Clin Microbiol Infect 2011; 17: 223–231
Hepatitis B virus (HBV) is classified into eight major genotypes, A–H, which are geographically distributed worldwide. The aim of this work was to describe the clinical characteristics associated with the HBV genotypes circulating in Buenos Aires city. The study included 139 patients infected with HBV, whose clinical courses were classified as acute symptomatic self-limiting hepatitis, inactive carrier state and chronic active hepatitis (HBV e-antigen (HBeAg)-positive and HBeAg-negative). The HBV genotypes were determined in 128 patients by PCR–restriction fragment length polymorphism and phylogenetic analysis. Biochemical, virological, clinical and histological features were analysed. A differential distribution of genotypes between acute symptomatic and chronic infections was found. Among the acute cases, genotype F was predominant (65.2%, 30/46) and genotype D was rare (4.3%, 2/46), whereas among the chronic infections, a homogeneous distribution of genotypes A (26.8%, 22/82), D (31.7%, 26/82) and F (36.6%, 30/82), with an unusual presence of genotypes B (1.2%, 1/82) and C (3.7%, 3/82), was observed. Regarding the liver histology of chronically infected patients, genotype F tended to display higher histological activity indexes. Mutations related to HBV surface antigen immunoreactivity, antiviral resistance and HBeAg-negative status were studied. This work constitutes, to our knowledge, the first description of the clinical characteristics related to HBV genotypes in Argentina, where the distribution of genotypes in patients with acute infection has not been reported previously. Finally, it was established that genotype F is the prevalent genotype among the acute symptomatic infections in Buenos Aires city, and that it shows a tendency to cause an adverse disease outcome among the chronic cases.
Hepatitis B virus (HBV) affects 350–400 million people worldwide and accounts for one million deaths annually from cirrhosis, liver failure and hepatocellular carcinoma [1,2].
The virus has been classified into eight major genotypes, A–H, and has a worldwide geographical distribution .
In Argentina, the presence of genotypes A, D and F has been previously reported [8,9]. However, there is no survey of the clinical courses in relation to the different genotypes.
Thus, the aim of this work was to describe the clinical characteristics associated with the HBV genotypes circulating in Buenos Aires city.
Materials and Methods
This was a retrospective study (from July 1999 to March 2006) of adults, unrelated patients infected with HBV who attend the F. J. Muñiz Hospital in Buenos Aires city. In patients who underwent treatment, the sample analysed corresponded to the pretreatment period. Co-infections with hepatitis C virus, human immunodeficiency virus or hepatitis D virus were excluded.
Patients were divided into three groups. For the classification, previous reports [10,11] were taken into account.
Group 1 comprised 51 patients with acute symptomatic self-limiting hepatitis (acute infections) with confirmed anti-HBV core antigen IgM and who had been HBV surface antigen (HBsAg)-positive for less than 6 months from the beginning of symptoms. The most common symptoms were jaundice, loss of appetite, nausea, and pain in the right upper part of the abdomen. The presence of anti-HBsAg antibodies was confirmed; however, 12 patients interrupted their follow-up after the HBsAg and viral load became undetectable.
Group 2 comprised 21 patients defined as inactive carriers, who had been HBsAg–positive for at least 6 months, were HBV e-antigen (HBeAg)-negative (HBeAg–, had viral loads under 104 copies/mL, and had persistently normal alanine aminotransferase (ALT) levels.
Group 3 comprised 67 patients with chronic active infections (chronic hepatitis), who had been HBsAg-positive for at least 6 months, had viral loads above 104 copies/mL, had elevated ALT, had fibrosis (F) scores >2, and had histological activity index (HAI) scores >3. This group was further subdivided into HBeAg-positive (HBeAg+) (n = 32) and HBeAg– (n = 35).
The study was carried out according to the World Medical Association Declaration of Helsinki and was approved by the Ethical Committee of the School of Pharmacy and Biochemistry, University of Buenos Aires.
Laboratory testing and evaluation of liver biopsy
HBV serological markers were analysed with the Axsym Abbott system (MEIA) (HBsAg v2.0, HBeAg and IgM anticore; Abbott Diagnostics, Wiesbaden, Germany), HBV DNA in serum was quantified by using the Amplicor HBV Monitor Test v1.0 (Roche Diagnostics, Meylan, France), and ALT levels were expressed as multiples of the upper limit of the normal value (41 UI/L).
Liver biopsy specimens were evaluated according to the modified Knodell score, and categorized as mild or advanced fibrosis (F ≤2 or F >2, respectively) and as low or high necroinflammation (HAI ≤3 or HAI >3, respectively). HAI ≥7 was considered to be a histological indication for treatment. All patients gave informed consent to undergo liver biopsy.
Genotyping of the samples by nested PCR–restriction fragment length polymorphism on the S gene was based on that reported by Zeng et al. , modified to improve the detection of genotype F.
Briefly, DNA was extracted from a 200-μL serum sample by using the QIAamp DNA Mini Kit (Qiagen GmbH, Hilden, Germany). The PCR was perfomed with primers 5′-CTGCTGGTGGCTCCAGTTC-3′ (nucleotides 57–75) and 5′-AGAAAATTGGTAACAGMGGYA-3′ (nucleotides 815–795) in the first round, and 5′-GCGGKGTKTTTCTTGTTG ACAA-3′ (nucleotides 203–224) and 5′-GGGACTCAAGATGYTGYACAG-3′ (nucleotides 787–767) in the second round, with the following conditions: 4 min at 95°C, 35 cycles of 1 min at 95°C, 45 s at 57°C (first round) or 45 s at 60°C (second round) and 1 min at 72°C, with a final extension of 10 min at 72°C. The amplified product spanned 585 nucleotides. The restriction of the PCR product was performed with the enzymes EaeI, MspI, StyI, MboI and BsrI.
Amplification of the basal core promoter/precore gene (BCP-pC) regions
Primers 5′-ATGGAGACCACCGTGAACGC-3′ (nucleotides 1608–1627) and 5′-CCCACCTTATGAGTCCAAGG-3′ (nucleotides 2484–2465) and primers 5′-TGCCAACAGTCTTACATAAGMG-3′ (nucleotides 1639–1660) and 5′-GAGTTCTTCTTCTAGGGGACCTG-3′ (nucleotides 2381–2359) were used in the first and second rounds, respectively, of a nested PCR to amplify the BCP-pC of the chronic samples. The reaction conditions in both rounds were: 5 min at 94°C, 35 cycles of 30 s at 94°C, 30 s at 53°C and 45 s at 72°C, and a final extension of 10 min at 72°C. The amplified product spanned 742 nucleotides.
Sequencing and phylogenetic analysis
The samples were sequenced in an ABI3130XL sequencer (Applied Biosystems, Carlsbad, California, USA). Sequences were aligned with ClustalX v1.83 , and edited with Bioedit v22.214.171.124 software . Phylogenetic analyses were performed using the maximum-likelihood method with tree bisection and reconnection branch swapping (PAUP* v4.0b10 ). The model of nucleotide substitution estimated by the Modeltest 3.7 program  was GTR+I+γ. The robustness of the phylogenetic grouping was evaluated by a neighbor-joining bootstrap analysis with 10 000 replicates (PAUP* v.4.0b10).
Quantitative variables were analysed by non-parametric tests (Mann–Whitney), and categorical variables by chi-squared or Fisher’s exact tests. The data were stratified into 2 × 2 tables, and measures of association were tested. The strength of the relationship was estimated by using ORs with 95% CIs. Multivariate analyses with logistic regression were used to determine the independent factors associated with the clinical course and the HBeAg status of the infection. Dummy variables were created for the variable ‘genotype’ (with more than two classes). The Spearman correlation test was used to evaluate the related variables. Differences were considered to be significant for p-values <0.05. Statistical analyses were performed using STATA 9.1 software .
Samples were genotyped by nested PCR–restriction fragment length polymorphism (n = 128) and validated by phylogenetic analysis (n = 121). Eleven PCR-negative samples were excluded from further analysis. Genotyping discrepancies between the methodologies were found in nine of the 121 sequenced samples (7.4%).
The distribution of genotype F sequences in the phylogenetic tree showed that a unique source of viruses was not supported, either for acute or for chronic isolates (Fig. 1). The analysis of the diversity of genotype D isolates showed that phylogeographical relationships of Argentinean samples were not evident, dismissing the existence of a local cluster (Fig. 2).
Comparative epidemiological, biochemical, virological, clinical and histological features of the studied population are shown in Tables 1 and 2.
|Characteristic||Total patients||Acute infections||Chronic infections||Patients with chronic hepatitis||Inactive carriers|
|Not detectable by PCR||11||5||6||0||3||3|
|Gender (men/women)||88 : 40||33 : 13||55 : 27||23 : 9||19 : 13||13 : 5|
|Age (years), median (range),hk||39 (18–79)||36.5 (18.0–78.0)||41 (20–79)||36 (20–79)||44 (20–65)||38 (22–58)|
|ALT (× normal value), median (range), ni||2.4 (1.0–166.9), 119||25.5 (1.2–166.9), 38||1.4 (1.0–10.2), 81||2.8 (1.0–10.2), 31||1.20 (1.00–6.76), 32||1.00|
|Viral load (log10 copies/mL), median (range), nj,k||5.6 (3.0–7.6),79||7.6 (5.3–7.6),32||4.7 (3.0–7.6),29||3.0 (3.0–3.8),18|
|Fibrosis, n F ≤2/F > 2||51 : 29||13 : 19||21 : 10||17 : 0|
|HAI, n HAI ≤3/HAI >3, HAI ≥7a||28 : 50, 14||5 : 26, 10||6 : 24, 4||17 : 0, 0|
|Characteristic||Total patients||Acute infection||Total chronic||Patients with chronic hepatitis||Inactive carriers|
|A||25 : 11||8 : 6||17 : 5||8 : 2||5 : 3||4 : 0|
|D||17 : 11||0 : 2||17 : 9||1 : 3||10 : 4||6 : 2|
|F||43 : 17||25 : 5||18 : 12||11 : 4||4 : 5||3 : 3|
|Age (years), median (range), nc–e|
|A||41.5 (36.0–78.0), 36||42.5 (24–78), 14||40.5 (21.0–59.0), 22||42.5 (21.0–58.0), 10||42 (24–59), 8||37.5 (34.0–52.0), 4|
|D||43.0 (20.0–65.0), 28||27.5 (27–28), 2||45 (20–65), 26||29.5 (20.0–52.0), 4||45 (32–65), 14||49.5 (29.0–52.0), 8|
|F||36.0 (18.0–79.0), 60||33.5(18–59), 30||38.5 (20.0–79.0), 30||33 (20–79), 15||47 (27–62), 9||36 (25–58), 6|
|ALT (× normal value), median (range), nf,g|
|A||25 (3–90), 11||1.2 (1.0–7.7), 21||4.0 (1.4–7.7), 9||1.0 (1.0–1.4), 8||1.00, 4|
|D||4.45 (3.90–5.00), 2||1.2 (1.0–6.8), 26||3.3 (1.2–6.0), 4||1.7 (1.0–6.8), 14||1.00, 8|
|F||33 (1.2–166.9), 25||1.6 (1.0–10.2), 30||2.3 (1.2–10.2), 15||1.0 (1.0–5.5), 9||1.00, 6|
|Viral load (log10 copies/mL), median (range), nh–j|
|A||5.3 (3.0–7.6), 21||7.6 (5.3–7.6), 10||4.2 (3.0–6.5), 7||3.25 (3.0–3.5), 4|
|D||4.7 (3.0–7.6), 25||7.6 (7.0–7.6), 4||5.2 (3.0–7.6), 13||3.0 (3.0–3.8), 8|
|F||6.4 (3.0–7.6), 29||7.6 (6.0–7.6), 15||4.6 (3.0–7.6), 8||3.0 (3.0–3.1), 6|
|Fibrosis, n F ≤2/F >2|
|A||11 : 10||3 : 7||4 : 3||4 : 0|
|D||20 : 5||3 : 1||10 : 4b||7 : 0|
|F||18 : 12||6 : 9||6 : 3b||6 : 0|
|HAI, n HAI ≤ 3/HAI > 3, HAI ≥7a|
|A||9 : 12, 4||1 : 9, 4||4 : 3, 0||4 : 0|
|D||11 : 14, 3||2 : 2, 1||2 : 12, 2||7 : 0|
|F||8 : 20, 7||2 : 12, 5||0 : 8, 2||6 : 0|
The overall prevalence of HBV genotypes in the 128 cases analysed was as follows: genotype F, 46.9% (60/128); genotype A, 28.1% (36/128); genotype D, 21.9% (28/128); genotype C, 2.3% (3/128); and genotype B, 0.8% (1/128).
A differential distribution of genotypes with regard to the clinical course of infection was found. Among the 46 acute infections, the distribution was as follows: genotype F, 65.2% (30/46); genotype A, 30.4% (14/46); and genotype D, 4.3% (2/46). However, among the 82 chronically infected patients, a homogeneous distribution of genotypes A, D and F was observed (26.8% (22/82), 31.7% (26/82) and 36.6% (30/82), respectively), as well as a few cases of genotypes B (1.2%, 1/82) and C (3.7%, 3/82). Univariate statistical analysis showed that genotype F was significantly more common among acute infections than among the total number of chronic cases; in contrast, genotype D was barely present in this group (Table 1). For a multivariate analysis, genotype, gender and age were considered as variables. The only one that independently associated with the course of infection was genotype. Genotypes F and A were about 13-fold and eight-fold more associated with the acute course of infection than genotype D (Table 1).
For the analysis of factors associated with HBeAg status, a multivariate model including the inactive cases in the HBeAg– group was built, with genotype, gender and age as variables. As a result, only genotype remained independently associated with HBeAg status, genotypes F and A being about five-fold more associated with HBeAg+ status than genotype D (Table 1).
Additionally, in a further analysis, we included other virological and clinical variables such as viral load, liver fibrosis and HAI. As these variables partly define the chronic inactive carriers, these patients were excluded. Also, the ALT level was not considered in the analysis, because it showed a correlation with viral load (r = 0.63, 95% CI 0.44–0.77, p <0.0001). Then, high viral load levels and young ages associated with HBeAg+ status, and genotype F was about 11-fold more associated with HBeAg+ status than genotype D (Table 1).
Even though genotype distribution in relation to liver histology was not statistically significant, it was noteworthy that 91% of the chronic hepatitis cases caused by genotype F presented high necroinflammatory activity (HAI >3, 20/22). In addition, genotype F was responsible for seven of 14 cases that reached HAI ≥7, and was found in two of the four cases of cirrhosis detected (Table 2).
Mutations in three different genetic regions (ORF-S, ORF-P and BCP-pC) were studied (Table 3).
|Genetic region||ORF S (100–180 aa)||ORF P (rt39–rt205)||BCP-pC (1690–1900 nt)|
|Genotype||Mutation||No. of isolates||Description||Mutation||No. of isolates||Description||Sequences analysed||Mutation||No. of isolates (HBeAg+/−)|
|A (n = 32)||T116N||1||a||rtS202I||1||d||A (n = 14)||A1762T||4 (1/3)|
|M133T||1||b||HBeAg+/−: 7/7||G1764A||3 (1/2)|
|del (21 nt)||2 (0/2)|
|B (n = 1)||T126A||1||a, c|
|C (n = 3)|
|D (n = 27)||T/V118A||2||c||rtL180M||1||e||D (n = 19)||A1762T||9 (1/8)|
|R122Q||2||b*||rtS202C||1||e||HBeAg+/−: 4/15||G1764A||7 (1/6)|
|G130N||1||c, b||A1838G||1 (0/1)|
|S174N||2||ins T1848||1 (0/1)|
|F (n = 58)||T114S||1||b*||rtL180M||1||f||F (n = 23)||A1762T||6 (1/5)|
|S117I||1||rtM204V||1||f||HBeAg+/−: 11/12||A1762T||6 (1/5)|
|Q129L||1||a*, b*, c*||A1819T||2 (0/2)|
|Q178P||1||ins G1774||2 (1/1)|
|del T1847||1 (0/1)|
Analysis of the amino acid substitutions within the major hydrophilic region of HBsAg was performed by genotype. For genotype A, only three of the seven mutations found were previously recognized as affecting the immunoreactivity of HBsAg. For genotype D, six of 20 mutations were previously described, and for genotype F, most of the mutations detected were previously characterized.
In addition, three patients displayed antiviral resistance mutations in the polymerase gene.
We also investigated the presence of mutations associated with the HBeAg seroconversion. Despite the fact that most of the mutations were found among the HBeAg– cases, three HBeAg+ patients presented the A1762T/G1764A mutations. For genotype A, the most common mutations were A1762T/G1764A, whereas G1896A was not found. In contrast, for genotypes F and D, G1896A was the most abundant mutation, followed by A1762T/G1764A and G1899A. It is important to note that all of the genotype F and D sequences had a T in position 1858, whereas all of the genotype A samples had a C.
The distribution of genotypes described in this survey is a reflection of the demographic history of our region. The high prevalence of genotype F is in agreement with previous reports [8,18] and might be related to its autochthonous origin, as it was predominantly found among the original People of America [19–21]. The circulation of genotypes A and D is explained by the presence of Europeans who have migrated to Argentina since the colonization period [9,22]. It was noteworthy to find genotypes B and C, which were isolated from patients from Asian countries. The presence of an Asian population in Argentina is related to a more recent immigration process that occurred since 1960 with increments during the 80s and 90s.
When the distribution of genotypes was analysed with respect to the different clinical courses, we found that genotypes F and A were more associated with acute infections than genotype D, whereas genotype D was associated with chronic infections.
In order to explain these results, several proposals are discussed.
In the first place, the differences found between acute and chronic infections are not attributable to different risk factors. Although the transmission modes were unknown in many cases, sexual contact was suspected as the main route of infection, and intravenous drug use was excluded in most patients.
Additionally, a common source of acute infections by genotype F was dismissed. First, samples were taken from epidemiologically unrelated individuals. Second, samples were dated over a 4-year period (2001–2005), and most of them were distant from each other. Among the samples separated by less than 6 months, we found that many of them belonged to different subgenotypes, dismissing the existence of a common source. However, for those samples belonging to the same subgenotype, a common origin could not be rejected or confirmed. Finally, neither acute nor chronic infections grouped as separated clusters in the phylogenetic tree.
Another argument to explain the different distribution of genotypes between acute and chronic cases is that if the infections by genotype D were preferentially asymptomatic, they would be under-represented among the acute cases reported here. However, infections with genotype D have been described as acute symptomatic in other countries [23,24]. Furthermore, the existence of local variants with particular characteristics and behaviours was dismissed at least by the phylogenetic analysis of the Argentinean genotype D isolates, which were intermingled with sequences from all around the world.
With regard to the predominance of genotype F among the acute infections, we consider that this could not only be the result of the prevalence of genotype F in the general population, but also could be determined by its presence in the highly contagious HBeAg+ chronic group. In line with this, genotype D was found to be associated with HBeAg– status, and was hardly found in the acute infections. These observations might suggest a temporal variation in the distribution of genotypes, whereby acute infections could be an expression of the genotypes currently being transmitted to new hosts (mainly from the chronic HBeAg+ hosts).
Another point to be argued is the high prevalence of genotype F that we found among chronic HBeAg+ infections, as another group has found it to be associated with HBeAg– status . However, the differences between the populations analysed suggest that the earlier medical assistance received by the patients analysed here allows the detection of HBV infections in early stages, even before the HBeAg seroconversion event.
The molecular bases for HBeAg seroconversion involve different characterized mutations. Whereas, for genotype A, the A1762T/G1764A mutations were mainly found, for genotypes F and D the G1896A mutation was most common. It is noteworthy that, in spite of genotypes F and D having the ability to seroconvert by the G1896A change (that is, they display a T at position 1858 opposite to position 1896 in the stem loop of the encapsidation signal), both genotypes were not equally distributed among the HBeAg– infections. This suggests different behaviours regarding the seroconversion process.
Other analysed mutations involve the MHR of HBsAg. These variants have been reported for their clinical importance, and their prevalence could be about 6–12%  (in our survey, the prevalence was 14.9%). Even though the patients were not previously vaccinated, three of them displayed vaccine escape mutations, indicating that these variants are currently circulating in the population. It is interesting to note that the S140T mutation, related to immunotherapy escape, has been frequently found in genotype F (although patients were not treated) and mainly in the acute infections.
Amino acid substitutions in the polymerase gene related to antiviral resistance were scarce, as expected, because the analysed samples are from the pretreatment stage.
Despite the fact that genotype distribution in relation to liver histology did not have statistical support, we found that genotype F displayed higher HAIs among the chronic patients. These findings might suggest a more adverse outcome for genotype F in chronic infection. Although fewer studies have analysed the clinical characteristics related to genotype F, their findings are in line with our results [20,26].
In conclusion, this work presents, to our knowledge, the first description of the clinical characteristics related to HBV genotypes in Argentina. Furthermore, the distribution of HBV genotypes in patients with acute infection has not been previously reported in our region. According to this, it was established that genotype F is the prevalent genotype among the acute symptomatic infections in Buenos Aires city, and that it shows a tendency to cause an adverse disease outcome among the chronic cases.
Finally, more and different approaches are needed to continue unravelling the behaviour of HBV genotypes, particularly the genotype F prevalent in Argentina.
This work was supported by grants from Universidad de Buenos Aires, Consejo Nacional de Investigaciones Científicas y Técnicas and Agencia Nacional de Promoción Científica y Tecnológica.
All authors declare no conflicts of interest.