| |
Team Identifies Human Antibodies that Prevent Hepatitis C Virus Infection in Mouse Model
|
| |
| |
By Mika Ono Benedyk
"A key question is whether broadly neutralizing AR3-specific antibodies can protect against infection by heterologous HCV quasispecies ......despite the enormous diversity of HCV, the prospects for developing a vaccine against this virus, perhaps targeting both conserved B and T cell epitopes11, 12, seem favorable."
A team of researchers has found that certain antibodies can prevent hepatitis C virus infection in a "humanized" mouse model, opening the door to the use of antibodies as a human therapeutic and the development of a preventive vaccine for the disease.
The work was published in the advance, online edition of the journal Nature Medicine on December 6, 2007. See full text below
The study's authors, who include Scripps Research Professor Dennis Burton, University of Alberta Professor Norman Kneteman, and an international team of colleagues, identified a group of special antibodies that can broadly neutralize hepatitis C virus. Antibodies (proteins produced by the body's immune system in response to a foreign substance) are considered to be broadly neutralizing when they are effective against many different strains of a pathogen.
"The findings that neutralizing human antibodies attack a region commonly found on the surface of many hepatitis C virus strains and [that these antibodies] protect against infection by the virus in an animal model provide proof of principle for the viability of passive immunotherapy against this extremely variable virus," says Mansun Law, a research associate at Scripps Research who was first author of the paper. "The results also raise hopes for the development of an effective vaccine to prevent the disease from taking hold."
Preventing Hepatitis C Virus Infection
Hepatitis C is a potentially fatal disease of the liver caused by the hepatitis C virus, which is spread by contact with the blood of an infected person. Those at highest risk include people with a history of injecting illegal drugs, participating in unsafe sexual practices, receiving a blood transfusion or solid organ transplant before 1992, or being on long-term kidney dialysis. According to the Centers for Disease Control and Prevention (CDC) and the World Health Organization (WHO), an estimated 4.1 million Americans have been infected with the virus and two to three percent of the world's population is chronically infected.
The course of the disease varies. A small fraction of people contracting the virus develop acute viral infection but produce a strong enough immune response to clear it. Most people contracting the virus, however, become chronically infected without being aware they have the disease. Chronic infection can result in loss of liver function, cirrhosis (scarring of the liver), and liver cancer.
While treatment of chronic hepatitis C infection with a combination therapy of pegylated interferon and the anti-viral drug ribavirin is successful about half the time, hepatitis C remains the leading cause of liver cancer and the leading indication for liver transplants in the United States.
"There is an urgent need to discover means to prevent new infection," said Law, "and to treat those infected [more effectively]."
Currently, no vaccine exists to prevent hepatitis C. A major problem in developing vaccines has been the virus's extreme variability. For a vaccine to be practical, it must evoke antibodies that counteract many of the strains of the hepatitis C virus circulating in the human population, not only a small subset.
In the current Nature Medicine study, for the first time scientists were able to identify just such antibodies, define the regions on the virus that these powerful antibodies target, and demonstrate their activity in animals.
The researchers began the study by isolating a panel of antibodies from a donor chronically infected with hepatitis C virus. Using cell culture-based assays, the researchers challenged each of these antibodies with a variety of strains of hepatitis C virus to see which, if any, were effective across the board.
The results showed that broadly neutralizing antibodies do exist, and that they target a part of the hepatitis C virus that the researchers called antigenic region 3. The other antibodies, which the researchers isolated more frequently, failed to neutralize as many different strains and targeted other parts of the virus, antigenic regions 1 and 2.
Law notes that antigenic region 3 is a part of the virus that is "relatively conserved," in other words unchanged despite many other mutations that distinguish among different strains. Often, a region of a virus is conserved because it is necessary for the virus's survival, for example, essential for the ability to reproduce.
In the next part of the study, the researchers tested the antibodies that were broadly neutralizing in cell culture in a small animal model designed to mimic human response-mice with livers containing human liver cells. One group of these mice received an injection of broadly neutralizing antibodies, while a control group did not. Both groups were then exposed to a high dose of hepatitis C virus quasispecies from another infected donor. Virus quasispecies, or a population of closely related virus strains, are produced during an infection by viruses that have an exceptionally high replication and mutation rate like hepatitis C virus. The production of virus quasispecies is a powerful mechanism for viruses to escape immune surveillance and anti-viral drugs.
The mice who had received the broadly neutralizing antibodies were either fully protected from contracting the disease ("sterilizing protection") or protected until levels of the antibody decayed. All animals in the control group were infected and unable to prevent the virus from proliferating.
Toward Vaccines and Therapies
The researchers plan to follow up on these results, pursuing potential vaccine candidates, developing antibodies that might be administered directly as a therapy in a technique called "passive immunization," and laying groundwork for future drug development.
On the vaccine front, the scientists will investigate how to induce the body to produce the specific type of broadly neutralizing antibodies discovered in the recent study. On the passive immunization front, the researchers will see if they can increase the antibodies' anti-viral activity. In addition, the team will seek to identify additional conserved targets on the virus, which could prove useful for vaccine, antibody, and drug development.
"In order to cure highly variable viruses, such as hepatitis C virus and HIV," said Law, "it will probably be necessary to use a combination of highly potent antibodies and anti-viral molecules to attack multiple conserved regions in the virus."
Law believes that antibody administration alone would be most useful in cases in which a health care worker has been accidentally stuck with an infected needle, or after a liver transplant, to prevent re-infection of the new organ.
The study was not supported by specific funding agencies, but partial salary support was provided by the Elizabeth Glaser Pediatric AIDS Foundation, the Sjogren's Syndrome Foundation, the Arthritis Foundation, the Alberta Heritage Foundation for Medical Research, and a Wyeth Canada-Canadian Institutes for Health Research Clinical Research Chair in Transplantation.
Brief Communication
Nature Medicine 14, 25 - 27 (2007)
Published online: 6 December 2007 | doi:10.1038/nm1698
Broadly neutralizing antibodies protect against hepatitis C virus quasispecies challenge
Mansun Law1, Toshiaki Maruyama1,9, Jamie Lewis4, Erick Giang1, Alexander W Tarr5, Zania Stamataki6, Pablo Gastaminza3, Francis V Chisari3, Ian M Jones7, Robert I Fox8, Jonathan K Ball5, Jane A McKeating6, Norman M Kneteman4 & Dennis R Burton1,2
Abstract
A major problem in hepatitis C virus (HCV) immunotherapy or vaccine design is the extreme variability of the virus. We identified human monoclonal antibodies (mAbs) that neutralize genetically diverse HCV isolates and protect against heterologous HCV quasispecies challenge in a human liver-chimeric mouse model. The results provide evidence that broadly neutralizing antibodies to HCV protect against heterologous viral infection and suggest that a prophylactic vaccine against HCV may be achievable.
We isolated a total of 115 clones showing specific binding to HCV E2 glycoprotein from an antibody antigen-binding fragment (Fab) phage-display library generated from a donor chronically infected with HCV (Supplementary Methods online). DNA sequence analysis identified 36 distinct Fabs with 13 unique heavy chain sequences (Supplementary Fig. 1 online). The binding properties of soluble Fabs prepared from the phage-Fab clones were characterized (Supplementary Fig. 2a-c online) to allow the Fabs to be sorted into three groups recognizing three antigenic regions of HCV E2 (Supplementary Fig. 3 online).
A total of seven Fabs from different heavy-chain groups recognizing the three different antigenic regions were converted into full-length IgG1s (Table 1a). All recombinant mAbs bound the genotype 1a HCV E1-E2 complex with approximately similar apparent affinities, in the range of 0.4-6 nM, but only antigenic region 3 (AR3)-specific mAbs reacted with genotype 2a HCV, suggesting that epitopes in AR3 are highly conserved. mAbs AR1A and AR3A-D inhibited the binding of E1-E2 to the virus co-receptor CD81 (refs. 1,2) at nanomolar concentrations, suggesting that these antibodies could potentially block HCV interaction with CD81 and thereby inhibit infection. We then studied the neutralizing activities of the mAbs against cell-culture HCV (HCVcc), JFH-1 virus3 and a panel of HCV pseudotype virus particles (HCVpp)4, 5 displaying E1-E2 from diverse genotypes (Table 1b). We observed, first, that antibodies that bind E2 in an ELISA did not necessarily neutralize the corresponding virus. The AR1-specific antibodies bound recombinant E1-E2 from genotype 1a HCV isolate H77 with a similar or higher affinity than AR3-specific antibodies, but they did not neutralize the virus, suggesting that the AR1 epitopes are available on isolated envelope proteins but not on infectious virions. Of note, the Fab fragments of antibodies AR1A and AR1B (that is, B2 and D1, Table 1a) did neutralize HCVpp-H77 (Supplementary Fig. 4 online), indicating that steric hindrance, possibly by E1 (Supplementary Fig. 2a), prevents virus neutralization by whole AR1-specific antibodies. Second, the ability of the antibodies to inhibit E1-E2 binding to CD81 in the 'neutralization of binding' assay6 did not fully predict virus neutralization. Third, and most notably, the AR3-specific antibodies bound E1-E2 from both genotypes 1a and 2a at nanomolar affinities and cross-neutralized JFH-1 virus and many HCVpps tested. These results show that AR3 is a relatively conserved neutralizing site on HCV E2.
The antigenic regions were mapped by competition ELISA (Fig. 1a) and alanine-mutagenesis scanning (Fig. 1b). The results confirm the broad designation of the antigenic regions and suggest that the discontinuous epitopes in AR3 are formed by at least three segments between amino acids 396-424, 436-447 and 523-540; the first and third segments also contribute to the CD81-binding domain of E2 (ref. 7), and the conserved residues Ser424, Gly523, Pro525, Gly530, Asp535, Val538 and Asn540 (ref. 7) are probably involved in the binding of the AR3-specific antibodies (Fig. 1c).
A key question is whether broadly neutralizing AR3-specific antibodies can protect against infection by heterologous HCV quasispecies. As a first step to evaluate the mAbs and establish the essential parameters for passive antibody protection, we used the human liver-chimeric Alb-uPA/SCID mouse model8, 9. Although this animal model is not suitable for studying virus pathogenesis, owing to its lack of a functional adaptive immune system, the question of whether antibodies can protect against HCV challenge seems appropriate. We first established the kinetics and tolerability of the antibodies AR3A AR3B and a human isotype control IgG1 to HIV-1, b6, in the model. The antibodies did not show adverse effects in control mice, and a dose of 200 mg/kg given through intraperitoneal injection was required to achieve mean serum titers approximately 100 times higher than in vitro neutralization titers (Supplementary Fig. 5 online). Such titers have previously been found necessary to achieve sterilizing immunity in other viral disease models. The half-lives of mAbs AR3A, AR3B and b6 are 6.0 plusminus 2.2 d, 9.0 plusminus 1.3 d and 7.3 plusminus 1.8 d (mean plusminus s.d.), respectively, and their specific neutralizing activities (that is, neutralizing activity relative to serum mAb concentration) are stable for at least 10 d in the mice (Supplementary Fig. 6 online).
In the passive transfer experiments, we administered the mAbs intraperitoneally to mice with high levels of human liver chimerism (Supplementary Methods), and the mean serum titers of mAbs AR3A, AR3B and the control mAb b6, at 24 h after injection were approx2.5 plusminus 0.3 mg/ml, 3.1 plusminus 0.5 mg/ml and 2.6 plusminus 0.3 mg/ml, respectively (Supplementary Fig. 7 online). To simulate a natural human exposure to virus, we inoculated genotype 1a HCV-infected human serum (Supplementary Fig. 8 online) intravenously into the mice. Infection was monitored by assessing serum viral load up to 6 weeks after inoculation (Fig. 1d). Protection in this mouse model is defined as the absence of serum HCV RNA as detected by quantitative PCR at or after 6 d after virus challenge. All mice in the control group (n = 4) were infected, and serum viral load was maintained at >10,000 RNA copies/ml until the completion of the study. In mice that received mAb AR3A (n = 5) or AR3B (n = 4), HCV was detected the day after challenge in five of nine mice but was cleared 6 d after virus challenge. High levels of HCV RNA were detected in four mice between weeks 2 and 4, indicating virus replication concurrent with the decay of antibody in these mice. By week 6, when the mAbs would have decayed to <10% of the initial serum level (Supplementary Figs. 5 and 7), two of five mice receiving mAb AR3A and three of four mice receiving mAb AR3B were still protected. The protection was highly significant compared to the isotype control antibody group (two-tailed log-rank test: AR3A, P = 0.0298; AR3B, P = 0.0171). The experiments ended at week 6 because two mice became morbid and were killed on day 41 and day 45, respectively, but the remaining mice were monitored to week 8, and a signal below the sensitivity of the quantitative PCR assay (6.0 times 102 international units/ml) was noted in one additional mouse in each neutralizing antibody-treated group (mice N681 and N697).
In summary, (i) it is possible to use mAbs against AR3 to protect from challenge with a heterologous HCV quasispecies swarm, consistent with the notion that AR3 is the principal conserved neutralizing antibody determinant of HCV; (ii) high concentrations of the mAbs were required for protection, suggesting that more potent antibody preparations will likely be required in immunotherapy, but that the mAbs described will be useful for comparative in vitro studies with newly identified mAbs and combinations of mAbs; and (iii) considering that one-third of the 115 phage-Fab clones isolated in this study are AR3 specific and are diverse in their heavy-chain sequences (Supplementary Figs. 1, 2, 3), and similar mAbs were isolated from different HCV-infected donors elsewhere (Fig. 1a)10, AR3 seems to be relatively immunogenic in humans and thus a favorable target for vaccine design. So, despite the enormous diversity of HCV, the prospects for developing a vaccine against this virus, perhaps targeting both conserved B and T cell epitopes11, 12, seem favorable.
|
|
| |
| |
|
|
|
