INTRODUCTION

The importance of regulatory DNA in long-term evolution is well recognized. Major evolutionary advances such as the origin of the bacteria, vertebrates, and mammals can be interpreted largely in terms of regulation of gene expression (1,2). Short-term evolution also is mediated substantially by changes in gene expression, and polymorphism in regulatory DNA provides a major component of genetic variation in natural populations (3–5). Such variation has been particularly well studied in respect to disease susceptibility in human populations, including susceptibility to infection. The immune system is a rich source of regulatory genetic variation, notable in the following four areas:

1. Inflammation, where variation in the expression of pro- and anti-inflammatory cytokines is conspicuous.

2. Th1/Th2/Treg balance, where differentiation into distinct T-cell subsets is regulated by the timing and level of gene expression, as can be modeled by Hopff bifurcation (6).

3. In the constitutive immune system, where multiple copies of, e.g., interferon genes occur.

4. In the generation of diverse antigen receptor repertoires, where variability among signal sequences mediating V(D)J recombination may regulate ordered assembly and allelic exclusion of the H, β, and δ loci and may bias the preselection repertoire.

In short, the immune system does not know what it will need to cope with next and uses regulatory DNA to provide some of the flexibility needed to function effectively. The same means are also used elsewhere to provide balance and flexibility, e.g., in the cardiovascular system (7).

The problem is that variation is harder to interpret in regulatory than in coding DNA.

INTRODUCTION

Analysis of Plasmodium spp. genome sequences would allow for the discovery of thousands of new genes and proteins, providing a unique opportunity for understanding the complex biology of malarial parasites. However, the identification of these new genes will be followed by the same old questions that researchers have faced for nearly three decades: how variable are the newly identified genes? How is such variation generated and maintained? How can this diversity affect intervention efforts?

Understanding the origin and extent of malarial parasites' genetic diversity, and the implications of these on the development of new intervention strategies, requires a close collaboration between biomedical researchers and evolutionary biologists. In the case of malaria research, as is the case in the study of many other infectious diseases, biomedical researchers/public health professionals and evolutionary biologists have traditionally worked in isolation.

Biomedical researchers and public health managers often seek answers to the following questions: What are the specific clinical end points in malaria? What level of efficacy can be expected from a multivalent vaccine? How does drug-resistance emerge? How quickly do drug-resistant parasites disperse? What is the impact of transmission pressure on the dispersal of drug-resistant parasites? Is it better to use one drug at a time or to use a “cocktail” before any of the drugs become ineffective? What is the impact of bed-nets on the selection of parasite lines?

I don't know who first had the idea that disease is a very potent agent of natural selection. I do know, however, that one of the first was J. B. S. Haldane. In The Causes of Evolution (Haldane 1932), he wrote: “A study of the causes of death in man, animals, and plants leaves no doubt that one of the principal characters possessing survival value is immunity to disease. Unfortunately, this is not a very permanent acquisition, because the agents of disease also evolve, and on the whole more rapidly than their victims.” Insightful as it was, this statement seems to have had little impact on evolutionary thought at the time (Sarkar 1992).

A further development of the subject came a few years later in a now-famous paper “Disease and evolution” (Haldane 1949b). Recognizing that density is often a limiting factor in population growth, he argued that the most important density-dependent limiting factor is a parasite. Overcrowding favors the parasite by, among other things, weakening the hosts and facilitating transmission among them. As an example, he discussed the detailed causes of death at various times in the life cycle of the gall insect, Urophora jaceana, and its various parasitoids. He noted that genetic diversity is much greater for disease resistance than, for example, resistance to predators. When a variety of wheat has been exposed to rust, a new resistant strain soon appears, followed by a new parasite strain that attacks the resistant host, and so on.

INTRODUCTION

The discovery of blood groups by Landsteiner (1900) coincided with the rediscovery of Mendelism. During the next generation, immunology and genetics developed together, but the ideas and even the vocabulary that are essential today were slow to evolve. The term population genetics was not used by Fisher, Wright, or Haldane but was introduced by C. C. Li (1948) in his magisterial synthesis that anticipated division of the field into what we now call evolutionary genetics and genetic epidemiology. Neel and Schull (1954) devoted a chapter of their book to epidemiological genetics, which evolved into genetic epidemiology to reflect the dual origins of disease (Morton et al., 1967). A characteristic of the time was that definitions appeared first in books rather than in articles. However, the pace quickened as human genetics rose from the ashes of eugenics. Haldane (1949) suggested that thalassemia minor might be protective against malaria. The young regional societies of human genetics and the first international congress of human genetics were excited by critical evidence for the link between hemoglobin S and falciparum malaria (Allison, 1954). At the same time, Watson and Crick (1953) created molecular biology, which provided the techniques and markers for genetic epidemiology of infectious disease. The essential vocabulary was slowly invented: linkage disequilibrium (Ohta and Kimura, 1969), haplotype (Ceppellini et al., 1967), diplotype (Morton, 1983), and linkage disequilibrium unit or LDU (Maniatis et al., 2002).

ABO BLOOD GROUPS

Fifty years ago polymorphisms were generally thought to be rare and maintained by selection.

THE INFECTIOUS THREAT

This opening century will appear as both “the golden age of genetics and the dark age of infectious diseases” (Tibayrenc, 2001a). On the battlefront of infectious diseases, the situation is more than just a concern, due to the threat of emerging and reemerging infectious diseases (ERID). In developing countries, infectious diseases still are the main demographic regulating factor. In particular, Africa is more than ever afflicted with sleeping sickness, malaria, bilharziosis, and other major parasitoses. The three “diseases of poverty,” namely malaria, tuberculosis, and AIDS, have become the top priority of the World Health Organization. The industrial world has not been spared. In France, 12,000 people die every year of nosocomial infections. In New York City, 25% of the Mycobacterium tuberculosis strains are resistant to antibiotics.

ENVIRONMENTAL AND BIOLOGICAL FACTORS

Transmission and severity of infectious diseases are the result of a complex interplay between environmental and biological (built-in) parameters. There is no doubt that environmental factors play a major role in the present resurgence of infectious diseases, through climatic changes, massive migrations, economic inequalities, and political instability. However, even in acting on these environmental factors, control is more efficient when sophisticated knowledge of the biology of the disease under survey is available. For example, in Latin America, Chagas disease is a parasitic disease caused by the flagellate Trypanosoma cruzi and transmitted by triatomine bugs (hematophagous tree bugs).

INTRODUCTION

Toxigenic strains of the Gram-negative bacterium Vibrio cholerae cause the most severe form of dehydrating diarrhea known as cholera and have been responsible for at least seven distinct pandemics of cholera (Faruque et al., 1998a). Since the first recorded pandemic, which began in 1817, V. cholerae strains associated with different pandemics are assumed to have undergone phenotypic and genetic changes with time. For example, although the seventh pandemic was caused by the El Tor biotype of V. cholerae, the sixth and possibly earlier pandemics were caused by the classical biotype. The genetic changes in V. cholerae associated with epidemics and pandemics, however, were not fully appreciated until the development and application of molecular techniques to analyze strains.

Recent application of molecular approaches has enabled extraordinary progress in our understanding of the evolution of V. cholerae. Like other bacteria, V. cholerae can be assumed to have existed well before human evolution and therefore to have originated primarily as a free-living aquatic microorganism. The fact that over a period of a million years some clones (serogroups) of this organism have acquired the ability to colonize transiently the human intestine and become an efficient human pathogen reflects the progressive acquisition of the genetic capability to become pathogenic for humans. They could have acquired this capability by embracing another “life style,” albeit accidentally, by carrying out a function in the human host that it performs either symbiotically or commensally for its nonhuman host but is, in effect, pathogenic to the human host, or alternatively, by finding a niche whereby the organism could amplify and perpetuate itself as effectively or more effectively, than it could in a free-living state.

SUSCEPTIBILITY TO SEVERE DISEASE DURING AN OUTBREAK OF VISCERAL LEISHMANIASIS IN A SUDANESE VILLAGE

Visceral leishmaniasis (VL) is caused by protozoan parasites of the Leishmania genus that are transmitted to humans by infected sandflies (Figure 10.1a). Parasites rapidly invade host phagocytes and multiply inside phagolysosomes. Clinical disease is primarily due to the uncontrolled multiplication of the parasite in many organs including the spleen and the liver; clinical symptoms include recurrent fever, considerable splenomegaly, hepatomegaly, and adenopathy (1). Death is certain if the patient is left untreated. Violent outbreaks of VL have occurred in regions of eastern Africa (2) (Kenya, Sudan) and in India (Bihar). VL is endemic in South America and in the Mediterranean basin. VL is caused by three Leishmania species: Leishmania donovani, L. chagasi/infantum, and L. archibaldi (L. donovani being the most pathogenic) (3). To identify the principal risk factors in VL, we carried out a five-year longitudinal study on 1,600 subjects from a village located on the Sudanese–Ethiopian border. The study was initiated in 1995 when the number of VL cases had just begun to rise in the village (4). Within five years, 28% of the population had been affected by VL. Most of these VL patients were treated and cured. Unfortunately, a small percentage either failed to respond to treatment or were not treated because of the difficulties involved in reaching them during the rainy season.

INTRODUCTION

Haldane (1949) was one of the first to speculate that the genetic outcome of parasitic interactions may have profound implications, in particular, that frequency-dependent selection on host and parasite polymorphisms can maintain genetic variation and promote sexual reproduction. Much theoretical work since then has supported his early insights: dynamic genetic polymorphisms or arms races are a common outcome of computer simulations of host–parasite coevolution. Recent theory now prompts us to ask precise questions about parasite-associated dynamics: Do host and parasite alleles cycle with regularity or as bursts followed by long periods of stasis? Do arms races result from directional selection on mutational input or does selection maintain alleles to antiquity (Hill et al., 1992; Hughes and Nei, 1992; Stahl et al., 1999)? The other major line of theory on host–parasite interactions, also foreshadowed by Haldane, concerns the evolution of virulence (the damage to the host caused by parasites), and this work has now developed explicit predictions concerning the conditions under which parasites ought to evolve towards doing greater or less harm to their hosts (Anderson and May, 1982; Ewald, 1983).

Probing these questions and predictions will shed light on the evolutionary significance of parasitism as envisioned by Haldane (1949) and others, because the exact tempo, mode, and outcome of selection matter if we are to properly evaluate the evolutionary significance of parasitism. However, empirical evidence lags behind theory, and it remains uncertain just how widespread and strong is genetic change associated with parasitic interactions in the wild.

INTRODUCTION

Pathogen evolution poses the critical challenge for infectious disease management in the twenty-first century. As is already painfully obvious in many parts of the world, the spread of drug-resistant and vaccine-escape (epitope) mutants can impair and even debilitate public and animal health programs. But there may also be another way in which pathogen evolution can erode the effectiveness of medical and veterinary interventions. Virulence- and transmission-related traits are intimately linked to pathogen fitness and are almost always genetically variable in pathogen populations. They can therefore evolve. Moreover, virulence and infectiousness are the target of medical and veterinary interventions. Here, we focus on vaccination and ask whether large-scale immunization programs might impose selection that results in the evolution of more-virulent pathogens.

The word virulence is used in a variety of ways in different disciplines. We take a parasite-centric view as follows. We use “disease severity” (morbidity and/or mortality) to mean the harm to the host following infection. Disease severity is thus a phenotype measured at the whole-organism (host) level that is determined by host genes, parasite genes, environmental effects, and the interaction between those factors. One component of this is virulence, a phenotypic trait of the pathogen whose expression depends on the host. Thus, virulence is the component of disease severity that is due to pathogen genes, and it can be measured only on a given host. We assume no specificity in the interaction between host and pathogen (more-virulent strains are always more virulent, whatever host they infect).

We have investigated the population structure of P. falciparum by analyzing several genes and conclude that the extant world populations of this parasite have evolved from a single strain within the past several thousand years. The evidence is based on a lack of synonymous polymorphisms among nuclear antigenic and nonantigenic genes; among introns; and among mitochondrial loci. Coalescence calculations for silent nucleotide variation converge in each case into a single ancestral allele. The extensive polymorphisms observed in the highly repetitive central region of the Csp gene, as well as the apparently very divergent two classes of alleles at the Msp-1 gene, are consistent with this conclusion.

Understanding the population structure and evolution of Plasmodium has important implications for the control of human malaria.

The human toll of malaria is stunning, perhaps the greatest of all human afflictions (Sherman, 1998). Malaria is caused by species of Plasmodium, a parasitic protozoan. Four species of Plasmodium are parasitic to humans: P. falciparum, P. malariae, P. ovale, and P. vivax. P. falciparum is the most pervasive and malignant human malarial parasite. It causes yearly 300 million to 500 million cases of clinical illness and 1.5 million to 2.7 million deaths in sub-Saharan Africa, plus 5–20 million clinical cases and 100,000 deaths elsewhere in the world, 80% of them in Asia (Trigg and Kondrachine, 1998).

The genus Plasmodium consists of nearly 200 named species that parasitize reptiles, birds, and mammals.

In 1948, J. B. S. Haldane proposed that the selective agent for maintaining the high frequency of thalassemia in the Mediterranean races might be malaria. Thus was born what, in the human hemoglobin field at least, later became known as the malaria hypothesis. In this short essay, I examine the origins of this hypothesis, ask how it has stood the test of time, and summarize the broader field of research which it has spawned and which attempts to understand the genetic basis of variability in individual susceptibility to infection in general. Many of these issues have been reviewed in detail over recent years and are only summarized here.

THE MALARIA HYPOTHESIS

Although the first clinical descriptions of the thalassemias appeared in the 1920's, it only became apparent that these conditions follow a Mendelian recessive or co-dominant pattern of inheritance in the period immediately preceding World War II. At this time relatively simple methods for carrier screening became available and hence it was possible to carry out population studies to attempt to determine gene frequencies.

The extraordinarily high frequency of the genes for thalassemia puzzled population geneticists, particularly those who had become interested in mutation rates as a result of their studies of the survivors of the atomic bombs that had been dropped on Hiroshima and Nagasaki. Extensive studies of gene frequencies of thalassemia were carried out quite independently in the 1940's by workers in the United States and Italy (Neel and Valentine, 1947; Silvestroni and Bianco, 1947), but because of lack of communication during the war and its aftermath it was not until the early 1950's that this broad body of work could be integrated and assessed.

INTRODUCTION

Over the course of the past century, medical science has produced rapid advances in human health. At first, the germ theory and Koch's postulates framed the search for pathogens causing major infectious diseases, and later, Mendelian genetics and modern molecular techniques were used to identify the genes responsible for inherited syndromes. A number of major human diseases are of unknown origin, including, but not limited to atherosclerosis, multiple sclerosis, rheumatoid arthritis, schizophrenia, manic depression, ulcerative colitis, many cancers, and diabetes mellitus.

In some cases, these diseases have plagued humanity for evolutionarily long periods. These diseases present a puzzle to evolutionary biologists. What selective pressures can maintain a high frequency for these deleterious syndromes over evolutionarily long periods? Using an adaptationist point of view, we have suggested that if a syndrome has been common in a well adapted population for an evolutionarily long period and maintains a large negative selective effect on the population, then the syndrome is likely to be caused, directly or indirectly, by infection (Cochran et al., 2000).

Diabetes may be one such disease. It is ancient, has a severe selective effect, and is relatively common; this combination of factors leads us to suspect that diabetes may be caused by infection. In this study, using an evolutionary perspective and with an eye on infection, we examine the etiology of diabetes mellitus.

EVOLUTIONARY PRESSURE, EPIDEMIOLOGY, AND FITNESS

What are the evolutionary pressures driving human disease syndromes?

INTRODUCTION

Humans have been plagued by malarial parasites for centuries and reference to maladies associated with malaria may be found in antiquities over the past 5,000 years (1). For much of this time the cause of the intermittent chills and fevers, splenomegaly, and mortality associated with malaria was unknown. However, with consistent identification of the brownish black pigment (hemozoin) found during autopsies of malaria victims from the early 1700s on, scientific discovery methodically began to dissect malarial parasites from the various secret hiding places of their complex life cycle. Alphonse Laveran first observed the tiny ring-stage parasites of the malaria blood-stage infection in 1880 (1), and Ronald Ross would reveal that the female anopheline mosquito was responsible for malaria transmission in 1897 (1). During the late 1800s and ending in 1922 individual discoveries illustrated that malaria in humans was caused by four distinct species of Plasmodium – P. falciparum, P. vivax, P. malariae, and P. ovale (2), and that fevers resulting from infection by these parasites would soon find their way into successful, albeit unorthodox, treatment of neurosyphilis.

The era of malaria therapy, launched in earnest by Julius Wagner von Jauregg in 1917 (3), paved the way for important advances in malaria research as observations resulting from thousands of treated patients provided the opportunity to study early stages of infection, development of immunity and characteristics of the immune response, and the efficacy of various antimalarial drugs (4).