Tracing hepatitis B virus to the 16th century in a Korean mummy

Authors


  • Potential conflict of interest: Nothing to report.

  • The study in Israel was partially funded by the Liver Unit and the Hadassah-Hebrew University Salzberg family endowment for research of hepatocellular carcinoma and the Deutsche Forschung Gemeinschaft (DFG). The study in Korea was supported by a grant (NRICH-1107-BO9F) from the National Research Institute of Cultural Heritage, Korea.

Abstract

A rare find of a mummified child from the 16th century AD, in Korea, with relatively preserved organs, enabled a search for ancient hepatitis B virus (aHBV) DNA sequences from laparoscopic-derived liver biopsies. Analysis of the complete aHBV genome (3,215 base pairs) revealed a unique HBV genotype C2 (HBV/C2) sequence commonly spread in Southeast Asia, which probably represents an HBV that infected the Joseon Dynasty population in Korea. Comparison of the aHBV sequences with contemporary HBV/C2 DNA sequences revealed distinctive differences along four open reading frames. Genetic diversity between contemporary and recovered aHBV/C2 DNA may be the result of immunologic, environmental, and/or pharmacologic pressures. The calculated time of most recent common ancestor suggests that the Korean HBV sequence origin dates back at least 3,000 years and possibly as long as 100,000 years. This isolate most likely represents the earliest human HBV sequence that colonized Southeast Asia by human migration. Conclusion: This study describes the complete sequence of the oldest HBV isolate and the most ancient full viral genome known so far. (HEPATOLOGY 2012;56:1671–1680)

Chronic Hepatitis B virus (HBV) infection is one of the most common viral infections worldwide and is an important cause for cirrhosis and hepatocellular carcinoma (HCC), leading to ∼700,000 HCC deaths annually.1-3 Human HBV belongs to the Hepadnavirus family, which includes related agents identified in nonhuman primates, rodents, and birds.4, 5 HBV is a highly infectious, hepatotropic, partially double-stranded DNA virus transmitted by contact with body fluids. The main modes of transmission include vertical transmission from infected mothers to their babies, sexual contact, and intravenous drug abuse. The incubation period varies between several weeks to 6 months, depending on viral load of the inoculum. Duration of infectivity in blood and body fluids after death is unknown, but HBV remains infectious outside the body for at least 7 days and can be found on objects in concentrations of 102-103 virions/mL.6 HBV has a lipid envelope that is susceptible to solvent-detergent treatment.

At least 10 HBV genotypes (A-J) and several subgenotypes with specific geographical distribution have been reported.4, 7, 8 These are defined by an >8% sequence divergence for genotypes and 4%-8% for subgenotypes. These differences seem to correlate with observed human migration patterns.9, 10 Currently, it is unclear whether HBV genotypes and subgenotypes evolved separately of their human host population or coevolved together with their hosts. The observations that individual genotypes are associated with a different outcome of disease and variable response to antiviral therapy generated a number of hypotheses regarding the genetic evolution of HBV.11 One theory suggests a primate origin of HBV with subsequent cross-species transmission from primates to humans in South America, South Saharan Africa, and Southeast Asia.9 These geographical areas harbor specific HBV genotypes (F, E, B, and C, respectively). Another theory suggests a common ancestor of HBV for primates and humans, which gave rise to two unrelated viral lineages,5 suggesting an evolutionary history of HBV that corresponds to the migration of modern humans from Africa approximately 100,000 years ago.

Genotype C, prevalent in East Asia, has 11 subgenotypes correlating with its geographic distribution.4, 12-17 The genetic variability of the 11 subgenotypes reflects the emergence of mutations and divergence of viral genomes in different populations. A number of additional factors may have an effect on the genetic diversity of HBV DNA, including interaction between host and pathogen (i.e., through an immune response or after vaccination), DNA recombination, as well as the effect of antiviral treatment, especially in recent years.12 Analysis of HBV genomes isolated from distinct geographic regions may shed light on the parallel or independent evolution of different HBV species and may be used as a marker indicating the differences among human populations and migrants.1, 18

Korea is endemic for HBV infection and almost exclusively for genotype C (>95%).12 Moreover, the majority of HBV isolates in Korea are classified as genotype C2 (HBV/C2) with a high degree of sequence conservation.19 The findings of one conserved genotype contradicts the observation in many regions worldwide where mixtures of genotypes are present in populations residing in distinct geographic regions.1 The high incidence of HBV/C2 infection unique to Korea may reflect distinct epidemiological and anthropological circumstances. The availability of a tissue sample from an HBV-infected ∼16th century mummified child in Korea enabled a comparison of an ancient HBV (aHBV) DNA isolate with contemporary HBV DNA.

Abbreviations:

aDNA, ancient DNA; aHBV, ancient hepatitis B virus; bp, base pairs; DRs, direct repeats; HBV, hepatitis B virus; HBV/C2, HBV genotype C2; HCC, hepatocellular carcinoma; H&E, hematoxylin and eosin; IRB, institutional review board; ME, minimum evolution; ML, maximum likelihodd; ORFs, open reading frames; PCR, polymerase chain reaction; tMRCA, time of most recent common ancestor; UV, ultraviolet.

Materials and Methods

The entire study was conducted in dedicated ancient DNA (aDNA) laboratories at different campuses of the Hebrew University, in Korea, and in London according to institutional biohazard guidelines (Supporting Materials). All investigators tested negative for HBV infection. The study was approved by the institutional review board (IRB) of the Dankook University Hospital (Cheonan, Korea), and this approval was accepted by the Hadassah-Hebrew University IRB (Jerusalem, Israel).

Histological Studies.

An endoscopic examination of a mummified child from the 16th century AD excavated in Yangju, Korea,20, 21 identified internal organs, including a dry parenchymatous pliable brownish tissue with a nodular surface, located at the upper right abdomen.22 Biopsies taken from the dry parenchymatous organ were immersed in sterile tubes to prevent contamination, were opened only in the aDNA laboratories, and were subjected to microscopic analysis.

aDNA Analysis.

aDNA analysis was performed in two steps, first to determine the presence of HBV DNA and second to amplify the entire genome.

The exploratory phase for HBV DNA detection was conducted independently in three laboratories in Korea, Israel, and the United Kingdom using different methods of DNA extraction amplifying different gene fragments (Supporting Materials). In Korea, two gene regions (the PreC/Core and the preS/S) were amplified, cloned, and sequenced using published primers (Supporting Materials).23, 24 In Israel, amplification of the PreC/Core region (245 base pairs; bp) was performed with an HBV Monitor kit (Roche Diagnostics, Branchburg, NJ). In addition, a short fragment of the PreS/S gene (223 bp) was amplified using two published primer sets.25, 26 An HBV Genotyping kit (INNO-LIPA HBV Genotyping Assay; Roche Diagnostics) was used for amplification of a fragment of the polymerase gene.27 Positive amplifications were directly sequenced and cloned (Supporting Materials). In London, DNA was extracted by a commercial kit (Qiagen GmbH, Hilden, Germany), amplified, and detected a 98-bp fragment from the S gene by commercial Taqman polymerase chain reaction (PCR) assay.

A control specimen of liver tissue obtained from an adult mummy dated to the 17th century AD28 was used in the initial experiments. No HBV sequences were detected using the methodology described above.

To demonstrate the viral hepatotropism, several attempts were carried out to amplify the aHBV from a lung-tissue biopsy taken from the same mummy. DNA was extracted using the same methodology as applied for the liver biopsy. All HBV PCR amplifications from lung tissue were negative. The extracted DNA was used also to characterize the mummy human DNA to determine authenticity and preservation of the extracted aDNA. The nuclear and mitochondrial partial profile indicates an Asian origin of the mummy (Supporting Materials, Supporting Table I1).

Whole Genome Analysis.

Once HBV was identified in the liver biopsies, we proceeded to determine the complete genome sequence. Liver biopsies divided into small equal portions (∼10 mg each) were used for separate extractions of DNA. Before DNA extraction, samples were incubated in 1,000 ppm bleach for 5 minutes, followed by incubation in double-distilled water (15 minutes) to eliminate contemporary contamination during sampling. The entire genome was amplified from the various DNA extracts using 44 overlapping primer sets designed for this study (Supporting Materials, Supporting Table I2). To increase the overlapping of the sequences, we amplified larger fragments, using combined primer sets (Supporting Materials, Supporting Table I2). PCR products were separated on 2.5% agarose gel electrophoresis and visualized under ultraviolet (UV) by staining with ethidium bromide and Vista Green. Positive PCR amplifications were purified, directly sequenced, and/or cloned and sequenced (Supporting Materials).

Sequence Verification and Alignment.

The sequences obtained were initially verified as HBV using the National Center for Biotechnology Information BLAST algorithm (http://www.ncbi.nlm.nih.gov/blast/Blast.cgi). Chromatograms were individually examined to confirm the quality of sequences, using Sequencher 4.9 (Gene Codes Corp., Ann Arbor, MI). The sense and antisense sequences generated from each primer set (direct or cloned sequences) were assembled into a contig in Sequencher 4.9. Each individual contig was visually inspected and verified; any ambiguities were visually resolved. Sequences with poor-quality chromatograms were excluded from the study. A final contig of all sequences was generated using a published reference HBV/C2 (accession no.: AB033553). The complete genome of the studied mummy was determined based on 161 sequences and submitted into GenBank (accession no.: JN315779).

Genetic Analysis.

We investigated the phylogenetic relationships of the obtained ancient HBV genome with other published HBV genotype C sequences from different geographical locations (Supporting Materials, Supporting Table I3). Phylogenetic analysis of the entire genome, as well as for separate genes, was performed with PAUP* 4.0b10,29 applying maximum likelihood (ML) and minimum evolution (ME) methods. Settings for the model of DNA sequence evolution were estimated initially using Modeltest.30 In addition, amino-acid phylogenetic analysis of the entire genome as well as for separate genes, applying ML and ME methods, were performed with MEGA version 5.31 Allele sharing was computed through Network 4.6 software, and results are presented as median-joining network algorithm. Genetic diversity of the mummy HBV genome, compared to other published genomes, was calculated in MEGA version 5.31 Time of most recent common ancestor (tMRCA) was calculated using BEAST software.32

Precautions Against Contamination.

The entire study DNA extraction and amplification procedures were performed under stringent conditions to ensure the authenticity of the aDNA findings.33 The pre-PCR experimental steps (e.g., sampling, DNA extraction, and PCR setting up) were physically separated from the post-PCR analyses. Each pre-PCR step was performed in a different fully equipped UV hood. The PCR mixture, excluding the Taq polymerase, was UV cross-linked to destroy any possible contaminants. Multiple blank extraction and PCR controls were included to ensure the absence of contamination during extraction and amplification procedures. PCR amplification and post-PCR analyses were carried out in a modern DNA laboratory.34 Contemporary, nonancient HVB/C2 DNA was never identified in the Israeli laboratory as far as we know; therefore, no positive control was used during the study in Israel.

Results

Investigators in Korea identified a rare finding of a dry parenchymatous organ in the right upper abdomen of a “naturally” mummified body of a child 4.5-6.6 years in age (Fig. 1).22, 35 The laparoscopically recovered organ later identified as an ancient human liver was dated by 14C testing to the 16th century (mean, 330 ± 70 years ago) during the ruling of the Joseon Dynasty. Although the precise morphology of hepatocytes could not be verified, histologic analysis of a wedge biopsy from the tissue recovered from the right upper abdomen was carried out. Overall, the histological analysis provided indirect evidence that the tissue was of hepatic origin. No hepatic or lymphoid cells could be identified microscopically on hematoxylin and eosin (H&E) staining. However, the overall organization by Masson trichrome and reticulin stains, with minimal fibrotic expansion of portal-tract remnants and condensation of the sinusoidal walls, as might be expected after the lysis of hepatocytes, resembled the gross morphology and organization of liver tissue (Fig. 2).

Figure 1.

Yangju child mummy at the excavation site.

Figure 2.

Macro- and microscopic evaluation of the mummy-derived liver tissue recovered from the right upper abdomen obtained by laparoscopy. (A) Liver surface identified during laparoscopic examination.22 Tissue biopsy with a leather-like consistency was fixed in formaldehyde, then embedded in paraffin and slides. (B) Reticulin stain. (C) Masson's trichrome stain. (D) H&E stain.

Initial attempts to identify HBV DNA present in liver biopsies were carried out in three different laboratories (in Korea, United Kingdom, and Israel), using conventional diagnostic DNA assays. Identification and genotyping of HBV DNA in tissue samples was carried on by direct sequencing of positive amplifications: in Jerusalem, a short amplicon along the preC/C region (104 bp) and the polymerase gene (234 bp); in Korea, a fragment along the core gene (239 bp); and in London, amplification of a 98-bp fragment from the S gene. These results confirmed the presence of HBV DNA sequences (genotype C) in material removed from the mummy.

After the repeated identification of HBV DNA in biopsy samples in all three laboratories, we proceeded to sequencing the entire HBV genome. The first attempt, using next-generation sequencing (SOLiD; at the Center for Genomic Technologies, The Hebrew University of Jerusalem, Jerusalem, Israel) failed to amplify the viral genome, indicating the low-preservation status of the virus. In addition, treatment of extracted DNA with Uracil N-glycosylase, the enzyme that removes deaminated cytosine from DNA, preventing DNA amplification and confirming that the ancient DNA was damaged. The partial genotyping of the mummy genome further supported degradation of the DNA (Supporting Materials, Supporting Table I1).

The entire genome of the virus (3,215 bp) was sequenced from 24 liver-tissue samples. The sequence was obtained from overlapping sequences using 44 primer sets (Supporting Materials, Supporting Table I2). A total of 161 sequences of 150-350-bp long, both sense and antisense, which were found to be of high quality, were used to establish the entire consensuses aHBV DNA genome extracted from the mummified liver (accession no.: JN315779; Fig. 3). Every region of the genome was represented by several overlapping sequences (Supporting Fig. I1). The replicable sequences of each region were highly conserved (99.4% similarity), supporting the authenticity of the results (Supporting Materials, Supporting Table I4). The aHBV genome was classified as genotype C2, based on a comparison with a genotype-matched reference sequence obtained from GenBank (accession no.: AB033553). Four open reading frames (ORFs; polymerase, envelope, precore/core, and hepatitis B virus x) were identified, together with the direct repeats (DRs), enhancer, TATA box, and PolyA regions (Fig. 3).

Figure 3.

Complete consensus sequence of HBV/C2 aDNA recovered from a liver biopsy of a Yangju mummified child. Red: starting codon; green: stop codon; gray: GRE, enhancer; purple: DR, blue (PolyA region); olive: TATA box. Annotation of genes along the consensus sequence (GenBank accession no.: JN315779): S gene: starting codon 155, stop codon 833; X gene: starting codon 1374, stop codon 2450; P gene: starting codon 2307, stop codon 1620; PreS1 gene: starting codon 2848, stop codon 833; PreS2 gene: starting codon 3205, stop codon 833. Published contemporary sequences: AY247030 (Korea): Y18857 (China); and AB033553 (Japan) (Supporting Materials, Supporting Table I3).

The obtained 161 aHBV sequences differed among themselves by four point mutations: two synonymous mutations along the PreS/S genes (C354T; C498T AB033553) and two nonsynonymous mutations along the PreS1 gene (T3149C AB033553; Ser-Pro) and the Polymerase gene (C1138A AB033553; His-Thr). These mutations were determined as possibly representing authenthic substitutions because they were found in more than one sequence among the overlapping sequences, ruling out the possibility of nucleotide misincorporations during PCR resulting from deamination (Supporting Materials). Consequently, four representative sequences of the mummy aHBV DNA were used for further study. The substitution found in the polymerase gene (C1138A AB033553) is not unique to the obtained aHBV, as was previously reported.19 Other published mutations identified in recent years in tissue and serum samples of infected patients were not found in the aHBV DNA sequences (Fig. 3).7, 36, 37

Phylogenetic and allele-sharing analysis of complete sequences indicates that the aHBV DNA sequences from the mummy are unique HBV/C2 sequences (Fig. 4). The aHBV DNA consensus sequences cluster within published HBV/C2 from Korea, Japan, and China and differ significantly from published HBV/C1 sequences obtained from Thailand, Vietnam, and Myanmar (Fig. 4). Among the HBV/C2 sequences, the aHBV consensus sequences clustered together as a separate clade (Fig. 4). These phylogenetic relationships were found also in the phylogenetic analysis conducted for every gene, representing the different ORFs (Supporting Fig. I2A-F). The differences among the HBV/C2 DNA sequences and the recovered aHBV sequences were also expressed at the amino-acid level. The aHBV amino-acid sequence clustered together with other published HBV/C2 genomes and acquired all amino-acid substitutions that distinguish HBV/C2 from HBV/C1.38 In addition, 18 amino acids were found to differentiate the aHBV DNA sequence from 20 HBV/C2 published sequences (Table 1). Among them, 15 were unique to the recovered genome from the mummy and three shared with two published sequences, from China (Y18857) and Japan (AB033553), but differ from all others (Table 1). Recombination events were not detected along the aHBV DNA, similar to a report of other HBV genomes,36 suggesting the establishment of the aHBV DNA from a common ancestor.

Figure 4.

Phylogenetic placement and historical context for aDNA HBV sequences and relationship to contemporary HBV sequences. (A) Phylogenetic relationships inferred for aHBV sequences with various contemporary published HBV full genomes (Supporting Materials, Supporting Table I3). Apart from the C genotype sequences, four other genotypes were chosen to be included in the analysis as an outgroup. Phylogenetic analysis was conducted using partitioned ML with the time-reversal generation with gamma and invariant model. Most nodes received strong bootstrap support. Scale bar denotes 0.001 substitutions per site. Korean mummy sequences are marked as aHBV. (B) Median network of aHBV and modern HBV based on 1,063 variant positions in modern genomes. Colored circles represent different clades as defined in phylogenetic tree. Colored circles represent the various sequences: red = aHBV; yellow = Korea; blue = China; brown = Japan; orange = Thailand; pink = Myanmar; olive = Vietnam; and gray = outgroups.

Table 1. Amino-Acid Comparison Between the aHVB DNA Sequences and Contemporary Published HBV/C2 DNA Sequences From Korea, Japan, and China
GeneAmino-Acid PositionaHBV DNA (Mummy)Published HBV/C2Comments
  1. Published HBV/C2: representative contemporary published sequences from Korea, China, and Japan, as indicated in Supporting Materials Table SI3.

C66LM 
PreC95ID 
 205PQ 
Polymerase154GW 
 159RL 
 272RS 
 757QR 
 830VD 
PreS1101LS 
 218EG 
 334KRExcept Y18857, AB033553
PreS299EG 
 215KR 
S44EG 
 160KRExcept Y18857, AB033553
 184AVExcept Y18857, AB033553
X130KM 
 131VI 

Overall, nucleotide and amino-acid analyses support a high similarity of the recovered aHBV DNA sequences to published HBV/C2 DNA from Korea, China, and Japan (Fig. 4; Table 1; Supporting Fig. I2). Among these genomes, the highest similarity of the aHBV sequence was found with the contemporary sequences from Korea (97% similarity). It should be remembered that our findings and characterization of the aHBV DNA were compared to contemporary reported sequences from South Korea, which may not necessarily be representative for North Korea.13, 19

Calculation of the divergence time between the studied aHBV DNA and other modern HBV DNA C2 sequences suggests that the mummy-derived aHBV sequence evolved at least 3,000 years ago and, most probably, earlier, representing one of the ancestral human HBV sequences (Table 2). This assumption is based on published estimations of HBV mutation rates per site per year ranging between <2*10−4 and 4.5*10−5, but not exceeding 10−6.10, 39, 40 Regardless of the calculation method employed, our findings are in line with previous estimates that human HBV DNA existed in East Asia ∼3,000 years ago or more.11, 39 In addition, tMRCA was calculated under the relaxed molecular clocks model. Results suggest that the aHBV DNA sequences from the mummy are clustered together, different from all other genotype C2 sequences with a posterior probability, and tMRCA estimates are similar to the outgroup (Supporting Materials; Supporting Fig. I3).

Table 2. Estimation of Divergent Time Between the aHBV DNA Sequence and Contemporary Published Sequences Representing the HBV/C2 Subgenotype From East Asia
GeneNo. of bpNo. Amino AcidsGroupNo. of Nucleotide VariationsNo. of Amino Acid VariationsDivergent Time*Divergent TimeAverage Divergent Time
  • *

    4.5*10−5.38

  • 2.1*105.41

  • Published HBV/C2 - representative contemporary published sequences from Korea, China and Japan indicated in Supporting Material Table SI3.

X465155All C278403,993.867,987.715,990.78
 465155Korea35191,792.113,584.232,688.17
 465155Japan31181,587.303,174.602,380.95
 465155China45252,304.154,608.293,456.22
Polymerase2,536845All C23791463,558.287,116.575,337.43
 2,536845Korea137541,286.242,572.481,929.36
 2,536845Japan215792,018.554,037.103,027.83
 2,536845China199781,868.333,736.672,802.50
PreS11,203401All C2143762,830.235,660.454,245.34
 1,203401Korea4527890.631,781.261,335.95
 1,203401Japan75381,484.382,968.772,226.58
 1,203401China76391,504.183,008.352,256.26
PreS2846282All C294522,645.505,291.013,968.25
 846282Korea3221900.601,801.191,350.89
 846282Japan45231,266.462,532.931,899.70
 846282China49271,379.042,758.082,068.56
S680226All C267382,345.944,691.883,518.91
 680226Korea2014700.281,400.561,050.42
 680226Japan34191,190.482,380.951,785.71
 680226China39241,365.552,731.092,048.32
PreC646215All C293643,427.696,855.375,141.53
 646215Korea43311,584.843,169.692,377.27
 646215Japan41331,511.133,022.262,266.70
 646215China45341,658.563,317.122,487.84
C558186All C286263,669.577,339.145,504.35
 558186Korea40121,706.783,413.552,560.16
 558186Japan37141,578.773,157.542,368.15
 558186China43111,834.783,669.572,752.18
Complete3,2221,074All C2     
 3,2221,074Korea190881,404.042,808.082,106.06
 3,2221,074Japan2821202,083.894,167.783,125.83
 3,2221,074China2961222,187.344,374.693,281.01
Average divergent time (all genes, all countries) 2,315.234,630.473,472.85

Discussion

Our results indicate the presence of aHBV DNA in the dry parenchymatous tissue recovered from the upper right abdomen of a mummified child from the 16th century AD who died from an unknown cause. Using overlapping primer sets, we amplified the entire genome and identified the four ORFs of HBV. The clustering of the aHBV DNA consensus sequences within the HBV/C2 clade, together with sequences from China and Japan and not only from Korea, the close similarity of the amino-acid sequences to two published HBV/C2 genomes from China and Japan and the lack of recombination events suggest that the origin of the common ancestor was in China and/or Japan.

Studies of contemporary HBV/C2 DNA in China and Japan suggest a geographical distribution of the same strain and possible transmission between past closely related communities.39 Similar distribution patterns were detected in the study of BK polyomavirus.41 Moreover, recent whole genome autosomal single-nucleotide polymorphism analysis study provided evidence for human migration between South Korea, Japan, and China during settlement in East Asia, supporting the historical knowledge of human migration to the region in earlier and later periods.42 These studies, and the finding that the obtained mummy hypervariable region I sequence and short tandem repeat partial profile matches human haplogroup D of an Asian origin probably from Japan (Supporting Materials), further supporting the thesis that the mummy-derived aHBV genome originated from Chinese/Japanese communities that migrated to Korea.

Our results should be discussed also in the context of the various hypotheses regarding the origin and evolution of HBV.11 Different theories date the origin of HBV over a period between millions of years9 to approximately 3,000 years.4, 39 All these hypotheses are mainly based on the number of nucleic-acid and amino-acid substitutions observed in different isolates of HBV, recovered over a short period of only a few decades from humans and nonhuman primates. In the present analysis, the calculation of the divergence time between the studied aHBV DNA and other modern HBV/C2 DNA sequences suggests that the mummy-derived aHBV evolved at least 3,000 years ago and most probably earlier, representing one of the ancestral human HBV sequences. Further support for this assumption may be derived from tMRCA analysis using relaxed molecular clock models. The mummy-derived viral aDNA was found to be an ancestral sequence similar to the outgroup. Based on published estimates, the outgroup genotypes (A, B, E, F, and C1) evolved ∼100,000 years ago.5 In contrast, Gunther et al. and Orito et al. suggested that HBV genotypes evolved later and independent of human migration.10, 24 In this context, it should be emphasized that the mummy-derived HBV-DNA is only 400-500 years old, most likely representing an ancient HBV strain approximately 3,000-100,000 years old, possibly in line with other estimates.10, 11, 24

The retrieval and evaluation of aDNA sequences from archeological specimens is associated with significant difficulties. The use of PCR assists in restoration of traces of DNA. However, aDNA is frequently severed into fragments of <200 bp because of autolytic or diagenic processes as well as deanimation of the nucleotides. Deaminated cytosine will result in nucleotide misincorporation (C<>T and G<>A), leading to false interpretation of sequences. Consequently, direct sequencing (and not cloning) of several PCR attempts, using different extracted DNA, has been used to minimize the effect of misincorporation. Another pitfall of aDNA research is the high risk of contamination with contemporary DNA. The authenticity and absence of contamination of the recovered aHBV DNA is supported by the fact that tissue samples were tested in three independent laboratories. Furthermore, HBV/C2 was never detected in the Israeli laboratory nor ever studied before the present research. The biopsies were treated with bleach before extraction of DNA, and all researches tested negative for hepatitis B surface antigen. The identified aHBV DNA genome contains all ORFs and lacking mutations that are causing termination of the coding protein. Although the aHBV DNA is identified as subgroup C2, nucleic-acid differences were found in comparison with contemporary HBV/C genomes representing different geographical regions in Southeast Asia.38

In spite of the major progress made in the field of aDNA, only one complete ancient viral genome (the 1918 Spanish influenza virus) has been described so far.43 In addition, small fragments of human papilloma virus, human T-cell lymphotropic virus type 1, and a fragment of an Anelloviridae were reported.44-46 To the best of our knowledge, the complete aHBV DNA genome reported herewith is the, so far, oldest described ancient episomal viral genome. The fact that we were able to sequence the entire HBV genome, including the single-stranded portion of the virus, suggests that the isolated aHVB DNA was derived from intranuclear covalently closed circular HBV DNA residing originally in the mummy hepatocytes.47

In summary, this report describes the first full viral genome of HBV extracted from an ancient human specimen. The complete sequence obtained probably represents a wild-type HBV pathogen without precore mutations, which infected the population during the Joseon Dynasty in Korea approximately 400-500 years ago. Based on our analysis, the virus most probably originated in China and/or Japan and spread to Korea. The observed genetic diversity of the recovered aHBV DNA, as compared to contemporary isolates, is most likely a result of a natural evolutionary process and not an iatrogenic pharmacologic pressure, as observed in patients treated with modern antiviral agents.1, 13 Additional analysis of aHBV genomes may shed further light on the evolutionary pathway of HBV/C in East Asia.

Acknowledgements

The authors thank Prof. C. Greenblatt for his support, Prof. H. Norder and Dr. B. Kessing for their advice and comments, Dr. S.S. Park, Director of Nansa Traditional Costume Heritage Institute, Seokjuseon Memorial Museum for providing the mummy's photo, L. Hadas and D. Hermon for their assistance in laboratory work.

Ancillary