Sonoporation-mediated gene transfer into adult rat dorsal root ganglion cells
© Lin et al; licensee BioMed Central Ltd. 2010
Received: 26 February 2010
Accepted: 3 June 2010
Published: 3 June 2010
Gene transfer into many cell types has been successfully used to develop alternative and adjunct approaches to conventional medical treatment. However, effective transfection of postmitotic neurons remains a challenge. The aim of this study was to develop a method for gene transfer into rat primary dorsal root ganglion neurons using sonoporation.
Dissociated cells from adult rat dorsal root ganglion (DRG) cells were sonicated for 1-8 s at 2.5-10 W to determine the optimal ultrasound duration and power for gene transfection and cell survival. Transfection efficiency was compared between sonoporation, liposome and lentiviral vector gene transfer techniques.
The optimum ultrasound intensity was 5 W for 2 s and yielded an efficiency of gene transfection of 31% and a survival rate of 35%.
Sonoporation can be optimized to minimize cell death and yield a high percentage of transfected neurons and that this technique can be easily applied to primary cultures of rat dorsal root ganglion neurons.
Methods for altering gene expression are widely used to elucidate molecular mechanisms involved in cellular physiopathology. Gene modification also has potential as a therapeutic modality for treating many diseases. Many gene transfer methods have been developed in the past two decades, including calcium phosphate coprecipitation, microinjection, recombinant viruses, liposome-mediated gene transfer, lipids that do not form liposomes, high molecular weight cationic polymers, particle bombardment (biollistics) and electroporation.
Ultrasound-mediated gene transfer has recently emerged as a promising technique with a broad range of potential applications. Ultrasound can be used to modify the permeability of the cell membrane to facilitate the uptake of RNA [1–5] and DNA into the cell [6–11]. Low-frequency ultrasound increases membrane permeability to many drugs, including high molecular weight proteins . The degree of macromolecule uptake is correlated with the acoustic energy and frequency of the stimulus . Miller et al.  demonstrated that the uptake of fluorescent dextran by Chinese hamster ovary cells is similar at ultrasound frequencies of 1.0 MHz and 3.3 MHz, but is greatly reduced at 5.3 MHz and 7.15 MHz. Huber and Pfisterer reported that focused ultrasound enhanced the transfer of DNA plasmids into several cell lines in vitro and into a Dunning prostate tumor after direct DNA injection in vivo. They showed that the pressure amplitude and duration of sonication affect transfection ratio and cell survival .
Transfection of postmitotic neurons has been a major challenge in the past. With few exceptions, neuronal transfections have been unreliable, cytotoxic, labor-intensive and inefficient . Although sonoporation-mediated gene transfer into cell lines [8, 17, 18], silkworm larvae , the ovary and uterus , muscle cells , the salivary gland , the joint synovium  and the chicken embryo  has been described, there are few reports on sonoporation-mediated gene transfer into neuronal cells. Shimamura et al. used ultrasound and microbubble-mediated cavitation to transfect rat brain cells . They successfully transfected meningeal and glial cells, but failed to transfect neurons. Manome et al. reported that they transfected neuronal cells in brain slices, but in very small amounts . Fischer et al. used sonoporation to transfect plasmids into the retinal neurons, dorsal forebrain and optic tectum of the chicken, into the cerebellar neurons of the rat and into the hippocampal neurons of the mouse. Although the ability of ultrasound to enhance gene transfer has been demonstrated in numerous studies, its efficacy for gene transfer into the dorsal root ganglion (DRG) has not been determined.
This study was designed to investigate the feasibility of gene transfer into the DRG using a standard laboratory sonicator and to determine the optimal ultrasound duration and power for gene transfection and DRG survival. We also compared transfection (transduction) efficiency between the sonoporation, liposome and lentiviral vector gene transfer techniques.
Materials and methods
The animals were used in accordance with the guidelines of the Chang Gung University.
DRG cells were prepared as previously described . DRGs were dissected in sterile Hanks' buffered saline solution (HBSS) containing 3% D-glucose and 0.01 M HEPES buffer (HBSS+). DRG cells were dissociated by mild trituration using a Pasteur pipette after incubation for 10 min at 37°C in Ca2+/Mg2+-free HBSS containing 0.05% trypsin. The cell suspensions were initially plated into noncoated plastic tissue culture flasks and left for 3 h at room temperature to minimize the number of fibroblasts in the cultures. The supernatant, which contained DRG neurons and satellite glial cells, was then aspirated. Cell density was determined using a hemacytometer. Between 100,000 and 200,000 cells were plated onto 12 mm glass cover slips that were coated sequentially with poly-D-lysine and laminin (Sigma) diluted 1:100 in HBSS. Cell cultures were maintained at 37°C under a 5% CO2 atmosphere in culture medium (Neurobasal Media, Gibco) containing N-2 supplement (Gibco), 100 U/ml penicillin, 100 mg/ml streptomycin and 2% fetal bovine serum (Gibco BRL).
Enhanced green fluorescence protein (E-GFP)-containing lentiviral vectors were obtained from the National RNAi Core Facility, Academia Sinica, Taiwan. Lentiviruses were prepared according to a standard protocol. Titers were assayed using 3T3 cells and serial dilutions of the vector preparations. The titers of these vector stocks were also estimated by measuring the level of viral p24gag antigen using a human immunodeficiency virus-1 p24 antigen assay kit (Beckman Coulter, Fullerton, CA).
Liposome-mediated gene transfer
Lipofectamine 2000 (Invitrogen) was mixed with plasmid E-GFP DNA (pE-GFP C1) in the ratio of 3 μl per 2 μg, respectively, in 50 μl of HBSS+. The vector is described elsewhere . The optimum ratio of lipofectamine to pE-GFP C1 was determined according to the manufacturer's instructions. These reagents were incubated at room temperature for 15 min and were then added to a solution containing freshly dissected DRG cells. The medium was replaced with Lipofectamine 2000 after 12 h.
We used a Sonics and Materials VC130 sonicator (Newtown, CT), which produces continuous wave ultrasound at 20 kHz and has an adjustable output range of 2.5-130 W and a 12 mm diameter probe tip. pE-GFP C1 was diluted in HBSS+, added to the wells at concentrations of 0.5-20 μg/ml and incubated for 5-10 min at 37°C. The sonicator probe was sterilized by sonication in 70% ethanol followed by sonication in sterile water. It was then placed into the wells, each of which contained 1000 μl of medium, and activated for 1-8 s, which delivered 2.5-10 W of energy. Cells were cultured for 1-14 d after sonoporation and were then fixed and processed for immunocytochemical labeling using standard methods .
Cell survival assay
Cell survival after sonoporation was assessed using the Live/Dead assay (Molecular Probes) according to the manufacturer's specifications. DRG cells were maintained in vitro for 48 h after sonication. The medium was then aspirated and cells were rinsed twice with HBSS+. After adding 200 μl of HBSS+ containing 1 μM ethidium homodimer and 1 μM calcein AM, the cells were incubated for 30 min at 37°C, rinsed twice in HBSS+ and counted using a fluorescence microscope.
Fixation and immunocytochemistry
Cells were fixed and immunolabeled as described previously . The primary antibodies were mouse anti-β-tubulin isotype III (1:1000; Sigma) and rabbit anti-E-GFP (1:1000; Sigma). The secondary antibodies were goat anti-rabbit Alexa568, goat anti-mouse Alexa568 and goat anti-rat Alexa488 (Molecular Probes Inc., Eugene, OR) and were diluted 1:1000 in PBS containing 0.2% Triton X-100. Cells were stained using 1 μg/ml of DAPI (Sigma) in PBS and then mounted on cover slips using Fluoromount-G.
Means for experimental variables were compared using ANOVA and Student's t-test post hoc. Means and standard errors for cell numbers were calculated by counting cells in at least five view fields from at least three cover slips per treatment.
The optimal power for sonoporation of DRG cells is 5 W
The optimal duration of sonoporation of DRG cells is 2 s
To determine the optimal duration of sonoporation, we conducted another series of experiments using DRG cells. Dissociated cells from adult rat DRGs were suspended at a concentration of 10,000 cells per cm3 and 10 μg/ml of pE-GFP C1 was added to the medium. The suspensions were then sonicated at 5 W for 1-8 s. The cells were maintained in vitro for 48 h after sonoporation. The number of GFP-positive cells increased up to a sonoporation duration of 2 s (Figure 1E).
The survival rate of DRG cells after sonoporation for 2 s at 5 W is 35%
The number of dead cells after sonoporation was assessed according to the accumulation of ethidium homodimer and the number of live cells was assessed according to the accumulation of calcein AM. The number of dead cells after sonication for 1 s at 2.5 W was 35% of that of the unsonicated control. Sonication for 1 s at 5 W or 10 W increased the number of dead cells to 43% of that of the control. At 5 W, sonication for 2 s, 4 s and 8 s increased the number of dead cells to 65%, 87% and 93%, respectively, of that of the control (Figure 2).
Comparison of transfection (transduction) efficiency between the sonoporation, liposome and lentiviral vector methods
Percentage of neuronal cells transfected using sonoporation
Transfection of postmitotic neurons is labor-intensive, inefficient, unreliable and may have cytotoxic effects. The inability to express foreign proteins in postmitotic neurons has hampered neuroscience research . Our results show that sonoporation is a feasible in vitro method for gene transfer into cultured DRG cells from adult rats. These results were achieved using a standard laboratory sonicator of the type used to disrupt cells and homogenize solutions. Sonoporation for 2 s at 5 W resulted in optimum transfection efficiency (31%) and a cell survival rate equivalent to 35% of that of the control.
Virus-based methods are the most successful of effecting gene transfer to neuronal cells [16, 28]. The increasing use of viral vectors to transfer DNA into neurons has arisen because of their high infection efficiencies compared with nonviral methods. However, preparation of recombinant viruses is expensive and labor-intensive. Other limitations of this method are its potential toxicity for neurons, a DNA expression cassette of limited size and the production of severe immune reactions in vivo. These approaches also constitute a potential health hazard for laboratory personnel [29–32]. Furthermore, serious concerns about the insertional mutagenesis have arisen, especially concerning the use of viral vectors when clinical trials are involved [11, 33]. By contrast, nonviral methods such as naked plasmid DNA injection, electroporation, and sonoporation should have a higher potential for clinical application, even although their efficacy for gene delivery is lower .
Liposome-mediated gene transfer involves the fusion of synthetic lipids into the plasmid membrane, which may affect cell membrane proteins. Therefore, sonoporation is preferable to liposome-mediated gene transfer for studying expression of transmembrane proteins. Furthermore, the efficacy of liposomal transfection is less than 1%.
Although gene transfer via electroporation in vivo is effective using DNA injection followed by the application of electric fields, the tissue damage caused by the electric pulse is problematic for cell survival. Ultrasound, on the other hand, makes the cell membrane porous and enhances the intracellular delivery of naked DNA in vitro. The membrane damage induced by ultrasound is transient and the holes (or pores) can reseal and allow survival of the cells. During sonoporation, large molecules in the medium can leak into the cells and remain trapped there after the membrane reseals [17, 35]. Sonoporation has opened tremendous opportunities for targeted gene transfer. Conceptually, gene vectors mixed with ultrasound contrast agents could be injected into animal cells and targeted gene transfer could be achieved by selective application to a predefined area. Indeed, promising results have now been reported in animal models . By using this approach, the risk of systemic exposure (a major drawback of current clinical gene transfer protocols) could be reduced substantially reduced.
Sonoporation of freshly dissected DRG cells was highly selective for neuronal cells. Neuronal cells constituted a much higher percentage of the total number of transfected DRG cells with the sonoporation method than with the lentivirus method. The percentage of sonoporated neuronal cells depended on energy level; under optimal sonoporation conditions, 75% of sonoporated DRG cells were neuronal cells. The mechanism underlying the preferred transfection of neuronal cells by sonoporation remains unknown. Sonoporation facilitates the entry of macromolecules into cells via microbubble-mediated cavitation and transient disruption of the plasma membrane [37, 38]. As the average diameter of neuronal cells is larger than that of glial cells, it is possible that the effects of ultrasound frequency and intensity depend on cell diameter.
Sonoporation is an alternative method for transferring naked plasmid DNA into neuronal cells and may avoid side effects associated with other methods. We found that sonoporated cells maintained transgene expression for at least 2 weeks after treatment. In addition, sonoporation did not harm normal cellular functions such as the outgrowth of dendrites and axons, and all of these cell types had well-developed dendritic and axonal arbors. Neuronal cells with elaborate neurites can be transfected via sonoporation without physical disruption . Thus, sonoporation does not result in genomic instability or other forms of permanent cellular damage that would limit sonoporation to short-term applications.
However, our study had several limitations. We have not evaluated a comprehensive comparison of ultrasound modalities and contrast agents. Indeed, the number of possible combinations of ultrasound field characteristics (e.g., continuous versus pulsed wave, frequency or duty cycle) and contrast microbubbles is enormous. Therefore the optimal conditions for gene transfer by sonoporation require further investigation.
This study demonstrates that sonoporation enables delivery of plasmids into dorsal root ganglion cells in vitro. Sonoporation is a simple and economic method for studying physiological pathways in DRG cells in vitro. Sonoporation of dorsal root ganglion cells allows the preferential transfection of neuronal cells. Therefore, we propose that sonoporation could be applied to the intact nervous system to transfer foreign DNA for physiological research.
This work was supported in part by grant No. 870641, 860231, and 860232 from Chang Gung Memorial Hospital Research, Kaohsiung, Taiwan, and by grant No. 95-2745-B-182A-004, 96-2628-B-182A-005-MY3, and 98-2314-B-182A-035-MY2 from the Taiwan National Science Council Research, Taipei, Taiwan.
- Cheon SH, Lee KH, Kwon JY, Choi SH, Song MN, Kim DI: Enhanced delivery of siRNA complexes by sonoporation in transgenic rice cell suspension cultures. J Microbiol Biotechnol. 2009, 19: 781-786.PubMedGoogle Scholar
- Sakai T, Kawaguchi M, Kosuge Y: siRNA-mediated gene silencing in the salivary gland using in vivo microbubble-enhanced sonoporation. Oral Dis. 2009, 15: 505-511. 10.1111/j.1601-0825.2009.01579.x.View ArticlePubMedGoogle Scholar
- Kinoshita M, Hynynen K: Key factors that affect sonoporation efficiency in in vitro settings: the importance of standing wave in sonoporation. Biochem Biophys Res Commun. 2007, 359: 860-865. 10.1016/j.bbrc.2007.05.153.PubMed CentralView ArticlePubMedGoogle Scholar
- Saito M, Mazda O, Takahashi KA, Arai Y, Kishida T, Shin-Ya M, Inoue A, Tonomura H, Sakao K, Morihara T, Imanishi J, Kawata M, Kubo T: Sonoporation mediated transduction of pDNA/siRNA into joint synovium in vivo. J Orthop Res. 2007, 25: 1308-1316. 10.1002/jor.20392.View ArticlePubMedGoogle Scholar
- Tsunoda S, Mazda O, Oda Y, Iida Y, Akabame S, Kishida T, Shin-Ya M, Asada H, Gojo S, Imanishi J, Matsubara H, Yoshikawa T: Sonoporation using microbubble BR14 promotes pDNA/siRNA transduction to murine heart. Biochem Biophys Res Commun. 2005, 336: 118-127. 10.1016/j.bbrc.2005.08.052.View ArticlePubMedGoogle Scholar
- Newman CM, Bettinger T: Gene therapy progress and prospects: ultrasound for gene transfer. Gene Ther. 2007, 14: 465-475. 10.1038/sj.gt.3302925.View ArticlePubMedGoogle Scholar
- Manome Y, Nakayama N, Nakayama K, Furuhata H: Insonation facilitates plasmid DNA transfection into the central nervous system and microbubbles enhance the effect. Ultrasound Med Biol. 2005, 31: 693-702. 10.1016/j.ultrasmedbio.2005.01.015.View ArticlePubMedGoogle Scholar
- Liang HD, Lu QL, Xue SA, Halliwell M, Kodama T, Cosgrove DO, Stauss HJ, Partridge TA, Blomley MJ: Optimisation of ultrasound-mediated gene transfer (sonoporation) in skeletal muscle cells. Ultrasound Med Biol. 2004, 30: 1523-1529. 10.1016/j.ultrasmedbio.2004.08.021.View ArticlePubMedGoogle Scholar
- Zarnitsyn VG, Prausnitz MR: Physical parameters influencing optimization of ultrasound-mediated DNA transfection. Ultrasound Med Biol. 2004, 30: 527-538. 10.1016/j.ultrasmedbio.2004.01.008.View ArticlePubMedGoogle Scholar
- Niidome T, Huang L: Gene therapy progress and prospects: nonviral vectors. Gene Ther. 2002, 9: 1647-1652. 10.1038/sj.gt.3301923.View ArticlePubMedGoogle Scholar
- Taniyama Y, Tachibana K, Hiraoka K, Aoki M, Yamamoto S, Matsumoto K, Nakamura T, Ogihara T, Kaneda Y, Morishita R: Development of safe and efficient novel nonviral gene transfer using ultrasound: enhancement of transfection efficiency of naked plasmid DNA in skeletal muscle. Gene Ther. 2002, 9: 372-380. 10.1038/sj.gt.3301678.View ArticlePubMedGoogle Scholar
- Mitragotri S, Blankschtein D, Langer R: Ultrasound-mediated transdermal protein delivery. Science. 1995, 269: 850-853. 10.1126/science.7638603.View ArticlePubMedGoogle Scholar
- Guzman HR, Nguyen DX, McNamara AJ, Prausnitz MR: Equilibrium loading of cells with macromolecules by ultrasound: effects of molecular size and acoustic energy. J Pharm Sci. 2002, 91: 1693-1701. 10.1002/jps.10156.View ArticlePubMedGoogle Scholar
- Miller DL, Bao S, Morris JE: Sonoporation of cultured cells in the rotating tube exposure system. Ultrasound Med Biol. 1999, 25: 143-149. 10.1016/S0301-5629(98)00137-9.View ArticlePubMedGoogle Scholar
- Huber PE, Pfisterer P: In vitro and in vivo transfection of plasmid DNA in the Dunning prostate tumor R3327-AT1 is enhanced by focused ultrasound. Gene Ther. 2000, 7: 1516-1525. 10.1038/sj.gt.3301242.View ArticlePubMedGoogle Scholar
- Washbourne P, McAllister AK: Techniques for gene transfer into neurons. Curr Opin Neurobiol. 2002, 12: 566-573. 10.1016/S0959-4388(02)00365-3.View ArticlePubMedGoogle Scholar
- Bao S, Thrall BD, Miller DL: Transfection of a reporter plasmid into cultured cells by sonoporation in vitro. Ultrasound Med Biol. 1997, 23: 953-959. 10.1016/S0301-5629(97)00025-2.View ArticlePubMedGoogle Scholar
- Miller DL, Quddus J: Sonoporation of monolayer cells by diagnostic ultrasound activation of contrast-agent gas bodies. Ultrasound Med Biol. 2000, 26: 661-667. 10.1016/S0301-5629(99)00170-2.View ArticlePubMedGoogle Scholar
- Lee JM, Takahashi M, Mon H, Koga K, Kawaguchi Y, Kusakabe T: Efficient gene transfer into silkworm larval tissues by a combination of sonoporation and lipofection. Cell Biol Int. 2005, 29: 976-979. 10.1016/j.cellbi.2005.07.007.View ArticlePubMedGoogle Scholar
- Lei L, Liu ZH, Ohta S, Yamada G: [In vivo transduction of EGFP into female mouse reproductive system by electroporation and microbubble-enhanced sonoporation]. Fen Zi Xi Bao Sheng Wu Xue Bao. 2006, 39: 378-382. 10.3736/jcim20060412.PubMedGoogle Scholar
- Tsai KC, Liao ZK, Yang SJ, Lin WL, Shieh MJ, Hwang LH, Chen WS: Differences in gene expression between sonoporation in tumor and in muscle. J Gene Med. 2009Google Scholar
- Sakai T, Kawaguchi M, Kosuge Y: siRNA-mediated gene silencing in the salivary gland using in vivo microbubble-enhanced sonoporation. Oral Dis. 2009Google Scholar
- Ohta S, Suzuki K, Tachibana K, Yamada G: Microbubble-enhanced sonoporation: efficient gene transduction technique for chick embryos. Genesis. 2003, 37: 91-101. 10.1002/gene.10232.View ArticlePubMedGoogle Scholar
- Shimamura M, Sato N, Taniyama Y, Yamamoto S, Endoh M, Kurinami H, Aoki M, Ogihara T, Kaneda Y, Morishita R: Development of efficient plasmid DNA transfer into adult rat central nervous system using microbubble-enhanced ultrasound. Gene Ther. 2004, 11: 1532-1539. 10.1038/sj.gt.3302323.View ArticlePubMedGoogle Scholar
- Lin CR, Amaya F, Barrett L, Wang H, Takada J, Samad TA, Woolf CJ: Prostaglandin E2 receptor EP4 contributes to inflammatory pain hypersensitivity. J Pharmacol Exp Ther. 2006, 319: 1096-1103. 10.1124/jpet.106.105569.View ArticlePubMedGoogle Scholar
- Lin CR, Tai MH, Cheng JT, Chou AK, Wang JJ, Tan PH, Marsala M, Yang LC: Electroporation for direct spinal gene transfer in rats. Neurosci Lett. 2002, 317: 1-4. 10.1016/S0304-3940(01)02354-0.View ArticlePubMedGoogle Scholar
- Lin CR, Wu PC, Shih HC, Cheng JT, Lu CY, Chou AK, Yang LC: Intrathecal spinal progenitor cell transplantation for the treatment of neuropathic pain. Cell Transplant. 2002, 11: 17-24.PubMedGoogle Scholar
- Berry M, Barrett L, Seymour L, Baird A, Logan A: Gene therapy for central nervous system repair. Curr Opin Mol Ther. 2001, 3: 338-349.PubMedGoogle Scholar
- Slack RS, Miller FD: Viral vectors for modulating gene expression in neurons. Curr Opin Neurobiol. 1996, 6: 576-583. 10.1016/S0959-4388(96)80088-2.View ArticlePubMedGoogle Scholar
- Simonato M, Manservigi R, Marconi P, Glorioso J: Gene transfer into neurones for the molecular analysis of behaviour: focus on herpes simplex vectors. Trends Neurosci. 2000, 23: 183-190. 10.1016/S0166-2236(99)01539-8.View ArticlePubMedGoogle Scholar
- Ehrengruber MU, Hennou S, Bueler H, Naim HY, Deglon N, Lundstrom K: Gene transfer into neurons from hippocampal slices: comparison of recombinant Semliki Forest Virus, adenovirus, adeno-associated virus, lentivirus, and measles virus. Mol Cell Neurosci. 2001, 17: 855-871. 10.1006/mcne.2001.0982.View ArticlePubMedGoogle Scholar
- Janson CG, McPhee SW, Leone P, Freese A, During MJ: Viral-based gene transfer to the mammalian CNS for functional genomic studies. Trends Neurosci. 2001, 24: 706-712. 10.1016/S0166-2236(00)01954-8.View ArticlePubMedGoogle Scholar
- Marshall E: Gene therapy death prompts review of adenovirus vector. Science. 1999, 286: 2244-2245. 10.1126/science.286.5448.2244.View ArticlePubMedGoogle Scholar
- Baque P, Pierrefite-Carle V, Gavelli A, Brossette N, Benchimol D, Bourgeon A, Staccini P, Saint-Paul MC, Rossi B: Naked DNA injection for liver metastases treatment in rats. Hepatology. 2002, 35: 1144-1152. 10.1053/jhep.2002.32709.View ArticlePubMedGoogle Scholar
- Bao S, Thrall BD, Gies RA, Miller DL: In vivo transfection of melanoma cells by lithotripter shock waves. Cancer Res. 1998, 58: 219-221.PubMedGoogle Scholar
- Shohet RV, Chen S, Zhou YT, Wang Z, Meidell RS, Unger RH, Grayburn PA: Echocardiographic destruction of albumin microbubbles directs gene delivery to the myocardium. Circulation. 2000, 101: 2554-2556.View ArticlePubMedGoogle Scholar
- Marmottant P, Hilgenfeldt S: Controlled vesicle deformation and lysis by single oscillating bubbles. Nature. 2003, 423: 153-156. 10.1038/nature01613.View ArticlePubMedGoogle Scholar
- Miller DL, Pislaru SV, Greenleaf JE: Sonoporation: mechanical DNA delivery by ultrasonic cavitation. Somat Cell Mol Genet. 2002, 27: 115-134. 10.1023/A:1022983907223.View ArticlePubMedGoogle Scholar
- Fischer AJ, Stanke JJ, Omar G, Askwith CC, Burry RW: Ultrasound-mediated gene transfer into neuronal cells. J Biotechnol. 2006, 122: 393-411. 10.1016/j.jbiotec.2005.10.006.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.