J. Perinat. Med. 2015; 43(2): 191–199

Junni Wei, Shulian Xue, Junfeng Zhang, Suping Wang* and Bo Wang

Study of the relationship in pregnant women between hepatitis B markers and a placenta positive for hepatitis B surface antigen Abstract Aims: A placenta with hepatitis B virus (HBV) is one of the main reasons for transplacental transmission during pregnancy. This study aims to explore the factors influencing the presence of hepatitis B surface antigen (HBsAg) in the placenta and the synergistic effect of these factors. Methods: A total of 155 placentae and blood specimens were collected from HBsAg-positive mothers and their newborns. HBsAg in placenta was detected using the immunohistochemistry method. HBV serum markers were detected using enzyme-linked immunosorbent assay (ELISA) and polymerase chain reaction (PCR) methods. Results: The results showed that hepatitis B e antigen (HBeAg) positive, or HBV DNA positive status, is significantly associated with an HBsAg-positive placenta. A synergistic effect was present. The hazard ratio for a HBsAgpositive placenta in mothers with HBeAg and HBV DNA was 1.97 times higher than the sum of the independent relative risk of each separate effect (synergy index, S = 1.97). There was a statistically significant association between HBsAg in newborns and HBsAg in placenta, and the risk of newborns with HBsAg was greater (odds ratio values 3.33 and 5.31, respectively) when placental cells close to the fetal side were HBsAg positive. Conclusions: Being positive for HBeAg and/or HBV DNA are significant risk factors for HBsAg in the placenta. HBsAg can pass through the placenta via cellular transfer, possibly contributing to transplacental transmission. Keywords: Fluorescent quantitative PCR; hepatitis B; immunohistochemistry; placenta.

*Corresponding author: Suping Wang, Department of Epidemiology, Public Health College, Shanxi Medical University, No. 56 Xinjian South Road, Shanxi (030001), Taiyuan, China, E-mail: [email protected] Junni Wei and Shulian Xue: Department of Epidemiology, Public Health College, Shanxi Medical University, Taiyuan, China Junfeng Zhang: Department of Health Statistics, School of Public Health, Shanxi Medical University, Taiyuan, China Bo Wang: Department of Obstetrics and Gynecology, Infectious Disease Hospital, Taiyuan, China

DOI 10.1515/jpm-2014-0056 Received February 17, 2014. Accepted June 13, 2014. Previously published online July 11, 2014.

Introduction Hepatitis B virus (HBV) infection is a serious, worldwide problem. The world is divided into regions of high, intermediate, and low endemicity. Transmission of HBV from hepatitis B surface antigen (HBsAg) positive mothers to their infants plays an important role in hyperendemic areas, such as China, for HBV infection [6, 8]. Infants born to mothers infected with HBV are at high-risk of infection during gestation. Neonatal HBV infection is related to the HBV status of mothers. When a mother is positive for hepatitis B e antigen (HBeAg) and/or viral DNA in her serum, the transmission rate is estimated to be 90%. Conversely, if a mother is negative for HBeAg and/or viral DNA in the serum, the transmission rate ranges from 10% to 30% [24]. A mother and her fetus, in the uterus, are separated by the placenta. The placenta provides an interface for the exchange of substances between a mother and her fetus during pregnancy. Molecular, histological, and functional rearrangements of the placenta are necessary throughout pregnancy. The placenta consists of trophoblasts and villous core tissue. The trophoblast acquires a villous and an extravillous phenotype; the villous cytotrophoblast enters the syncytial pathway, while the extravillous trophoblast invades the maternal decidua [11, 12, 35]. Therefore, the placenta, from maternal side to fetal side, includes the following main cell types: decidual cells (DC), trophoblastic cells (TC), villous mesenchymal cells (VMC), and villous capillary endothelial cells (VCEC) [27, 28]. Among these cell types, the trophoblast, as a major cell population, is one of the earliest to differentiate and shows extensive proliferation and/or differentiation up to the formation of the normal placenta [3]. Therefore, a human placenta-derived cell line, Bewo, cultured in vitro, has been extensively used to simulate placenta and avoid the limitations of in vivo studies [4, 17]. If HBV in the mother’s body is to infect the fetus, it must first, through the maternal blood, directly contact

Brought to you by | Karolinska Institute Authenticated Download Date | 5/24/15 5:00 AM

192      Wei et al., Hepatitis B mothers and HBsAg-positive placentae placental cells then enter the fetal blood circulation. Numerous studies suggest that the main route of HBV transmission from mother to fetus is via the placenta [27, 32, 34]. That is, HBV from the maternal side of the placenta passes to the fetal side. Therefore, placental HBV infection may play an important role in neonatal HBV infection. Many researchers have detected HBsAg in placental tissue in population-based and in vitro experiments, and have focused on demonstrating the relationship between placental HBsAg status and intrauterine HBV infection [26, 28, 33]. Obviously, HBsAg in the placenta is closely related to infection status and viral load of the mother. However, the relationship between hepatitis B infection status in pregnant women and the antigen status of the placenta, as well as the interaction effect among risk factors, remain unknown. These risk factors are of great significance for elucidating the mechanism of mother-to-child transplacental transmission of hepatitis B. In order to minimize the risk of transplacental transmission of the virus and decrease the infection rate in high-risk newborns, this study explored the factors influencing the presence of HBsAg in the placenta, as well as the synergistic effect of these factors. At the same time, in order to verify the risk effect of these factors on HBV infection in placenta cells, trophoblast-derived Bewo cells were used to detect HBV cccDNA after co-culture with HBV DNA positive serum, whose concentration of HBV DNA was determined as the basis of our population-based study. Results of this study provide guidance for serum screening of hepatitis B-infected pregnant women who wish to bear children.

Materials and methods Study subjects and specimen collection From January 2005 to February 2009, a total of 155 HBsAg-positive pregnant women, who were undergoing prenatal examination and delivery in the Department of Obstetrics and Gynecology of Taiyuan Infectious Disease Hospital in Shanxi Province, Taiyuan, China, were recruited to the study. These mothers had not received any antiviral therapy during their pregnancy. The study was approved by the Ethics Committee of Shanxi Medical University, Taiyuan, China, and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from each mother included in the study. Maternal venous blood was collected before delivery. Neonates’ venous blood was collected at birth before the hepatitis B virus immune globin (HBIG) and hepatitis B vaccine were given. Serum was separated from 2 mL of non-anticoagulated blood. Serum samples were stored at –20°C for laboratory testing. Placental tissue (1.5  cm × 1.5 cm × 0.3 cm) was collected under sterile conditions from mothers during childbirth, and was immediately immersed in 4% paraformaldehyde for overnight fixation. The tissue was then embedded in paraffin according to standard procedures and sectioned to 4–5 μm.

The inclusion criteria of pregnant women were as follows: (i) single pregnancy; (ii) gestational age   ≥  37 weeks; (iii) serum HBsAg positive; (iv) normal liver and kidney functions; (v) negative serial tests for hepatitis A virus (HAV), hepatitis C virus (HCV), hepatitis D virus (HDV), and hepatitis E virus (HEV); (vi) absence of fetal anomalies shown by B-ultrasonography; (vii) did not receive anti-virals, immunomodulating, cytotoxic, or steroid hormone therapy during pregnancy; and (viii) husband was not a HBV carrier nor HBsAg positive patient.

Determination of HBV markers in serum An enzyme-linked immunosorbent assay (ELISA) was used to detect HBV markers, including the hepatitis B e-antigen (HBeAg, eAg), the antibody to hepatitis B e protein (anti-HBe), and the antibody to hepatitis B c protein (anti-HBc) in 155 serum samples of mothers. In addition, 155 serum samples from newborns were tested for HBsAg and anti-HBc. Testing was performed in strict accordance with the instructions of the kit (Shanghai Kehua Biotechnology, Shanghai, China) for all samples. The experiment was performed in 96-well micrometer plates, and the OD value of each well was detected by an ELISA reader at a 450 nm wavelength after zeroing with a blank well. Results were determined according to the ratio of the sample OD value to the mean OD values of the negative control. Samples were considered HBsAg positive when the ratio was more than 2.1.

Quantitative detection of HBV DNA levels in serum Nucleic acids were isolated from serum with the classic method of phenol-chloroform extraction. HBV DNA in nucleic acid preparations was detected with fluorescent quantitative polymerase chain reaction (PCR) according to the manufacturer’s instructions (Stratagene, South San Francisco, CA, USA). The diagnostic kit for quantification of HBV DNA (Da An Gene Diagnostic Center, Zhongshan Medical University, Guangzhou, China) was used in reaction volumes of 26  μL in a model Mx3000P qPCR machine (Stratagene). The PCR mixture contained 24 μL of reaction mixture and 2 μL of sample DNA or quality control DNA (negative control, positive control of 104 copies/mL of the HBV genome, and positive controls of 107, 106, 105, and 104 copies/mL of the HBV genome). The reaction mixture was composed of HBV-PCR reaction mixture, Taq polymerase and a solid capping agent. After initial activation at 93°C for 2 min, thermal conditions were 10 cycles of 93°C for 45 s and 55°C for 60 s, followed by 30 cycles of 93°C for 30 s and 55°C for 45 s. All information was collected by the thermal cycler and stored by the associated analysis software. After the reaction, the standard curve was drawn automatically, based on the amplification of the positive quantitative standard template. The copy number of HBV DNA in the test sample was calculated according to the standard curve generated by the qPCR software. An HBV DNA load  > 1 × 103 ­copies/mL was considered positive.

Avidin-biotin complex method of immunohistochemistry Tissue samples, including placenta from 155 HBsAg-positive ­mothers, were tested for HBsAg by immunohistochemistry with the avidin-biotin complex (ABC) technique. Confirmed HBsAg-positive

Brought to you by | Karolinska Institute Authenticated Download Date | 5/24/15 5:00 AM

Wei et al., Hepatitis B mothers and HBsAg-positive placentae      193 liver from the pathology department was used as a positive control, and placenta from HBV-negative mothers was used as a negative control. Blank control and alternative control were done in parallel. Instead of primary antibody, blank control used phosphate-buffered saline (PBS) and alternative control used normal mouse serum. Each experiment had three parallel samples. Paraffin sections were incubated at 55°C for 12 h in an electric incubator, and then conventionally dewaxed in water. Slices were then incubated with 3% hydrogen peroxide (H2O2) in methanol for 30 min at room temperature to block endogenous peroxidase activity, washed three times in PBS, blocked with normal goat serum for 30  min and finally incubated at 4°C overnight with a mouse anti-human HBsAg antibody (1:100, Zymed, South San Francisco, CA, USA) in the humidor. Sections were washed with PBS, incubated with biocatalytic sheep anti-mouse IgG (Vector, Burlingame, CA, USA) then peroxidase-conjugated streptomycin for 1 h at room temperature. Diaminobenzidine (DAB) kits (Vector) were used as a substrate for the peroxidase reaction. After color development for 3–5 min, the sections were counterstained with hematoxylin. Sections were dehydrated with a gradient of alcohol, cleared with dimethylbenzene, covered with a glass cover slip, and observed by light microscopy. Immunohistochemical evaluation of the presence of HBsAg was performed independently by two pathologists blinded to the samples’ source information. Discrepancies between the pathologists were resolved by consensus.

HBV cccDNA detection in vitro experiment A human placenta-derived cell line, Bewo (3111C0001CCC000156), was purchased from the cell culture center of the School of Basic Medicine of Peking Union Medical College (Beijing, China), and was cultured at 37°C and in a 5% CO2-humidified atmosphere in F12 medium (Hyclone, Logan, UT, USA). The medium contained 1% glutamine and was supplemented with 10% fetal bovine serum (FBS; Tianhang Biological Technology Co., Ltd, Hangzhou, Zhejiang, China), streptomycin (100 U/mL), and penicillin (100 U/mL). After 24 h, the cells were subcultured in culture bottles or plates. When cell growth covered 70% of the bottom of the culture bottle or plate, HBV infection experiments were performed using serum from hepatitis B patients with HBV DNA at a concentration of 5 × 106 copies/mL and positive for both HBsAg and HBeAg. In the infected group, 3 mL 10% F12 culture solution and 100 μL HBV positive serum were added into the culture bottle or plate. In the negative control group, 3 mL 10% F12 culture solution and 100 μL healthy serum were added. Cells were cultured for 48 h, and were washed six times with 0.01 M PBS. The last wash solution was collected for detection; 3 mL 10% F12 culture solution was then added to continue the cultures. Cells were sampled at 24 h and 48 h to detect HBV cccDNA. DNA from cultured cells and from the last PBS wash solution was extracted by the classic method of proteinase K and phenol-chloroform. The primer sequences of HBV cccDNA were designed as previously described [15]: sense, 5′-CCG ACC ACG GGG CGC ACC TCT CTT TAC G-3′ (1515-1542) and antisense, 5′-CAA GGC ACA GCT TGG AGG CTT GAA CAG T-3′ (1888-1861), yielding a product of 373 bp. PCR was performed with: 94°C for a 5 min initial denaturation, 35 cycles of 94°C for 50 s, 55°C for 50 s, 72°C for 55 s, and 72°C for a 10  min final extension. After the reaction, the amplification products, HBV cccDNA positive control, negative control, and DNA marker were electrophoresed together on a 1.5% agarose gel (containing 0.5 μg/mL ethidium bromide).

Statistical analysis SPSS version 16.0 (Chicago, IL, USA) was used to analyze the data; positive rates, odds ratio (OR) values, OR 95% confidence intervals (CI) and Chi-square (χ2) tests were calculated. Differences were considered statistically significant when P-values were