Hepatitis A virus (HAV) is one of the major causes of acute hepatitis throughout the world and causes substantial morbidity in both developed and developing countries [1, 2]. On the basis of genome sequence divergence, all viruses infecting humans have been classified in three genotypes (I, II, III), further divided into two sub-genotypes (A and B). Genotypes and subtypes tend to show different geographic distribution [3].

HAV is mainly transmitted by the faecal-oral route. HAV can survive for long in water and numerous waterborne epidemics have been observed following consumption of contaminated drinking water [4], food produce [5, 6], or shellfish [7].

The infection is usually asymptomatic or mild in children under five years, while in adults more frequently occurs with symptoms and jaundice. Clearance of infection confers lifelong immunity. In high endemicity countries, HAV infection is acquired in early childhood and most adult population is positive for anti-HAV IgG and protected from re-infection. In contrast, in low endemicity countries most adult population is susceptible: as a result, infections are more likely to occur in adults, in which are frequently symptomatic [8].

In Europe, the epidemiology of the HAV infection is characterized by countries with variable endemicity. In 2012, the European Centre for Disease Prevention and Control (ECDC) reported a mean rate of infection in Europe of 2.6 cases per 100,000 inhabitants (13,156 total confirmed cases). The majority of them were from an endemic area which included Eastern European and Balkan region countries [9]. Among them, Bulgaria is a country with intermediate hepatitis A endemicity, with annual incidence between 25 and 94 cases per 100,000 population (data since 2000). In 2011 and 2012 the cases increased to 75.88/100,000 (5588 cases) and 67.13/100,000 (4919 cases), respectively, then decreased to 25.05/100,000 in 2013 (1825 cases) [10].

In 2011–2013, the most affected population group was children aged 5–9 years, followed by the 10–14 years and 1–4 years age groups, which are the main risk groups for the disease in Bulgaria in general. During this time, 28 outbreaks of hepatitis A were detected in different Bulgarian districts, affecting dominantly minority Roma neighbourhoods [9, 10]. The prevalence of anti-HAV antibodies in general population in Bulgaria is poorly known. The only available data were from a small study carried out in Plovdiv, the second largest town in Bulgaria, in which two quarters with very different hygienic and sanitary conditions were sampled [11]. The mean seroprevalence was 68.3%, but it reached 90.2% in the quarter with poor hygienic and sanitary conditions and 44.8% in the other with normal conditions [11]. Vaccination in Bulgaria is recommended for people at risk [12]; however, targeted immunization campaigns are launched in response to outbreaks, such as in 2006 [13].

Molecular analysis of HAV isolates has proved to be crucial for understanding the demographic, social and environmental factors that have given rise to the current epidemiological patterns in Europe. Several studies described circulation of different HAV genotypes from both autochthonous and travel-related cases [14–21]. Nevertheless, HAV strains from Bulgaria and other European endemic countries have never been described. This gap delays the comprehension of the local transmission routes and the implementation of control measures in areas that urgently need them.

In the present study, serum samples from 105 patients with acute hepatitis A, collected in different towns and villages in Bulgaria during 2012–2014, were analysed by nested RT-PCR and sequencing to characterize HAV isolates. Phylogenetic analysis of sequences was performed to describe, for the first time, HAV genotypes and strains circulating in Bulgaria.

The ECDC surveillance report of food- and waterborne diseases, published in 2014, defined Bulgaria as the European country that reported the highest number of hepatitis A cases in 2011 (5587 cases) and 2012 (4919 cases). Investigation of a large outbreak in Bulgaria, occurred in 2012 in the Iskra village in the Plovdiv region, indicated that unregulated and/or frequently failing water supply and sewage system are associated with virus spread in this area [9]. Despite the level of endemicity, molecular epidemiology of HAV in Bulgaria has never been described. Thus, the studied samples represent an extremely rare resource, if not unique, to obtain information on HAV strains that have circulated and might still be circulating in an Eastern Europe country. The results of the present study represent the first analysis of the HAV genotypes and strains circulating in Bulgaria.

Analysis of the age distribution showed that patients were younger in SCs than in LMCs, suggesting different endemicity among these areas. In SCs, high HAV endemicity results in a large proportion of adults protected by antibodies developed in childhood and, thus, HAV circulation mainly in children. High endemicity in SCs is likely related to poor hygienic conditions and health education of the population, as suggested by a study investigating an outbreak occurred in 2012 in a SC (Iskra village) [25].

Molecular characterization of HAV from 105 acute hepatitis A cases, collected between 2012 and 2014, indicated that both IA and IB sub-genotypes circulate in Bulgaria; however, about 74% of the isolates were assigned to sub-genotype IA that, thus, seems to be dominant.

Interestingly, IA strains grouped in two distinct phylogenetic clusters. The major cluster, that included 79.5% of the IA cases and 59% of the total cases, contained highly related viruses. Sequence analysis indicated that isolates were either identical (at least over the analysed genomic region), as observed in strain 1 or strain 2 groups, or showed very few (1 to 4) nucleotide variations. Strain 1 and strain 2 groups differed each other for a single specific nucleotide, that was T in strain 1 and C in strain 2. The very low level of variability observed in the IA major cluster might be the result of a common origin of the two main strains and of their spread in the local population, with random generation of new variants, highly related to the main strains, in the course of endemic and epidemic circulation.

The distribution of strains according to urban size (Fig. 3b) and the geographical localization of the involved urban centers (Fig. 4) indicate that strain 1 preferentially circulated in small urban centers and villages south-east of Sofia, where transmission was likely maintained by poor water supply and lack of infrastructures. Although the relative frequency of strain 2 was higher in LMCs than in SCs (Fig. 3b), strain 2 showed wider geographical distribution than strain 1 (Fig. 4) and it was detected both in large urban centers, such as Sofia, Pernik and Shumen, and in small villages.

A completely different pattern of circulation was observed for isolates from the IA minor cluster and from the IB cluster. In fact, all but one patients infected with strains from the IA minor cluster and two thirds of cases infected with IB strains were from LMCs (Fig. 3a), suggesting that different transmission routes, linked to urbanization, might operate in these cases (e.g. consumption of shellfish or other food, contaminated in the same or other countries).

Assuming the distribution of HAV genotypes and strains reported in the present study roughly represents isolates circulating in Bulgaria in 2012–2014, it is conceivable that the IA strains from the IA major cluster, accounting for 59% of sequenced cases, were responsible for the epidemic peak observed in Bulgaria in 2011 and 2012 and, most likely, for the multiple outbreaks detected in different Bulgarian districts during those years [9, 10].

This is the first study describing the HAV genotypes and strains circulating in Bulgaria, one of the European countries with intermediate endemicity. Genotype IA predominates, with several highly related strains. Further studies need to be carried out to increase the number of molecularly characterised isolates from Bulgaria and from other countries of Eastern Europe that, by a phylogenetic approach, can be useful to investigate the factors underlying the spread of HAV infections and to implement appropriate preventing measures, at both national and European level.