To elucidate needlestick transmission of hepatitis B virus (HBV), strains isolated from 1 physician who acquired HBV infection through a needlestick accident and 3 patients with chronic hepatitis B (donor patients A, B, and C) were tested using molecular evolutionary analysis based on full-length HBV genomic sequences. Nucleotide sequences of these isolates were aligned with 55 previously reported full-length genomic sequences. Genetic distances were estimated using the 6-parameter method, and phylogenetic trees were constructed using the neighbor-joining method. Strains isolated from patient A and the recipient pair were clustered within a closer range of evolutionary distances than were strains recovered from the recipient pair and patients B and C. Furthermore, strains from patient A and the recipient were also clustered on the S gene sequences of HBV. These results demonstrated that patient A alone was the source of direct transmission to the recipient. This approach can be used to investigate the transmission route of HBV.
Infection with hepatitis B virus (HBV) leads to a wide spectrum of liver conditions, including acute self-limited infection, fulminant hepatitis, and chronic hepatitis with progression to cirrhosis and hepatocellular carcinoma. Transmission of HBV from patients to health care workers is a serious problem for medical institutions. Accidental exposure to HBV by needlestick is a well-known route of transmission in medical staff. According to previously reported studies, the average risk of HBV infection after a needlestick accident involving HBVinfected blood ranges from 7% to 30% , which is higher than risk estimates for infection with hepatitis C virus (HCV) (2%–4%) [2, 3] and HIV infection (0.3%) .
Molecular techniques have been used to determine the transmission route of HBV infection. Intrafamilial transmission of HBV was demonstrated by using sequence analysis of mutant HBV DNA . Nosocomial spread of HBV in a hemodialysis unit  and in a health care setting  and HBV contamination of a cryopreservation tank  were confirmed by comparison of results of DNA sequencing and serotyping. Molecular biological techniques including antigenic subtyping (adw, adr, ayw, and ayr) and oligonucleotide pattern analysis may provide supporting evidence for the transmission source, but it is often difficult to discriminate between various HBV strains in a country such as Japan where similar subtypes of HBV are predominant . In the present study, we investigated a case of HBV infection acquired through a needlestick accident by performing molecular evolutionary analysis using a combination of methods, including determination of genetic distance, phylogenetic tree analysis, and bootstrap analysis based on full-length nucleotide sequences of HBV.
Patients and Methods
Study group. Three patients suspected of direct HBV transmission were enrolled in the present study. The first patient (donor patient A) was a 36-year-old man with chronic HBV infection who was admitted to our hospital because of acute exacerbation of chronic hepatitis B (table 1). The second patient (donor patient B) was a 32-year-old man with chronic HBV infection (table 1). He was admitted to our hospital at the same time as donor patient A because of acute exacerbation of chronic hepatitis B. None of the other inpatients posed a risk of HBV transmission at that time. The third patient (donor patient C) was a 57-year-old woman with chronic HBV infection who was the recipient's mother (table 1). She was receiving treatment at another outpatient department. She was in frequent contact with the recipient at his home.
The recipient of HBV infection was a 25-year-old male physician who had normal results of liver biochemistry tests and was seronegative for all markers of HBV and HCV at the time of the accidents (figure 1). He sustained an incidental needlestick injury with a 21-gauge needle on his finger after obtaining a blood sample from donor patient A. About 1 week after the first accident, he sustained another incidental needlestick injury with a 21-gauge needle on his finger after administering an iv drip infusion to donor patient B. He bled after the puncture and washed his hand immediately. He did not receive vaccination or iv hyperimmune Ig to HBV. He developed acute self-limited hepatitis B about 3 months later (figure 1). The physician had no risk factors for hepatitis B, including history of blood transfusion, injection drug abuse, and hemodialysis.
To determine who among the 3 suspected donor patients was the source of direct HBV transmission to the physician, we obtained blood specimens from the 3 donor patients and the recipient during the acute phase of HBV infection (figure 1).
Serological markers for hepatitis B. All serum samples were tested for the following: hepatitis B surface antigen by reverse passive hemagglutination assay (Fuji Rebio, Tokyo); antibody to hepatitis B surface antigen by passive hemagglutination assay (Fuji Rebio, Tokyo); hepatitis B e antigen, antibody to hepatitis B e antigen, antibody to hepatitis B core antigen, IgM antibody to hepatitis B core antigen, antibody to hepatitis A antigen, and IgM antibody to hepatitis A antigen by RIA (Dainabbott, Tokyo); and antibody to HCV antigen by second-generation EIA (Abbott Laboratories, North Chicago, IL). Alanine aminotransferase levels were also measured in each sample.
Amplification of HBV DNA by PCR analysis. All serum samples were stored at −80°C until assayed. Serum DNA was extracted from 100 µL of serum by using a DNA extractor kit (SUMITOMO, Tokyo). HBV DNA was detected by nested PCR analysis with use of the primers listed in table 2. PCR analysis was initiated by means of the hot-start technique. The amplification reaction was done in a 96-well cycler (GeneAMP9600, Perkin-Elmer Cetus, Norwalk, CT). The first round of PCR analysis was performed with an outer primer set for 35 cycles (94°C for 1.5 min, 55°C for 1 min, and 72°C for 1 min), which was followed by an extension reaction at 72°C for 7 min. The second round was performed with an inner primer set for 30 cycles and was also followed by an extension reaction. The PCR products were analyzed by electrophoresis on 2.0% agarose gels stained with ethidium bromide; an ultraviolet transilluminator was used for visualization. The standard precautions for avoiding contamination during PCR analysis were observed. A control serum sample negative for HBV was also included in each run to ensure specificity. The quantity of HBV DNA was determined by using transcription-mediated amplification and hybridization protection assay (Fuji Rebio, Tokyo).
Sequencing of full-length genomic sequences. PCR analysis with use of the primers listed in table 2 was also performed to amplify the full-length genomic sequence of HBV. PCR-amplified HBV DNA was sequenced directly according to the dideoxy method by means of a Taq Dye Deoxy Terminator cycle sequencing kit (Perkin-Elmer Applied Biosystems, Foster City, CA) with a fluorescent 373A DNA sequencer (Applied Biosystems, Foster City, CA).
Molecular analysis of HBV DNA. The nucleotide sequences of these isolates were then aligned with 55 previously reported full-length genomic sequences and 91 S gene sequences of HBV that were obtained from international DNA databases (DNA Data Bank of Japan [DDBJ]; European Molecular Biology Laboratory [EMBL], Hinxton, UK; GenBank, Bethesda, MD). Genetic distances were estimated by using the 6-parameter method , and phylogenetic trees were constructed by means of the neighbor-joining method . To further confirm the reliability of phylogenetic tree analysis, bootstrap resampling and reconstruction were carried out . These analyses were performed by using the ODEN program of the National Institute of Genetics (Mishima, Japan). HBV genotypes were classified as A to F according to previous reports [13, 14].
All 4 strains that had a nucleotide length of 3215 bases and belonged to genotype C were analyzed. A precore-stop mutation at codon 28 (TGG to TAG) was found only in the strain recovered from donor patient B. Within the core gene, mutation E to G at codon 182 was found in the strains isolated from donor patient A and the recipient, and mutation S to P at codon 183 was found in the strain recovered from donor patient C. The sequence homology between strains recovered from the recipient and donor patient A was 99.8%, the recipient and donor patient B was 97.7%, and the recipient and donor patient C was 95.9%. A phylogenetic tree was constructed based on 55 full-length genomic sequences of HBV that were obtained from DDBJ, EMBL, and GenBank, including 45 genotype C sequences (figure 2). The strains isolated from donor patient A and the recipient pair were clustered together within a closer range of evolutionary distances than were the strains recovered from the recipient pair and the other donor patients.
Bootstrap analysis was also performed to estimate the reliability of the phylogenetic tree. The aligned sequences were resampled 1000 times, and the phylogenetic trees were reconstructed for all these alignments. These results suggested that there was a close evolutionary relationship between the strain isolated from donor patient A and the strain recovered from the recipient. However, there was also a close evolutionary relationship between the strain isolated from donor patient C and the strain recovered from the recipient pair, because there was only a 1-strain difference between them. To confirm close relationships, we included 91 genotype C nucleotide sequences of the S gene of HBV in our analysis that were obtained from DDBJ, EMBL, and GenBank (including 81 genotype C sequences) and aligned them to maximize homology, and another phylogenetic tree was constructed based on the S gene sequences of HBV (figure 3). The strains recovered from donor patient A and the recipient were also closely clustered. On the other hand, with regard to the strain isolated from donor patient C and the strain isolated from the recipient pair, there was a 6-strain difference between them. These results demonstrated that the source of direct transmission of HBV to the recipient was donor patient A, not the other donor patients.
Molecular evolutionary techniques including determination of genetic distance and phylogenetic tree analysis have been used recently to determine the transmission route of viral infection. Molecular evolutionary analysis of the route of infection was first used in a study of the transmission pattern of HIV in one dental practice . After that study, HCV transmission through needlestick accidents  and vertical transmission  was investigated by means of this method. Transmission of HBV in health care settings also was analyzed by using this method [7, 18]. The occurrence of genetic variation is necessary for studying the molecular epidemiology of infectious diseases. Although HBV is a DNA virus, the reverse transcriptase activity of this virus may be responsible for its high rate of mutation. In fact, the average annual evolutionary rate of HBV strains is 10-5–10-6 mutations per site . For a virus with substantial genomic variation, the identification of strains with a high degree of genetic relatedness may be considered demonstration of a molecular epidemiological linkage between different patients infected with these strains . To our knowledge, this is the first study demonstrating HBV transmission through a needlestick accident by molecular evolutionary analysis (by use of a combination of methods, including estimation of nucleotide substitutions, phylogenetic tree analysis, analysis of nucleotide sequence diversity, and bootstrap resampling) with full-length genomic sequences of HBV.
The complete nucleotide sequence of the HBV genome contains 4 open reading frames designated S, C, P, and X. The rate of synonymous substitutions in the S gene is less than in other open reading frames . Because the P region partially overlaps with the S gene, it encodes a DNA polymerase, including a functional domain of reverse transcriptase. In a previously reported study , we demonstrated that results of HBV genotyping with use of the S gene sequence were consistent with those of genetic analysis with use of the full-length genomic sequence. Our results of phylogenetic analysis with use of the S gene sequence of HBV for investigating the relationship between strains isolated from donor patients and the recipient were also consistent with those of genetic analysis with use of the full-length genomic sequence. These findings indicate that the S gene contains nucleotide sequences with sufficient variability to distinguish between strains and therefore provides a feature essential for establishing direct transmission of HBV. Furthermore, the HBV sequence database contains a relative abundance of sequences for comparative purposes.
These results demonstrate that the nucleotide sequence of S gene is a useful tool for analyzing routes of HBV transmission by molecular evolutionary analysis. It is often difficult to discriminate between various HBV strains on the basis of short-length nucleotide sequences. The S gene sequence (with its sufficient length) and a large number of HBV strains for comparison with other strains are recommended for use in molecular evolutionary analysis, because the S gene is a comparatively conserved region as described previously .
A core and precore variant of HBV has been found to cause fulminant hepatitis in various instances such as HBV outbreaks . A precore-stop mutation at codon 28 (TGG to TAG) was found only in the strain isolated from donor patient B. Within the core gene, mutation E to G at codon 182 was found in the strains recovered from donor patient A and the recipient, and mutation S to P at codon 183 was found in the strain isolated from donor patient C (data not shown). These mutations are noteworthy because their association with fulminant hepatitis has been suggested . In the present study, the clinical course of acute hepatitis in the recipient was severe. A relationship between the severity of his clinical course and the core gene variant was suspected.
In conclusion, we report a case of HBV infection acquired through a needlestick accident in which direct transmission was determined by molecular evolutionary analysis (via a combination of methods, including determination of genetic distance, phylogenetic tree analysis, and bootstrap analysis based on the full-length nucleotide sequences of HBV). This approach is considered useful for analyzing routes of HBV transmission.