4
Diagnosis and Control of Zoonotic Infections

PATHOLOGY AND EARLY RECOGNITION OF ZOONOTIC DISEASE OUTBREAKS

Tracey S. McNamara, D.V.M.

Head, Department of Pathology, Wildlife Conservation Society Bronx Zoo

In Zambia, in 1987, thousands of animals—hippos, giraffes, wild dogs, lions—turned up dead. What was killing them? Was there a risk for villag-ers who harvest the meat and use the skins of local animals? In New York City, in 1999, crows and zoo birds were dying. People were dying. What was killing them? Would the risk spread? In both situations, as in many others, diagnosis of the disease outbreaks relied entirely on diagnostic pathology, that is, on analyzing dead animals.

This ability of diagnostic pathology to help in recognizing and understanding diseases—both old and emerging, in humans and in animals— often is overlooked. Basic anatomic pathology involves analyzing tissues from dead specimens, making observations, interpreting those findings, and following up with histopathology studies of samples under a microscope. In recent years, the advent of molecular pathology has heightened the power of diagnostic pathology. Using such tools as immunohistochemistry, in situ hybridization, and polymerase chain reaction (PCR) assays, pathologists now can identify the etiology, or cause of death, faster than ever before and,



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary 4 Diagnosis and Control of Zoonotic Infections PATHOLOGY AND EARLY RECOGNITION OF ZOONOTIC DISEASE OUTBREAKS Tracey S. McNamara, D.V.M. Head, Department of Pathology, Wildlife Conservation Society Bronx Zoo In Zambia, in 1987, thousands of animals—hippos, giraffes, wild dogs, lions—turned up dead. What was killing them? Was there a risk for villag-ers who harvest the meat and use the skins of local animals? In New York City, in 1999, crows and zoo birds were dying. People were dying. What was killing them? Would the risk spread? In both situations, as in many others, diagnosis of the disease outbreaks relied entirely on diagnostic pathology, that is, on analyzing dead animals. This ability of diagnostic pathology to help in recognizing and understanding diseases—both old and emerging, in humans and in animals— often is overlooked. Basic anatomic pathology involves analyzing tissues from dead specimens, making observations, interpreting those findings, and following up with histopathology studies of samples under a microscope. In recent years, the advent of molecular pathology has heightened the power of diagnostic pathology. Using such tools as immunohistochemistry, in situ hybridization, and polymerase chain reaction (PCR) assays, pathologists now can identify the etiology, or cause of death, faster than ever before and,

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary in many cases, where it would otherwise have been impossible. However, current disease surveillance systems, for human diseases and zoonoses alike, fail to make adequate use of diagnostic pathology. Veterinary pathologists are highly attuned to zoonotic diseases, because most of the diseases that threaten humans—brucellosis, Q-fever, leptospirosis, tularemia, rabies, and many more—are diseases that go hand in hand with being a veterinarian. Nor is this a one-way street: some diseases of humans can be passed to animals. This animal–human link is particularly critical for veterinary pathologists who work in zoos. These individuals regularly work with a number of high-risk species, such as macaques, which can transmit herpes B virus that can be fatal to humans. In addition, zoos typically are located in urban areas, which have their own forms of indigenous wildlife that present a constant threat of introducing diseases to the zoo’s collection. These introduced diseases hold potential to spread rapidly throughout an entire herd or flock, and from there to the veterinary staff, the keeper staff, and, in the worst-case scenario, to the public. Yet only six zoos nationwide have a full-time pathologist on staff. Thus, zoo pathologists lack the generations of information, as well as the arsenal of drugs and vaccines, that other veterinarians take for granted. As a result, zoo and wildlife pathology is still very much a frontier. This means veterinary pathologists always must expect the unexpected. In today’s changing world, with unknown disease threats, this is perhaps a good model for other fields to follow. At the Bronx Zoo, this philosophy of expecting the unexpected leads us to perform a necropsy on every animal that dies and to use a variety of techniques to study tissue samples. Before making a final diagnosis, we consider all known possibilities, including diseases in both domestic animals and wildlife, as well as the chance that an unknown agent may be responsible. When our analyses are complete— with, it is hoped, a definitive diagnosis—we “bank” microscope slides of representative sections of every organ. The zoo has accumulated a massive library of samples, some dating to the 1930s, and we now are incorporating the slides on CD-ROM. We also are computerizing a variety of other data from as far back as 1895. In this way, members of the larger veterinarian community will be able to readily share these valuable, often unique, resources. By taking all of the above steps, every time and in an ordered fashion, we greatly improve our ability to reach a definitive diagnosis of etiology. This approach also adds to the scientific community’s knowledge base; enables zoo staff members—and, with the new database, other researchers as well—to carry out powerful retrospective studies; increases the zoological community’s ability to detect disease trends; and enables researchers to

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary more fully detect and define diseases that threaten captive and free-ranging wildlife. Indeed, the value of this systematic approach to diagnostic pathology was demonstrated during the 1999 emergence of West Nile virus in birds, and ultimately in humans, in New York. The Bronx Zoo played a major role in identifying the virus, which had never before been detected in the United States. In this effort, members of the zoo’s pathology department worked with several other researchers, including another veterinary pathologist and a virologist, as well as with a variety of local, state, and federal organizations, including the Centers for Disease Control and Prevention (CDC) and the U.S. Army Medical Research Institute of Infectious Diseases (USAMRIID). Several observations emerged from this experience. First, although the West Nile outbreak highlighted the need for expanded disease surveillance, there is doubt about who will be responsible for such effort. In the medical community today, many unexplained deaths are not being subjected to rigorous pathology studies. In more than 31 states, local officials, many of them not skilled in the latest techniques in diagnostic pathology, perform postmortems. Second, zoos can be important in disease surveillance. This will require more zoos to add a veterinary pathologist to their staff, as well as to train their staff members in how to carry out such studies and how to take necessary safety precautions. Third, the nation has no Biosafety Level-4 laboratories devoted to veterinary research, a situation that can impede the process of identifying unknown pathogens. Perhaps the most fundamental need is for improved collaboration and cooperation among government agencies at all levels—local, state, and federal. When our zoo’s staff first began investigating the disease outbreak that ultimately would be linked to West Nile virus, scientific exchanges were relatively free. But as more organizations and people became involved—and, ironically, as the magnitude of the problem escalated—the situation degenerated: some states seemed unwilling to work with other states or with federal agencies; some organizations did not seem willing to work with other organizations. In remedying this situation, the CDC’s new program to provide states with funding to develop strategies and capabilities to cope with bioterrorism may provide a model. In particular, Montana, North Dakota, and South Dakota have developed what appears to be an effective regional surveillance system that integrates both veterinary and human public health. Lessons from their experience may help as the nation seeks to improve its ability to detect, prevent, and control zoonotic diseases.

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary MOLECULAR AND OTHER TECHNOLOGIES FOR RAPID DIAGNOSIS OF ZOONOTIC AGENTS Alfred D. Steinberg, M.D. Consulting Physician-Scientist, MITRETEK In recent years a number of previously uncharacterized or unknown diseases have affected humans in various regions of the world. In addition, diseases previously confined to certain parts of the globe have appeared in new areas. Thus, we have a need to be able to recognize new diseases and to identify their causative agents. Some successes in this process have included the human immunodeficiency virus (HIV) as a cause of AIDS, Legionella pneumophila as the cause of Legionnaires’ disease, Nipah virus as the cause of the 1998–99 outbreak of disease in Malaysia, and West Nile virus as the agent of disease emerging in the New York City area in 1999. The most promising method of assessing new disease involves parallel and multidisciplinary approaches: Clinical recognition. Some astute individual or group of people recognizes a disorder that is not typical of any known disorder. It may be an atypical presentation for a previously described disease, or it may be a new disease. In order to foster such recognition, it is helpful to educate emergency room physicians, hospital infectious disease experts, and other individuals in the line of battle to the possibility of a new disease or one not previously encountered in the region. Standard laboratory assessments. An important finding may occur as a result of clinical recognition when special attention is given to laboratory analyses of material from subjects identified as having an unusual presentation or when astute laboratory personnel notice something unexpected in material from a patient(s) not yet considered to have an unusual disorder. Epidemiology studies. These studies may disclose the geographic region encompassed by the new disorder; the age, gender, and races of susceptible individuals; the immune status of resistant and susceptible individuals; potential vectors or animal hosts; and the incubation period and modes of transmission. Laboratory tests. When there is evidence of a new and important disease, all possible laboratory tests should be done early in the process. These tests include: Microscopy. Electron microscopy.

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary Culture. Culture everything—all mucosal surfaces, all secretions and excretions, including: Bacteria—both routine and special cultures. Viral cultures. Rickettsial cultures. Measurement of antibody titers in host sera. Antigen capture assays. Use antibodies to putative agents to identify potential pathogens. Nucleic acid amplification (RNA, DNA) by probe hydrolysis PCR or reverse transcriptase PCR and use of chip arrays. Mass spectrometry. Genetic analyses—prions, etc. Inoculation of test animals. A few additional issues of importance: Host range of pathogen. A pathogen that once infected only dogs or swine or cats but now causes life-threatening disease in humans. Change in tissue tropism. A virus that previously caused only mild enteritis now causes severe encephalitis or severe myocarditis. Immune avoidance. Large DNA viruses, such as orthopox viruses, have many genes that serve to avoid or subvert the host’s immune system. Mutations in these genes have the potential to change the host range of the virus. Even smaller viruses, including RNA viruses, may jump species as a result of mutations. And some RNA viruses, such as Rift Valley fever virus, are segmented and can recombine so as to produce new variants that evade the immune systems of both humans and animal reservoirs. Human actions on the environment. Such actions include building roads through the jungle and the movement of animals or vectors. Zoonoses not requiring a vector. For example, in the case of hantavirus Sin Nombre, the virus persists in deer mice for months despite antibodies and is transmitted to humans when they inhale virus-containing excretions from the deer mice. A few comments on laboratory tests: For viruses especially, typing sera may determine that a pathogen is in a particular virus group or family even if the specific virus is previously unknown. A polyclonal antibody reagent is far more likely to cross-react and identify an unknown than is a monoclonal typing reagent. Such relatedness helped in the diagnosis of Nipah virus in Malaysia (the new virus reacted strongly with antisera to Hendra virus but not to antisera against

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary other paramyxoviruses) and of West Nile virus in New York City (the virus reacted with antisera to other flaviviruses). Electron microscopy may point to a specific type of virus (e.g., size, shape, enveloped versus nonenveloped). Mass spectrometry is a relatively new and emerging technology for identifying bacteria and viruses. Each molecule of the organism is measured in terms of mass. As libraries of patterns for various organisms become available, mass spectrometry will become a more and more powerful analytical tool for pathogen identification. PCR reactions can use primers that will amplify even distantly related members of a virus family. PCR can be coupled with sequencing of the amplicons to provide additional information on relatedness to known viruses. In summary, a systematic approach to analysis of specimens from individuals suspected of having a novel zoonosis will facilitate pathogen identification. The higher the index of suspicion early on in the outbreak, the greater the likelihood of identifying the pathogen in time for suitable public health measures. METHODS AND MODELS FOR PATHOGEN DISCOVERY W. Ian Lipkin, M.D. Louise Turner Arnold Professor of Neuroscience, Director, Emerging Diseases Laboratory University of California-Irvine Director, Laboratory for Immunopathogenesis and Infectious Diseases, and Director, Center for Developmental Neuroscience Mailman School of Public Health, Columbia University Establishing a causal relationship between infection with an infectious agent and a specific chronic disease may be complex. In most acute diseases, the responsible agent is readily implicated because it replicates at high levels in the affected tissue at the time the disease is manifest, morphological changes consistent with infection are evident, and the agent is readily cultured with standard microbiological techniques. Implication of infectious agents in chronic diseases may be confounded because persistence requires restricted gene expression, classical hallmarks of infection are absent, and/or mechanisms of pathogenesis are indirect or subtle. Methods for cloning nucleic acids of microbial pathogens directly from clinical specimens offer new opportunities to investigate microbial associations in chronic

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary diseases. The power of these methods is that they can succeed where methods for pathogen identification through serology or cultivation may fail due to absence of specific reagents or fastidious requirements for agent replication. Various methods are employed or proposed for cultivation-independent characterization of infectious agents. These can be broadly segregated into methods based on direct analysis of microbial nucleic acid sequences (e.g., DNA microarrays; consensus polymerase chain reaction, or PCR; representational difference analysis; differential display), direct analysis of microbial protein sequences (e.g., mass spectrophotometry), immunological systems for microbe detection (e.g., expression libraries, phage display), and host response profiling. Over the past decade, the application of molecular pathogen discovery methods resulted in identification of novel agents associated with both acute and chronic diseases, including Borna disease virus, hepatitis C virus, Sin Nombre virus, HHV-6, HHV-8, Bartonella henselae, and Tropherema whippeli. Despite these achievements and the fact that novel pathogens and new relationships between known pathogens and diseases are dependent on this type of research, few academic scientists pursue work in the field of pathogen discovery. Funding is limited and pathogen discovery projects are high risk. In contrast to hypothesis-driven projects where virtually all outcomes are reportable, pathogen discovery projects may fail to yield meaningful data within the timeframe of a traditional 3- to 5-year fellowship or grant allocation. Pathogen discovery in central nervous system diseases is particularly challenging. Syndromes may blend, particularly in psychiatry. Thus, associations between infectious agents and diseases may be obscured unless discrete analyses of specific syndromic subsets are performed. Another difficulty is access to clinical material. One cannot readily biopsy brain or spinal cord. It is also difficult to obtain short postmortem interval tissue. Therefore, human materials are frequently not optimal for microbiological or molecular biological studies. Additional potential confounders include the fact that an agent may cause damage through an indirect mechanism without replicating in the tissue showing signs of disease, or it may disappear by the time tissue becomes available for analysis (“hit-and-run” mechanisms for pathogenesis). Nucleic acid-based methods for pathogen detection can be broadly defined as binary or not. Subtractive cloning methods like representational difference analysis are ideal for examining a homogeneous population such as a tumor or a cell line, or animal models exposed to an environmental factor such as an infectious agent or toxin, where pre- and postadministration comparisons are feasible. They may fail, however, to elucidate polyfactorial disorders in which some individuals may be infected but not show disease or where more than one agent can result in a similar disorder. To

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary address such nonbinary problems, the investigator needs methods that enable recognition of patterns that may not be uniform in disease and control samples. The approach we have used for such population analyses is based on consensus PCR and the differential display method of Liang and Pardee. Radiolabeled cDNA libraries are prepared by PCR from different nucleic acid populations using specific but arbitrary primers. The libraries are then size fractionated in acrylamide gels or by using capillary electrophoresis. Patterns of amplification products from several individuals with or without disease are displayed in side-by-side comparisons. Those that appear to be specific for the disease are sequenced and used to probe databases to yield a variety of outcomes: identification as known host gene, known pathogen, or something novel. This method has been modified to emphasize virus detection using primers representing viral families. The first practical test of this method, termed domain-specific differential display, was in the context of the CDC’s Unexplained Encephalitis Project, a collaborative network comprised of investigators in New York, California, and Tennessee, as well as investigators at CDC’s laboratories in Atlanta. This group has determined that despite use of state-of-the-art microbiological methods, 50 percent to 70 percent of encephalitis is unexplained. In 1999 the New York State Department of Health asked that we examine brain samples from victims of what was then described as a St. Louis encephalitis outbreak, because efforts to isolate virus or viral nucleic acids had been unsuccessful. Within 3 days of receipt, flaviviral sequences were amplified from four of five patients. At approximately the same time, work was pursued with a cultured bird isolate by investigators at CDC’s laboratories in Fort Collins, Colorado. The two independent lines of evidence converged on September 24, when the information from our group, based on sequence from human brain, and from the Fort Collins group, based on analysis of a cultured bird isolate, indicated that the New York City outbreak was a zoonosis due to West Nile virus. Neither group initially called the agent West Nile virus. It was called West Nile-like virus by the Fort Collins group. We called it a Kunjin/West Nile-like virus. The Kunjin/West Nile distinction is now moot because Kunjin virus was subsequently classified as a West Nile virus. After the entire viral genome was sequenced, it became apparent that the New York virus was virtually identical to a lineage I West Nile virus isolated from a goose in Israel in 1998. In contrast to West Nile virus, Borna disease virus, an agent associated with persistent infection, required considerably more effort to isolate, characterize, and implicate in human disease. Our experience in these two systems illustrates the differences in strategy required to investigate the pathogenesis of acute and chronic diseases. Borna disease was originally described as an encephalitis of ungulates in the early 1800s. Nonetheless, interest was modest until immunoreactivity to the virus in subjects with

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary bipolar disorder was reported in 1985. In response to this finding, after classical methods for isolation had failed, we cloned Bornaviral nucleic acids by subtractive hybridization. We reasoned that introduction of tools based on molecular reagents would rapidly resolve the prevalence of Bornavirus infection and its role in human disease; however, this was not to be the case. Since the first reports of human infection based on nucleic acid studies appeared in 1995, Bornaviruses have been implicated in an improbably long list of disorders, including major depressive disorder, bipolar disorder, panic disorder, multiple sclerosis, amyotrophic lateral sclerosis, schizophrenia, and chronic fatigue syndrome. Most of these reports are based on nested PCR, a sensitive method, yet prone to artifact. Because Bornavirus sequences are highly conserved, it is difficult to recognize contamination of assays with laboratory strains. To address these challenges, the National Institutes of Health is supporting a multicenter project comprised of investigators in North America, Europe, and Asia, wherein samples are collected from subjects and yoked case controls following standardized interviews, encoded, and shipped to a central site for blinded analysis using automated molecular and serological methods. The structure of this study provides a model for research into the pathogenesis and epidemiology of chronic infectious diseases. The future of microbial epidemiology in public health and clinical medicine depends on development of new tools for rapid, sensitive, molecular detection of infectious agents. Toward this end, we are establishing software programs to facilitate automated retrieval, filtration, and curation of microbial sequences from databases. In turn, these bioinformatics data will be used to design hybridization reagents and optimize methods for signal amplification. Beyond the differential display and real-time PCR methods described above, we are now emphasizing higher throughput strategies, including DNA microarrays and multiplexed, bead-based, flow cytometric assays. The application of these types of technologies to acute diseases will be straightforward. More challenging will be investigation of chronic illnesses, such as neoplastic, rheumatologic, and neuropsychiatric disorders, where host genetic context, toxins, nutritional status, and timing of exposure to an agent (e.g., multiple sclerosis or schizophrenia) may influence expression of disease. A key resource in the success of such research will be large banks of well-characterized clinical materials collected prospectively over the lifespan. Equally important will be an openness within the biomedical community to invest in new concepts in microbial pathogenesis.

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary VACCINES FOR EMERGING ZOONOSES: MARBURG VIRUS PARADIGM Alan L. Schmaljohn,* Ph.D. Chief, Department of Viral Pathogenesis and Immunology Virology Division U.S. Army Medical Research Institute of Infectious Diseases A well-educated immune system is the body’s last and best defense against viral disease, and vaccines were almost singularly responsible for reducing the impact of many human viruses in the United States during the 20th century. However, the complete cycle of vaccine development and licensure is time-consuming and relatively expensive: one published estimate cites 10 to 18 years and $30 million to $100 million per vaccine, when success proves achievable at all; other estimates are higher. The time and cost are expected to be reduced if unambiguous precedents for scientific rationale, safety, manufacture, potency testing, and assessment of efficacy have been established with one or more closely related agents. For the total array of emerging viral zoonoses, the stockpiling of licensed vaccines is scientifically daunting, a controversial use of resources, and likely unrealistic politically. Marburg virus (MBGV) provides an archetype that is useful in considering many of the issues involved in the development of vaccines for emerging viruses, including viral zoonoses. The first issue regarding development of any vaccine is determining actual need. For MBGV the answer emerges from a series of questions. First, is the virus currently a major global health problem? No. While there is an ongoing outbreak with high mortality rates in the Democratic Republic of Congo, other outbreaks have been localized, sporadic, and of low global impact. Second, does MBGV have significant potential for mutation to pandemic spread? Unknown. Human-to-human spread occurs but has proven self-limiting to date. Third, does MBGV have potential for import and establishment in new ecological niches, including disruption of agriculture? Unknown. The virus’s host species has not been identified, and domestic species have not been tested. Fourth, does MBGV pose a risk to laboratory and health care workers? Yes, but the number of people at risk is relatively small, and risk can be minimized by safe laboratory procedures *   Opinions and statements made herein are solely those of the author and are not to be construed as official USAMRIID or Department of Defense positions or policy unless supported as such by separate documentation.

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary and appropriate barrier medical care. Fifth, is MBGV a potential strategic, tactical, or terrorist weapon? Yes. The virus is highly infectious and stable in aerosol form, causes a terrifying disease with high morbidity and mortality rates, can be readily grown using low-tech methods, and exhibits significant if ultimately limited human-to-human spread. Sixth, are there any known therapies or alternatives to vaccine? No, none that have proven effective in nonhuman primates against either MBGV or Ebola viruses; state-of-the-art palliative care may reduce mortality rates. And seventh, what is the cost of ignoring the risk altogether, deprioritizing even the discovery phase of vaccine development that could establish a rationale and seed source for manufacture of an efficacious vaccine? Incalculable. The fifth question and the seventh question, along with published reports that MBGV has already been weaponized, are sufficient for the Department of Defense to seek countermeasures. Responsibility for developing such countermeasures, including a vaccine, naturally falls to USAMRIID. In 1995, USAMRIID began to determine the requirements for a safe and efficacious MBGV vaccine. We anticipated a number of problems. First, because of its high human mortality rate and known aerosol infectivity, MBGV is restricted to Biosafety Level 4 laboratories; this automatically restricts the sites in which live virus research can be conducted and increases the overall time and expense. Second, there was a paucity of baseline scientific information to guide vaccine development: protective antigens had not been identified, the nature of protective immunity was almost completely unknown, a nonhuman primate model of viral disease was incompletely understood and a guinea pig model incompletely developed, and prior reports of classical (formalin-inactivated) vaccine formulations had failed to protect more than half of animals. Third, an MBGV vaccine should protect against all routes of infection, including aerosol infection, since this is a potential route of infection in both laboratories and hospitals, as well as an obvious risk with intentionally released virus. Fourth, MBGV is a single virus only in a taxonomic sense: it actually is a constellation of antigenically distinct viruses, all highly pathogenic, and a vaccine should be efficacious against all. Fifth, related to the sparse understanding of immunity, there were no established predictors or correlates of immune status. Sixth, due to the unpredictability of outbreaks and the severity of disease, it will be either impossible or unethical to test any vaccine’s efficacy in humans in a placebo-controlled fashion; this means that licensure must ultimately proceed under new guidelines issued by the Food and Drug Administration for such circumstances. While the problems are formidable, a number of factors also favor successful vaccine development. First, this is a transitional period in vaccine technology, with several new strategies and platforms available to the vaccinologist. Second, it has been relatively easy to establish two animal

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary models, guinea pigs and nonhuman primates, in which viral disease not only approaches 100 percent lethality within 14 days but also resembles human disease. Third, there is a relatively solid mandate and funding in an institute that has the research expertise and vaccine pragmatism to discover and develop a vaccine. Fourth, as a not-for-profit venture, there are relatively few inherent biases that necessarily exclude any promising vaccine approaches. Given the paucity of previous information on the subject, these new efforts provided a rationale and a logical need to look simultaneously at several possible vaccine approaches, ranging from the classical to the most newly available. This provided a limited opportunity for direct comparison of several approaches against a single agent. To discern further the relative merits of different vaccine approaches, we examined the capacity of each prototypical vaccine to protect guinea pigs not only against subcutaneous infection but also against aerosol infection. Classical Viral Vaccines Classical viral vaccines have exploited either killed viruses as immunogens or live but relatively benign “attenuated” versions of the virus. For MBGV, the efficacy of using killed viruses was in doubt: tests had shown that inactivated MBGV protected only about 40 percent of guinea pigs and 50 percent of nonhuman primates, an observation consistent with the dubious efficacy of killed Ebola virus vaccines. Taking a somewhat different approach to using killed viruses, we recently showed that purified, irradiated MBGV administered in adjuvant (an additive intended to potentiate immune responses) could evoke solid immunity in small numbers of guinea pigs. However, the immune response to killed viruses may well be qualitatively different and possibly inferior to that elicited by live virus infection. Thus, while it seems highly unlikely that a live-attenuated MBGV vaccine could be proven acceptable for human use, it is useful for experimental purposes to test whether such a vaccine might elicit a qualitatively superior form of immunity. To this end, we have obtained a derivative of MBGV that causes viremia but no overt disease or deaths over a wide range of doses in one particular strain of guinea pigs, providing a useful virus with which to immunize animals and confer a “natural” form of immunity. Baculovirus Recombinant, Soluble Envelope Antigen Only a single MBGV antigen, the glycoprotein (GP), is known to reside on the exterior of virions and the surfaces of MBGV-infected cells. We used a variety of recombinant DNA methods to produce GP and other antigens of MBGV to test as vaccines. In tests with guinea pigs, soluble GP antigen,

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary made by a recombinant insect virus in insect cells, could immunize most of the animals against lethal subcutaneous infection, though some 20 percent of the animals remained unprotected. This observation became the basis for testing GP as a protective antigen that could be delivered via either of two new alternative vaccine approaches: DNA immunization or packaged RNA replicon. DNA Vaccines DNA immunization involves the delivery into animals of DNA that contains genes for foreign antigens, uptake of the genetic material into cells and their nuclei, and expression of the vaccine antigen in much the same manner that other cell proteins are made (that is, DNA makes RNA makes protein). In our work with the MBGV glycoprotein, we chose as our leading approach and benchmark the “gene gun”—that is, delivery of genes into skin cells ballistically, on gold microspheres. Our experience, as well as a variety of other published results, has confirmed this approach to represent fairly the potency of this antigen delivery method. This is a relatively new technology that is in a state of continual improvement. RNA Vaccines Using the same GP antigen gene as for the DNA immunization, this approach bypasses the DNA requirement by reshaping a virus that consists of RNA and copies itself and its proteins in a wholly RNA mode (thus, RNA makes RNA makes protein). We have focused on a molecule of RNA, called an RNA replicon, derived from the virus that causes Venezuelan equine encephalitis (VEE). In brief, this RNA replicon directs synthesis of its own polymerase within the cytoplasm of cells, replicates itself in a single cell, and produces abundant quantities of a second RNA that, in turn, produces the MBGV glycoprotein. For delivery as a vaccine, the replicon is provided the coat proteins of VEE so that the RNA can easily enter cells; however, since the genes for the VEE coat are not packaged, the replicon particles can infect only one set of cells, produce MBGV antigen, and spread no farther. Our studies suggest that genetic approaches—DNA vaccines and replicon-based vaccines among them—are particularly attractive, for several reasons. Such vaccines avoid the production biohazard and safety issues of both the classical killed and lived attenuated vaccines. Safety, production, and quality assurance issues, once optimized for a single vaccine, are expected to be highly similar for each succeeding vaccine that uses the same technology. The vaccines focus the immune system’s attention on one or more important antigens, thus avoiding incorporation of superfluous or

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary even harmful antigens. They express antigen via host cellular pathways and thus effectively elicit several types of immune system response, including both the production of T cells and antibodies. They are theoretically amenable to simultaneous delivery of vaccines to multiple unrelated pathogens. If insufficiently potent on their own, the vaccines may provide immunological priming that improves the performance of a “boosting” antigen. Despite major conceptual breakthroughs and proofs of vaccine feasibility that we have published recently on the above approaches, much remains to be done in bringing a safe and effective MBGV vaccine to fruition. There are questions, for example, about the basic formulation of the ultimate vaccine (precisely which antigens should be used?), about immunology (what markers predict efficacy?), and about the optimum delivery system (the RNA replicon is the leading candidate because of its impressive protection of nonhuman primates, but it is not the certain winner). There also remain important unresolved issues regarding development of the manufacturing process (how can production best be scaled up?), and numerous regulatory requirements will have to be satisfied before a new vaccine can enter general use. Based on this work in developing a Marburg virus vaccine, combined with lessons emerging from work on other vaccines, a number of needs stand out if the nation is to move ahead in preparing to deal with zoonotic diseases: Public support is needed, and this should be rooted in science education. Education focused on emerging viruses and viral zoonoses is critical and should be expanded. An alarmed but poorly informed Congress, responding to an alarmed but poorly informed public, will rarely if ever craft anything but an inefficient emergency response. Broad support of vaccine discovery efforts is crucial. The threat of emerging zoonoses will be reduced if there is preexisting knowledge of the antigens required in a vaccine and of the immune responses to be elicited. A cadre of experts in these diseases cannot be maintained, nor steady progress made in vaccine development, without consistent support. In an emergency, the experiments to obtain such answers cannot be done more quickly than a succession of vaccine regimens and viral incubation periods, results could be unexpected, and the delay could be catastrophic. Emphasize vaccinology, and reward basic research efforts that team to make actual vaccine progress. The National Institute of Allergy and Infectious Diseases took a significant step in this direction in 1999 by requesting applications for “Vaccine Immunology Basic Research Centers.” However, this is a relatively modest initiative, and its emphasis is on vaccine immunology and high-impact human diseases, not on emerging or zoonotic diseases.

OCR for page 64
The Emergence of Zoonotic Diseases: Understanding the Impact on Animal and Human Health - Workshop Summary Set priorities for vaccine manufacture. The lengthy process of advanced vaccine development and licensure should be reserved for agents for which vaccines are thought to be needed within the next several years. For other agents, secure the scientific underpinnings. The antigen requirements of a modern vaccine will remain relatively unchanged once discovered and well defined, as will the phenotype of a protective immune response. However, the optimal vaccine platforms by which such antigens will be constructed and delivered may change almost as rapidly as computer hardware.