Genetic response to low‑intensity ultrasound on mouse ST2 bone marrow stromal cells.

Although low‑intensity ultrasound (LIUS) is a clinically established procedure, the early cellular effect of LIUS on a genetic level has not yet been studied. The current study investigated the early response genes elicited by LIUS in bone marrow stromal cells (BMSCs) using global‑scale microarrays and computational gene expression analysis tools. Mouse ST2 BMSCs were treated with LIUS [ISATA, 25 mW/cm2 for 20 min with a frequency of 1.11 MHz in a pulsed‑wave mode (0.2‑s burst sine waves repeated at 1 kHz)], then cultured for 0.5, 1 and 3 h at 37˚C. The time course of changes in gene expression was evaluated using GeneChip® high‑density oligonucleotide microarrays and Ingenuity® Pathway Analysis tools. The results were verified by reverse transcription‑quantitative polymerase chain reaction (RT‑qPCR). A single exposure of LIUS did not affect cell morphology, cell growth or alkaline phosphatase activity. However, 61 upregulated and 103 downregulated genes were identified from 0.5 to 3 h after LIUS treatment. Two significant gene networks, labeled E and H, were identified from the upregulated genes, while a third network, labeled T, was identified from the downregulated genes. Gene network E or H containing the immediate‑early genes FBJ osteosarcoma oncogene and early growth response 1 or the heat shock proteins heat shock protein 1a/b was associated mainly with the biological functions of bone physiology and protein folding or apoptosis, respectively. Gene network T containing transcription factors fos‑like antigen 1 and serum response factor was also associated with the biological functions of the gene expression. RT‑qPCR indicated that the expression of several genes in the gene networks E and H were elevated in LIUS‑treated cells. LIUS was demonstrated to induce gene expression after short application in mouse ST2 BMSCs. The results of the present study provide a basis for the elucidation of the detailed molecular mechanisms underlying the cellular effects of LIUS.

Introduction

Ultrasound technology has many applications for diagnosis and therapy in medicine. Low-intensity ultrasound (LIUS) has also been used to treat fractures. The LIUS protocol for fracture consists of signals of 30 mW/cm2 spatial average-temporal average (ISATA) intensity with a frequency of 1.5 MHz in pulsed-wave mode (0.2-s burst sine waves repeated at 1.0 kHz) for 20 min per day (1,2). The ability of LIUS to promote fracture healing has been widely investigated in both experimental vertebrate animal models (3–6) and randomized clinical trials (7,8). Fracture healing is a complex physiological process involving various cell types. A number of in vitro studies have clearly demonstrated LIUS-associated bone-formative responses in cells related to fracture healing (9–21). However, the mechanisms underlying the ability of bone cells to detect and respond to LIUS irradiation remain poorly understood.

To elucidate these mechanisms, we consider that monitoring of the early cellular response to LIUS is crucial. The combination of global-scale DNA microarray analysis and bioinformatics analysis tools has provided a view of the gene expression profiles, biological functions, and gene networks relevant to the LIUS response of cells. Several groups have reported that the gene expression patterns were affected within 3 to 24 h after LIUS-treatment in several cell types (18–23). However, there has been no report about earlier changes in gene expression profiles shortly after exposure of the cells to LIUS.

Immediate-early genes (IEGs) were reported to be induced within minutes after stimulation and have essential roles in stress responses (24). Previous findings indicated that FBJ osteosarcoma oncogene (Fos), a well-characterized IEG, is induced by mechanical loadings including LIUS (9–13). The Fos protein, a leucine zipper-containing domain, heterodimerizes with the jun proto-oncogene (Jun) protein to form the dimeric transcription factor activator protein-1 (AP-1) (24), which plays an important role in bone formation (25). More recently, we showed that LIUS at an intensity of 25 mW/cm2 significantly and transiently increased the expression levels of 4 IEGs - Fos, early growth response 1 (Egr1), Jun, and prostaglandin-endoperoxide synthase 2 (Ptgs2) - in mouse ST2 bone marrow stromal cells (BMSCs) (13).

Here, to elucidate the early cellular response to LIUS in cells, the early response genes elicited by LIUS [ISATA, 25 mW/cm2 for 20 min with a frequency of 1.11 MHz in a pulsed-wave mode (0.2-s burst sine waves repeated at 1 kHz)] in mouse ST2 BMSCs were investigated using GeneChip® oligonucleotide microarrays and computational gene expression analysis tools.

Materials and methods

Cell culture

Mouse ST2 BMSCs (RCB0224) were obtained from the RIKEN BRC through the National Bio-Resource Project of the Ministry of Education, Culture, Sports, Science, and Technology of Japan (MEXT). The cells were maintained in RPMI-1640 medium (Nacalai Tesque, Inc.) supplemented with 10% fetal bovine serum (Equitech-Bio, Inc.) at 37°C in humidified atmosphere of 95% air with 5% CO2. For the LIUS experiments, the cells (2×105 cells) were seeded on a 35-mm plastic culture dish (Nippon Genetics Co., Ltd.) with 2 ml of culture medium and cultured at 37°C for 24 h.

LIUS treatment

LIUS treatment was applied by an ultrasound irradiating system, which consists of an irradiator SCI-p1-35 (Medical Ultrasound Co., Ltd.) with a ceramic transducer (28 mm diameter, 2 mm thickness) and SCI-D100 driving equipment (Nepa Gene Co., Ltd.). The LIUS signal had a ISATA of 25 mW/cm2, with a frequency of 1.11 MHz in a pulsed-wave mode (0.2-s burst sine waves repeated at 1 kHz) (13). LIUS was transmitted through the bottom of the culture dish in a CO2 incubator at 37°C (Fig. 1). The cells were irradiated with LIUS for 20 min, followed by incubation for 0, 0.5, 1, 3 or 24 h at 37°C. For heat treatment, cells were exposed to 38, 39, 40, 41 and 42°C for 20 min in a water bath. After heat treatment, the cells were incubated for 0 or 0.5 h at 37°C. The temperature of the culture medium was monitored with a digital thermometer (SK-1110, Sato Keiryoki MFG Co., Ltd.) coupled to a type K thermocouple sensor (SK-K020, Sato Keiryoki MFG Co., Ltd.) (13).

Figure 1.

Set-up for low-intensity ultrasound exposure.

Morphological and cell growth analyses

Cell morphology was observed using a phase-contrast microscopy. For cell growth, Cell Count Reagent SF (Nacalai Tesque Inc.), a water-soluble tetrazolium salt WST-8 based assay, was used. Cells were incubated with WST-8 solution at 37°C. Thirty min later, the produced formazan dye concentration was determined from the absorbance at 450 nm (26).

Assay of alkaline phosphatase (ALP) activity

Cellular material was placed into 50 mM Tris-HCl buffer (pH 7.2) containing 0.01% Triton X-100 and homogenized by an ultrasonic disruptor (UD-200, Tomy Co.). ALP activity of the cell homogenate was measured by using a protocol supplied by Sigma (Sigma-Aldrich; Merck KGaA). The absorbance at 405 nm was read on a plate reader (13). The protein concentration was determined with a bicinchoninic acid assay.

RNA isolation

Total RNA was isolated from cells using a NucleoSpin® RNA Plus isolation kit (MACHEREY-NAGEL GmbH & Co.) and treated with NucleoSpin® genomic DNA removal column (MACHEREY-NAGEL GmbH & Co.) to remove residual genomic DNA. The concentration of RNA was measured by spectroscopy with an expected A260/A280 ratio close to 2. Qualitative assessment of the RNA was also checked using a Bioanalyzer 2100 (Agilent Technologies, Inc.). RNA samples (RNA integrity number values: >9.5) were used (26).

Microarray gene expression and gene network analyses

Microarray analysis was performed using a GeneChip® system with a Mouse Genome 430 2.0 array (Affymetrix, Inc.) spotted with 45,101 probe sets. Array samples were prepared as described in the Affymetrix GeneChip® Expression Technical Manual. Total RNA from three experiments was pooled, and 500 ng of the RNA was used to synthesize cRNA with a GeneChip® 3′ IVT Express Kit (Affymetrix, Inc.). The array was hybridized with biotin-labeled cRNA at 45°C for 16 h. After the treatment with phycoerythrin-labeled streptavidin, the array was scanned using a prove array scanner (Affymetrix GeneChip® Scanner 3000). The obtained hybridization intensity data were further analyzed using GeneSpring® GX (Agilent Technologies, Inc.) to extract the significant genes and Ingenuity® Pathway Analysis tools (Tomy Digital Biology, Co., Ltd.) to examine gene ontology, including biological processes, cellular components, molecular functions, and gene networks (21,27).

Real-time quantitative polymerase chain reaction (qRT-PCR) assay

qRT-PCR was carried out on an Mx3005P real-time PCR system (Agilent Technologies, Inc.) as described previously (13). The specific primer and probe sequences are listed in Table I. β-actin (Actb) was used as an internal control.

Table I.

Nucleotide sequences of the primer pairs and a probe for target genes.

Gene	Orientation	Nucleotide sequence (position)	GenBank accession no.
Actb	Sense	ACCCTAAGGCCAACCGTGAA (441–460)	NM_007393
	Antisense	TGTGGTACGACCAGAGGCATAC (556–535)
Bag3	Sense	CCAAATCGGGAGAAGCCGAAAC (1514–1535)	NM_013863
	Antisense	TCGTCCTTCAGGGTCTACGGAA (1710–1689)
Bglap	Sense	GGCAATAAGGTAGTGAAC (169–187)	NM_007541
	Antisense	GCCATACTGGTCTGATAG (297–279)
	Probe	FAM-CGCTACCTTGGAGCCTCAGT-TAMRA (196–216)
Dnajb1	Sense	AGATCTACAGCGGCTGTACC (678–697)	NM_018808
	Antisense	AGCCCCTCTTCACTTCGATG (791–772)
Egr1	Sense	ATGAGCACCTGACCACAGAG (563–582)	NM_007913
	Antisense	GAGGCAGAGGAAGACGATGA (830–811)
Fos	Sense	TGCAAGATCCCCGATGACCT (749–768)	NM_010234
	Antisense	TCAGCTCCACGTTGCTGATG (932–913)
Hspa1a	Sense	GAACGCGCTCGAATCCTATG (1848–1867)	NM_010479
	Antisense	GAGATGACCTCCTGGCACTT (1917–1899)
Hspb1	Sense	CTGGCAAGCACGAAGAAAGG (514–533)	NM_013560
	Antisense	AGGGGATAGGGAAGAGGACA (629–610)
Ptgs2	Sense	TTCTCCCTGAAGCCGTACAC (1559–1578)	NM_011198
	Antisense	GGCAGGGTACAGTTCCATGA (1669–1650)

Actb, β-actin; Bag3, BCL2-associated athanogene 3; Bglap, bone γ carboxyglutamate protein; Dnajb1, DnaJ heat shock protein family (Hsp40) member B1; Egr1, early growth response 1; Fos, FBJ osteosarcoma oncogene; Hspa1a, heat shock protein 1A; Hspb1, heat shock protein 1; Ptgs2, prostaglandin-endoperoxide synthase 2.

Separation of total cellular, cytoplasmic and nuclear fractions

For separation of total cellular fraction, cells were lysed with lysis buffer (50 mM NaCl, 1% Nonidet P-40 and 50 mM Tris-HCl, pH 8.0) containing protease inhibitor cocktail (Nacalai Tesque, Inc.). For separation of cytoplasmic and nuclear fractions, cells were lysed in the fractionation buffer [phosphate-buffered saline containing 0.1% Nonidet P-40 and the protease inhibitor cocktail (Nacalai Tesque, Inc.)]and centrifuged at 15,000 × g for 10 sec at 4°C to obtain the cytosolic fraction (supernatants). The insoluble pellets were resuspended in the fractionation buffer and centrifuged at 15,000 × g for 10 sec at 4°C to obtain the nuclear fraction (pellets). Either glyceraldehyde 3-phosphate dehydrogenase (Gapdh) or fibrillarin (Fbl) was used as the cytoplasmic or nuclear marker protein, respectively (26).

SDS-polyacrylamide gel electrophoresis (PAGE) and Western blotting

SDS-PAGE and Western blotting were carried out according to our previous reports (26). Proteins were detected using the following primary antibodies: rabbit polyclonal anti-heat shock transcription factor 1 (Hsf1) antibody (1:2,000 dilution, cat. no. 4356; Cell Signaling Technology, Inc.), mouse monoclonal anti-Gapdh antibody (1:2,000 dilution, cat. no. 60004-1-Ig; Proteintech Group, Inc.) and rabbit monoclonal anti-Fbl antibody (1:2,000 dilution, cat. no. 2639; Cell Signaling Technology, Inc.). Secondary fluorescent IRDye-conjugated anti-rabbit and anti-mouse antibodies (1:10,000 dilution, LI-COR Biosciences) were also used. Fluorescence images were acquired using an Odyssey Infrared Imager (LI-COR Biosciences), and the band density was quantified using Image Studio 5.1 software (LI-COR Biosciences).

Statistical analysis

Statistical analysis was performed using KaleidaGraph version 4.1 software (Hulinks, Inc.). Data were expressed as the means ± standard deviation (SD) of ≥3 repeated experiments. Differences between groups were analyzed by ANOVA, and correction for multiple comparisons was made using Tukey's post hoc test. Comparisons between two groups were made by using Student's t-test. Statistically significance was set at P-values <0.05.

Results

Effects of LIUS on the morphology, growth and ALP activity of ST2 BMSCs

ST2 BMSCs were exposed to a single LIUS irradiation (25 mW/cm2 for 20 min) followed by culturing at 37°C for 24 h. Microscopic investigations demonstrated that ST2 BMSCs had a fibroblast-like morphology. LIUS treatment did not change the ST2 morphology (Fig. S1). Moreover, LIUS treatment did not affect cell growth or ALP activity, a marker for osteoblast differentiation (Fig. S2), as it did in our previous investigations (13).

Gene expression analysis

After a single LIUS treatment, the cells were cultured at 37°C for 0.5, 1, and 3 h. Global-scale gene expression analysis was carried out using a GeneChip® Mouse Genome 430 2.0 oligonucleotide array. Of the 45,101 probe sets analyzed, approximately 20,000 probe sets were significantly expressed in either control or LIUS-treated cells. The complete lists of genes from the ST2 cell samples have been deposited in the Gene Expression Omnibus, a public database (accession number: GSE135935). GeneSpring® software-based expression analysis of cells exposed to LIUS demonstrated many probe sets that were differentially regulated by a factor of 2.0 or greater. As shown in a hierarchical clustering heatmap (Fig. 2), 83 and 111 probe sets were found to be upregulated and downregulated, respectively, in ST2 cells 0.5, 1 and 3 h after LIUS treatment.

Figure 2.

Hierarchical clustering of 194 probe sets differentially expressed by >2.0 in the cells exposed to LIUS. After treatment with LIUS at 25 mW/cm2 for 20 min, the cells were cultured for 0.5, 1 and 3 h at 37°C. Clustering was carried using GeneSpring® GX software. Non-treated cells served as the control. The number in parentheses indicates the number of probe sets. LIUS, low-intensity ultrasound; Ctr, control; Down, downregulated; Up, upregulated.

Biological function and gene network analyses

Next, biological function and gene network analyses were conducted by using the Ingenuity® Pathways Knowledge Base. A total of 61 upregulated and 103 downregulated, functionally annotated genes were identified in the 83 and 111 differentially expressed probe sets, respectively. The top 10 upregulated and top 10 downregulated genes are shown in Tables II and III, respectively. Moreover, 2 significant gene networks, labeled E (early-response) and H (Hsf1-regulated), were identified from the upregulated genes (Figs. 3 and 4), while a third network, labeled T (transcription-related), was identified from the downregulated genes (Fig. 5). Gene network E included 11 genes - 3 IEGs (Fos, Egr1 and Ptgs2) plus activating transcription factor 3 (Atf3), bone γ carboxyglutamate protein (Bglap), dual specificity phosphatase 1 (Dusp1), endothelin1 (Edn1), fibromodulin (Fmod), gap junction protein, α 1 (Gja1), Kruppel-like factor 10 (Klf10) and transcriptional repressor GATA binding 1 (Trps1) - and was associated with the biological functions of differentiation of bone cells, remodeling of bone, and ossification of bone (Fig. 3, Table IV). Gene network H included 10 genes - 7 heat shock proteins (HSPs) [DnaJ heat shock protein family (Hsp40) member A1 (Dnaja1), DnaJ heat shock protein family (Hsp40) member A4 (Dnaja4), DnaJ heat shock protein family (Hsp40) member B1 (Dnajb1), heat shock protein 1a/b (Hspa1), heat shock protein 8 (Hspa8), heat shock protein 1 (Hspb1) and heat shock 105kDa/110kDa protein 1 (Hsph1)] plus BAG cochaperone 3 (Bag3), αB-crystallin (Cryab) and ubiquitin C (Ubc) - and was associated with the biological functions of protein folding, protein refolding, and apoptosis (Fig. 4, Table IV). Ingenuity® pathway analysis also indicated that the Hsf1 protein was predicted to be activated at 0.5 to 3 h after LIUS stimulation (Fig. 4). In addition, we identified 9 genes in gene network T - 6 transcription factors [euchromatic histone lysine N-methyltransferase 2 (Ehmt2), ets variant 5 (Etv5), fos-like antigen 1 (Fosl1), GLIS family zinc finger 2 (Glis2), snail family zinc finger 2 (Snai2) and serum response factor (Srf)] plus gelsolin (Gsn), spleen tyrosine kinase (Syk) and trefoil factor 1 (Tff1). This network was associated with the biological functions of transcription, expression of RNA and activation of DNA endogenous promoter (Fig. 5, Table IV).

Table II.

Top 10 upregulated genes.

	Fold-change
Gene symbol	0.5 h	1 h	3 h	Description
Hspa1a	73.0	102.6	14.5	Heat shock protein 1A
Fos	26.8	11.0	1.9	FBJ osteosarcoma oncogene
Hspa1b	10.5	11.0	7.3	Heat shock protein 1B
Egr1	10.0	5.6	1.5	Early growth response 1
Hspb1	6.2	6.3	9.1	Heat shock protein 1
St3gal6	5.8	1.8	3.5	ST3 beta-galactoside α-2,3-sialyltransferase 6
Dnajb1	5.4	6.4	1.2	DnaJ (Hsp40) homolog, subfamily B, member 1
Arc	5.3	4.0	0.9	Activity regulated cytoskeletal-associated protein
Dnaja4	5.3	7.9	5.4	DnaJ (Hsp40) homolog, subfamily A, member 4
Hspa8	5.1	2.6	0.7	Heat shock protein 8

Table III.

Top 10 downregulated genes.

	Fold-change
Gene symbol	0.5 h	1 h	3 h	Description
Ddah1	0.10	0.04	0.06	Dimethylarginine dimethylaminohydrolase 1
Gsn	0.13	0.15	0.11	Gelsolin
Flnc	0.18	0.18	0.60	Filamin C, γ
Itgb1	0.19	0.18	0.30	Integrin beta 1
Ddx55	0.20	0.52	0.61	DEAD (Asp-Glu-Ala-Asp) box polypeptide 55
Riok1	0.30	0.77	0.88	RIO kinase 1
Pcsk4	0.31	0.79	0.62	Proprotein convertase subtilisin/kexin type 4
Cerk	0.36	0.66	0.67	Ceramide kinase
Tbca	0.36	0.64	0.84	Tubulin cofactor A
Nlgn2	0.37	0.38	0.50	Neuroligin 2

Figure 3.

Early-response gene network. Upregulated genes were analyzed using Ingenuity® Pathway Analysis tools. The network is presented graphically as nodes (genes) and edges (the biological relationships between the nodes). Atf3, activating transcription factor 3; Bglap, bone γ carboxyglutamate protein; Dusp1, dual specificity phosphatase 1; Egr1, early growth response 1; Edn1, endothelin 1; Fmod, fibromodulin; Fos, FBJ osteosarcoma oncogene; Gja1, gap junction protein, α 1; Klf10, Kruppel-like factor 10; Ptgs2, prostaglandin-endoperoxide synthase 2; Trps1, transcriptional repressor GATA binding 1.

Figure 4.

Hsf1-regulated gene network. Upregulated genes were analyzed using Ingenuity® Pathway Analysis tools. The network is presented graphically as nodes (genes) and edges (the biological relationships between the nodes). A, activation; E, expression; LO, localization; P, phosphorylation/dephosphorylation; PD, protein-DNA binding; PP, protein-protein binding; T, transcription. Bag3, BAG cochaperone 3; Cryab, αB-crystallin; Dnaja1, DnaJ heat shock protein family (Hsp40) member A1; Dnaja4, DnaJ heat shock protein family (Hsp40) member A4; Dnajb1, DnaJ heat shock protein family (Hsp40) member B1; Hsf1, heat shock transcription factor 1; Hspa1, heat shock protein 1a/b; Hspa8, heat shock protein 8; Hspb1, heat shock protein 1; Hsph1, heat shock 105kDa/110kDa protein 1; Ubc, ubiquitin C.

Figure 5.

Transcription-related gene network. Downregulated genes were analyzed using Ingenuity® Pathway Analysis tools. The network is presented graphically as nodes (genes) and edges (the biological relationships between the nodes). E, expression; PP, protein-protein binding; PD, protein-DNA binding; T, transcription. Ehmt2, euchromatic histone lysine N-methyltransferase 2; Etv5, ets variant 5; Fosl1, fos-like antigen 1; Glis2, GLIS family zinc finger 2; Gsn, gelsolin; Snai2, snail family zinc finger 2; Srf, serum response factor; Syk, spleen tyrosine kinase; Tff1, trefoil factor 1.

Table IV.

Biological functions in gene networks.

Name	P-value	Genes
Gene network E
Differentiation of bone cells	2.83×10−8	Bglap, Edn1, Fos, Gja1, Klf10 and Ptgs2
Remodeling of bone	5.19×10−10	Bglap, Edn1, Egr1, Fos, Gja1 and Ptgs2
Ossification of bone	2.39×10−9	Fmod, Gja1, Klf10, Ptgs2 and Trps1
Gene network H
Folding of protein	6.37×10−13	Cryab, Dnaja1, Dnaja4, Hspa8, Hspa1 and Hspb1
Refolding of protein	3.56×10−11	Dnaja1, Dnaja4, Hspa8 and Hspa1
Apoptosis	2.00×10−6	Bag3, Cryab, Dnaja1, Dnajb1, Hspa8, Hspa1, Hspb1, Hsph1 and Ube
Gene network T
Transcription	1.84×10−9	Ehmt2, Etv5, Fosl1, Glis2, Gsn, Snai2, Srf, Syk and Tff1
Expression of RNA	3.70×10−9	Ehmt2, Etv5, Fosl1, Glis2, Gsn, Snai2, Srf, Syk and Tff1
Activation of DNA endogenous promoter	1.76×10−6	Ehmt2, Etv5, Fosl1, Glis2, Snai2a and Srf

Verification of upregulated genes by qRT-PCR assay

Eight upregulated genes were chosen: Bglap, Egr1, Fos, and Ptgs2 from gene network E, and Bag3, Dnajb1, Hspa1 and Hspb1 from gene network H. As shown in Fig. 6, treatment of cells with LIUS at 25 mW/cm2 for 20 min significantly elevated the expression levels of all 8 genes. The expression levels peaked at 0.5–1.0 h after LIUS stimulation in 7 genes, i.e., Egr1, Fos, Ptgs2, Bag3, Dnajb1, Hspa1 and Hspb1. On the other hand, the peak expression level of Bglap was observed 3 h after LIUS exposure (Fig. 6).

Figure 6.

Verification of the upregulated genes determined via RT-qPCR. After treatment with low-intensity ultrasound at 25 mW/cm2 for 20 min, cells were cultured for 0.5, 1 or 3 h at 37°C. RT-qPCR was performed with specific primers for (A) Bglap, (B) Egr1, (C) Fos, (D) Ptgs2, (E) Bag3, (F) Dnajb1, (G) Hspa1 and (H) Hspb1. Levels were normalized to that of Actb. Non-treated cells were served as the Ctr. Data are presented as the mean ± SD (n=4). *P<0.05 vs. the Ctr group. RT-qPCR, reverse transcription-quantitative PCR; Ctr, control. Bglap, bone γ carboxyglutamate protein; Egr1, early growth response 1; Fos, FBJ osteosarcoma oncogene; Ptgs2, prostaglandin-endoperoxide synthase 2; Bag3, BAG cochaperone 3; Dnajb1, DnaJ heat shock protein family (Hsp40) member B1; Hspa1, heat shock protein 1a/b; Hspb1, heat shock protein 1.

Effects of LIUS on the Hsf1 activation

Under heat-stress conditions, the induction of HSPs principally occurs through the activation of Hsf1. It is well known that the mobility shift and nuclear translocation of Hsf1 due to its phosphorylation indicates activation of the molecule (28). Here, LIUS (25 mW/cm2 for 20 min) induced the temperature rise of the culture medium from 35.3±0.23°C (control) to 38.1±0.29°C (mean ± SD, n=4). Under control conditions, the ratio of phosphorylated Hsf1 (mobility shift of Hsf1) to non-phosphorylated Hsf1 was very low. On the other hand, a slight but significant mobility shift of Hsf1 was observed at 0 h after the LIUS treatment (Fig. 7A and C). Under heat-treated conditions, the mobility shift of Hsf1 0 h after the heat exposure was markedly and significantly increased in a temperature-dependent manner (Fig. 7B and D). As shown in Fig. 8, Hsf1 was principally localized in the cytosolic compartments in the control cells. Treatment of cells with LIUS significantly induced the nuclear localization of Hsf1 (Fig. 8A and C). This translocation was comparable to that in the cells treated with heat at 40°C for 20 min. On the other hand, the ratio of nucleus Hsf1 to cytoplasm Hsf1 was dramatically elevated in the cells treated with heat at 42°C for 20 min (Fig. 8A and D). These data suggested that Hsf1 is activated in the cells treated with LIUS.

Figure 7.

Effects of LIUS and heat on the activation of Hsf1 in ST2 cells. Cells were exposed to LIUS (25 mW/cm2) or heat (38, 39, 40, 41 or 42°C) for 20 min, and cultured at 37°C. (A and C) Cells were harvested 0, 0.5 or 1 h after LIUS exposure. (B and D) Cells were harvested immediately after heat exposure. Western blotting was carried out using specific primary antibodies against Hsf1 and Gapdh. Each Hsf1 band density was quantified, and the ratio (phosphorylated Hsf1 to non-phosphorylated Hsf1) was calculated. Gapdh served as a loading control. Green arrow indicates nHsf1. Red arrow indicates pHsf1. Data are presented as the mean ± SD (n=4–6). *P<0.05 vs. the Ctr group. LIUS, low-intensity ultrasound; Ctr, control; Hsf1, heat shock transcription factor 1; p, phosphorylated; n, non-phosphorylated; Gapdh, glyceraldehyde 3-phosphate dehydrogenase.

Figure 8.

Effect of LIUS and heat on the intracellular localization of Hsf1 in ST2 cells. Cells were exposed to (A and C) LIUS (25 mW/cm2) or (B and D) heat (40 or 42°C) for 20 min. Immediately after stress treatment, cells were harvested. Either the cytoplasmic or nuclear fraction was separated. Western blotting was carried out using specific primary antibodies against Hsf1, Gapdh and Fbl. (C and D) Each band density of Hsf1 was quantified, and the ratio (nucleus to cytoplasm) was calculated. Gapdh and Fbl served as marker proteins for the cytoplasm and nucleus, respectively. Data are presented as the mean ± SD (n=4). *P<0.05 vs. the Ctr group. LIUS, low-intensity ultrasound; Hsf1, heat shock transcription factor 1; Fbl, fibrillarin; Ctr, control; Cy, cytoplasm; Nu, nucleus; Gapdh, glyceraldehyde 3-phosphate dehydrogenase.

Discussion

LIUS has clear clinical potential for bone fracture healing. However, the molecular mechanisms underlying the ability of bone cells to detect LIUS stimulation have not been well understood. Here, when we used our developed irradiating system (13) to apply a single LIUS irradiation at an ISATA of 25 mW/cm2 for 20 min, which is approximately equivalent to the clinical condition (30 mW/cm2 for 20 min) (1,2), we identified many differentially expressed genes and 3 unique gene networks in LIUS-treated mouse ST2 BMSCs. To our knowledge, this is the first report to identify changes in cellular gene expression as an early cellular response to LIUS.

In the present study, a number of genes (61 upregulated and 103 downregulated) that responded to LIUS at an early time point (0.5 to 3 h) were found to be differentially expressed. Of these 164 genes, the expression levels of 3 genes (Bag3, Dnajb1 and Hspa1) and 4 genes [pentraxin related gene (Ptx3), Gja1, integrin beta 1 (Itgb1) and methionine aminopeptidase 2 (Metap2)] were reported to be affected 3 h after LIUS at 300 mW/cm2 for 1 min in human leukemia Molt-4 (22) or lymphoma U937 cells (23), respectively. Except for these 7 genes, however, none of the genes identified here were reported in previous microarray investigations in which the gene expression patterns were monitored 3–24 h after LIUS (30–300 mW/cm2) in several cell types (18–23).

Particularly noteworthy in this study was the detection of gene network E, which contained 3 IEGs, i.e., Fos, Egr1, and Ptgs2, that were found to be associated mainly with the biological functions of bone physiology. IEGs respond to a wide variety of stresses, including LIUS (9–14). In our previous study, qRT-PCR analysis confirmed the up-regulation of the transient expression of these 3 IEGs in LIUS-treated ST2 cells (13). Fos, Egr1, and Ptgs2 proteins are reported to have essential roles in bone formation (25,29–31). Our results showed that network E also included Bglap (32), Edn1 (33) and Gja1 (34), which are known to be involved in the differentiation of bone cells. It is thus intriguing that interactions have been reported between Fos on the one hand and Bglap (35), Edn1 (36), Gja1 (37) or Ptgs2 (38) on the other.

In this study, gene network H containing Hsf1-regulated genes - 7 HSP genes (39,40), Bag3 (39), Cryab (41) and Ubc (42) - was obtained from up-regulated genes and was associated principally with the biological functions of protein folding and apoptosis. The products translated from these genes are known to exert chaperonic and anti-apoptotic activity (39,40). To our surprise, Ingenuity® pathway analysis indicated that the Hsf1 protein was activated at 0.5 to 3 h after LIUS stimulation (Fig. 4). We confirmed the transient activation of the Hsf1 protein by using Western blot analysis (Figs. 7 and 8). Under our experimental conditions, the medium temperature was increased from 35.3°C (control) to 38.1°C in LIUS-treated cells (25 mW/cm2, 20 min). On the other hand, our previous study showed that heat treatment of the cells at 38°C for 20 min did not affect Fos expression (13). Zhang et al (16) previously reported that upregulation of the Hsp70 protein was observed in the human adipose-derived cells treated with LIUS at 30 mW/cm2 for 30 min, and the temperature in the medium increased from 37°C to 40°C during the LIUS exposure. Very weak but nonetheless significant heat-inducible heat shock element-binding activity of Hsf1 has been reported in cells incubated at 39°C for 20 min (43). Our previous paper also clearly indicated a transient activation of Hsf1 at 0.5–1 h after mild hyperthermia treatment (41°C, for 30 min) in human lymphoma U937 cells (44). It is well known that Hsf1 can be activated by a wide variety of stressors, such as oxidative stress, heavy metals, toxins, and bacterial infections, in addition to heat (39). In addition, under nonthermal-LIUS conditions (300 mW/cm2 for 1 min; temperature rise in medium: 0.3°C), the expression of HSP genes increased significantly in human leukemia Molt-4 cells (22). We considered that the induction of genes such as HSPs in gene network H may have been attributable, at least in part, to the activation of Hsf1 by the slight rise in temperature.

In the present study, LIUS decreased the expression levels of many genes. Previous findings demonstrated that heat activates the transcription of HSPs coincident with a bulk decrease in mRNA and protein syntheses, and this overall reprogramming of gene expression permits the selective synthesis of HSPs in mammalian cells (45). In the present study, we detected gene network T, which contained many transcription factors in down-regulated genes. Functional analysis using the Ingenuity® Pathways Knowledge Base demonstrated that this network was associated with the biological functions of gene expression. For example, Etv5 (46), Fosl1 (47), Glis2 (48) and Srf (49) proteins increase the transcription of promoters by RNA polymerase II complex.

In conclusion, LIUS as shown to elicit gene expression even after short application in mouse ST2 BMSCs. The present results provide a basis for elucidation of the detailed molecular mechanisms underlying the cellular effects of LIUS. However, the biological roles of genes induced by LIUS in either cell or animal models remain a subject for further investigation. Therapeutic LIUS has been used to enhance bone healing caused by fracture in humans (1,2). The effects and possible advantages of ultrasound have been a subject of increasing interest in other fields of medicine as well, including oncology, surgery, gene therapy and regenerative medicine (50–56). Therefore, genetic effects should also be considered to be a factor in these other fields.