genotype diversity of hepatitis c virus (hcv) in hcv-associated liver disease patients in indonesia

CL IN ICAL STUDIES

Genotype diversityof hepatitis Cvirus (HCV) inHCV-associatedliver disease patients in IndonesiaAndi Utama1, Navessa Padma Tania1, Rama Dhenni1, Rino Alvani Gani2, Irsan Hasan2, Andri Sanityoso2,Syafruddin A. R. Lelosutan3, Ruswhandi Martamala3, Laurentius Adrianus Lesmana2, Ali Sulaiman4 and Susan Tai1

1 Molecular Epidemiology Division, Mochtar Riady Institute for Nanotechnology, Lippo Karawaci, Tangerang, Banten, Indonesia

2 Department of Internal Medicine, Hepatology Division, Faculty of Medicine, University of Indonesia, Jakarta, Indonesia

3 Department of Internal Medicine, Gastroentero-Hepatology Division, Gatot Soebroto Hospital, Jakarta, Indonesia

4 Klinik Hati ‘Prof. Ali Sulaiman’, Jakarta, Indonesia

Keywords

core region – genotype – hepatitis C virus –

Indonesia – liver disease

Correspondence

Andi Utama, PhD, Molecular Epidemiology

Division, Mochtar Riady Institute for

Nanotechnology, Jalan Boulevard Jend,

Sudirman 1688, Lippo Karawaci, Tanggerang

15810, Banten, Indonesia

Tel: 162 21 542 10123

Fax: 162 21 542 10110

e-mail : [email protected]

Received 4 March 2010

Accepted 26 April 2010

DOI:10.1111/j.1478-3231.2010.02280.x

AbstractBackground: Hepatitis C virus (HCV) genotype distribution in Indonesia hasbeen reported. However, the identification of HCV genotype was based on50-UTR or NS5B sequence. Aims: This study was aimed to observe HCV coresequence variation among HCV-associated liver disease patients in Jakarta,and to analyse the HCV genotype diversity based on the core sequence.Methods: Sixty-eight chronic hepatitis (CH), 48 liver cirrhosis (LC) and 34hepatocellular carcinoma (HCC) were included in this study. HCV corevariation was analysed by direct sequencing. Results: Alignment of HCV coresequences demonstrated that the core sequence was relatively varied amongthe genotype. Indeed, 237 bases of the core sequence could classify the HCVsubtype; however, 236 bases failed to differentiate several subtypes. Based on237 bases of the core sequences, the HCV strains were classified into genotypes1 (subtypes 1a, 1b and 1c), 2 (subtypes 2a, 2e and 2f) and 3 (subtypes 3a and3k). The HCV 1b (47.3%) was the most prevalent, followed by subtypes 1c(18.7%), 3k (10.7%), 2a (10.0%), 1a (6.7%), 2e (5.3%), 2f (0.7%) and 3a(0.7%). HCV 1b was the most common in all patients, and the prevalenceincreased with the severity of liver disease (36.8% in CH, 54.2% in LC and58.8% in HCC). These results were similar to a previous report based on NS5Bsequence analysis. Conclusion: Hepatitis C virus core sequence (237 bases)could identify the HCV subtype and the prevalence of HCV subtype based oncore sequence was similar to those based on the NS5B region.

Hepatitis C virus (HCV) infection is known to be a majorcontributor to chronic liver diseases including chronichepatitis, liver cirrhosis (LC) and hepatocellular carcino-ma (HCC). Variability of the RNA genome of HCV hasmade it possible to distinguish six genotypes and over 70subtypes (1). Relatively well-conserved regions of thegenome (50-UTR, C/E1 and NS5B) have been used asthe basis for the HCV classification (1, 2). The variedgenotypes differ in distribution both by geographicalregion and by mode of transmission. Based on NS5Band 50-UTR sequences, we demonstrated previously thatthe genotypes 1 (subtype 1a, 1b and 1c), 2 (mainlysubtype 2a) and 3 (mainly subtype 3k) are found amongblood donors and HCV-associated liver disease patientsin Indonesia (3), similar to others’ observations (4–6).

The current standard treatment for patients withchronic hepatitis C consists of pegylated a interferon(Peg-IFN) in combination with the nucleoside analogueribavirin for 24–48 weeks, and leads to a sustained

virological response in 54–56% of cases (7, 8). Sustainedvirological response is defined as undetectable HCV RNAby a sensitive assay (lower detection limit of o 50 IU/ml)at the end of a 24-week follow-up period after the end oftreatment. Patients who do not achieve a sustainedvirological response may be found to be HCV RNAnegative during therapy but may relapse thereafter, ormay be non-responders who have detectable HCV RNAthroughout the treatment period. Virological responserates have been shown to vary with host and viral factorssuch as age, weight, sex, race, liver enzymes and stage offibrosis, HCV genotype and HCV RNA concentration atbaseline (7–11). In addition, each genotype also displaysparticular features such as resistance to Peg-IFN andribavirin combination treatment, for example, genotype1-infected patients respond less efficiently to therapythan those infected with genotype 2 and 3 viruses (7).

Previous studies have reported that, especially in HCVgenotype 1b, polymorphisms of amino acid (aa) no. 70 of

Liver International (2010)1152 c� 2010 John Wiley & Sons A/S

Liver International ISSN 1478-3223

mailto:[email protected]

arginine (arg/R) or glutamine (gln/Q) and aa no. 91 ofleucine (leu/L) or methionine (met/M) in the HCV coreprotein were significantly associated with the response totreatment with the standard Peg-IFN combined withribavirin (12–17). In addition, amino acid substitutionsin the core region of HCV genotype 1b are also theimportant predictor of hepatocarcinogenesis as well assevere insulin resistance in patients without cirrhosis anddiabetes mellitus (18, 19). It is also reported recently thatsubstitutions in the amino acids of the HCV genotype 1bcore region are associated with hepatic steatosis andoxidative stress in patients with chronic hepatitis C (20).Therefore, analysis of the sequence, including particularmutations, in the HCV core region is definitely necessary.

In this study, we analysed the HCV core sequence inserum samples obtained from patients with CH, LC andHCC, and observed the variation of the region. Weadditionally tested whether partial HCV core sequenceallowed the identification of HCV subtype, and com-pared the HCV subtype based on HCV core and NS5Bsequences.

Methods

Samples

Blood samples were obtained from 68 CH patients, 48 LCpatients and 34 HCC patients. Blood of CH, LC andHCC patients were collected from Cipto Mangunkusu-mo Hospital, Gatot Soebroto Hospital and Klinik Hati,Jakarta, Indonesia, from May 2006 until June 2009.Blood samples were collected from each patient at thetime of their clinical evaluation, then separated into seraand stored at � 70 1C until use. All sera were positive forantibody to HCV antibody as determined using a third-generation microparticle enzyme immunoassay (AxSYMHCV, version 3.0; Abbott Laboratories, Chicago, IL,USA). This study was approved by the InstitutionalEthics Committee and informed consent was obtainedfrom each patient.

Viral RNA extraction and amplification of hepatitis Cvirus core region

Serum samples that had been stored at � 70 1C wereretrieved for analysis. HCV RNA was extracted from200 ml serum using a High Pure Viral RNA kit (RocheDiagnostics, Mannheim, Germany) according to themanufacturer’s protocol. The RNA was finally eluted in50 ml of ribonuclease-free water and was used for genomeamplification of the partial core region by a nestedreverse transcriptase-polymerase chain reaction (RT-PCR). First-round RT-PCR was performed using aSuperScriptTM III One- Step RT-PCR System (Invitro-gen, Carlsbad, CA, USA) in 25 ml aliquots containing5 ml RNA, 1x reaction mix, 1 unit of Superscript III RT/Platinum Taq mix and 0.5mM of 50-UTR1 (50-CCC-TGTGAGGAACTWCTGTCTTCACGC-30) and Core-R1(50-AAGATAGARAARGAGCAACC-30) primers. The

following cycling parameters were used for the first-roundRT-PCR: cDNA synthesis at 55 1C (35 min), enzymeinactivation at 95 1C (2 min), 35 cycles of DNA amplifica-tion: denaturation at 95oC (1 min), annealing at 45oC(30 s) and elongation at 72 1C (1 min). Three microlitresof PCR product from first-round PCR was used as atemplate for second-round PCR. The second-round PCRwas performed using the Go Taq PCR Core System(Promega, Madison, WI, USA) in 25ml aliquots contain-ing 2 mM MgCl2, 0.2 mM dNTPs, 1x Green Go Taq FlexiBuffer, 0.025 units of Go Taq DNA Polymerase and 0.5mMof 50-UTR2 (50-TCTAGCCATGGCGTTAGTAYGAGTGT-30) and Core-R2 (50-ATGTACCCCATGAGGTCGGC-30)primers. The following cycling parameters were usedfor 35 cycles of second-round PCR: denaturation at95 1C (1 min), annealing at 45 1C (30 s) and elongation at72 1C (1 min). A PCR product of 674 bp, which included264 bp of 50-UTR and 410 bp of HCV core sequences,was purified using a Wizard SV Gel and PCR Clean Upsystem kit (Promega), according to the manufacturer’sprotocol.

Sequencing of hepatitis C virus core region andphylogenetic analysis

Purified DNA fragments were directly sequenced withprimer 50-UTR2 or Core-R2 using ABI 3130 xl GeneticAnalyzers (Applied Biosystem, Foster City, CA, USA)with the Big Dye Terminator V3.1 Cycle Sequencingkit (Applied Biosystem). Amplification products thatshowed an ambiguous sequence were inserted intopBluescript II SK (1), and five independent clones ofeach were sequenced. For comparison, NS5B region wasalso sequenced as described previously (3). The nucleo-tide sequences of the HCV and a panel of sequencesretrieved from the GenBank were aligned using BIOEDIT

Sequence alignment Editor software. Phylogenetic treeswere constructed using the neighbour-joining method(21) with Kimura’s two-parameter (22) and 1 000 repli-cates of bootstrap resampling as implemented in MEGA 4.1(Center for Evolutionary Functional Genomics, Tempe,AZ, USA) (23).

Measurement of hepatitis C viral load

Levels of HCV viral load were assessed using a commer-cially available kit (Diagnostic Kit for Quantification ofHepatitis C Virus RNA; PG Biotech, Shenzhen, China)according to the manufacturer’s instructions. The lowestdetectable titre with this kit was 5.0� 102 IU/ml. A viralload higher than 1.0� 104 IU/ml was regarded as a highvirus titre, and that of 1.0� 104 IU/ml or lower wasregarded as a low titre.

Statistical analysis

All statistical analyses were performed using SPSS Statistics17.0 (SPSS Inc., Chicago, IL, USA). All continuous

Liver International (2010)c� 2010 John Wiley & Sons A/S 1153

Utama et al. Genotype diversity of HCV in Indonesia

variables were compared using Mann–Whitney’s test.The w2-test was used to compare categorical data.Po 0.05 was considered significant.

Results

Demographical data of patients are shown in Table 1. Themean age of patients was significantly higher in LC andHCC groups compared with the CH group. Male andfemale ratio was increased from CH to LC and from LCto HCC groups; however, the increment was not statisti-cally significant. AFP value was tremendously high inHCC group than LC or CH group. A high virus titrewas detected in almost all of the samples. The virus titrewas significantly higher in CH than HCC (P = 0.01), butthere was no significant difference of virus titre betweenCH and LC (P = 0.32) and between LC and HCC(P = 0.09).

Hepatitis C virus core sequence variation

As the HCV 50-UTR region is well conserved, forwardprimers (50-UTR1 and 50-UTR2) used in a previous studyfor the amplification of the 50-UTR region (3) were used.For reverse primers, we designed new primers as described

in the ‘Methods’. As a result, the length of the nestedRT-PCR product was 674 bp (nucleotide no. 78–751),which consisted of 264 bp (nucleotide no. 78–341) of partial50-UTR and 410 bp (nucleotide no. 342–751) of coresequences. In most samples, the sequences of RT-PCRproduct could be read for 620 bases (nucleotide no.115–734), consisting of 227 bases (nucleotide no. 115–341)of 50-UTR and 393 bases (nucleotide no. 342–734) of HCVcore regions. However, in a few samples the sequence couldonly be read for 415 bases (nucleotide no. 207–621),consisting of 135 bases (nucleotide no. 207–341) of 50-UTR and 280 bases (nucleotide no. 342–621) of coreregions. Therefore, we initially analysed 415 bases ineach of the sequences. Alignment of sequences from 150samples demonstrated the deletion of three codons in theHCV core (codon no. 28–30) in one CH sample(06.10.071), which belongs to HCV subtype 1c (data notshown). Alignment of 280 bases of HCV core sequencesdemonstrated that several nucleotides were correlated witha specific HCV genotype/subtype (Fig. 1). These resultssuggested that these partial HCV core sequences might besufficient to identify HCV genotype. On the other hand,amino acid sequence of HCV core was less specific for eachsubtype (data not shown).

Table 1. Demographical characteristics of patients

All patients CH, n = 68 LC, n = 48 HCC, n = 34 Total, n = 150

Age (years), mean� SD� 54.0�15.2 64.3� 10.0 66.9�8.8 60.2�13.6Male/female (%male) 36/32 (52.9) 28/20 (58.3) 23/11 (67.6) 87/63 (58.0)AFP (ng/ml), median (range)w 4.7 (0.4–8900.0) 9.5 (0.9–386.4) 235.0 (2.0–501757.4) 9.1 (0.4–501757.4)Viral load (log10 IU/ml), median (range)z,‰ 6.1 (2.9–7.8) 5.8 (3.3–6.8) 5.5 (4.2–7.2) 5.8 (2.9–7.8)Viral load (log10 IU/ml), mean� SDz,z 6.6� 6.9 6.1�6.2 6.0� 6.5 6.4�6.8

Viral load group, high/lowz 62/1 40/1 29/0 131/2HCV subtype 1b-infected patients CH, n = 25 LC, n = 26 HCC, n = 20 Total, n = 71Age (years), mean� SD 62.2�8.8 63.7� 10.6 65.1�9.4 63.6�9.6Male/female (%male) 10/15 (40.0) 14/12 (53.9) 14/6 (70.0) 38/33 (53.5)AFP (ng/ml), median (range)k 10.2 (0.4–8900.0) 8.7 (0.9–341.4) 214.3 (2.0–277659.6) 13.7 (0.4–277659.6)Viral load (log10 IU/ml), median (range)�� 6.0 (4.7–7.2) 5.8 (3.3–6.8) 5.3 (4.6–7.2) 5.8 (3.3–7.2)Viral load (log10 IU/ml), mean� SD ��,ww 6.4� 6.6 6.0�6.2 6.2� 6.6 6.2�6.5Viral load group, high/low�� 23/0 20/1 16/0 59/1Amino acid substitutions in HCV core (%)

R70Q/H 9 (36.0) 7 (26.9) 7 (35.0) 23 (32.4)L91M 10 (40.0) 6 (23.1) 7 (35.0) 23 (32.4)R70Q/H and L91M 4 (16.0) 2 (7.7) 3 (15.0) 7 (9.9)R70Q/H and/or L91M 15 (60.0) 11 (42.3) 11 (55.0) 37 (52.1)75A 15 (60.0) 16 (61.5) 10 (50.0) 41 (57.8)75T 7 (28.0) 9 (34.6) 6 (30.0) 22 (31.0)75S 3 (12.0) 1 (3.9) 2 (10.0) 6 (8.5)

�Differences were statistically significant between CH vs LC and CH vs HCC.

wDifferences were statistically significant between CH vs LC, CH vs HCC, and LC vs HCC.

zAvailable for 63 CH, 41 LC, and 29 HCC patients (total 133 patients).

‰Differences were statistically significant between CH vs HCC.

zDifferences were statistically significant between CH vs HCC.

kDifferences were statistically significant between CH vs HCC and LC vs HCC.��Available for 23 CH, 21 LC, and 16 HCC patients (total 60 patients).

wwDifferences were statistically significant between CH vs HCC.

AFP, a-fetoprotein; CH, chronic hepatitis; LC, liver cirrhosis; HCC, hepatocellular carcinoma; SD, standard deviation.


Genotype diversity of HCV in Indonesia Utama et al.

Hepatitis C virus core sequence for genotyping

To observe the region within the HCV 50-UTR and coreregions that could be used for HCV subtype identifica-tion, different lengths of the 50-UTR and core region thathad been partially sequenced were examined. Analysis of416 bases sequence of 50-UTR and core regions as well as280 bases of the core region allowed the identification ofthe HCV subtype (data not shown). Hence, this subtypecould be assigned with the minimal requirement ofsequence; we were interested in determining the mini-

mum length of core sequence that could be used for theidentification of the HCV subtype. Analysis of 237 bases(nucleotide no. 342–578) of the HCV core sequencecould classify the HCV into a subtype (Fig. 2a); however,236 bases (nucleotide no. 342–577) of the core regionfailed to differentiate subtypes 3a, 3b and genotype 6(Fig. 2b). Thus, we propose that 237 bases of the HCVcore region was the minimum length that can be used forthe identification of HCV subtype. By the phylogeneticanalysis of 237 bases core region, the HCV strains from150 samples were classified into genotypes 1 (withsubtypes 1a, 1b and 1c), 2 (with subtypes 2a, 2e and 2f)and 3 (with subtypes 3a and 3k) (Table 2). The subtype1b was the most prevalent, accounting for 47.3% of thetotal samples, followed by subtypes 1c (18.7%), 3k(10.7%), 2a (10.0%), 1a (6.7%), 2e (5.3%), 2f (0.7%)and 3a (0.7%) (Table 2).

Hepatitis C virus genotype and clinical diagnosis

When the genotype distribution was compared withclinical diagnosis, it was found that in the CH group theHCV subtype prevalence was as follows: 1b (36.8%)41c (22.1%)4 3k (14.7%)4 1a = 2a (8.8%)42e (7.4%)4 3a (1.5%). On the other hand, in the LCgroup, the percentage of HCV subtype was in thefollowing order: 1b (54.2%)4 1c (16.7%)4 2a(10.4%)4 1a = 3k (6.3%)4 2e (4.2%)4 2f (2.1%),and no subtype 3a was found. In addition, in the HCCgroup, the following order of HCV prevalence wasdetected: 1b (58.8%)4 1c (14.7%)4 2a (11.8%)4 3k(8.8%)4 1a = 2e (2.9%), and subtypes 2f and 3a werenot found. If the prevalence was classified into threegroups [group I (the most prevalent), 4 30%; group II(moderately prevalent) and 4–30%; group III (less pre-valent), o 4%], the prevalence pattern of HCV subtypewas most likely the same in all clinical diagnosis ofsamples, except for HCV subtype 1a, which was classifiedas moderately prevalent in CH and LC, but was classifiedas less prevalent in HCC. Particularly for the HCVsubtype 1b, it was the most common in all clinicaldiagnosis status and the prevalence was increased withthe severity of liver disease (36.8% in CH, 54.2% in LCand 58.8% in HCC) (Fig. 3). However, statistical analysisshowed a significant difference only between CH andHCC (P = 0.035), but not between CH and LC(P = 0.064) and LC and HCC (P = 0.677).

Amino acid substitutions in the hepatitis C virus 1b coreregion

Of 150 samples, 71 samples were classified into HCVsubtype 1b. Particularly in HCV 1b, three amino acidsubstitutions (aa 70, 75 and 91) were frequently detected.It was found that in 32.4% of HCV 1b, arginine (R) wassubstituted by glutamine (Q) or histidine (H) in aa 70,and leucine (L) was replaced by methionine (M) in aa 91(Table 1). Either mutation in aa 70 or aa 91 was detected

Fig. 1. Nucleotide polymorphism in hepatitis C virus core region.Dots indicate nucleotides identical to those of the subtype 1asequence, used as a reference sequence (GenBank accession no.AF009606, H77 isolate). Nucleotides specific to genotypes orsubtypes are highlighted.



in 52.1% of patients infected with HCV 1b. Amino acidsin position 70 and 91 of the core region have beenproposed as determinants for patients to Peg-IFN andribavirin therapy. These results suggested that more thanhalf of Indonesian patients infected with HCV 1b arepredicted to be resistant to Peg-IFN and ribavirin ther-apy. Furthermore, in aa 75, alanine (A) was found to bereplaced by threonine (T) in 31.0% of samples (Table 1).Amino acid substitution in this position has been re-ported previously (20); however, the role of this aminoacid remains unknown.

Putative hepatitis C virus recombinants

To confirm the accuracy of HCV genotypes based on thepartial HCV core sequence, the HCV genotyping resultsfrom the HCV core sequence were compared with thosefrom the NS5B region. Of 150 samples, which successfullygenotyped by using the HCV core sequence, the genotypebased on the NS5B sequence of 73 samples (accessionnumber GQ418294–GQ418298, GQ418300–GQ418324,GQ418326–GQ418334, GQ418336, GQ418338–GQ418339,GQ418341–GQ418344, GQ418346–GQ418349, GQ418352–GQ418354, GQ418356–GQ418358, GQ418360–GQ418363, GQ418365–GQ418366, GQ418368, GQ418370–GQ418371 and GQ418381–GQ418388) have been reportedpreviously (3). The remaining 77 samples were analysed inthis study; however, only 71 samples were successfullyamplified and sequenced (accession number GU441387-GU441457). Thus, the data of the HCV genotype-basedNS5B sequence of 144 samples were publicly available.Comparison of HCV identification based on the HCV core(237 bases) and NS5B (367 bases) of these 144 samplesrevealed that the genotyping results based on the HCV coreand NS5B regions were different in nine samples, whichcomposed of five samples from CH (06.10.034, 06.10.080,08.40.072, 08.80.014 and 08.80.070), one sample from LC(08.80.075) and three samples from HCC (06.10.009,P.X00.15 and P.X00.16) (Table 3). Three samples(06.10.080, 08.40.072 and P.X00.15) were identified assubtype 1b based on the HCV core region, but based onthe NS5B region they were classified as HCV genotype 1c, 1aand 2a, which resulted in putative recombinants of 1b/1c,1b/1a and 1b/2a respectively. Similarly, three samples(06.10.034, 08.80.014 and 06.10.009) were classified as

A B

Fig. 2. Phylogenetic analysis of 237 bases (A) and 236 bases (B) of the hepatitis C virus core region. Genotyping based on 236 bases of the coreregion could not differentiate between subtypes 3a and 3b and genotype 6 as indicated by an asterisk.

Table 2. Hepatitis C virus (HCV) genotype prevalence in differentclinical diagnosis based on 237 bases of HCV core region

Genotype

Number (%) of samples in each clinicaldiagnosis

Total, n = 150CH, n = 68 LC, n = 48 HCC, n = 34

Genotype 146 (67.6) 37 (77.1) 26 (76.5) 109 (72.7)1a 6 (8.8) 3 (6.3) 1 (2.9) 10 (6.7)1b 25 (36.8) 26 (54.2) 20 (58.8) 71 (47.3)1c 15 (22.1) 8 (16.7) 5 (14.7) 28 (18.7)Genotype 211 (16.2) 8 (16.7) 5 (14.7) 24 (16.0)2a 6 (8.8) 5 (10.4) 4 (11.8) 15 (10.0)2e 5 (7.4) 2 (4.2) 1 (2.9) 8 (5.3)2f 0 (0.0) 1 (2.1) 0 (0.0) 1 (0.7)Genotype 311 (16.2) 3 (6.3) 3 (8.8) 17 (11.3)3a 1 (1.5) 0 (0.0) 0 (0.0) 1 (0.7)3k 10 (14.7) 3 (6.3) 3 (8.8) 16 (10.7)All 68 (100.0) 48 (100.0) 34 (100.0) 150 (100.0)

CH, chronic hepatitis; LC, liver cirrhosis; HCC, hepatocellular carcinoma.



subtype 3k based on the HCV core region; however, theywere grouped into respectively HCV subtype 2a, 1b and 2abased on the NS5B region. Thus, these three putativerecombinants pattern were 3k/2a, 3k/1b and 3k/2a respec-tively. Two samples, 08.80.070 and P.X00.16, demonstratedrespectively 2a/1a and 2a/1b recombination patterns. An-other one sample (08.80.075) had a putative recombinationof 1a/2a (Table 3). Analysis of the core and NS5B sequencesof these nine samples was repeated three times to avoidexperimental errors, and each independent experiment gavethe same results.

Discussion

In addition to the complete genome sequence, relativelyshort sub-genome regions such as 50-UTR, core, core/E1,NS5B have been used as the basis for HCV classification(1, 2). The 50-UTR sequence is well conserved, and henceprimers hybridize, making it easy to amplify, but the lackof variation limits its usefulness for determining differ-ences among various subtypes (24). The nearby core

region is also relatively well conserved, with nucleotidesequence similarity ranging from 81 to 88% in isolates ofdifferent genotypes (25). Because of their high degree ofconservation, these regions have been chosen for design-ing generic primer sequences, enabling direct PCR am-plification in all HCV types. Normally, classification ofHCV using a relatively long sequence of core region(360–405 bases) is consistent with that using the NS5Bregion (26–29). In this study, we have tried to use regionsof the core sequence that are as short as possible, andfound that a 237-base region could distinguish allsubtypes, at least for viruses which were transmitted inJakarta, Indonesia. However, if the length of sequencewas reduced, even only by one base (to be 236 bases), itcould not differentiate between the HCV virus subtypes,especially subtypes 3a and 3b and genotype 6. Thus, itwas shown that a relatively short segment of HCV coresequence (237 bases) could be used for the identificationof HCV subtype. Sometimes it is difficult to obtain longsequence data because of technical problems; hence, theshorter length required for the identification of HCVsubtype will improve the likelihood of obtaining positivedata. Furthermore, in the present study, forward primersused previously for the amplification of 50-UTR region(3) were used for the amplification of the core region inorder to increase the probability of positive amplifica-tion. This is because the 50-UTR region is more con-served compared with the core region, or even the mostconserved region in the HCV genome.

Phylogenetic analysis of 237 bases of the HCV coreregion demonstrated that the prevalence of HCV sub-type was in the following order: 1b (47.3%)4 1c(18.7%)4 3k (10.7%)4 2a (10.0%)4 1a (6.7%)4 2e(5.3%)4 2f (0.7%) = 3a (0.7%). In general, these resultsare similar to our previous data of HCV genotype basedon the NS5B region. In our previous study, we reportedthe HCV subtype based on the NS5B region fromasymptomatic and liver disease patients, and found thatsubtype 1b (36.5%) was the most prevalent in oursamples, followed by subtypes 3k (15.4%), 2a (14.4%), 1a(12.5%), 1c (12.5%), 2e (4.8%), 2f (1.0%), 3a (1.0%), 3b(1.0%) and 4a (1.0%) (3). If the asymptomatic carrierswere excluded, the prevalence of HCV subtype was asfollows: 1b (45.2%)4 2a (15.5%)4 1c (14.3%)4 3k(9.5%)4 1a (7.1%)4 2e (6.0%)4 2f (1.2%) = 3a(1.2%), which is slightly different with previous report.However, if the prevalence was classified into three groups(group I (mostly prevalent),4 30%; group II (moderatelyprevalent), 4–30%; and group III (less prevalent),o 4%),the subtypes will fall into in the same pattern. In otherwords, the order of prevalence will be the same in bothstudies; group I (1b)4 group II (1a, 1c, 2a, 2e,3k)4 group III (3a, 2f). Thus, the 237 bases of the HCVcore sequence could be used as an alternative for HCVsubtype classification. The slight different in HCV subtypeclassification observed in the present study might bebecause of the possible recombination of the HCV withdifferent genotype/subtype.

0

10

20

30

40

50

60

70

CH (n=68) LC (n=48) HCC (n=34)

Clinical diagnosis

HC

V 1

b p

reva

len

ce (

%)

P=0.035

P=0.064 P=0.677

Fig. 3. Prevalence of hepatitis C virus subtype 1b in the samplesfrom different clinical diagnosis. CH, chronic hepatitis; LC, livercirrhosis; HCC, hepatocellular carcinoma.

Table 3. Samples which showed different genotype from analysisbased on hepatitis C virus core and NS5B regions

Report number Clinical diagnosis

HCV genotype based on

Core NS5B

06.10.080 CH 1b 1c08.40.072 CH 1b 1a08.80.070 CH 2a 1a06.10.034 CH 3k 2a08.80.014 CH 3k 1b08.80.075 LC 1a 2aP.X00.15 HCC 1b 2aP.X00.16 HCC 2a 1b06.10.009 HCC 3k 2a

CH, chronic hepatitis; LC, liver cirrhosis; HCC, hepatocellular carcinoma.



When genotype distribution was analysed based onclinical diagnosis (CH, LC and HCC), it was foundthat HCV subtype 1b was the most prevalent in alldifferent clinical diagnosis. The prevalence pattern ofHCV subtype was similar in all clinical diagnosis ofsamples, except for the HCV subtype 1a, which wasclassified as moderately prevalent group in CH and LC,but was classified as the less prevalent group in HCC.Particularly for HCV subtype 1b, it was the mostprevalent for all of the different clinical diagnosis, andthe prevalence increased with the severity of liver disease(36.8% in CH, 54.2% in LC and 58.8% in HCC),although the statistical analysis only demonstrated asignificant difference between CH and HCC (Table 2,Fig. 3). Similar results were observed from the analysis ofthe NS5B region (3).

Several reports demonstrated that amino acid substi-tutions at position 70 and/or 91 in the HCV core regionof the HCV subtype 1b could be used as predictors ofpoor virological response to 48-week Peg-IFN and riba-virin combination therapy (12–16, 30, 31), of risk factorsfor hepatocarcinogenesis (18, 19) and of hepatic steatosisand oxidative stress (20). We found that the prevalence ofamino acid substitutions at position 70 and/or 91 inHCV core was high (52.1%) (Table 1). These resultssuggest that a high percentage of Indonesian patientswho were infected with HCV 1b are predicted to beresistant to Peg-IFN and ribavirin therapy; however, acohort study has to be performed to confirm thishypothesis. In addition, we also found relatively frequentsubstitutions of amino acid at position 75, similar with aprevious report (20). However, the role of this aminoacid substitution has not been reported and warrantsfurther investigation.

Comparison of phylogenetic analysis of HCV core andNS5B regions showed differences of HCV genotype innine samples, implying a possible recombination be-tween different genotypes (intergenotypic) or differentsubtypes in the same genotype (intragenotypic). Theexperiment was repeated at least three times in these ninesamples, and the same results were obtained. There wereeight patterns of putative recombination including twointragenotypic recombinations (1b/1a and 1b/1c) and sixintergenotypic recombination (1a/2a, 1b/2a, 2a/1a, 2a/1b, 3k/1b and 3k/2a). Only two samples were with thesame pattern 3k/2a, whereas all samples had a differentrecombination pattern (Table 3). Further sequence ana-lysis of full-length genome is needed to prove that arecombination event has occurred. Nevertheless, priorinvestigators reported the natural recombination of HCVfrom samples in Russia, Uzbekistan and Ireland (2k/1b),Peru (1b/1a), the Philippines (2b/1b), Vietnam (2i/6p)and France (2/5) (32–38). Recently, the recombination ofHCV 2a/1a, 3a/1b and 2b/6w was detected in drug usersco-infected with HIV in Taiwan (39). Among nineputative recombinants found in our samples, five, threeand one samples were CH, HCC and LC groups respec-tively. Thus, the percentage of putative recombinant was

relatively high in CH; however, it is still not knownwhether there is an association between the prevalenceof recombination and the stage of the disease. In general,a virus recombination event could naturally occur duringvirus replication, and the occurrence of recombinationwill depend on the presence of different genotypes in apopulation (32). Although the recombinant is not neces-sarily more virulent, it will contribute to HCV evolution.Thus, recombination may play a role in creating geneticdiversity of the virus, which may have important im-plications for the pathogenesis, laboratory diagnosis andtreatment of HCV infection. For instance, the genotypesof HCV are considered predictors for the outcome ofinterferon treatment, and interferon resistance has beenattributed to sequence variability of the NS5A, E2 andcore proteins. If the HCV genotypes are identified basedon one particular subgenomic region, it may inaccuratelyidentify the genotype and it may lead to misprediction ofthe interferon response. In other words, genotypingbased on one subgenomic region may not always be avalid means of predicting the response to interferon.Therefore, analysis of more than one subgenomic regionis necessary to avoid missing recombinant strains.

In conclusion, our results indicate that a partialsequence of HCV core region could identify the HCVsubtype as the prevalence of HCV subtype based on thecore sequence was similar to those based on the NS5Bregion. Because the HCV core sequence might provideinformation for patients’ response to Peg-IFN and Riba-virin combination therapy while it might also predict therisk for HCC, hepatic steatosis and oxidative stress in chronicHCV, we therefore recommend the analysis of HCV coresequence as the basis of HCV genotyping in the future.

Acknowledgements

The authors thank Griskalia Christine and Shinta Sorayafor sample collection and Theresia Imelda Octavia fortechnical assistance. The authors also thank Dr DavidVaux (La Trobe University, Australia) for critical readingof the manuscript. This work was supported by MRINResearch Funding (budget no. cc041/2009).

References

1. Simmonds P, Bukh J, Combet C, et al. Consensus proposalsfor a unified system of nomenclature of hepatitis C virusgenotypes. Hepatology 2005; 42: 962–73.

2. Simmonds P, Smith DB, McOmish F, et al. Identification ofgenotypes of hepatitis C virus by sequence comparisons in thecore, E1 and NS-5 regions. J Gen Virol 1994; 75: 1053–61.

3. Utama A, Budiarto BR, Monasari D, et al. Hepatitis C virusgenotype in blood donors and associated liver disease inIndonesia. Intervirology 2008; 51: 410–6.

4. Hotta H, Handajani R, Lusida MI, et al. Subtype analysis ofhepatitis C virus in Indonesia on the basis of NS5B regionsequences. J Clin Microbiol 1994; 32: 3049–51.



5. Soetjipto, Handajani R, Lusida MI, et al. Differentialprevalence of hepatitis C virus subtypes in healthy blooddonors, patients on maintenance hemodialysis, and pa-tients with hepatocellular carcinoma in Surabaya, Indone-sia. J Clin Microbiol 1996; 34: 2875–80.

6. Inoue Y, Sulaiman HA, Matsubayashi K, et al. Genotypicanalysis of hepatitis C virus in blood donors in Indonesia.Am J Trop Med Hyg 2000; 62: 92–8.

7. Manns MP, McHutchison JG, Gordon SC, et al. Peginter-feron alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C:a randomized trial. Lancet 2001; 358: 958–65.

8. Fried MW, Shiffman ML, Reddy KR, et al. Peginterferonalfa-2a plus ribavirin for chronic hepatitis C virus infection.N Engl J Med 2002; 347: 975–82.

9. Myers RP, Patel K, Pianko S, et al. The rate of fibrosisprogression is an independent predictor of the response toantiviral therapy in chronic hepatitis C. J Viral Hepat 2003;10: 16–22.

10. Hadziyannis SJ, Sette H Jr, Morgan TR, et al. Peginterferon-alpha2a and ribavirin combination therapy in chronichepatitis C: a randomized study of treatment duration andribavirin dose. Ann Intern Med 2004; 140: 346–55.

11. Jeffers LJ, Cassidy W, Howell CD, et al. Peginterferon alfa-2a (40 kd) and ribavirin for black American patients withchronic HCV genotype 1. Hepatology 2004; 39: 1702–8.

12. Akuta N, Suzuki F, Sezaki H, et al. Association of aminoacid substitution pattern in core protein of hepatitis C virusgenotype 1b high viral load and non-virological response tointerferon-ribavirin combination therapy. Intervirology2005; 48: 372–80.

13. Akuta N, Suzuki F, Sezaki H, et al. Predictive factors ofvirological nonresponse to interferon-ribavirin combinationtherapy for patients infected with hepatitis C virus ofgenotype 1b and high viral load. J Med Virol 2006; 78: 83–90.

14. Akuta N, Suzuki F, Kawamura Y, et al. Prediction ofresponse to pegylated interferon and ribavirin in hepatitisC by polymorphisms in the viral core protein and very earlydynamics of viremia. Intervirology 2007; 50: 361–8.

15. Akuta N, Suzuki F, Kawamura Y, et al. Predictive factors ofearly and sustained responses to peginterferon plus ribavir-in combination therapy in Japanese patients infected withhepatitis C virus genotype 1b: amino acid substitutions inthe core region and low-density lipoprotein cholesterollevels. J Hepatol 2007; 46: 403–10.

16. Akuta N, Suzuki F, Kawamura Y, et al. Predictors of viralkinetics to peginterferon plus ribavirin combination ther-apy in Japanese patients infected with hepatitis C virusgenotype 1b. J Med Virol 2007; 79: 1686–95.

17. Mori N, Imamura M, Kawakami Y, et al. Randomized trialof high-dose Interferon-a-2b combined with Ribavirin inpatients with chronic hepatitis C: correlation between aminoacid substitutions in the Core/NS5A region and virologicalresponse to interferon therapy. J Med Virol 2009; 81: 640–9.

18. Akuta N, Suzuki F, Kawamura Y, et al. Amino acidsubstitutions in the hepatitis C virus core region are theimportant predictor of hepatocarcinogenesis. Hepatology2007; 46: 1357–64.

19. Akuta N, Suzuki F, Hirakawa M, et al. Amino acid substitutionsin the hepatitis C virus core region of genotype 1b are the im-portant predictor of severe insulin resistance in patients withoutcirrhosis and diabetes mellitus. J Med Virol 2009; 81: 1032–9.

20. Tachi Y, Katano Y, Honda T, et al. Impact of amino acidsubstitutions in the hepatitis C virus genotype 1b core regionin liver steatosis and hepatic oxidative stress in patients withchronic hepatitis C. Liver Int 2010; 30: 554–9.

21. Saitou N, Nei M. The neighbor-joining method: a newmethod for reconstructing phylogenetic trees. Mol Biol Evol1987; 4: 406–25.

22. Kimura M. A simple method for estimating evolu-tionary rate of base substitutions through comparativestudies of nucleotide sequences. J Mol Evol 1980; 16: 111–20.

23. Tamura K, Dudley J, Nei M, et al. MEGA4: MolecularEvolutionary Genetics Analysis (MEGA) software version4.0.. Mol Biol Evol 2007; 24: 1596–9.

24. Hraber PT, Fischer W, Bruno WJ, et al. Comparativeanalysis of hepatitis C virus phylogenies from coding andnon-coding regions: the 50 untranslated region (UTR) failsto classify subtypes. Virol J 2006; 3: 103.

25. Cantaloube JF, Laperche S, Gallian P, et al. Analysis of the 50

noncoding region versus the NS5b region in genotypinghepatitis C virus isolates from blood donors in France.J Clin Microbiol 2006; 44: 2051–6.

26. Mellor J, Walsh EA, Prescott LE, et al. Survey of type 6group variants of hepatitis C virus in Southeast Asia byusing a core-based genotyping assay. J Clin Microbiol 1996;34: 417–23.

27. Ohno O, Mizokami M, Wu RR, et al. New hepatitis C virus(HCV) genotyping system that allows for identification ofHCV genotypes 1a, 1b, 2a, 2b, 3a, 3b, 4, 5a, and 6a. J ClinMicrobiol 1997; 35: 201–7.

28. Lole KS, Jha JA, Shrotri SP, et al. Comparison of hepatitis Cvirus genotyping by 50 noncoding region- and core-basedreverse transcriptase PCR assay with sequencing and use ofthe assay for determining subtype distribution in India.J Clin Microbiol 2003; 41: 5240–4.

29. Chinchai T, Labout J, Noppornpanth S, et al. Com-parative study of different methods to genotype hepatitisC virus type 6 variants. J Virol Methods 2003; 109: 195–201.

30. Donlin MJ, Cannon NA, Yao E, et al. Pretreatmentsequence diversity differences in the full-length hepatitis Cvirus open reading frame correlate with early response totherapy. J Virol 2007; 81: 8211–24.

31. Okanoue T, Itoh Y, Minami M, et al. Guidelines for theantiviral therapy of hepatitis C virus carriers with normalserum aminotransferase based on platelet counts. HepatolRes 2008; 38: 27–36.

32. Kalinina O, Norder H, Mukomolov S, et al. A naturalintergenotypic of hepatitis C virus identified in St. Peters-burg. J Virol 2002; 76: 4034–43.

33. Colina R, Casane D, Vasquez S, et al. Evidence of intratypicrecombination in natural populations of hepatitis C virus.J Gen Virol 2004; 85: 31–7.

34. Kageyama S, Agdamag DM, Alesna ET, et al. A naturalinter-genotypic (2b/1b) recombinant of hepatitis C virus inthe Philippines. J Med Virol 2006; 78: 1423–8.



35. Noppornpanth S, Lien TX, Poovorawan Y, et al. Identifica-tion of a naturally occurring recombinant genotype 2/6hepatitis C virus. J Virol 2006; 80: 7569–77.

36. Moreau I, Hegarty S, Levis J, et al. Serendipitous identifica-tion of natural intergenotypic recombinants of hepatitis Cvirus in Ireland. Virol J 2006; 3: 95.

37. Legrand-Abravanel F, Claudinon J, Nicot F, et al. Newnatural intergenotypic (2/5) recombinant of hepatitis Cvirus. J Virol 2007; 81: 4357–62.

38. Kurbanov F, Tanaka Y, Avazova D, et al. Detection ofhepatitis C virus natural recombinant RF1_2k/1b strainamong intravenous drug users in Uzbekistan. Hepatol Res2008; 38: 457–64.

39. Lee YM, Lin HJ, Chen YJ, et al. Molecular epidemiologyof HCV genotypes among injection drug users in Taiwan:full-length sequences of two new subtype 6w strains and arecombinant form_2b6w. J Med Virol 2010; 82: 57–68.



genotype diversity of hepatitis c virus (hcv) in hcv-associated liver disease patients in indonesia

Documents