Analysis of serum specimens collected from two remote rural communities in North Central Nigeria revealed that ~11% of their residents were actively infected with HBV. Although this high prevalence concurs with the rate of HBV infection of 10–40% found in other populations across Nigeria 
, it is surprising that limited outside contacts did not shield these communities from HBV infection. Taking into consideration the isolation of these communities from the general population and the high prevalence of HBV infection, it is conceivable that only a few HBV strains could have been introduced and effectively transmitted among the residents. However, phylogenetic analysis of the S
-gene sequences from 55 HBV isolates recovered in this study showed the presence of many HBV variants that belong to two genotypes, A and E ( and ).
HBV/E was the most prevalent genotype, being detected among 96.4% of the isolates. Phylogenetic analysis showed that all the HBV/E variants were closely related to HBV/E strains circulating in West and Central Africa ( and ). Detection of the most frequent HBV/E variant () in many African countries strongly supports close genetic relatedness between HBV/E variants in these communities and other regions of Africa. These findings are in agreement with the previous observation of HBV/E in the majority of HBV-infected individuals in Nigeria 
. Additionally, it confirms that Nigeria is part of the HBV/E crescent spanning countries from Senegal to Namibia 
. Only 2 (3.6%) isolates belonged to HBV/A3, and both preferentially clustered with the recently identified HBV/A3 isolates from Cameroon 
rather than with HBV/A3 isolates from Gabon 
. Cameroon could have been a possible source of HBV/A3 into Nigeria owing to the substantial volume of trade and intermarriage between these two neighboring countries (J.C.F. personal observation). None of the HBV/A in this study grouped with the African/Asian A1 or European/American A2 subgenotypes.
Of the 53 HBV/E isolates, 51 (96.2%) were predicted to belong to serotype ayw4
and 2 (3.8%) to ayw2
, while the 2 HBV/A3 isolates belonged to ayw1
. These data are in disagreement with previous findings that all HBV/E from Africa belonged to serotype ayw4 
. No evidence of recombination was found in the HBV/E ayw2
isolates. The geographical distribution and frequency of HBV/E ayw2
are not known.
Phylogenetic analysis of the whole-genome sequences () did not reveal any village-specific clustering and showed extensive intermixing of these sequences with other HBV/E variants identified in different countries in Africa. Genetic diversity between the whole-genome sequences identified for HBV/E isolates circulating in the 2 communities was 2.1%, which is similar to the range found among all other HBV/E variants 
. Such low genetic diversity among these local HBV/E isolates indicates that the virus might have only been recently introduced into these rural populations. Additionally, the presence of many HBV/E variants in the 2 communities and apparent intermixing with variants from other African countries suggest that all these HBV variants were introduced rather than having independently evolved in these communities. Because of limited contacts with outside populations, few opportunities were available for extraneous transmission, and, therefore, these HBV variants were most probably introduced to these villages simultaneously or within a very short timespan.
Analysis of HBV quasispecies from 24 residents of these 2 communities revealed that each individual was infected with many different HBV variants (). Identification of the genotype D and G sequences in 5 individuals () adds another level of complexity to HBV populations circulating in the 2 communities. However, the consensus sequences identified by direct sequencing of PCR fragments ( and ) did not accurately reflect the genotype complexity of the HBV population in each infected person. Among 22 residents infected with HBV/E, 20 shared HBV variants with 2–14 other individuals. Variant sharing was so extensive that a large network of shared HBV sequences linking 20 residents could be constructed (). Among these 20 residents, 10 were completely interlinked through shared HBV sequences to each other. Should variant sharing be considered as the proof of transmission 
, this network reveals a very complex pattern of HBV transmission, which could be related to frequent infections with more than one HBV variant or to extensive superinfections with different HBV variants. The network contains male and female individuals of different ages residing in both villages, which suggests that HBV transmission occurred across both communities via a mode affecting the entire population.
The predominant mode of transmission leading to such significant HBV variant sharing is not known. Analysis of HBV quasispecies shared within 2 groups of siblings suggests that intra-familial transmission is not the only possible mode of transmission. These numerous HBV strains could have been maintained within the population through socio-cultural practices like facial or body scarification, traditional birth attendance and shaving by local barbers using unsterilized sharp instruments, all of which have the potential for the transmission of blood-borne pathogens and which have been associated with the transmission of human immunodeficiency virus (HIV) in Nigeria 
Phylogenetic analysis also indicated a very close genetic relatedness between predominant and frequently shared HBV/E variants. These observations in conjunction with the star-like MJN topology suggest a very dynamic evolution of HBV/E variants found in these communities. The close connections between the HBV/E variants ( and ) and frequent sharing of HBV sequences between individuals did not allow for a clear division of these HBV/E variants into separate strains, suggesting that the population of variants can be seen as a single swarm evolving among many hosts. Given the significant intermixing between the HBV/E variants found in this study and HBV/E variants identified in other African countries, together with the low heterogeneity of all these variants, this suggestion seems applicable to the entire HBV/E. The existence of this swarm of closely related HBV variants may reflect its recent origin followed by diversifying selection and adaptation to its particular transmission mode in the West/Central African HBV/E crescent.
The origin of HBV/E remains unclear. The very rare detection of HBV/E infections outside of Africa suggests that HBV/E became prevalent in the West/Central Africa only after the trans-Atlantic slave trade 
. This hypothesis of the recent HBV/E origin was supported by tMRCA analysis. The MRCA for the 47 HBV/E full-length isolates identified in this study was estimated to appear in ~1952 (95% HPD: 1927–1970). Surprisingly, a similar tMRCA (~1948; 95% HPD: 1924–1966) was estimated for the entire HBV/E. These findings indicate that the HBV/E variants identified in the 2 villages and in the entire HBV/E crescent in Africa have a similar time of origin.
Uncertainty in establishing the rate of substitutions for HBV may significantly affect estimates of tMRCA. Recently, tMRCA for HBV/E has been estimated to range between 30 to 1536 years depending on the used substitution rate 
. The rate of 2.97×10−4
substitutions per site per year 
used in the present analysis is most consistent with estimates made in several studies 
. The analysis conducted here dated the HBV/E MRCA within a timeframe consistent with the hypothesis that modern HBV/E lineages emerged after the cessation of the trans-Atlantic slave trade 
Although the modern HBV/E variants have a recent origin, the HBV/E variants were probably present long before the calculated tMRCA. Therefore, HBV/E could still have been introduced to the other parts of the world through the slave trade. The recent discovery of HBV/E variants among individuals of African descent in one isolated community in Colombia, South America 
provides some support to this hypothesis. The coalescent analysis conducted in that study suggested that the closely related HBV/E variants found in that community originated from a single variant that existed a few years ago, thus indicating that only a single HBV/E strain was introduced to that community 
. Considering its isolation and lack of contacts with persons traveling from Africa over the period of time exceeding the calculated tMRCA for these variants 
, it can be speculated that the ancestral HBV/E strain was introduced to that community many years ago, possibly even during the time of the slave trade.
The 2 isolated Nigerian communities in the current study were infected with and maintained a diverse population of HBV/E variants. This diversity is of the same degree as for all HBV/E isolates identified in different countries of Africa. Thus, HBV/E variants from Colombia 
and the 2 communities in this study most probably had different evolutionary and epidemiological histories. Although the actual epidemiological and evolutionary processes leading to such a difference in the HBV/E sequence diversity are not known, it may be speculated that the conditions for massive introduction of multifarious HBV variants to the Nigerian communities did not exist when HBV was introduced to the Colombian community.
Strikingly, the estimated divergence dates for the identified HBV/E variants from the MRCA coincide with the period of intense mass public-health campaigns conducted in West/Central Africa. During 1967–1969, the World Health Organization mounted a large-scale program to eradicate smallpox and measles in West and Central Africa by arm-to-arm injections using jet injectors that have been recognized as presenting a significant risk for infection by blood-borne pathogens 
. The geographical area stretching from Mauritania to the Congo (Zaire) river and from the Bight of Benin to the Sahara desert, covering 20 countries, constituted a single contiguous territory with ~120 million inhabitants at that time. Between January 1967 and December 1969, one hundred million persons living in this belt were vaccinated against smallpox 
. However, the mass vaccination practices were already adopted more than 50 years before the campaigns began. Mass inoculations against smallpox was carried out in the late 19th
century in Nigeria, Benin, Ghana, Guinea and Burkina-Faso that involved serial exchanges of blood and lymph 
. These could have constituted a route along which HBV/E was transmitted. It has been proposed that mass vaccination campaigns were associated with the dissemination of HBV/E in this region of Africa 
. In Egypt, unsafe injections used during nationwide campaigns against schistosomiasis between 1920 and 1980 have been associated with the country's high prevalence (>40%) of hepatitis C virus (HCV) infection 
. With HBV estimated to be ~10 times more transmissible than HCV 
, unsafe injections are a possible route of HBV/E transmission.
The data obtained in this study, although providing no direct indication on the actual events leading to the high HBV prevalence in the 2 communities, are consistent with massive transmission that resulted in rapid expansion of the HBV/E-infected population. In addition to the Bayesian coalescent, phylogenetic and MJN analyses described above, the Tajima's D and Fu's statistical tests 
provided support to this hypothesis. Analysis of the skyline plot () for HBV/E variants identified in these 2 communities showed increase in the effective number of HBV/E infections ~30–40 years ago. Identification of HBV/E variants identical in the S-gene sequence to the HBV/E MRCA in many countries in Africa also suggests a rapid HBV/E spread. The conditions facilitating such massive HBV/E expansion are not known. It is possible that HBV/E was the most prevalent genotype in this region of Africa before the expansion and/or the most adapted to the mode of transmission leading to such expansion. Analysis of selection pressures acting in the 2 Nigerian communities identified only few HBV/E sites under the positive selection (). Identification of a unique positively selected site at the anchor position of the potential T-cell epitope in the polymerase 
supports a possible unique adaptation of HBV/E at the population level. All these findings suggest a dramatic shift in the epidemiological factors and evolutionary trends affecting the presentation of HBV/E in the West and Central Africa sub-region over the last century.