Characterization of genetic mutations in hepatitis B virus isolated from HBsAg+/HBcAb+/HBsAb-/HBV DNA + Japanese blood donors

Abstract

In Japan, a universal hepatitis B (HB) vaccination was introduced in 2016. Through continuous analysis of HBV DNA mutations, the changes in HBV prevalence before and after the introduction of the universal HB vaccine can be monitored. In this study, we conducted mutational analysis of HBV DNA in HB small protein antigen-positive/HBV core antibody-positive/HB small protein antibody-positive/HBV DNA-positive donor blood samples, with the aim of establishing a baseline prior to the introduction of universal vaccination. We successfully sequenced the full-length HBV DNA in 32 of 33 samples. Immune-escape mutations in the S protein were frequently detected in genotypes B and C. In genotype B, a 1896 nonsense mutation in the precore protein-coding region, which enhances HBV replication and is associated with HB e antigen negativity, was detected at high frequency. In genotype A, the 1858 mutation, which suppresses the 1896 nonsense mutation, was detected at a high frequency. The 1762/1764 double mutation, a risk factor for hepatocellular carcinoma (HCC), was detected frequently in genotype C. These findings provide baseline data and indicate the need for continued monitoring to assess whether universal vaccination influences mutation patterns over time.

Introduction

The hepatitis B virus (HBV) is a DNA virus belonging to the family Hepadnaviridae that causes hepatitis in humans. Based on phylogenetic analysis of its DNA sequences¹, HBV can be classified into at least eight genotypes (A–I), the distribution of which exhibits regional differences, with genotypes A, B, and C being prevalent in Japan^2,3. HBV is transmitted through blood and body fluids either vertically (mother-to-child transmission) or horizontally (blood transfusions or sexual activity). In Japan, a project aimed at preventing mother-to-child HBV transmission was initiated in 1986, resulting in the amelioration of such cases^4,5. In addition, by combining hepatitis B core antibody (HBcAb) screening with nucleic acid amplification tests for blood donations, infection through blood transfusions has decreased to 0.19 per million⁶. Therefore, new HBV infections in Japan are mainly attributed to horizontal transmission through sexual intercourse, which can also be prevented by vaccination. In 1992, the World Health Organization recommended hepatitis B (HB) vaccination, leading to the implementation of a universal vaccination program worldwide. The HB vaccine mainly targets the hydrophilic region (MHR) of the S protein in the envelope; however, mutations in this region, such as P120Q and G145R, enable the virus to escape the immune system^7–9. There are growing concerns that a universal HB vaccination may trigger an increase in vaccine-induced immune-escape mutations. However, the relationship between the HB vaccine and the incidence of vaccine-induced immune-escape mutations is unclear^10–13.

The risk of developing HBV-related liver damage differs depending on the HBV genotype. Among the aforementioned genotypes prevalent in Japan, genotype A has a low risk of HBV-related chronic liver damage but is prone to causing persistent infection¹⁴; genotype B is a risk factor for fulminant hepatitis (FH)¹⁵; and genotype C is a risk factor for hepatocellular carcinoma (HCC)^16,17. In addition, mutations that increase the risk of HBV-related liver damage have been identified. The 1762/1764 double mutation in the core promoter region increases HBV replication and induces FH^17–23. The nonsense mutation at position 1896 in the precore region suppresses the production of the e antigen (HBeAg), increases HBV replication, and is a risk factor for HBV-related liver diseases, including FH^9,23–26. HBV mutational analysis has primarily focused on HBV DNA isolated from individuals with HBV infection who have developed hepatitis or other related conditions. However, blood donors who test positive for HBsAg are likely unaware that they are infected with HBV. In such cases, although the timing of exposure of the HBsAg + blood donor to HBV is unclear, a subclinical infection followed by a persistent infection is likely to have occurred. Mutations that increase the risk of HBV-related chronic liver disease and immune escape mutations were also detected in HBV DNA isolated from HBsAg + donor blood samples⁵. This finding indicated that, in addition to HBV DNA isolated from individuals who developed symptoms, analyzing mutations in HBV DNA isolated from HBsAg + blood donor samples is crucial for gaining insight into the current HBV prevalence.

In Japan, a universal HB vaccine for infants was introduced only in 2016; therefore, the majority of Japanese population does not have antibodies against HBV. The aim of this study was to provide baseline data by examining mutations in strains isolated from HB surface antigen-positive (HBsAg+)/HB surface antibody-negative (HBsAb-)/HBcAb+/HBV DNA-positive (HBV DNA+) blood donors prior to the introduction of the universal vaccination.

Results

Full-length HBV DNA sequencing and phylogenetic analysis of HBsAg+/HBcAb+/HBsAb-/HBV DNA + donor blood samples

We quantified HBV DNA copy number in 33 donor blood samples with HBsAg+/HBcAb+/HBsAb-/HBV DNA + status using real-time qPCR (Fig. 1a). We detected HBV DNA in all 33 donor blood samples. The copy numbers are listed in Table 1. We amplified full-length HBV DNA in 32 of the 33 donor blood samples. The sample that tested negative for full-length HBV DNA was successfully PCR-amplified and sequenced for the segment of the HBV DNA that encodes the S protein, followed by analysis of the amino acid sequence of the S protein for mutations. However, the PCR product was short and insufficient for phylogenetic analysis; therefore, we performed no such analysis. Phylogenetic analysis of the 32 full-length HBV DNA sequences (Fig. 1b) revealed the following genotype distribution: 1 (3.1%) genotype A1 (Asian/African type), 6 (15.6%) genotype A2 (European type), 6 (18.8%) genotype B1 (Japanese type), 2 (6.3%) genotype B2 (Asian type), 1 (3.1%) genotype B4 (Southeast Asian type), 15 (46.9%) genotype C2 (East Asian type), and 1 (3.1%) genotype D2 (European, Bangladesh, and Indian types).

Fig. 1.

Schema of HBV structure. (a) Summary of the HBV DNA sequencing of the 33 HBsAg+/HBcAb+/HBsAb-/HBV DNA + donor blood samples. (b) Phylogenetic analysis of full-length HBV DNA.

Table 1.

Results of phylogenetic and mutational analysis of HBV whole genome sequences isolated from HBsAg positive donor blood samples.

No.	Genotype	length	HBV DNA (copies/3 µL)	PreS1/PreS2/S protein			Polymerase		X protein	precore/core protein			accession No.
				preS1	preS2	S	deletion	DRM		core promoter	precore	core protein
T2	A2	3221 bp	4163982.8			–	–	–	–	–	1858T > C		LC851032
T9	A1	3179 bp	147.4	S17_F25del	G138_Y140del	sM1K*** sF20S sR24K sI68T sC76Y sF85C sM103I sI110L sS113T	G17_T18del S198_I206del S321_F323del	–	–	–	1858T > C	cR152_D153del (2354_2359del) cS183P(2441 C > A)	LC851033
T11	A2	3221 bp	248090469.2	–	–	–	–	–	–	–	1858T > C	–	LC851034
T12	A2	3221 bp	20	–	–	–	–	–	–	–	–	–	LC851035
T24	A2	3221 bp	1517.4	–	–	–	–	–	–	–	1858T > C	–	LC851036
T26	A2	3176 bp	47.3	–	N123_R137del F141L	sF20S sS167L sL216*	V304_S318del	–	H94Y	1653 C > T	–	cS183P (2441 C > A)	LC851037
T32	A2	3221 bp	102.4	–	–	–	–	–	–	–	eM1*(1814 A > C)* 1858T > C 1862G > T		LC851038
T5	B1	3215 bp	115.8	–	–	–	–	–	I127M	1754T > G	eW28* (1896G > A)*	2002 C > T cS181P (2441T > C)	LC851443
T6	B2	3215 bp	151	W4P	–	sL21S sM133L sY161F	–	–	I127V	1752 A > G	eW28* (1896G > A)*	1915 C > G cI97L (2189 A > C)	LC851444
T7	B1	3215 bp	63	–	–	sA159V	–	rtA194T	I127T	1753T > G	–	2002 C > T cS181P (2441T > C)	LC851445
T8	B1	3215 bp	101.1	–	–	sL21S sG145A	–	–	I127M	1754T > G 1613G > A	eW28* (1896G > A)*	2002 C > T cS181P (2441T > C)	LC851446
T15	B1	3215 bp	158.6	–	–	–	–	–	–	–	1858T > C	2002 C > T cS181P (2441T > C)	LC851447
T18	B1	3215 bp	367744.3	–	–	sC76Y	–	–	I127M	1754T > G	eW28* (1896G > A)*	2002 C > T cS181P (2441T > C)	LC851448
T21	B2	3215 bp	14.9	–	–	sL95W	–	–	K130M/V131I	1762 A > T/1764G > A	1858T > C	1915T > G cI97F (2189 A > T)	LC851449
T25	B1	3215 bp	97.9	–	–	sT126A	–	–	–	–	eW28* (1896G > A)*	2002 C > T cS181P (2441T > C)	LC851450
T33	B4	3215 bp	89.6	–	–	sK122R	–	–	–	–	eG29D (1899G > A)	1915 C > G cI97L (2189 A > C)	LC851451
T1	C2	3215 bp	1.2	–	–	sLI68T sL95W	–	–	I127N K130M/V131I	1674T > C 1753T > A 1762 A > T/1764G > A	–	–	LC852214
T4	C2	3215 bp	363754086.6	–	–	–	–	–	–	–	–	–	LC852215
T10	C2	3215 bp	50.5	–	–	sF20S sV96A sL98V sI126S	–	–	–	1613G > A	–	cI97L (2189 A > C) cQ182* (2444 C > T*)	LC852216
T14	C2	3215 bp	4113.4	–	–	sF20S sL77R	–	–	K130M/V131I	1762 A > T/1764G > A	eG29D (1899G > A)	1915G > A cI97L (2189 A > C) cL100I (2198 C > A)	LC852217
T16	C2	3215 bp	410.9	–	–	–	–	–	H94Y K130M/V131I	1653 C > T 1762 A > T/1764G > A	eW28* (1896G > A)*	cL60V(2078 C > G)	LC852218
T17	C2	3215 bp	57.9	H51Q	–	sV96A	–	–	–	–	–	2159 A > G cI97L (2189 A > C)	LC852219
T19	C2	3215 bp	382.8	–	–	sF80S			H94Y K130M/V131I	1653 C > T 1762 A > T/1764G > A	–	–	LC852220
T20	C2	3215 bp	222784834.2	–	–	–	–	–	K130M/V131I	1762 A > T/1764G > A	–	–	LC852221
T22	C2	3155 bp	192.8	–	M1del S124_P142del	sI68T	G299_G318del	–	–	–	–	2002 C > A cI97F (2189 A > T)	LC852222
T23	C2	3215 bp	7614.1	–	–	sF20S	–	–	I127S K130M/V131I	1753T > G 1762 A > T/1764G > A	eW28* (1896G > A)*	–	LC852223
T27	C2	3215 bp	61739067.4	–	–	–	–	–	–	–	–	2002 C > A	LC852224
T28	C2	3215 bp	532	H51Q	–	sI68T	–	–	I127T K130M/V131I	1753T > C 1762 A > T/1764G > A	–	2159 A > G cI97L (2189 A > C)	LC852225
T30	C2	3215 bp	130.1	–	–	–	–	–	–	1766 C > G	–	–	LC852226
T31	C2	3182 bp	463,650	–	M120I *** L131_P142del	–	A312_S322del	–	–	–	eW28* (1896G > A)*	cS181P (2441T > C)	LC852227
T34	C2	3215 bp	184.3	–	–	–	–	–	V131I	1764G > A 1766 C > T	–	cI97L (2189 A > C)	LC852228
T29	D2	3182 bp	493.3	–	–	sV118T sT127P sV128A	–	–	I127L	1752 A > C 1766 C > G	–	1915T > A	LC852229
T3	A	217 bp	8.1	N/A	N/A	sS167L	N/A	N/A	N/A	N/A	N/A	N/A	LC867244

***Start codon mutation.

DRM: drug-resistant mutation, VEM: vaccine-escape mutation, N/A: not applicable.

Immune escape mutations in the preS1/preS2/S proteins

Amino acid sequence alignments are shown in Fig. 2a–d. In genotype A, mutations accumulated in the preS1/preS2/S proteins of sample T9 (Fig. 2a). The N-terminal amino acid sequence of preS1 in T9 was deleted (S17_F25del). Additionally, three amino acids were deleted from preS2 (G138_Y140del). We detected a missense mutation in which the first methionine in the S protein of T9 was mutated to lysine (sM1K). We also detected nine other missense mutations (sF20S, sR24K, sI68T, sC76Y, sF85C, sM103I, SI110L, and sS113T). Sample number T26 also showed an amino acid deletion in the preS2 protein, one missense mutation (F141L) in the preS2 protein, three missense mutations (sF20S, sA128V, and sS167L), and a nonsense mutation near the C-terminus of the S protein (sL216*), resulting in the deletion of 11 amino acids from the C-terminus. We only observed the immune-escape mutations detected for genotype A in samples with deletions, at an incidence of two out of seven cases (28.6%). In genotype B, we detected a missense mutation near the N-terminus of preS1 in one sample (W4P) (Fig. 2b). W4P/R missense mutations have been implicated in severe liver diseases, such as cirrhosis, chronic hepatitis, and HCC^27,28. In addition, seven of nine cases (77.8%) of genotype B had immune-escape mutations (sL21S, sC76Y, sL95W, sK122R, sT126A, sM133L, sG145A, sA159V, and sY161F). These immune-escape mutations were concentrated in the MHR. In genotype C, we detected the mutation H51Q, which is associated with HCC²⁹, in two preS1 proteins (Fig. 2c). Moreover, in genotype C, we observed the deletion of the first methionine in preS2 (M1del) and a missense mutation (M120I). In the two samples lacking the first methionine residue of the preS2 protein, we observed amino acid deletions in the protein (S124_P142del, L131_P142del). Unlike genotype A, genotype C of the S protein showed no tendency for immune-escape, even with preS2 amino acid deletions. However, in genotype C, missense mutations were scattered throughout the S protein similar to those in genotype A (sF20S, sI68T, sL77R, sF80S, sL95W, sL98V, and sl126S), without evidence of localization to the MHR, as observed in genotype B. We observed these immune-escape mutations in the S protein of genotype C in 8 out of 15 cases (53%). In genotype D, we found no mutations related to HCC in the preS1/preS2 proteins (Fig. 2d). In addition, we detected the immunogenicity-evading mutation sT127P in the MHR of the S protein in this strain. In the sample that tested negative for full-length HBV DNA, amino acid sequence analysis revealed the immune escape mutation sS167L (Table 1). A list of immune escape mutations is provided in Table 2, and the frequency of immune escape mutations by genotype is shown in Table 3.

Fig. 2.

Amino acid mutations in the preS1/preS2/S protein. The schematic diagram shows the amino acid sequence of the preS1/preS2/S protein. The HBV receptor, composed of amino acids 2–48 at the N-terminus of the preS1 protein, is shown in blue. The major hydrophilic region (MHR, amino acids 99–169 of the S protein) is shown in pink. (a) Alignment of amino acid sequences of genotype (A) (b) Alignment of amino acid sequences of genotype (B) (c) Alignment of amino acid sequences of genotype (C) (d) Alignment of amino acid sequences of genotype (D) Slashes marked in pink, amino acid deletion; green, missense mutation; blue, start codon mutation.

Table 2.

Mutations clinically associated with liver diseases.

	Immune escape mutations	Liver cirrhosis-associated mutations	Chronic hepatitis-associated mutations	HCC-associated mutations	FH-associated mutations
preS1		W4P	W4P	W4P H51Q
preS2		F141L M120I		F141L M120I
S	sL21S sR24K sI68T sC76Y sL77R sF80S sF85C sL95W sL98V sM103I sI110L sS113T sK122R sT126A sL126S sT127P sA128V sM133L sG145A sA159V sY161F sS167L
Core promoter				1613G > A 1653 C > T 1674T > C 1752 A > G 1753T > A/G/C 1762 A > T/1764G > A 1766 C > G/T	1754T > G 1762 A > T/1764G > A
Precore/core		1753T/1899G 2441T > C (S183P)	2441T > C (S183P)	1814 A > C 1858T > C 1862G > T 1915G > A 2002 C > A 2159 A > G 2189 A > C (cI97L) 2189 A > T (cI97F) 2444 C > T (cQ182*)	1896G > A 1753T/1896G 1899G > A
X				H94Y I127V/T/M/L I127N/S/T K130M/V131I

Table 3.

Frequency of clinically relevant mutations according to HBV genotype.

	Genotype A (n = 7)	Genotype B (n = 9)	Genotype C (n = 15)	Genotype D (n = 1)
Immune escape mutations
S	sR24K sI68T sC76Y sF85C sM103I: 28.6% sI110L sS113T sA128V sS167L	sL21S sC76Y sL95W sK122R sT126A: 77.8% sM133L sG145A sA159V sY161F	sF20S sI68T sL77R sF80S: 53% sL95W sL98V sl126S	sV118T sT127P: 100% sV128A
Liver cirrhosis-associated mutations
preS2	F141L: 14.3%		M120I: 6.7%
Chronic hepatitis-associated mutations
preS1		W4P: 11.1%
precore/core	2441T > C (S183P) : 28.6%
HCC-associated mutations
preS2	F141L: 14.3%		M120I: 6.7%
Core promoter	1653 C > T: 14.3%	1613G > A: 11.1% 1752 A > G/1753T > G/1754T > G: 55.6% 1762 A > T/1764G > A: 11.1%	1613G > A: 6.7% 1653 C > T : 13.3% 1674T > C: 6.7% 1753T > A/G/C: 20% 1753/1896: 6.7% 1762 A > T/1764G/A: 46.7% 1764G/A: 6.7% 1766 C > G/T: 13.3%	1752 A > C 100% 1766 C > G 100%
Precore/core	1814 A > C (eM1*) : 14.3% 1858T > C: 71.4% 1862G > T: 14.3% 2441T > C (S183P) : 28.6%	1858T > C: 22.2% 1896G > A: 55.6% 1899G > A: 11.1% 1915 C > G: 22.2% 2002 C > T: 66.7% 2189 A > C (cI97L) : 22.2% 2189 A > T (cI97F) : 11.1% 2441T > C (cS181P) : 55.6%	1896G > A: 20% 1899G > A: 13.3% 1915G > A: 6.7% 2002 C > A: 6.7% 2159 A > G: 13.3% 2189 A > C (cI97L) : 40% 2444 C > T (cQ182*): 6.7% 2441 A > C (cS181P) : 6.7%	1915T > A: 100%
X	H94Y: 14.3%	I127V/T/M: 55.6% K130M/V131I: 11.1%	H94Y: 13.3% I127N/S/T: 20% V131I: 6.7% K130M/V131I: 46.7%	I127L: 100%
FH-associated mutations
preS1		W4P: 11.1%	H51Q: 13.3%
Core promoter		1754T > G: 33.3% 1762 A > T/1764G > A: 11.1%	1762 A > T/1764G/A: 46.7%
Precore/core		1896G > A: 55.6%	1896G > A: 20%

Core promoter and X protein mutations in HBsAg+/HBcAb+/HBsAb-/HBV DNA + donor blood samples

The core promoter regulates the expression of the precore/core protein, including HBeAg^23,30. The mutations observed in the core promoter are listed in Table 1. Certain core promoters encode the X protein, and mutations within these promoters result in amino acid mutations in the X protein²¹ (Fig. 3). In genotype A, we observed the core promoter mutations 1653 C > T and H94Y in the X protein (1/7, 14.3%). In genotype B, X protein exhibited the core promoter mutations 1754T > G and I127M (3/9, 33.3%). In addition, we found the core promoter mutations 1752 A > G and 1753T > G, which resulted in the I127V and I127T mutations in the X protein, respectively, in one case each (11.1%). In genotype B, we frequently observed the 1752 A/1753T/1754T mutations linked to I127V/T/M (5/9, 55.6%). Moreover, we detected the 1762 A > T/1764G > A mutation in the core promoter linked to the K130M/V131I mutations in the X protein in one out of nine cases (11.1%). Furthermore, we detected the 1613G > A mutation that does not cause a missense mutation in the X protein (1/9, 11.1%). In genotype C, we detected the core promoter mutations 1753T > A/G/C, which cause the I127N/S/T mutations in the X protein, in 3 out of 15 cases (20%). We frequently detected the core promoter double mutation 1762 A > T/1764G/A along with K130M/V131I in the X protein, in 7 out of 15 genotype C cases (46.7%). When the 1764G/A (V131I) mutation alone (1/15, 6.7%) and the 1766 C > G/T mutation without amino acid changes in the X protein (2/15, 13.3%) were included, the incidence of the 1762/1764/1766 mutation in genotype C reached 10 out of 15 cases (66.7%). We also detected the 1653 C > T promoter mutation causing the H94Y mutation in the X protein in 2 out of the 15 cases (13.3%). In addition, we detected the core promoter mutations 1613G > A (1/15, 6.7%) and 1674T > C (1/15, 6.7%), not leading to missense mutations in the X protein. In genotype D, we detected the 1752 A > C mutation (1/1, 100%), which causes I127L, and the 1766 C > G mutation (1/1, 100%), not causing any missense mutations. A list of mutations associated with liver disease in the core promoter and X protein is provided in Table 2, and the frequency of mutations by genotype is shown in Table 3.

Fig. 3.

Amino acid mutations observed in X protein. Alignment of the amino acid sequence of the X protein. Green symbols indicate missense mutations. The schematic diagram represents the amino acid sequence of the X protein, including the X-box binding domain, BH-3-like motif, and zinc-finger motif. The left side of the box represents the N-terminus of the amino acid sequence, whereas the right side represents the C-terminus of the amino acid sequence.

Precore/core mutations in HBsAg+/HBcAb+/HBsAb-/HBV DNA + donor blood samples

The precore region, starting at HBV DNA position 1814, encodes HBeAg, whereas the core protein is encoded from position 1901. Precore mutations spanning the 1858−1896 region increase the amount of HBV DNA in the serum^23,30. In addition, a nonsense mutation in the precore region at position 1896 inhibits HBeAg expression. In genotype A, we did not detect the 1896G > A nonsense mutation; however, we found a nonsense mutation (1814 A > C, eM1*) in which the first methionine of HBeAg was replaced by a stop codon (1/7, 14.3%) (Fig. 4a). Additionally, we detected the 1858T > C (5/7, 71.4%) and 1862G > T (1/7, 14.3%) mutations without amino acid changes. Furthermore, we detected cR152_D153del (2354_2359del) in the core protein from genotype A samples. These two amino acids are characteristic of genotype A, and their deletion makes genotype A similar to genotypes B and C. The S183P (2441T > C) mutation at the C-terminus of the core protein is a risk factor for transition from an inactive carrier to chronic HB (CHB) or liver cirrhosis (LC)^31–33; we detected it in two of the seven cases of genotype A (28.6%). In genotype B, we observed a 1896G > A (eW28*) nonsense mutation in five of the nine cases (55.6%) (Fig. 4b). We also found the 1899G > A (eG29D) missense mutation in one of the nine cases (11.1%). Further, we observed the 1858T > C mutation (2/9, 22.2%). In the core protein from genotype B cases, we identified two 2189 A > C (cI97L, 22.2%), one 2189 A > T (cI97F, 11.1%), and five 2441T > C (cS181P, 55.6%) missense mutations. In addition, we detected the 2002 C > T (6/9, 66.7%) and 1915 C > G (2/9, 22.2%) mutations without amino acid changes. In the precore region of genotype C, we observed the 1896G > A (eW28*) nonsense mutation (3/15, 20%) and 1899G > A (eG29D) missense mutation (2/15, 13.3%) (Fig. 4c). Additionally, in the core protein from genotype C cases, we detected the 2078 C > G (cL60V) (1/15, 6.7%), 2189 A > C (cI97L) (6/15, 40%) and 2441 A > C (cS181P) missense mutations (1/15, 6.7%). Furthermore, we observed the 2444 C > T (cQ182*) nonsense mutation (1/15, 6.7%). Moreover, we detected the 1915G > A (1/15, 6.7%), 2002 C > A (1/15, 6.7%), and 2159 A > G (2/15, 13.3%) mutations not resulting in any amino acid mutations. In genotype D, we detected the 1915T > A mutation in the core protein (Fig. 4d). A list of variants associated with liver disease in the precore/core region is shown in Table 2, and the frequency of variants by genotype is shown in Table 3.

Fig. 4.

Amino acid mutations observed in precore/core protein. The schematic diagram shown above the alignment represents the amino acids of the precore/core protein. (a) Alignment of amino acid sequences of genotype (A) (b) Alignment of amino acid sequences of genotype (B) (c) Alignment of amino acid sequences of genotype (C) (d) Alignment of amino acid sequences of genotype (D) Slashes marked in pink, amino acid deletion; green, missense mutation; blue, start codon mutation.

Discussion

In this study, we successfully detected HBV DNA in all 33 HBsAg+/HBcAb+/HBsAb-/HBV DNA + donor blood samples. Full-length HBV DNA was sequenced using the Sanger method for 32 of the 33 samples, and the presence of disease-related mutations in the amino acid sequences of the preS1/preS2/S, X, and precore/core proteins was analyzed.

In this study, genotype C and genotype B were detected at similar frequencies, and genotype A was detected less frequently, which is consistent with the results reported in 2009³⁴. The frequency of genotype A2 was greater than that of genotype A1, which was also consistent with the results of a previous study in 2006. Similarly, as reported in the previous study, genotype B1 was detected at a greater frequency than genotype B2. This study revealed that the frequency of each genotype in HBsAg + donor blood samples in Tokyo, Chiba, and Kanagawa has not changed significantly over the past 10 years.

The HBV envelope comprises three types of proteins: a large protein composed of preS1/preS2/S proteins, a medium-sized protein comprising the preS2/S proteins, and a small protein composed of the S protein. Amino acids 2–48 of preS1 bind to the host receptor, sodium-taurocholate cotransporting polypeptide³⁵. The W4P mutation in preS1 is associated with severe liver disease, such as chronic hepatitis, cirrhosis, and HCC; notably, only male carriers have been reported^27,28. In the present study, W4P was detected in one case of genotype B; however, owing to ethical restrictions, the sex of the patient was unknown. The first methionine missense mutation (M120I) and F141L in preS2 are also associated with chronic liver diseases, such as cirrhosis and HCC^36–39. In this study, M120I deletion and F141L were detected in genotypes A and C. However, all of these strains exhibited amino acid deletions, and none of these mutations occurred in isolation. Amino acid mutations in the S protein may be associated with immune escape. In particular, HB vaccines target the MHR of the S protein, and amino acid mutations in this region are associated with immune escape⁸. In addition, mutations in the amino acid sequence from the first methionine to just before the MHR of the S protein can result in HBsAg retention within the Golgi apparatus and prevent its release or induce failure to cross-react with antibodies owing to changes in antigenicity. This can lead to false negative HBsAg test results even when HBV DNA is detectable^40,41. In the present study, in addition to amino acid mutations in the MHR, we also analyzed amino acid changes that have been reported to affect the results of clinical tests, such as those related to immune escape^38,42–52. In genotype B, immunogenicity-evading mutations were scattered throughout the S protein, whereas in genotype C, such mutations were concentrated at the N-terminal region of the S protein, including the MHR, and were not observed in the C-terminal region. The effects of amino acid mutations in the S protein on the intracellular vesicular transport of the HBV envelope and the formation of viral particles are not fully known; however, the differences between genotype C, which is a risk factor for HCC in Japan, and genotype B, associated with FH in Japan, is particularly intriguing. Future studies should investigate the causal relationship between amino acid mutations in the S protein and the amount of HBsAg secreted in vitro. In genotype D, immune-escape mutations were observed only in the MHR. Genotype D was observed in only one case in this study. However, the occurrence of immune-escape mutations in other locations than the MHR in genotypes B and C is noteworthy.

In the mutational analysis of the preS1/preS2/S protein, two strains with particularly notable mutations were identified to belong to genotype A. One of the strains with a deletion in preS1, in which the first methionine in the S protein was mutated to lysine, also had a deletion in preS2. Additionally, the S protein exhibited an accumulation of immune escape-related mutations. However, the mechanism by which changing the first methionine to lysine alters its antibody binding ability remains unclear. Exploring the structural morphology of the viral particles would be very interesting. We also detected a case of a genotype A strain with deletions at the C-termini of the preS2 and S proteins; this strain also carried several immune-escape mutations in its S protein. Additionally, the preS2 protein of this strain carried the F141L mutation, which is a risk factor for HCC³⁶. Notably, in genotype A, immune-escape mutations in the S protein occurred only in strains with these deletions, but not in other strains.

Core promoter mutations are associated with HBV-related liver damage^{16,18,20,23,30,53–55}. In Japan, genotype C is associated with a high risk of HCC, and almost all patients with HCC in Japan harbor genotype C^16,54. Therefore, mutations have possibly accumulated in the core promoter of genotype C isolated in Japan, prompting a detailed analysis in this study. The mutation at position 1613 decreases HBeAg expression but increases HBV DNA levels and thus is a risk factor for HCC; however, in the present study, we detected it in only one case each for genotypes B and C^25,54. Compared with the 1613 single mutation, the 1613/1653 double mutation increases the HCC risk^23,54 but we did not detect in this study. The 1653 mutation, which is a risk factor for HCC, was observed in one case of genotype A and two cases of genotype C; however, in this study, we did not detect the 1753/1899 double mutation that increases the risk of cirrhosis. The 1753 mutation is a risk factor for alcoholic liver disease and HCC^16,18; in this study, we observed it in one case of genotype B and two cases of genotype C. The 1762/1764 double mutation associated with alcoholic liver disease and HCC^16,18 was found in genotypes C and B. The incidence of the 1762/1764 double mutation in genotype C was strikingly higher than that in genotype B, and may be a major risk factor for HCC in genotype C cases in Japan. In this study, we rarely observed core promoter mutations in genotype A, but were frequently observed in genotypes B and C. The low incidence of core promoter mutations in genotype A was consistent with previous studies examining HBsAg-positive donor blood samples⁵.

Precore mutations at the region spanning 1858−1899 in the HBV DNA are associated with increased serum HBV DNA levels and an increased risk of HBV-related liver damage^19,20,22,33. We found the 1896 mutation to be more frequent in genotype B. In Japan, genotype B is known to be a risk factor for severe hepatitis and FH¹⁵, and this could be attributed to the frequent detection of the 1896 mutation in genotype B. The simultaneous occurrence of the 1762/1764 double mutation in the core promoter along with the 1896 mutation increases the risk of FH because of the enhanced replication of HBV^22,23,30. Although we did not observe this in genotype B, we detected it in two cases of genotype C. The simultaneous occurrence of 1753 and 1896 mutations in the core promoter increases the risk of developing FH or ALF^20,26,30; however, we observed this in only one case of genotype C. The occurrence of the 1858 mutation confers a protective effect that suppresses the 1896 mutation. In this study, we detected the 1858 mutation at a high frequency in genotype A. Owing to its high frequency, the 1896 mutation rarely occurs in genotype A despite the occurrence of HBeAg-negative-related mutations in genotype A. The 1858 mutation in genotype A associated with HBeAg-negative is induced by a nonsense mutation in the first methionine of the precore protein. In this study, we detected this mutation in one case of genotype A. We detected the 1899 mutation, a risk factor for HCC⁵⁶, in one case each of genotypes B and C.

We found the HCC-related mutation cI97L/F in the core protein^33,57 in genotypes B and C but not in genotypes A and D. The incidence of cI97L/F in the core protein of genotype C was similar to that in genotype B. We detected the cS183P mutation, which is associated with progression from inactive carrier to CHB and LC^31,32, at the C-terminus of the core protein in genotype (A) In Japan, genotype A cases often follow a relatively mild course, and patients often become inactive carriers¹⁴; thus, careful observation of the patient may be necessary in case of a cS183P mutation. The 2002 and 2159 mutations detected in this study, which do not involve amino acid sequences in the core protein, are also risk factors for HCC⁵⁸. The 2002 mutation was particularly common in genotype (B) Furthermore, all such mutations in genotype B accompanied cS181P at the C-terminus of the core protein; cS181P is also a risk factor for HCC³³. The significance of the simultaneous occurrence of the 2002 mutation and cS181P is unknown; however, this relationship needs to be elucidated in future in vitro studies. We also observed the 2002 mutation in genotype C, but it was not accompanied by mutations at the C-terminus of the core protein. Mutations at the C-terminus of the core protein were also observed in genotype C, but their incidence was substantially lower than that in genotype B.

This study had some limitations. Although we detected several immune-escape mutations and mutations that increase the risk of HCC in HBsAg+/HBsAb-/HBcAg+/HBV DNA + donor blood samples, the Japanese Red Cross Society was unable to provide any medical information other than blood type owing to ethical restrictions. Consequently, investigating the relationship between mutations and diseases in greater detail was not possible. In addition, because of ethical restrictions, we were unable to perform any experiments on the provided donor blood samples other than the isolation and sequence analysis of HBV DNA. In the future, we will attempt further experiments to investigate the effects of the identified mutations in donor blood.

In conclusion, we sequenced full-length HBV DNA using samples from Japanese HBsAg+/HBsAb-/HBcAg+/HBV DNA + blood donors to identify immune-escape, core promoter, and precore/core mutations associated with chronic liver injury. Immune-escape mutations were particularly common in genotypes B and C. Mutational analysis revealed that mutations that increased the HCC risk were relatively common in both genotypes B and C; however, the incidence of the 1762/1764 double mutation was substantially higher in genotype C than in B. Several studies have reported HCC risk factors linked to clinical symptoms; however, the 1762/1764 double mutation may be the most significant risk factor. Similarly, mutations that increase the risk of FH were also found at high frequencies in both genotypes B and C; however, the 1896 mutation was particularly prominent in genotype B. We also detected multiple mutations in HBV isolated from asymptomatic carriers that were associated with the risk of HBV-related liver damage. Understanding HBV mutations in asymptomatic carriers unaware of their infection status is crucial for determining the prevalence of the virus. For disease progression risk management and hepatocellular carcinoma surveillance in asymptomatic carriers, mutation analysis using full-length HBV DNA is recommended if HBV DNA is positive in blood donation. The results of this study highlight the need for continuous monitoring of HBV mutations for the identification of emerging mutations and individuals at risk of transitioning from an inactive carrier state to chronic stages, and implementing effective measures to prevent future transmission. Continuous monitoring is essential to provide baseline data for this study and to evaluate whether universal vaccination affects mutation patterns over time.

Methods

Source of HBsAg+/HBcAb+/HBsAb-/HBV DNA + donor blood samples

The 33 HBsAg+/HBcAb+/HBsAb-/HBV DNA + donor blood samples were obtained from the Japanese Red Cross Society, the only organization providing a nationwide blood donation service in Japan. These samples were donated by individuals in Tokyo, Chiba, and the Kanagawa prefecture from March to August 2019. In Japan, individuals who have not traveled abroad in the past 4 weeks, are in good health, and free from infectious diseases are eligible to donate blood. For ethical reasons, medical information, such as age, sex, and HBV treatment history of blood donors, is not provided. Molecular analysis of the blood samples was performed according to the flowchart shown in Fig. 1A. This study was approved by the Research Ethics Committee of the University of Tokyo (No. 2022-62-1227). Written informed consent was obtained from all participants. This study was conducted in accordance with the Helsinki Declaration.

HBV DNA extraction and quantification using qPCR

HBV DNA was extracted from plasma using SMITEST EX-R&D (GS-J0201, Medical & Biological Laboratories Co. Ltd., Nagoya, Japan) according to the manufacturer’s instructions. The samples were incubated at 55 °C using an Eppendorf ThermoMixer C (Eppendorf, Hamburg, Germany) with stirring at 2000 rpm. DNA pellets were dissolved in 20 µL of PCR-grade distilled water. The primer sets and probes used for the qPCR have been previously described⁵. qPCR was performed using the THUNDERBIRD Probe qPCR Mix (TOYOBO Co., Ltd., Osaka, Japan) in the CFX Connect (Bio-Rad, Hercules, CA, USA). The program comprised an initial incubation step at 95 °C for 1 min, followed by 50 cycles at 95 °C for 10 s, and 55 °C for 30 s.

Amplification of HBV DNA for sequencing

The primers used for target HBV DNA amplification have been previously described⁵. Full-length HBV DNA was amplified as follows: the first PCR program comprised an initial incubation at 95 °C for 1 min, followed by 30 cycles at 98 °C for 10 s, 60 °C for 5 s, and 68 °C for 20 s, with a final extension at 68 °C for 1 min and a hold at 15 °C indefinitely. The second PCR program consisted of an initial step at 95 °C for 1 min, followed by 25 cycles at 98 °C for 10 s, 60 °C for 5 s, and 68 °C for 20 s, with a final extension at 68 °C for 1 min and a hold at 15 °C indefinitely. A partial sequence encoding the S protein was amplified using the following program: an initial incubation at 95 °C for 1 min, followed by 30 cycles at 98 °C for 10 s, 60 °C for 5 s, and 68 °C for 10 s, with a final extension at 68 °C for 1 min and a hold at 15 °C indefinitely. The second PCR program consisted of an initial incubation at 95 °C for 1 min, followed by 25 cycles at 98 °C for 10 s, 60 °C for 5 s, and 68 °C for 10 s, followed by an extension at 68 °C for 1 min and a hold at 15 °C indefinitely.

Sequencing, alignment, and phylogenetic analysis

The PCR products were purified using a PCR Purification Kit (Qiagen, Hilden, Germany) and subjected to direct sequencing using the FASMAC software (Nucleics, Sydney, Australia). The primers used for the direct sequencing have been previously described⁵. The sequence data were analyzed using ATGC-MAC Ver. 9.0.1 (GENETYX Corp., Tokyo, Japan) and GENETYX-MAC Ver. 22.0.1 (GENETYX Corp.). The text files for alignment were converted to the FASTA format using MAFFT Ver. 7 (https://mafft.cbrc.jp/alignment/software/). Phylogenetic analysis was performed using the neighbor-joining (NJ) method with Molecular Evolutionary Genetics Analysis (MEGA) Version 11.0.13⁵⁹.

Author contributions

A.S. contributed to specimen processing; data acquisition and analysis; and manuscript writing and revision. K.T. provided the technology used in this study. T.T. contributed to data acquisition. K.A. contributed to specimen processing and data acquisition. F.N., K.I., and E.A. processed the specimens. H.Y. conceived the study. All authors have read and approved the final version of the manuscript.

Data availability

All HBV DNA sequences identified in this study are registered in DDBJ, and sequencing data can be accessed using the accession number described in Table 1.

Declarations

Competing interests

The authors declare no competing interests.