Characterization of antibodies for quantitative determination of spiggin protein levels in male and female three-spined stickleback (Gasterosteus aculeatus)
- Håkan Berg†1,
- Nikolai Scherbak†1,
- Harri Liimatta†1,
- Erik Hoffmann2,
- Johnny Karlsson1 and
- Per-Erik Olsson1Email author
© Berg et al; licensee BioMed Central Ltd. 2009
Received: 02 March 2009
Accepted: 14 May 2009
Published: 14 May 2009
Spiggin is an adhesive glycoprotein produced in the kidney of sticklebacks during the breeding season and is subsequently secreted into the urinary bladder from where it is employed for nest building. Since the production of the protein has been shown to be under androgenic control, spiggin has been suggested to be a useful biomarker for androgenic substances in the environment. In this study, two polyclonal spiggin antibodies based on synthetic peptides and one polyclonal antibody directed against native spiggin have been characterized. The antibodies ability to identify spiggin was investigated by quantitative immunoassay. For both peptide antibodies the quantification range was determined to be between 1 and 80 ng spiggin and determination of renal spiggin levels from immature and mature males displayed a 15-fold increase in total spiggin content of the kidney resulting in a 6-fold increase in male kidney weight due to hypertrophy. The kidney somatic index (KSI) was found to correlate well with the total renal spiggin content and therefore it appears that KSI in sticklebacks could be used as an initial method to identify substances displaying androgenic effects. Furthermore, western blot analysis revealed that the polyclonal antibodies recognize different spiggin isoforms and that spiggin can be detected in the urinary bladder and kidney of both males and female sticklebacks. In order to develop a quantitative detection method for native spiggin it is necessary to produce a standard that can be used in a bioassay. Due to the adhesive and polymerization characteristics of spiggin the protein is difficult to use as a standard in bioassays. So far spiggin has been shown to exist in at least 14 isoforms, all of which contain polymerization domains. To overcome the solubility problem we have produced recombinant spiggin gamma, with only one polymerization domain, that can be expressed in E. coli. Western blot analysis demonstrated that the polyclonal antibodies were able to detect recombinant spiggin gamma protein in bacterial cell lysate, suggesting that it may be developed into a useful source of standard spiggin to be used for quantitative determination of androgen induced spiggin production in sticklebacks.
Three-spined stickleback (Gasterosteus aculeatus) are small fish (~10 cm) with three spikes on their back and two abdominal spikes that are widely distributed throughout the northern hemisphere, and live and reproduce in fresh, brackish and salt water.
During the reproductive season, androgens control the development of male stickleback secondary sexual characteristics such as blue eyes, a red belly and hypertrophied kidneys. Androgens has also been suggested to initiate male reproductive behavior in sticklebacks, such as territorial establishment and nest building using a combination of plant fragments and renal secretions made up of an adhesive glycoprotein called spiggin .
Spiggin is a glycoprotein complex consisting of a multitude of isoforms, which are formed by alternative splicing, and their expression is regulated by 11-ketoandrogens . Structural analysis of the spiggin subunits shows that the protein is highly hydrophobic and thereby insoluble in water. Due to its regulation by androgens, spiggin has been suggested to be a valuable biomarker for environmental androgen and antiandrogen exposure and the three-spined sticklebacks abundance makes it a good candidate as a species for environmental monitoring of androgenic substances. Sticklebacks are easy to keep and breed, which make sticklebacks a good choice as a monitoring species for mechanisms and effects of androgenic exposure. .
Spiggin is synthesized from at least 5 gene-loci and multiple subunits occur as a result of alternative splicing . Initially we identified three spiggin subunits, α, β and γ , but recently, it has been shown that spiggin is encoded by a multi-gene family that give rise to at least 14 protein isoforms. This is suggested to contribute to the effective synthesis of large amount of the glue-like spiggin, during the sticklebacks reproductive season .
The spiggin isoforms all display amino acid sequence similarities to other adhesive proteins such as mucin from Xenopus (28%), rat (27%), human (27%) and to the human von Willebrand factor (vWF) protein D-domain (26%) . The vWF D-domain plays an important role in protein multimerization and for the function of vWF in blood coagulation while mucin plays an important role in the protection of epithelial cells by creating a viscous surface cover . Unlike the vWF and the vWF-related proteins, the D-domain within each subunit of spiggin is organized into non-tandem repeats separated by a number of cystein-rich regions . The shortest of all spiggin subunits identified so far is spiggin γ which is 472 residues long and contain 2 D-like domains coupled by a cystein-rich region. Native spiggin protein is difficult to purify due to multimerization and its hydrophobic nature. Development of a recombinant spiggin protein would facilitate the production of spiggin standard to allow quantitative analysis of spiggin protein. Spiggin γ is a good candidate for the development of a recombinant spiggin protein as it is the smallest isoform and only has one polymerization domain (CGLCG) .
In the present study three polyclonal spiggin antibodies have been characterized and used to develop a quantitative immunoassay. The antibodies have been screened to determine which spiggin isoforms they recognized. In order to achieve knowledge of the basal expression level of spiggin in males and females spiggin levels in immature and mature fish of both sexes were determined. Furthermore, the uses of recombinant spiggin γ protein, as a standard for spiggin bioassays were examined.
Polyclonal antibody production
Synthetic peptides KTK (KTKEIQTYTCRTFGS-C), corresponding to amino acid 24 to 38 in the three-spined stickleback spiggin protein (Jones et al., 2001) and HRD (HRDELIRDSKLHDHR-C) corresponding to amino acid 173 to 188, were produced by Innovagen AB (Lund, Sweden). The polyclonal antisera against subunits KTK and HRD of spiggin were produced by Agrisera AB (Vännäs, Sweden).
Experimental animals and samplings
Adult three-spine sticklebacks were caught in Oresund and were kept in a 700 L tank with brackish water (0.5% salinity) at low temperature (4–7°C), short photoperiod of 8 hrs light and 16 hrs darkness (L:D 8:16) to maintain them in non breeding condition. In order to induce maturation, the sticklebacks were transferred to 200 L aquaria and held under long photoperiod L:D 16:8 which resulted in visible secondary sexual characteristics in males. Prior to sampling the fish were euthanized by a swift strike to the head, weighed and bladders and kidneys were excised and weighed. The excised tissues were immediately transferred to 2.0 ml Eppendorf tubes (Eppendorf, Germany), containing 400 μl of an ice cold 10% saline solution, homogenized, and kept on ice or at 4°C until analyzed.
For the spiggin induction studies, non breeding stickleback (1.39 ± 0.3 g) were injected intraperitoneally with 5 μg 11-ketotestosterone (11-KT) per kg bodyweight. For the negative control, carrier alone (peanut oil) was injected. Following the injection the fish were kept under normal conditions for 72 hours before sampling. At sampling the stickleback were sacrificed, weighed and the kidneys were excised and immediately homogenized in an ice cold 10% saline solution prior to performing an ELISA.
ELISA was performed using NUNC MaxiSorp 96-well plates, (Nunc A/S, Denmark), synthetic peptides were diluted to a final concentration of 20 ng in 100 μl coating buffer (0.1 M Na2CO3, pH 9. 6). The plates were incubated at room temperature (RT, 22°C) for 1 hr followed by a 3 × 3 minutes wash with 200 μl PBST (0.05 M phosfate buffered saline pH 7.4 with 0.5% Tween 20) and subsequent addition of 200 μl 5% dry milk dissolved in PBS, for 1 hour at RT. The plate was subsequently washed with PBST, 3 × 3 minutes before addition of 100 μl primary antibody to each well and incubated at RT for 1 h. Following the incubation the plate was washed with PBST, 3 × 3 minutes prior to addition of a HRP conjugated secondary antibody, anti-rabbit Ig (Amerersham, England, 1:2000). The plate was incubated at RT for 30 minutes and followed by a 3 × 3 minutes wash in TBST. Detection was enabled by addition of 100 μl 2,2'-azino-bis (3-ethylbenzthiazoline-6-sulfonic acid (ABTS, Pierce, Rockford, IL, USA), absorbance was measured with a plate-reader, (TD-20/20 luminometer, Turner designs), at 405 nm after 60 minutes. The linear range of detection was calculated and used for measurements on serial dilutions of the peptides.
Protein concentration determination
Protein concentration were determined by using the Bio-Rad protein assay, (Bio-Rad, USA), according to the manufacturers instructions. Briefly, a standard dilution series of bovine serum albumin (BSA) was made in PBS, and Bio-Rad protein assay reagent (Bio-Rad, USA), diluted 1:3 in PBS, was used as detection reagent. Absorbance was measured in a plate-reader at 595 nm.
Western blot procedures
SDS-PAGE was performed using a 5% (w/v) stacking gel and a 7.5% (w/v) separating gel. Prior to electrophoresis the samples were mixed with sample buffer and heated at 65°C, 10 minutes and immediatley put on ice. 6 μg of total protein was added to each well. Electrophoresis was carried out at 145 V, 45 minutes at RT. Following SDS-PAGE, the separated proteins were electroblotted onto PVDF membranes (Hybond-P, Amersham, England). Electro-transfer of proteins was performed overnight at 5°C, 30 mV in a regular transfer buffer containing 20% methanol. Upon completion of transfer, the membranes were incubated in tris buffered saline (TBS) containing 5% dry milk for 2 hours. The membrane was rinsed in TBST (TBS with 0.1% (v/v) Tween 20) for 3 × 3 minutes prior to a 1 hour incubation in primary antibody. Subsequently the membranes were washed 3 × 15 minutes in TBST and incubated with the secondary antibody, diluted 1:3000 in TBST, for 30 minutes. Prior to detection the membrane was washed, 3 × 15 minutes in TBST. The detection was made by using ECL (Amersham, England), according to the manufacturers instructions.
Cloning and transformation
The open reading frame (ORF) of spiggin γ was amplified from a pBK-CMV-spiggin γ vector construct using the following PCR (primers: forward – 5'-CAC CAT GAC AAC CCA GCG ATG GAT TC-3'; reverse-5'-CGG TGT TTC TGG GTT ATC AC-3') using proof-reading VENT DNA-polymerase (New England Biolabs, USA). Optimized PCR conditions were as follows: 97°C for 3 minutes → [95°C for 30 seconds, 55°C for 30 seconds, 72°C for 3 minutes] × 35 cycles → 72°C for 10 minutes. Following DNA electrophoresis the PCR product was isolated from the gel using QIAquick Gel Extraction Kit (Qiagen, Germany). TOPO-TA directional cloning reaction into pET101 vector, (Invitrogen, USA), and further transformation into Top10 E. coli strain, (Invitrogen, Carlsbad, CA, USA) were performed according to the manufacturer's manual. For the over-expression purpose the purified pET101/spiggin γ vector was transformed into BL21o (DE3) pLysS E. coli (Invitrogen, USA). Transformed bacteria were grown on LB plates containing 50 mg/L carbenicillin. Screening of the colonies was performed using PCR under the previously mentioned conditions using forward T7 and spiggin reverse primers pair.
Protein over-expression and bacterial cell lysis
20 μL of bacterial glycerol stock was added to 20 ml LB supplied with 50 mg/L carbenicillin and cultivated on shaker at 37°C. Protein expression was induced by addition of 1 mM IPTG at OD600 = 0.8. Following overnight incubation on shaker at 30°C, the bacteria were pelleted by centrifugation for 20 min at 4000 g. Pelleted bacterial cells were lysed by the addition of 600 μL of CelLytic B 2× buffer, (Sigma, USA), supplied with 20 mM 2-mercaptoethanol, DNase and Complete™ protease inhibitor cocktail (Roche, Basel). In order to detect the over expressed protein the supernatant was mixed with Laemli loading buffer , boiled for 5 min and was resolved on a 12% SDS-PAGE prior to western blot analysis as described above. The immunodetection was performed using a polyclonal antiserum raised against N-terminal peptide of spiggin, or the polyclonal antiserum against native spiggin protein (1:5000 working dilution), alternatively – the mouse monoclonal anti-polyhistidine antibody (1:2500 dilution; Sigma-Aldrich, St. Louis, MO). Proteins were visualized using Western Blue® Stabilized Substrate for Alkaline Phosphatase (Promega, USA).
Data are presented as mean ± standard deviation and all statistical analysis were performed with Graphpad Prism V 3.03 (GraphPad Software, USA). Statistical analysis of differences in spiggin levels was performed by one-way Anova, followed by Tukey's post hoc test.
Spiggin protein isoforms. The calculated primary molecular mass and sequence similarity (%) to spiggin γ based on protein sequences obtained from the indicated GeneBank accession numbers are shown.
% similarity to spiggin γ
Body weight, kidney weight and KSI of male and female stickleback.
Non-spawning females (n = 5)
1.90 ± 0.32
14.1 ± 3.5
0.76 ± 0.24
Non-spawning males (n = 6)
1.23 ± 0.11
5.8 ± 1.5
0.48 ± 0.13
Pre-spawning males (n = 5)
1.69 ± 0.12
36.5 ± 7.9
2.17 ± 0.52
During breeding male stickleback use the spiggin protein as an underwater adhesive to build their nest. The physical properties of spiggin protein render it extremely difficult to purify as multimerization of the protein results in aggregate formation and precipitation. On the other hand, spiggin would be a valuable tool to be able to measure, since it is one of the few proteins that so far are known to be under androgenic control and therefore could be used as a measurable parameter for androgen exposure.
This paper presents the development of sensitive quantitative antibody based spiggin assays useful for the detection of androgenic and anti-androgenic substances in the environment using three-spined stickleback as a model. Furthermore, it describes the use of recombinant spiggin γ as a standard for spiggin immunoblotting. The isoform composition of expressed spiggin protein complexes may vary in kidney and in urinary bladder . Since the spiggin isoforms share the amino acids corresponding to the spiggin γ protein (Fig. 1), this isoform was chosen as a good candidate to use as a standard. The differences in the region for which the synthetic peptide KTK was raised is shown, with spiggin 2 and 3 lacking part of the sequence, whereas the sequence for the synthetic peptide HRD is present in all the sequences. The polyclonal native spiggin antibody used in the present study exhibits a high specificity as observed by ELISA and Western blot analysis. The ELISA showed that the spiggin antibody cross-reacts with the native adhesive protein in a linear range and in a concentration-dependent manner (Fig. 2). Spiggin polyclonal antibody against the native protein cross-reacted with native spiggin and produced several bands of different molecular weigh on the Western blot (Fig 4b). These bands correspond well with the expected sizes of the individual subunits and multimerized spiggin. Thus, the polyclonal antibody directed against native spiggin is able to cross-react with all spiggin subunits on Western blot analysis and also detects spiggin in ELISA analysis. Detection of native spiggin from the urinary bladder indicated a high level of multimerization with a strong band around 122 kDa and additional bands around 70 kDa and 52 kDa. In the kidney from female fish the protein is less abundant than in the males, including the multimerized isoforms. As in the males, bands corresponding to the sizes of all spiggin isoforms were detected in the kidney and the urinary bladder. (Fig. 4b). In contrast to males no multimerized spiggin could be identified in the urinary bladder of female stickleback. In male kidney there are bands corresponding to 230 kDa, 120 kDa, 100 kDa and 70 kDa. While the 230 kDa band corresponds to multimerized spiggin isoforms, the other bands correspond well with the calculated molecular weights of individual spiggin isoforms (Table 1).
The use of a synthetic peptide standard allows for quantification of spiggin and we have shown that the HRD peptide can be used in a competitive ELISA for detection of spiggin in methyltestosterone induced male and female three spined sticklebacks . In this study both peptide antibodies were shown to be able to identify and quantify spiggin in biological samples. By using KTK and HRD in an ELISA it was found that a three day 11-KT exposure gave rise to a 1.73 and 1.79 spiggin increase respectively, indicating that the two antibodies can be used individually to measure androgen induced spiggin production.
Due to the native spiggin proteins highly hydrophobic properties, which are instrumental for its use by stickleback males as underwater glue, it poses challenges in handling and detection. Therefore, the use of recombinant V5/His6-tagged spiggin γ as a standard for spiggin measurements offers an alternative in order to obtain quantitative measurements of spiggin in biological samples. Addition of V5/His6 epitopes to recombinant proteins has shown to increase their solubility  and recombinant spiggin can be produced in bacteria in necessary amounts and in a short period of time which makes it highly available.
On the western blot, the spiggin antibodies against the native protein and antibodies against the synthetic peptides cross-reacted with the recombinant protein in bacterial cell lysate enabling detection of a recombinant protein of approximately 50 kDa. The molecular weight of the recombinant protein on Western blot was less than the theoretical protein weight which is in agreement with previously reported results . Since E coli lack glycosylation properties, discrepancies in theoretical and experimental molecular weight for the recombinant spiggin could be explained by the physical properties of the spiggin. The high hydrophobicity of spiggin might also contribute to this phenomenon due to higher affinity to SDS .
Spiggin shows great potential as a screening tool for endocrine disruptors, and has been used for this purpose in kidney cell cultures , tissue slice cultures  and in whole animal exposures . However, there are many factors in the environment that might affect the production of the adhesive protein. For instance, parasite harbouring fish should be excluded to avoid errors, as infection of Schistocephalus solidus in male three-spined stickleback result in reduced levels of spiggin in the kidney . The correlation between total spiggin content in all three groups with KSI suggests that KSI measurements could be used as an initial rapid screening method to detect possible androgenic disturbances.
Due to the adhesive and multimerization properties of spiggin it is of interest to develop methods to prevent polymerisation of both native and recombinant protein upon purification. We have tested the effets of saline solution as solvent to simplify handling of samples. Adjustments of pH was not found to improve the solubility of spiggin whereas a saline solution kept spiggin soluble for several days in +4°C (data not shown). Even so, the positive effect of salinity on spiggin soulubility could not be used in the ELISA procedures since the small volumes of the samples added to the buffer used in ELISA results in low final concentrations of salt and a rapid mutlimerization of the protein in the wells.
The seasonal cycling in plasma levels of 11-ketotestosterone in the male three-spined stickleback show an marked increase in May with a peak from June to July, this peak is also mirrored in the kidney epithelial height, (KEH), which increases from approximately 15 μm to 40 μm in the spawning season [15, 16]. Seasonal variations in KSI showed an increase from March to April, peaking in May, followed by a decrease in June to July . Measurements of spiggin in conjunction with measurements of KSI would enable trend monitoring of local populations over time to obtain data on possible increases in accumulation of endocrine disrupting substances, and the appearance of previously undetected environmental androgens. The cloning and characterization of the stickleback nuclear androgen receptor  will be a useful tool for future work on identification of androgenic substances in the aquatic environment.
In the present study we have characterized three spiggin antibodies and demonstrated that they are highly specific and able to detect basal levels of spiggin in both male and female fish. All three antibodies were able to detect protein corresponding to all the known sizes of spiggin, suggesting that they will give a total spiggin value when used in ELISA assays. We have also developed two quantitative spiggin ELISA using peptide standards and have shown that posibility of producing recombinant spiggin protein for use as a standard with native spiggin antibodies.
The study was funded by the Knowledge Foundation, Sweden.
- Pall MK, Mayer I, Borg B: Androgen and behavior in the male three-spined stickleback, Gasterosteus aculeatus. II. Castration and 11-ketoandrostenedione effects on courtship and parental care during the nesting cycle. Horm Behav. 2002, 42: 337-344. 10.1006/hbeh.2002.1820.View ArticlePubMedGoogle Scholar
- Jones I, Lindberg C, Jakobsson S, Hellqvist A, Hellman U, Borg B, Olsson PE: Molecular cloning and characterization of spiggin. An androgen-regulated extraorganismal adhesive with structural similarities to von Willebrand Factor-related proteins. J Biol Chem. 2001, 276: 17857-17863. 10.1074/jbc.M101142200.View ArticlePubMedGoogle Scholar
- Jakobsson S, Borg B, Haux C, Hyllner SJ: An 11-ketotestosterone induced kidney-secreted protein: the nest building glue from male three-spined stickleback, Gasterosteus aculeatus. Fish Physiology and Biochemistry. 1999, 20: 79-85. 10.1023/A:1007776016610.View ArticleGoogle Scholar
- Kawahara R, Nishida M: Extensive lineage-specific gene duplication and evolution of the spiggin multi-gene family in stickleback. BMC Evol Biol. 2007, 7: 209-10.1186/1471-2148-7-209.PubMed CentralView ArticlePubMedGoogle Scholar
- Kawahara R, Nishida M: Multiple occurrences of spiggin genes in sticklebacks. Gene. 2006, 373: 58-66. 10.1016/j.gene.2006.01.008.View ArticlePubMedGoogle Scholar
- Moniaux N, Nollet S, Porchet N, Degand P, Laine A, Aubert JP: Complete sequence of the human mucin MUC4: a putative cell membrane-associated mucin. Biochemical Journal. 1999, 338: 325-333. 10.1042/0264-6021:3380325.PubMed CentralView ArticlePubMedGoogle Scholar
- Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970, 227: 680-685. 10.1038/227680a0.View ArticlePubMedGoogle Scholar
- Sanchez W, Goin C, Brion F, Olsson PE, Goksoyr A, Porcher JM: A new ELISA for the three-spined stickleback (Gasterosteus aculeatus L.) spiggin, using antibodies against synthetic peptide. Comp Biochem Physiol C Toxicol Pharmacol. 2008, 147: 129-137. 10.1016/j.cbpc.2007.08.007.View ArticlePubMedGoogle Scholar
- Dyson MR, Shadbolt SP, Vincent KJ, Perera RL, McCafferty J: Production of soluble mammalian proteins in Escherichia coli: identification of protein features that correlate with successful expression. Bmc Biotechnology. 2004, 4: 32-10.1186/1472-6750-4-32.PubMed CentralView ArticlePubMedGoogle Scholar
- Miyake J, Ochiaiyanagi S, Kasumi T, Takagi T: Isolation of a Membrane-Protein from R-Rubrum Chromatophores and Its Abnormal-Behavior in Sds-Polyacrylamide Gel-Electrophoresis Due to a High Binding-Capacity for Sds. Journal of Biochemistry. 1978, 83: 1679-1686.PubMedGoogle Scholar
- Jolly C, Katsiadaki I, Le Belle N, Mayer I, Dufour S: Development of a stickleback kidney cell culture assay for the screening of androgenic and anti-androgenic endocrine disrupters. Aquat Toxicol. 2006, 79: 158-166. 10.1016/j.aquatox.2006.06.005.View ArticlePubMedGoogle Scholar
- Bjorkblom C, Olsson PE, Katsiadaki I, Wiklund T: Estrogen- and androgen-sensitive bioassays based on primary cell and tissue slice cultures from three-spined stickleback (Gasterosteus aculeatus). Comp Biochem Physiol C Toxicol Pharmacol. 2007, 146: 431-442. 10.1016/j.cbpc.2007.05.004.View ArticlePubMedGoogle Scholar
- Andersson C, Katsiadaki I, Lundstedt-Enkel K, Orberg J: Effects of 17alpha-ethynylestradiol on EROD activity, spiggin and vitellogenin in three-spined stickleback (Gasterosteus aculeatus). Aquat Toxicol. 2007, 83: 33-42. 10.1016/j.aquatox.2007.03.008.View ArticlePubMedGoogle Scholar
- Rushbrook BJ, Katsiadaki I, Barber I: Spiggin levels are reduced in male sticklebacks infected with Schistocephalus solidus. Journal of Fish Biology. 2007, 71: 298-303. 10.1111/j.1095-8649.2007.01495.x.View ArticleGoogle Scholar
- Borg B: Seasonal Effects of Photoperiod and Temperature on Spermatogenesis and Male Secondary Sexual Characters in the 3-Spined Stickleback, Gasterosteus-Aculeatus L. Canadian Journal of Zoology-Revue Canadienne De Zoologie. 1982, 60: 3377-3386. 10.1139/z82-427.View ArticleGoogle Scholar
- Mayer I, Borg B, Schulz R: Seasonal-Changes in and Effect of Castration Androgen Replacement on the Plasma-Levels of 5 Androgens in the Male 3-Spined Stickleback, Gasterosteus-Aculeatus L. General and Comparative Endocrinology. 1990, 79: 23-30. 10.1016/0016-6480(90)90084-Y.View ArticlePubMedGoogle Scholar
- Hellqvist A, Schmitz M, Mayer I, Borg B: Seasonal changes in expression of LH-beta and FSH-beta in male and female three-spined stickleback, Gasterosteus aculeatus. Gen Comp Endocrinol. 2006, 145: 263-269. 10.1016/j.ygcen.2005.09.012.View ArticlePubMedGoogle Scholar
- Olsson PE, Berg AH, von Hofsten J, Grahn B, Hellqvist A, Larsson A, Karlsson J, Modig C, Borg B, Thomas P: Molecular cloning and characterization of a nuclear androgen receptor activated by 11-ketotestosterone. Reprod Biol Endocrinol. 2005, 3: 37-10.1186/1477-7827-3-37.PubMed CentralView 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.