Complete genome of a European hepatitis C virus subtype 1g isolate: phylogenetic and genetic analyses
© Bracho et al; licensee BioMed Central Ltd. 2008
Received: 30 January 2008
Accepted: 05 June 2008
Published: 05 June 2008
Hepatitis C virus isolates have been classified into six main genotypes and a variable number of subtypes within each genotype, mainly based on phylogenetic analysis. Analyses of the genetic relationship among genotypes and subtypes are more reliable when complete genome sequences (or at least the full coding region) are used; however, so far 31 of 80 confirmed or proposed subtypes have at least one complete genome available. Of these, 20 correspond to confirmed subtypes of epidemic interest.
We present and analyse the first complete genome sequence of a HCV subtype 1g isolate. Phylogenetic and genetic distance analyses reveal that HCV-1g is the most divergent subtype among the HCV-1 confirmed subtypes. Potential genomic recombination events between genotypes or subtype 1 genomes were ruled out. We demonstrate phylogenetic congruence of previously deposited partial sequences of HCV-1g with respect to our sequence.
In light of this, we propose changing the current status of its subtype-specific designation from provisional to confirmed.
Hepatitis C virus (HCV), a single-stranded positive-sense RNA virus belonging to the Flaviviridae family, is the leading etiologic agent of chronic liver disease. According to WHO, about 180 million people, an estimated 3% of the world population, are infected with HCV . Its genome, which is approximately 9600 nucleotide (nt) long, contains two short untranslated regions at each end (5'UTR and 3'UTR) and a single ORF of about 9000 nt, known as polyprotein, encoding three structural (core, E1 and E2) and seven non-structural proteins (P7, NS2, NS3, NS4A, NS4B, NS5A and NS5B). Based mainly on phylogenetic analyses, all HCV isolates are currently grouped into six genotypes (from 1 to 6) , and within each genotype, closely related isolates cluster in a varying number of subtypes (designated with letters a, b, c and so on) . Provisional designation of subtypes requires rigorous phylogenetic analysis of sequences from both the core/E1 region and the NS5B region obtained from three or more different infections. Confirmed designation status is acquired after intensive phylogenetic analysis including, at least, one complete genome sequence of the candidate subtype. Before a new subtype is confirmed, rigorous recombination and phylogenetic analyses should preclude both recombination events between subtypes and significant grouping within any of the confirmed subtypes .
Thirteen subtypes of HCV genotype 1 have been described so far (from 1a to 1m). However, only three subtypes (1a, 1b and 1c), for which the complete genome sequence has been obtained, have the status of confirmed subtype. The remaining subtypes (from 1d to 1m), from which only partial sequences are known, have been denoted as provisional. In addition, a complete genotype 1 sequence from an Equatorial Guinea isolate with unassigned subtype is also available .
The HCV genotype infecting a patient is important as it influences dose and duration of current antiviral therapy (pegylated alpha interferon plus ribavirin); patients infected with genotype 2 or 3 respond better than those infected with genotype 1 or 4 [5, 6]. Apart from being an excellent method for reliable genotyping, phylogenetic and genetic analysis of appropriate sequence data, is an important tool for epidemiological surveys, including deep outbreak studies , novel transmission risks , viral evolution [9, 10] and origin and spread of HCV epidemics [11–14].
In the present study, viral RNA was isolated from a specimen (serum) obtained from a 56-year-old Spanish female patient, who seroconverted to HCV after undergoing surgery and receiving a blood transfusion in 1996. No other recognizable risk factor could be identified for acquiring HCV infection. Serum was obtained before pegylated alpha interferon plus ribavirin treatment, to which the patient did not respond. Initially, HCV genotype was determined by means of two genotyping assays. First, an assay using the Trugene® 5'NC genotyping kit (TRUGENE 5'NC; Bayer HealthCare) based on the sequencing of a fragment of the 5'UTR, led to an ambiguous subtype 1a/1c. Secondly, use of the Abbott Real Time HCVTM kit (Abbott Diagnostics), which targets the NS5B region for genotype 1 but only distinguishes subtypes a and b, led to an unambiguous subtype 1a. Accurate identification as subtype 1g could only be determined after partial sequencing of the NS5B gene followed by both sequence comparison against sequence databases and phylogenetic analysis. Many authors have pointed out some discordant subtyping results on comparing the results obtained using different genotyping assays based on the 5'UTR [15–17] or on comparing results from these assays with results from genotyping in-house methods, based on NS5B sequences [18, 19]. Furthermore, a more relevant point has definitively been demonstrated concerning the intrinsic limitations of the 5'UTR. Due to this region's high level of conservation, its power to reproduce phylogenetic trees obtained using complete genome is limited, and consequently, it fails to discriminate subtypes or even genotypes . As a result of inefficient genotyping and subtyping in most commercial assays, the presence of some subtypes could have been underestimated, or some of them even ignored, in epidemiological investigations of circulating HCV variants. An important consequence of accurate assignation of HCV subtypes based on appropriate sequence data is that it turns routine genotyping into a reliable tool for molecular epidemiology studies in which, apart from a clear description of circulating subtypes, putative new subtypes and/or genotypes can be detected .
Results and Discussion
Here we report the first complete genome of a hepatitis C virus subtype 1g isolate. To demonstrate this we have both performed phylogenetic analysis with representative complete genomes of all genotypes, including the confirmed subtypes 1a, 1b and 1c, and also with all of the partial subtype 1g sequences deposited in sequence databanks.
Potential recombination events between genotypes or subtype 1 genomes were investigated following two approaches and, finally, ruled out. Firstly, phylogenetic reconstructions using the same representative sequences as in Fig. 1 were performed separately for the 10 protein-coding genes (from core to NS5B). With respect to the grouping of HCV-1g within HCV-1 subtypes, all the phylogenetic trees congruently reproduced the same topology obtained from the complete genome analysis. Moreover, subtype 1g still retained its basal position with respect to the other HCV-1 sequences in topologies based on E1, E2, NS2, NS4A, NS4B and NS5A genes (data not shown). Secondly, potential recombination events using the complete sequence alignment were investigated using the RDP 3.0b03 software [ and references therein]. This program implements several methods to identify of recombinant sequences and recombination breakpoints. All the recombination analyses based on the complete genome alignment showed no evidence that our subtype 1g sequence had participated in recombination events (data not shown).
Mean genetic distances among HCV subtype 1 representative sequences
Unassigned subtype 1
subtype 1a (n = 10)
subtype 1b (n = 12)
subtype 1c (n = 2)
Unassigned subtype 1 (n = 1)
subtype 1g (n = 1)
In the analyses of the NS5B region, three short deposited sequences were not included, because after nucleotide alignment the overlapping region was too short to be analysed. These three early deposited sequences, then considered subtype 1c and later assigned as subtype 1g, (Z70375, Z70392 and X88710) were obtained from sera dated between 1994 and 1995 in Germany  and would represent the first subtype 1g isolates detected in Europe. Although birthplace of these three patients could not be checked, the authors mentioned that some patients participating in the study had recently emigrated from Egypt and Sudan. The phylogenetic tree obtained using the NS5B region also includes 2 sequences from Lebanon , deposited in 1993 as subtype 1c and later assigned to subtype 1g (in fact, the two first subtype 1g sequences detected worldwide). The tree also includes one sequence from a Sudanese individual, detected in a study of unpaid blood donors in the Netherlands . Interestingly, the patient in our study was born and resided in Spain, which is evidence of local transmission of subtype 1g.
In summary, we have determined the complete genome sequence of an HCV-1g isolate, we have verified its grouping within HCV-1 and differentiation from other subtypes of this group by rigorous phylogenetic analyses, we have verified that this genome does not result from recombination events and that it is the most basal subtype among those belonging to HCV-1 for which a complete genome sequence is currently available. Taking this into account, we propose changing the status of subtype 1g from proposed to confirmed subtype.
Viral purification, RT-PCR and sequencing
Viral RNA was extracted from 200 μl of serum using High Pure Viral RNA Kit (Roche). Retrotranscription of viral RNA was performed in a final volume of 20 μl containing 10 μl of eluted RNA, 4 μl retrotranscription buffer, 500 μM of each dNTP, either 0.5 μg of random hexadeoxynucleotides (Promega) or 1 μM antigenomic sense primer, 100 U of M-MLV reverse transcriptase (Promega) and 20 U of rRNasin® Ribonuclease Inhibitor (Promega). The mixture was incubated at 42°C for 60 min, followed by 3 min at 95°C. Table showed in additional file 1, lists the oligonucleotide primers used to obtain and/or sequence the overlapping RT-PCR products, which covered almost the whole genome. Primers denoted with "g" or "a" indicate genomic or antigenomic sense, respectively. Primers named with "h" refer to primers used in first round PCR followed by a hemi-nested PCR. Sequence primers named with "R" were directly designed from our sequence of subtype 1g. The genome was covered by 10 overlapping fragments using primer pairs: H28g-COA1a, COS2g-E1E2a, E1E2A2g-NS1a, NS1g2R-NS3a3, NS3g2R-305a2R, 1503g-577a, 5600gR-NS5a2R, KUg2-NS5B1a, NS5B1g-1279a, 1327gR-3utra2. When necessary, additional PCR fragments were also obtained by using combinations of the primers listed in additional file 1.
First round and hemi-nested amplifications were performed in a 50 μl volume containing either 5 μl of the RT product (in the case of first round PCR) or 1 μl of the first round PCR product (in the case of hemi-nested PCR), 5 μl of 10× PCR buffer, 100 μM of each dNTP, 200 nM of the genomic sense primer, 200 nM of the antigenomic sense primer and 5 U of Taq DNA Polymerase (Amersham). All PCRs were performed in a GeneAmp® PCR system 2700 (Applied Biosystems) thermal cycler with the following profile: 95°C for 2 min, then 35 cycles at 95°C for 30 sec, 50–65°C (depending on the primers used) for 30 sec and 72°C for 3 min, and a final extension at 72°C for 10 min.
Amplified products were purified with High Pure PCR Products Purification Kit (Roche). These purified DNAs were sequenced using the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kit v 3.1 in a 3700 automated sequencer (Applied Biosystems). Sequencing primers are also listed in additional file 1. Chromatogram files were assembled, verified and edited using the Staden Package . The newly characterised sequence has been deposited in EMBL with accession number AM910652.
Phylogenetic reconstructions and genetic distances
Two sets of nucleotide sequences were analysed: one corresponding to the complete polyprotein (Fig. 1) and the other corresponding to all sequences provisionally designed as subtype 1g and deposited at the HCV sequence database in Los Alamos  (Fig. 2). In the first set, the nucleotide sequence coding for the polyprotein (Fig. 1) was included in phylogenetic reconstructions along with 29 homologous complete genome sequences representative of the main HCV genotypes and subtypes (see accession numbers, genotypes and subtypes in Fig. 1). Selected sequences fulfil the condition of containing less than 15 ambiguities. In the second set, partial sequences belonging to four regions of HCV genome (5'UTR, core, core/E1 and NS5B) were analysed separately along with the corresponding homologous fragment of our subtype 1g complete genome. Alignments of partial sequences of HCV-1g used in phylogenetic reconstructions included (number of nucleotides in parenthesis): nine 5'UTR sequences (186 nt), four core sequences (217 nt), eighteen core/E1 sequences (220 nt) and thirteen NS5B sequences (222 nt). (see accession numbers, specimen name, and subtypes in Fig. 2). In addition, representative sequences used in the complete genome analysis for subtypes 2a, 3a, 4a, 5a and 6a (Fig. 1) and referred as outgroup were also included in the phylogenetic analyses.
ClustalW  implemented in MEGA version 4  was used to obtain a multiple alignment of the corresponding amino acid sequences from which a codon-based nucleotide alignment was derived, except for the 5'UTR alignment. All phylogenetic trees were constructed by maximum likelihood in PHYML with the nucleotide substitution model that best fit the data according to Akaike Information Criterion (AIC)  for which we used the procedure implemented in Modeltest 3.8 . The robustness of the tree topology was assessed by bootstrap analysis with 1000 replicates implemented in PHYML .
Estimates of mean distances between subtypes of HCV genotype 1 and between these and the new subtype 1g sequence were obtained with the maximum likelihood distance (see above) with PAUP*4.0b10 . For this, we used 26 complete genomes from EMBL: our sequence for subtype 1g [EMBL: AM910652], ten sequences representing subtype 1a [EMBL: D10749, EMBL: M62321, EMBL: M67463, EMBL: AF009606, EMBL: AF011751, EMBL: AF011752, EMBL: AF290978, EMBL: AF271632, EMBL: AJ278830, EMBL: EU155214], twelve sequences for subtype 1b [EMBL: D11168, EMBL: D14484, EMBL: D45172, EMBL: L02836, EMBL: AB080299, EMBL: AB016785, EMBL: AB049095, EMBL: AF139594, EMBL: AF165048, EMBL: AF333324, EMBL: AJ000009, EMBL: AY045702), two sequences for subtype 1c [EMBL: D14853, EMBL: AY051292] and one sequence that corresponds to an unassigned subtype 1 [EMBL: AJ851228].
This work is supported by Conselleria de Sanitat i Consum, Generalitat Valenciana (Spain) and project BFU2005-00503 from Ministerio de Educación y Ciencia (Spain).
This work is also partially supported by grant PI051131 from Instituto de Salud Carlos III-Fondo de Investigaciones Sanitarias, grant CD05/00258 (EM) (contratos postdoctorales de perfeccionamiento) from the Ministerio de Sanidad y Consumo, within the Plan Nacional de Investigación científica, Desarrollo e Innovación Tecnológica (I+D+I); and by grant 2007FIC00550 (VS) from Comissionat per a Universitats i Recerca del Departament d'Innovació, Universitats i Empresa de la Generalitat de Catalunya i del Fons Social Europeu (Spain).
- WHO: Hepatitis C – global prevalence (update). Weekly Epidemiological Record 1999, 49: 425-427.Google Scholar
- Robertson B, Myers G, Howard C, Brettin T, Bukh J, Gaschen B, Gojobori T, Maertens G, Mizokami M, Nainan O, Netesov S, Nishioka K, Shin-I T, Simmonds P, Smith D, Stuyver L, Weiner A: Classification, nomenclature, and database development for hepatitis C virus (HCV) and related viruses: proposals for standardization. Arch Virol 1998, 143: 2493-2503. 10.1007/s007050050479View ArticlePubMedGoogle Scholar
- Simmonds P, Bukh J, Combet C, Deleage G, Enomoto N, Feinstone S, Halfon P, Inchauspe G, Kuiken C, Maertens G, Mizokami M, Murphy DG, Okamoto H, Pawlotsky JM, Penin F, Sablon E, Shin-I T, Stuyver LJ, Thiel HJ, Viazov S, Weiner AJ, Widell A: Consensus proposals for a unified system of nomenclature of hepatitis C virus genotypes. Hepatology 2005, 42: 962-973. 10.1002/hep.20819View ArticlePubMedGoogle Scholar
- Bracho MA, Carrillo-Cruz FY, Ortega E, Moya A, González-Candelas F: A new subtype of hepatitis C virus genotype 1: complete genome and phylogenetic relationships of an Equatorial Guinea isolate. J Gen Virol 2006, 87: 1697-1702. 10.1099/vir.0.81666-0View ArticlePubMedGoogle Scholar
- Carrat F, Bani-Sadr F, Pol S, Rosenthal E, Lunel-Fabiani F, Benzekri A, Morand P, Goujard C, Pialoux G, Piroth L, Salmon-Céron D, Degott C, Cacoub P, Perronne C, ANRS HCO2 RIBAVIC Study Team: Pegylated interferon alfa-2b vs standard interferon alfa-2b, plus ribavirin, for chronic hepatitis C in HIV-infected patients: a randomized controlled trial. JAMA 2004, 292: 2839-2848. 10.1001/jama.292.23.2839View ArticlePubMedGoogle Scholar
- Heathcote EJ: Antiviral therapy: chronic hepatitis C. J Viral Hepat 2007, (Suppl 1):82-88. 10.1111/j.1365-2893.2007.00921.xGoogle Scholar
- Wróbel B, Torres-Puente M, Jiménez N, Bracho MA, García-Robles I, Moya A, González-Candelas F: Analysis of the overdispersed clock in the short-term evolution of hepatitis C virus: Using the E1/E2 gene sequences to infer infection dates in a single source outbreak. Mol Biol Evol 2006, 23: 1242-1253. 10.1093/molbev/msk012View ArticlePubMedGoogle Scholar
- Bronowicki JP, Venard V, Botté C, Monhoven N, Gastin I, Choné L, Hudziak H, Rihn B, Delanoë C, LeFaou A, Bigard MA, Gaucher P: Patient-to-patient transmission of hepatitis C virus during colonoscopy. N Engl J Med 1997, 337: 237-240. 10.1056/NEJM199707243370404View ArticlePubMedGoogle Scholar
- Jiménez-Hernández N, Torres-Puente M, Bracho MA, García-Robles I, Ortega E, del Olmo J, Carnicer F, González-Candelas F, Moya A: Epidemic dynamics of two coexisting hepatitis C virus subtypes. J Gen Virol 2007, 88: 123-133. 10.1099/vir.0.82277-0View ArticlePubMedGoogle Scholar
- Salemi M, Vandamme AM: Hepatitis C virus evolutionary patterns studied through analysis of full-genome sequences. J Mol Evol 2002, 54: 62-70. 10.1007/s00239-001-0018-9View ArticlePubMedGoogle Scholar
- Njouom R, Nerrienet E, Dubois M, Lachenal G, Rousset D, Vessière A, Ayouba A, Pasquier C, Pouillot R: The hepatitis C virus epidemic in Cameroon: genetic evidence for rapid transmission between 1920 and 1960. Infect Genet Evol 2007, 7: 361-367. 10.1016/j.meegid.2006.10.003View ArticlePubMedGoogle Scholar
- Pybus OG, Charleston MA, Gupta S, Rambaut A, Holmes EC, Harvey PH: The epidemic behavior of the hepatitis C virus. Science 2001, 292: 2323-2325. 10.1126/science.1058321View ArticlePubMedGoogle Scholar
- Pybus OG, Cochrane A, Holmes EC, Simmonds P: The hepatitis C virus epidemic among injecting drug users. Infect Genet Evol 2005, 5: 131-139. 10.1016/j.meegid.2004.08.001View ArticlePubMedGoogle Scholar
- Simmonds P: Genetic diversity and evolution of hepatitis C virus – 15 years on. J Gen Virol 2004, 85: 3173-3188. 10.1099/vir.0.80401-0View ArticlePubMedGoogle Scholar
- Germer JJ, Majewski DW, Rosser M, Thompson A, Mitchell PS, Smith TF, Elagin S, Yao JDC: Evaluation of the TRUGENE HCV 5'NC genotyping kit with the new GeneLibrarian module 3.1.2 for genotyping of hepatitis C virus from clinical specimens. J Clin Microbiol 2003, 41: 4855-4857. 10.1128/JCM.41.10.4855-4857.2003PubMed CentralView ArticlePubMedGoogle Scholar
- Martró E, González V, Buckton AJ, Saludes V, Fernández G, Matas L, Planas R, Ausina V: Evaluation of a new assay for hepatitis C virus (HCV) genotyping targeting both 5'NC and NS5b genomic regions, in comparison with reverse hybridization and sequencing methods. J Clin Microbiol 2008, 46: 192-197. 10.1128/JCM.01623-07PubMed CentralView ArticlePubMedGoogle Scholar
- Schutzbank TE, Sefers SE, Kahmann N, Li H, Tang YW: Comparative evaluation of three commercially available methodologies for hepatitis C virus genotyping. J Clin Microbiol 2006, 44: 3797-3798. 10.1128/JCM.01159-06PubMed CentralView ArticlePubMedGoogle Scholar
- Cantaloube JF, Laperche S, Gallian P, Bouchardeau F, de Lamballerie X, de Micco P: Analysis of the 5' noncoding region versus the NS5b region in genotyping hepatitis C virus isolates from blood donors in France. J Clin Microbiol 2006, 44: 2051-2056. 10.1128/JCM.02463-05PubMed CentralView ArticlePubMedGoogle Scholar
- Laperche S, Lunel F, Izopet J, Alain S, Dény P, Duverlie G, Gaudy C, Pawlotsky JM, Plantier JC, Pozzetto B, Thibault V, Tosetti F, Lefrère JJ: Comparison of hepatitis C virus NS5b and 5' noncoding gene sequencing methods in a multicenter study. J Clin Microbiol 2005, 43: 733-739. 10.1128/JCM.43.2.733-739.2005PubMed CentralView ArticlePubMedGoogle Scholar
- Hraber PT, Fischer W, Bruno WJ, Leitner T, Kuiken C: Comparative analysis of hepatitis C virus phylogenies from coding and non-coding regions: the 5' untranslated region (UTR) fails to classify subtypes. Virol J 2006, 3: 103. 10.1186/1743-422X-3-103PubMed CentralView ArticlePubMedGoogle Scholar
- Murphy DG, Willems B, Deschênes M, Hilzenrat N, Mousseau R, Sabbah S: Use of sequence analysis of the NS5B region for routine genotyping of hepatitis C virus with reference to C/E1 and 5' untranslated region sequences. J Clin Microbiol 2007, 45: 1102-1112. 10.1128/JCM.02366-06PubMed CentralView ArticlePubMedGoogle Scholar
- Lu L, Li C, Fu Y, Thaikruea L, Thongswat S, Maneekarn N, Apichartpiyakul C, Hotta H, Okamoto H, Netski D, Pybus OG, Murphy D, Hagedorn CH, Nelson KE: Complete genomes for hepatitis C virus subtypes 6f, 6i, 6j and 6m: viral genetic diversity among Thai blood donors and infected spouses. J Gen Virol 2007, 88: 1505-1518. 10.1099/vir.0.82604-0View ArticlePubMedGoogle Scholar
- Martin D, Rybicki E: RDP: detection of recombination amongst aligned sequences. Bioinformatics 2000, 16: 562-563. 10.1093/bioinformatics/16.6.562View ArticlePubMedGoogle Scholar
- Kuiken C, Hraber P, Thurmond J, Yusim K: The hepatitis C sequence database in Los Alamos. Nucleic Acids Res 2007, 36: D512-D516. 10.1093/nar/gkm962PubMed CentralView ArticlePubMedGoogle Scholar
- Feucht HH, Schröter M, Zöllner B, Polywka S, Nolte H, Laufs R: The influence of age on the prevalence of hepatitis C virus subtypes 1a and 1b. J Infect Dis 1997, 175: 685-688.View ArticlePubMedGoogle Scholar
- Simmonds P, Holmes EC, Cha TA, Chan SW, McOmish F, Irvine B, Beall E, Yap PL, Kolberg J, Urdea MS: Classification of hepatitis C virus into six major genotypes and a series of subtypes by phylogenetic analysis of the NS-5 region. J Gen Virol 1993, 74: 2391-2399. 10.1099/0022-1317-74-11-2391View ArticlePubMedGoogle Scholar
- Laar TJ, Koppelman MH, Bij AK, Zaaijer HL, Cuijpers HT, Piel CL, Coutinho RA, Bruisten SM: Diversity and origin of hepatitis C virus infection among unpaid blood donors in the Netherlands. Transfusion 2006, 46: 1719-1728. 10.1111/j.1537-2995.2006.00961.xView ArticlePubMedGoogle Scholar
- Staden R, Beal KF, Bonfield JK: The Staden Package, 1998. Methods Mol Biol 2000, 132: 115-130.PubMedGoogle Scholar
- Thompson JD, Higgins DG, Gibson TJ: CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucl Acids Res 1994, 22: 4673-4680. 10.1093/nar/22.22.4673PubMed CentralView ArticlePubMedGoogle Scholar
- Tamura K, Dudley J, Nei M, Kumar S: MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 2007, 24: 1596-1599. 10.1093/molbev/msm092View ArticlePubMedGoogle Scholar
- Akaike H: A new look at the statistical model identification. IEEE Trans Automatic Control 1974, 19: 716-723. 10.1109/TAC.1974.1100705View ArticleGoogle Scholar
- Posada D, Crandall KA: Selecting the best-fit model of nucleotide substitution. Syst Biol 2001, 50: 580-601. 10.1080/106351501750435121View ArticlePubMedGoogle Scholar
- Guindon S, Gascuel O: A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst Biol 2003, 52: 696-704. 10.1080/10635150390235520View ArticlePubMedGoogle Scholar
- Swofford DL: PAUP*. Phylogenetic analysis using parsimony (* and other methods). 4th edition. Sunderland, MA, Sinauer Associates; 2002.Google 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.