Sequence diversity of hepatitis D virus in Mongolia

Abstract

Introduction: The Hepatitis Delta Virus (HDV) is a defective, single-stranded RNA virusoid encoding for a single protein, the Hepatitis Delta Antigen (HDAg), which requires the hepatitis B virus (HBV) envelope protein (HBsAg) for its transmission. Currently, hepatitis D is the most aggressive form of viral hepatitis and treatment options are limited. Worldwide 12 million people are chronically infected with HDV being at high risk for progression to cirrhosis and development of liver cancer.

Objectives: Although it is well established that Mongolia is the country with the highest prevalence of HDV infections, the information on the molecular epidemiology and factors contributing to HDV sequence diversity are largely unclear. The aim of the study was to characterize the sequence diversity of HDV in rural areas from Mongolia and to determine the extent of HLA class I-associated selection pressure.

Patients and methods: From the HepMongolia cohort from rural areas in Mongolia, 451 HBsAg-positive individuals were selected and anti-HDV, HDV-RNA and the sequence of the large HDAg was determined. For all individuals the HLA class I locus was genotyped. Residues under selection pressure in the presence of individual HLA class I types were identified with the recently published analysis tool HAMdetector.

Results: Of 431 HBsAg positive patients, 281 were anti-HDV positive (65%), and HDV-RNA could be detected in 207 of 281 (74%) of patients. The complete large HDAg was successfully sequenced from 131 samples. Phylogenetic analysis revealed that all Mongolian HDV isolates belong to genotype 1, however, they separate into several different clusters without clear regional association. In turn, from phylogeny there is strong evidence for recent local transmission events. Importantly, we found multiple residues with strong support for HLA class I-associated selection pressure consistent with a functional CD8⁺ T cell response directed against HDV.

Conclusion: HDV isolates from Mongolia are highly diverse. The molecular epidemiology suggests circulation of multiple subtypes and provides evidence for ongoing recent transmissions.

Keywords: HDV subtypes; HLA class I-associated selection pressure; Mongolia; hepatitis D virus; molecular epidemiology; sequence diversity.

Figure 1

Recruitment of the HepMongolia cohort. (A) Sampling regions for the HepMongolia cohort. The map was created with data from open street map and visualized with datawrapper. (B) HDV viral load in the different sampling regions.

Figure 2

Phylogenetic analysis of viral sequences covering the large HDAg. A total of 131 HDV sequences covering the large HDAg were aligned with reference sequences [from (11)] using MAFFT. A phylogenetic tree was calculated with the Mr. Bayes Plugin (34) in Geneious 10.2.8 using the GTR genetic distance model and GT3 as outgroup. (A) Left: unrooted tree all sequences from the HepMongolia cohort together with reference sequences with assigned genotypes. Genotypes 2–8 are colored red, GT1 samples in black. Right: Inset showing only GT1 samples. Sequences from the HepMongolia cohort are colored black, GT1 subtypes described by [(11)] are color coded. (B) Disperse geographical distribution of HDV GT1 subtypes in Mongolia. The color code is according to the sampling region.

Figure 3

Sites under selection predicted by Bayesian regression model. (A) Shannon entropy map of the 131 L-HDAg Sequences. (B) Posterior probability values >0.5 of the Bayesian regression model from the HAMdetector for each HLA-molecule on every amino acid in the L-HDAg are shown. (C) Posterior probability values of an HAMdetector run with the same alignment however with randomly assigned HLA-type.

Figure 4

Frequency of HLA-associated viral polymorphisms in the epitope region predicted by the HAMDetector. The frequency of variations from the prototype sequence of the epitope region are shown for patients carrying the corresponding HLA-allele (black) and patients not carrying the corresponding HLA-allele (white). Positions with significant differences in polymorphism frequencies in the absence or presence of the corresponding HLA-molecule are marked, and p values (Fisher exact test) and the most frequent variant amino acid are indicated. (A) HLA-B*37 epitope region aa98–110. The confirmed epitope is outlined. (B) HLA-A*68 epitope region aa78–90. The predicted epitope is outlined in dashed lines. (C) HLA-A*33 epitope region aa13–25. The predicted epitope is outlined in dashed lines.