[The global spread and evolution of hepatitis B virus genotypes, subgenotypes and recombinants]. (Doctoral thesis)

Despite efficient vaccines, hepatitis B virus (HBV) continues to be a global public health problem. Especially in low income regions, such as sub-Saharan Africa or South-East Asia, HBV is highly endemic. In these regions perinatal or early childhood transmission is considered the most important route of infection for HBV, frequently leading to chronic HBV carriage.
Based on the genetic variability of the DNA genome, hepatitis B virus has evolved into at least eight classified genotypes A–H and a tentative genotype I. Recently a tenth genotype J has been proposed. With the exception of genotypes E, G, H and J, HBV genotypes are divided into subgenotypes with more or less distinct geographic distributions. In highly endemic regions where different HBV genotypes and subgenotypes co-circulate, mixed infections and recombinations are frequent.
In the first parts of the present dissertation, genotype E is further characterized. This genotype has an extensive spread throughout a vast West African crescent where it is highly prevalent. Nevertheless, identical sequences have been found several thousand kilometres apart. We show that genotype E has a surprisingly low genetic diversity, despite its wide spread and dominance in most countries in sub-Saharan Africa. These results indicate an only short time of evolution for genotype E, while genotype A, the minority genotype in this region, is highly divers and has a long time of evolution. Due to its wide spread, high prevalence and low genetic diversity, genotype E represents an unsolved puzzle that we investigated in multiple studies.
As the spread of pathogens is closely linked to their natural hosts, African slaves that were force-migrated to the Americas from the 17th to the early 19th century during the transatlantic slave trade would have disseminated African HBV strains in the New World. Nevertheless, we show that genotype E is only sporadically found in the Americas, corroborating an only recent emergence of this genotype. We furthermore demonstrate that in Haiti, where >90% of the population are descendants of African slaves, more than 40% of hepatitis B virus infections were caused by subgenotype A1, which today is found mainly in eastern Africa. Another 20% belong to a rare subgenotype, A5, which has been found only in the former Bight of Benin, a former primary slave trading post. Haitian A subgenotypes appear to have separated early from the African subgenotypes, while the most prevalent genotype and subgenotype in West Africa today (E and A3, respectively) are rare in Haiti. These results provide evidence that the dominant sub/-genotypes in West-Africa emerged in the general population only after the end of the transatlantic slave trade, within the last 100 to 200 years, and explain the low genetic diversity of genotype E. The high prevalence of genotype E in much of Africa further suggests that HBV hyperendemicity is a recent phenomenon, probably resulting from well-intended injection mass campaigns using unsafe needles, performed by the French and Belgian colonial powers.
Our phylogenetic analyses also provide evidence that genotype E may have emerged in the Northern part of the genotype E crescent before spreading to the South, as the Southern countries of the genotype E crescent exhibit lower genetic distances than the Northern countries. We confirm this by phylogeographic analyses that suggest an origin of genotype E in the area of Nigeria, before rapidly spreading throughout sub-Saharan Africa.
We find a substitution rate of 1.9x10-4 substitutions per site and year and demonstrate that genotype E emerged only recently, most likely within the last 130 years. Together with the strong increase in the effective number of genotype E infections over time, our results corroborate the recent introduction in sub-Saharan Africa. Interestingly, genotype E strains found in Haiti seem to be the result of multiple introductions only in the second half of the 20th century, corroborating an absence of a significant number of genotype E strains in West Africa when and where the slaves were rounded up. Taken together, the results provide strong evidence that the hyperendemicity of HBV and genotype E in today`s Africa is a surprisingly recent phenomenon and likely the result of dramatic changes in the routes of viral transmission.
In an additional study in sub-Saharan Africa we characterize hepatitis delta virus (HDV), a satellite virus of hepatitis B, in samples from Burkina Faso, the Central African Republic, Chad and Nigeria. We demonstrate that up to 50% of hepatitis B surface antigen positive carriers were infected with HDV, with prevalences varying between cohorts and countries. Children from Burkina Faso had a surprisingly high prevalence of HDV, with 20.5% being infected. This prevalence is almost 10-times higher than the one observed in their mothers and these children are at high risk to develop severe fulminant or chronic hepatitis. We furthermore found 50% of liver patients in the Central African Republic to be HDV-antibody positive, indicating that in the Central African Republic, in contrast to Nigeria with only 1.3% and 6.8% prevalences in liver patients, HDV super-infection might be an important cause of chronic hepatitis with possibly frequent severe liver conditions.
Of the eight characterized hepatitis delta clades, we detect clades 1, 5 and 6 in this region, with a predominance of HDV-1. Interestingly, more than 50% of genotypable HDV and HBV pairs revealed both HDV-1 and HBV/E. However, while our results indicate that HDV might have originated in Africa, we demonstrate that a close co-evolution of HDV clade 1 and HBV genotype E seems unlikely.
In additional studies, we analysed sera from Lao PDR, were about 9% of the population are chronic carriers of HBsAg, the primary marker of hepatitis B virus infection. Additionally, multiple subgenotypes of genotypes B, C and I co-circulate in this country.
We demonstrate that about 6% of HBsAg positive rejected blood donors are infected with multiple sub-/genotypes. We furthermore provide evidence, that about 65% of the mixed infected donors showed recombinations in the surface gene that had surprisingly diverse recombination breakpoints and involved the predominant genotypes B and C. We show that recombinant virus strains were largely distinct and varied between donors and within individuals. In addition, our analyses indicate that recombinant strains largely evolved within the lifetime of the donor. In addition, based on the analyses of viral quasipecies, we propose a substitution rate of 1.58x10-4 substitutions per site and year for strains in Lao PDR. Thus, our results provide evidence that, at least in Laos, hepatitis B virus mixed infections lead to frequent recombinations that so far have not led to new dominant strains in the population.
Finally, in serologic and phylogenetic analyses on sera from randomly selected blood donors in Lao PDR, we show that 45.5% of the blood donors were positive for at least one of the hepatitis B virus markers. Our results indicate that about half of the Laotian population has been in contact with this virus at one point during their lives. We furthermore provide evidence, that blood screening procedures in Lao PDR need to be reconsidered. We show that the screening of donations solely for HBsAg leaves a sizable risk of hepatitis B virus transmission by HBsAg-negative occult infected donors in at least 3.9% of blood donations. These are potentially DNA positive and could transmit hepatitis B virus.