Acute Hepatitis C Virus Infection Induces Anti-Host Cell Receptor Antibodies With Virus-Neutralizing Properties Rajiv G. Tawar,1,2 Che C. Colpitts,1,2 J€org Timm,3 Tanja Fehm,4 Michael Roggendorf,5 Helga Meisel,6 Nicolas Meyer,7 Franc¸ois Habersetzer,1,2,8 Franc¸ois-Lo€ıc Cosset,9 Thomas Berg,10 Mirjam B. Zeisel,1,2 and Thomas F. Baumert1,2,8 Hepatitis C virus (HCV) causes persistent infection in the majority of infected individuals. The mechanisms of persistence and clearance are only partially understood. Antibodies (Abs) against host cell entry receptors have been shown to inhibit HCV infection in cell culture and animal models. In this study, we aimed to investigate whether anti-receptor Abs are induced during infection in humans in vivo and whether their presence is associated with outcome of infection. We established an enzyme-linked immunosorbant assay using a recombinant CD81-claudin-1 (CLDN1) fusion protein to detect and quantify Abs directed against extracellular epitopes of the HCV CD81-CLDN1 coreceptor complex. The presence of anti-receptor Abs was studied in serum of patients from a well-defined cohort of a single-source HCV outbreak of pregnant women and several control groups, including uninfected pregnant women, patients with chronic hepatitis B and D virus (HBV/HDV) infection, and healthy individuals. Virus-neutralizing activity of Abs was determined using recombinant cell culture–derived HCV (HCVcc). Our results demonstrate that HCVinfected patients have statistically significantly higher anti-CD81/CLDN1 Ab titers during the early phase of infection than controls. The titers were significantly higher in resolvers compared to persisters. Functional studies using immunoadsorption and HCV cell culture models demonstrate that HCV-neutralizing anti-receptor Abs are induced in the early phase of HCV infection, but not in control groups. Conclusion: The virus-neutralizing properties of these Abs suggest a role for control of viral infection in conjunction with antiviral responses. Characterization of these anti-receptor Abs opens new avenues to prevent and treat HCV infection. (HEPATOLOGY 2015;62:726-736)
N
early 170 million people are infected with hepatitis C virus (HCV) worldwide.1 Approximately 85% of infected individuals develop persistent viral infection with a risk of developing liver
cirrhosis and hepatocellular carcinoma; the remaining 15% clear the virus spontaneously.2 Despite intensive efforts to understand the immune correlates of HCV clearance, the underlying mechanisms are still only
Abbreviations: Abs, antibodies; ANAs, anti-nuclear antibodies; ASMAs, anti-smooth muscle antibodies; CLDN, claudin; ECL, extracellular loop; ELISA, enzyme-linked immunosorbent assay; GM-CSF, granulocyte macrophage colony-stimulating factor; HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; HCV, hepatitis C virus; HCVcc, cell culture–derived HCV; HDV, hepatitis D virus; HIV, human immunodeficiency virus; HRP, horseradish peroxidase; IFN, interferon; IgG, immunoglobulin G; IL, interleukin; IPTG, isopropyl b-D-1 thiogalactopyranoside; kD, kilodalton; LEL, large extracellular loop; mAb, monoclonal antibody; OD, optical density; oxLDL, oxidized low-density lipoprotein; PBS-T, phosphate-buffered saline with Tween 20; PPI, posterior probability of inferiority; PPS, posterior probability of superiority; ROC, receiver operating characteristic analysis; SR-BI, scavenger receptor class B type I; TGF-b3, transforming growth factor beta 3; TMB, 3,30 ,5,50 –tetramethylbenzidine; TNF-a, tumor necrosis factor alpha. From the 1Inserm, U1110, Institut des Maladies Virales et H e patiques, Strasbourg, France; 2University of Strasbourg, Strasbourg, France; 3Institute for Virology, usseldorf, HeinUniversity Hospital D€ usseldorf, Heinrich-Heine-University, D€ usseldorf, Germany; 4Department of Gynecology and Obstetrics, University Hospital D€ rich-Heine-University, D€ usseldorf, Germany; 5Institute of Virology, University of Duisburg-Essen, Essen, Germany; 6Institute of Medical Virology, Helmut-RuskaHaus, Charite University Medicine, Berlin, Germany; 7P^ ole de Sant e Publique, H^ opitaux Universitaires de Strasbourg, Strasbourg, France; 8Institut HospitaloUniversitaire, P^ ole H e pato-digestif, H^ opitaux Universitaires de Strasbourg, Strasbourg, France; 9CIRI–International Center for Infectiology Research, Lyon, France; Inserm, U1111, Lyon France; Ecole Normale Sup e rieure; CNRS UMR 5308, Lyon, France; LabEx Ecofect, University of Lyon, Lyon, France; 10Department of Internal Medicine, Neurology and Dermatology, Gastroenterology and Rheumatology Clinic, Section of Hepatology, University of Leipzig, Leipzig, Germany Received December 2, 2014; accepted May 19, 2015. Additional Supporting Information may be found at http://onlinelibrary.wiley.com/doi/10.1002/hep.27906/suppinfo. 726
HEPATOLOGY, Vol. 62, No. 3, 2015
partially understood, hampering the development of vaccines and immunotherapeutics. Broad, sustained T-cell responses along with intrahepatic production of interferon (IFN)-c have been associated with spontaneous HCV clearance.3,4 Furthermore, studies have shown a strong correlation between induction of neutralizing antibodies (Abs) in the early phase of infection and viral clearance.5-7 HCV entry is a complex process mediated by the viral envelope glycoproteins, E1 and E2, and involving several host factors. Whereas some of these host factors directly interact with the viral envelope glycoproteins, others interact with one another to promote the formation of a receptor complex required for efficient internalization of the virus.8-10 Indeed, formation of the CD81-claudin-1 (CLDN1) coreceptor complex, regulated by receptor tyrosine kinase signaling, plays a central role in HCV entry.9,10 Immunologically raised Abs targeting CLDN1, scavenger receptor BI (SR-BI), and CD81 exhibit a broad, potent anti-viral effect in cellculture models of HCV infection.11-13 Although the roles of neutralizing antiviral responses in HCV infection has been extensively studied within the past years, it is currently not known whether any anti-receptor Abs are generated during HCV infection. Thus, we questioned whether anti-receptor Abs are induced during HCV infection and, if so, whether they play a role in viral clearance. Information on the quality of early immune responses is limited in HCV infection owing to the fact that most cases are not clinically apparent. This has led to a paucity of patient samples from the early phase of infection. Moreover, the heterogeneity of viral strains and patients within a cohort also represent drawbacks for the study of antiviral immune responses. In this context, the AD78 cohort is unique because it consists of a homogeneous patient group of young pregnant women infected by a single source of HCV 1b strain.14,15 Thus, we used patient samples from this well-defined AD78 cohort to assess the role of antireceptor Abs in HCV infection by employing a novel,
TAWAR ET AL.
727
sensitive enzyme-linked immunosorbent assay (ELISA) and immunoadsorption experiments.
Patients and Methods For more experimental details, please refer to the Supporting Information. Patients. Serum samples from the HCV-AD78 cohort were used with written informed consent and approval from the ethics committee, as previously described.6,16,17 The HCV-AD78 cohort consisted of 2,867 young pregnant women infected with a single genotype 1b virus-contaminated anti-D immunoglobulin G (IgG) between August 1978 and March 1979. Early- (1-6 months postinfection) and late-phase (10-17 years postinfection) serum samples from 48 patients of this cohort were available for this study.6,14,15 Spontaneous viral clearance in these patients was characterized by undetectable serum HCV RNA during the first 6 months of infection, whereas chronic infection was classified as serum that tested positive for HCV RNA as long as 17 years postinfection. None of these patients received antiviral treatment before sampling of latephase samples. Characteristics of patients are presented in Table 1. Serum samples from 30 noninfected pregnant women were obtained from the University Hospital D€ usseldorf (D€ usseldorf, Germany) after written informed consent from individuals and approval from the hospital ethics committee (study no.: #4995). Samples tested negative for hepatitis B virus (HBV), HCV, and human immunodeficiency virus (HIV)21 and 22. Serum samples from 15 patients with chronic HBV/ hepatitis delta virus (HDV) infection enrolled with written informed consent in a clinical trial (NCT01732081) were obtained from the University Hospital Strasbourg (Strasbourg, France). These patients were diagnosed by positive hepatitis B surface antigen (HBsAg) and HBV DNA in serum.18 Serum samples from 50 noninfected healthy individuals were obtained from the Etablissement Franc¸ais du Sang of Alsace and Besanc¸on, France (Convention
This study was supported through funding by the European Union (ERC-2008-AdG-233130-HEPCENT, FP7 HepaMab and Interreg IV FEDER-HepatoRegio-Net 2012), the Agence Nationale de Recherche sur le SIDA (ANRS; 2012/239, 2012/318, and 2013/108), and the Direction G e n e rale de l’Offre de Soins (A12027MS). This work has been published under the framework of the LABEX ANR-10-LABX-0028_HEPSYS and benefits from a funding from the state managed by the French National Research Agency as part of the Investments for the Future program. Address reprint requests to: Thomas F. Baumert, M.D., Inserm U1110, Institut de Recherche sur les Maladies Virales et H e patiques, Universit e de Strasbourg, 3 Rue Koeberle , Nouvel H^ opital Civil, F-67000 Strasbourg, France. E-mail: [email protected]; fax: (133) 3 68 85 37 24. C 2015 by the American Association for the Study of Liver Diseases. Copyright V View this article online at wileyonlinelibrary.com. DOI 10.1002/hep.27906 Potential conflict of interest: Nothing to report.
728
TAWAR ET AL.
HEPATOLOGY, September 2015
Table 1. Characteristics of Patients Used in this Study N Median age at infection, years Sex Source of HCV Mode of transmission HCV genotype Samples Therapy
Chronic
Resolved
29 24 Female anti-D IgG* Intravenous 1b Early†, late‡ Na€ıve
19 24 Female anti-D IgG* Intravenous 1b Early†, late‡ Na€ıve
*HCV-contaminated. † 0-6 months postinfection. ‡ 10-17 years postinfection.*0-6 months post-infection, **10-17 years postinfection
PSNT2014 Inserm). These samples were negative for HCV, HBV, HIV-1 and 22, human T-lymphotropic virus, cytomegalovirus, Plasmodium falciparum, and Treponema pallidum. Reagents. Anti-CLDN119 and anti-SR-BI20 sera have been described. Anti-CD81 serum was raised by genetic immunization of a Wistar rat.13 Anti-CD81 monoclonal antibodies (mAbs) JS-81 and 5A6 were from BD Biosciences (San Jose, CA) and Santa Cruz Biotechnology (Santa Cruz, CA), respectively. Anti-E2 mAbs AP3321 and CBH-522 have been described. The anti-HBsAg-specific mAb (NCL-HBsAg-2, clone 1044/ 341) was from Leica Biosystems (Nussloch, Germany). mAbs specific for CLDN1, SR-BI, claudin 6 (CLDN6), and claudin 9 (CLDN9) have been described.11,12 Antihuman horseradish peroxidase (HRP) and anti-mouse HRP Abs were obtained from GE Healthcare (Little Chalfont, UK). Anti-rat HRP Abs were from SigmaAldrich (St. Louis, MO). The following cytokines were used: IFN-a2a (Roche Diagnostics, Indianapolis IN); interleukin (IL)21a, IL2 (Novartis AG, Basel, Switzerland), IL4, IL6, IL7 (Cytheris, Issy-les-Moulineaux, France), IL15, and IL21 (PeproTech, Rocky Hill, NJ); tumor necrosis factor alpha (TNF-a); transforming growth factor beta 3 (TGF-b3); and granulocyte macrophage colony-stimulating factor (GM-CSF). Oxidized low-density lipoprotein (oxLDL) was obtained from Euromedex (Souffelweyersheim, France). CD81/CLDN1 Protein and Abs. Extracellular loops (ECL1 and ECL2) of CLDN1 and the large extracellular loop (LEL) of CD81 were produced as a fusion protein (Fig. 1A) in the periplasmic space of Escherichia coli BL21(DE3). To facilitate downstream purification, a C-terminus strep-tag was added. Loops were cloned into pET-20b(1) vector under the pelB signal sequence. The fusion protein was expressed in BL21(DE3) cells after induction with 1 mM of isopropyl b-D-1 thiogalactopyranoside (IPTG). Cells were
pelleted and the periplasmic fraction was isolated by treating with 1 mg/mL of polymyxin B sulfate, as previously described23; buffers were adapted for streptactin purification. The periplasmic fraction was clarified by centrifugation at 4,500 rpm for 1 hour, loaded on a preequilibrated strep-tactin column, and eluted with 2.5 mM of desthiobiotin. As a control, recombinant ECL2 of tetraspanin CD9 was expressed and purified using the same protocol. Immunoblotting. Purified protein (5 mg) was allowed to migrate on a 15% sodium dodecyl sulfate polyacrylamide gel electrophoresis gel and transferred to HyBond-P nitrocellulose membranes (GE Healthcare). To detect CD81 LEL, the membrane was probed with anti-CD81 Ab 5A6 and bands detected using antimouse HRP Abs. To detect the strep tag, the membrane was directly probed with streptavidin-HRP. IgG Purification. IgGs from serum of a subset of HCV-infected patients (resolved and chronic), where sufficient quantity was available, were purified using the MAbTrap kit (GE Healthcare), following the manufacturer’s instruction. ELISA. Ninety-six-well Nunc maxisorp plates (Thermo Fisher Scientific, Inc., Waltham, MA) were coated with 50 mL of 20 mg/mL of CD81/CLDN1 or CD9 ECL2 recombinant protein by incubating for 2 hours at 378C. After washing once with phosphatebuffered saline with Tween 20 (PBS-T), wells were blocked with PBS-T containing 5% milk for 1 hour at 378C. Subsequently, 50 mL of 20 mg/mL of mAbs specific for CD81 or CLDN1 or 100 mg/mL of purified IgG or serum, serially diluted in blocking buffer, were added and incubated for 2 hours at 378C. Bound Abs were detected by HRP-labeled anti-species Abs—antihuman HRP (for patient IgG and serum), anti-mouse HRP (for anti-CD81 Ab), and anti-rat HRP (for antiCLDN1 Ab) Abs—and TMB (3,30 ,5,50 –tetramethylbenzidine) substrate (Thermo Fisher Scientific). AntiCD81 and anti-CLDN1 mAbs11 were used as controls. For CD9 ECL2, effective coating on the ELISA plate was detected by streptavidin-HRP targeting the streptag on the protein. Specific absorbance was calculated by subtracting the value of control PBS-coated well from CD81/CLDN1 or CD9 coated well. Cut-Off Determination. Receiver operating characteristic (ROC) analysis was used to determine the optimal cutoff to calculate the endpoint titers, as described previously.24 ROC curve defines sensitivity and specificity of an assay based on the true-positive and false-positive rates. Accordingly, for our analysis, we chose healthy individuals as the negative population (n 5 50) and patients with resolved HCV as the positive
HEPATOLOGY, Vol. 62, No. 3, 2015
TAWAR ET AL.
729
Fig. 1. Expression of recombinant CD81/CLDN1 protein as an ELISA antigen. (A) Approach used for expression of the recombinant CD81/ CLDN1 protein. The two extracellular loops of CLDN1 were fused with the LEL of CD81 by (G4S)3 linker and were placed under the signal sequence for periplasmic expression. A strep-tag was added to facilitate downstream purification. (B) Expression of recombinant CD81/CLDN1 protein. Bacterial strain BL21(DE3) was transformed with pET20b1 plasmid containing the construct described in (A); protein expression was induced by adding 1.0 mM of IPTG. Periplasmic fraction was then extracted by treating cells with 1 mg/mL of polymyxin B sulfate, which was subsequently clarified by centrifugation and purified on an equilibrated strep-tactin column. Western blotting was performed using anti-CD81 Ab and streptavidin-HRP. Two different protein preparations are shown in the two lanes. LEL: long extracellular loop, ECL: extracellular loop, (G4S)3: linker sequence consisting of four glycine and one serine residues.
population (n 5 19) because these two groups exhibited the highest difference for mean optical density (OD)450nm values at 1:100 dilution. Using this information, ROC curves were generated using GraphPad Prism software (GraphPad Software Inc., La Jolla, CA). The optimal test cut-off value was determined after plotting sensitivity and specificity against OD450nm values. Sensitivity and specificity in ROC analysis are defined as the ability to correctly identify those with disease as positive and those without disease as negative, respectively. We gave similar importance to sensitivity and specificity in our assay, given that there was no reason to favor either of the two parameters, and selected a cut-off value of 0.343 corresponding to sensitivity and specificity of 72% and 74%, respectively. It should be noted that we were not attempting to develop a diagnostic test, which would require a sensitivity and specificity greater than 95%. Endpoint Titration. Anti-CD81/CLDN1 Ab titers were determined by endpoint dilution of serum samples from healthy individuals and HCV-infected patients using the optimal cut-off value chosen from ROC analysis. Serum samples from healthy individuals and HCV-infected patients were tested by 2-fold endpoint serial dilution from 1:100 to 1:1,600. Serum samples from healthy individuals that gave absorbance of
N
early 170 million people are infected with hepatitis C virus (HCV) worldwide.1 Approximately 85% of infected individuals develop persistent viral infection with a risk of developing liver
cirrhosis and hepatocellular carcinoma; the remaining 15% clear the virus spontaneously.2 Despite intensive efforts to understand the immune correlates of HCV clearance, the underlying mechanisms are still only
Abbreviations: Abs, antibodies; ANAs, anti-nuclear antibodies; ASMAs, anti-smooth muscle antibodies; CLDN, claudin; ECL, extracellular loop; ELISA, enzyme-linked immunosorbent assay; GM-CSF, granulocyte macrophage colony-stimulating factor; HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; HCV, hepatitis C virus; HCVcc, cell culture–derived HCV; HDV, hepatitis D virus; HIV, human immunodeficiency virus; HRP, horseradish peroxidase; IFN, interferon; IgG, immunoglobulin G; IL, interleukin; IPTG, isopropyl b-D-1 thiogalactopyranoside; kD, kilodalton; LEL, large extracellular loop; mAb, monoclonal antibody; OD, optical density; oxLDL, oxidized low-density lipoprotein; PBS-T, phosphate-buffered saline with Tween 20; PPI, posterior probability of inferiority; PPS, posterior probability of superiority; ROC, receiver operating characteristic analysis; SR-BI, scavenger receptor class B type I; TGF-b3, transforming growth factor beta 3; TMB, 3,30 ,5,50 –tetramethylbenzidine; TNF-a, tumor necrosis factor alpha. From the 1Inserm, U1110, Institut des Maladies Virales et H e patiques, Strasbourg, France; 2University of Strasbourg, Strasbourg, France; 3Institute for Virology, usseldorf, HeinUniversity Hospital D€ usseldorf, Heinrich-Heine-University, D€ usseldorf, Germany; 4Department of Gynecology and Obstetrics, University Hospital D€ rich-Heine-University, D€ usseldorf, Germany; 5Institute of Virology, University of Duisburg-Essen, Essen, Germany; 6Institute of Medical Virology, Helmut-RuskaHaus, Charite University Medicine, Berlin, Germany; 7P^ ole de Sant e Publique, H^ opitaux Universitaires de Strasbourg, Strasbourg, France; 8Institut HospitaloUniversitaire, P^ ole H e pato-digestif, H^ opitaux Universitaires de Strasbourg, Strasbourg, France; 9CIRI–International Center for Infectiology Research, Lyon, France; Inserm, U1111, Lyon France; Ecole Normale Sup e rieure; CNRS UMR 5308, Lyon, France; LabEx Ecofect, University of Lyon, Lyon, France; 10Department of Internal Medicine, Neurology and Dermatology, Gastroenterology and Rheumatology Clinic, Section of Hepatology, University of Leipzig, Leipzig, Germany Received December 2, 2014; accepted May 19, 2015. Additional Supporting Information may be found at http://onlinelibrary.wiley.com/doi/10.1002/hep.27906/suppinfo. 726
HEPATOLOGY, Vol. 62, No. 3, 2015
partially understood, hampering the development of vaccines and immunotherapeutics. Broad, sustained T-cell responses along with intrahepatic production of interferon (IFN)-c have been associated with spontaneous HCV clearance.3,4 Furthermore, studies have shown a strong correlation between induction of neutralizing antibodies (Abs) in the early phase of infection and viral clearance.5-7 HCV entry is a complex process mediated by the viral envelope glycoproteins, E1 and E2, and involving several host factors. Whereas some of these host factors directly interact with the viral envelope glycoproteins, others interact with one another to promote the formation of a receptor complex required for efficient internalization of the virus.8-10 Indeed, formation of the CD81-claudin-1 (CLDN1) coreceptor complex, regulated by receptor tyrosine kinase signaling, plays a central role in HCV entry.9,10 Immunologically raised Abs targeting CLDN1, scavenger receptor BI (SR-BI), and CD81 exhibit a broad, potent anti-viral effect in cellculture models of HCV infection.11-13 Although the roles of neutralizing antiviral responses in HCV infection has been extensively studied within the past years, it is currently not known whether any anti-receptor Abs are generated during HCV infection. Thus, we questioned whether anti-receptor Abs are induced during HCV infection and, if so, whether they play a role in viral clearance. Information on the quality of early immune responses is limited in HCV infection owing to the fact that most cases are not clinically apparent. This has led to a paucity of patient samples from the early phase of infection. Moreover, the heterogeneity of viral strains and patients within a cohort also represent drawbacks for the study of antiviral immune responses. In this context, the AD78 cohort is unique because it consists of a homogeneous patient group of young pregnant women infected by a single source of HCV 1b strain.14,15 Thus, we used patient samples from this well-defined AD78 cohort to assess the role of antireceptor Abs in HCV infection by employing a novel,
TAWAR ET AL.
727
sensitive enzyme-linked immunosorbent assay (ELISA) and immunoadsorption experiments.
Patients and Methods For more experimental details, please refer to the Supporting Information. Patients. Serum samples from the HCV-AD78 cohort were used with written informed consent and approval from the ethics committee, as previously described.6,16,17 The HCV-AD78 cohort consisted of 2,867 young pregnant women infected with a single genotype 1b virus-contaminated anti-D immunoglobulin G (IgG) between August 1978 and March 1979. Early- (1-6 months postinfection) and late-phase (10-17 years postinfection) serum samples from 48 patients of this cohort were available for this study.6,14,15 Spontaneous viral clearance in these patients was characterized by undetectable serum HCV RNA during the first 6 months of infection, whereas chronic infection was classified as serum that tested positive for HCV RNA as long as 17 years postinfection. None of these patients received antiviral treatment before sampling of latephase samples. Characteristics of patients are presented in Table 1. Serum samples from 30 noninfected pregnant women were obtained from the University Hospital D€ usseldorf (D€ usseldorf, Germany) after written informed consent from individuals and approval from the hospital ethics committee (study no.: #4995). Samples tested negative for hepatitis B virus (HBV), HCV, and human immunodeficiency virus (HIV)21 and 22. Serum samples from 15 patients with chronic HBV/ hepatitis delta virus (HDV) infection enrolled with written informed consent in a clinical trial (NCT01732081) were obtained from the University Hospital Strasbourg (Strasbourg, France). These patients were diagnosed by positive hepatitis B surface antigen (HBsAg) and HBV DNA in serum.18 Serum samples from 50 noninfected healthy individuals were obtained from the Etablissement Franc¸ais du Sang of Alsace and Besanc¸on, France (Convention
This study was supported through funding by the European Union (ERC-2008-AdG-233130-HEPCENT, FP7 HepaMab and Interreg IV FEDER-HepatoRegio-Net 2012), the Agence Nationale de Recherche sur le SIDA (ANRS; 2012/239, 2012/318, and 2013/108), and the Direction G e n e rale de l’Offre de Soins (A12027MS). This work has been published under the framework of the LABEX ANR-10-LABX-0028_HEPSYS and benefits from a funding from the state managed by the French National Research Agency as part of the Investments for the Future program. Address reprint requests to: Thomas F. Baumert, M.D., Inserm U1110, Institut de Recherche sur les Maladies Virales et H e patiques, Universit e de Strasbourg, 3 Rue Koeberle , Nouvel H^ opital Civil, F-67000 Strasbourg, France. E-mail: [email protected]; fax: (133) 3 68 85 37 24. C 2015 by the American Association for the Study of Liver Diseases. Copyright V View this article online at wileyonlinelibrary.com. DOI 10.1002/hep.27906 Potential conflict of interest: Nothing to report.
728
TAWAR ET AL.
HEPATOLOGY, September 2015
Table 1. Characteristics of Patients Used in this Study N Median age at infection, years Sex Source of HCV Mode of transmission HCV genotype Samples Therapy
Chronic
Resolved
29 24 Female anti-D IgG* Intravenous 1b Early†, late‡ Na€ıve
19 24 Female anti-D IgG* Intravenous 1b Early†, late‡ Na€ıve
*HCV-contaminated. † 0-6 months postinfection. ‡ 10-17 years postinfection.*0-6 months post-infection, **10-17 years postinfection
PSNT2014 Inserm). These samples were negative for HCV, HBV, HIV-1 and 22, human T-lymphotropic virus, cytomegalovirus, Plasmodium falciparum, and Treponema pallidum. Reagents. Anti-CLDN119 and anti-SR-BI20 sera have been described. Anti-CD81 serum was raised by genetic immunization of a Wistar rat.13 Anti-CD81 monoclonal antibodies (mAbs) JS-81 and 5A6 were from BD Biosciences (San Jose, CA) and Santa Cruz Biotechnology (Santa Cruz, CA), respectively. Anti-E2 mAbs AP3321 and CBH-522 have been described. The anti-HBsAg-specific mAb (NCL-HBsAg-2, clone 1044/ 341) was from Leica Biosystems (Nussloch, Germany). mAbs specific for CLDN1, SR-BI, claudin 6 (CLDN6), and claudin 9 (CLDN9) have been described.11,12 Antihuman horseradish peroxidase (HRP) and anti-mouse HRP Abs were obtained from GE Healthcare (Little Chalfont, UK). Anti-rat HRP Abs were from SigmaAldrich (St. Louis, MO). The following cytokines were used: IFN-a2a (Roche Diagnostics, Indianapolis IN); interleukin (IL)21a, IL2 (Novartis AG, Basel, Switzerland), IL4, IL6, IL7 (Cytheris, Issy-les-Moulineaux, France), IL15, and IL21 (PeproTech, Rocky Hill, NJ); tumor necrosis factor alpha (TNF-a); transforming growth factor beta 3 (TGF-b3); and granulocyte macrophage colony-stimulating factor (GM-CSF). Oxidized low-density lipoprotein (oxLDL) was obtained from Euromedex (Souffelweyersheim, France). CD81/CLDN1 Protein and Abs. Extracellular loops (ECL1 and ECL2) of CLDN1 and the large extracellular loop (LEL) of CD81 were produced as a fusion protein (Fig. 1A) in the periplasmic space of Escherichia coli BL21(DE3). To facilitate downstream purification, a C-terminus strep-tag was added. Loops were cloned into pET-20b(1) vector under the pelB signal sequence. The fusion protein was expressed in BL21(DE3) cells after induction with 1 mM of isopropyl b-D-1 thiogalactopyranoside (IPTG). Cells were
pelleted and the periplasmic fraction was isolated by treating with 1 mg/mL of polymyxin B sulfate, as previously described23; buffers were adapted for streptactin purification. The periplasmic fraction was clarified by centrifugation at 4,500 rpm for 1 hour, loaded on a preequilibrated strep-tactin column, and eluted with 2.5 mM of desthiobiotin. As a control, recombinant ECL2 of tetraspanin CD9 was expressed and purified using the same protocol. Immunoblotting. Purified protein (5 mg) was allowed to migrate on a 15% sodium dodecyl sulfate polyacrylamide gel electrophoresis gel and transferred to HyBond-P nitrocellulose membranes (GE Healthcare). To detect CD81 LEL, the membrane was probed with anti-CD81 Ab 5A6 and bands detected using antimouse HRP Abs. To detect the strep tag, the membrane was directly probed with streptavidin-HRP. IgG Purification. IgGs from serum of a subset of HCV-infected patients (resolved and chronic), where sufficient quantity was available, were purified using the MAbTrap kit (GE Healthcare), following the manufacturer’s instruction. ELISA. Ninety-six-well Nunc maxisorp plates (Thermo Fisher Scientific, Inc., Waltham, MA) were coated with 50 mL of 20 mg/mL of CD81/CLDN1 or CD9 ECL2 recombinant protein by incubating for 2 hours at 378C. After washing once with phosphatebuffered saline with Tween 20 (PBS-T), wells were blocked with PBS-T containing 5% milk for 1 hour at 378C. Subsequently, 50 mL of 20 mg/mL of mAbs specific for CD81 or CLDN1 or 100 mg/mL of purified IgG or serum, serially diluted in blocking buffer, were added and incubated for 2 hours at 378C. Bound Abs were detected by HRP-labeled anti-species Abs—antihuman HRP (for patient IgG and serum), anti-mouse HRP (for anti-CD81 Ab), and anti-rat HRP (for antiCLDN1 Ab) Abs—and TMB (3,30 ,5,50 –tetramethylbenzidine) substrate (Thermo Fisher Scientific). AntiCD81 and anti-CLDN1 mAbs11 were used as controls. For CD9 ECL2, effective coating on the ELISA plate was detected by streptavidin-HRP targeting the streptag on the protein. Specific absorbance was calculated by subtracting the value of control PBS-coated well from CD81/CLDN1 or CD9 coated well. Cut-Off Determination. Receiver operating characteristic (ROC) analysis was used to determine the optimal cutoff to calculate the endpoint titers, as described previously.24 ROC curve defines sensitivity and specificity of an assay based on the true-positive and false-positive rates. Accordingly, for our analysis, we chose healthy individuals as the negative population (n 5 50) and patients with resolved HCV as the positive
HEPATOLOGY, Vol. 62, No. 3, 2015
TAWAR ET AL.
729
Fig. 1. Expression of recombinant CD81/CLDN1 protein as an ELISA antigen. (A) Approach used for expression of the recombinant CD81/ CLDN1 protein. The two extracellular loops of CLDN1 were fused with the LEL of CD81 by (G4S)3 linker and were placed under the signal sequence for periplasmic expression. A strep-tag was added to facilitate downstream purification. (B) Expression of recombinant CD81/CLDN1 protein. Bacterial strain BL21(DE3) was transformed with pET20b1 plasmid containing the construct described in (A); protein expression was induced by adding 1.0 mM of IPTG. Periplasmic fraction was then extracted by treating cells with 1 mg/mL of polymyxin B sulfate, which was subsequently clarified by centrifugation and purified on an equilibrated strep-tactin column. Western blotting was performed using anti-CD81 Ab and streptavidin-HRP. Two different protein preparations are shown in the two lanes. LEL: long extracellular loop, ECL: extracellular loop, (G4S)3: linker sequence consisting of four glycine and one serine residues.
population (n 5 19) because these two groups exhibited the highest difference for mean optical density (OD)450nm values at 1:100 dilution. Using this information, ROC curves were generated using GraphPad Prism software (GraphPad Software Inc., La Jolla, CA). The optimal test cut-off value was determined after plotting sensitivity and specificity against OD450nm values. Sensitivity and specificity in ROC analysis are defined as the ability to correctly identify those with disease as positive and those without disease as negative, respectively. We gave similar importance to sensitivity and specificity in our assay, given that there was no reason to favor either of the two parameters, and selected a cut-off value of 0.343 corresponding to sensitivity and specificity of 72% and 74%, respectively. It should be noted that we were not attempting to develop a diagnostic test, which would require a sensitivity and specificity greater than 95%. Endpoint Titration. Anti-CD81/CLDN1 Ab titers were determined by endpoint dilution of serum samples from healthy individuals and HCV-infected patients using the optimal cut-off value chosen from ROC analysis. Serum samples from healthy individuals and HCV-infected patients were tested by 2-fold endpoint serial dilution from 1:100 to 1:1,600. Serum samples from healthy individuals that gave absorbance of
