A Combination of Low-Intensity Pulsed Ultrasound and Nanohydroxyapatite Concordantly Enhances Osteogenesis of Adipose-Derived Stem Cells From Buccal Fat Pad

Abstract

The osteogenic induction of adipose-derived stem cells (ADSCs) has been regarded as an important step in bone tissue engineering. In the present study, we focused on the buccal fat pad (BFP) as a source of adipose tissue, since BFPs are encapsulated by adipose tissue and are often coextirpated during oral surgery. Low-intensity pulsed ultrasound (LIPUS) is effective in the treatment of fractures, and nanohydroxyapatite (NHA) is known as a bone substitute material. Here we investigated the synergistic effects of LIPUS and NHA in the osteogenesis of ADSCs. A combination of LIPUS irritation and NHA as a scaffold significantly increased the osteogenic differentiation of ADSCs in vitro, and in our in vivo study in which ADSCs were transplanted into calvarial bone defects of nude mice, the combinational effect greatly enhanced the new bone formation of the margin of the defects. These results demonstrate that synergistic effects of LIPUS and NHA are capable of effectively inducing the differentiation of ADSCs into osteoblasts, and they suggest a novel therapeutic strategy for bone regeneration by the autotransplantation of ADSCs.

INTRODUCTION

Tissue engineering is an emerging field that combines the principles of bioengineering, cell transplantation, and biomaterial engineering to enable the regeneration with restitution of lost tissues. The continuously increasing biological knowledge regarding human development will most likely result in novel therapies for many clinical applications. For example, adipose-derived stem cells (ADSCs) are able to differentiate into multiple mesenchymal tissue cell types (2), such as osteoblasts, chondrocytes, adipocytes, myocytes, cardiomyocytes, and endothelial cells, and growing evidence suggests that ADSCs are capable of giving rise to cells from other lineages, such as those of the ectoderm and endoderm (6).

In addition, adipose tissue secretes a variety of angiogenic and antiapoptotic growth factors, making fat a promising source for reconstructive surgery (2). Since ADSCs have been used successfully in several experimental studies in the bone tissue engineering field, adipose tissue is an interesting source for cell-based therapy. The oral cavity in humans contains a mass of specialized fatty tissue, the buccal fat pad (BFP), which is distinct from subcutaneous fat (3). Since BFPs are easily accessed and have rich vascularization, BFP tissue can provide useful graft material, and in fact this tissue has been used widely in oral surgery for the repair of bone and periodontal defects (3,14). The harvesting of BFPs is a simple procedure that requires a minimal incision with local anesthesia and causes minimal donor-site morbidity.

A variety of different mechanical stimuli to induce osteogenic differentiation have been actively studied, among which is low-intensity pulsed ultrasound (LIPUS) (5,15). LIPUS has been shown to enhance bone growth during the healing of nonunion alignments, fractures, and other osseous defects. In addition, biomaterial scaffolds that are employed for bone tissue engineering provide temporary structural and functional support within a bone defect (5,15). The advantages of a combinational effect of LIPUS and biomaterial scaffolds in the osteogenic differentiation of ADSCs have thus been suggested, but few relevant studies have been reported. In the present study, we investigated the combinational effect of LIPUS as a mechanical stimulus and nanohydroxyapatite (NHA) as a biomaterial scaffold. We observed a synergistic effect of these, which contributes to efficient bone regeneration both in vitro and in vivo.

MATERIALS AND METHODS

Tissue Preparation and Cell Isolation

BFP was obtained from five healthy individuals undergoing elective orthognathic surgery procedures at Showa University Dental Hospital with the patients’ informed consent and the approval of the ethical committee of Showa University (Approval No. 2010-004). The age range of the patients, all of whom were female, was 18–42 years (n = 5). All patients were in good health, and no diabetes or other systemic complications were reported (Table 1). Raw oral fat tissue was washed several times with sterile phosphate-buffered saline (PBS; Mitsubishikagaku Yatoron, Chiba, Japan), minced into small pieces, and treated with 0.075% collagenase (Sigma-Aldrich, St. Louis, MO, USA) for 60 min at 37°C with shaking. After incubation, adipose tissue was centrifuged at 1,200 × g for 10 min twice to separate the adipocytes and lipid droplets from the stromal vascular fraction (SVF). The cell pellets were resuspended in red blood cell lysis buffer (0.83 % NH₄Cl in PBS), and then SVF cells were resuspended in Dulbecco’s modified Eagle’s medium nutrient mixture F-12 HAM (Sigma-Aldrich) containing 10% fetal bovine serum (FBS) and 100 units/ml antibiotic solution (penicillin; Life Technologies, Carlsbad, CA, USA). Thereafter, the cells were seeded in a 10-cm tissue culture dish (Techno Plastic Products, Trasadingen Switzerland) and maintained in a humidified incubator at 37°C and 5% CO₂.

Table 1.

The Five Patients From Whom Buccal Fat Pads Were Removed and the Number of Cells per Weight of Fat

Age (years)	Weight of Fat (g)	Cell Number per Weight (g)
18	4.5	3.9 × 104
33	2.6	7.7 × 104
18	4.7	5.5 × 104
42	4.0	6.5 × 104
37	3.4	2.0 × 104

Flow Cytometry

For flow cytometry with a FACScan argon laser cytometer (BD Biosciences, San Jose, CA, USA), cells were harvested and washed in cold PBS and then analyzed by FACScan flow cytometry, using anti-CD29 (1:200 dilution), CD44 (1:200), CD271 (1:200), CD105 (1:200), CD90 (1:100), CD34 (1:100), and CD45 (1:200) (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Aliquots of cells were incubated with isotype-matched mouse anti-human IgG (BD Bioscience) as a negative control. Data were analyzed using CellQuest software and FACSDIVA (BD Biosciences), as previously described (14).

Cell Staining

The cells were seeded into six-well tissue culture plates at a density of 3 × 10⁴ cells/well. On days 7, 14, and 21, cells were fixed with 4% paraformaldehyde solution (Wako, Osaka, Japan) and subjected to Oil red O (Wako) staining as well as alkaline phosphatase (ALPase) activity (Primary Cell, Hokkaido, Japan) and Alizarin red (Wako) staining, as previously described (2,3). Stained cells were examined by a microscope (BX-51; Olympus, Tokyo, Japan), and photos were taken by a CCD camera (DP-71; Olympus).

Cell Culture and NHA Constructs

ADSCs were grown in osteogenic medium (OM; Lonza, Tokyo, Japan) containing growth supplement (ascorbate, l-glutamine, dexamethasone, β-glycerophosphate, and penicillin–streptomycin; Lonza). A commercially available NHA (SHAp NHA; Sofsera, Tokyo, Japan) was used as a scaffold. The cells were grown in the presence or absence of NHA, as previously described (11).

LIPUS Stimulation

The cells in a six-well tissue culture plate (Corning, Corning, NY, USA) were subjected to LIPUS (OSTEOTRON D²; Ito, Tokyo, Japan) for 10 min everyday. The transducers were applied, with gel, directly to the bottom of the culture dish containing the cells. The frequency and spatial and temporal averaged intensity was set at a burst of 3.0 MHz sine waves repeated at 100 Hz and 60 mW/cm², respectively.

Real-Time PCR Analysis

Real-time PCR was performed with the SYBR green system (MyiQ2; Bio-Rad, Hercules, CA, USA), as previously described (1). One nanogram of each cDNA was used as a template, with 10 nM of each primer pair and 25 µl of 2 × iQ SYBR Green Supermix (Bio-Rad) in a total volume of 50 µl. The primer sequences are shown in Table 2. The time course condition was 95°C for 7 min; 1 cycle, 94°C for 30 s, 55°C for 30 s, 72°C for 1 min; 45 cycles. The statistical analyses were performed using the Bio-Rad iQ5 analysis software. The condition was gene expression was first normalized to that of glyceraldehyde phosphate dehydrogenase (GAPDH) within each sample group by a Ct^ΔΔ method.

Table 2.

Genes, Primer Sequences, and Annealing Temperatures Used in Real-Time PCR

Gene	Primer Sequences	Annealing Temp (°C)
Osteocalcin	5′-AGCAAAGGTGCAGCCTTTGT-3′ 5′-GCGCCTGGGTCTCTTCACT-3′	55
RUNX2	5′-AACCCACGAATGCACTATCCA-3′ 5′-CGGACATACCGAGGGACATG-3′	55
Osterix	5′-CCCCACCTCTTGCAACCA-3′ 5′-CCTTCTAGCTGCCCACTATTTCC-3′	55
PPARγ	5′-TGCCTTGCAGTGGGGATGT-3′ 5′-ATCGCCCTCGCCTTTGCTT-3′	58
GAPDH	5′-TGGTATCGTGGAAGGACTCATGAC-3′ 5′-ATGCCAGTGAGCTTCCCGTTCAGC-3′	55

Upper sequences are the sense primers, and lower sequences are the antisense primers. RUNX2, runt-related transcription factor 2; PPARγ, peroxisome proliferator-activated receptor γ; GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

Biochemical Assays

The cells were washed with saline, lysed with saline containing 0.2% Triton X-100 (Sigma-Aldrich), and subjected to DNA measurement, ALPase activity, and calcium and PO₄ deposition assays, as previously described (1).

Western Blot Analysis

Western blot analysis was carried out according to the method described previously (10). Cell lysates were prepared by centrifugation at 12,000 × g at 4°C. The supernatant was collected, and the protein concentrations of each lysate were measured by a BCA protein assay kit (Thermo Fisher Scientific, Waltham, MA, USA). Ten micrograms of each lysate proteins were heated at 95°C for 5 min in the SDS sample buffer (Bio-Rad) in the presence of 5% 2-melcaptoethanol (Wako), separated in 12.5% SDS-PAGE (Bio-Rad), and transferred to a polyvinylidene difluoride (PVDF) membrane (iBlot 2 Gel Transfer Device; Life Technologies). The blot was incubated with 1:1,000 dilution of primary antibodies for anti-phospho-p42/p44 mitogen-activated protein kinase [MAPK; extracellular signal-regulated kinase (ERK) 1/2] (Promega, Madison, WI, USA), anti-phospho-p38 (Promega), anti-p42/p44 MAPK (ERK1/2) (Cell Signaling Technology, Beverly, MA, USA), anti-phospho–c-Jun NH2-terminal kinase (Cell Signaling Technology), anti–c-Jun NH2-terminal kinase (Cell Signaling Technology), and anti-p38 (Merck Millipore, Billerica, MA, USA). Thereafter, the blot was incubated with 1:5,000 dilution of peroxidase-conjugated secondary antibodies [rabbit anti-goat IgG antibody (R&D Systems, Minneapolis, USA), goat anti-rabbit IgG antibody (Santa Cruz)], and then, proteins were visualized using enhanced chemiluminescence detection reagents (GE Health Care, Buckinghamshire, UK).

Animals

The Animal Experimentation Committee of Showa University approved this study (AP13062). Twelve 4-week-old female nude mice (Balb/c nu-nu; CLEA Japan, Tokyo) were used. Experiments were performed in n = 3. Therefore, we used a total of 12 mice. After the induction of anesthesia with intramuscular sodium pentobarbital (0.64 ml/kg body weight; Kyoritsu Seiyaku, Tokyo, Japan), a 2-cm-long incision was made with a scalpel down the midline of the calvarium, and the skin was peeled off. Symmetrical bone defects (5 mm dia.) were chipped away with a dental carbide bar in the dorsal parietal bones on each side of the midsagittal suture. Of note, the midsagittal suture was not included in the bone defects, in order to exclude its contribution to bone healing and limit the risk of damage to the superior sagittal sinus. During the surgical procedure, care was taken to avoid damaging the dura mater and puncturing the superior sagittal sinus (4,7,8). The ADSCs were admixed with porcine atelocollagen (Koken, Tokyo, Japan), and the cells were transplanted into the bone defects with or without NHA as a scaffold. Each transplantate was divided into two for two subgroups; one part was subjected to LIPUS irradiation every other day, and the other was not. Also, everyday LIPUS stimulation in vivo is more stressful for nude mice, leading to them being occasionally ill treated to death.

Histological Analysis

The defect sites were removed along with the surrounding bone and soft tissue, and then fixed in 10% formalin (Wako). The specimens were decalcified by soaking in 10% EDTA (Wako) for 1 week, and embedded in paraffin (Thermo Fisher Scientific). Coronal sections (5 µm) were cut through the centers of the circular defects and processed for hematoxlin (Wako) and eosin (Sakura Finetek Japan, Tokyo, Japan) (H&E) staining, and immunohistochemistry (IHC) for human-specific osteocalcin (OC) antibody (1:50 dilution; Alfa Aesar, Ward Hill, MA, USA), and mouse-specific osteocalcin antibody (1:100 dilution; Alfa Aesar), as previously described (7). The histological examination was then performed under a light microscope, and photos were taken.

Statistical Analysis

Unless otherwise specified, all experiments were repeated at least three times, and similar results were obtained in the repeated experiments. The data were subjected to a one-way analysis of variance (ANOVA) and multiple post hoc Bonferroni-corrected t-tests to compare differences among the various stimulation groups. In all cases, data were expressed as the mean ± standard deviation (SD), and differences were considered significant at p < 0.05.

RESULTS

Characteristics of ADSCs in a Monolayer Culture

Freshly isolated and cultured SVF was analyzed for cell morphology and membrane marker profiles to observe whether the cells derived from BFP share characteristics with ADSCs from another fat depot (Fig. 1A). After 1 week of culture, adherent cells were counted and reseeded in a tissue culture dish. Some adherent cells were observed on day 1 after reseeding. Thereafter, the adherent cells began to grow rapidly, approaching confluence as a monolayer of large, flat, and spindle-shaped cells.

Figure 1.

Pluripotency of ADSCs derived from BFP. (A) Morphology of ADSCs in culture. The cells were isolated from the patients’ BFP, as described in Materials and Methods. After 1 day or 7 days, the cells were examined by phase microscopy. (B) Flow cytometric analysis for various stem cell-specific CD antigens. ADSCs were cultured for 7 days and then subjected to flow cytometry for CD29, CD44, CD271, CD105, CD90, CD34, and CD45. The horizontal and vertical axes represent the expression of each target protein and the number of the cells, respectively. Negative control represents the condition by using isotype control IgGs, and relative values are shown at the right side of each CD antigen versus negative control. (C) Adipogenic and osteogenic differentiation of ADSCs. ADSCs were isolated and then cultured in an adipogenic (AM) or osteogenic (OM) medium for 7, 14, or 21 days. The cells were then subjected to Oil red O, ALPase activity, and Alizarin red staining.

After 7 days, the cells exhibited a more fibroblast-like morphology, which is a characteristic of ADSCs. At this point, the cells were subjected to flow cytometry for stem cell-specific cell-surface marker proteins (Fig. 1B). Several mesenchymal stem cell-specific markers were significantly positive, that is, CD29 (+169.2%), CD44 (+140.2%), CD271 (+144.3%), CD105 (+178.7%) and CD90 (+159.0%) versus control IgG. In contrast, none of the hematopoietic stem cell-specific markers (CD34 and CD45 antigens) were detected.

The multiple lineage-differentiative potency of ADSCs was then investigated (Fig. 1C). After the adipogenic induction of ADSCs (AM), intracellular lipid vacuoles were observed, and they were increased in size and number during culture. As well, the Oil red O staining revealed multiple intracellular lipid-filled droplets. We also cultured ADSCs in OM in order to assess the osteogenic potential (OM). ALPase, an early marker of osteoblastic differentiation, was determined after day 7, and its activity was increased in a time-dependent manner. The cells changed their morphology from spindle shaped to more polygonal shaped, accompanied by an increase in ALP activity, and the extracellular matrix was also stained intensely with Alizarin red, indicating calcification in a time-dependent manner. Taken together, these results demonstrated the pluripotency of ADSCs, and they suggest the usefulness of ADSCs in bone regeneration by transplantation.

LIPUS Irritation and NHA as a Scaffold Concordantly Increased the Osteogenesis of ADSCs

To investigate the roles of LIPUS irritation and NHA as a scaffold in the osteogenic differentiation of ADSCs, we conducted histocytochemical staining of ALPase activity and Alizarin red staining (Fig. 2A). ALPase activity was broadly observed in the OM and additional LIPUS irritation groups, regardless of the presence or absence of NHA. LIPUS irritation alone moderately enhanced the calcium deposition of extracellular matrix, which was stained by Alizarin red. It is noteworthy that the combination of LIPUS and NHA increased the calcium deposition very markedly.

Figure 2.

Concordant effects of NHA and LIPUS on the osteogenic differentiation of ADSCs. ADSCs were grown in the presence or absence of NHA as a scaffold and/or in the presence or absence of LIPUS irritation as a mechanical stimulation. On days 7, 14, and 21, the cells were subjected to Alizarin red staining (A), to real time RT-PCR for osteocalcin, Runx 2, osterix, and PPARγ (B), and to biochemical assays for DNA, ALPase activity, calcium and PO₄ deposition (C). (A, B, and C) Open, closed, hatched, and shaded boxes represent the conditions of control, NHA, LIPUS, and the combination of NHA + LIPUS, respectively. *p < 0.05 versus control.

We next used a real-time PCR for the quantification of osteogenesis- and adipogenesis-specific gene expressions (Fig. 2B). The expression of OC gene was increased by the combined stimulation of LIPUS and NHA on day 14. The expression of Runx 2 was also very markedly increased by this combination. However, these observations were uniformity attenuated in a time-dependent manner, and on day 21, they were not detectable any longer. The expression of osterix was observed in the same manner as of OC.

PPARγ, a major marker of adipocytes, was decreased in all conditions in a differentiation-dependent manner. To further quantitatively assess the results shown above, we carried out biochemical assays including measurements of the DNA content, ALP activity, calcium deposition, and PO₄ (Fig. 2C). In the DNA and ALPase activity measurements, significant differences were barely observed among the conditions. However, of importance, the combination of LIPUS + NHA increased the calcium and PO₄ deposition on day 21.

As seen in Western blot analysis (Fig. 3), however, the difference between the NHA− LIPUS-stimulated and NHA+ LIPUS-stimulated condition for three subgroups of MAPK pathways was not observed at these time (<60 min) periods. The apparent discrepancy observed between other biochemical assays and Western blot analysis might be due to the time scale difference between two assays. The results of these in vitro studies indicated that ADSCs from patients’ BFPs are inducible into osteoblasts when the optimal combination of LIPUS and NHA is used, and they will then be available for cell transplantation for bone defects. The results of our in vivo studies, in which these cells were transplanted to artificial bone defects in nude mice, are discussed next.

Figure 3.

Western blot analysis. The effect of NHA on LIPUS-induced activation of three subgroups of MAPK pathways was not observed within 60 min.

LIPUS Irritation + NHA as Scaffolds Concordantly Induced New Bone Formation of Transplanted ADSCs In Vivo

Atelocollagen alone or ADSCs admixed with atelocollagen were transplanted into bone defects of nude mouse calvarium, and each transplantate was subjected to LIPUS irritation or not. Figure 4 shows the H&E staining and IHC results for the human-specific OC of transplantates with surrounding calvarial cortical bone. The atelocollagen alone (ADSC⁻) transplantates showed no new bone formation at the bone defect margins, whereas the transplantation of ADSC (ADSC⁺) admixed atelocollagen showed obvious new bone formation along with bone defect margins.

Figure 4.

(A) Histochemistry of transplanted ADSCs. Porcine atelocollagen alone (ADSC⁻) or ADSCs admixed with atelocollagen (ADSC⁺) in the presence of NHA were transplanted into bone defects of mouse calvaria. Thereafter, the calvaria were irritated by LIPUS (see Materials and Methods). After day 14, mice were sacrificed, and their calvaria were subjected to H&E staining and IHC for the identification of human-specific OC and mouse-specific OC, after proper demineralization. The area with double-arrowed bars indicates the thickness of newly formed bone positive for human-specific OC. OC, osteocalcin; T, transplantate; C, calvarial cortical bone; NB, new bone. (B) Quantitative analysis of the new bone data of NHA, LIPUS, NHA + LIPUS group (ADSC⁺).

The IHC showed that the bone matrix contained human-specific OCs but no mouse-specific OCs. This observation revealed that the new bone was synthesized by the transplanted human ADSCs, not by the mouse endogenous osteoblasts. The transplantates with NHA as a scaffold showed only slightly increased effects. It is noteworthy that LIPUS irritation + NHA as a scaffold concordantly showed significantly thick new bone formation. These results indicate that a combination of LIPUS irritation + NHA as a scaffold can have a very positive effect on the osteogenesis of transplanted ADSCs, and they suggest a novel therapeutic method for bone regeneration using the autotransplantation of these cells into bone defects.

DISCUSSION

Adult stem cells are present in various organs, and they develop important functions in tissue maintenance and homeostasis. One major goal of tissue engineering medicine is to find a tissue source that can provide adequate numbers of stem cells for clinical applications but be accompanied by minimal morbidity. Adipose tissue holds great promise in regenerative medicine; it is available in large quantities as a waste material, and it contains more progenitor cells that give rise to different cell populations than does bone marrow.

Unlike subcutaneous fat, the BFP is a specialized mass of adipose tissue considered an ideal flap for oral surgery. BFPs are easy to harvest and are reliable, and they contain a rich blood supply; the use of BFPs has shown minimal donor-site morbidity and a low rate of complications (3). In the present study, the separated cells from human BFP expressed various differential markers of ADSCs, shown by cytohistochemical staining and flow cytometry (Fig. 1). The results indicated that ADSCs are pluripotent per se, and that the cells might have a great advantage for bone regeneration.

Hydroxyapatite is widely used as a bone substitute material in dentistry (9), and NHA is also widely used as a scaffold of the cell as a scaffold is more efficient (16). LIPUS is often used as a mechanical stimulation for therapies in orthopedic setting and in wide range of other medical fields (14). We therefore investigated the combinational effects of NHA and LIPUS for the differentiation of ADSCs.

First, our in vitro study (Fig. 2) showed that the combination of NHA + LIPUS promoted the osteogenic differentiation of ADSCs in an osteogenetic medium.

Although the effects in an adipogenic medium were investigated in the present study, the results strongly indicated the direct differentiation of ADSCs into osteoblasts. However, although the promotive effect of LIPUS alone on Alizarin red staining was minimum compared with NHA (Fig. 2A), LIPUS alone promoted both calcium and phosphate deposition as much as NHA (Fig. 2C). The apparent discrepancy observed between biochemical assays and Alizarin red staining might be due to the sensitivity and systematic difference between these assays. Since biochemical assays are more sensitive than Alizarin red staining, the results between these assays may not be comparable, and our present results are somewhat expected and are not contradictory. To understand the molecular mechanism by which NHA increases LIPUS-induced osteogenic differentiation, we investigated the effects of NHA on LIPUS-induced activation of three subgroups of MAPK pathways. As shown in Figure 3, the NHA activation of three subgroups of MAPK pathways was not observed within 60 min. It was plausible that NHA increases LIPUS-induced signaling at prolonged time scale. Unlike the conventional HA, such as micro-HA, nanosize HA may be considered more efficient for their improved osseointegrative properties (12,13,16). Based on this hypothesis, we next carried out an in vivo study (Fig. 4A, B), and it was revealed that the combination of NHA as a scaffold and LIPUS irritation effectively reinforces new bone formation of transplanted ADSCs at the margins of bone defect, as was observed in our in vitro study. This result suggests an optimal condition of scaffold plus stimulation that compels the ADSCs to form new bone in a transplantation area.

In the medical field, a focus has been placed on the regeneration of human tissue that was lost, and a goal of regenerative medicine is to establish the optimal culture conditions for the induction of functional cells and/or specific cell proliferation by medical engineering. The development of aspects of sustained-release technology such as growth factors and the differentiation of scaffolding is another challenge (9,16). HA is well known as a scaffold that can induce the proliferation and differentiation of various cells, and LIPUS irritation is usually applied for wound healing (12) and to stimulate the proliferation and differentiation of cells (13).

However, to the best of our knowledge, a combinational effect on the osteogenic differentiation of transplanted ADSCs had not been reported prior to the present study, the results of which revealed that a combination of NHA and LIPUS is a potential novel application for the autotransplantation of ADSCs to bone defects. Of course, more detailed and extensive studies should be carried out before any preclinical or clinical application of NHA + LIPUS. Further relevant investigations are under way in our laboratory.