Regulation of PMP22 mRNA by G3BP1 affects cell proliferation in breast cancer cells
© Winslow et al.; licensee BioMed Central Ltd. 2013
Received: 21 May 2013
Accepted: 27 November 2013
Published: 9 December 2013
Regulation of mRNAs is one way to control protein levels and thereby important cellular processes such as growth, invasion and apoptosis. G3BPs constitute a family of mRNA-binding proteins, shown to be overexpressed in several cancer types, including breast, colon and pancreas cancer. G3BP has been reported to both stabilize and induce degradation of specific mRNAs.
Here, we show that G3BP1, but not G3BP2, supports proliferation of several breast cancer cell lines. Global gene expression analyses of G3BP1- and G3BP2-depleted cells indicate that primarily G3BP1, and much less G3BP2, influences mRNA expression levels. Peripheral myelin protein 22 (PMP22) was one gene that was significantly influenced by G3BP1 depletion which led to a 2–3 fold increased expression. Depletion of PMP22 resulted in increased proliferation and the G3BP1-mediated effect on proliferation was not seen upon PMP22-depletion.
This indicates a novel role for G3BP1 in the regulation of cell proliferation in breast cancer cells, perhaps via a regulatory effect on PMP22 expression.
KeywordsG3BP Breast cancer cells PMP22 Proliferation Gene expression
Ras-GTPase activating protein SH3 domain binding proteins
Peripheral myelin protein 22
RNA recognition motif
- PNGase F:
Peptide -N-Glycosidase F
Regulation of mRNA levels and mRNA translation is important since these processes to a large extent determine protein expression . Regulation of mRNAs is frequently mediated by factors binding to AU-rich elements (ARE) in the 3’UTR . Ras-GTPase activating protein SH3 domain binding proteins (G3BPs) constitute one group of such mRNA-binding proteins and consist of a family of three homologous proteins (G3BP1, G3BP2a and G3BP2b) . The name originates from early reports of G3BP1 interacting with the SH3 domain of RasGAP , even though more recent studies have questioned this finding . All G3BP proteins contain a RNA recognition motif (RRM) and have been shown to have both mRNA-stabilizing effects, exemplified by TAU mRNA  as well as mRNA-degrading effects, as demonstrated for c-MYC, BART, CTNNB1, ATP5B, IGF-II, and GAS5.The degrading effect has been indicated to be mediated by endonuclease activity of the G3BPs themselves .
The interactions between G3BPs and their target mRNAs have been shown to be important for regulation of processes that can influence cancer, such as cell growth and motility [7, 8]. Overexpression of G3BP1 can promote S-phase entry in a RNA-binding domain-dependent way in fibroblasts  and up-regulation of G3BP1 has been detected in proliferating retinal pigment epithelial cells . A potential role for G3BPs in cancer is indicated by the finding that they have been found to be expressed at high levels in many different tumor types e.g. breast [13, 15, 16], pancreas , thyroid, colon, head and neck tumors  as well as in several cancer cell lines [8, 13].
The aim of this study was to investigate a putative role for G3BPs in breast cancer cell growth. G3BP1 and/or G3BP2 were depleted and the effects on cell growth and global gene expression were analyzed. We found that G3BP1 to a larger extent than G3BP2 influences mRNA expression levels and breast cancer cell proliferation. Peripheral myelin protein 22, PMP22, was one gene that was influenced by G3BP1 levels and potentially mediates the G3BP1 effect on cell proliferation.
G3BP1 depletion decreases cell proliferation
In addition, the total number of viable cells was quantified after 96 hours in culture in the presence of siRNAs targeting G3BP1 and/or G3BP2 (Figure 1G and H). For both MCF-7 and MDA-MB-468 cells, depletion of G3BP1 + G3BP2 led to reduced number of cells. Targeting only G3BP1 led to a significant reduction of MDA-MB-468 cells and a tendency to lower MCF-7 cell number. Downregulation of G3BPs in MCF-7 cells was confirmed by Western blot (Figure 1I-K). Thus, G3BP1, but not G3BP2, seems to be a proliferation-promoting factor in breast cancer cells.
G3BP-depletion does not influence cell death in breast cancer cell lines
Gene expression profiling indicates different roles for G3BP1 and G3BP2 in breast cancer cells
G3BPs are considered to be RNA-binding proteins. To analyze whether altered mRNA expression may be involved in the effects of G3BP knockdown seen on cell proliferation, global gene expression in G3BP1- and/or G3BP2-depleted MCF-7 breast cancer cells was analyzed with microarray. Clustering of the expression data revealed that control cells and siG3BP2-treated cells largely clustered together (Additional file 1: Figure S1). Only one control sample diverged from this pattern. Likewise, cells treated with siG3BP1 or the combination of siG3BP1 and siG3BP2 oligonucleotides co-clustered, except for one siG3BP1 sample. The pattern was also seen when the expression levels of individual genes were compared (Additional file 2: Table S1). No genes were found to be significantly altered by downregulation of G3BP2. On the other hand, downregulation of G3BP1 led to alterations in the expression of several genes. The effect of siG3BP1 was seen both when it was used as a single agent and when it was added together with siG3BP2. This suggests that G3BP1 and G3BP2 differ in their effects on mRNA expression pattern. G3BP2 only seems to have minor effects on mRNA levels.
G3BP1 suppresses PMP22 mRNA expression
To analyze whether the effect on PMP22 mRNA is seen in other breast cancer cells as well, three additional cell lines (MDA-MB-231, MDA-MB-468 and BT549) were treated with siRNAs targeting G3BP1 and/or G3BP2 (Figure 3C-E). Knockdown of G3BP1 alone or in combination with G3BP2-depletion lead to a significant increase in PMP22 mRNA levels in all cells.
Several efforts were made to analyze PMP22 protein levels. However, two separate antibodies, or treatment with PNGase F to deglycosylate the protein did not lead to detection of specific bands in Western blot analyses.
Knockdown of G3BP does not influence the mRNA stability of PMP22
Determination of PMP22 mRNA stability in G3BP-depleted cells after actinomycin D treatment
(Mean ± SEM, n = 3)
4.4 ± 0.67
4.2 ± 0.09
3.9 ± 0.26
4.3 ± 0.48
G3BP increases cell proliferation
Suppression of PMP22 expression facilitates proliferation
Functional studies of G3BPs have indicated a role for these RNA-binding proteins in cell growth. Here we show that G3BP1 is important for optimal breast cancer cell proliferation. This is in line with findings that downregulated levels of G3BP1 lead to suppressed growth in colon carcinoma cells  and that mice with G3BP1 gene deletion have decreased fetal growth and higher embryonic lethality . However, we could not find any evidence for a survival role for G3BP1 as has been suggested in the same studies. In concordance with the effect on cell growth, G3BP1 [8, 13, 15], but also G3BP2  has been reported to be up-regulated in various tumor types and higher levels of G3BP1 have been reported in proliferating retinal epithelial cells . We could only see effects on proliferation following G3BP1 depletion and not upon G3BP2 depletion, indicating that G3BP1 and G3BP2 may have different functions in breast cancer cells.
G3BP1 and G3BP2 have around 70% sequence similarity , suggesting they may have similar functions. Both G3BP1 and G3BP2 promote stress granule assembly [21, 22] and interact with protein kinase Cα  and both G3BPs have RNA-binding motifs. However, effects on post-transcriptional regulation has this far only been reported for G3BP1 [7–12]. In most cases this results in downregulation of the target mRNAs, except for TAU which was reported to be stabilized by G3BP1 . None of the reported G3BP1 targets were significantly altered in our global gene expression analysis, suggesting that the effects may depend on cellular context.
So far little is known about G3BP2 effects on gene expression. Our gene expression data indicate that G3BP1, but not G3BP2, influences mRNA levels. Thus, both proliferation and levels of mRNAs are primarily influenced by G3BP1 and not by G3BP2. There are not many other reports demonstrating differences between G3BP1 and G3BP2. However, there may be subtle differences in their subcellular localization. Although both proteins are primarily cytosolic, G3BP2 can localize to the nucleus after serum stimulation  whereas G3BP1 can be found in the nucleus of quiescent cells as well [7, 12].
The fact that G3BP1-specific effects on expression levels of mRNAs, as assayed by a global analysis, parallels effects on cell proliferation raises the hypothesis that some of the affected mRNAs may mediate the effects of G3BP1 on proliferation.
In this study we identified PMP22 mRNA levels as being influenced by G3BP1 in breast cancer cell lines. PMP22 is a 22 kDa transmembrane glycoprotein first identified as a growth arrest specific gene (Gas3). The protein is a major component of myelin  and has been associated with demyelinating neuropathies [25–28]. However, there are also reports connecting PMP22 to cancer. For instance, the PMP22 gene was found to be amplified in two glioblastoma cell lines and an unusual transcript was found in the cells , PMP22 mRNA was higher in pancreatic adenocarcinomas than in normal tissue and was detected in cell lines , and high PMP22 mRNA in tumors correlated to poorer survival of breast cancer patients . When it comes to cellular effects, mRNA expression has primarily been shown to be inversely correlated to cell growth [18, 32–34] and overexpression of PMP22 leads to delayed G0/G1 to S-phase transition in Schwann cells . Here, we show that PMP22 depletion results in increased proliferation in breast cancer cells, which supports the role of PMP22 as a growth arrest-specific gene. The effect on proliferation by PMP22 is also in line with a hypothesis that decreased PMP22 levels may mediate G3BP1-promoting effects on proliferation.
PMP22 has been reported to be post-transcriptionally regulated in the 3’UTR in Schwann cells by miRNA  and G3BP1 has been shown to affect mRNA levels by regulating their stability. However, our results with actinomycin D indicate no variation in stability of PMP22 mRNA in G3BP-depleted cells compared to control siRNA. The effect on PMP22 mRNA levels may therefore be due to a more indirect effect on transcriptional regulation.
In conclusion, our study demonstrates specific effects of G3BP1 and G3BP2 in breast cancer cells. G3BP1, but not G3BP2, supports proliferation and influences mRNA levels of different genes. We identified PMP22 as a gene that is suppressed by G3BP1 and this could be one mechanism that mediates the G3BP1 effect on proliferation.
MCF-7, MDA-MB-231, MDA-MB-468 and BT549 cell lines were all cultured in RPMI 1640 medium (Sigma) supplemented with 10% fetal bovine serum (EuroClone), 100 IU/ml penicillin (Gibco), and 100 μg/ml streptomycin (Gibco) and 1% sodium pyruvate (Gibco). BT549 medium was additionally supplemented with 0.023 IU/ml insulin. Cells were maintained in a humidified atmosphere with 5% CO2 at 37°C and seeded at a density of 1.5×105 cells/35-mm dish or 0.5×105 cells/12-well plate. For siRNA transfections, cells were seeded at a density of 1.2×105 cells/35-mm dish. Actinomycin D (Sigma) was used at 4 μg/ml when indicated and cells from three experiments were harvested at 0, 2, 4 or 8 hours. PMP22 mRNA half-times were calculated by doing a non-linear curve fit to the equation y = a*2-x/b using the nls function in R.
Plasmids and transfection
Control 48% GC
Cells were lysed for 15 minutes on ice with 150 μl RIPA buffer (10 mM Tris–HCl pH 7.2, 160 mM NaCl, 1% Triton X-100, 1% Na-deoxycholate, 0.1% SDS, 1 mM EGTA and 1 mM EDTA) supplemented with complete protease inhibitor cocktail without EDTA (Roche). Lysates were cleared by centrifugation at 14000 × g for 15 min at 4°C and sample buffer was added to the supernatant. Proteins were separated by SDS-PAGE and transferred to a PVDF membrane (Millipore). Membranes were pre-incubated with 5% dried milk in PBS followed by incubation with primary antibodies. Membranes were washed and incubated with horseradish peroxidase-labeled secondary antibody, which were detected with the SuperSignal system (Biological Industries), enhanced chemiluminescence detection system (GE Healthcare) or SuperSignal (Pierce). The chemiluminescence was captured with a charge-coupled device camera (Fujifilm). Anti-G3BP2 polyclonal antibody has been described . Antibodies against G3BP1, FLAG and actin were purchased from BD Bioscience, Sigma and Santa Cruz, respectively.
Proliferation assay and analysis of cell number
For the proliferation assay, cells were seeded in triplicate at a density of 5×104 cells/well in 12-well plates and transfected with 1 μg plasmid DNA or 40 nM siRNA in 1 ml antibiotics-free medium for 24 or 72 hours. Cells were incubated with 1 μCi/ml [3H]-thymidine for 6 hours before harvesting the cells with 10 mM EDTA. The amount of radioactivity was measured with a Tri-carb 2810TR liquid scintillation analyzer (Perkin Elmer). The number of viable cells was quantified following, suspending the cells, addition of trypan blue (Sigma-Aldrich), and counting of trypan blue-negative cells.
Cells were detached from the dish using trypsin and washed twice in PBS prior to Annexin V-APC staining according to supplier’s protocol (BD, Pharmingen). 10000 events were acquired in channel FL-4 with a FACSCalibur (Becton Dickinson). Analyses were performed with CellQuest software (Becton Dickinson).
Cells were cultured on coverslips for 24 hours prior to transfection with 40 nM siRNA according to standard procedure. After 72 hours, cells were washed with PBS and fixed with 4% paraformaldehyde in PBS for 4 minutes followed by two additional washing steps. Coverslips were mounted onto slides with VECTASHIELD mounting medium containing DAPI as counterstain (Vector Laboratories) and sealed with nail polish. For each cover slip 200 nuclei were scored for apoptotic morphology.
Quantitative real-time PCR
Primers for qPCR
Sequence 5′ to 3′
Illumina DirectHyb HumanHT-12 v4.0 whole genome microarray was performed at the SCIBLU facility at Lund University. RNA was extracted from MCF-7 cells as described above, and evaluated on a Bioanalyzer 2100 (Agilent). Microarray data were background-corrected and normalized using the lumiB and lumiN functions from the lumi package in R . Clustering was done using the AGNES (Agglomerative Hierarchical clustering algorithm) function in the cluster package. Differences in expression levels of individual genes were analyzed by fitting to a linear model followed by ranking using an empirical Bayes method (lmFit and eBayes functions in the limma package of R).
This work was supported by grants from the Swedish Research Council, the Swedish Cancer Society, the Gunnar Nilsson, Ollie and Elof Ericsson, and Kock foundations, and Malmö University Hospital research funds.
- Schoenberg DR: Mechanisms of endonuclease-mediated mRNA decay. Wiley Interdiscip Rev RNA. 2011, 2: 582-600. 10.1002/wrna.78PubMed CentralView ArticlePubMed
- Wurth L: Versatility of RNA-Binding proteins in cancer. Comp Funct Genomics. 2012, 2012: 178525-PubMed CentralView ArticlePubMed
- Kennedy D, French J, Guitard E, Ru K, Tocque B, Mattick J: Characterization of G3BPs: tissue specific expression, chromosomal localisation and rasGAP(120) binding studies. J Cell Biochem. 2001, 84: 173-187.View ArticlePubMed
- Parker F, Maurier F, Delumeau I, Duchesne M, Faucher D, Debussche L, Dugue A, Schweighoffer F, Tocque B: A Ras-GTPase-activating protein SH3-domain-binding protein. Mol Cell Biol. 1996, 16: 2561-2569.PubMed CentralView ArticlePubMed
- Annibaldi A, Dousse A, Martin S, Tazi J, Widmann C: Revisiting G3BP1 as a RasGAP binding protein: sensitization of tumor cells to chemotherapy by the RasGAP 317–326 sequence does not involve G3BP1. PLoS One. 2011, 6: e29024- 10.1371/journal.pone.0029024PubMed CentralView ArticlePubMed
- Atlas R, Behar L, Elliott E, Ginzburg I: The insulin-like growth factor mRNA binding-protein IMP-1 and the Ras-regulatory protein G3BP associate with tau mRNA and HuD protein in differentiated P19 neuronal cells. J Neurochem. 2004, 89: 613-626. 10.1111/j.1471-4159.2004.02371.xView ArticlePubMed
- Gallouzi IE, Parker F, Chebli K, Maurier F, Labourier E, Barlat I, Capony JP, Tocque B, Tazi J: A novel phosphorylation-dependent RNase activity of GAP-SH3 binding protein: a potential link between signal transduction and RNA stability. Mol Cell Biol. 1998, 18: 3956-3965.PubMed CentralView ArticlePubMed
- Taniuchi K, Nishimori I, Hollingsworth MA: The N-terminal domain of G3BP enhances cell motility and invasion by posttranscriptional regulation of BART. Mol Cancer Res. 2011, 9: 856-866. 10.1158/1541-7786.MCR-10-0574View ArticlePubMed
- Bikkavilli RK, Malbon CC: Arginine methylation of G3BP1 in response to Wnt3a regulates beta-catenin mRNA. J Cell Sci. 2011, 124: 2310-2320. 10.1242/jcs.084046PubMed CentralView ArticlePubMed
- Ortega AD, Willers IM, Sala S, Cuezva JM: Human G3BP1 interacts with beta-F1-ATPase mRNA and inhibits its translation. J Cell Sci. 2010, 123: 2685-2696. 10.1242/jcs.065920View ArticlePubMed
- Zekri L, Chebli K, Tourriere H, Nielsen FC, Hansen TV, Rami A, Tazi J: Control of fetal growth and neonatal survival by the RasGAP-associated endoribonuclease G3BP. Mol Cell Biol. 2005, 25: 8703-8716. 10.1128/MCB.25.19.8703-8716.2005PubMed CentralView ArticlePubMed
- Tourriere H, Gallouzi IE, Chebli K, Capony JP, Mouaikel J, van der Geer P, Tazi J: RasGAP-associated endoribonuclease G3Bp: selective RNA degradation and phosphorylation-dependent localization. Mol Cell Biol. 2001, 21: 7747-7760. 10.1128/MCB.21.22.7747-7760.2001PubMed CentralView ArticlePubMed
- Guitard E, Parker F, Millon R, Abecassis J, Tocque B: G3BP is overexpressed in human tumors and promotes S phase entry. Cancer Lett. 2001, 162: 213-221. 10.1016/S0304-3835(00)00638-8View ArticlePubMed
- Kociok N, Esser P, Unfried K, Parker F, Schraermeyer U, Grisanti S, Toque B, Heimann K: Upregulation of the RAS-GTPase activating protein (GAP)-binding protein (G3BP) in proliferating RPE cells. J Cell Biochem. 1999, 74: 194-201. 10.1002/(SICI)1097-4644(19990801)74:2<194::AID-JCB5>3.0.CO;2-MView ArticlePubMed
- Barnes CJ, Li F, Mandal M, Yang Z, Sahin AA, Kumar R: Heregulin induces expression, ATPase activity, and nuclear localization of G3BP, a Ras signaling component, in human breast tumors. Cancer Res. 2002, 62: 1251-1255.PubMed
- French J, Stirling R, Walsh M, Kennedy HD: The expression of Ras-GTPase activating protein SH3 domain-binding proteins, G3BPs, in human breast cancers. Histochem J. 2002, 34: 223-231. 10.1023/A:1021737413055View ArticlePubMed
- Schneider C, King RM, Philipson L: Genes specifically expressed at growth arrest of mammalian cells. Cell. 1988, 54: 787-793. 10.1016/S0092-8674(88)91065-3View ArticlePubMed
- Manfioletti G, Ruaro ME, Del Sal G, Philipson L, Schneider C: A growth arrest-specific (gas) gene codes for a membrane protein. Mol Cell Biol. 1990, 10: 2924-2930.PubMed CentralView ArticlePubMed
- Zhang H, Zhang S, He H, Zhao W, Chen J, Shao RG: GAP161 targets and downregulates G3BP to suppress cell growth and potentiate cisplaitin-mediated cytotoxicity to colon carcinoma HCT116 cells. Cancer Sci. 2012, 103: 1848-1856. 10.1111/j.1349-7006.2012.02361.xView ArticlePubMed
- Irvine K, Stirling R, Hume D, Kennedy D: Rasputin, more promiscuous than ever: a review of G3BP. Int J Dev Biol. 2004, 48: 1065-1077. 10.1387/ijdb.041893kiView ArticlePubMed
- Tourriere H, Chebli K, Zekri L, Courselaud B, Blanchard JM, Bertrand E, Tazi J: The RasGAP-associated endoribonuclease G3BP assembles stress granules. J Cell Biol. 2003, 160: 823-831. 10.1083/jcb.200212128PubMed CentralView ArticlePubMed
- Matsuki H, Takahashi M, Higuchi M, Makokha GN, Oie M, Fujii M: Both G3BP1 and G3BP2 contribute to stress granule formation. Genes Cells. 2013, 18: 135-146. 10.1111/gtc.12023View ArticlePubMed
- Kobayashi T, Winslow S, Sunesson L, Hellman U, Larsson C: PKCalpha binds G3BP2 and regulates stress granule formation following cellular stress. PLoS One. 2012, 7: e35820- 10.1371/journal.pone.0035820PubMed CentralView ArticlePubMed
- Snipes GJ, Suter U, Welcher AA, Shooter EM: Characterization of a novel peripheral nervous system myelin protein (PMP-22/SR13). J Cell Biol. 1992, 117: 225-238. 10.1083/jcb.117.1.225View ArticlePubMed
- Spreyer P, Kuhn G, Hanemann CO, Gillen C, Schaal H, Kuhn R, Lemke G, Muller HW: Axon-regulated expression of a Schwann cell transcript that is homologous to a ‘growth arrest-specific’ gene. EMBO J. 1991, 10: 3661-3668.PubMed CentralPubMed
- Muller HW: Tetraspan myelin protein PMP22 and demyelinating peripheral neuropathies: new facts and hypotheses. Glia. 2000, 29: 182-185. 10.1002/(SICI)1098-1136(20000115)29:2<182::AID-GLIA12>3.0.CO;2-KView ArticlePubMed
- Naef R, Suter U: Impaired intracellular trafficking is a common disease mechanism of PMP22 point mutations in peripheral neuropathies. Neurobiol Dis. 1999, 6: 1-14. 10.1006/nbdi.1998.0227View ArticlePubMed
- Suter U, Nave KA: Transgenic mouse models of CMT1A and HNPP. Ann N Y Acad Sci. 1999, 883: 247-253. 10.1111/j.1749-6632.1999.tb08586.xView ArticlePubMed
- Huhne K, Park O, Liehr T, Rautenstrauss B: Expression analysis of the PMP22 gene in glioma and osteogenic sarcoma cell lines. J Neurosci Res. 1999, 58: 624-631. 10.1002/(SICI)1097-4547(19991201)58:5<624::AID-JNR3>3.0.CO;2-NView ArticlePubMed
- Li J, Kleeff J, Esposito I, Kayed H, Felix K, Giese T, Buchler MW, Friess H: Expression analysis of PMP22/Gas3 in premalignant and malignant pancreatic lesions. J Histochem Cytochem. 2005, 53: 885-893. 10.1369/jhc.4A6546.2005View ArticlePubMed
- Tong D, Heinze G, Pils D, Wolf A, Singer CF, Concin N, Hofstetter G, Schiebel I, Rudas M, Zeillinger R: Gene expression of PMP22 is an independent prognostic factor for disease-free and overall survival in breast cancer patients. BMC Cancer. 2010, 10: 682- 10.1186/1471-2407-10-682PubMed CentralView ArticlePubMed
- Taylor V, Welcher AA, Program AE, Suter U: Epithelial membrane protein-1, peripheral myelin protein 22, and lens membrane protein 20 define a novel gene family. J Biol Chem. 1995, 270: 28824-28833. 10.1074/jbc.270.48.28824View ArticlePubMed
- van Dartel M, Hulsebos TJ: Characterization of PMP22 expression in osteosarcoma. Cancer Genet Cytogenet. 2004, 152: 113-118. 10.1016/j.cancergencyto.2003.11.003View ArticlePubMed
- Roux KJ, Amici SA, Fletcher BS, Notterpek L: Modulation of epithelial morphology, monolayer permeability, and cell migration by growth arrest specific 3/peripheral myelin protein 22. Mol Biol Cell. 2005, 16: 1142-1151. 10.1091/mbc.E04-07-0551PubMed CentralView ArticlePubMed
- Zoidl G, Blass-Kampmann S, D’Urso D, Schmalenbach C, Muller HW: Retroviral-mediated gene transfer of the peripheral myelin protein PMP22 in Schwann cells: modulation of cell growth. EMBO J. 1995, 14: 1122-1128.PubMed CentralPubMed
- Verrier JD, Lau P, Hudson L, Murashov AK, Renne R, Notterpek L: Peripheral myelin protein 22 is regulated post-transcriptionally by miRNA-29a. Glia. 2009, 57: 1265-1279. 10.1002/glia.20846PubMed CentralView ArticlePubMed
- Du P, Kibbe WA, Lin SM: lumi: a pipeline for processing Illumina microarray. Bioinformatics. 2008, 24: 1547-1548. 10.1093/bioinformatics/btn224View ArticlePubMed
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.