From the Department of Psychiatry, The Univeristy of Texas Southwestern Medical Center at Dallas.

Presented to internal medicine housestaff at Baylor University Medical Center on August 11, 1999, as second part of pathogens and parasites lecture. The first part was published in BUMC Proceedings 1999;12:175-187.

Corresponding author: Gregory G. Dimijian, MD, 3277 Brincrest Drive, Dallas, Texas 75234. (dimijian@connect.net).

New and unique ecologies are providing new opportunities for microbes. Rivers and oceans play a prominent role in these new ecologies. Nutrient runoff into coastal waters is causing many more algal blooms (“red tides”) than ever before, some of which are frightening in their toxicity. One dinoflagellate, Pfiesteria piscicida (the “cell from hell”), has at least 24 life-forms and elaborates a neurotoxin that has caused neurocognitive deficits in fishermen and laboratory workers (3).

Pathogens from human feces are now being found alive and well in deep ocean water, where they survive for long periods of time at the low temperatures. Poliovirus and rotavirus have been identified in ocean water samples below 1000 meters. Enteric bacteria have been found in good condition at sewage sludge sites >170 km offshore from New York City 30 months after the sites had been closed to dumping. These bacteria were resistant to several antibiotics, a sure sign that they had originated from humans or their food animals. A 17-km pipeline is being built to take sewage from Boston out to sea. Although the sewage will be treated, some bacteria and viruses are resistant to chlorine (4).

Introduced species of plants and animals, transported by humans from one part of the world to another, serve as a dramatic metaphor for the similar spread of microbes. Exotic plants, insects, and animals are transforming environments around the world and include Africanized bees, fire ants, Asian tiger mosquitoes, zebra mussels, the European house sparrow, and uncounted marine species discharged with ballast water from ships in every port. A global “McEcosystem” is being created (5).

While microbes are being disseminated into new habitats, scientists cloning and sequencing RNA from microbes living in their natural environments have discovered a biomass and genetic diversity beyond the dreams of earlier researchers whose tools were limited to microscopy and cell culture.

Earth's deep biosphere, from the surface down to several miles, may contain a biomass of microorganisms rivaling that of all organisms on the earth's surface (6). Microbes—bacteria and archaea and probably viruses—appear to be ubiquitous throughout the upper part of the earth's crust, living in fluid-filled pores and interstices of rocks and deriving energy and nutrients from hydrogen, methane, sulfur, and petroleum. Growing in utter darkness and reproducing at a snail's pace, the microbes comprise a new biome, receiving energy from the internal heat of the earth and from chemicals. The depth limit is ultimately determined by the increasing temperature. We may share chemically fueled subsurface life with other bodies of the solar system and the cosmos.

In surface waters of the oceans, bacteria typically number a million per milliliter of seawater. They are outnumbered by viruses, which attain 10 million per milliliter (7). Viruses are consistently the most abundant biological entities in the sea, from the tropics to the poles and from the sea floor to the surface, and even in sea ice. How did they get there? By infecting cells—mostly bacteria—and arranging for their own replication. As most viruses undergo rapid deterioration, viral infections of marine organisms, especially bacteria, must be continuous and ubiquitous. The same dominance of viruses may also occur in surface and subsurface terrestrial environments but has not yet been demonstrated.

The following sections discuss topics related to viruses, genes and natural selection, strategies of parasites, co-evolution of hosts and parasites, and new challenges.

VIRUSES

Important viral agents in humans

Viruses of greatest concern to us include the following:

• HIV-1. The prevalence of AIDS is skyrocketing in sub-Saharan Africa and southeast Asia, with little being done to prevent or treat it. Life expectancy is falling at an ever-increasing rate.
• Rotavirus. Rotavirus diarrhea kills 800,000 children a year worldwide and is the most important cause of severe childhood diarrhea before the age of 2 years.
• Influenza A. New strains of influenza A viruses appear every year. Influenza A is one of the deadliest infectious diseases of developed countries, causing 20,000 to 40,000 deaths each year in the USA. The combined deaths of the 2 world wars (with atomic bombs dropped on 2 cities) did not attain the mortality of the 1918 influenza epidemic.
• Hemorrhagic fever viruses, including
• Hantaviruses (e.g., 4 Corners syndrome)
• Filoviruses (Marburg, Ebola)
• Dengue fever (the most important viral infection transmitted by mosquitoes)

Virus classification

One way of classifying viruses is to group them broadly into animal viruses, plant viruses, or bacterial viruses. A more fundamental classification separates them into 2 categories: DNA viruses and RNA viruses (Table 1).

A DNA virus: bacteriophage. Bacteriophages, viruses that infect bacteria, can convert a bacterial cell into a factory for manufacturing new copies of themselves. In the process they sometimes emerge with a small package of genes acquired from the bacterium. When they infect another bacterial cell, those genes may be transferred to the genome of the new host, which acquires a new trait. That trait may be toxin production or antibiotic resistance, either of which may be useful to the bacterial host and be favored by natural selection. Bacteria sometimes invite the phages in, offering pili for attachment along with free copying services.

Bacteria can transfer toxin genes intraspecifically (within the same species), as in the case of Vibrio cholerae, as well as interspecifically (between different species). E. coli 0157:H7 owes its virulence to a toxin gene almost identical to that of Shigella dysenteriae type I and probably acquired the gene through phage transfer.

Bacteria have invented several ways of transferring genes “laterally” among themselves. Conjugation between bacteria enables passage of genes between them, and there is free movement of genes from a disrupted bacterium to a live one. A newly discovered secretion system actively transfers plasmid DNA from one cell to another (9). Vibrio cholerae possesses a versatile genetic element called an integron, which actively integrates newly acquired foreign genes into its chromosome. Over 40 different antibiotic resistance genes have been found in integrons (10). There are thus passive and active mechanisms of gene movement among bacteria, both intra- and interspecifically—an Internet of sorts for rapid information exchange.

Who needs sex if you're a bacterium or a virus? Microbes accomplish efficient diversification of progeny in at least 3 ways: through gene exchange, through high mutation rates with copy errors, and through reshuffling of gene segments.

RNA viruses. RNA viruses are the only known genomes with genes made of RNA. Cellular genomes first translate their coding DNA into RNA, which serves as the molecule transferring the genetic information to the ribosomes, where amino acids are assembled into protein molecules. For RNA viruses, this step is not needed, as their RNA genomes provide direct instructions for assembly of protein molecules.

RNA viruses also have genes coding for RNA polymerases, which are used to replicate their RNA genomes. These polymerases lack the proofreading abilities of DNA polymerases and thus have a high error rate. Mutations are introduced into the viral genome during replication. A swarm of variants called a quasispecies is produced in some RNA viruses. Some RNA viruses have the highest mutation rates known for any genomes.

HIV and influenza A, both RNA viruses, also have high recombination rates, involving reassortment between 2 strains infecting the same host. The influenza A virus reassorts its genome when avian and mammalian strains infect pigs, and HIV strains reassort their genes in human hosts.

One type of RNA virus, the retrovirus, is unusual in that it transcribes its RNA genes into DNA and integrates them seamlessly into the host cell genome. Integration is a required intermediate step in retrovirus replication. Most other viruses hijack the cell's replicative machinery without inserting their genes into the host chromosome. The newly inserted genes—now called endogenous retroviruses or proviruses—are inherited in the lineage of the cell.

If only somatic cells are infected, the provirus is not passed on to the host's progeny. If germ cells are infected, proviruses become part of the genetic dowry of the host species, inherited as classical Mendelian genes and evolving in concert with the host genome. As far as we know, neither HIV-1 nor HIV-2 has ever become integrated into human germ cells. Such an event might just be a dead-end experiment for a lethal virus.

Retroviruses are effective gene-delivery vectors. If some of the viral RNA is replaced with RNA coding for a useful protein, integration of the provirus may provide missing genetic instructions for making the protein in a patient with a genetic disease.

Eukaryote evolution has taken place against a background of constant retroviral infection. Proviruses are found in all of the vertebrate genomes in which they have been sought. We and other vertebrates have hundreds of proviruses in our genomes, and where they came from we may never know. Most appear to be harmless baggage and are not expressed, but some appear to encode infectious virus. Some may remain quiescent until activated sporadically by stress or some other trigger, and there is reason to suspect a role in autoimmune disease, as some retroviral proteins share amino acid sequences with normal cellular proteins and may trigger an antibody response to host proteins through molecular mimicry.

Present concern over xenotransplantation stems from the ubiquity of proviruses in vertebrates. Moving tissue or organs across the species barrier into intimate association with a new host may release proviruses held in check by the natural host. How can we provide donor organs free of proviruses? Unfortunately, proviruses are part of an animal's genome, and even captive-bred “clean” animals harbor them. To make matters worse, we intentionally suppress the immune system in transplant recipients in order to facilitate graft acceptance. There is a public health risk as well from possible person-to-person spread. This could become a runaway risk with a global black market fueled by the enormous unmet need for organs; more than 4000 Americans die each year waiting for an organ transplant.

Are viruses “alive”?

Writing in The New York Times about the disinterment of frozen human bodies above the Arctic Circle in the search for fragments of the 1918 influenza A virus, John Noble Wilford emphasized that precautions were being taken in case the virus “is still alive.” We all know what Wilford meant, but the facile use of the term “alive” betrays our ambivalence about how to regard viruses. They are, after all, far from being rocks. They program their own replication using genes, and those genes also code for proteins that specifically manipulate their cellular hosts. However, viruses differ from cells by having no cytoplasm, organelles, cell wall, or metabolism.

The choice of how to regard viruses is not easy. A size comparison with bacteria (Figure 1) shows that some viruses are larger than the smallest bacteria. In addition, some large viruses, like poxviruses, are so complex that they come close to carrying out their own replication—inside cells, of course. And some bacteria, like mycoplasmas, are so tiny and almost devoid of cell walls that they are obligate intracellular parasites, like viruses.

John Maynard Smith, emeritus professor of biology at the University of Sussex, offered an interesting argument for saying that viruses are alive. We have an alternative, he said, to the phenotypic definition of life (based on the phenotype, the translation of the genes into morphology, biochemistry, and behavior). Maynard Smith's alternative is to define as alive any entities that have the properties of multiplication, variation, and heredity. The logic behind this definition is that entities with these properties will evolve by natural selection and can be expected to acquire the complex adaptations for survival and reproduction characteristic of living things (11).

In a sense, viruses are satellites of living organisms, orbiting at a distance and plunging in when they can, integrating their genetic language with that of a host cell, sometimes becoming a part of the genome of the cell. They are not whole, but they consist of the very guts of life and are a celebration of the ability of genes to code for their own progeny. As Stephen Jay Gould has aptly written, nature abhors boundaries.

GENES AND NATURAL SELECTION

The fluidity of genomes

Genomes appear to open their arms to foreign genes, whether those genes come from an invading virus or a syringe in a laboratory. The molecular and cell biologist Robert Pollack has written:

Prokaryotic and eukaryotic cells require little convincing to incorporate foreign genes: they take in all DNA with alacrity. One way to get DNA into a cell's nucleus is to inject it through a fine glass needle. The nucleus swells as if a mosquito had bitten it and within hours the cell is expressing the injected gene. . . . The monkey kidney cell draws in the membrane where the SV40 virus is attached, bringing virus and membrane coat into the cytoplasm. The cell begins to dissolve it with proteinases, which is just what the virus wants. As its coat is dissolved, its genome is released into the cell. Now seen by the cell as a set of genes that has lost its way, the cell carts the viral genome through the cytoplasm into the nucleus. There the tiny viral genome is picked up and incorporated into the nuclear DNA, and replicated. In a few days the nucleus is packed with millions of . . . virus particles (12).

Why are cellular genomes so receptive to foreign genes? Whatever the reason, it seems that genes go both ways with ease—into the genome of a cell, inserted by a virus, or out, sometimes to become a virus or a part of a viral genome. Many large viruses have pirated host genes and made them a permanent part of their own genome, thus acquiring the ability to direct the synthesis of their own versions of cellular proteins. Human cytomegalovirus, for example, produces its own version of cellular proteins that interfere with transcription of genes of the major histocompatibility complex (MHC) (13).

Virologists have been called vertebrate biologists who study fugitive fragments of their subject. Christian de Duve calls viruses “gypsy genes” that have wandered far from home (14).

Viruses take on many roles in the cell, which seems to be their playground. At times viruses appear to be just hitchhikers, stowaways, freeloaders. If they are harmless baggage, however, sometimes the baggage carries a bomb, waiting until a particular time when a bacterial host cell is damaged or not reproducing well, at which time the virus becomes activated and destroys the cell in a burst of its own replication. Viruses remind us that natural selection acts directly on genes, which are the only real replicators in the long run.

Mobile genetic elements (“jumping genes”)

Mobile genetic elements, or transposons, are stretches of DNA in eukaryotic genomes that are capable of excising themselves out of the host chromosome and moving to another location on the chromosome or even to another chromosome. They are flanked by DNA sequences that encode instructions for their own movement. Some jump about so regularly that they have been dubbed “mariner transposons.” When Barbara McClintock first described jumping genes in the 1940s, she didn't know what a Pandora's box she had so brilliantly opened (Figure 2).

Some mobile genes are more prolific than others. Called retrotransposons, they carry a gene for reverse transcriptase, so that when their DNA is transcribed into RNA, they can reverse-transcribe the RNA into DNA progeny that broadcast themselves around the genome and splice themselves in at any point. The process can lead to a massive increase in copy number over a short period of time. Incredible as it may seem, up to 35% of the human genome and over 50% of the maize genome are believed to consist of retrotransposon DNA (15).

Retrotransposons often land in a “harmless” place in the genome (a noncoding site). If <5% of the human genome consists of coding DNA, this is not surprising. But sometimes they land inside a functional gene, disrupting the gene and producing disease. Muscular dystrophy is believed to be one such hereditary disease.

The yeast genome contains regions believed to be “landing pads” for retrotransposons. Perhaps this is a case of “if you can't beat `em, at least show `em to a safe seat.”

Some retrotransposons may be distant progeny of proviruses. The term retrotransposon, however, should not be confused with retrovirus, as these are not (yet?) viruses but highly adapted elements of the genome.

Transposable elements appear to have been components of eukaryotic genomes since the Cambrian (16). On the one hand, they seem to be “selfish genes,” DNA parasitic upon DNA, coding only for their own replication. On the other hand, they may provide a fertile ground for genome rearrangements and gene duplications, opening doors to evolutionary change.

The genome has long been thought of as an archival blueprint of life, a relatively permanent record. Mobile genetic elements are replacing that view with one of an ephemeral environment, undergoing continuous remodeling. Genomes look more like a melting pot of immigrants, welfare recipients, restless teenagers, outright crooks, potential pathogens, and—here and there—a few honest, hard-working genes.

Ernst Mayr, one of the architects of modern evolutionary theory, once called the genome a republic of parts. Today it seems anything but a republic. The biologist William Hamilton has written (paraphrased):

There came to me the realization that the genome isn't the monolithic data bank devoted to one project (keeping oneself alive, having babies) that I had imagined it to be. Instead it was beginning to seem more like a company boardroom, a theatre for a power struggle of egotists and factions. . . . I was an ambassador ordered abroad by some fragile coalition, a bearer of conflicting orders from the uneasy masters of a divided empire. . . (17).

On the other hand, the small minority of protein-coding genes seems to be a more stable blueprint. They provide a framework for studying molecular phylogenetic history and identifying complex gene families that have evolved over hundreds of millions of years.

Natural selection and the immune system

When an unfamiliar antigen enters the body and remains in the extracellular space, B lymphocytes initiate rapid replication and reshuffle their antibody-coding genes at a high rate. Called somatic hypermutation, this occurs within individual B cells. An army of B cell precursors moves to the outer zones of lymph nodes, where the B cells “compete,” in a sense, to make an antibody that best binds to the new antigen. After several rounds of hypermutation, one of the B cells by chance hits upon an antibody that binds to the antigen and is stimulated to produce a rapidly spreading clone of itself. Its progeny will secrete a soluble form of the antibody into the plasma.

The genes coding for this antibody will be inherited by the progeny of that B cell, often for the life of the host. But the code is inherited only in the B-cell line, not in the host's germ cells. The antigen has acted as the criterion for selection within a somatic cell line. It's a vertebrate's way of doing what some viruses do—reshuffling gene segments to create new genotypes virtually overnight. But our B cells do it inside their own genome, instead of trading off gene segments with other B cells, as viruses do between strains.

The result of somatic hypermutation, together with point mutations, is remarkable: theoretically up to 18 billion different antibody molecules can be created from only 300 genes. Like microbes, the vertebrate immune system is capable of generating rapid and extensive genetic diversity (Table 2).

Antibodies are the weapons of the so-called humoral immune system, which first appeared with the evolution of vertebrates. We also have the “innate” immune system and the MHC. The innate system is a first line of defense and can destroy some microbes outright. It is crucial as an early defense while the humoral system and MHC are building their specific responses to each challenge.

The humoral and innate systems detect only extracellular pathogens; once a virus or bacterium moves into a cell it is beyond their reach. The term adaptive immune system refers to both the humoral and cell-mediated arms, as both respond selectively to new antigens.

MHC molecules are specialized to recognize foreign peptides inside the cell, bind to them, and carry them to the cell surface. There they are recognized by T cells, which either destroy the invader or initiate apoptosis, which destroys the infected cell and prevents further replication of an invading virus. A number of viruses can down-regulate the transcription of MHC genes, thus interfering with the cell-mediated immune response (13). This provides another clue to the usefulness of MHC polymorphism.

Gene polymorphism in the malaria parasite seems to represent a countervailing strategy in the arms race. When inside red blood cells, the parasite exports some of its own polypeptides to the surface of the cell, and here it is open to immune attack. The polypeptides are encoded by an estimated 200 genes, the largest gene family in Plasmodium falciparum. This polymorphism enables extreme antigenic variation in the polypeptides inserted in the red cell membrane, thus keeping the host immune system scrambling to make new antibodies (18).

This abbreviated approach to the immune system is misleading unless the bridges between the 3 “arms” are appreciated. For example, the humoral response is primed by activities of the cell-mediated arm. This occurs in at least one way: when an antigen-presenting cell takes in foreign peptide fragments by endocytosis, the fragments are transported by MHC molecules to the cell surface and “presented” to T cells. Antigen binding by T cells triggers release of cytokines by helper T cells. The cytokines activate B cells, which make antibodies. The humoral and cell-mediated arms are thus interdependent.

There are tantalizing new hints that some autoimmune diseases may result from reduced exposure to parasites. Inflammatory bowel disease is most prevalent in countries with a high standard of living, where people are at low risk for helminthic infections; in turn, human populations with chronic helminthic infections tend to have a low incidence of inflammatory bowel disease. Experimental exposure to helminthic parasites in mice was shown to blunt the inflammatory responses that result in intestinal mucosal injury, and a preliminary study of human patients with refractory inflammatory bowel disease showed clinical improvement after oral administration of helminth eggs, with down-modulation of the inflammatory response (19). The vertebrate adaptive immune system has evolved over hundreds of millions of years in a parasite-rich milieu. Do autoimmune diseases reflect immune function gone wrong in a novel, hygienic environment?

Molecular biology is providing an embarras de richesses in the form of insights into immune system evolution and function. We may owe the remarkable repertoire of our humoral immune system to a transposon insertion in the germ line of B lymphocytes early in vertebrate history. The recombination of gene segments in developing B lymphocytes (somatic hypermutation) is enabled by the same splicing and joining that takes place in transposon splicing and integration and is mediated by the same enzymes encoded by a transposon that inserted itself in early vertebrate chromosomes. According to this hypothesis, antibody formation becomes a dramatic expression of genetic restructuring by transposons. The change was presumably immortalized by natural selection acting on a beneficial adaptation (20).

“STRATEGIES” OF PARASITES

Virulence

Suppose that 3 strains of a pathogen are circulating in a host population. Strain 1 is the most virulent strain, with a high reproductive rate. It kills the host quickly, curtailing its own spread. Strain 2 has intermediate virulence and moderate reproductive rate. The host becomes ill but remains intermittently mobile and continues to spread the strain. Strain 3 is the least virulent strain, with a low reproductive rate. The host is mildly ill and fully mobile, but there is less shedding of the agent into host secretions.

When would strain 1 be likely to outcompete the other strains? When transmission is easy, as with open sewers, with trench warfare, or in refugee camps. Easy transmission means host death is relatively unimportant to the parasite. The most virulent strain wins, as it reproduces most rapidly and moves more genes into the future than other strains, and those progeny have no trouble finding hosts. By similar reasoning, strain 3 wins if transmission is difficult, requiring live, mobile hosts who can serve as vectors as long as possible.

The relation between virulence and ease of transmission, explored in depth by biologist Paul Ewald, provides valuable predictive power, all else being equal (21, 22). Although only in laboratory tests can all else be kept equal, the correlation provides important information. Serial passage experiments, in which an inoculum of a pathogen is passed from laboratory host to laboratory host, show that competition among strains drives parasite adaptation and the evolution of virulence (23). These experiments, however, do not reflect the natural history of host-parasite relations. Transmission relies on the experimenter, and hosts are usually of low genetic diversity or even clonal.

How can a parasite “get away” with high virulence, without seriously compromising transmission and losing out to other strains? One way is to employ a vector, such as a mosquito, tick, or flea, to provide transmission to the next host. Then the parasite can afford to replicate rapidly and immobilize the host—provided, of course, that the vector can get to an immobilized host. Another way is to delay the onset of illness, enabling spread during a pre-illness phase while the host is mobile. Another way is to infect some hosts as carriers only, as in the case of typhoid and hepatitis A. Still another way is to utilize cultural vectors, such as sewage, food, or hospitals, all maintained free of charge by human hosts. An uncommon but effective way is survival for long periods in the environment. Many fungal pathogens survive well outside of a host, but most bacterial and viral pathogens don't. Finally, the parasite could establish a wildlife reservoir, so that a source of an inoculum would exist indefinitely. Smallpox could not have been eradicated if the virus could hide in a wildlife reservoir. In any of these ways a parasite may retain its virulence and a high reproductive rate and minimize the cost of truncated spread caused by early host immobility or death.

Contrary to Lewis Thomas' argument that there is nothing to be gained, in an evolutionary sense, by the capacity to cause illness and death, there is now ample reason to believe that pathogenicity is as reasonable a strategy as any other, given certain ecological conditions or imperatives. One such imperative is competition with other strains, and one such ecological condition is a new crossing of the species barrier into a novel host species whose immune system is na?ve to the new pathogen. Such a pathogen may not have had time to evolve more effective adaptations to the new host.

One bacterial pathogen has found a way to repair the damage it inflicts, in the interests of its own replication in the living host. Salmonella typhimurium first disrupts an intestinal epithelial cell and achieves into it and then elaborates a protein that helps the cell restore its disrupted cytoskeleton. Both the disruption and restoration of cellular architecture are under bacterial control (24).

Pathogenicity and high virulence are here to stay. Yet many highly successful parasites make a perfectly good living without being virulent most of the time, such as common cold viruses, herpes simplex viruses, and Chlamydia bacteria.

Host death begins to look like an incidental byproduct of infection, not an outcome that is useful to the parasite. It is a price that both host and parasite pay. One fundamental cause may be competition between parasite strains, in which the more virulent strains move more genes into the future; as Ewald emphasizes, more virulent strains are favored when transmission is made easier. Alternatively the cause may be a newly emerged pathogen that has recently crossed the species barrier and happens to be highly virulent to the new host. Host death may occasionally benefit a pathogen, such as anthrax bacteria, which spread by spores formed and released upon death of the host organism. Anthrax spores can remain viable in the soil for a decade or longer.

Unreasonable specificity of pathogens: zoonoses and the species and tissue barriers

The term “zoonosis” refers to a disease in humans caused by a pathogen of other animals that has crossed the species barrier to humans. The organism may or may not cause illness in the other animal. Examples include rabies, toxoplasmosis, cutaneous larva migrans, brucellosis, trichinellosis, and hantaviral infection.

Herpesviruses have a clear “preference” for certain host species and not others. For each herpesvirus there exists a host for which the virus is almost always fatal and reservoir hosts in which the virus produces little or no clinical illness. HSV-1, for example, is usually latent in man and fatal in Aotus monkeys, gibbons, and marmosets.

The tiny flagellated protozoan Trypanosoma brucei brucei causes a disease in cattle resembling African sleeping sickness in humans but does not harm immunocompetent humans. It causes an initial infection in humans that is quickly controlled by host defenses. In 1995 two apolipoproteins from human serum were found that destroy trypanosomes by inducing oxidative damage.

Infectious diseases believed to be unique to humans include cholera, typhoid fever, smallpox, rubella, pertussis, syphilis, and gonorrhea. As far as we know, the pathogens of these diseases may infect other animals only if those animals are immunocompromised.

The opposite of zoonoses, humorously dubbed “humanoses,” include tuberculosis spread to domestic and zoo animals by humans, and salmonellosis transmitted to penguins by researchers in the Antarctic.

The tissue barrier is as remarkable as the species barrier. Pathogens attack one kind of cell and utterly ignore another in the same host. The rabies and polio viruses infect anterior horn cells and cerebral neurons, respectively, and disregard most other cell types. Yet there are also pantropic pathogens, like tubercle bacilli, which infect lungs, bone, genitourinary tract, meninges, peritoneum, and skin.

Both species and tissue barriers tumble down in an immunocompromised host. Pathogens that normally infect other species move in. The line between fungal pathogens of plants, for example, and fungal pathogens of humans becomes blurred in immunocompromised human hosts, who become living Petri dishes. The fungi consume host tissues like they would eat a plant leaf.

The Centers for Disease Control and Prevention recommends that immunocompromised pet owners keep their pets indoors, take them for regular checkups, keep vaccinations current, declaw cats, and avoid very young pets, which are more likely to shed enteric organisms in their stool. Immunocompromised individuals should avoid jobs that require intensive exposure to animals, such as veterinary medicine or zookeeper employment.

Species barriers are rarely absolute. Given a fighting chance, barriers may be breached. As we invade the last tropical rain forests on Earth, we are laying out a red carpet for novel microbes to explore new host territory. There is, in fact, evidence that species barriers are being crossed at an accelerating rate around the world. In closely monitored groups of marine mammals, new diseases and disease epidemics are occurring with increasing frequency. Rather than reflecting the appearance of new pathogens, these diseases appear to reflect an expansion of the host range of previously known pathogens. A new disease, aspergillosis of sea fan corals, has been attributed to transport of terrestrial fungi in runoff waters. Morbillivirus infections, including distemper, have caused mass mortality in seals and porpoises around the world and have been acquired in some cases from domestic dogs. Influenza viruses from aquatic or migratory birds have caused mortality among seals and whales. Human activities have modified marine and terrestrial ecosystems, opening avenues for microbial spread to new hosts (25). Terms such as “primary host,” “reservoir host,” and “zoonosis” may become more tentative as we look more closely at the changing specificities of parasites.

“Extended phenotype”

A remarkable study has shown that mosquitoes carrying malaria parasites bite more frequently and more aggressively than parasite-free mosquitoes (26). The researchers speculate that the parasite interrupts afferent signals from the mosquito's abdominal stretch receptors, blocking a sensation of a full blood meal. For once, parasites do not treat their vectors with kindness! This finding is a novel example of Richard Dawkins' “extended phenotype,” the expression of genes in the body or behavior of another organism (27, 28). The concept of an extended phenotype has great teaching merit, as it provides a new way of understanding how a parasite manipulates the phenotype (body or behavior) of another organism in its own interests. The mosquito would normally limit its biting frequency to match its own needs for blood, but it is driven to more aggressive biting by the parasite, which benefits from the greater number of contacts.

Rats parasitized with Toxoplasma gondii also seem to be manipulated by their parasite into behavior of unilateral benefit to the parasite and clearly dangerous to the rats. Parasitized rats in a recent study were found to lose their natural fear of new objects and odors, and some, in a suicidal turn, were actually attracted to feline scents. Only in its definitive host, the cat, can T. gondii undergo sexual reproduction and complete its life cycle. The authors of the study suggest that the rats' changed behavior results from the presence of the parasite in the rat's brain, where it promotes behavior that makes the rats more susceptible to predation by cats (29).

The biting behavior of rabid mammals represents another example of manipulative behavior on the part of a pathogen, in this case a virus that infects not only the brain but also the salivary glands of its mammalian host. The host's brain is modified to induce aggressive behavior, and the salivary glands supply copies of the virus to be spread by the biting behavior. In like manner upper respiratory viruses and Mycobacterium tuberculosis achieve spread by causing their hosts to disseminate them in respiratory droplets. Some birds lay their eggs in the nest of the same or another species and leave them there to be tended by an unwitting foster parent who also feeds the chick after it hatches; called brood parasitism, this deceptive strategy is yet another example of the extended phenotype, in which the true parent tricks the foster parent into performing the chores of parenting. It is as if the victimized bird were an extension of the true parent's phenotype and motor neurons of the perpetrator extended to the muscles of the victim.

When you see a behavior you don't understand, ask yourself whose genes it is benefiting (Figure 3).

Parasite life history and ecology

The ability of bacteria to adapt to chemical and environmental challenges is legendary. At the Homestake gold mine in South Dakota, bacteria were found that not only survived the cyanide effluent but utilized the carbon and nitrogen of the cyanide as nourishment (30). The bacterium Deinococcus radiodurans survives exposure to several million rads of ionizing radiation, which breaks down the toughest of glass containers. It repairs its DNA by lining up the broken parts with their homologues on other chromosomes (31). Alkaliphilic bacteria thrive at a pH of 10 in soda lakes of the African Great Rift Valley, and acidophilic bacteria thrive near a pH of 0 in Yellowstone's sulfur springs. Thermophilic bacteria survive and reproduce between 90C (194?F) and 113C (235?F) in hydrothermal vent ecosystems, and cold-loving bacteria grow at 0C (32?F) in the winter ice cover of high mountain lakes (32, 33). Three vignettes follow that further illustrate adaptation to the environment.

Herpesviruses. Herpesviruses are past masters at producing latent infections in humans, for the life of the host. This is not a simple task. Consider the steps taken by HSV-1 and HSV-2, which produce cold sores and genital herpes, respectively (with a little overlap):

• The virus must enter the terminal branches of sensory nerves on the lips or genitalia.
• It must climb up to the neuronal cell bodies (Greek herpein, to creep) and hide inside them, escaping immune surveillance (neurons are “immune-privileged” cells with reduced MHC activity).
• The target neurons are in the trigeminal ganglion (for HSV-1) and the sacral ganglion (for HSV-2); the virus resides in those neurons for the life of the host.
• The virus must insert its DNA into the neuron's chromosomes and, because these neurons are nonreplicating, the virus must find a way to get itself replicated periodically, permitting sporadic bursts of infectivity. During these episodes of replication, it moves out to the lips or genitalia where it produces lesions, which enable shedding and spread of virus.

The unique ecology of these herpesviruses enables them to survive and spread without being virulent. Since the host is both virus factory and vector, the virus has achieved the ideal state of harmlessly infecting a healthy, active host which can broadcast it far and wide.

Some herpesviruses have even managed to erase their own tracks. A high proportion of persons reactivate HSV-1 and –2 subclinically, i.e., without discernible lesions or symptoms. The virus bursts out of the ganglia and spreads out on mucosal surfaces for 2 or 3 days, without producing any fever blisters or genital ulcers, then disappears again. While on mucosal surfaces it is easily spread by contact to anyone else (34).

Lyme disease (Lyme Borreliosis). Lyme disease is the most common vector-borne disease in the USA. White-footed mice constitute the wildlife reservoir of the spirochaete Borrelia burgdorferi, and the tick, Ixodes ricinus, is the vector. Larval and nymphal ticks are infected by the mice, and adult ticks in turn infect humans and white-tailed deer, both of which are incidental hosts. The deer are important in Lyme disease epidemiology, however, because ticks of reproductive age live on deer, not mice.

The mice feed on acorns and the pupae of gypsy moths. The moths are an introduced species that periodically defoliates the oak forests, causing a crash of acorn crops and a consequent population crash of mice. With few mice, the incidence of Lyme disease plummets.

Oaks produce large acorn crops every 2 to 5 years in episodes of “mast fruiting.” Acorns are rich in proteins and lipids and set off an “ecological chain reaction.” Mice and deer gorge on acorns and their populations grow; tick populations increase along with their host numbers, and moth numbers decline as they fall prey to mice.

Still other variables come into play, such as rainfall and competing parasites. A chaotic network of interactions generates multiple feedback loops, each loop having a different time course. Prediction of Lyme disease risk becomes a major challenge for models and computers (35, 36). The lesson is that ecology must be taken into account. Virulence, an important variable, is but one of many.

Helicobacter pylori—a highly specialized symbiont. Colonization of the human stomach by the bacterium Helicobacter pylori is believed to be a causative factor in up to 80% of gastric and 95% of duodenal ulcers and is strongly implicated in the pathogenesis of chronic gastritis and gastric adenocarcinoma. At least half (perhaps two thirds) of the world's human population is infected, making it the most widespread chronic human bacterial infection known. Low socioeconomic status and crowding are associated with higher prevalence, and evidence points to intrafamilial spread, possibly by the oro-oral or feco-oral route. The bacterium can survive on the body and in the gut of houseflies, suggesting a possible mode of spread.

H. pylori is well adapted to the hostile environment of the stomach at extreme acid concentrations, despite vigorous humoral and cellular immune responses mounted against it. It has a relatively small genome—1.7 megabase pairs compared with 4.6 for E. coli and 5.8 for Pseudomonas aeruginosa, bacteria that can live in a wide range of habitats. Features of its small genome reflect its highly specialized adaptation. It has very few regulatory genes for switching on and off genes needed for moving between different environments, supporting epidemiologic evidence that it lives mostly in the stomach. The enzymatic pathways needed for survival in this harsh milieu are continuously switched on. The most abundant enzyme produced by H. pylori is a urease that breaks down urea and releases ammonia, making its immediate environment less acidic. It eventually migrates below the mucous layer where the cellular environment is less acidic.

Are gene sequences too reductionistic to reveal the ecology of an organism? With H. pylori, we have a dramatic refutation of this claim. Approximately one third of the 1590 genes identified in H. pylori have no equivalents in databases of other bacteria. They provide a clue to proteins that can be targeted for therapeutic drugs or vaccines specific to this microbe, sparing the normal gastrointestinal flora (37).

It has been suggested that H. pylori may actually have beneficial effects on infected carriers who are heavily exposed to other gastrointestinal pathogens. A recent study has found that H. pylori possesses antibacterial activity to which it is itself resistant (38). If H. pylori is an innocuous gut symbiont for most people and may actually benefit some, it may be an example of a longstanding mutualism, a symbiotic relationship in which both parties benefit more than they lose—that is, the association helps both to survive and reproduce. Natural selection would favor a “cooperative” interaction on the part of both organisms, if this is the case. The colonist would be selected to maintain low virulence, and the host would not mount a damaging immune response.

COEVOLUTION OF HOSTS AND PARASITES

Like hosts, parasites have evolved from earlier ancestors. Time lines of free-living organisms can be thought of as accompanied by a co-evolving cloud of colonists—pathogens, parasites, and symbiotic microorganisms (Figure 4). Family trees of hosts and contemporary symbionts often reveal a congruity in branching patterns, which suggests parallel evolution of both partners over long time intervals (39). The clouds overlap, and there is a dynamic movement of some colonists across species boundaries (Figure 5).

The co-evolutionary arms race may at times be more like “trench warfare,” with recurring advances and retreats in resistance gene frequencies in a host population (40). Recurring epidemics drive up resistance gene frequencies, which fall during intervening epidemic-free periods. This dynamic gene polymorphism is long-lived and maintained by natural selection.

High pathogenicity and mutualism span an unbroken continuum along which organisms may move dynamically over evolutionary time. The surprisingly common world of symbioses and mutualisms is reflected in family trees of co-evolving partners and will be covered in a future review.

NEW CHALLENGES

Prions

If prions are infectious proteins, as we think they are, they are an anomaly in modern biology, as they are transmissible disease agents without a genome. They cannot replicate themselves or (like viruses) manipulate a cell into replicating them. How does natural selection maintain them? The more important question should probably be how selection acts on the genes coding for them. For the genes to be maintained (especially in the face of the lethality of the protein in the disease-causing conformation), there may be an unrelated advantage conferred on the host that spreads them, one that we can now only guess at. Genes predisposing to hereditary diseases such as sickle cell disease and diabetes mellitus seem to be maintained in this manner; they confer advantages on the host that outweigh the disadvantages, in an evolutionary sense. Some studies suggest such an advantage conferred by prion protein genes, such as a possible role in long-term survival of Purkinje neurons (41, 42).

“Noninfectious diseases”

Are all diseases infectious? asks the title of an article in The Annals of Internal Medicine (43). The tongue-in-cheek question seems less far-fetched the more we discover about disease etiology. There is no longer doubt that some viruses are oncogenic, triggering the onset of neoplastic change in cells. Some viruses infect a cell and code for oncogenic proteins that remove restraints on cell division; others insert their genome into the host genome adjacent to a proto-oncogene, switching it to an oncogene, which begins to code for oncogenic protein. Helicobacter pylori, isolated for the first time in 1982 from a human gastric biopsy, has radically changed our view of the etiology of gastric and duodenal ulcers and gastric adenocarcinoma. Seroepidemiologic and anatomical studies have implicated Chlamydia pneumoniae in coronary artery disease, myocardial infarction, carotid artery disease, and cerebrovascular disease, although a strict causative role has yet to be established. Campylobacter jejuni infection has been implicated in up to 75% of cases of Guillain-Barr? syndrome. Molecular mimicry, involving the induction of self-directed immunity by microbial antigens, may help explain the role of streptococci in rheumatic heart disease, chlamydia in atherosclerosis, and C. jejuni in Guillain-Barr? syndrome.

Bacterial biofilms

Bacterial biofilms (44) may be a common cause of persistent infections. Bacteria can adhere to solid surfaces and form a slippery coat; these microbial communities display an inherent resistance to antimicrobial agents and constitute a protected mode of growth. Cells in different regions of a biofilm show different patterns of gene expression. The complexity of biofilm structure has prompted a comparison to the social behavior of multicellular organisms. Using a remarkable molecular “language” called quorum sensing, bacteria in the company of others sense both their own numbers and those of other species, altering their behavior and collectively forming biofilms (45).

Genetic vaccines

DNA (genetic) vaccines, in which naked DNA is injected in the form of plasmids, are a promising new approach to prevention. Target cells take up the injected plasmids and begin manufacturing proteins coded for by the new DNA. Strong humoral and cell-mediated immune responses usually result. A potential complication of DNA vaccination is incorporation of vaccine DNA in fetal or germ-line cells, which might induce immunological tolerance in the progeny, with resulting susceptibility to infection and to development of a carrier state (46).

An especially intriguing form of genetic vaccination is expression-library immunization, pioneered by Stephen Johnston at the University of Texas Southwestern Medical Center. This technique breaks up the entire genome of a pathogen into expression libraries representing a portion of the pathogen's genome. DNA from these libraries is injected into mice, which are later challenged with the pathogen. Those particular fragments that protected a subset of the host population are further broken down and the new fragments injected into another population of mice. The final individual gene fragments most effective in providing protection are then incorporated into a multi-component vaccine. The most remarkable feature of the technique is that the immune system has been used to screen candidate genes; nothing needs to be known of the pathogen's biology (47).

Bioterrorism

Bioterrorism has suddenly (and belatedly) taken center stage. The threat of biological warfare is decades old, and bioweapons are potentially more devastating than nuclear weapons. Smallpox or plague can leapfrog over large geographic areas like a biological crown fire, and current supplies of antibiotics and vaccines are inadequate to quench the conflagration. According to Ken Alibek, an expert in Russia's bioweapons program who defected to the USA, at least 70 different types of bacteria, viruses, rickettsiae, and fungi can be turned into weapons (48). One panel of experts listed smallpox, plague, anthrax, and botulism as the most likely choices for bioweapons (49). For decades the Russian program produced genetically altered agents that were uncommonly virulent and antigenically modified so that traditional vaccines would not provide protection. A gene for myelin toxin, which destroys the myelin sheath around neurons, was successfully introduced by Russian scientists into Yersinia pestis, the agent of plague. An Ebola-smallpox chimera may also have been engineered, marrying the virulence of both agents to the airborne spread of smallpox.

Bioterrorism and emerging infectious diseases call for a new awareness of infectious disease epidemiology, including how pathogens and parasites evolve and how we may be subverting the evolutionary process by genetic engineering. Laboratory-induced virulence and epidemic spread by human design are new kinds of threats with no counterpart in the evolutionary past.

Acknowledgements: Miriam Muallem, librarian at Medical City Dallas Hospital, provided skilled guidance acquiring reference material. Mary Beth Dimijian, wife of the author, and Cynthia D. Orticio, managing editor of BUMC Proceedings, provided invaluable editing advice and assistance.

1. Dow U. Birth registration: the `first' right. The Progress of Nations 1998. UNICEF (United Nations International Children's Emergency Fund), 1999:5.
2. Dimijian GG. Pathogens and parasites: insights from evolutionary biology. BUMC Proceedings 1999;12:175–187.
3. Morris JG Jr. Pfiesteria, “the cell from hell,” and other toxic algal nightmares. Clin Infect Dis 1999;28:1191–1196.
4. Ezzell C. It came from the deep. Sci Am 1999;280:22, 24.
5. Enserink M. Biological invaders sweep in. Science 1999;285:1834–1836.
6. Whitman WB, Coleman DC, Wiebe WJ. Prokaryotes: the unseen majority. Proc Natl Acad Sci USA 1998;95:6578–6583.
7. Fuhrman JA. Marine viruses and their biogeochemical and ecological effects. Nature 1999;399:541–548.
8. Blower SM, Porco TC, Darby G. Predicting and preventing the emergence of antiviral drug resistance in HSV-2. Nat Med 1998;4:673–678.
9. Vogel JP, Andrews HL, Wong SK, Isberg RR. Conjugative transfer by the virulence system of Legionella pneumophila. Science 1998;279:873–876.
10. Mazel D, Dychinco B, Webb VA, Davies J. A distinctive class of integron in the Vibrio cholerae genome. Science 1998;280:605–608.
11. Maynard Smith J, Szathmary E. The Major Transitions in Evolution. Oxford: Oxford University Press, 1995.
12. Pollack R. Signs of Life: The Language and Meanings of DNA. Boston: Houghton Mifflin, 1994.
13. Ploegh HL. Viral strategies of immune evasion. Science 1998;280:248–253.
14. de Duve C. Vital Dust: Life as a Cosmic Imperative. New York: Basic Books, 1995.
15. Kidwell MG, Lisch DR. Hybrid genetics. Transposons unbound. Nature 1998;393:22–23.
16. Burke WD, Malik HS, Lathe WC III, Eickbush TH. Are retrotransposons long-term hitchhikers? Nature 1998;392:141–142.
17. Hamilton WD. Narrow Roads of Gene Land: The Collected Papers of W.D. Hamilton. New York: Oxford University Press, 1996.
18. Wahlgren M, Bejarano MT. A blueprint of `bad air.' Nature 1999;400:506–507.
19. Elliott DE, Li J, Crawford C, Blum A, Metwali A, Khurram Q, Urban J, Weinstock JV. Exposure to helminthic parasites protect mice from intestinal inflammation [abstract]. Gastroenterology 1999;116:A706.
20. Agrawal A, Eastman QM, Schatz DG. Transposition mediated by RAG1 and RAG2 and its implications for the evolution of the immune system. Nature 1998;394:744–751.
21. Ewald PW. Evolution of Infectious Disease. Oxford: Oxford University Press, 1994.
22. Ewald PW. Portrait of a pathogen. Sci Am 1997;276:112–116.
23. Ebert D. Experimental evolution of parasites. Science 1998;282:1432–1435.
24. Fu Y, Galan JE. A Salmonella protein antagonizes Rac-1 and Cdc42 to mediate host-cell recovery after bacterial invasion. Nature 1999;401:293–297.
25. Harvell CD, Kim K, Burkholder JM, Colwell RR, Epstein PR, Grimes DJ, Hofmann EE, Lipp EK, Osterhaus AD, Overstreet RM, Porter JW, Smith GW, Vasta GR. Emerging marine diseases—climate links and anthropogenic factors. Science 1999;285:1505–1510.
26. Morell V. How the malaria parasite manipulates its hosts. Science 1997;278:223.
27. Dawkins R. The Selfish Gene, new ed. Oxford: Oxford University Press, 1989:234–266.
28. Dawkins R. The Extended Phenotype. Oxford: WH Freeman and Co, 1982.
29. Webster JP, Brunton CF, MacDonald DW. Effect of Toxoplasma gondii upon neophobic behaviour in wild brown rats, Rattus norvegicus. Parasitology 1994;109:37–43.
30. Canby TY. Bacteria: teaching old bugs new tricks. National Geographic 1993;184:36–61.
31. Huyghe P. Conan the bacterium. The Sciences 1998;38:16–19.
32. Madigan MT, Marrs BL. Extremophiles. Sci Am 1997;276:82–86.
33. Psenner R, Sattler B. Life at the freezing point. Science 1998;280:2073–2074.
34. Posavad CM, Koelle DM, Corey L. Tipping the scales of herpes simplex virus reactivation: the important responses are local. Nat Med 1998;4:381–382.
35. Ostfeld RS, Keesing F, Jones CG, Canham CD, Lovett GM. Integrative ecology and the dynamics of species in oak forests. Integrative Biology 1998;1:178–186.
36. Jones CG, Ostfeld RD, Richard MP, Schauber EM, Wolff JO. Chain reactions linking acorns to gypsy moth outbreaks and Lyme disease risk. Science 1998;279:1023–1026.
37. Lee A. The Helicobacter pylori genome—new insights into pathogenesis and therapeutics. N Engl J Med 1998;338:832–833.
38. Putsep K, Branden C, Boman HG, Normark S. Antibacterial peptide from H. pylori. Nature 1999;398:671–672.
39. Moran N, Baumann P. Phylogenetics of cytoplasmically inherited microorganisms of arthropods. Trends in Ecology and Evolution 1994;9:15–20.
40. Stahl EA, Dwyer G, Mauricio R, Kreitman M, Bergelson J. Dynamics of disease resistance polymorphism at the Rpm1 locus of Arabidopsis. Nature 1999;400:667–671.
41. Kuwahara C, Takeuchi AM, Nishimura T, Haraguchi K, Kubosaki A, Matsumoto Y, Saeki K, Matsumoto Y, Yokoyama T, Itohara S, Onodera T. Prions prevent neuronal cell-line death. Nature 1999;400:225–226.
42. Sakaguchi S, Katamine S, Nishida N, Moriuchi R, Shigematsu K, Sugimoto T, Nakatani A, Kataoka Y, Houtani T, Shirabe S, Okada H, Hasegawa S, Miyamoto T, Noda T. Loss of cerebellar Purkinje cells in aged mice homozygous for a disrupted PrP gene. Nature 1996;380:528–531.
43. Lorber B. Are all diseases infectious? Ann Intern Med 1996;125:844–851.
44. Costerton JW, Stewart PS, Greenberg EP. Bacterial biofilms: a common cause of persistent infections. Science 1999;284:1318–1322.
45. Strauss E. A symphony of bacterial voices. Science 1999;284:1302–1304.
46. Gorecki DC, Simons JP. The dangers of DNA vaccination. Nat Med 1999;5:126.
47. Barry MA, Lai WC, Johnston SA. Protection against mycoplasma infection using expression-library immunization. Nature 1995;377:632–635.
48. Alibek K. Biohazard. New York: Random House, 1999:281.
49. Henderson DA. The looming threat of bioterrorism. Science 1999;283:1279–1282.