Hepatitis B virus genotype D can be found in many parts of the world and is the most prevalent strain in south-eastern Europe, the Mediterranean Basin, the Middle East, and the Indian sub-continent. The epidemiological history of the D genotype and its subgenotypes is still obscure because of the scarcity of appropriate studies. We retrieved from public databases a total of 312 gene P sequences of HBV genotype D isolated in various countries throughout the world, and reconstructed the spatio-temporal evolutionary dynamics of the HBV-D epidemic using a Bayesian framework.
The phylogeographical analysis showed that India had the highest posterior probability of being the location of the tree root, whereas central Asia was the most probable location of the common ancestor of subgenotypes D1–D3. HBV-D5 (identified in native Indian populations) diverged from the tree root earlier than D1–D3. The time of the most recent common ancestor (tMRCA) of the tree root was 128 years ago, which suggests that the common ancestor of the currently circulating subgenotypes existed in the second half of the XIX century. The mean tMRCA of subgenotypes D1–D3 was between the 1940s and the 1950–60s. On the basis of our phylogeographic reconstruction, it seems that HBV-D reached the Mediterranean area in the middle of the XX century by means of at least two routes: the first pathway (mainly due to the spread of subgenotype D1) crossing the Middle East and reaching north Africa and the eastern Mediterranean, and the second pathway (closely associated with D2) that crossed the former Soviet Union and reached eastern Europe and the Mediterranean through Albania. We hypothesise that the main route of dispersion of genotype D was the unsafe use of injections and drug addiction.
Citation: Zehender G, Ebranati E, Gabanelli E, Shkjezi R, Lai A, Sorrentino C, et al. (2012) Spatial and Temporal Dynamics of Hepatitis B Virus D Genotype in Europe and the Mediterranean Basin. PLoS ONE 7(5): e37198. https://doi.org/10.1371/journal.pone.0037198
Editor: Marco Salemi, University of Florida, United States of America
Received: March 23, 2012; Accepted: April 16, 2012; Published: May 25, 2012
Copyright: © 2012 Zehender et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Hepatitis B virus (HBV) is an enveloped DNA virus belonging to the Hepadnaviridae family: its genome is a small, circular partially double-stranded DNA molecule of about 3.7 kilobases including four open reading frames (S, C, P and X) coding for seven proteins (three surface proteins, two core antigens, the polymerase and the X protein).
HBV is the leading cause of liver disease and infects an estimated 240 million people worldwide . Despite the recent decrease in the rate of new cases, about 7–8,000 new diagnoses are made every year in Europe . The prevalence of HBV infection in Europe varies widely: in general, it is higher in south-eastern than north-western countries . The highest prevalence rates are in Turkey, Romania, Bulgaria, Greece, Albania and southern Italy , .
HBV is characterised by a high degree of genetic heterogeneity due to the use of a reverse transcriptase during viral replication. Eight main genotypes (A–H) that differ genetically by at least 8% have so far been identified , some of which further segregate into subgenotypes with a mean genetic distance of about 4% . The genotypes and subgenotypes have a distinct ethno-geographical distribution. The ubiquitous genotypes A and D are respectively the most prevalent in north-western and south-eastern Europe/Mediterranean countries . Seven main HBV-D subgenotypes have so far been described (D1–D7). Considering only Europe and the Mediterranean area, D1 is the most prevalent subgenotype in Greece, Turkey, and north Africa –, D2 the most prevalent in north-eastern Europe (Russia, Belarus, Estonia) and Albania , , and D3 the most prevalent in Italy and Serbia –. Some specific D strains are restricted to more limited geographical areas: D7 in Tunisia and Morocco , , D6 in Papua and Indonesia , and D5 in India, where all of the main D subgenotypes are also found . Finally, D4 is the dominant subgenotype in Oceania .
The epidemiological history of the HBV-D genotype is still unclear because of the scarcity of appropriate studies. The aim of this study was to use a recently developed Bayesian approach to virus phylogeography analysis  in order to reconstruct the spatial and temporal dynamics of HBV-D genotype, with particular reference to the origin and the geographic dispersion of the principal D subgenotypes circulating in south-eastern Europe, the Middle East and the Mediterranean basin.
HBV genotype D data set
We retrieved from public databases (Genbank at http://www.ncbi.nlm.nih.gov/genbank/) a total of 312 P gene sequences of HBV-D isolated in various countries of the world, with particular attention being given to eastern Europe, the eastern Mediterranean and the Middle East (including 63 Italian and 59 Albanian isolates previously characterised by us) , . The viral sequences were selected on the basis of the following inclusion criteria: 1) they had been published in peer-reviewed journals; 2) there was no uncertainty about the sub-type assignment of each sequence and all were classified as non-recombinant; and 3) the city/state of origin were known and clearly stated in the original publication. The sequences were aligned using ClustalX software  and cropped to the same length of 990 nucleotides coding for a region of the P protein between amino acids 305 and 634, and the overlapping S, PreS2 and PreS1 regions . They were manually edited using Bioedit software (Tom Hall, 2011; available at http://www.mbio.ncsu.edu/bioedit/bioedit.html). The sampling locations were Albania (n = 59), central Asia (n = 10), Egypt/Israel (n = 5), the Far East (n = 8), Greece (n = 4), India (n = 26), Iran (n = 13), Italy (n = 63), Russia (n = 53), Serbia (n = 3), South Africa (n = 6), Spain (n = 4), Tunisia (n = 41), and Turkey (n = 17).
Likelihood mapping analysis
In order to obtain an overall impression of the phylogenetic signals in the HBV P sequences, we used likelihood mapping to analse 10,000 random quartets generated using TreePuzzle , . A likelihood map consists of an equilateral triangle containing dots representing the likelihoods of the three possible unrooted trees for a set of four sequences (quartets) randomly selected from the data set: the dots near the corners or sides respectively represent tree-like (fully resolved phylogenies in which one tree is clearly better than the others) or network-like phylogenetic signals (three regions in which it is not possible to decide between two topologies). The central area of the map represents a star-like signal (the region in which the star tree is optimal).
Estimated evolutionary rates and time-scaled phylogeny reconstruction
To estimate the evolutionary rates of the HBV P gene sequences, we used molecular clock models based on the isolation dates of the tips of a tree. As sampling dates were not available for all of the sequences included in the analysis, we used an external calibration approach previously adopted by other authors , . The evolutionary rates were estimated on a smaller data set including only isolates for which the sampling years were known. The estimated mean evolutionary rate was used to define a priors probability distribution during the analysis of the main data set. The smaller data-set included 216 isolates from the same places as those of the main data set and the sampling dates ranged from 1980 to 2007. The evolutionary model that best fitted the data was selected using an information criterion implemented in JModelTest , which is freely available at http://darwin.uvigo.es/software/jmodeltest.html.
The evolutionary rates were estimated using a Bayesian Markov Chain Monte Carlo (MCMC) method implemented in BEAST 1.5.4  under a strict and relaxed molecular clock, and an uncorrelated log normal rate distribution model. As coalescent priors, we compared three simple parametric demographic models of population growth (constant size, and exponential and logistic growth) and a piecewise-constant model, called Bayesian skyline plot (BSP). The same Bayesian MCMC method, with the same substitution, molecular clock and demographic models were used to reconstruct the time-scaled phylogeny of the main data set, and statistical support for specific clades was obtained by calculating the posterior probability of each monophyletic clade.
The MCMC chains were run for at least 50 million generations, and sampled every 5000 steps. Convergence was assessed on the basis of the effective sampling size (ESS) after a 10% burn-in using Tracer software version 1.5 ( http://tree.bio.ed.ac.uk/software/tracer/ ). Only ESS's of >250 were accepted. Uncertainty in the estimates was indicated by 95% highest posterior density (95% HPD) intervals, and the best fitting models were selected by means of a Bayes factor (BF, using marginal likelihoods) implemented in BEAST . In accordance with Kass and Raftery , the strength of the evidence against H0 was evaluated as 2lnBF<2 = no evidence; 2–6 weak evidence; 6–10 strong evidence; and >10 very strong evidence. A negative 2LnBF indicated evidence in favour of H0. Only values of ≥10 were considered significant.
The spatial reconstruction was obtained by means of the same Bayesian framework using a continuous time Markov Chain (CTMC) implemented in BEAST  over discrete sampling locations. The Bayesian Stochastic Search Variable Selection (BSSVS) model was implemented, which allows the diffusion rates to be zero with a positive prior probability. Comparison of the posterior and prior probabilities of individual rates being zero provided a formal BF to test the significance of the linkages between locations. Rates yielding a BF of >6 were considered well supported and formed the migration pathway. The maximum clade credibility (MCC) tree (i.e. the tree with the largest product of posterior clade probabilities) was selected from the posterior tree distribution using the program TreeAnnotator (included in the Beast package) after a 10% burn in. The final trees were manipulated in FigTree v.1.3 for display purposes. The significant migration rates were analysed and visualised by SPREAD  a recently developed application available at http://www.kuleuven.be/aidslab/phylogeography/SPREAD.html.
Likelihood mapping analysis of the data set
The phylogenetic noise of the data set was investigated by means of likelihood mapping. The evaluation of 10,000 random quartets showed that more than 92.7% of the randomly chosen quartets fell in the corners, and only 2.7% in the central area of the likelihood map, thus indicating that the alignment contained sufficient phylogenetic information (Figure S1).
Estimated evolution rates
The mean evolutionary rate of the HBV-D P sequences was evaluated on a data-set including 216 isolates with known sampling dates.
Comparison of the strict and relaxed clock models using the BF test showed that the relaxed clock fitted the data significantly better than the strict clock (2lnBF = 316.4 in favour of the relaxed clock). Comparison of the coalescent priors showed that the BSP model was better than the constant (2lnBF = 222.27) or exponential growth model (2lnBF = 140.9).
Under the relaxed clock model, the estimated mean value of the 990-nt fragment of the P gene was 4.4×10−4 subs/site/year (95%HPD 2.6×10−4–6.2×10−4), and this external rate estimate was used for the subsequent analyses.
Figure 1 shows the maximum credibility tree of the main data set, using the previously estimated mean evolutionary rate of the HBV P sequences as a prior probability distribution. The figure shows the most probable location of each branch in different colours, and the tMRCA estimates under relaxed molecular clock. Analysis of the tree showed that the 312 HBV genotype D isolates segregated into a number of significant clades corresponding to the main subgenotypes D present in the geographic area considered (D1, D2, D3, D5 and D7). In particular, HBV D5 and D7 formed two highly significant clades (pp = 1 each one) that diverged from the tree root earlier than the other subgenotypes. The isolates belonging to the three major subgenotypes (D1–D3) formed three clades (with pp values of respectively 0.73 and 1) sharing two highly significant ancestors: the first common to all three subgenotypes (pp = 0.89) and the second shared only by D1 and D2 (pp = 0.88).
The branches are coloured on the basis of the most probable location state of the descendent nodes (see colour codes in upper left inset). The numbers on the internal nodes represent posterior probabilities, and the scale at the bottom of the tree represents the years before the last sampling time (2007). The clades corresponding to the main HBV-D subgenotypes (D1, D2, D3, D5, D7) are highlighted.
Phylogeographical analysis showed that India had the highest posterior probability to be the location of the tree root (pp = 0.30), significantly higher than that of the second most probable location, the Central Asia (pp = 0.22) (see histogram in Figure 1 for more details), which was the most probable location of the deepest nodes of the tree: i.e. the common ancestors of D1–D3 and D1–D2 (pp values of respectively 0.33 and 0.43).
Two clades included isolates from homogenous geographical areas: clade D5 encompassed only Indian isolates, and D7 isolates from Tunisia (Table 1). As all of the other clades included isolates from different parts of the world, we analysed each subgenotype separately.
The strains in clade D1 were isolated in India, Central Asia (Kazakhstan and Uzbekistan), the Far East (China and Vietnam), Tunisia, Iran, Egypt, north-eastern Europe, and Italy, and formed only small groups of no more than five isolates. Turkey had the greatest posterior probability (pp = 0.40) of being the location of the D1 MRCA, followed by Central Asia (pp = 0.30) and Iran (pp = 0.22) (Figure S2A). There was no significant geographical structure.
The clade corresponding to the D2 subgenotype (Figure S2B) included a large prevalence of isolates from Russia, Belarus and Estonia, a large group of Albanian isolates, a number of Italian strains, and sporadic isolates from Greece and Turkey. Central Asia and Russia (state pp = 0.33 and 0.30 respectively) shared the vast majority of posterior distribution with a 0.63 cumulative probability of being the location of the MRCA (followed by Albania: pp = 0.22). All of the D2 isolates tended to segregate significantly on the basis of their geographical origin. In particular, there were two highly significant sub-clades: sub-clade D2a (pp = 0.99) included all of the Albanian and several Italian strains, indicating Albania as the most probable MRCA location (pp = 0.96); the D2b sub-clade (pp = 0.81), which included all of the north-east European and a few Italian strains, had Russia as its most probable MRCA location (pp = 0.96).
The D3 clade included a number of Italian isolates, several Albanian and north-east European strains, three isolates from Serbia, four from India and one from Mongolia; a group of six South African isolates significantly segregated from the other strains. One single Indian isolate (Ne20D3) was at the outgroup (Figure S2C). Central Asia and India shared the highest posterior probability of being the location of the D3 MRCA (cumulative pp = 0.54, 0.27 for both). All of the D3 isolates tended to segregate significantly on the basis of their geographical origin.
Using a BF test under BSSVS, we identified the main linkages between the different locations. Figure 2 (see Table S1 for more details) shows all of the rates with a non-zero expectancy and a BF of ≥6. Interestingly, among the most significant rates, India linked only with central Asia (BF = 27.3), whereas central Asia was related to five locations (Russia, the Far East, Turkey, Iran and India), thus suggesting it played a central role in the global dispersion of HBV-D. Other locations with several significant linkages were Russia (with Italy, Spain, and central Asia) and Turkey (with central Asia, Iran and Egypt/Israel), and Italy (with Russia, Albania and South Africa).
Estimated times of the most recent common ancestors (tMRCA)
Using the relaxed molecular clock model described in Methods, we estimated the tMRCA of all of the internal nodes of the Bayesian tree. In particular, the estimated mean tMRCA of the tree root was 128 years ago (95%HPD 66–210 years), thus suggesting that the origin of the D subgenotypes goes back to 1880 (credibility interval 1798–1942). HBV-D5 and D7 were the first subgenotypes to diverge from the root, although their radiation is relatively recent (40 and 52 years ago, corresponding to 1968 and 1956; see Table 1). On the basis of our estimates, the ancestor shared by D1–D3 existed a mean 86 years ago (about 1922) and that shared by D1 and D2 68 years ago (1940). Once again, the time of subgenotype radiation was more recent: 65 years ago (95%HPD 37–101) for HBV-D3 (1943), and 52 years ago (1956) for D2 and 48 years ago (1960) for D1. On the basis of the tMRCA estimates, the two D2 sub-clades (A and B) originated in 1963 (95%HPD 1943–1981) and 1965 (95%HPD 1945–1981) respectively.
HBV genotype D is found throughout the world: it is the most prevalent genotype in north-eastern Europe, the Mediterranean basin, northern Africa, and the Middle East; it is highly prevalent in the Indian sub-continent and a group of island in the Indian Ocean with high endemic levels of HBV (Nicobare and Andaman) ; and it has also been identified in Oceania .
It is characterised by a high degree of heterogeneity. Using a relaxed molecular clock model, we estimated the evolutionary rates of the P sequences in a sub-set of dated isolates as ranging from 2.6 to 6.2×10−4 subs/site/year. Previous studies have suggested evolutionary rates varying between 1.5×10−5 and 7.72×10−4 subs/site/year –. Recently, we showed that the average substitution rate of genotype D was higher than that of the other genotypes (particularly genotype A) and suggested that these differences were mainly due to population-related factors such as the wider spread of the HBV-D epidemic and the different main routes of transmission of the two genotypes . In the last year, Harrison et al.  have reported slower viral evolutionary rates during hepatitis B e antigen positive state, compared to hepatitis B e antigen negative state. It is well known that HBV genotypes influence the HBeAg state, as a number of mutations that reduce HBeAg expression (such as those affecting nucleotide 1896 of the Pre-core region) occur more frequently in genotype D, than among others . This may further explain the higher evolutionary rates of HBV-D.
On the basis of the evolutionary rate estimate obtained, we inferred the tMRCA of every internal node within the Bayesian tree of the entire data set. Interestingly, our time-scaled reconstruction supported a relatively recent history of the currently circulating HBV-D genotype with the time of the tree root being 128 years ago, in line with previous estimates , . This estimate suggests that the common ancestor of the D genotype existed in the second half of the XIX century.
The origin of D genotype and subgenotypes and their spread throughout the world is still obscure. Ours is the first attempt to reconstruct the epidemiological history of HBV-D genotype phylodynamically and phylogeographically. Our reconstruction shows that the most probable location of the tree root is India, which suggests that the Indian sub-continent was the place in which HBV-D originated. This is supported by the fact that genotype D is highly prevalent in India, which hosts at least four HBV-D subgenotypes with very different local distributions: D1 is the most prevalent in northern India, D2 in western India, D3 is as prevalent as D2 in eastern India , and the recently reported D5 (the only subgenotype present in a primitive tribe in South-East India) . As shown by our analysis and previous reports , D5 was probably the first subgenotype to diverge from the root of the phylogenetic tree.
As we were interested in studying the HBV-D subgenotypes mainly circulating in Europe and the Mediterranean area, no D4 isolates were included in our analysis. One recent study has confirmed that D5 was the first subgenotype emerging from the tree , and found that D4 formed a significant group with D7, a finding that is supported by a preliminary review of our own data which include a few D4 sequences (Figure S3). These observations and the very high degree of HBV-D genetic divergence suggest that India was the cradle of the HBV-D genotype.
HBV D1–D3 share more recent ancestors which, on the basis of our phylogeographical reconstruction, were most probably located in Central Asia. Our tMRCA estimate suggests that the D1–D3 MRCA existed in the first decades of the XX century (1922), with D3 being the first subgenotype to diverge in the early 1943 while the radiation of subgenotypes D1 and D2 probably occurred between the mid- 1950s and early 60s. Only a few data are available concerning HBV genotype prevalence in central Asia. HBV-D is the most prevalent genotype in Uzbekistan, Tajikistan and Kazakhstan –, and one of these studies  found that D1 was the most prevalent subgenotype. Much more data support the greater prevalence of D1 than the other subgenotypes in Iran , , Turkey , Pakistan ,  and Afghanistan . In line with these data, Turkey had the highest posterior probability of being the location of origin of D1, followed by Central Asia and/or Iran. The lack of any significant geographical segregation of D1 strains suggests the relatively recent penetration of HBV-D1 in this area as a result of multiple introduction events, probably due to frequent exchanges with neighbouring countries.
HBV D2 (the most prevalent subgenotype in north-eastern Europe) has two distinct clades, one originating in Russia and the other in Albania , , , , . In our reconstruction, the D2 clade as a whole probably originated somewhere between central Asia and Russia as they had similar posterior probabilities.
The phylogeography of the D3 subgenotype remains elusive although, once again, central Asia and India seem to be the most likely locations of origin. The reason for this uncertainty was the scarcity of dated HBV D3 sequences collected in places other than Italy, which represented the majority of the available strains. It has been shown that subgenotype D3 (and particular strains carrying specific mutations such as the T125M in the a determinant) is prevalent among IVDUs in various countries, including Italy, Denmark, the USA, British Columbia, Belgium and Serbia , , –. These observations, partially strengthened by the high prevalence of this subgenotype in India or Pakistan , suggest that the international drug trade may have played a role in its spread, at least in recent times. Our recent study suggested the presence of D3 in Italy even before the epidemic of intravenous drug addiction in the early 1970s, rising questions about other possible routes of geographic dispersion of this strain . Nevertheless our analysis suggests that D3 was the earlier subgenotype diverging in 1940s.
Combining our temporal and spatial reconstructions of the evolutionary dynamics of HBV-D, our data suggest that, by the end of the XIX century, its ancestor left India and reached central Asia, where subgenotypes D1–D3 diverged. Central Asia was the source of their further spread to Europe and the Mediterranean area in the first decades of the XX century by means of at least two routes: a south-western route (mainly due to the diffusion of subgenotype D1) crossing the Middle East (where most probably D1 diverged) and reaching north Africa and the south-eastern Mediterranean, and a second route (closely associated with D2) towards the north-west, which crossed the former Soviet Union and reached eastern Europe and the Mediterranean through Albania . A third possible epidemic of HBV in Eurasia is represented by the diffusion of D3, which still remains obscure.
The spread of the main European HBV subgenotypes dates to a period of time including the First and especially the Second World War. The importance of war in the history of hepatitis viruses has been efficiently explained by Reuben . Times of war provide many opportunities for infection, such as massive vaccines administration to soldiers, wounds, blood transfusions and other violence-related factors. The largest recorded outbreak of viral hepatitis occurred in 1942, when more than 28,000 US soldiers developed jaundice (and 62 died) after being inoculated by an HBV-contaminated yellow fever vaccine , . Subsequent studies of veterans have shown that 97% of the subjects who received this vaccine were positive for serum markers of HBV infection . We can hypothesise that this or similar events, such as the outbreaks of jaundice following the intravenous injection of arsphenamine (the main anti-syphilis treatment from the early 1920s to the late 1940s) led to the spread of some HBV D subgenotypes in Europe , . The spread of D genotype could have been subsequently amplified locally by other routes.
In particular, the rate of vertical transmission is relatively low in populations in which genotype D is predominant probably because of the high rate of mutations causing HBeAg negativity in this genotype. Consequently, the majority of infections are acquired horizontally within these populations , mainly as a result of the use of unsterilised needles and syringes, and household contacts , . In line with our spatio-temporal reconstruction of HBV-D epidemics, the unsafe use of injections in medical practice was still a problem in India and the Middle East in the year 2000, when about 70% of the equipment was reused , and about 57% of HBV infections in India were attributable to this route of transmission , . Unsafe injections have also been indicated as a possible cause of the spread of HBV in east European countries such as Moldavia  and Romania .
One further event that probably contributed to the rapid diffusion of HBV-D is the intravenous drug use epidemic that has affected Europe since the early 1970s. A recent literature review has pointed out that the greatest prevalence of HBsAg among IVDU populations is in the eastern Mediterranean and eastern Europe , where the most prevalent genotype is HBV-D.
In conclusion, HBV-D is highly prevalent in eastern and southern Europe, where HBV is highly endemic and the main transmission route is unsafe injections. On the contrary, genotype A is highly prevalent in central and northern Europe, where HBV is mainly sustained by sexual transmission , , . This observation recalls the epidemiological dichotomy previously described by us in Italy , and suggests that similar population-related mechanisms may have played a role in the different distribution and evolutionary dynamics of the two main HBV genotypes throughout Europe.
Likelihood map of the 312 HBV-D P gene sequences. Each dot represents the likelihoods of the three possible unrooted trees per quartet randomly selected from the data set: the dots near the corners or sides respectively represent tree-like (fully resolved phylogenies in which one tree is clearly better than the others) or network-like phylogenetic signals (three regions in which it is not possible to decide between two topologies). The central area of the map represents a star-like signal (the region in which the star tree is optimal tree). The numbers indicate the percentage of dots in the centre of the triangle.
Part of the MCC tree shown in Figure 1 focusing on the D1 (a), D2 (b) and D3 (c) clades. The branches are coloured on the basis of the most probable location state of the descendent nodes (see colour codes in upper left inset). The numbers on the internal nodes represent posterior probabilities, and the scale at the bottom of the tree represents the years before the last sampling time (2007). Subclades D2a and D2b are highlighted (panel b).
The maximum clade credibility (MCC) tree, including also three D4 isolates. The branches are coloured on the basis of the most probable location state of the descendent nodes (see colour codes in upper left inset). The numbers on the internal nodes represent posterior probabilities, and the scale at the bottom of the tree represents the years before the last sampling time (2007).
Conceived and designed the experiments: GZ EE AL MC. Performed the experiments: EG RS. Analyzed the data: GZ EG RS CS ALP. Contributed reagents/materials/analysis tools: GZ RB ET SB MB. Wrote the paper: GZ EE AL MC MG.
- 1. Ott JJ, Stevens GA, Groeger J, Wiersma ST (2012) Global epidemiology of hepatitis B virus infection: New estimates of age-specific HBsAg seroprevalence and endemicity. Vaccine. 30(12): 2212–9.
- 2. European Centre for Disease prevention and Control (2010) Hepatitis B and C in the EU neighbourhood: prevalence, burden of disese and screening policies. Stockholm. ECDC.
- 3. Da Villa G, Romano L, Sepe A, Iorio R, Paribello N, et al. (2007) Impact of hepatitis B vaccination in a highly endemic area of south Italy and long-term duration of anti-HBs antibody in two cohorts of vaccinated individuals. Vaccine 25: 3133–3136.
- 4. Rantala M, van de Laar MJ (2008) Surveillance and epidemiology of hepatitis B and C in Europe-a review. (21).Euro Surveill 13.
- 5. Schaefer S (2007) Hepatitis B virus genotypes in Europe. Hepatol Res 37: S20–26.
- 6. Schaefer S (2007) Hepatitis B virus taxonomy and hepatitis B virus genotypes. World J Gastroenterol 13: 14–21.
- 7. Bozdayi G, Turkyilmaz AR, Idilman R, Karatayli E, Rota S, et al. (2005) Complete genome sequence and phylogenetic analysis of hepatitis B virus isolated from Turkish patients with chronic HBV infection. J Med Virol 76: 476–481.
- 8. Cox LE, Arslan O, Allain JP (2011) Characterization of hepatitis B virus in Turkish blood donors, and the prevalence of the SP1 splice variant. J Med Virol 83: 1321–1325.
- 9. Fylaktou A, Papaventsis D, Daoudaki M, Moskophidis M, Reiberger T, et al. (2011) Molecular epidemiology of chronic hepatitis B virus infection in Greece. J Med Virol 83: 245–252.
- 10. Garmiri P, Rezvan H, Abolghasemi H, Allain JP (2011) Full genome characterization of hepatitis B virus strains from blood donors in Iran. J Med Virol 83: 948–952.
- 11. Tallo T, Tefanova V, Priimagi L, Schmidt J, Katargina O, et al. (2008) D2: major subgenotype of hepatitis B virus in Russia and the Baltic region. J Gen Virol 89: 1829–1839.
- 12. Zehender G, Shkjezi R, Ebranati E, Gabanelli E, Abazaj Z, et al. (2011) Reconstruction of the epidemic history of hepatitis B virus genotype D in Albania. Infect Genet Evol 12(2): 291–8.
- 13. De Maddalena C, Giambelli C, Tanzi E, Colzani D, Schiavini M, et al. (2007) High level of genetic heterogeneity in S and P genes of genotype D hepatitis B virus. Virology 365: 113–124.
- 14. Kimbi GC, Kramvis A, Kew MC (2004) Distinctive sequence characteristics of subgenotype A1 isolates of hepatitis B virus from South Africa. J Gen Virol 85: 1211–1220.
- 15. Lazarevic I, Cupic M, Delic D, Svirtlih NS, Simonovic J, et al. (2007) Distribution of HBV genotypes, subgenotypes and HBsAg subtypes among chronically infected patients in Serbia. Arch Virol 152: 2017–2025.
- 16. Kitab B, El Feydi AE, Afifi R, Derdabi O, Cherradi Y, et al. (2011) Hepatitis B genotypes/subgenotypes and MHR variants among Moroccan chronic carriers. J Infect 63: 66–75.
- 17. Meldal BH, Moula NM, Barnes IH, Boukef K, Allain JP (2009) A novel hepatitis B virus subgenotype, D7, in Tunisian blood donors. J Gen Virol 90: 1622–1628.
- 18. Lusida MI, Nugrahaputra VE, Soetjipto , Handajani R, Nagano-Fujii M, et al. (2008) Novel subgenotypes of hepatitis B virus genotypes C and D in Papua, Indonesia. J Clin Microbiol 46: 2160–2166.
- 19. Banerjee A, Kurbanov F, Datta S, Chandra PK, Tanaka Y, et al. (2006) Phylogenetic relatedness and genetic diversity of hepatitis B virus isolates in Eastern India. J Med Virol 78: 1164–1174.
- 20. Norder H, Courouce AM, Coursaget P, Echevarria JM, Lee SD, et al. (2004) Genetic diversity of hepatitis B virus strains derived worldwide: genotypes, subgenotypes, and HBsAg subtypes. Intervirology 47: 289–309.
- 21. Lemey P, Rambaut A, Drummond AJ, Suchard MA (2009) Bayesian phylogeography finds its roots. PLoS Comput Biol 5: e1000520.
- 22. Zehender G, Maddalena CD, Giambelli C, Milazzo L, Schiavini M, et al. (2008) Different evolutionary rates and epidemic growth of hepatitis B virus genotypes A and D. Virology 380: 84–90.
- 23. Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 4673–4680.
- 24. Schmidt HA, Strimmer K, Vingron M, von Haeseler A (2002) TREE-PUZZLE: maximum likelihood phylogenetic analysis using quartets and parallel computing. Bioinformatics 18: 502–504.
- 25. Strimmer K, von Haeseler A (1997) Likelihood-mapping: a simple method to visualize phylogenetic content of a sequence alignment. Proc Natl Acad Sci U S A 94: 6815–6819.
- 26. Hue S, Pillay D, Clewley JP, Pybus OG (2005) Genetic analysis reveals the complex structure of HIV-1 transmission within defined risk groups. Proc Natl Acad Sci U S A 102: 4425–4429.
- 27. Ciccozzi M, Ciccaglione A, Presti AL, Yalcinkaya T, Taskan ZP, et al. (2011) Reconstruction of the evolutionary dynamics of the hepatitis C virus 1b epidemic in Turkey. Infect Genet Evol 11(5): 863–8.
- 28. Posada D (2008) jModelTest: phylogenetic model averaging. Mol Biol Evol 25: 1253–1256.
- 29. Drummond AJ, Rambaut A (2007) BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol Biol 7: 214.
- 30. Suchard MA, Weiss RE, Sinsheimer JS (2001) Bayesian selection of continuous-time Markov chain evolutionary models. Mol Biol Evol 18: 1001–1013.
- 31. Kass RE, Raftery AE (1995) Bayes factors. Journal of American Statistical Association 90: 773–795.
- 32. Bielejec F, Rambaut A, Suchard MA, Lemey P (2011) SPREAD: spatial phylogenetic reconstruction of evolutionary dynamics. Bioinformatics 27: 2910–2912.
- 33. Murhekar MV, Murhekar KM, Sehgal SC (2008) Epidemiology of hepatitis B virus infection among the tribes of Andaman and Nicobar Islands, India. Trans R Soc Trop Med Hyg 102: 729–734.
- 34. Fares MA, Holmes EC (2002) A revised evolutionary history of hepatitis B virus (HBV). J Mol Evol 54: 807–814.
- 35. Hannoun C, Horal P, Lindh M (2000) Long-term mutation rates in the hepatitis B virus genome. J Gen Virol 81: 75–83.
- 36. Okamoto H, Tsuda F, Sakugawa H, Sastrosoewignjo RI, Imai M, et al. (1988) Typing hepatitis B virus by homology in nucleotide sequence: comparison of surface antigen subtypes. J Gen Virol 69: 2575–2583.
- 37. Orito E, Mizokami M, Ina Y, Moriyama EN, Kameshima N, et al. (1989) Host-independent evolution and a genetic classification of the hepadnavirus family based on nucleotide sequences. Proc Natl Acad Sci U S A 86: 7059–7062.
- 38. Osiowy C, Giles E, Tanaka Y, Mizokami M, Minuk GY (2006) Molecular evolution of hepatitis B virus over 25 years. J Virol 80: 10307–10314.
- 39. Zhou Y, Holmes EC (2007) Bayesian estimates of the evolutionary rate and age of hepatitis B virus. J Mol Evol 65: 197–205.
- 40. Harrison A, Lemey P, Hurles M, Moyes C, Horn S, et al. (2011) Genomic analysis of hepatitis B virus reveals antigen state and genotype as sources of evolutionary rate variation. Viruses 3: 83–101.
- 41. Lindh M, Andersson AS, Gusdal A (1997) Genotypes, nt 1858 variants, and geographic origin of hepatitis B virus–large-scale analysis using a new genotyping method. J Infect Dis 175: 1285–1293.
- 42. Chandra PK, Biswas A, Datta S, Banerjee A, Panigrahi R, et al. (2009) Subgenotypes of hepatitis B virus genotype D (D1, D2, D3 and D5) in India: differential pattern of mutations, liver injury and occult HBV infection. J Viral Hepat 16: 749–756.
- 43. Ghosh S, Banerjee P, RoyChoudhury A, Sarkar S, Ghosh A, et al. (2011) Unique hepatitis B virus subgenotype in a primitive tribal community in eastern India. J Clin Microbiol 48: 4063–4071.
- 44. Kato H, Ruzibakiev R, Yuldasheva N, Hegay T, Kurbanov F, et al. (2002) Hepatitis B virus genotypes in Uzbekistan and validity of two different systems for genotyping. J Med Virol 67: 477–483.
- 45. Khan A, Kurbanov F, Tanaka Y, Elkady A, Sugiyama M, et al. (2008) Epidemiological and clinical evaluation of hepatitis B, hepatitis C, and delta hepatitis viruses in Tajikistan. J Med Virol 80: 268–276.
- 46. Tallo T, Norder H, Tefanova V, Krispin T, Priimagi L, et al. (2004) Hepatitis B virus genotype D strains from Estonia share sequence similarity with strains from Siberia and may specify ayw4. J Med Virol 74: 221–227.
- 47. Mohebbi SR, Amini-Bavil-Olyaee S, Zali N, Damavand B, Azimzadeh P, et al. (2012) Characterization of hepatitis B virus genome variability in Iranian patients with chronic infection, a nationwide study. J Med Virol 84: 414–423.
- 48. Ahmed CS, Wang ZH, Bin Z, Chen JJ, Kamal M, et al. (2009) Hepatitis B virus genotypes, subgenotypes, precore, and basal core promoter mutations in the two largest provinces of Pakistan. J Gastroenterol Hepatol 24: 569–573.
- 49. Baig S, Siddiqui A, Chakravarty R, Moatter T (2009) Hepatitis B virus subgenotypes D1 and D3 are prevalent in Pakistan. BMC Res Notes 2: 1.
- 50. Amini-Bavil-Olyaee S, Alavian SM, Adeli A, Sarrami-Forooshani R, Sabahi F, et al. (2006) Hepatitis B virus genotyping, core promoter, and precore/core mutations among Afghan patients infected with hepatitis B: a preliminary report. J Med Virol 78: 358–364.
- 51. Deterding K, Constantinescu I, Nedelcu FD, Gervain J, Nemecek V, et al. (2008) Prevalence of HBV genotypes in Central and Eastern Europe. J Med Virol 80: 1707–1711.
- 52. Olinger CM, Lazouskaya NV, Eremin VF, Muller CP (2008) Multiple genotypes and subtypes of hepatitis B and C viruses in Belarus: similarities with Russia and western European influences. Clin Microbiol Infect 14: 575–581.
- 53. Lazarevic I, Cupic M, Delic D, Svirtlih NS, Simonovic J, et al. (2010) Prevalence of hepatitis B virus MHR mutations and their correlation with genotypes and antiviral therapy in chronically infected patients in Serbia. J Med Virol 82: 1160–1167.
- 54. Panessa C, Hill WD, Giles E, Yu A, Harvard S, et al. (2009) Genotype D amongst injection drug users with acute hepatitis B virus infection in British Columbia. J Viral Hepat 16: 64–73.
- 55. Pourkarim MR, Amini-Bavil-Olyaee S, Verbeeck J, Lemey P, Zeller M, et al. (2009) Molecular evolutionary analysis and mutational pattern of full-length genomes of hepatitis B virus isolated from Belgian patients with different clinical manifestations. J Med Virol 82: 379–389.
- 56. Swenson PD, Van Geyt C, Alexander ER, Hagan H, Freitag-Koontz JM, et al. (2001) Hepatitis B virus genotypes and HBsAg subtypes in refugees and injection drug users in the United States determined by LiPA and monoclonal EIA. J Med Virol 64: 305–311.
- 57. Reuben A (2002) The thin red line. Hepatology 36: 770–773.
- 58. Frierson JG (2010) The yellow fever vaccine: a history. Yale J Biol Med 83: 77–85.
- 59. Oliphant JW (1944) Jaundice Following Administration of Human Serum. Bull N Y Acad Med 20: 429–445.
- 60. Seeff LB, Beebe GW, Hoofnagle JH, Norman JE, Buskell-Bales Z, et al. (1987) A serologic follow-up of the 1942 epidemic of post-vaccination hepatitis in the United States Army. N Engl J Med 316: 965–970.
- 61. Mortimer (1995) Arsphenamine jaundice and the recognition of instrument-borne virus infection. Genitourin Med 71(2): 109–19.
- 62. Zuckerman JN, Zuckerman AJ (1999) THE EPIDEMIOLOGY OF HEPATITIS B. Clinics in Liver Disease 3: 179–187.
- 63. Hadziyannis SJ (2011) Natural history of chronic hepatitis B in Euro-Mediterranean and African countries. J Hepatol 55(1): 183–91.
- 64. Alavian SM, Fallahian F, Lankarani KB (2007) The changing epidemiology of viral hepatitis B in Iran. J Gastrointestin Liver Dis 16: 403–406.
- 65. Custer B, Sullivan SD, Hazlet TK, Iloeje U, Veenstra DL, et al. (2004) Global epidemiology of hepatitis B virus. J Clin Gastroenterol 38: S158–168.
- 66. Hutin YJ, Hauri AM, Armstrong GL (2003) Use of injections in healthcare settings worldwide, 2000: literature review and regional estimates. BMJ 327: 1075.
- 67. Kane A, Lloyd J, Zaffran M, Simonsen L, Kane M (1999) Transmission of hepatitis B, hepatitis C and human immunodeficiency viruses through unsafe injections in the developing world: model-based regional estimates. Bull World Health Organ 77: 801–807.
- 68. Simonsen L, Kane A, Lloyd J, Zaffran M, Kane M (1999) Unsafe injections in the developing world and transmission of bloodborne pathogens: a review. Bull World Health Organ 77: 789–800.
- 69. Hutin YJ, Harpaz R, Drobeniuc J, Melnic A, Ray C, et al. (1999) Injections given in healthcare settings as a major source of acute hepatitis B in Moldova. Int J Epidemiol 28: 782–786.
- 70. Dentinger C, Pasat L, Popa M, Hutin YJ, Mast EE (2004) Injection practices in Romania: progress and challenges. Infect Control Hosp Epidemiol 25: 30–35.
- 71. Nelson PK, Mathers BM, Cowie B, Hagan H, Des Jarlais D, et al. (2011) Global epidemiology of hepatitis B and hepatitis C in people who inject drugs: results of systematic reviews. Lancet 378: 571–583.
- 72. van Houdt R, Bruisten SM, Geskus RB, Bakker M, Wolthers KC, et al. (2010) Ongoing transmission of a single hepatitis B virus strain among men having sex with men in Amsterdam. J Viral Hepat 17: 108–114.