The possible role of S gene mutations of hepatitis B virus in intrauterine transmission

Background: Many hepatitis B virus (HBV) substances could inevitably enter fetuses and occurred neonatal intrauterine transmission. HBV often occurs mutation, especially S gene, and may lead to different outcomes on intrauterine transmission. We explored the associations between HBV S gene mutations of hepatitis B surface antigen positive (HBsAg-positive) mothers and intrauterine transmission. Methods: A total of 399 HBsAg-positive mothers and neonates were recruited and their general demographic information was collected between June 2011 and July 2013. The mothers with HBV DNA levels ≥ 10 6 IU/ml were selected, 22 mothers whose neonates occurred HBV intrauterine transmission were in the HBV intrauterine transmission group (GT) and 22 mothers were randomly selected from the remaining controls were in the control group (GC). Maternal whole-genome HBV DNA was extracted, amplified, cloned, and sequenced. Obtained sequences were adjusted, genotyped, and analyzed for mutation rates. A case-control study was designed to analyze the relationship between mutations in the S gene of HBV and intrauterine transmission. Results: Fifty-five neonates were found to have experienced intrauterine transmission (13.78%). Genotype B (4.55%), genotype C (88.64%) and inter-genotype B/C (6.81%) were found in the 44 HBsAg-positive mothers. The mutation rates of the S gene, in both genotypes B (0.58% vs 1.41%, P = 0.040) and C (7.56% vs 14.71%, P 0.001), were lower in group T than in group C. Missense substitutions such as L84I, P47S, K10Q, A41P, M133L, A60V, and I42T only existed in group C. The mutation rates of G73S, I126T, and I126S in group C were higher ( P < 0.001, P < 0.001, P = 0.010). Deletions occurred in the S gene. <

The incidence of intrauterine transmission in China is 3.00%-43.33%,(4-6) of which 80%-90% of neonates (7,8) develop chronic HBV infection after one year of age. In the past few decades, scholars have done much research to ascertain the complex mechanisms, but much remains unclear. Both the load and the characteristics of the virus may affect this process. (9,10) HBV is a small, enveloped 3.2-kb DNA virus with four open reading frames (ORFs): S gene, C gene, P gene, and X gene. (11) It is classified into eight genotypes (A to H). (12)(13)(14)(15) The genotypes of B and C, which primarily transmit vertically, are most common in China; the genotype B is mainly found in central and southern China, whereas genotype C predominates in the northern areas. (16,17) HBV has a higher mutation rate than other DNA viruses, owing to its lack of a proof-reading function for viral polymerase and the greatly increasing pressure of host immunities which cause sequence heterogeneity. (18,19) So far, published data on the S gene, including the Pre-S1, the Pre-S2, and the S region,(11) which changes hepatitis B surface antigen (HBsAg) antigenicity, seems to be highly variable. (20) It is known that mutations of the S gene can directly affect the abilities of the HBV infection and its immunoresponse to hepatitis B vaccines. The "α" region (aa 124-147), one of the major mutated regions in the S region, as well as the Major Hydrophilic Region (MHR)(aa 32-76 and aa 100-160), are highly valuable in HBsAg detection and virus clearance as the most important functional epitopes. (21,22) It seems more likely that the "mirror changes" of the genetic functions occur when overlapped with the reverse transcriptase (RT) region.(23) These mutations give rise to HBV immune escape, occult infection, resistance to drug therapy, and tolerance of immunoprophylaxis. (24)(25)(26) Presently, little reports (27)(28)(29)(30) showing that some S gene mutations are related to intrauterine infection. This study concentrates on the characteristics of the virus in neonates at birth, constitutes the first comprehensive analysis of the relationship between different maternal mutations of different genotypes (mutation hotspots and non-hotspots, and deletions) and intrauterine transmission before neonatal immunoprophylaxis. Further initially investigate the possible role of meaningful mutations in intrauterine transmission.
There is hope that we will find virological evidence of intrauterine transmission. It will point us in the right direction for utilizing timely, effective, and targeted immunoprophylaxis to reduce the risk of HBV infection in neonates. The face-to-face interviews and electronic medical records were obtained. The study was reviewed and approved by the Ethics Committee of Shanxi Medical University, Taiyuan, China, and conducted in accordance with the Declaration of Helsinki. Informed written consent was obtained from all participants. Three ml maternal peripheral blood before delivery within 24 hours and three ml neonatal femoral venous blood within the 24 hours after birth (before giving the HBIG and hepatitis B vaccine) were collected.

Serological markers and serum HBV DNA assays
Serological markers were detected by electrochemiluminescence immunoassay (ECLIA) kits (Roche diagnostics GmbH, Germany) and expressed as a COI (COI: cut-off index).
Serum HBV DNA levels were tested by a real-time PCR-TaqMan kit (DAAN Gene Co. Ltd., Sun Yat-sen University, Guangdong, China) and expressed as IU/ml.
There is no exact definition of intrauterine transmission worldwide. We want to explore the characteristics of the HBV in the fetus as much as possible to achieve the purpose of studying the mechanism of intrauterine transmission. The definition we have chosen that excluded the effects of some factors after birth on the study, even if it was not possible to completely exclude it. HBsAg-positive (> 1.00 COI) and/or HBV DNA-positive (> 200 IU/ml) in neonatal femoral venous blood that within 24 hours of birth, before breastfeeding, and before giving the HBIG and hepatitis B vaccine was defined as HBV intrauterine transmission.(5)

HBV DNA extraction and amplification
We selected mothers with HBV DNA levels ≥ 10 6 IU/ml whose neonates were affected by  (Table 1).

HBV DNA genotyping
The DNAStar software, SeqMan software (DNASTAR, Madison, WI, USA) (31) and Mega 6.0 software package were used to edit, splice, and compare with reference sequences downloaded from NCBI by constructing Neighbor-joining (NJ) phylogenetic trees, respectively. Kimura's two-parameter model calculated distances among the sequences.

Sequence analysis
A nucleotide substitution which means contrasted with the reference sequence. The mutation rate of the S gene was calculated by the number of mutation sites dividing the total sites, and more than ten percent were defined as mutation hotspots. Finally, a casecontrol study was used to analyze the association between S gene mutations and intrauterine transmission. GT was set as the case group and GC was set as the control group.

Statistical analysis
Data was analyzed with the SAS statistical package (Version 9.4, SAS Institute Inc, Cary, North Carolina, USA). Data for continuous variables is presented as mean ± standard deviation (SD) or as median and range, and differences between data were analyzed by a student's t test or Wilcoxon signed-rank test. The dichotomous data was compared by a Chi-square test or Fisher's exact test. Odds ratios (OR) and 95% confidence intervals (CI) were estimated using unconditional logistic regression to analyze the relationship between maternal factors and HBV intrauterine transmission. P < 0.05 was considered statistical significance for the test. years) in GC (t = 2.578, P = 0.011, Student's t test). In the end, 44 HBsAg-positive mothers whose HBeAg tests were all positive were selected and the general characteristics were summarized in Table 2. The median ages of the 44 mothers were 24.0 years and 26.5 years in GT and GC, respectively (20 to 45 years)(P = 0.015). One of the mothers in GC was 45 years old. There was difference in the mode of delivery distribution between the two groups (P = 0.001), but no significant statistical difference was observed in other characteristics of the mothers and neonates (P > 0.05).

HBV genotypes of sequences
A total of 442 HBV clones were obtained including 220 clones from GT and 222 from GC.
The mean ± SD (range) number of clones per sample was 10 ± 1.3 (6 to 14). There was no difference between the two groups (10 ± 1.7 vs 10 ± 0.4, P = 0.814). The genotype C accounted for the largest proportion of the clones, about 88.64% (39/44). Two mothers (4.55%) of genotype B and three mothers (6.81%) of inter-genotype B/C were found as well ( Figure 1). No relationships between genotypes and intrauterine transmission were found (Fisher's exact test, P = 0.622 ).

Mutations of S gene
In genotype B, the mutation rate of GT was 0.58% (7/1203), lower than that of the GC were only in GT. The mutation rate of A90V in GT was significantly higher than in GC (P < 0.001), but I68T was lower than in GC (P = 0.310) ( Table 3). The mutations in genotype C below 10% were analyzed (Table 4). A60V and I42T only occurred in GC. The mutation rates of G73S, I126T, and I126S were all higher in GC (P < 0.001, P < 0.001, P = 0.010).
The mutation rate of L98V was higher in GT, but no great statistical significance was observed (P = 0.128). Two mutations were found in the RT region that overlapped with the S region: C446G and T531G.

Deletions of S gene
Deletions were found in the Pre-S1 region. Eight clones in GT began at nt 2847-2850 (12 to 18 bp) and two clones began at nt 3047 (39 bp). Ten clones in GC began at nt 2847-2849 (12 to 21 bp) and ten clones began at nt 3016 (183 bp). Deletion rates in GC were not statistically higher than GT (4.5% vs 9.01%, χ 2 = 3.48, P = 0.062).

The possible role of A90V in intrauterine transmission
We placed the S gene mutation hotspots as well as variables of the factors that were associated with intrauterine transmission into multivariate logistic regression models.
Maternal HBeAg tests were equivalent between the two groups so that possible influence on intrauterine transmission could be eliminated. The study demonstrated that maternal mutation A90V could associated with a greater risk of intrauterine transmission (Table 5).
It was observed that median neonatal HBsAg with maternal mutation A90V was higher than with the non-mutation (1.43 COI vs 0.65 COI, Z = 2.30, P = 0.022, Wilcoxon signedrank test).

Discussion
Neonates born to HBsAg-positive mothers are more at risk of immunity failure and chronic infection with HBV, even after immunoprophylaxis. Many substances still inevitably enter fetuses, although the placental barrier can block some maternal HBV antigens. These substances may jeopardize neonatal immunity and cause lifelong dysfunction of the immune system.(32) These special groups have drawn more attention from physicians and public health experts about whether they were affected by intrauterine transmission or not.
HBV has been constantly evolving and unpredictably mutating in the process of copying.
HBV genotypes appear during this long-term accumulated mutation process. The major genotype among the forty-four mothers recruited from Taiyuan was C (88.63%), which was consistent with the distributions reported in northern China. Current reports (33)  infection, which is also unfavorable for neonates. Mothers with the mutation A90V need to pay more attention because they may be prone to intrauterine transmission, and the same with L85F. We recommend them to take antiviral drugs during pregnancy to reduce HBV DNA, cut down the proteins expressed which could enter the fetus. Other mutations that mainly found in GC are indicate the changes in the structure of HBsAg caused by the mutation affect its assembly and secretion, making it unable to be expressed normally.
Moreover, deletions in this area were found and may change HBV sequences, possibly causing interference or termination of HBsAg expression at the transcriptional level. (37) Even though no difference of deletion rate was found, it was higher in the GC still had clues.
The "α" region and the Major Hydrophilic Region (MHR), are two hydrophilic regions playing leading roles in detection of HBsAg and the clearing of viruses. (38) The mutations P47S, A41P, M133L, A60V, G73S, I42T, and I126T/S are found in GC. The mutation T531G overlapped with RT region. Some scholars have pointed out that the mutations in the RT region overlapping with the S region may cause changes in the epitope of the S protein(34) and may alter the antigenicity, immune recognition, and kinetics of viral replication, especially in "α" determinants.(10) Previous studies have held views that the mutation I126T affects the structure of HBsAg, that M133T is more inclined to appear in HBV occult infection and HCC, (39) and that G145R correlatives with immune escape. (22,40) These views provide some clues for our research. Maternal mutations in these areas that viral particles may not easily enter can cause neonates to lose the ability to assemble or normally secrete HBsAg, even if the viral particles enter them. Neonatal HBsAg tests in GC are still negative, which means HBsAg is lower than one COI; however, neonatal HBV DNA is still at a low level of replication and may become occult HBV infection in the future, which also requires our attention. For mothers, although the HBV DNA are at high levels, their expressed HBsAg structures have changed. Altered antigens have impaired the formation of the HBsAg-anti-HBs complex and have inhibited the immune response, resulting in immune escape. It is easier for this to turn into HBV chronic infection, which eventually leads to HCC, for it cannot be completely eliminated. Our baseline data also shows that HBV was detected earlier in the mothers.
It should be noted that although there were differences in maternal age between the two groups in Table 2, it may be due to the fact that a 45-year-old mother in GC. Meanwhile, we found that the mother had only two mutation sites in the S gene, which were A60V and I42T (non-hotspots), which indicated that the high mutation rates in GC was not because of the age. There are still some limitations: we only explored the S gene while the wholegenome mutations played their part in intrauterine transmission together. The results will be beneficial in guiding our later research to further explore the influence of mutations in other ORFs on intrauterine transmission.

Conclusions
In conclusion, this study has comprehensively explored the relationships between S gene mutations and intrauterine transmission. It was speculated that the mutation of A90V may be closely correlated with intrauterine transmission. L85F, I4T, and F80S may be useful biomarkers for predicting intrauterine transmission. L84I, A41P, M133L, P47S, A60V, and I42T were also first reported as biomarkers in this study. G73S and I126T/S may reduce the incidence of intrauterine transmission. It must be noted that this study was significantly different from the others. We mainly studied the genotype C and different types of mutations. The number of cloned sequences was somewhat larger and may have caused the different results we found. We may need to further confirm our discoveries in the genotype B. Availability of data and materials: The datasets generated and analyzed during the current study are not publicly available due the data will continue to do genetic and epigenetic analysis but are available from the corresponding author on reasonable request.

Competing interests:
The authors declare that they have no competing interests in this section.