What exactly is meant by a “logic of disease”? Webster’s defines the word logic variously: (1) “a statement of the formal principles of a branch of knowledge,” and (2) “interrelationships or connections or sequences of facts and events seen by rational analysis as inevitable, necessary or predictable.” The questions here are, Are there principles of disease that flow by means of predictable connections and sequences from the facts of evolution and natural selection? Are there regularities in the nature, frequency, and characteristics of diseases that can be shown to be consequences of the human genetic and cultural condition? and Have the whole range of diseases some common properties that stem from the human substrate they all share? If such a logic can be demonstrated it should prove useful, perhaps principally to the medical teacher in offering the students both a framework for the facts and a philosophy, defining that word as embodying principles and causes of reality.38 For the practitioner it should be useful in demystifying imponderables that must be confronted even in “straightforward” cases. The idea of such a logic of disease is well within the Garrodian canon and altogether foreign to that of Osler. Indeed, it might be said with reason that Inborn Factors 6 is an argument for just such a logic.
A critic might protest that human physiology and the pathogenesis of disease are far too complex to allow the elaboration of such a logic. There is too much randomness. Living organisms are not simple linear systems; their moment-to-moment behavior depends on their own state. So predictions can be only probabilistic at best. But even in chaotic systems like the weather, where conditions may change hourly, long-range predictions of climate are usually fulfilled. It is the rules of “climate” rather than “weather” that constitute the logic of disease. A critic might also say that any such logic must defy composition. The facts are too many and too diverse, the diseases too varied, the homeostasis too complex to be embraced in a set of principles. If such principles are truly comprehensive, they must be banal, and if not, then no logic will be apparent. But the authors of Molecular Biology of the Cell, a book of 1200 pages that comprehends much of what is known about the structure and function of cells, say in their preface, “There is a paradox in the growth of scientific knowledge. As information accumulates in ever more intimidating quantities, disconnected facts and impenetrable mysteries give way to rational explanations and simplicity emerges from chaos.”42 If simplicity emerges from chaos in the biology of normal cells, why not from the biology of cells caught up in the processes of disease?
Such a logic can evolve only when the thinking of biologists and that of physicians are so interlaced as to have attained the identity of a new discipline. This evolution is occurring rapidly as contrasts between the aims of cell and molecular biology and those of medicine are resolved. Some of these contrasts and convergences are reviewed below.
Biology and Medicine: Some Contrasts
In the practice of medicine the physician is confronted by outcomes of individual incongruence. Medical thought is not primarily concerned with such general biological aims as how the species has become attuned to the environment but with how patients have come to represent discordant notes in nature’s harmony. So the physician tries to describe the behavioral, physiological, biochemical, and molecular attributes of individuals dissonant with life. That is not to say that physicians are altogether indifferent to the evolutionary forces that have led to adaptation and congruence or that biologists are indifferent to incongruence. Rather, the rules for a logic of disease must originate in an emphasis unlike that of either molecular biology or medicine. It is an emphasis perhaps more congenial to the population biologist or geneticist, whose views accommodate both the characterization of species and genetic individuality. If so, it will be useful to begin by contrasting the aims of biology with those of medicine and to show how they are coming to coincide.
Biological thinking perceives individuals as vehicles for the transmission of genes and a means for generation of the variability required to maintain the species. In this view individuals represent the variegated materials from which an indifferent nature creates the fabric of species. In sharp contrast, medical thought perceives each human being as an individual person with an option to pursue life, liberty, and happiness, and it is the business of medicine to correct individual encumbrances to those pursuits. Molecular genetics is modifying this view of the individual to include uniqueness: a Garrodian chemical individuality that is expressed in disease no less than in other aspects of life.
In studying the elements of congruence, the questions biologists ask are given largely typological answers: this is how replication is achieved in Escherichia coli, homeotic genes control body segmentation, and so on. Further, although physiology is a strongly quantitative discipline, biochemistry is less so, and molecular biology is almost altogether qualitative. So for molecular biology, the classes are the species, and for the quantitative disciplines, individuality is disguised in means and deviations. Typological thinking in these disciplines is natural; variation is noise and biologists are interested in elaborating general rules of development and homeostasis. But a logic of disease requires population thinking; individuals are ill because their particular qualities are at odds with their particular experiences. The differences in goals between biology and medicine are also reflected in the language employed. In traditional medical thinking much attention is given to proximate causes. This preoccupation is revealed in our classification of diseases as “genetic” or “environmental.” Although these typological classes are losing their sharp definition as the genetic contribution to “nongenetic” disorders is affirmed, we still call monogenic phenotypes “genetic,” and we often refer to “genes for” even such ill-defined conditions as hypertension or schizophrenia. Molecular biologists are less likely to think of their phenotypes as genetic or nongenetic. Their interest in proximate causes derives from a need, not to circumscribe and classify, but to show how gene-specific molecules integrate in structure and function.42 But when both are working at the level of the DNA and its products, even when pursuing quite different ends, differences in language and concept disappear. One result of this confluence of purpose is a growing interest among molecular biologists in “medical” problems as they are expressed in molecular aberration. This infusion was observed in the sixth edition of MMBID, in which, in the “General Themes” section, authors with nonmedical degrees were nearly equal in number to those with MDs, while among the rest, people with nonmedical degrees made up only 25 percent of the authors.
A central feature of biology is the union of two gametes to engender a zygote. Population biology is much taken up with mating systems and families, reproductive fitness, ethnicity, and migration, all marginal to traditional medical thought. Physicians focus on people one at a time and deal only exceptionally (e.g., pediatrics or obstetrics) with couples or families; examples are couples plagued by infertility or whole families exposed to malign infections. But consanguinity, ethnicity, and history of illness among relatives are now staples of the encounter of patient and doctor. They represent the recognition of remote causes with their significance in preventive thinking. Whole families are also the basis for linkage studies that lead to the detection and description of genes of medical moment.
A salient difference between biological and medical thought lies in the significance of time. To the biologist everything is “time-bound.”43 The organism “is not a particular expression of an ideal organism, but one thread in an infinite web of all living forms, all interrelated and all interdependent.”43 Biologists perceive ontogeny as a prelude to phylogeny; the development of each individual represents a test of the robustness and viability of the species. But in traditional medical thinking, species and the history of the DNA are of marginal interest or are irrelevant. The physician’s attention is held by the here and now, by what must be done now. In biology, “To every thing there is a season, and a time to every purpose under the heaven, a time to be born and a time to die … ” (Eccles. 3:1, 2). Not so in medicine, in which if imminent birth will be harmful, the time to be born must be changed, and a primary aim is to defer death as long as possible. To these ends medicine is practiced cross sectionally, the lifetime is stratified; obstetrics, pediatrics, adolescent medicine, internal medicine, and geriatrics all represent compartments wherein the nature of the work is determined by the age at onset of the disease. Such a narrow focus is likely to minimize the physician’s appreciation of the unity of a lifetime that proceeds from zygote to dissolution, as well as to obscure the historical quality of development in which the attributes of today carry the stamp of the past.
But here too our thinking is beginning to change, and in two ways. The first is in the recognition of the predictive powers of genetics, which provide the wherewithal for reproductive decisions for individuals and families found to harbor undesirable mutants, as well as for probabilities for recurrence of many diseases associated with specific genes known to contribute to diseases of complex origin. These perceptions of the influence of genetic variation on disease are affecting the compartmentalization of medicine. Divisions and departments of genetics cross age boundaries to reveal patients as people with a past that begins in the lives of their ancestors and with a future that will be colored by genes inherited from those predecessors. Here the biologist’s idea of time is imposed on the pragmatism of medicine.
The second way our thinking is changing is in our recognition that predispositions to diseases of late onset reside not only in the genes but in experiences of early life, beginning with conception and including intrauterine life.44 Once the natural history of disease included only its history from some overt onset to its end. Now the natural history may be traced to conditions of hereditary susceptibility with early onset of subtle expressions that show a steady progression from genetic potential to pathogenesis. It is as if people grow into their diseases. Atherosclerosis and hypertension are examples.
What is it specifically about evolution that suggests a logic of disease? After defining evolution, the argument will be presented in a series of propositions, each based on some aspect of the definition and each capable of expansion to include subsections and examples.
How is evolution defined? It is the process biological organisms undergo to attain the morphological and physiological characteristics that identify the species. It proceeds by descent with such modification as may be necessary to achieve adaptive flexibility in unstable environments. The modifications are a result of a process of sorting through stocks of genetic variation to find that which facilitates the broad range of congruence required for development and reproduction, and for the continuity of life both between and within species. The success or failure of such variants in fulfilling this mission is likely to be reflected in their frequencies in subsequent generations. Words in this definition most pertinent to the logic of disease are (1) species identity, (2) adaptive flexibility and development, (3) descent with modification, and (4) congruence and continuity. These key words and phrases can be elaborated into propositions that embody the concepts of the logic.
Evolution is the record of biological history. Life has proceeded from the simple to the complex by integrating simple systems into more complicated aggregates, leading, at each new level of integration, to new constraints as well as new qualities.45,46 The constraints are imposed because the new level of integration represents only one of several or many that might have occurred but did not, and new qualities emerge because old systems have been aggregated to interact in new and not necessarily predictable ways.45,46 The basic biochemical elements of these systems took form very early in evolution, and diversification was made possible by new variation on old themes, including the differential use of existing biochemical mechanisms, the elaboration of new genes and gene families, and the development of more complex and refined controlling elements. Thus, phylogeny represents an uninterrupted continuum in which both change and stability are reflected in the sequences of units in the molecules of both DNA and protein, and despite the morphological and reproductive discontinuity that is observed in species, the molecular continuity remains. So, although species identity is sharply defined, shared properties that transcend species identity are retained, with implications for the logic of disease. This chemical continuity both between and within species was noted by Garrod.5
Lessons from Phylogeny: The Constraints of Evolution.
Toward the end of Pope’s “An Essay on Man,” appears the line, “Whatever is, is right.” We need not subscribe to the 18th-century ideology that informed the “Essay” to see the line as a metaphor for the relationship between a species and its environment. If a quality exists, it has stood the test of selection and is likely to serve, or to have served, some useful function.47 It need not be the best possible adaptation; evolution is a process of tinkering. But whatever is selected is likely at some time and condition to represent some advantage. The search for such useful functions is the essence of what evolutionary biologists call “The adaptationist program.”47
Adaptation is defined in relation to some environment or particular experiences, and however flexible the homeostatic defenses, adaptation to one environment or experience is likely to represent a constraint for another. Some such constraints may lead to incongruences: Some are species-specific; others, because of shared molecular identity, may cross species. There is preventive significance in such observations. That is, where particular experiences lead to disease in one species, they may do so in others. Similarly, environments and experiences that are foreign to one species are likely to be alien to individuals of related species. Two examples are given below. In one, a popular human experience shared by no other species has proven to be harmful to many individuals. In a second, a species-specific adaptation has proved to exert an unexpected constraint.
Nowhere among plants and animals do we find any who seek or tolerate an atmosphere of smoke; no species other than Homo sapiens purposefully dilutes inspired oxygen with carbon monoxide. A biologically alert society might have made these observations and taken warning; for example, today no substance capable of the degree of damage plainly caused by tobacco could possibly be widely advertised and legitimately sold—except tobacco. So, if it had been observed that inhalation of smoke is natural to no organism, the possibility of harm to human beings might have been carefully examined. But, as it happened, the hazard of smoking became apparent in Britain only when it was observed to be strongly associated with an otherwise unaccountable rise in the incidence of bronchogenic cancer, and in the United States only when a surgeon observed the frequency of heavy smokers among his patients with lung cancer.48–50 The observations were entirely empirical; no one had tested the hypothesis that smoking might be damaging. Evidence of the relationship of smoking and chronic obstructive pulmonary disease and heart attack came later, after suspicion about the bad effects of smoking had been aroused.
Today we know that smoking is associated with cancer of several organs other than the lung—with chronic pulmonary disorders, atherosclerosis and heart attack, reproductive hazards, and indeed with abnormality of nearly every organ system. It is an imposing list, and in an appraisal of data gathered from 31 “developed” countries, Peto et al. concluded that about 30 percent of deaths of individuals between 35 and 69 years of age in these countries during the 1990s are attributable to smoking.51 So it is now a principal contributor to early mortality. Medical educators might do well to teach the virtues of nature’s examples.
No one would deny that human milk is the “natural” food for newly born human babies; again, evolution has seen to that.52 To begin with, breast-feeding leads to temporary amenorrhea in the mother and a natural spacing of pregnancies of around 2 years.53 Failure of this natural contraception and more frequent childbearing may promote disease and mortality for all infants embraced within the narrowed interval.53,54 But for many reasons, some necessary, some trivial, infants may be fed nonhuman milk.
Although we act as if birth represents a discrete discontinuity from intrauterine life, it is in fact a far less abrupt change than it seems; it is more like doing the same business under new conditions. At birth the infant assumes responsibility for its own physiology and development but must still conform to the trajectory set in utero and supported by the maternal environment. Some of the infant’s physiological equipment needs time to adapt to the new environment, and it is human milk that has been designed by nature to supply the antibodies, nutrients, and other substances that the infant lacks. Cow’s milk is the most popular substitute, but is it appropriate? Perhaps not, at least not for all infants. For example, the rate of development of the brain of the human newborn infant is well behind that of a newborn calf, so we may question whether cow’s milk, which is adapted to the needs of calves, is attuned to the dynamics of early postnatal human development. That it is not is suggested by the discovery that average attainments on tests of cognitive development of premature infants fed cow’s milk were significantly less than their breast-fed coevals.55 Further, a comparison of learning-disabled children with controls showed that significantly fewer of the former were breast-fed, 56 and scores in language attainment tests of 15-year-old children were significantly higher among those who had been breast-fed.57 Table 2-1 is a sample of a growing literature on the protective effects of human as opposed to cow’s milk. The reasons for these effects are not always clear. Protection against diabetes is attained by avoiding cow’s milk proteins, 58 but the evolutionary origin of protection against diarrhea is demonstrated by the discovery in human milk of a glycolipid receptor analogue that binds to Shiga toxin, thereby preventing its adherence to its target receptors.59 Another example of protection against microorganisms is a human milk mucin that specifically binds rotaviruses, 60 and still another is an enhancement of response to Haemophilus influenzae vaccine in breast-fed babies.71 So human milk is best for human babies. No data exist to attest to the dangers of human milk when fed to calves, but we have every reason to suppose they exist. Oddly enough, the evolutionary significance of human milk has been remarked on for years, but until recently medical thinking managed to ignore it, so we have been given all sorts of milk substitutes, some facsimiles, some representing “improvements.” But human milk is a cardinal example of Pope’s principle, or possibly of that of a later, less elegant poet who said, “If it ain’t broke, don’t fix it.”
Table 2-1: Diseases Prevented or Ameliorated by Human Milk |Favorite Table|Download (.pdf) Table 2-1: Diseases Prevented or Ameliorated by Human Milk
|Disorder ||References |
|Type I diabetes mellitus ||168–170 |
|Allergies ||61,62 |
|Necrotizing enterocolitis ||63 |
|Urinary tract infection ||64 |
|Infectious diarrhea ||59,60,65 |
|α1-Antitrypsin and infantile liver disease ||66,67 |
|Otitis media ||68 |
|Childhood lymphoma ||69 |
|Sudden infant death syndrome ||70 |
Adaptive Flexibility: Homeostasis and Development.
Human beings are open systems in and of the world around them, so the species has evolved physiological mechanisms to ensure protective responses to environmental fluctuation. The qualities of the molecules that compose these systems reflect relationships of individuals to family, populations, species, and phylogeny. These same molecules are also engaged in the processes of development and change as individuals move through life. So homeostasis has several faces.
It is amusing to think of Claude Bernard, W. B. Cannon, and L. J. Henderson whiling away an afternoon at the bar of some celestial Harvard Club examining the sixth edition of MMBID. Who can doubt their pleasure in observing that although the word homeostasis does not appear in the index, the ideas are so deeply woven into the thinking of the authors of all the articles that each disease is perceived as a genetic flaw in some homeostatic system. All this they would approve with enthusiasm, even while envying today’s biological scientist the ideas, the information, and the technology available to pursue aims they could only imagine. Bernard, who perceived a milieu interieure that defended the integrity of the cells, and Henderson, who saw the blood as an agent of integration that governed the fluxes of gases, water, and ions into and out of Bernard’s milieu, would be delighted with mechanisms of transport that include gates and channels through which the flow of ions and metabolites is controlled by the contortions of proteins responding to signals that start and stop the flow. And Cannon would rejoice in the complexity of the endocrine system, the peptide hormones that penetrate to the nuclei of cells to influence genes to transcribe RNA. And they might above all applaud the modern description of the integration of components to form and protect the whole organism, since it was the whole organism that was the object of their interest.72–76 Henderson in particular emphasized that stability lay in complexity; the greater the number of components and the more intricate their interrelationships, the greater the stability of the whole.74,75
But they might regret the lack of attention to the relationships of social homeostasis to disease.72,75,76 They would say that since physiological homeostasis is the means whereby an open system maintains its integrity in its environment, some knowledge of how that environment is organized is essential to a full understanding of the internal milieu. After all, it is the outside that tests the inside. All three were much taken by the idea that societies were organized around mechanisms of self-regulation.72,74–76 They recognized the necessity for integration of the elements of society in a government that controls communication, regulation, and protection; individuals, families, and communities can survive only by organizing and integrating systems to preserve the common weal.
At some point they might be joined by C. H. Waddington, who would add that physiological homeostasis would be at risk were it not for the genes whose strategy it is to guide and preserve homeostasis through a lifetime and to connect each person with an ordered past and a possible future.77 Waddington was interested in working out a synthesis of genetics, evolutionary biology, and embryology.78 Lacking knowledge of the molecules involved, he envisioned this interaction in the form of “epigenetic landscapes,” in which he attempted to give dynamic visual form to the differentiation, under genetic guidance, of homogeneous cells into diverse cell types.77 Today, insights into determination, differentiation, and morphogenesis have been given molecular and biochemical body, and the synthesis Waddington visualized in the abstract is a reality.79,80 So there is a homeostasis of the genes that constrains the physiological homeostasis of the individuals who compose the species. Accordingly, Waddington would say, no biological story can ever be complete unless it is told in three time frames at once—that of the moment, that of the lifetime, and that of evolution.77
In introducing the genes into the discussion Waddington would be adding a dimension not emphasized by the others. The study of the mechanisms of physiological homeostasis is the study of proximate causes, but by connecting the momentary time frame with that of the past, Waddington introduces remote causes that regulate whether and how the proximate causes can work. It is through evolution, genetic variation, selection, and adaptation that physiological homeostatic systems have attained the perfection we observe. And it is in these same processes that the remote causes of disease are to be found.
Of the three time frames, Waddington was primarily interested in that of the lifetime.77 Here the strategy of the genes embraces yet another homeostasis, that of ontogeny, wherein the genes help both to promote and to constrain development through early life and on into aging, wherein the integration of physiological homeostasis is loosened. This relationship of development to physiological homeostasis was observed by Cannon, who remarked that people “grow into” their regulatory processes.76 And, he might have added, in aging they grow out of them. So, since physiology changes throughout development, moment-to-moment homeostasis could be said to represent a cross-sectional account of developmental homeostasis. That is, as development proceeds, the matrix wherein the interactions of the genes and experiences take place changes. Early in development, newly activated genes begin their mission, adjusting later to new conditions. Thus, moment-to-moment phenotypes are outcomes of hierarchies of causes derived from three systems of organization—the genes, the environment, and development—each with its own imperatives. And inevitably, each such phenotype exerts a reciprocal influence on the forces that shaped it; there is a dialectical process. This idea has been expounded by Lewontin, who has pointed out how organisms not only adapt to their surroundings but do much to shape them.81
1. The Present: The Phenotype.
The word phenotype is being used here as a stand-in for moment-to-moment homeostasis and its effects. It may be taken to include all the enzyme pathways and cascades, receiving and signaling systems, transporting mechanisms, and other means whereby metabolic business is done. It embraces also the means whereby such systems are integrated in the whole organism to protect the interior from adverse experiences. The molecular details of moment-to-moment homeostasis is the principal subject of the authors of the sections of this book. Less attention has been given to the relationships of physiological homeostasis to the elements of the other time frames.
2. The Past: Genetic and Cultural Homeostasis.
Genetic homeostasis—The purpose of life, as opposed to that of individual lives, is reproduction. To this end each species has a gene pool whose stability is maintained by reproduction, through which both losses and replenishment are attained. The instruments of this stability are the individuals of the species, who as recipients of aliquots of the pool act as filters, transmitting whatever genes and combinations of genes are workable and discarding those that are not by reproductive failure. So there is a genetic homeostasis, which in serving the purpose of reproducing the species serves that of individual lives too, but with the proviso that the preservation of the gene pool comes first. That is, the necessity for replenishment and variation brings in new mutants and new combinations of genes, many inviable, some conditionally tolerable, and perhaps a few even useful. All of these must be submitted to the test of life, so the necessity to replenish and preserve the stability of the gene pool is an overriding remote cause of diseases whose proximate causes are the mutants themselves.
The means whereby the stability of the gene pool is both conserved and promoted is the province of evolutionary biologists. Genetics is first of all a predictive discipline, so it can be no surprise that theories of natural selection and evolution were advanced even before there were data to test them. These data were supplied by Lewontin for Drosophila and Harris for H. sapiens when, after examining the mobility of soluble enzymes in electrophoresis, they found that nearly a third of the loci specifying soluble enzymes of both species were polymorphic and that there was a great deal of rare variation as well.82,83 These studies of enzyme polymorphism have been extended to many hundreds of species with similar results, 84 so we know, at least in regard to soluble enzymes, how variable the species are, and it is well attested that the average human being is heterozygous for 7–10 percent of such loci.82,83 And now we know that the polymorphism of the DNA is far greater still.
It is not yet clear to what degree all this variation, particularly that of proteins, has been attained by selection and how much by chance. But theory suggests it is mainly by chance, and theory is made plausible by the observation that allelic differences in enzyme mobility are not usually reflected in significant differences in activity.83 But the question remains whether there is selective advantage in some heterozygotes.84,85 That is, substantial heterozygosity would reduce species variability, enhancing conformity to the average and reducing the number of anomalous outliers. This is an old question that has engendered much controversy, and it is still unsettled, although studies of many species have demonstrated the reduced variability and enhanced developmental stability anticipated in the hypothesis.84–86 The alternative, that heterozygosity contributes to a “genetic load,” or debt to be paid in illness or death, while clearly not applicable to much of the common polymorphism, retains some validity in regard to heterozygosity for a few mutants associated with recessive disease.87,88 That is, although Kacser and Burns have demonstrated that heterozygotes for most recessive mutants show no deficiency in the flux through a pathway, there are a few such heterozygotes that express modified forms of the diseases observed in homozygotes or unexpected susceptibility to other disorders.88–91
Cultural homeostasis—If remote causes of disease are engendered by the requirement of species for genetic variation, the organization of the environment must be no less a source of such causes. Both the gene pool and the cultural traits that characterize societies evolve. Cultural traits are the ideas, beliefs, and values by which we live, and they motivate society, so there are analogies between cultural traits and genes on the one hand and between social organization and biological integration on the other.92–95 Culture is to society what genes are to physiology. This analogy between the evolution of biological qualities and that of cultural traits has been given serious attention by several workers in the past decade.92–95 Some of their thinking is summarized below.
Cultural traits may be perceived as units of transmission and as being capable of mutation. They may be transmitted vertically like the genes, but also horizontally, as between sibs or coevals, and obliquely, as between aunt and nephew. Other departures from the Mendelian-like pattern involve the transmission of ideas or behaviors from one central figure to many pupils or fans as well as from many to one, as in the pressure on individuals to accept the conventional wisdom. The cultural units may also vary in frequency as genes do and are subject to selection, proliferating to prevalence or diminishing to extinction. And their frequencies may be changed by migration—adoption of foreign styles, say—or by chance. But unlike biological qualities, which change at glacial rates, cultural change may be very rapid, a difference in tempo likely to create incongruence and risks for disease.
There is much interaction between the behaviors determined by culture and the homeostatic qualities promoted by the genes; the former may constitute risk factors for disease. And there is a complex relationship in the evolution, or coevolution, of both. Apposite examples are given by Durham.95 One is the persistence of lactase in the intestinal mucosa of some populations. This persistence varies directly with latitude and inversely with skin pigmentation and ultraviolet radiation exposure. The hypothesis is that in areas of low ultraviolet radiation precursors may not be converted to active vitamin D and there is the threat of rickets, which, untreated, may reduce reproductive fitness. But rickets can be averted by lactose, which promotes the intestinal absorption of Ca2+; hence the persistence of lactase. But where the intensity of ultraviolet radiation is high, rickets is not a threat, so lactase is unneeded. So cultural and biological traits have evolved together.
Durham’s second example is that of the well-known advantage of hemoglobin S in malarial areas. In some parts of Africa the frequency of the hemoglobin S gene is influenced by diet. The highest frequency of the gene is found where yams are cultivated for food. Yams require sunlight, which means cleared forests and warm sunlit pools wherein the mosquitos thrive. Populations that live in forests and eat other food have fewer mosquitoes, less malaria, and less hemoglobin S.
So there are intimate relationships between gene frequencies and the way people live. Gene-determined molecules work in cellular environments, themselves influenced by cultural traits that determine social environments. Sometimes these are at cross purposes and therefore of interest in medicine. The tobacco industry and human health are again an example. The tobacco enterprise has its own internal homeostasis that includes farmers, manufacturers, wholesalers, retailers, advertisers, marketers, and lobbyists. And it is a part of a larger homeostatic system consisting of community, state, and international trade. The livelihoods of all the workers named above, as well as taxes and the economy, all depend on smokers who imperil their health by smoking and do so in the face of growing social disapprobation. So in this case, physiological homeostasis, subject to injury by smoke, is opposed by a social homeostasis, and we are observing how difficult it is, when there is a clash of values, to change the prevalent custom, even when it is manifestly harmful. Social homeostatic systems are as stubborn and defensive as the physiological version.
Medical educators often stress the need for an infusion of “humanities” into the medical curriculum. The inheritance and evolution of cultural traits, their influence on social homeostasis, and how these interact with their biological counterparts would seem an appropriate answer to their concern.
3. The Lifetime: Developmental Homeostasis.
Nature imposes three tests on us all: (1) We must survive long enough to (2) attain fertility, which should lead to (3) successful reproduction. To fail any is to fail nature’s intent, but only the first compromises individual life irretrievably. Obviously, nature’s expectations can be fulfilled only by a fully developed person acting independently, so human development may be seen as a path to independence. Although it is useful to think of development in stages, it is in fact seamlessly continuous. Birth and puberty, however apparently discrete, are merely transitional phases that lead to fertility and reproduction. Aging is more insidious, more indistinct, and miscellaneous in its choice of organ systems. There is some question whether aging is a part of development at all, but if not, it is inalienably coupled with it.
The human path to independence lasts longer than that of even our nearest phylogenetic relatives. It is characterized by neoteny, an evolutionary program wherein juvenility is much prolonged and, in comparison with related organisms, retained in adult features.96 For example, not only is the time of growth and maturation of human beings longer than that of, say, the chimpanzee, but as adults we have features that resemble those of their young. This prolonged growth and maturation has promoted bipedal posture as well as enlargement and increased complexity and plasticity of the brain; the latter persists into adult life, even into old age. The advantages are obvious. Extended exposure of a gradually maturing nervous system to experiences of a variable environment, together with the mental resiliency to continue to learn at all ages, is a recipe for the adaptive agility that has enabled human beings to live in all latitudes and climates and so to exploit the earth’s resources to construct civilizations and to be aesthetically creative. These achievements are a consequence of the nature of learning. Learning is not a static process of piling block on block. Rather it is integrative; what we learn today has been determined by what we learned yesterday, and tomorrow’s learning will be influenced by today’s. So the store of learning accumulates and takes the learners into diverse fields of information and behavior. Learning is a metaphor for development, which is also a historical process.97 The effects of experience are accumulative and, within constraints set by diverse genotypes, lead to variable developmental paths—“canalization,” Waddington called it. So development is a formative process in which we move through life from the general to the specific, the unspecialized to the particular, the simple to the complex, defining and refining our distinctions and risking, in old age, becoming caricatures of ourselves.
Early life—It is the special triumph of molecular biology to have replaced the phenomenology of embryology with molecular explanations of mechanisms.98 Induction, fields, and gradients have given way to descriptions of the molecules engaged in fertilization and cell division and in cell determination and differentiation.98,99 Molecular mechanisms of intercellular communication, migration, adhesion, association, morphogenesis, and selective cell death have been discovered, and lineages of cells have been traced to their final roles in the coordinated assembly of specialized cells in the tissues and organs of the embryo.99 All these processes are under rather strict genic control, and while chance enters in, for example, in the choice of which cells migrate in which direction and which die off (called “developmental noise” by Waddington), there is little leeway for mistakes that would result in distortions. So embryogenesis is more nearly programmed than later fetal and postnatal development, in which the opportunities for variation due to experiences are progressively greater. And postnatal development shows it. For example, attainment of postnatal milestones is increasingly variable with age; nearly all babies sit unaided at or around 6 months, but walking and talking vary by several months, and even broader limits are observed around the onset of menses and puberty. Social development is more variable still. Very young infants respond only to stimuli imposed on them; later, babies learn to evoke responses and still later to create for themselves an environment in which to react.100 So by slowing the process of our development into whatever each of us is to become, neoteny provides opportunities for shaping that outcome, some of it by choice and some by the imposition of outside influences, adding thereby to individuality and to the diversity of the species. Neoteny allows us to escape the tyranny of our genes, or at least to modulate their rule.
Maternal effects on gestational life—There is another source of individuality in development. Although fetal life is shielded from the outside by the uterus and maternal circulation, intrauterine life is marked by maternal effects, some of which leave a lifelong stamp. Table 2-2 lists some of these.
Table 2-2: Maternal Effects on Development of Embryo, Fetus, and Newborn |Favorite Table|Download (.pdf) Table 2-2: Maternal Effects on Development of Embryo, Fetus, and Newborn
| || In Utero ||Postnatal |
|Maternal genes ||Mitochondria || |
| ||Imprinting || |
| ||Maternal–fetal incompatibility || |
|Maternal environment ||Cytoplasm of zygote ||Breast milk |
| ||Nutrition ||Cultural environment |
| ||Maternal age || |
| ||Birth rank || |
| ||Disease || |
| ||Teratogens || |
| ||Drugs || |
(1) Since all of the cytoplasm of the fertilized ovum is maternal, the events of the first cell divisions are governed not by the embryo but by the mother. (2) The transfer of maternal mitochondria may have an impact on developmental direction and disease. (3) Maternal-fetal incompatibilities make the embryo a potentially incompatible graft. Results of blood group antigen incompatibility are well known, and while polymorphism of the MHC antigens makes inconsonance the rule, paradoxically, maternal-paternal compatibility for these proteins reduces fertility.101 (4) Imprinting of parental genes is reflected in inheritance of disease according to the sex of one or the other parent, the imprinting having created a species of haploidy.102,103 If imprinting is a consequence of the necessity to defend the integrity of membranes and embryo, then such a biological necessity is a remote cause of disorders whose proximate causes are traceable to the haploid effect. (5) Variations in the intrauterine environment may have profound effects on development; maternal age and birth rank are examples with effects that may not be apparent until postnatal life. For example, onset of both insulin-dependent diabetes mellitus (IDDM) and schizophrenia is influenced by the age of the mother at the birth of the patient. Further, maternal disease may reduce fertility or have adverse fetal consequences. Phenylketonuria (PKU) and IDDM are examples. (6) Disorders due to external influences are too well known to review here. Their often profound effects prejudice the lives of their victims, and the differences in effects depending on the timing of the insult is further evidence of the historical nature of development. (7) Neotenous delay means that the newborn human infant, as compared with those of other primate species, is still a fetus, so the continuation of an intimate maternal-infant tie is paramount. The virtues of human milk in this regard have been reviewed above.
These aspects of fetal and early infant life have been recounted one at a time, but in fact every embryo and fetus is subject to some version of each of these variables and to many others besides, including its own genes. And the course such an individual takes must depend both on the qualities of all these influences and on how they interact with one another at which times in ontogeny. What happens in embryogenesis is the coalescence of parts into a unique individual. Each embryonic phase represents another choice, a further canalization, another irretrievable ordination to be supplanted by yet another, each reducing the range of choices for the next. The fertilized ovum represents the earliest and most versatile array of matrixes for homeostatic development and change. As embryogenesis and postnatal growth continue, each matrix is influenced by the last, and each influences the next so that no characteristic can escape the mark of individual history. So a salient feature of ontogeny is the continuous creation of new matrixes for the interaction of genes and experiences, an internal environment that is strongly influential in the shaping of phenotypes, that they cannot be analyzed apart from it.
Later life—The path to independence peters out somewhere in early adult life; the first two tests have been passed and the third is or is not an option. All systems are at peak performance and seemingly ready to go on forever. But, insidiously and variously, aging begins in the third decade, hardly recognized by the individual and for most purposes compensated by homeostasis, whose only sign of aging is a modest delay in return to baseline after perturbation.104 Early expressions are followed by increasing homeostatic disorder; some see aging as a kind of entropy. The uniqueness of individual aging is composed of individual patterns of affected homeostatic systems, indeed of individual cells, 104 so if it is dust from which we arise and dust to which we return, each aliquot thereof will have traveled a different route. Evidences of this uniqueness are derived from both cross-sectional and longitudinal studies.105 The former reveal increasing variances with age for physiological measurements, while the latter make clear that patterns and rates of aging in different organ systems vary independently both within and among individuals. That is, few individuals follow the patterns of change predicted by averages of many subjects, and the age of an individual is a poor predictor of performance in tests of physical function. Evidently, cleaving to the central tendency is a criterion for passing the three tests; afterward we are at liberty to pursue an individual path compounded of not entirely wholesome gene effects and possibly destructive experiences.
Why do we age and what causes it? More than 300 theories of aging have been reported.106 If anything is favored, it is that selection bears most heavily on life before reproduction ends and, as it were, loses interest after that.107,108 Williams has suggested that genes that are at a selective advantage in youth are less useful, even damaging, in later life.107 Kirkwood has explained this seeming paradox by proposing that nature wisely invests energy resources in youth and reproduction rather than in the very expensive enzymatic and metabolic equipment required to continue indefinitely to repair the random and cumulative ravages of life’s experiences.108 And it is this damage, whether due to oxidants, free radicals, radiation, or other causes, that leads to such metabolic aberrations as reduced protein synthesis and degradation, changes in posttranslational modification, and failure of DNA repair.109 So is aging itself a form of disease, or does it provide a favorable substrate for the agents of disease? This is a moot question and may always be.110,111 Homeostatic abnormality, due either to mutations in the germ line or to aging, creates a substrate for disease, which becomes overt, perhaps as a consequence of other mutants or of cumulative damage or, more likely, of both. For example, do the dominant diseases of later years emerge because of aging? The mutants date to conception but may express themselves only when some homeostatic inhibition is weakened by aging.112 This relationship of disease to aging was foreseen by Garrod, who proposed that susceptibility might become overt disease in individuals who encounter “the innumerable minimal insults that constitute the wear and tear of living.”
A logic of disease must be based on relationships between physiological, genetic, and developmental homeostasis and on prevalent cultures. Susceptibility, made up of accumulating effects of genes and experiences, becomes a disease whose form depends on the specificity of the genes and experiences as well as on the individuality of ontogeny. These elements, which can be seen to comprehend Waddington’s three time frames—the present, the lifetime, and the biological past—can be listed as in Table 2-3, in an arrangement that allows the physician to analyze each case according to the individual array of constituents, which is to say, to the individuality of the case, which itself should help in the choice of treatment. Is it not also plausible that so individual an analysis of a patient’s disease must lead the doctor to observe the specificity of the patient as a person?
Table 2-3: Elements Contributing to the Individuality of Phenotypes |Favorite Table|Download (.pdf) Table 2-3: Elements Contributing to the Individuality of Phenotypes
|Genes ||Developmental Matrix ||Environment (Experiences) |
|Major gene ||Age ||Geography |
|Modifiers ||Sex ||Time |
| ||Parental effects ||Climate |
| ||Ethnic group ||Education |
| ||Cognition ||Occupation |
| ||Behavioral attributes ||Diet |
| || ||Habits |
| || ||Socioeconomic status |
| || ||Disease |
Descent with Modification.
If evolution proceeds by descent with modification, the basis for the latter is the mutable gene. Genetic novelty arises in point mutations and the meiotic reconstitution of chromosomes. Variations in chromosomal conformation and number occur also and may be important in speciation.
If the inborn error is a central theme in medicine, the gene must be at the heart of the logic that gives the inborn error such a position. What is a gene? Although the word has not changed since Johanssen’s coinage, the ideas it embodies have evolved, always in consonance with the techniques available to pursue the question. So this section on descent with modification must begin with some definitions.
It is significant that the ideas of the gene originated in an evolutionary context. For example, Mendel did not set out to invent genetics; he was attempting to give mathematical rigor to experiments on hybrids that were expected to elucidate speciation, and the Ancestrian-Mendelist controversy of the first decade of this century turned on whether the genetic contribution to evolution was continuous or discontinuous.113,114 Genetic explanations of natural selection and evolution were the subject of the lucubrations of Fisher, Wright, and Haldane in the 1920s and 1930s, and Delbruck, the physicist, was drawn to study the replication of genes in bacteriophages because therein might lie the answer to the “riddle of life.”115 Indeed, it is generally accepted that nothing in biology makes sense apart from evolution. So the salience of mutable genes and chromosomes in the origins of disease locks the logic of disease into evolution. Accordingly, this section will be devoted to how this need for variation leads to inborn errors that, broadly defined, originate the pathogenesis of disease.
Initial definitions of the gene were statistical and abstract. It is not easy today to grasp the thinking of Bateson, whose heritable “unit character” was actually a phenotype, not a gene, or the ideas of Johanssen, who distinguished genotype from phenotype but could not accept that the gene represented more than a statistical concept. But the drosophilists and maize geneticists, whose experiments led to an operational definition of the gene, located heritable units in chromosomes and demonstrated linkage, recombination, allelism, and mutation, which, in turn, led them to propose that specific entities would be found in the chromosomes that would add a functional definition to the statistical and operational versions.116 Such a concept was clarified by Beadle and Tatum’s elaboration of Garrod’s inborn error to include a unitary relationship of specificity between genes and enzymes. And finally, the acceptance of DNA as the genetic material, together with the recognition of the information latent in its structure, led to a structural definition, with the collinearity of nucleotide sequence in DNA with amino acid sequence in protein at its core.117 Later discoveries of new features of the DNA—exons, introns, flanking sequences, promoters, enhancers, repeating sequences, and the like—led to an enlarged definition of the gene to include everything that is transcribed into RNA, which, in omitting that which goes untranscribed, omits a good deal of “genetic material.”118
The Uses of These Definitions.
In analyzing the genetic contribution to a disease phenotype we follow the historical path of the definition of the gene. We begin with a comparison of the recurrence rate of the disorder in families with the incidence in the general population. This is often followed by tests of twin concordance, path analysis, and calculation of heritability. These statistical processes ignore the physical reality of genes. Segregation analysis and tests of phenotypic heterogeneity or even of linkage with another defined phenotype provide operational definition—evidence that a physical gene exists. Next, the association of a biochemical attribute that segregates with the phenotype provides a functional dimension directing attention to some abnormal protein, which in turn channels the search for chromosomal location and a structural dissection of its originator gene. So the diagnostic process, defined in this instance as characterization of the phenotype by specific cause, has imitated the historical in its progression from a statistical definition by way of operational and functional descriptions to a structural reality. Diagnosis is an ontogenetic process that recapitulates the phylogeny of gene definition.
In the process of descent of the definition of the gene from an abstract heritability to DNA, a concept of gene action evolved in which information was shown to flow in the opposite direction, that is, from base-pair sequences to physiological levels of ever-increasing complexity, ending with some property of biologic, social, and evolutionary significance. Most of the inborn errors listed in MMBID have been characterized by following the path of phenotype to DNA, but today it is possible to omit the steps between the phenotype and the DNA and, beginning with the latter, to proceed to the discovery of an affected protein and so to elucidate pathogenesis. Examples are muscular dystrophy, chronic granulomatous disease, and cystic fibrosis. These successes suggest that as the fruits of the human genome initiative become available the latter pathway may come to prevail. But even then, history will not be denied. Phenotypic descriptions will always be necessary, and statistical and operational evidences of genetic influences will always be useful in detecting linkage with some marker of known genetic provenance. So this example of cultural evolution resembles its biologic counterpart in patching together something useful from that which history offers as well as that which is new. If there is a revolution in genetics it is in the methods; the ideas have evolved.
Mutation: Whatever Is, Is Variable.
If the gene is the central focus of evolution, it is because of its capacity for variation. The mutability of the DNA, its versatility in response to the agencies of mutation, and the studied imprecision of the instruments of its repair provide the wherewithal for nature the tinkerer to try out what she will.46 Perhaps most of the trials turn out to be errors that remain in the untranslated DNA or are lost by chance, but those that remain in translatable DNA are constrained by the consequences of previous tinkering, which, having turned out to work in particular conditions, is likely to be intolerant of further novelty. But some changes, representing neither better nor worse in the homeostasis of the moment, do attain such population frequencies as to constitute a reserve of variation against unforeseen need.
Mutation is both random and extravagant in its variety. These properties are reviewed in Chap. 3; it is enough to observe here that since nothing we know exempts any part of the DNA, or any part of any chromosome, no gene product is immune to genetic variation. So the question for medicine is how random mutation is reflected in overt or potential disease. Many of the answers to this question are given in Chap. 1. Here I wish to emphasize some implications of the observation that all DNA is susceptible to mutation.
The first such implication is the vulnerability of all elements in all homeostatic systems. A second is that the specificity of expression of disease and the severity thereof must depend on both the nature of the mutation involved and how critical it is to optimal homeostasis—for example, whether it is highly conserved or rate-limiting, and so on. These implications are under test. Figure 2-1 shows the rate at which single-gene inborn errors have been accumulated in the editions of MMBID. The curve will surely continue to rise, and at some point it will be useful to ask questions about which genes are proving to be most prone to cause what kinds of disease. In the meantime an implication for the medical educator is the necessity to bring about a change in mentality; the expectation of type causes variation to be squeezed and kneaded into existing classes, whereas the flexibility to see variation rather than type leads to the description of new disorders. A visceral grasp of the extent and variety of mutation promotes the latter frame of mind.
Inborn errors being described at a rapid rate.
A further implication of the randomness of mutation is the heritability of genetic heterogeneity. The subject is given appropriate prominence in Chap. 1; it is enough to say here that unilocal heterogeneity will certainly be found in all monogenic diseases, while multilocal heterogeneity is equally inevitable whenever more than one gene is required for expression of disease.
We are accustomed to thinking of heterogeneity as distinguishing families. At the clinical level heterogeneity is suspected when there is within-family likeness and between-family difference for any of many manifestations of a disorder. But when a disease is a consequence of the effects of two or more genes, heterogeneity becomes more and more difficult, at the clinical level, to sort out, partly because there is likely to be allelic variety for each locus and partly because the heterogeneity may be as much within families as between.
Nor is that all. There is heterogeneity in experiences of the environment as well. Smoking is again an example. There are smokers, former smokers, and nonsmokers, and the smokers and former smokers vary according to intensity of exposure to smoke—for example, the number and type (filter, nonfilter) of cigarettes smoked per day, how completely they are smoked, whether or not the smoke is inhaled, and the duration of the smoking experience, sometimes expressed as pack-years. And of course other experiences, with their own variations in kind, duration, and intensity, act as modifiers of the heterogeneity of the experience of smoking—to say nothing of providing variables in which genetic heterogeneity may be expressed.
In this section we have been dealing with the means to attain a balance between a teeming variety in the DNA and the necessity for homeostatic stability. Most of the alternatives proffered by the DNA cannot contribute to such an end, so the remote causes of individual disease reside in the freedom of the DNA to experiment in the interest of a resilient homeostasis for the species.
Sexual Reproduction and its Consequences.
Only at the level of gene-protein specificity is bimodality commonly observed in human biology. Most human phenotypes are readily accommodated in continuous distributions; evidently nature prefers continuity, which is to say, variability. But sex is a nearly complete dichotomy, and paradoxically, the complete differentiation of the sexes is an important reason for the continuity and variability of other phenotypes and as such must be included among remote causes of disease.
Diploidy enables eukaryotes to tolerate a greater range of mutational variation than is allowed to prokaryotes, and so to submit a broader compass of options to selective test. The consequences of diploidy expressed in dominant, recessive, and sex-linked phenotypes are discussed in detail in Chap. 1. The emphasis here is on the origins of human disease as a consequence of the possession of a double chromosome set.
A diploid zygote is compounded from sexually differentiated gametes. This differentiation has three kinds of implications for disease: (1) Sex chromosome differences are compensated by inactivation of one X chromosome in females. But even though such inactivation equates the number of active genes in each cell, populations of cells may reflect both allelic differences and variable inactivation, giving females sources of variation denied males. (2) Physiological differentiation leads to increasing divergence of the sexes in body conformation and structure, as well as in the function of homeostatic systems susceptible to hormonal influence, some of which work in ways not directly related to reproduction or sexually differentiated behavior. (3) Behavioral development is codified in sexual characteristics that differentiate the outcomes of the kinds, intensities, and durations of experiences to which the sexes are exposed.
So genetic, physiological, and social influences create aspects of the developmental matrix that condition and modify not only normal homeostasis but how, where, and when it breaks down under the influence of mutant genes and experiences. That is why in characterizing the clinical and physiological qualities of diseases at all ages, sex is always included as an important variable.
The Frequency of Disease.
What accounts for the frequency of a disease? Why are some rare while others retain frequencies of 1–10 percent of a population, occasionally even more, despite their harmful effects?
The answer is in no way surprising if asked in an evolutionary context. Given a species that is well accommodated to environments commonly encountered, rare diseases must be a consequence of (1) breaches of adaptation, due to genetic mistakes that invariably make their victims unfit, or (2) chance encounters with “foreign” environments infrequently met. The common diseases, for their part, are due mainly to encounters with environments that have been systematically changed from those to which the species had become reconciled or changed in response to social and cultural imperatives that are at odds with biological adaptation and that lead to both deficiency and intolerance. Some of these changes are poverty, increased population densities that have changed the population dynamics of microorganisms, and dietary habits for which human beings are not genetically prepared.31 These social changes represent provocations that single out the genetically vulnerable, who may be fairly frequent for two reasons—because the genes that contribute to their vulnerability are not intrinsically as noxious as the rare genes and because such genes may have, under other conditions, contributed something useful.119,120 Of course, assortative mating and inbreeding play some part, but perhaps less and less as the world becomes increasingly panmictic.
Why do people not have several diseases at once? The answer is some do, but where genetic differences are influential the frequencies are multiples of those of the genes involved. So there should be a frequency distribution in the population in which unlucky outliers on one side have several or many genetic susceptibilities, some or several of which are overtly expressed, while their more fortunate brethren at the other end have none at all. The former die prematurely, the latter live to a healthy old age. Since it is the nature of such distributions to decline from the mean, outliers are infrequent. The modal position is currently unknown. Do we all have some genes that put us at risk for something? Perhaps, but maybe only under conditions that require other genes, too, or complicated concatenations of genes and experiences acting only at particular phases of the lifetime. It is a promise of the human genome initiative to provide lists of such genes. But whether or not such lists are either desired or desirable remains to be seen.121
So, if disease is to be perceived as a consequence of incongruence between homeostasis and experiences of the environment, the forces that account for the qualities and frequencies of both the specific genes and the experiences that have interacted to produce a disease represent remote causes that have set the stage for proximate causes that lead to pathogenesis. The conditions that engender those forces have been recounted in the section on “Homeostasis” above. The point to be made here is that the frequency of most diseases is not randomly attained. Its variability is commensurate with the selective values of genes, values that are determined by the congruence of their effects with experiences of the environment, a congruence that may be nil from conception or, of more consequence in medicine, be overwhelmed by the consequences of choices offered in a society all of whose diverse motivations are not necessarily compatible with human biology.