformaldehyde (Himedia) for 5 mins at RT and washed with MQ H2O. The cells were incubated with a mixture of naphthol AS-BI alkaline solution with fast red violet LB at RT for 30 mins in dark and washed with MQ H2O for 2 mins. The resulting red insoluble deposit indicates site of ALP activity, which were then photographed.
von Kossa staining
To demonstrate the presence of mineral deposits containing salts of calcium following osteogenic differentiation, von Kossa (VK) staining was performed with minor variations (Ogawa, et al., 2004). The cells were rinsed with phosphate-buffered saline and fixed in 4% paraformaldehyde for 1 h. They were then incubated in 2% silver nitrate (Qualigens) for 30 min in the dark, after which they were washed thrice with MQ H2O. The plates were then exposed to light for 15 min on a white background and then dehydrated with 100% ethanol before microscopical observation. Secretion of calcified extracellular matrix was confirmed as black nodules with VK staining.
Alizarin Red S staining and quantification
Assessment of in vitro mineralization was performed using alizarin red S (AR-S) (Himedia) staining on BMSCs cultured in various conditions (with and without treatment) for 14 days (Stanford, et al., 1995). Cells were briefly rinsed with PBS followed by fixation with ice-cold 70% ethanol (Merck) for 1 h. Following MQ H2O rinsing, the cells were stained for 10 min with 50 mM AR-S (pH 4.2) at RT. Cultures were then rinsed five times with MQ H2O succeeded by a 15-min wash with PBS with rotation to reduce nonspecific staining following which the stained cells were photographed. A quantitative destaining procedure was performed using 10% (w/v) cetylpyridinium chloride (CPC) (Himedia) in 10 mM sodium phosphate (pH 7) for 15 min at RT which were then diluted 10-fold in 10% CPC solution. AR-S concentration was determined by absorbance measurement at 560 nm on a multiplate reader (Molecular Devices) using an AR-S standard curve in the same solution. Values were normalized to total protein levels determined with a BCA assay kit (Pierce), performed in micro titer plates as per manufacturer's instructions.
Oil Red O staining and quantification
To detect adipogenic differentiation following induction, Oil Red O (ORO) staining was performed as previously reported to monitor lipid accumulation with slight modifications (Ramírez-Zacarías, et al., 1992). 0.3 g of Oil Red O (Sigma) was dissolved in 100 ml of isopropanol (Merck) and left overnight at RT. The solution was filtered from the formed precipitate. 40 ml of MQ H2O was added to 60 ml of that solution (to make 0.1% working solution) and left overnight at 4⁰C. The working solution was filtered twice before addition to cells. Adherent cells were rinsed with PBS and fixed in 4% paraformaldehyde for 1 h at RT, followed by rinsing with DPBS once and with MQ H2O twice. After the rinsing, staining with 0.1% Oil Red O was performed for 10 mins at RT. Plates were washed with 60% isopropanol to eliminate non-specific staining and rinsed thrice with MQ H2O. Photographs of the cells on plate were taken in MQ H2O. For quantification, the dye was extracted from dried wells with 100% isopropanol for 10 mins at RT and absorbance was determined at 520 nm.
Nile Red staining
Nile Red (NR) staining was performed as previously reported to monitor lipid accumulation following adipogenic differentiation (Szabo, et al., 2008). The dye stock solution of 0.5 mg/ml in acetone was diluted (1:500) in DPBS (0.05 ml of Nile red stock solution in 1 ml of DPBS). Adherent cells were rinsed with PBS and fixed in 4% paraformaldehyde for 20 mins at RT, followed by rinsing with DPBS twice. After the rinsing, staining with Nile Red working solution was performed for 30 mins at RT. Plates were washed with DPBS twice to eliminate non-specific staining. Photographs of the cells on plate were taken in DPBS. Hoechst 33342 (Sigma) was added to the working solution for nuclear counterstaining.
Sirius red staining
Sirius red staining was done to detect the presence of collagen fibers in the cardiogel with slight modifications in counter staining (Junqueira, et al., 1979, Kiernan, 1999, Whittaker, et al., 1994). For Sirius red stain, 0.1% (w/v) of Sirius red F3B was dissolved in saturated aqueous solution of picric acid. Acidified water was prepared by adding 1 volume acetic acid (glacial) to 2000 liter of MQ water. Iron haematoxylin was prepared by mixing 2 volumes of 5% ferric chloride in water with 5 volumes of 5% haematoxylin. 1 volume of conc. hydrochloric acid was added to 240 volumes of MQ water carefully which, was then mixed with 560 volumes of ethanol, to obtain acid alcohol. Unfixed coverslips with cardiogel and with BMSC were washed with PBS and stained with sirius red stain for two hour. The stain was washed off in two changes of acidified water and counterstained with freshly prepared iron haematoxylin for 10 mins. After decanting haematoxylin stain, coverslips were washed with acidified alcohol for 5 mins. Coverslips were dried and observed, under normal and differential interference contrast microscopy, and photographs were recorded.
Scanning Electron Microscopy (SEM)
Differentiation of BMSC was further validated using SEM (Li, et al., 2005). A modified protocol developed in house was developed for sample preparation and used in the present study (Sreejit and Verma, 2011). BMSC were seeded on glass cover slips coated with cardiogel and allowed to grow and differentiate as per protocols described. After attaining ~70% confluency, the cover slips were chemically fixed with 2.5% Glutaraldehyde at room temperature and then serially dehydrated in ethanol. The surface of the cover slip was sputter coated in a vacuum with an electrically conductive 5 nm thick layer of Gold – Palladium alloy using Precession Etching Coating system (Gatan Model 682). SEM images were then recorded with a High Resolution Scanning Electron Microscope (FEI Quanta 200).
Atomic Force Microscopy
The surface morphologies on cardiogel and its effect on BMSC were observed under atomic force microscope (AFM) (Song, et al., 2009). BMSC were seeded on glass cover slips coated with cardiogel and allowed to grow. After attaining ~70% confluency, the cover slips were fixed with 2.5% Glutaraldehyde at room temperature. Scanning probe microscopy was then done with AFM in tapping and phase mode on fixed BMSC on cardiogel.
DiI-Hoechst 33342 staining
The efficacy of BMSC to differentiate into cardiomyogenic lineage and generate cardiomyocytes like cells was assessed by enumerating the number of multinucleated cells using a modified staining protocol (Sahara, et al., 2010). DiI stain and Hoechst 33342 nuclear counter stain were added to obtain final concentration of 5 µM and 10 µM respectively to the media in the wells containing putative BMSC cell lines induced with 5-azacytidine grown on control and cardiogel coated plates for 30 days. After 12 hrs, the stains along with media were decanted and replenished with fresh media. The wells were observed under fluorescent microscope and photographs of 5 fields per well were taken. The percentage of multinucleated cells in the total cells was calculated to provide the multinucleation ratio, which was used to determine the efficacy of cardiomyogenic differentiation.
Electronic Supplementary Material - Results and Discussion
FACS analysis
The isolated BMSCs were characterized by surface marker analysis using FACS (Electronic Supplementary Material, Fig. S1). Isotype antibodies were always found to be absent in cells. The MSC-specific markers CD 29, CD 44, and Sca-1 were consistently expressed, whereas CD 90, 105 and 106 were partially expressed. The hematopoietic marker CD 34 and 45 was absent from the cell surface.
qRT-PCR analysis
To compare and quantify the gene expression profile of BMSC following cardiac specific induction (5-azaC treatment) and growth on cardiogel and validate the data obtained from semi-quantitative RT-PCR, qRT-PCR was performed (Fig. 2 b-g). β-Actin was used as a house-keeping control for normalization of the endogenous gene expression. Various genes were used to compare the expression profile in normal and differentiation induced BMSC. The increased gene expression clearly validates the data from semi-quantitative RT-PCR and ICC. The enhanced expression of cardiac specific transcription factors GATA-4 and MEF-2C suggests lineage specific differentiation of BMSC into cardiomyocyte like cells. The increased gene expression clearly validates the data from semi-quantitative RT-PCR. The enhanced expression of gap junction molecules Connexin 43 and 45 which are responsible for the functional characteristics, suggests lineage specific differentiation of BMSC into cardiomyocyte like cells.
BNP (Brain Natriuretic Peptide) was found to be homogenously expressed in both untreated and treated BMSC grown on normal (gelatin coated) and cardiogel coated plates. In contrast, enhanced expression of cardiac specific structural molecule α-cardiac actin, suggests lineage specific differentiation of BMSC into cardiomyocyte like cells. The increased gene expression clearly validates the data from ICC for α-Sarcomeric Actin, another structural component of cardiomyocytes.
Osteogenic induction
To assess the osteogenic potential of BMSCs, induction was performed for 14 days of culture in standard osteogenic media (Kretlow, et al., 2008). A large number of BMSC exhibited ALP positive aggregates, VK stain-positive nodules and AR-S staining, as shown in Electronic Supplementary Material, Fig. S2, suggesting osteogenic differentiation of mouse BMSCs into osteocytes. SEM analysis further demonstrated the presence of nodules and mineral deposits, thus validating the differentiation (Li, et al., 2005). However, untreated control cells showed very low levels of AR-S staining and hardly any ALP positive aggregates and VK stain-positive nodules. Similar observations were made on untreated control cells grown on cardiogel. However, induction of BMSC grown on cardiogel hardly showed any differentiation. This was further validated by the use of AR-S staining to quantify the calcium levels in mineral deposits formed following induction (Electronic Supplementary Material, Fig. S2). The extracellular calcium levels were normalized with total protein levels. Calcium concentration increased in treated cells grown on normal plates by as much as 9 fold when compared to uninduced plates (Fig. 4 g). But in cells grown on cardiogel following induction, there was a significant decrease in calcium levels when compared to normal plates (* = p< 0.005). Calcium levels in BMSC grown on normal plates following adipogenic induction, was found to be even lower than normal uninduced plates. This clearly suggests that following adipogenic induction, BMSC differentiated into adipogenic lineage and produced lower amounts of extracellular calcium in comparison with uninduced cells, where BMSC did not differentiate into any particular lineage due to absence of specific induction. Sample to sample variability was observed in calcium levels following induction.