Low-frequency, low-magnitude vibrations (LFLM) enhances chondrogenic differentiation potential of human adipose derived mesenchymal stromal stem cells (hASCs)

Description of the cell vibration generator prototype

The process of inducing vibrations was applied using custom-made vibration platforms, specially constructed device that allowed to induce mechanical motion of a 24-well culture plate. Movement of the plate was characterized by the harmonic sine of a given amplitude and frequency. The direction of plate translations was perpendicular to the main surface on which the cells were cultured.

A scheme of the stand is shown in Fig. 1C and photographs in Figs. 1A and 1B. The vibration generator, an electromagnetic actuator, is positioned on a stationary base. The principle of its operation is based on coil movement, which is generated by an alternating current flow. The design of the actuator is similar to that of a typical loudspeaker, with the main difference being that the moving parts are not a flexible membrane but a stiff plate.

The stiff plate of the actuator moves in a linear manner in relation to the stationary part. Fig. 1C depicts the displacement value as ‘x’. Between the actuator and the culture plate, a spacer has been mounted. The spacer is a rigid element made of polyethylene, placed to create distance from the cell culture actuator. This was done to eliminate the possible influence of the alternating magnetic field generated by the EM-actuator on the cell culture. The height of the spacer was about 10 cm. The strength of the magnetic field at this distance does not differ from the background.

The culture plate was attached to the top surface of the spacer in such a way to allow quick mounting. Such a method was dictated by the fact that the vibration stimulation was scheduled only for short periods of time each day. Movement of the culture plate was defined as a course of the sine function with a given value of frequency and amplitude of acceleration. A laser displacement sensor (KEYENCE LK-G157) was used to measure the translation of the culture plate. The acceleration signal was calculated according to the following formula: ${\displaystyle x=Asin\left(\omega t\right)\to \ddot{x}=-A{\omega }^{2}sin\left(\omega t\right)}$ where x—displacement, $\ddot{x}$—acceleration, A—amplitude of displacement, ω—frequency of vibrations (ω = 2πf), f—frequency, t—time, Aω²—amplitude of acceleration.

Vibration loading protocol of hASCs culture

The hASCs were seeded at a concentration of 3 × 10⁴ on 24-well plates and 5 × 10⁴ to a 15 ml tube. Tubes with 3D model were vibrated on tube rack. For each vibration model (25, 35 and 45 Hz) separate dishes were used. Plates/racks were placed securely onto the vibration device and oscillated vertically at 25, 35 and 45 Hz. The stimulus was sinusoidal and delivered with a peak acceleration of 0,3 g for 15 min once a day, for 14 consecutive days. Cells in the non-vibration group were placed on the same but stationary plate. After 15 min of vibration, the hASCs (both vibrated and non-vibrated groups) received fresh culture medium.

Isolation of human adipose derived mesenchymal stem cells (hASCs)

This study was approved by the local bioethics committee of Wroclaw Medical University, Poland (number KB-177/2014). Written informed consent was obtained from each patient prior to tissue collection during total hip arthroplasty. This study adhered to the Helsinki Declaration (1964) and its later amendments.

Subcutaneous adipose tissue was collected from 4 patients. From each patient we obtained 4 donor samples, representing of a total n = 16. The average age of the patients was 69 ± 1 years. Briefly, after collection, the tissue samples were placed in sterile Hank’s Balanced Salt Solution (HBSS). The isolation procedure of adipose-derived mesenchymal stem cells was conducted under aseptic conditions and in accordance with previously described protocol (Grzesiak et al., 2011; Marycz et al., 2013). Samples were washed with HBSS supplemented with a 1% antibiotic-antimicotic solution (penicillin/streptomycin/amphotericin b at a concentration of 0.017 mol/l, 0.01 mol/l and 0.0002 mol/l respectively; Sigma Aldrich, cat no A5955) and then cut into small pieces using surgical scissors. Next, the samples were placed in a sterile centrifuge tube and digested with type I collagenase (1 mg/ml, Sigma Aldrich, cat no C5894). After 30 min incubation at 37°C, the tissue homogenate was centrifuged at 1,200 g for 10 min. The supernatant was removed and the cell pellet was resuspended in growth media. The cell suspension was then transferred to the cell culture flask.

Immunophenotyping, Fluorescence-activated cell sorting (FACs) analysis, and multipotency test

Cells were plated on 24-well culture plates suspended in 500 µl of standard medium at a concentration of 8 × 10³ cells per well. The presence of specific antigens for ASCs, i.e., integrin beta-1 (CD29), HCAM (CD44), 5′-nucleotidase (CD73) and endoglin (CD105) and leukocyte common antigen (CD45) was examined after one week of culture by means of primary antibodies (all from Sigma Aldrich). Negative staining of CD45 was used to exclude hematopoietic origin. After fixation, cells were permeabilized with 0.2% Tween 20 for 15 min and washed three times with HBSS. The solution of primary antibody and 4% FBS in PBS was applied to every well and incubated overnight at 4°C. Next, the cells were washed three times and secondary goat anti-rabbit conjugated with Atto 488 antibody was added to appropriate wells at concentration 1:150. After incubation at room temperature for 1.5 h, the cells were washed again and photographed under a fluorescence microscope.

For the multipotency test, cells were cultured on chondrogenic and adipogenic media (STEMPRO^® Chondrogenesis/Osteogenesis Differentiation Kit and STEMPRO^® Adipogenesis Differentiation Kit, Life Technologies) for 14 days. The culture media was changed every second day.

After 3 passages, the ASCs were examined for surface protein molecule expression by flow cytometry. Cells were trypinized using a Trypsin-EDTA solution (TrypLE™, Life Technologies), centrifuged at 400 xg for 3 min, and then washed with PBS containing 2% FBS (fetal bovine serum) (Sigma Aldrich). A total of 5 × 10⁵ cells were labeled for 20 min (on ice and dark) with antibodies pre-conjugated with allophycocyanin (APC), peridinin chlorphyllprotein (PerCP), fluorescein isothiocyanate (FITC) or phycoerythrin (PE). The following CD surface markers were tested: CD34, CD45, CD105, CD90, CD73, CD44, CD29 and IgG1 as an isotype control antibody (BD Pharmingen). The samples were analyzed by a Becton Dickinson FACSCalibur flow cytometer. At least ten thousand events were acquired for each CD surface marker. The data was then analyzed using FlowJo X software (Treestar).

Cell culture

Throughout the experiment, hASCs were cultured in aseptic and constant conditions in an incubator at 37°C, 5% CO₂ and 95% humidity. The cell population was plated in T-75 culture flasks for primary culture and was maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) with nutrient F-12 Ham (Sigma Aldrich) supplemented with 10% FBS and 1% of antibiotic/antimicotic solution at a concentration of 0,017 mol/l, 0,01 mol/l and 0,0002 mol/l respectively (Sigma Aldrich, cat no A5955). The culture medium was changed every second day. Human ASCs were passaged using Trypsin-EDTA solution (TrypLE™, Life Technologies) in accordance with manufacturer’s instruction after reaching about 80–90% confluence. Cells were passaged three times before use in experiments.

Isolated adipose-derived mesenchymal stem cells were divided into 2 groups. The first one was stimulated with chondrogenic medium and the second one with adipogenic medium. The differentiation processes of hASCs were performed using the STEMPRO^® Chondrogenesis Differentiation Kit and STEMPRO^® Adipogenesis Differentiation Kit (Life Technologies), respectively. Cells were seeded at 3 × 10⁴ on 24-well plates and 5 × 10⁴ to a 15 ml tube. The stimulation of cells was performed in accordance with the manufacturers’ instruction.

The chondrogenic culture was maintained in two systems: 2D in 24-well plates for fluorescent, histochemical stainnings, rtPCR analysis and SEM, and 3D for ELISA tests and Focused Ion Beam Scanning Electron Microscope (FIB-SEM, Auriga Compact Crossbeam, Zeiss, Germany). For the 3D system, 2.5 × 10⁵ hASCs were seeded into 15 ml polypropylene tubes and pelleted. The hASCs were cultured for 14 days as 3D pellets in induction medium STEMPRO^® Chondrogenesis Differentiation Kit (Life Technologies). The experiment was repeated three times. Both 2D and 3D cultures were incubated two days before starting vibration stimulation.

Cell proliferation assay

The cell proliferation factor (PF) was evaluated using the Alamar Blue test (TOX-8, Sigma Aldrich) according to the manufacturer’s instructions. The culture media was replaced with a medium containing 10% of resazurin-based dye and incubated for two hours. Afterwards, the supernatants were collected and subjected to absorbance measurement by means of spectrophotometer (SPECTRO StarNano, BMG Labtech) at 600 nm of wavelength, with a distraction of 690 nm of background absorbance. The procedure was performed during the differentiation period at days 2, 5, 10, and 14.

A standard curve obtained during the experiment, allowed to estimate the amount of cells. The population doubling time (PDT) was assessed using an online calculator (http://www.doubling-time.com/compute.php).

Examination of hASCs morphology

Cell morphology, cellular composition, and culture growth pattern were analyzed using an inverted, fluorescence microscope (AxioObserverA1, Zeiss) and a scanning electron microscope (SEM; EVO LS15, Zeiss).

In order to begin observations, after the culture period, cells were fixed with 4% paraformaldehyde. After 15-min permeabilization with 0.2% Tween, cells were stained using atto-565-labeled phalloidin for 40 min to visualize actin filaments. After triple washing, diamidino-2-phenylindole (DAPI) staining was applied for 5 min to analyze the distribution of cell nuclei. All fluorescence staining was performed at room temperature in the dark. Additionally, Oil Red O staining for adipogenic cultures and Safranin-O staining for chondrogenic cultures were performed to observe fat droplets and chondrocytes, respectively. After fixation, adipogenic samples were treated with 0.1% Oil Red O solution for 5 min, Plates were rinsed 3×with water and images of cells on plate were taken. For quantification the percentage of Oil Red O absorption, the dye was extracted by isopropanol and absorbance was determined at 490 nm. Chondrogenic samples were treated with 1% acetic acid for 10 min and stained with Safranin-O for 5 min. Images were acquired using a Cannon PowerShot digital camera. In order to evaluate the chondrogenic differentiation efficiency, the concentration of proteoglycans was determined, basing on the binding Safranin O to glycosaminoglycans and spectrophotometric measurements at a 470 nm wavelength and assessment percentage of Safranin O absorption. As 100% control, we adopted reagent not added to culture. To analyze detailed morphological features of the cells, especially fat droplets, and chondrogenic nodules, SEM was performed. After fixation, the cells were washed in distilled water and dehydrated in ethanol (concentrations from 50 to 100%, every 5 min). Thoroughly dried cells were coated with gold (ScanCoat 6, Oxford), placed in a microscope chamber, and observed using the SE1 detector, at 10 kV of filament’s tension. To observe morphological features and measure diameters of nodules (n = 6) Ion Beam Scanning Electron Microscope (FIB-SEM, Auriga Compact Crossbeam, Zeiss, Germany) obsrvations were performed. Analysis was performed using a focus ion beam detector at magnification of 200X. Diameter of adipocytes (n = 6) was measured using Scanning Electron Microscope (SEM; EVO LS15, Zeiss).

Enzyme-linked immunosorbent assays

In order to evaluate the chondrogenic differentiation efficiency on the protein level, the concentration of chondrogenesis-specific markers was investigated. The total concentration of proteins from pellet cultures was determined with enzyme-linked immunosorbent assay (ELISA). For the analysis, cells homogenate were collected on the last day of the experiment. Chondrogenic media was subjected to a BMP-2 ELISA assay (Bone Morphogenic Protein 2 Quantikine ELISA Kit, R & D Systems), and a Col-1 and a Col-2 ELISA assay (Human Collagen alpha-1(I) and (II) chain ELISA Kit, EIAab). All steps of each ELISA tests were performed in accordance with the manufacturer’s protocol. Each sample was prepared in duplicate. Spectrofotometric determination was performed using a microplate reader (Spectrostar Nano, BMG Labtech) at a wavelength equal to 450 nm and with the correction wavelength of 540 nm. The concentration of proteins was presented as a ratio of protein weight and supernatant volume (w/v).

Quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR)

In order to analyze gene expression, cells after stimulation were rinsed with HBSS and were homogenized with 0.5 ml of TRI Reagent (Sigma Aldrich) directly in the culture well. The total RNA was isolated using a phenol-chloroform method as previously described (Chomczynski & Sacchi, 1987). After isolation, total RNA was diluted in DEPC-treated water. The concentration and purity of RNA preparations was determined by absorbance measured at 260 nm with a nanospectrophotometer (VPA biowave II). Preparation of DNA-free RNA was performed using DNase I RNase-free kit (Thermo Scientific). For each reaction, 100 ng of total RNA was used. Transcription of gDNA-free total RNA to a complementary DNA (cDNA) was reverse transcribed using Moloney Murine Leukemia Virus Reverse Transcriptase (M-MLV RT) and oligo(dT)15 primers (Novazym). RNA purification and cDNA transcription was performed according to manufacturers’ instructions. Quantitative Real-Time polymerase chain reaction (qRT-PCR) was performed using 5 µl of cDNA in total volume of 20 µl by means of SensiFast SYBR & Fluorescein Kit (Bioline). The reaction was performed at a 500 nM final concentration of primers. The primer sequences used are presented in Table 1. qRT-PCR was performed as described previously (Kim & Im, 2010). Expression levels of all analyzed genes were normalized for the expression level of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a housekeeping gene.

Statistical analysis

All experiments were performed with at least 3 (n = 3) independent experiments (biological replicates, n ≥ 4) measured as quadriplicate or more (technical replicates, n ≥ 4).

Statistical analysis was performed using GraphPad Prism 5 software. The statistical significance of results was calculated using the one-way analysis of variance (ANOVA) with post-hoc Dunnett’s test by means. A P-value of less than 0.05 was considered statistically significant.

hASCs—FACs analysis, immunophenotyping and multipotency test

Flow cytometry analysis revealed that hASCs showed positive labeling for CD29, CD44, CD73, CD105, and CD90 (Fig. 2B). The investigated cells were negatively labeled for two hematopoietic markers: CD34 and CD45 (Fig. 2B). Additionally, immunohistochemical staining confirmed the presence of mesenchymal markers (CD29, CD44, CD73, CD105) and excluded hematopoietic origin (CD45) (Fig. 2A).

Moreover, the multipotent character of hASC’s was confirmed by abundant osteogenic and adipogenic differentiation. In contrast to cells cultured under control conditions, the presence of osteo nodules, was observed in hASCs cultivated under osteogenic conditions (Fig. 2C). Moreover, mineral calcium deposits visualized by Alizarin Red staining were clearly detected after 3 weeks of osteogenic differentiation. After 2 weeks of culture with an adipogenic inducing media, hASCs developed Oil Red O positive lipid droplets (Fig. 2C), whereas control cultures grown in standard media failed to produce similar results (Fig. 2C).

Proliferation rate (PF) and population doubling time (PDT) of 3D chondroblasts originated from chondro induced hASCs

The proliferative activity, as well as the PDT, was analyzed during the 14 days of hASCs culturing in chondrogenic induction medium exposed to vibration frequencies of 25, 35 and 45 Hz and in non-vibration control conditions. The obtained data showed that all investigated vibration frequencies influenced the proliferative potential, as well as the PDT, of chondroblasts originated from hASCs (Figs. 3A and 3B). The percentage of Alamar Blue reduction decreased proportionally with cell count and activity.

Cells cultured under 25 Hz frequency reached the highest PF after 2 days of incubation and then declined, but on the 10th day of culture reached higher PF and declined again till the 14th day. Moreover, we noticed the longest PDT (277 ± 25 h) when cells were stimulated with a 25 Hz frequency. The cells exposed to 35 Hz vibrations had a higher PF than control cultures at all investigated time points (Fig. 3A). During the analysis, hASCs after chondrogenic differentiation using 25 Hz vibrations resulted in the lowest proliferation potential, as well as longest PDT, when compared to the other investigated groups.

The morphology of chondroblasts originated from hASCs

After 14 days, formation of chondroblast-specific nodules could be clearly observed (Figs. 4G–4V and 5A–5G) and showed a strong orange signal after Safranin O staining (Figs. 4O–4R). Characteristic chondrocyte-like cells were observed in all of the investigated cultures, however, the cells cultured under the influence of 35 Hz vibrations more efficiently induced chondrodifferentiation (Fig. 4T). In these samples, nodules had the largest size (Fig. 6A) as well as exhibited the highest absorption of Safranin O (Fig. 6B). Cells cultured under 25 Hz vibrations absorbed less Safranin O dye and formed nodules with a significantly smaller diameter than samples cultured with 35 Hz vibrations (Figs. 6A and 6B). Chondroinduction of cells treated with 45 Hz were comparable to the control group. Although chondro-nodules had similar diameters, the absorption of Safranin O by cells cultured with 45 Hz frequencies was significantly higher (Figs. 6A and 6B).

Quantitative Collagen 1 and 2 (Col-1, Col-2) and Bone Morphogenetic Protein 2 (BMP-2) assay and chondrogenic gene expression analysis (SOX-9, Col-X, Col-II, Runx, Col-I, ACAN)

The performed analysis showed an increase in collagen type 2 concentration in comparison to the amount of collagen type I in all groups where vibrations were applied (Figs. 5A and 5B). These results were additionally confirmed by positive chondrogenic differentiation of hASCs.

The highest difference on mRNA level between collagen type II and type I was observed in the cultures stimulated with 25 and 35 Hz vibrations, and in the control culture. The lowest concentration of collagen type II with respect to collagen type I was observed when 45 Hz frequency vibrations were applied (Figs. 5A and 5B). Cells treated with 35 Hz frequency vibrations tended to secrete significantly higher amounts of BMP-2 in comparison to the other groups (Fig. 5C). Exposure to the 25 Hz stimulation model resulted in secretion of lower a concentration of BMP-2 in comparison to the 35 Hz, however significantly higher when compared to the 45 Hz frequencies, as well as to the control culture.

The quantitative evaluation of the concentration of collagen type I and type II was additionally confirmed by gene expression analysis (Figs. 4C and 4E). The highest activity of collagen type II and its predominance over expression of collagen type I was observed in cells treated with 35 Hz vibrations. Similarly to the quantitative evaluation, exposure to 25 Hz vibrations resulted in lower expression of Collagen type II when compared to the 35 Hz vibrations model. The gene expression of SOX-9 and Col-X, the master transcription factors of chondrogenesis, gradually increased in 35 Hz treated cells compared to control group (Figs. 4A and 4B). Aggrecan (ACAN) and RUNX2, another chondrogenic markers, significantly increased after 25 Hz treatment (Figs. 4D and 4F).

Analysis of integrin expression in response to vibration stimulation

In order to sense and translate the applied external mechanical signals, cells express mechanoreceptors on their surface, such as integrins. In our study, we analyzed the expression changes of four alpha (α3, α4, α5 and αV) and three beta (β1, β3 and β5) integrin subunits (Fig. 7). qPCR analysis demonstrated a slight increase of integrin α3, α4, β1 and β3 subunit expression after 25 Hz stimulation in comparison to control (0 Hz). We also found that when cells were stimulated with 35 Hz vibrations, hASCs significantly upregulated integrin α3, α4, β1 and β3 subunit. Interestingly, after 35 Hz stimulation, the highest increase in expression of the β3 intergrin was observed. With respect to integrin subunits α5, αV, and β5, expression levels were similar between to stimulated groups, however down-regulated as compared to control.

Proliferation factor and population doubling time of human adipocytes originated from adipo induced hASCs

The proliferation factor and PDT were determined in the various groups after 14 days of adipogenic induction. The stimulated cultures were characterized by an irregular proliferation rate (Fig. 8A). Stimulation with a 25 Hz frequency resulted in an increase of PF when compared with the other groups. The highest PF was on day 5, and from that point on a decreasing trend was observed (Fig. 8A). The 25 Hz frequency group also has the shortest PDT in comparison to other experimental groups (Fig. 8B).

In the 35 Hz and 45 Hz treatment groups, proliferation remained decreased at days 2, 5 and 10 of culture in comparison to the control and 25 Hz group, although this difference was not as pronounced as that on day 14. These results were also reflected in the PDT calculations—the 35 Hz and 45 Hz cultures had longer time to achieving PDT (Fig. 8B).

Morphology of ASCs after adipogenic differentiation

Microscopic analysis of adipocytes that originated from hASCs revealed that the 25 Hz vibrations mostly enhanced the adipogenic differentiation in comparison to the other groups (Fig. 9L). The findings from gene expression analysis showed noticeable increase of PPAR-γ and adiponectin (ADIQ) (Figs. 9A and 9B) in 25 and 35 Hz stimulated groups compared to control. However, qRT-PCR findings also showed similar gene expression of leptin (LEP) (Fig. 9C). The adipocytes derived from 25 Hz stimulated group were highly abundant with large lipid vacuoles, positively stained by Oil Red O staining, and absorbed the highest percentage of dye (Fig. 10A). The smallest adipocyte diameter was observed in the control group (ranges between 60 and 92 um), while adipocytes treated with 35 Hz were characterized by the highest average size (ranges between 90 and 119 um) (Fig. 10B).