Screening for serum biomarkers in patients with chronic hepatitis B with hepatitis B surface antigen seroclearance, following pegylated interferon alpha therapy
- Authors:
- Published online on: October 16, 2014 https://doi.org/10.3892/mmr.2014.2670
- Pages: 427-433
Abstract
Introduction
Chronic hepatitis B (CHB) is one of the most common infectious diseases and leading causes of death, with ~400 million chronically infected individuals worldwide (1). Over the past three decades, progressively more patients have received treatment; initially with interferon alpha (IFN-α), and recently with the advent of nucleoside analogues (2). IFN-α has restricted efficacy, as not all patients will exhibit a therapeutic response. However, in some hepatitis B e antigen (HBeAg)-positive and -negative patients, the sustained suppression of viral replication and the seroclearance of hepatitis B surface antigen (HBsAg) may be attained (2).
Recombinant IFN-α is used clinically worldwide. However, the efficacy of the therapy is restricted by protein characteristics, including poor stability, a short half-life and immunogenicity. Recent advances in pegylation technology have resulted in the development of pegylated (peg) IFN-α, which leads to a reduced clearance of IFN-α, and therefore an increased circulatory time and sustained systemic exposure (2).
HBsAg was initially identified >40 years ago (3). The detection of HBsAg in the serum is the hallmark of hepatitis B virus (HBV) infection and remains the gold standard diagnostic marker of CHB. HBsAg has been proven to be a steady marker, that may be used to predict clinical outcomes (4). An advantage of peg-IFN-α, as compared with nucleotide analogues, is that HBsAg seroclearance is observed in numerous patients undergoing peg-IFN-α treatment (2). It would therefore be beneficial to identify serum biomarkers that can predict which patients with CHB would have HBsAg seroclearance following peg-IFN-α therapy. The biomarkers would not only be potential treatment outcome predictors, but may also provide a new basis for evaluating the molecular mechanisms underlying CHB pathogenesis.
In 1985, Smith et al (5) established a method for displaying polypeptides on the surface of filamentous M13-derived bacteriophage (phage). This phage display technique was originally developed to map epitope-binding sites of antibodies, through the panning of random peptide-phage libraries on immobilized immunoglobulins. Since then, phage display has been used to determine polypeptide binding with a diverse range of basic biological, as well as technical and medical applications (6,7). In the present study, a random phage display peptide library was used to screen for potential serum peptide biomarkers, in predicting which patients with CHB would exhibit HBsAg seroclearance following peg-IFN-α therapy.
Materials and methods
Subjects
Between January 2009 and February 2012, 1,120 Han Chinese patients with CHB, admitted to the Qingdao Infectious Diseases Hospital (Shangdong, China), were screened for HBsAg seroclearance following 48 weeks of peg-IFN-α therapy. A total of 30 patients with CHB who achieved HBsAg seroclearance following peg-IFN-α therapy and 30 age-, gender-, HBeAg status- and HBV genotype-matched patients with CHB without HBsAg seroclearance following peg-IFN-α therapy, were enrolled as the discovery cohort. A further 50 patients with CHB with HBsAg seroclearance and 50 randomly selected patients with CHB without HBsAg seroclearance, following peg-IFN-α therapy, were enrolled as a validation cohort, to validate the potential serum peptide biomarkers identified in the discovery cohort. The baseline characteristics of the discovery and validation cohorts are listed in Tables I and II, respectively. A further 50 randomly selected healthy volunteers were enrolled as a healthy control group, in the validation phase. The recruited patients with CHB were diagnosed according to the Asia-Pacific consensus statement on the management of CHB (a 2008 update), by the Asian Pacific Association for the Study of the Liver (8).
 | Table IBaseline characteristics of the discovery cohort of patients with chronic hepatitis B, with or without hepatitis B surface antigen seroclearance, following pegylated interferon-α therapy. |
| Table IIBaseline characteristics of the validation cohort of patients with chronic hepatitis B, with or without hepatitis B surface antigen seroclearance, following pegylated interferon-α therapy. |
All of the recruited patients with CHB underwent 48 weeks of peg-IFN-α therapy of 180–190 μg/week (Roche, Shanghai, China). All met the following criteria: HBV DNA load of ≥104 copies/ml, elevated serum alanine aminotransferase (ALT) level 2–10 times the upper limit of normal and no immunotherapy or HBV vaccine treatment for 6 months prior to enrollment. The exclusion criteria for the recruited patients were as follows: Co-infection with hepatitis A (HAV), C (HCV), E (HEV) or human immunodeficiency (HIV) viruses; evidence of other etiological liver co-morbidities (e.g., alcohol-related, drug-related, or autoimmune diseases); serious concurrent medical illnesses (e.g., malignancy, severe cardiopulmonary disease, psychiatric disorders, uncontrolled diabetes mellitus, or hyper- or hypothyroidism); and pregnancy or nursing. Blood samples were collected from all of the screened (n=1,120) patients with CHB at baseline, within 72 hours prior to peg-IFN-α therapy, at 48 weeks, the end of peg-IFN-α therapy, and finally at 80 weeks, 32 weeks post peg-IFN-α therapy. The patients with CHB who achieved consistent HBsAg loss, until 32 weeks post peg-IFN-α therapy, were included in the HBsAg seroclearance group in the discovery or validation cohorts. In the present study, only the baseline blood samples from the discovery and validation cohorts were used for the discovery/screening and validation tests with a random phage display peptide library. Blood samples were also collected from all of the healthy volunteers within 72 hours of enrollment (n=50). The present study was approved by the Ethics Committee of Qingdao Infectious Diseases Hospital. Written informed consent was obtained from all of the participating subjects prior to the study.
Reagents
The phage display peptide library and the ER2738 Escherichia coli host strain were purchased from New England Biolabs Inc. (Ipswich, MA, USA). The QIAprep Spin M13 kit was purchased from Qiagen (Valencia, CA, USA). The serum ALT levels were measured using a fully automatic biochemical analyzer (AU2700; Olympus, Tokyo, Japan) with a 40 IU/l upper limit of normal; HBV DNA was quantified by real-time polymerase chain reaction using an ABI PRISM 7000 thermocycler (Roche Molecular Systems Inc., Alameda, CA, USA) and a diagnostic kit (Daan Genetech, Guangzhou, China), with a linear range of 5.0×102–1.0×109 copies/ml. HBsAg and HBeAg were detected using an auto-electrochemiluminescence technique (E170; Roche Diagnostics Ltd., Basel, Switzerland), with a normal detection limit of <0.05 IU/ml for HBsAg and <1.0 S/CO for HBeAg.
Preparation of sera, biopanning and amplification of phages
All phage-related procedures in the present study were conducted as described by previous methods (9). Serum pools were obtained from the patients with CHB, with or without HBsAg seroclearance following peg-IFN-α therapy, and from the healthy volunteers. Immunoglobulin G (IgG) was purified from the serum pools. To increase the screening specificity, microtiter wells were coated with purified sera IgG from the patients with CHB without HBsAg seroclearance, in order to absorb non-specific phages from the random phage peptide library. The unbound phages were then incubated with the sera IgG from patients with CHB, with HBsAg seroclearance, to screen for phage containing potential peptide biomarkers. Briefly, microtiter wells were incubated at 4°C overnight with 100 μl of purified IgG (100 μg/ml) from the patients with CHB without HBsAg seroclearance. The plates were blocked with 3% nonfat milk for 2 hours at 37°C and then washed five times with 0.05% Tween® 20 in Tris-buffered saline. A 100 μl sample of the diluted random 12-peptide phage display library was added to the coated plates. Following a 1 h incubation at room temperature, the unbound phages were collected and added to a plate coated with sera IgG from the patients with CHB with HBsAg seroclearance (100 μl/well). Following a further 1 h incubation at room temperature, the bound phages were eluted, amplified in the host strain and purified by precipitation. A further two rounds of biopanning were carried out in the same way, except that 1:200 and 1:400 sera dilutions were added to 100 μl of the diluted phages from the previous round. The percentage of enrichment was calculated using the following formula: percentage enrichment of phage clones (%) = (eluted phages/added phages) × 100%.
Phage enzyme-linked immunosorbent assay (ELISA)
A 10 μl volume of eluted phages, from the third round of biopanning, was added to 200 μl of the host strain that had been cultured overnight and then incubated for 20 minutes at 37°C. The transformed host strain cells were then transferred to a culture tube with agarose, quickly vortexed and immediately poured onto a pre-warmed lysogeny broth plate, and incubated overnight at 37°C. A random selection of 20 phage clones were picked the following day. The ELISA wells were coated with 2×1011 phage particles and then blocked with 2% bovine serum albumin (BSA; Sigma-Aldrich, St Louis, MO, USA) in phosphate-buffered saline (PBS). Following blocking, 100 μl of a 1:100 dilution of purified sera IgG, from the patients with CHB with HBsAg seroclearance following peg-IFN-α therapy, was subsequently added to the wells, which were incubate for 1 hour at 37°C. The wells were then washed three times with PBS containing 0.05% Tween® 20. Horseradish peroxidase (HRP)-conjugated goat anti-human IgG antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) was added to the plate, and the specifically bound antibodies were visualized using 3,3′,5,5′-tetramethylbenzidine (TMB; Sigma-Aldrich). The absorbance value was measured at 450 nm. The original phage library was used as the negative control and phage clones were considered positive when their absorbance value was >2 that of the negative control.
Phage DNA sequencing
The phage clones considered to be positive in the phage-ELISA were precipitated with polyethanol glycol/NaCl. The phage DNA was then extracted for sequencing using a QIAprep Spin M13 kit (Qiagen) and sequenced with an ABI PRISM 377 sequencer, using the −96 g111 sequencing primer (5′-CCC TCA TAG TTA GCG TAA CG-3′)from the Phage Display Peptide Library kit (New England Biolabs, Ipswich, MA, USA).
Determination of the optimal reaction concentration
In order to determine the optimal reaction concentrations, a chessboard titration was performed with the diluted positive phage clones (1010 pfu/well, 1011 pfu/well, 1012 pfu/well, 1013 pfu/well and 1014 pfu/well) and purified sera IgG from the patients with CHB with HBsAg seroclearance following peg-IFN-α therapy (1:10, 1:100, 1:200, 1:400, 1:800). The plates were coated with 100 μl of diluted phages overnight at 4°C, and then washed and blocked with 2% BSA. The sera IgG dilutions were subsequently added and incubated for 1 hour at 37°C. After incubation with a HRP-conjugated goat anti-human IgG antibody for 1 h at 37°C, TMB was added to each well, and the enzymatic reaction was allowed to proceed for 15 minutes at room temperature. The absorbance values were measured at 450 nm, using a microplate reader (BioTek, Beijing, China). The original phage library and the undiluted purified sera IgG from the patients with CHB with HBsAg seroclearance following peg-IFN-α therapy, were used as the negative and the positive controls, respectively. Each dilution was repeated in triplicate.
Predictive validity of the positive peptide phage clone
To evaluate the predictive validity of the IFNC1 peptide phage clone, sera from 50 patients with CHB with HBsAg seroclearance and 50 patients with CHB without HBsAg seroclearance, following peg-IFN-α therapy, and 50 healthy volunteers were used to react with the IFNC1 clone using a phage-ELISA method. The ELISA wells were coated with 1011 pfu/well phage particles and incubated with 1:200 diluted sera IgG. Following the incubation, a 1:2,000 dilution of HRP-Staphylococal Protein A (Life Technologies, Carlsbad, CA, USA) was added, and TMB was used as the substrate for coloration. Absorbance values were measured at 450 nm. The original phage library was used as the negative control, and the phage-ELISA results were considered positive when the absorbance value was >2 times that of the negative control. Predictive validity indicators were calculated using the following formulae: sensitivity = true positive/(true positive + false negative); specificity = true negative/(true negative + false positive); positive predictive value (PPV) = true positive/(true positive + false positive); negative predictive value (NPV) = true negative/(true negative + false negative); accuracy = (true positive + true negative)/(true positive + false negative + true negative + false positive).
Statistical analyses
Statistical analyses were performed using SPSS for Windows version 13.0 (SPSS Inc., Chicago, IL, USA). All continuous variables are expressed as the means ± standard deviation. The comparisons of the means between two groups were performed using Student’s t-tests. The comparisons of the means between multiple groups were performed using a one-way analysis of variance, followed by post hoc pairwise comparisons using Tukey’s tests. Categorical variables were compared with χ2 tests. A two-tailed P<0.05 was considered to indicate a statistically significant difference.
Results
The discovery cohort consisted of 30 patients with CHB who exhibited HBsAg seroclearance following peg-IFN-α therapy and 30 age-, gender-, HBeAg status- and HBV genotype-matched patients with CHB without HBsAg seroclearance following peg-IFN-α therapy. Each of the two cohorts contained equal numbers (n=15) of HBeAg (+) and HBeAg (−) patients. As shown in Table I, there were no significant differences in the age, gender, body mass index (BMI), serum ALT level, HBV DNA load, serum HBsAg level and HBV genotype between the HBsAg clearance and non-clearance groups at baseline. A further 50 patients with CHB, with HBsAg seroclearance and 50 randomly selected patients with CHB without HBsAg seroclearance, following peg-IFN-α therapy, were enrolled as a validation cohort, to validate the potential serum peptide biomarkers identified in the discovery cohort. In the validation cohort, there were also no significant differences in the baseline characteristics between the HBsAg clearance and non-clearance groups, either in total or by the HBeAg (+)/(−) subgroups (Table II). In the validation phase, 50 randomly selected healthy volunteers were also enrolled as a control group, which had no significant differences in age (37.9±8.5 years) and gender (male, n=31; 62%), as compared with the HBsAg clearance and non-clearance groups.
As shown in Table III, the percentage enrichment of phage increased from 2.6×10−5% to 2.1×10−2% following three rounds of biopanning, an enrichment of ~1,000 folds. Purified sera IgG from the HBsAg clearance group at baseline (within 72 hours before peg-IFN-α therapy), had a positive reaction with 17/20 randomly picked phage clones (85%; Fig. 1A). The single-stranded DNA from the positive phage clones was then sequenced. Out of the 17 positive phage clones, 13 clones (C1–4, 6, 8–9, 11–13, 15–16, 20) had the same inserted DNA sequence 5′-GAA ACC TGT CGT GCC AGC TGC ATT AAT GAA TCG GCC-3′ (Fig. 1B). The corresponding peptide sequence was ETCRASCINESA (named IFNC1). Using NCBI Blast software (National Institutes of Health, Bethesda, MA, USA), the identified peptide sequence was searched for in numerous protein databases, including Swissprot (www.ebi.ac.uk/uinprot) and the Protein Data Bank (www.rcsb.org/pdb/home/home.do). No significant homology was determined with other protein sequences (score <50). A chessboard titration was performed to determine the optimal reaction concentrations for the IFNC1 peptide phage clone and sera IgG from the HBsAg clearance group. The optimal coating concentration of the IFNC1 peptide phage clone was 1011 pfu/well and the optimal dilution for sera IgG from the HBsAg clearance group was 1:200.
To evaluate the predictive validity of IFNC1, sera from the validation cohort at baseline and the healthy control group were used to react with the IFNC1 peptide phage clone using a phage-ELISA. The original phage library was used as a negative control. As shown in Table VI, sera from 46/50 patients with CHB with HBsAg seroclearance and 6/50 patients with CHB without HBsAg seroclearance, following peg-IFN-α therapy, and 4/50 healthy controls exhibited a positive reaction with IFNC1. The mean absorbance value of the HBsAg seroclearance group was significantly higher, as compared with the HBsAg non-clearance and healthy control groups, either in total or by the HBeAg (+)/(−) subgroups. The positive reaction rate of the IFNC1 peptide phage clone was 92.0% with the HBsAg seroclearance group. This was significantly higher as compared with the HBsAg non-clearance group (12.0%) and the healthy control group (8.0%).
As shown in Table V, the sensitivity of the IFNC1 peptide phage clone to detect patients with CHB with HBsAg seroclearance, following peg-IFN-α therapy, was 92.0%, specificity 86.4%, PPV 88.5%, NPV 91.7% and accuracy 90.0%, as compared with the non-clearance group. As compared with the healthy control group, the IFNC1 peptide had a sensitivity of 92.0%, specificity 92.0%, PPV 92.0%, NPV 92.0% and accuracy 92.0%, in detecting the patients with CHB with HBsAg seroclearance, following peg-IFN-α therapy.
Discussion
Hepatitis B is a potentially life-threatening, infectious liver disease caused by HBV. Worldwide, ~2 billion people have been infected with HBV, and >240 million have chronic liver infection (11). The covalently closed circular DNA (cccDNA) template of HBV transcription has a key role in the life cycle of the virus and promotes the persistence of infection. It has previously been suggested that the amount of circulating HBsAg reflects the concentration of cccDNA in the liver and therefore the number of infected hepatocytes (4). The aim of CHB antiviral therapy is to clear the cccDNA from the liver. Two treatment strategies may be used, either immune-mediated control, obtained with 48 weeks of peg-IFN therapy; or viral control, obtained with long-term nucleoti(si)de analogues, which are potent HBV inhibitors (4). One of the advantages of peg-IFN is that a finite course of therapy may potentially lead to sustained disease remission in subsequent decades (4). HBsAg seroclearance can reportedly be achieved in some patients with hepatitis B in response to peg-IFN-α treatment (2). In the present study, a mimic peptide was identified with a predictive value in differentiating patients with CHB with or without HBsAg seroclearance, following peg-IFN-α therapy. This was determined using a random phage display peptide library and sera from the patients with CHB prior to peg-IFN-α therapy.
In the discovery/screening phase, the HBsAg clearance group (n=30) was matched with the HBsAg non-clearance group in age, gender, HBeAg (+)/(−) status, and HBV genotype, in order to minimize background noise for the screening tests. In the validation phase however, patients in the HBsAg non-clearance group (n=50) were randomly selected, therefore the predictive validity of the IFNC1 peptide could be determined in a more realistic context. In the present study, the predominant HBV genotypes were B and C, this is in agreement with previous reports that genotypes B and C circulate in China (2,12).
The phage display peptide library technique was originally developed to map epitope-binding sites of antibodies, by panning random peptide-phage libraries on immobilized immunoglobulins (6,7). This technique involves specific screening and affinity selection of phages displaying peptides which may be ligands for particular proteins. A major advantage of the technique is the extensive control that may be exerted over the selection conditions, allowing the preservation of the three-dimensional structure of the peptide (13,14). The IFNC1 mimic peptide identified in the present study demonstrated high predictive validity in differentiating patients with CHB with or without HBsAg seroclearance, following peg-IFN-α therapy, as well as from healthy controls. Therefore, the IFNC1 mimic peptide may be a potential serum biomarker which could be used to predict the therapeutic outcome of peg-IFN-α and to guide the treatment choices for patients with CHB. This may be of particular use for HBeAg (−) patients, whose response rate to peg-IFN therapy is known to be low (<20%) (15). Since the IFNC1 mimic peptide exhibited no significant homology with any other protein sequences, it is likely to be derived from an unknown protein. Exploration into the identity and functional roles of IFNC1, and its parent protein(s), may provide novel insights into the molecular mechanisms underlying CHB pathogenesis.
There are some limitations to the present study; in China, patients with CHB have a high chance of co-infection with HCV (9, 16, 17). However, patients with CHB co-infected with either HAV, HCV, HEV or HIV were excluded from the present study to minimize background noise for the screening tests. Patients with CHB with a comprehensive disease background may be enrolled in future studies, to test the clinical application value of IFNC1. Additionally, only Han Chinese patients with CHB were enrolled in the present study, which minimized the background noise for the screening tests, however it may have limited the validity of the findings. Therefore, it may be interesting to test the predictive validity of IFNC1, or screen for its counterparts, in different ethnicities in future studies.
In conclusion, the IFNC1 mimic peptide was identified as being capable of differentiating between patients with CHB with or without HBsAg clearance, following peg-IFN-α therapy. IFNC1 showed high predictive validity for patients with CHB with HBsAg clearance, and therefore may be a potential serum biomarker to predict the treatment outcome of peg-IFN-α therapy and to guide the treatment choices for patients with CHB. In addition, the IFNC1 mimic peptide may provide a new basis for exploration into the molecular mechanisms underlying CHB pathogenesis.
References
|
Liaw YF and Chu CM: Hepatitis B virus infection. Lancet. 373:582–592. 2009. View Article : Google Scholar : PubMed/NCBI | |
|
Dusheiko G: Treatment of HBeAg positive chronic hepatitis B: interferon or nucleoside analogues. Liver Int. 33:137–150. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Blumberg BS, Alter HJ and Visnich S: A “new” antigen in leukemia sera. JAMA. 191:541–546. 1965. | |
|
Martinot-Peignoux M, Lapalus M, Asselah T and Marcellin P: HBsAg quantification: useful for monitoring natural history and treatment outcome. Liver Int. 34:97–107. 2014. View Article : Google Scholar : PubMed/NCBI | |
|
Smith GP: Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science. 228:1315–1317. 1985. View Article : Google Scholar : PubMed/NCBI | |
|
Samoylova TI, Morrison NE, Globa LP and Cox NR: Peptide phage display: opportunities for development of personalized anti-cancer strategies. Anticancer Agents Med Chem. 6:9–17. 2006.PubMed/NCBI | |
|
Sergeeva A, Kolonin MG, Molldrem JJ, Pasqualini R and Arap W: Display technologies: application for the discovery of drug and gene delivery agents. Adv Drug Deliv Rev. 58:1622–1654. 2006. View Article : Google Scholar : PubMed/NCBI | |
|
Liaw YF, Leung N, Kao JH, et al: Asian-Pacific consensus statement on the management of chronic hepatitis B: a 2008 update. Hepatol Int. 2:263–283. 2008. View Article : Google Scholar | |
|
Zhang Z, Xu L and Wang Z: Screening serum biomarkers for early primary hepatocellular carcinoma using a phage display technique. J Clin Lab Anal. 25:402–408. 2011. View Article : Google Scholar : PubMed/NCBI | |
|
Wang M, Yi XY, Li XP, Zhou DM, Larry M and Zeng XF: Phage displaying peptides mimic schistosoma antigenic epitopes selected by rat natural antibodies and protective immunity induced by their immunization in mice. World J Gastroenterol. 11:2960–2966. 2005. View Article : Google Scholar | |
|
Wang Y, Zhao C, Zhang L, et al: Predictive value of interferon-gamma inducible protein 10kD for hepatitis B e antigen clearance and hepatitis B surface antigen decline during pegylated interferon alpha therapy in chronic hepatitis B patients. Antiviral Res. 103:51–59. 2014. View Article : Google Scholar | |
|
Tanwar S and Dusheiko G: Is there any value to hepatitis B virus genotype analysis? Curr Gastroenterol Rep. 14:37–46. 2012. View Article : Google Scholar : PubMed/NCBI | |
|
Goffinet M, Chinestra P, Lajoie-Mazenc I, Medale-Giamarchi C, Favre G and Faye JC: Identification of a GTP-bound Rho specific scFv molecular sensor by phage display selection. BMC Biotechnol. 8:342008. View Article : Google Scholar : PubMed/NCBI | |
|
Nizak C, Monier S, del Nery E, Moutel S, Goud B and Perez F: Recombinant antibodies to the small GTPase Rab6 as conformation sensors. Science. 300:984–987. 2003. View Article : Google Scholar : PubMed/NCBI | |
|
Hadziyannis SJ: Treatment paradigms on hepatitis B e antigen-negative chronic hepatitis B patients. Expert Opin Investig Drugs. 16:777–786. 2007. View Article : Google Scholar : PubMed/NCBI | |
|
Merican I, Guan R, Amarapuka D, et al: Chronic hepatitis B virus infection in Asian countries. J Gastroenterol Hepatol. 15:1356–1361. 2000. View Article : Google Scholar : PubMed/NCBI | |
|
Sievert W, Altraif I, Razavi HA, et al: A systematic review of hepatitis C virus epidemiology in Asia, Australia and Egypt. Liver Int. 31(Suppl 2): 61–80. 2011. View Article : Google Scholar : PubMed/NCBI |
