Jump to ContentJump to Main Navigation
In Defense of SelfHow the Immune System Really Works$

William R. Clark

Print publication date: 2008

Print ISBN-13: 9780195336634

Published to Oxford Scholarship Online: September 2008

DOI: 10.1093/acprof:oso/9780195336634.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: null; date: 17 September 2019

First Defense: The Immune System and Bioterrorism

First Defense: The Immune System and Bioterrorism

(p.221) 14 First Defense: The Immune System and Bioterrorism
In Defense of Self

William R. Clark

Oxford University Press

Abstract and Keywords

During the first 48–72 hours of a bioterrorist attack using pathogenic microbes, the immune system will be our primary means of defense against potentially fatal disease. This chapter looks at the pathogens identified by the CDC as most likely to be used as bioweapons — anthrax, plague, smallpox, botulin toxin, tularemia, and Ebola-like viruses — and what is known of the human immune response to them. Progress in enhancing our ability to deal with these pathogens is also discussed, in particular strategies that go beyond standard vaccines.

Keywords:   anthrax, plague, small pox, humanized antibodies, innate immunity, Ebola, bioweapons

Bioterrorism is the use of biological organisms or their derivatives to sow terror in a civilian population. Bioterrorism is an offshoot of biological warfare, and like most progeny it differs from its parent. The main difference is that biological warfare is a highly organized aggressive activity carried out by one state against another, usually through a military arm, with the sole aim of killing or disabling people. Bioterrorism, while using many of the same agents and tactics as biological warfare, is a more ad hoc activity carried out by individuals or political groups against other political groups or states, with a mixture of objectives.

Biological warfare itself has a long if occasionally crude history, including dipping arrowheads and spear points into rotting cadavers or feces, or lobbing entire diseased corpses over town or castle walls. The perpetrators obviously had little understanding of what they were doing, so it may be less than accurate to call this biological warfare.

But once the basis for infectious diseases was uncovered in the second half of the nineteenth century, it didn't take long before biological warfare became a highly precise science. By World War I, and on through World War II, virtually every major world power had established scientific research units dedicated to the subject. In the United States the War Department (precursor of today's Department of Defense) established a special biological warfare facility at Fort Detrick, Maryland. Anthrax and plague were among the microbes of choice in the programs of most countries.

However, with the exception of Japan during its occupation of China and Manchuria, there was no extensive use of biological (p.222) agents against either military or civilian targets during World War II. President Nixon ended the active development of biological warfare agents in the United States in 1969. And finally, in 1972, over 100 countries signed a Biological Weapons Convention that outlawed biological weapons and mandated destruction of existing weapon stockpiles.

Bioterrorism has a more limited history. The first documented instance of bioterrorism in the United States was carried out by an Oregon cult (the Rajneeshees) in 1984, in an attempt to manipulate a local election. Over 700 people were made ill with Salmonella bacteria, though none died. In the early 1990s, the Japanese Aum Shinrikyo cult released anthrax spores in several Japanese urban settings, with fortunately few casualties. Shortly after the September 11 attacks in the United States, unknown individuals used the Postal Service to disseminate anthrax spores in letters.

The FBI had uncovered and foiled two additional bioterrorist plots before then. In 1992, an antigovernment group in Minnesota—the “Patriot's Council”—planned to use a toxic extract of castor beans called ricin to kill local and federal law enforcement personnel. In 1995, a member of a white supremacy group was arrested and sentenced to 18 months’ probation for stockpiling bacteria that cause the plague. He could produce no legitimate reason for possessing such quantities of a deadly pathogen. So bioterrorism has already arrived on American shores, and the enemy, so far, is us.

In the United States, the Centers for Disease Control (CDC), in Atlanta, Georgia, is the primary federal agency responsible for coordinating all scientific, medical, and public health aspects of the federal response to potential and actual bioterrorism. In 1999, even before the September 11 attacks on the World Trade Center and the almost immediately subsequent anthrax scares, the CDC commissioned a detailed study of agents that could be used in bioterrorist attacks. Those that proved to be of the greatest concern, based on factors such as lethality, ease of dissemination, and ability (p.223) to induce panic and social disruption, were given a “Category A” designation (Table 14.1).

Federal, state, and local governments have a wide range of programs poised to be activated at the first hint of a bioterrorism attack, including rapid identification of the biological agents involved, tracking and containing their spread, and identifying and treating affected individuals. That is all well and good, and would doubtless greatly reduce the potential damage from a bioterrorism attack.

But in the early stages of any such attack, your primary—your only—defense will be your own immune system, honed as we have seen over millions of years of evolutionary selection to respond rapidly and effectively to invasion of your body by potential microbial predators and their toxins. The microbes around which a bioterrorism attack could be mounted will likely be selected in part on the basis of a known poor immune response by humans to the agent involved as well as maximum debilitation—or panic—caused by the attack.

In the sections that follow we take a look at the microbes and toxins currently deemed by the CDC as most likely to be used as bioterrorism agents.


Anthrax is a disease caused by the bacterium Bacillus anthracis. It affects animals, mostly grazing herbivores such as sheep, goats, and cows. Humans are vulnerable to anthrax infection, but we have learned over the centuries how to avoid it, and veterinarians are skilled at keeping domestic livestock free of the disease. Fewer than 250 cases of naturally acquired anthrax in humans have been reported in the past 50 years in the United States. In less developed parts of the world, annual new cases of anthrax are considerably more.

Anthrax is arguably the most serious threat on the CDC's list of Category A bioterrorism agents. It is deadly: the mortality rate for (p.224)

table 14.1 Biological Agents Classified by the Centers for Disease Control as Category A Potential Bioterrorism Agents







Bacillus anthracis





Variola major





Yersinia Pestis


Antibiotics, antiserum



Francisella tularensis





Clostridium botulinum




Viral hemorrhagic fevers

Ebola, Marburga viruses




Among the criteria for classification as a Category A agent: high level of virulence or toxicity in humans, feasibility of large-scale production and dissemination as an aerosol, readily spread from person to person, lack of effective treatment and of public health preparedness, and potential for public panic and social and/or economic disruption. Category B agents include (but are not limited to) ricin toxin, Staphylococcus enterotoxin B, and encephalomyelitis virus. Category C agents include such things as hantavirus and multidrug-resistant tuberculosis.

(a) Other hemorrhagic fever viruses include Lassa fever virus, various arenaviruses, Rift Valley fever virus, yellow fever virus, Onsk hemorrhagic virus, and Kayasanur virus.

(b) This vaccine is not effective against pneumonic plague, however. Vaccines for this form of the plague are nearing readiness for clinical trials.

(p.225) untreated anthrax can range from 20% to 100%, depending on the form of infection (see below). The World Health Organization (WHO) has estimated that 50 kg (about 110 pounds) of anthrax spores released in the air over a population of 5 million could result in serious disease in 250,000 people and lead to as many as 100,000 deaths, putting such an incident on par with a nuclear bomb attack! There is already substantial public awareness of just how deadly anthrax is, and news of an anthrax attack in a dense urban area would doubtless create major panic and civil disruption, a major aim of any terrorist attack.

Part of what makes anthrax so deadly is that B. anthracis forms spores. Most bacteria, when they run out of food, simply starve to death. A few bacteria, however, are able to enter a state of suspended animation—to convert to bacterial spores. Spores do not carry out any metabolism, do not need water, and are extremely resistant to heat and many toxic chemicals. Properly prepared, they are hard, dry particles easily carried on wind, which makes them perfect for use as bioterrorism agents. When they land on a surface possessing moisture and nutrients—human skin or lungs, for example—they rapidly revert from spores to normal bacterial cells in a process called germination. Spores can survive in their dehydrated state for several decades.

Anthrax spores can be inhaled or can settle on exposed areas of skin. Both would likely occur in most exposed individuals, and both pathways of entry can result in disease. In inhalational anthrax, many of the spores are engulfed by lung macrophages. Initial symptoms are fever, achiness, and often a sore throat. Many of the ingested spores are able to germinate inside the macrophages, eventually destroying them and escaping into surrounding tissues. Other spores will settle directly on soft, wet lung tissue, germinate, and begin to divide. In either case, actively dividing bacteria quickly migrate through lymph and blood to other parts of the body. It doesn't take long before rabidly dividing, healthy bacteria have spread everywhere. Mortality in untreated inhalational anthrax can approach 100%.

(p.226) Spores settling on the skin (cutaneous anthrax) can enter the body through cuts or abrasions. Once inside, they follow a similar path. Some germinate locally and cause redness and itching that can develop into local skin ulcers (Figure 14.1); many will spread to other parts of the body, germinating as they go. However, death from this form of anthrax, untreated, rarely exceeds 25%.

Anthrax infections can be treated with the antibiotic Cipro. Effectiveness depends on the form of infection (it is most effective against cutaneous anthrax) and how long the infection has been in progress. There are no known cases of transmission of anthrax from one human being to another, an important factor in the management of an anthrax attack.

B. anthracis produces two deadly toxins that are responsible for the illness and death accompanying anthrax infections, regardless of the mode of entry. Edema toxin causes water to escape from host cells in the vicinity of anthrax bacteria, causing massive swelling, which interferes with normal tissue functions. Lethal toxin cripples the innate, and thus the adaptive as well, immune responses, allowing

                      First Defense: The Immune System and Bioterrorism

figure 14.1 Skin ulcer resulting from a cutaneous anthrax infection. (Courtesy NIAID Biodefense Image Library.)

(p.227) unimpeded replication of anthrax bacteria in the body. The resulting inflammation and accumulation of fluids and bacterial byproducts lead to rapid deterioration of host metabolic functions, profound shock, and death. After consuming whatever is left of the host and running out of food, the bacteria generate more spores and drift away on the wind looking for a new hotel. Cremation is recommended for any remains.

There have been few studies of the immune response to anthrax in humans, because natural infections are now so rare. Most of what we know about the immune response comes from studying infections in animals and human responses to anthrax vaccines. The most important response is the production of antibodies against the two anthrax toxins. These antibodies block the ability of the toxins to bind to cells and also “tag” both spores and bacteria for removal by macrophages. CD4 T cells necessary to help B cells make these antibodies are important, but since B. anthracis does not live inside cells (aside from their brief transit through macrophages), CD8 T cells probably play little role in immune defense.

Unfortunately, useful levels of antibodies rarely develop in natural infections, because the toxins so quickly knock out the key cells involved in starting an immune response. The loss of dendritic cells in particular, so crucial in triggering inflammation and activating T-helper cells, is perhaps the most serious damage caused by anthrax toxins.

Several vaccines against anthrax toxins work reasonably well to induce antitoxin antibodies. Unfortunately, all of the current vaccines require several injections over at least several weeks, and so would be of little use for the early victims of an anthrax attack. They also have uncomfortable side effects for some individuals.

However, people in the immediate vicinity of an attack would likely be given these vaccines anyway, since anthrax spores will be everywhere and can persist for decades. Researchers are working on improved vaccines, some based on the DNA technology discussed in chapter 7, that could generate protection much more quickly and with fewer side effects. Such a vaccine could be of more (p.228) use during an actual attack, but for economic and practical reasons would probably not be used to immunize the general population in the absence of an attack.

The CDC has recommended that we might revert to one of the oldest forms of immunization—passive immunization (chapter 2)—for anthrax. This involves the injection into one individual of antibodies made in another individual. Five of the 11 victims of the postal anthrax attack in 2001 died despite intense antibiotic treatment. The U.S. Department of Heath and Human Services (HHS) is currently contracting with several private companies for production of such antibodies. Passively transferred antibodies might work even more rapidly and effectively than antibiotics. The antibodies would be directed at the anthrax toxins and block them from attacking host cells. But passive immunization is intended only for the immediate treatment of victims, not for preventive immunizations of entire populations.


Today it is hard to imagine that smallpox was once one of the deadliest diseases on this planet, probably exceeding even the plague in the cumulative number of people killed throughout history. When contracted through the lungs by breathing in air into which an infected person had sneezed or coughed, it routinely killed 20% to 30% of unvaccinated individuals, well into the twentieth century, and left the rest badly disfigured for life. Smallpox can also be spread by person-to-person physical contact, although the resultant disease is usually less fatal.

Smallpox is caused by an orthopoxvirus called Variola major. Almost uniquely among human pathogens, V. major has no known animal or insect reservoir (a host in which it can reproduce without causing disease, or at least death). In its present form, it appears to be entirely dependent on human beings for its propagation and survival.

(p.229) Smallpox is the first disease-causing microbe to be purged from the human species, by a worldwide immunization campaign launched by the WHO in 1967. By 1972, routine vaccinations were discontinued in the United States because of a small risk of active disease from the vaccine itself. The fact that V. major could not retreat into an animal reservoir during this campaign was a major factor in its eradication. Today V. major officially exists only as frozen stockpiles at the CDC in Atlanta, and in a former biological warfare research center near Novosibirsk, in Russia.

Smallpox is on the CDC Category A list because of its high mortality rate and because, as a viral disease, it is essentially untreatable. It spreads very efficiently as an aerosol, and the virus is relatively stable. Also, like anthrax, there is enough residual public awareness of the deadliness of smallpox that news of its spread in a terrorist attack would likely generate considerable panic and social disruption. Since for the past 30 years almost no one in this country has been vaccinated against smallpox, the U.S. population is highly vulnerable to this disease.

Smallpox was used as a weapon by the British in the French and Indian Wars (1754–1767). Blankets that had been used to wrap infected British soldiers were distributed to Indian tribes cooperating with the French. Although this means of spreading smallpox is less deadly for the initial victims, they in turn spread it as an aerosol through coughing and sneezing. The overall fatality rate among the Indians was well over 50%.

In the course of a V. major infection, viruses settle into airway tissues and are swept along into regional lymph nodes where they provoke an immediate response by the innate immune defense system. This results in some combination of mild fever, chills, and achiness. When the virus reaches the skin (from the inside out, as it were), a rash appears, followed by the formation of multiple, closely packed blisters on all parts of the body, but particularly the face and neck. These blisters also form in the mouth and throat, where they break easily, dumping their viral load into the saliva. This aids in the further spread of the virus into the general (p.230) population through coughing and sneezing. The cause of death, in those cases that are fatal, is unclear but may be due to the enormous buildup of antigen–antibody complexes, which trigger rampant inflammation and tissue damage in kidneys and lungs.

As we saw in chapter 5, T cells are a major part of the immune defense against many viruses. Because smallpox was largely eradicated in humans by the early 1970s, when T cells were just beginning to be studied, we know almost nothing about T-cell immunity against smallpox. It would seem likely that CD8 killer T cells play a role in controlling human smallpox infections. But techniques for studying CD8 T cells in humans were not worked out until the mid-1970s. Moreover, our views of innate immune mechanisms, and their interaction with the adaptive immune system in the activation of T cells, have changed radically in the last 10 years and have never been examined in smallpox infections. Natural killer (NK) cells were not even discovered until 1975. The lack of an animal model for studying smallpox has long been a major drawback. Recently, however, V. major has been used to produce infections in macaque monkeys, and information about how this virus interacts with their immune systems may be useful in designing smallpox vaccines and antiviral drugs for smallpox infections in humans.

The few insights we do have into the possible course of the human cellular immune responses to V. major come from studies in the late 1970s on volunteers receiving smallpox vaccinations. Immunizations for smallpox over the years have never been carried out with V. major—it is too deadly—but rather with a closely related orthopoxvirus called vaccinia. The exact origins of this virus are unclear (chapter 7). Vaccinia is injected in a fully viable form. It induces a mild local reaction at the site of injection that usually resolves in 7 to 10 days. Protection from subsequent infection by V. major after vaccinia immunization is excellent, but about 1.6 cases per 1 million immunizations progressed from mild reaction to more serious disease, and occasional deaths, which is why vaccination for smallpox was abandoned in 1972.

(p.231) Nevertheless, a number of healthy individuals agreed to serve as volunteers in a study published in the late 1970s. In this study, both mice and humans produced killer cells against vaccinia-infected target cells. The killer cells from mice were clearly CD8 killer cells, but the killers produced by the human volunteers were not. The best guess at the time was that NK cells, and possibly neutrophils, were involved in killing vaccinia-infected cells in humans. Later studies suggested that CD8 killers are in fact produced at low levels in response to vaccinia. At any rate, we cannot be sure that the immune response in humans to infection by V. major, the actual pathogen in smallpox, would be exactly the same as that produced following immunization by vaccinia. For designing future vaccines, we would really like to know this. But it seems unlikely we ever will.

We do know that the amount of virus-neutralizing antibody produced in response to immunization with vaccinia directly correlates with subsequent protection to V. major. During a natural infection with V. major, antibody production peaks after about three weeks and remains high for several years. Most studies also suggest a role for vaccinia-induced antibody in tagging V. major for phagocytosis and destruction by macrophages and neutrophils. The Department of Health and Human Services is currently funding development of so-called third-generation vaccinia-based vaccines, specifically in response to concerns about bioterrorist attacks using smallpox.


References to the plague in human history date back to at least 500 B.C., although we don't really know if that plague was the same as the three pandemics that swept Europe and Asia in the Middle Ages, which is what we recognize as plague today. There were several major and many minor pandemics in Europe in the fourteenth through eighteenth centuries. They were deadly. Although we have no precise figures, as many as 200 million people are estimated to have died.

(p.232) The non–spore-forming bacterium Yersinia pestis has been associated with the European outbreaks, and Y. pestis DNA has actually been extracted from dental remains found in graves of persons dying from the plague. We still see occasional incidents of Y. pestis plague, with several thousand new cases arising annually throughout the world. There have been about 400 cases in the United States since 1950, mostly in the southwest.

Y. pestis can cause several types of disease in humans, depending on how the infection is acquired. Bubonic plague results when a human is bitten by an insect, usually a flea, carrying Y. pestis, which it acquired from previously biting an infected animal. In urban areas, the most common animal carriers are rats and squirrels and the occasional house cat. Y. Pestis can also jump from animal fleas into fleas more at home in humans, which greatly aids human-to-human spread of the disease. In the middle ages, most people had fleas in abundance. Prior to the introduction of antibiotics in the 1940s and 1950s, fatality rates of 50% or more for bubonic plague were not uncommon.

After transmission through a bite, Yersinia bacteria begin to replicate and are swept along to nearby lymph nodes in lymph fluid. At various stages in this journey they may be engulfed by macrophages, but are relatively resistant to being digested by them. A few days later the typical symptoms of a microbial infection set in: fever, chills, and general achiness, byproducts of activation of the innate immune system.

As the bacteria continue to replicate inside the lymph nodes, the nodes become greatly enlarged (“buboes”) and very tender. They can grow as much as four inches across. If the bite occurs in the lower part of the body, lymph nodes in the groin are preferentially affected. Bites in the upper body regions tend to deliver bacteria to lymph nodes in the armpit and neck. Untreated mortality rates are around 50% of those infected.

If the bacteria spill out of the lymph nodes and enter the general blood circulation, numerous other tissue compartments become involved and the infection is even more lethal (“septicemic (p.233) plague”). Blood vessels are destroyed, resulting in gangrene in the extremities. This is probably the origin of the term “Black Death” for plague. Prolonged infection can also trigger shock, a common cause of plague death. If the bacteria invade the lungs (secondary pneumonic plague), the infection is almost always fatal and the bacteria spread more readily from person to person through sneezing and coughing.

Primary pneumonic plague, the second major form of plague, is particularly deadly. It occurs when Y. pestis is taken in directly through the respiratory system as opposed to an insect bite. Untreated mortality rates approach 100%. Symptoms set in within a day or two after inhalation of infectious Y. pestis and are initially indistinguishable from other forms of aggressive pneumonia. Aerosolized Y. pestis and the pneumonic plague that results would likely be the choice of terrorists. Bubonic plague carried by fleas is difficult to spread over a large area and does not pass easily from person to person. Experience in diagnosing and treating pneumonic plague is very limited in the United States. Moreover, many currently used antibiotics have never really been tested against Y. pestis in humans.

The only modern-day use of plague for biological warfare was by the Japanese during occupation of China in World War II, when they released plague-infected fleas onto civilian populations. Both the United States and the Soviet Union pursued development of aerosolized Y. pestis, but these appear never to have been used. Aerosols of course would induce pneumonic plague and could be unbelievably deadly. The WHO estimates that 50 kg of aerosolized Y. pestis spread over an urban population of 5 million would infect at least 150,000 people, causing at least 36,000 deaths.

There have been no documented attempts to use Y. pestis as a bioterrorism agent. However, in 1995, a microbiologist was arrested for fraudulently obtaining large amounts of plague bacteria, with no obvious legitimate scientific purpose. And in 2004, a respected physician-scientist at Texas Tech University was sentenced to two years in prison for grossly mishandling and illegally (p.234) shipping to Tanzania vials containing infectious Y. pestis—on a commercial airliner, no less! No connection with bioterrorism was alleged or proved.

We know few details of the human immune response to Y. pestis, especially primary pneumonic plague infections, because of the scarcity of cases. Most of our recent insights into human immune responses have never been examined in plague patients. The standard vaccine for many years, based on a whole-cell, killed form of Y. pestis, is no longer used. A second vaccine based on a live but attenuated form of the bacterium induces good protection against bubonic plague but unfortunately little or no protection against pneumonic plague. Producing an effective vaccine for pneumonic plague is an active area of research, driven largely by concern about use of plague for terrorism.

From animal studies we know that in bubonic plague (and we assume in pneumonic plague as well), plague bacteria quickly make contact with host macrophages, dendritic cells, and neutrophils. But immediately upon contact, the bacteria inject substances into the cells that greatly reduce their phagocytic function. Even for those bacilli that are successfully engulfed, additional substances released inside the cell inhibit degradation of the bacilli and allow them to replicate. Another chemical released while the bacilli are still outside the cell kills off nearby NK cells, a vital component of the early response to viruses. So right off the bat, crucial elements of the innate immune system that otherwise would ordinarily slow the infection, and help kick off an adaptive response, are seriously disabled.

The B-cell response to Y. pestis requires CD4 T-cell help, and CD4 T cells also produce numerous cytokines that help fight the infection. Little is known about the role of CD8 cells in the response. Many Y. pestis bacteria remain extracellular during an infection, but many also manage to replicate within macrophages and perhaps dendritic cells. Whether these intracellular bacteria elicit a protective CD8 response is not known. Many people die without ever mounting an effective antibody or T-cell response

(p.235) As noted above, currently available vaccines induce a good antibody response that can control bubonic plague but offer little protection against pneumonic plague, which we expect would be the form we would encounter in a terrorist attack. Moreover, these vaccines require multiple injections to be effective, do not give long-lasting immunity without booster shots, and can have numerous undesirable side effects. There is currently an intensive laboratory campaign to develop vaccines effective against pneumonic plague. Hopefully one can be found that acts quickly enough to be of some use during an attack while having acceptable side effects. Several are based on recombinant DNA. Tests on animals look promising, and one of these has recently received Food and Drug Administration (FDA) fast-track approval for human clinical trials.


Botulism is caused by a protein toxin released by several strains of bacteria of the genus Clostridium, including the eponymous Clostridium botulinum. This toxin is the most lethal biological poison known—100,000 times more poisonous than sarin gas. One gram (about 1/28 of an ounce) in aerosol form could theoretically kill 1 million people.

Unlike other A-list bacterial agents, such as anthrax, plague, and tularemia, where the agent is the bacterium itself, in the case of C. botulinum, we cut right to the chase—the isolated, purified toxin produced by the bacterium, which is entirely responsible for the disease caused by this bacterium, is the agent. Like B. anthracis, C. botulinum is a spore-forming bacterium, and could readily be aerosolized as such. But since the object of delivering the bacterium is to deliver its toxin, it is more efficient to simply aerosolize the toxin itself, which is relatively easy to collect and concentrate.

The toxin produced by C. botulinum and related strains is a neurotoxin that prevents the brain from telling muscles to contract. Individuals poisoned by botulinum toxin through food experience (p.236) extreme muscular weakness and have difficulty seeing, speaking, and swallowing. Their brain and associated mental functions are not impaired, but their muscles just cannot do work, including the muscles that control breathing. Death comes mostly from respiratory failure.

Intact C. botulinum bacteria were fed by the Japanese to prisoners during their occupation of Manchuria in World War II. The results, as far as they are known, were uniformly lethal. The Aum Shinrikyo cult in Japan attempted to carry out attacks in Tokyo in the 1990s using botulinum toxin, but for technical reasons were unsuccessful. Research on botulinum toxin as a possible biological weapon was carried out by the United States and other countries over the years but never used. United Nations inspectors determined during the 1990s that Iraq had prepared about 5,000 gallons of concentrated toxin, some of which was found by inspectors to have been loaded onto missiles, ready for use. How effective these would have been as bioweapons is, however, unknown.

Natural infection by C. botulinum can occur by eating contaminated food, usually vegetables, though the name botulinum in fact derives from the Latin for sausage (botulus), a common food contaminated by C. botulinum in former times. The poisonous effect of food contaminated with this bacterium is due entirely to the toxin, which it readily secretes into its surroundings. C. botulinum can also enter through wounds, for example, in the foot, when walking in soil harboring this bacterium. It can also infect needle puncture wounds and can be a problem among intravenous drug users. The toxin itself will not penetrate unbroken skin.

Aerosolized toxin is rare in nature, but has been prepared in numerous laboratories. Whatever the mode of entry into the body, the toxin quickly gains access to the blood and lymph, from where it reaches the points at which nerves make contact with muscles (neuromuscular junctions). With the purified toxin there is no response by the innate immune system and none of the signs associated with a microbial infection (fever, chills, achiness, etc.). Common early signs of poisoning, which begin 12 to 72 hours after (p.237) ingestion (depending on dose), include difficulty in speaking and swallowing, dry mouth, blurred vision, and extreme muscular weakness. Treatment almost always requires extended stays in intensive care units, which in the event of a large-scale attack would rapidly overwhelm the capacities of even the best hospital systems.

A complicating factor with botulin toxin is that it has a number of useful medical applications when delivered in extremely low doses and in a highly targeted fashion. A number of disorders are characterized by excessive or involuntary muscle contractions, and botulin toxin (“Botox”) can be used to alleviate these spasms. The list of such disorders is quite long and includes many genuinely serious conditions. Thus, plans for general immunization of large populations could interfere with the use of Botox for therapeutic purposes. Botox has also been used in recent years for cosmetic purposes, although this is obviously not a major public health consideration.

The immune response to botulin toxin in humans is not well understood, again because of the scarcity of cases of natural infection available for study. Anyone exposed to botulin toxin must immediately undergo intensive treatment and is not an appropriate subject for being poked in his or her lymph nodes. So most of what we know about the immune response to the toxin comes from studying the response in animals to C. botulinum and its toxin, or the human response to botulin toxoid, a form of the toxin that has been crippled in its ability to cause disease without altering its ability to induce an immune response. Botulin toxoid indeed does induce a good immune response in humans, but to what extent this mirrors the response to native botulin toxin is not known.

Botulin toxin as a bioterrorism agent would involve purified protein with no contamination by viable microbes. So, as noted earlier, there would be no signs of an active microbial infection to signal its presence—no fever, no chills, no achiness. Moreover, the dose received by any individual would likely be extremely minute. Given its incredibly poisonous toxicity, such doses would be sufficient to cause severe disability or even death, but could be insufficient to (p.238) trigger a meaningful immune response. About half of the botulism patients who have had their blood analyzed do show antitoxin antibodies, but the response is not particularly strong and the majority of these patients who survive do not go on to develop an immunological memory of the toxin. This latter is attributed to the low levels of botulin toxin in their system.

The only current botulin toxin vaccine for use in humans in the United States was first produced in 1970. It consists of a mixture of various forms of the toxin that have been chemically treated to produce toxoids. Reasonable levels of antibody capable of neutralizing botulin toxin have been induced in volunteers after two to three injections. It is intended only for use in vaccinating military personnel in the event of a threat of biological warfare and people working in laboratories where botulin toxin is studied. A number of DNA vaccines, in which genes for various portions of the toxin known to trigger antibody production are introduced into individual subjects (Chapter 7), have shown great promise in animal studies. Recently, one such vaccine received clearance from the FDA for fast-track human clinical trials.

At present the only therapy for botulism is injection of toxin antibodies (antitoxin) produced in horses. This is not without side effects, mostly due to the fact that the horse antibodies are recognized as foreign by the human immune system, so multiple administrations of the antitoxin would not be possible. Researchers have focused on the production of toxin antibodies that are less likely to trigger an immune response in humans, but human trials for FDA approval lie some years in the future.


Tularemia is caused by the bacterium Francisella tularensis, named for its discoverer, Edward Francis, and the place of its discovery, in 1911, in Tulare County, California. It had been associated with a variety of plague-like diseases in animals such as deer-fly fever, (p.239) rabbit fever, and tick fever, among others, all of which are now grouped as various forms of tularemia. Humans can be infected by contact with F. tularensis in the wild, although such incidents are relatively rare, and humans do not readily transmit the resulting infection to others. Contact can result from handling of infected animals or through insect bites, but these infections are usually mild.

The most serious incidents of tularemia in humans come from inhalation of bacteria, often through handling of contaminated hay or other grains. “Inhalation tularemia” requires only a few bacteria, whereas infection through other routes usually requires exposure to millions of bacteria. If F. tularensis were to be used as a bioterrorism agent, it would almost certainly be in aerosol form. Because of the low incidence of inhalation tularemia in the United States, a large outbreak of this disease in a concentrated area would lead to an immediate suspicion of bioterrorism.

Tularemia was investigated by several countries between 1930 and 1970 as a potential biological warfare agent. The bacterium can be concentrated into a paste, which can be freeze-dried and then milled into a fine powder suitable for distribution through the air. A WHO study estimated that 50 kg of bacteria (about 110 pounds) in aerosolized form, spread over a population of 5 million people, would incapacitate about 250,000 people and cause nearly 20,000 deaths. The United States retained stocks of F. tularensis through the late 1960s, but these were destroyed in the general obliteration of such stockpiles in the early 1970s. Current military research with this microbe is restricted to defensive strategies.

Inclusion of F. tularensis by the CDC as a Category A bioterrorism agent is due largely to its effectiveness when spread in aerosolized form. The initial signs of inhalational tularemia infection are no different from the signs accompanying most microbial infections. Most of what we know about the immune response to tularemia has been gleaned from studies of this disease in rodents. The course of infection and the resulting immune response in rats and mice appear to mimic closely the events occurring in human (p.240) infection. F. tularensis, like Mycobacterium tuberculosum (chapter 6), invades and takes up residence in macrophages but manages to escape digestion and multiplies rapidly within the macrophage itself.

In the case of inhalation tularemia, the primary target is macrophages resident in the lungs, but the bacteria make their way to regional lymph nodes as well. The lung tissues become generally inflamed and can develop various forms of pharyngitis, bronchitis, and other forms of lung infection. One form or another of pneumonia is the most common cause of death in fatal cases. Tularemia is one of those diseases that have a “low index of suspicion” among doctors and laboratory personnel, which could also be a factor in its selection by bioterrorists.

Both innate and adaptive responses are mobilized in response to F. tularensis, but development of a vigorous adaptive response is absolutely essential to clearing an infection. Production of cytokines like IFN-γ and TNF-α are critical during the early innate immune response and are probably provided by dendritic cells, macrophages, and perhaps NK cells. Interestingly, neutrophils are able to scavenge and kill F. tularensis; apparently the tricks used by this bug in escaping lysosomal destruction in macrophages don't work in neutrophils.

As would be expected for an intracellular parasite, B cells and antibody play little role in the ensuing adaptive response. Effective, long-term immunity is provided almost entirely by T cells, both through enhanced production of IFN-γ and probably through direct T-cell–mediated killing as well, although there is little direct evidence for the latter in the current scientific literature.

There is a vaccine for tularemia, based on a live but relatively harmless strain of F. tularensis. But this vaccine is only marginally effective against inhalation tularemia, and it takes at least a week or two to build up a good level of protection after vaccination. In a bioterrorism attack using an aerosolized form of the bacterium, it is unlikely that this vaccine would be useful in protecting exposed individuals after the attack. Development of a more active vaccine, (p.241) perhaps based on DNA (Chapter 7), should be a goal of those concerned with homeland security.


Hemorrhagic fever viruses (HFVs) are by far the most deadly of human pathogenic microbes. The CDC has designated nine HFVs as potential bioterrorism agents (Table 14.1). We will focus here on only two of these, the Ebola and Marburg viruses (Figure 14.2). Both of these viruses are fairly recent additions to the repertoire of human pathogens, and not that much is known about their interaction with their human host. There have only been a dozen or so outbreaks of these viruses since their discovery in 1967 (Marburg) and 1976 (Ebola).

We do not know what animals serve as a reservoir for these viruses—hosts that harbor them without contracting serious disease. Several nonhuman primates, such as rhesus monkeys and macaques, are fully susceptible to the ravages of Marburg and Ebola infection and would be unlikely reservoirs. Cases of human infection tend to occur in clusters, the origins of which are not always clear; in several instances human infection seems likely to have originated from contact with infected monkeys. Once one

                      First Defense: The Immune System and Bioterrorism

figure 14.2 Ebola (left) and Marburg (right) virions. (Courtesy NIAID Biodefense Image Library.)

(p.242) person has been infected, however, transmission to others occurs through contact with fluids or tissues from previously infected individuals.

Ebola and Marburg viruses would certainly fulfill the CDC requirement that an agent have the potential to cause “public panic and social disruption.” Through books and films in the past two decades, plus regular media coverage of outbreaks, these viruses may have, along with anthrax, the highest public profile of the Category A agents. Because of the high mortality rate, the fear factor may be even greater than for anthrax.

As of mid-2005, 1,848 cases of hemorrhagic fever in humans caused by Ebola had been reported to the WHO, with 1,287 fatalities (69.9%). Three hundred fifty-four cases of Marburg fever had been reported, with 288 deaths (81.3%). Almost all of these cases arose in Africa. Many have been traced to transmission through unclean clinical syringes, a common problem in rural Africa; the resulting mortality in these cases was 100%. We might hope that mortality would be somewhat lower in industrialized countries, but make no mistake: these viruses are far and away the most lethal pathogens on the CDC's A list.

Both the United States and the Soviet Union produced aerosol versions of HFVs, including Ebola and Marburg viruses, for biological warfare. These were never used, and we have no idea how effective aerosolized HFVs would be. Aerosolized HFVs are relatively stable and cause disease and death in nonhuman primates, and it is presumed they would do the same in humans. Aum Shinrikyo traveled to Zaire to obtain samples of Ebola but was unable to procure enough stock to create a weapon.

Because there have been so few cases, usually occurring in remote areas, exactly what happens in the course of a naturally acquired Ebola or Marburg infection in humans is not entirely clear. The popular depiction of humans being literally melted away from the inside out contains a good deal of dramatic license, but these are undeniably ghastly diseases. Blood vessels as well as blood cells are a frequent target of the viruses and are rapidly (p.243) destroyed once the virus begins to spread in the body, causing massive internal bleeding. But organs such as the liver and kidneys are also severely damaged. Symptoms of hemorrhagic fever include the usual early signs of any microbial infection, but these are quickly followed by widespread body rashes and blood spots in the skin, blood seepage from various orifices, convulsions, delirium, and a rapid descent into shock and coma. There are no antiviral drugs effective against Marburg and Ebola. For the few people who survive, there is a long period of impairment of numerous body functions.

We know very little about the human immune response to Ebola and Marburg viruses. A 1996 study in Gabon showed that those infected with Ebola who subsequently died of circulatory collapse failed to develop a strong antibody response and had no CD8 killer cell response at all. Among close family members who did not die, about half had produced Ebola antibodies, indicating they had been exposed to the virus, and most also had activated CD8 T cells and a strong inflammatory response.

Why some people were able to mount a protective immune response to Ebola, while others weren't, is not presently known. In mice, we know that both antibodies and CD8 killer cells are induced by exposure to Ebola. Passive transfer of the antibodies to naïve mice did not provide protection against subsequent exposure to Ebola, but transfer of immune CD8 T cells did. Passive transfer of HFV antibodies in nonhuman primates has not generally provided much protection, and there is little hope that this would be an effective treatment in humans.

Attempts to produce an effective vaccine against Ebola and Marburg had been generally unsuccessful until 2005, when a research group centered in Canada developed a single, DNA-based vaccine that is very potent against both Ebola and Marburg. Just one injection protected monkeys from infection by either virus. The Ebola/Marburg gene was engineered to be delivered preferentially to macrophages and dendritic cells, to optimize rapid antigen presentation of viral peptides to T cells. It is possible that with further (p.244) work this vaccine could also be made effective for other A-list HFVs. More work needs to be done before human trials can begin, but this vaccine looks extremely promising.


So what do we know about the ability of this wall we hide behind—our immune system—when it comes to bioterrorism agents? Can it help us? Well, the first thing to remember is that, with the agents on the CDC's A list (not to mention lists B and C), if our immune systems could stop these agents dead in their tracks, they wouldn't be on the CDC lists. The real question is, is there anything we can do to help our immune systems do a better job?

One key to helping the immune system is to have the most thorough knowledge possible about how our immune system interacts with these pathogens once they have invaded our bodies. As we have seen, the diseases caused by A-list pathogens are so rare in the United States that we have had little opportunity to study how the immune system responds to them. And until recently we have expended little effort in developing effective clinical measures to guard against them. Most first-line health care responders have no experience with either these pathogens or their diseases, which can cost precious time in identifying the problem in a real attack.

This is quite different from the pathogen that causes AIDS—the HIV virus. We probably know more about every aspect of HIV and its interaction with human beings and their immune systems than any other pathogen on earth. If we are really concerned about bioterrorism with the agents described in this chapter, we need to know much more than we do at present about how they work, and most of all how they are handled by our immune systems. Research programs to answer these questions are currently under way.

The question is often asked, why wouldn't bioterrorists use HIV as a weapon? Why isn't it on the A list? Unquestionably, the release of HIV over a large metropolitan area could generate a (p.245) maximum fear effect. And as we know all too well, all but a tiny handful of us are defenseless against HIV, with no vaccine on the immediate horizon. So the fear factor probably extends to would-be terrorists themselves. They may be extremely reluctant even to get into the same room with HIV. Another factor is that the incubation period with HIV, before frank AIDS sets in, is 6 to 10 years. Suspiciously large numbers of new cases would likely not be apparent for several years at a minimum. The immediate public relations sensation so craved by terrorists would be lost.

Still, the overall psychological impact on affected populations could be enormous. In the end, the main thing preventing use of HIV is that it is an exceptionally fragile virus. Exposure to anything other than a warm, wet human body disables it within a matter of hours. Aerosolization would almost certainly cripple it. Laboratories working with HIV must take enormous care to keep their strains viable. It is, in fact, a poor candidate for even the CDC's C list.

There are three general strategies for helping our immune systems deal with the kind of pathogens we do find on the A list. The first is to produce enough of what we might call “traditional” vaccines that are effective enough to be of help warding off a bioterrorist attack. Preferably, we would like to have vaccines that could be of help after someone has already been exposed to a particular pathogen. But traditional vaccines are designed to work prophylactically—before contact with the pathogen. They are designed to generate protective adaptive immunity in order to boost responses to subsequent exposures to the pathogen. Normally, it doesn't matter that several weeks may be required to build up that memory.

Since we don't know which populations of people might be the target of an attack, in order for this approach to be effective we would have to immunize the entire nation—against six pathogens! We do something like that now, with our children, for the most common (and potentially crippling or lethal) childhood infections. And it works. But we do not have vaccines at present for any of (p.246) the A-list pathogens that are suitable for mass prophylactic immunization programs, for either children or adults.

And it is not obvious we would want to undertake such a program even if we had such vaccines. There use would likely be limited to selective immunization of “first responders”—health care personnel, police, fire, certain military units. In the case of anthrax, the causative spores of which could linger in the environment for a long time after a terrorist attack, such vaccines could be useful to immunize individuals present but not infected in an initial attack. A great deal of research is currently directed at making faster-acting vaccines for all the A-list pathogens, and almost certainly we will get some vaccines that induce adaptive immune responses more quickly. But the chances are slim they will be able to act fast enough to be of much use in treating already infected individuals once a bioterrorist attack has been unleashed.

A second approach, still in the largely theoretical stage, takes advantage of the knowledge we have gained over the past decade or so about the workings of the innate immune system. For microbial pathogens (less so for their protein toxins), we now know that the innate immune system plays a direct, cognitive role in the early stages of all infections. Dendritic cells, macrophages, neutrophils, and even B cells have receptors for the pathogen-associated molecular patterns (PAMPs) present on all microbes (chapter 5). The innate immune system, remember, is our first line of defense in any microbial infection, keeping the infection at bay long enough for the adaptive T- and B-cell response to get up and running.

So a great deal of effort is also being expended to find ways to stimulate and strengthen the innate immune response that all of us will be mounting within minutes of any pathogen invasion and lasting for as long as the pathogen remains a threat. The innate response is crucial for triggering inflammation and for processing and presenting forms of microbial antigens that will bring T and B cells into play. Instead of focusing on the antigen-specific, adaptive aspects of vaccination, more attention is being placed (p.247) on “vaccines” that provide as strong a stimulus as possible to pumping up the critical innate elements of the immune system at the beginning of the response. And because they are not directed at any particular agent, we would need only one such vaccine.

An incoming pathogen will of course trigger these responses on its own, but if substances can be quickly introduced into the body to accelerate that portion of the immune response, it is reasoned, we may be able to stave off the deadliest aspects of infection long enough for the more potent adaptive immune response to get off the ground. The studies of survivors of Ebola and Marburg outbreaks referred to in the last chapter provide strong motivation for this general approach.

The third approach to helping the immune system is to build better antibodies to use for passive immunization. Ready-made antibodies provide a powerful weapon against any bacterial and many viral infections and are also useful for neutralizing microbial toxins. If injected during the first 24 hours or so after someone is infected with a pathogen or toxin, infection could in many instances be enormously reduced, buying precious time for the innate system to complete its job and for the adaptive system to begin functioning.

The problem, as we have seen previously, is that most of these antibodies are made in animals, usually horses, and the antibodies themselves trigger an immune response in the person into whom they are injected. A single injection of horse antibodies into a person doesn't cause a problem, because by the time that person makes antibodies against the incoming horse antibodies, the horse antibodies are gone. Those not taking part in neutralizing microbes are cleared from the blood, like any other protein. But a subsequent administration of horse antitoxin antibodies into that same person would quickly encounter large quantities of that person's antihorse antibodies and be neutralized. And it is entirely possible in a terrorist attack with large amounts of a deadly pathogen that one injection of antibody might not be enough.

(p.248) Humans recovering from natural infections or planned immunizations with crippled pathogens are also a source for antibodies that could be used for passive immunization. Although a slight immune response would be triggered in the person receiving these antibodies, the response would be relatively mild compared to the reaction against horse antibodies and could be managed. But the number of persons from whom such antibodies could be harvested is vanishingly small compared to the huge numbers of people who might need treatment in the immediate aftermath of a bioterrorist attack with a particular pathogen.

Great effort is now being directed at producing “humanized” antimicrobial antibodies for use in passive immunization. This approach depends on the technique of monoclonal antibodies described in chapter 2, with a little genetic razzle-dazzle thrown in. Antibodies, say, to B. anthracis would first be produced in mice. Mouse B cells producing this antibody would then be isolated and converted to monoclonal B cells, which can be expanded enormously and used to produce theoretically unlimited amounts of monoclonal antibody specific for B. anthracis.

But these are still mouse antibodies. They will trigger the same kind of vigorous immune response in humans that horse antibodies do. This is where the razzle-dazzle comes in. It is possible to genetically engineer mouse B cells so that they produce monoclonal antibodies with most of their mouse portions replaced with a human counterpart (Figure 14.1). These humanized mouse antibodies will provoke a greatly reduced immune response in human recipients, one that will not wipe out a subsequent administration of humanized antibody. Initial trials of this concept in animal models have been highly encouraging.

So there is hope! Our immune systems clearly will need some help in building an effective immune response to pathogens used in a terrorist attack. Once our immune responses have a chance to get off the ground and make it to the adaptive response stage, they will be more than able to defend us against not only the first attack with a given pathogen, but any subsequent exposures as well. (p.249)

                      First Defense: The Immune System and Bioterrorism

figure 14.3 Creation of “humanized” mouse monoclonal antibodies.

It may be that none of the three approaches just described, by themselves, can provide the help we need, but by combining them in the way we combine different approaches to treating cancer, there is a very good chance that we can gain the most important thing we need to respond effectively to a bioterrorist attack—time! (p.250)