Ihe p nczples fcvo ution by n iturals4eGtionre finally bcgznnzng to inform medicine

Evolution and theOrigins of Disease

houghtful contemplation of the human body elicits awe—in equal measure

with perplexity. The eye, for instance, has long been an object of wonder,

with the clear, living tissue of the cornea curving just the right amount, the

iris adjusting to brightness and the lens to distance, so that the optimal quantity of light

focuses exactly on the surface of the retina. Admiration of such apparent perfection soon

gives way, however, to consternation. Contrary to any sensible design, blood vessels and

nerves traverse the inside of the retina, creating a blind spot at their point of exit.

The body is a bundle of such jarring contradictions. For each exquisite heart valve, we

have a wisdom tooth. Strands of DNA direct the development of the 10 trillion cells that

make up a human adult but then permit his or her steady deterioration and eventual

death. our immune system can identify and destroy a million kinds of foreign matter, yet

many bacteria can still kill us. These contradictions make it appear as if the body was de

signed by a team of superb engineers with occasional interventions by Rube Goldberg.

in fact, such seeming incongruities make sense but only when we investigate the origins

of the body’s vulnerabilities while keeping in mind the wise words of distinguished geneti

cist Theodosius Dobzhansky: “Nothing in biology makes sense except in the light of evo

lution,” Evolutionary biology is, of course, the scientific foundation for all biology, and bi

ology is the foundation for all medicine. To a surprising degree, however, evolutionary bi

86 Copyright 199S Scientific American, Inc.

Cor straw tE o lede 01 1 hah cy dtoa dspotar aNow fo d t chad rat aTI q iday isf ° ofsuc probem

QUIt 9 PU PN9EDN

DefensesExample Syiip orsuch ougk orar no def t bufocart b sefero acor

Trade offs

E or pe rdesig of nyorsyter sa a ofu break

ab r v uld ore tlee e00 r rxsfu c oniig

ConflictsE ol sa bcr a o

rt tt vit t o assist t

ebe I ed utio

0 aOt

Novel env mnments

Exa pe I- naodyk or c ly

r 0 t c1’ 1 r ts

uc g tb

Copyright 1998 Scientific American, Inc. SCIENTIFIC AMERICAN November 1998 87

ology is just flOW being recognized as abasic medical science, The enterprise ofstudying medical problems in an evolutionary context has been termed Darwinian medicine. Most medical researchtries to explain the causes of an individual’s disease and seeks therapies to cureor relieve deleterious conditions, Theseefforts are traditionally based on consideration of proximate issues, thestraightforward study of the body’sanatomic and physiological mechanisms as they currently exist. In contrast, Darwinian medicine asks why thehod is designed in a way that makesus all vulnerable to problems like cancer, atherosclerosis, depression andchoking, thus offering a broader context in which to conduct research.

The evolutionary explanations for thebody’s flaws fall into surprisingly fewcategories. First, some discomfortingconditions, such as pain, feser, cough,vomiting and anxiety, are actually neither diseases nor design defects hut rather are evolved defenses. Second, conflictswith other organ isms—Escherichia colior crocodiles, for instance—are a factof life. Third, some circumstances, suchas the ready availability of dietary fats,are so recent that natural selection hasnot yet had a chance to deal with them.Fourth, the body may fall victim to tradeoffs between a trait’s benefits and itscosts; a textbook example is the sickle

cell gene, which also protects againstmalaria. Finally, the process of naturalselection is constrained in ways thatleave us with suhoptimal design features,as in the case of the mammalian eye.

r ps the most onI ise rncch snism Is c ‘I o

I who cannot clear I c , i t

n th jr lungs ate likP t t r

worn eua, 1 he capac it’ for acertainis beneficial. The tart rndaP who cannot feel fljifl tail esen to x

perlcnce discomfort trm staving P mesame ostnn for long pcriuds. Tire runnatural stillneSs 101 airs the ki1

to their mints. hLh then Jutenotate. Such pain-free mopP usuabs ekehr e,irlx aduithm ui rov ti5cua damageand iflFect0ns (mitth or pa’n P usual1interpreted as d ease or irauma Jut

atualP part of the sOlution zarher thaz’the problem, fhese Jetenuve capab’Pnes, ‘hapud be naturai selection, arekept in reset’ until needed.

Lec widels reeipn,-d as detenses meICS er, ii susea, ti;ii1itifl. dPi’ :hea ancOt fatigue, nec’ng and ‘ntlamnsFsen some phvsciatis ‘ernain or

P fexer’s utilite \o rnr a Sc zc ahohc rate, feser is a e I

rise in the set poek I y

tat. Fh high

A’ r ir ; sickness ha 1on ctnt rtundu S

r y ie nausea, I ed i F eriod of

‘ten

b r A ausca si Ie Is ntakc o t) - I rip

o I h imful subst s i I si led ndepend

e M e I tofet zo hvpotlwsiii s of pregnanes is an ad r

F the mother protects t

(ii xposurz tO P iPins, Pi oter t t

i idea hr examiung pregnancs outfles. Sun enough. omen with molt

nausea nere less lkcle to siziter iniccartapes, I his e denee supports the h -

rIOrilesis hut N hardly coneusis e. if P”otet is correct, further resarch shouldJPcoeee that pregnant ten’ales o in in’

sCLiC show angus it’ food preferences. Her theory jlso pmdic an inmaw in birth dcieeis atm np nftsnring

of women who hase little or no morn

io seknec and thzo eat a wider Sarieta of t”od durinw pregnanc;.

Anoniercemmon coflct’tiflfl. in’oet’.uhv ousis or’ir’natcd -w a de’ense .ncugruus stuar ion’ or promoting ‘s

ipe and as oidance. A 1992 stud r.ec A. Dugatkin ot the f,n’vtas ty

ilk evaluated the henefits er en He grouped them d

tar or hold, dupe ndingr se presence of u h

DEATH

I—I-fl

0U

MINORINCONVENIENCE

iiSEVERE

RESPONSEWITH NO

REALTHREAT

1 hedcstruor \Iatthew I KIt e F

tute in Alhuqr e , A’ F- e that even cold h

c hen infected, mm e to art res u ini their hodes are sescr I es ahos e their usual tempe’i an e.

I rtwntcd from mm ing to the warnpart of their cage, thee are at increasedrisk of death from the infeeti in, insimilar studs hx Es eli ii Satinoft or theL ins ersItv ot Deaw are, e[der[y rats,who can no loupe’- ach’es e the highlevers r;f their eouuger lab corn anofls.

NO also instinct’s clv sought hotter eneiron—SEVERE ments when eIsa lIen ‘eel hs uieeton.

RESPONSE p p i

WITH REAL ‘- Feta,ceu 1ev ci of non in me rnoocz IS

THREAT another misunderstood detense rnechanism. People suffering from chronic in—t”ction Otten has e dec reased levels orblood iron. Although such low iron sometm’ blamed fot the iine’ss. P cc

tualls is a po ltCni5 e responsei duringinfection, iron Is seeiucstercd ill the ita

v hich prey cuts ins teling haueria in r

p adequatx stipplies of thi

MUCH SUTFFRINC is unnecessary but incitahkbtaause of the smoke-detector nature of our detenses The cost of a false alarm—a strong rnanon mcti

as lo’niting in the -absence of a true threat to life—isr sorirs unpleasantness. But the cost of no un

r I e resence of a true threat, such as a ft od xmo d r death. A lack of defensive respon

n A for example. could kill ft e u

Copyright 1998 Scientific American, Inc.

bass. The timid hid, the ordinary simply swam away, and the bold maintained their ground and eyed the bass.Each guppy group was then left alonein a tank with a bass. After 60 hours,40 percent of the timid guppies hadsurvived, as had only 15 percent of theordinary fish. The entire complement ofbold guppies, on the other hand, woundup aiding the transmission of bass genesrather than their own.

Selection for genes promoting anxiousbehaviors implies that there should hepeople who experience too much anxiety, and indeed there are. There shouldalso be hypophobic individuals whohave insufficient anxiety, either becauseof genetic tendencies or antianxiedrugs. The exact nature and frequencyof such a syndrome is an open question,as few people come to psychiatrists complaining of insufficient apprehension.But if sought, the pathologically non-anxious may he found in emergencyrooms, jails and unemployment lines,

The utility of common and unpleasant conditions such as diarrhea, feverand anxiety is not intuitive. If naturalselection shapes the mechanisms thatregulate defensive responses, how canpeople get away with using drugs toblock these defenses without doing theirbodies obvious harm? Part of the answeris that we do, in fact, sometimes do ourselves a disservice by disrupting defenses,

Herbert L. DuPont of the Universityof Texas at Houston and Richard B.Hornick of Orlando Regional MedicalCenter studied the diarrhea caused byShigella infection and found that peoplewho took antidiarrhea drugs stayed sicklonger and were more likely to havecomplications than those who took aplacebo. In another example, EugeneD. Weinberg of Indiana University hasdocumented that wel 1-intentioned attempts to correct perceived iron deficiencies have led to increases in infectiousdisease, especially amebiasis, in parts ofAfrica. Although the iron in most oralsupplements is unlikely to make muchdifference in otherwise healthy peoplewith everyday infections, it can severelyharm those who are infected and ma!nourished. Such people cannot makeenough protein to bind the iron, leavingit free for use by infectious agents.

On the morning-sickness front, anantinausea drug was recently blamed forbirth defects. It appears that no consideration was given to the possibility thatthe drug itself might be harmless to thefetus but could still he associated with

birth defects, by interfering with themother’s defensive nausea.

Another obstacle to perceiving thebenefits of defenses arises from the observation that many individuals regularly experience seemingly worthless reactions of anxiety, pain, fever, diarrheaor nausea. The explanation requires ananalysis of the regulation of defensive responses in terms of signal-detection theory. A circulating toxin may come fromsomething in the stomach. An organism can expel it by vomiting, but onlyat a price. The cost of a false alarm—vomiting when no toxin is truly present—is only a few calories. But the penalty for a single missed authentic alarm—failure to vomit when confronted witha toxin—may be death.

Natural selection therefore tends toshape regulation mechanisms with hairtriggers, following what we call thesmoke-detector principle. A smokealarm that will reliably wake a sleepingfamily in the event of any fire will necessarily give a false alarm every time thetoast burns. The price of the humanbody’s numerous “smoke alarms” ismuch suffering that is completely normal but in most instances unnecessary.This principle also explains why block-

Intermediary disease vectors(Mosquitoes, health care workers’hands, unsanitary water supplies)

Unprotected and/or promiscuous sex

ing defenses is so often free of tragic consequences. Because most defensive reactions occur in response to insignificantthreats, interference is usually harmless;the vast majority of alarms that arestopped by removing the battery fromthe smoke alarm are false ones, so thisstrategy may seem reasonable. Until.that is, a real fire occurs.

Conflicts with Other Organisms

Natural selection is unable to provide us with perfect protection

against all pathogens, because they tendto evolve much faster than humans do.E. coil, for example, with its rapid ratesof reproduction, has as much opportunitv for mutation and selection in oneday as humanity gets in a millennium.And our defenses, whether natural orartificial, make for potent selection forces. Pathogens either quickly evolve acounterdefense or become extinct. Amherst College biologist Paul W. Ewaldhas suggested classifying phenomenaassociated with infection according towhether they benefit the host, the pathogen, both or neither. Consider the runny nose associated with a cold, Nasalmucous secretion could expel intruders,

Casual human-to-human transmission(Sneezing, coughing, touch)

Protected and/or monogamous sex

SCIENTIFIC AMERICAN November 1998 89

Evolution of Virulence

C hanges in virulence relate to the life history of the infectious agent and itsmode of transmission. As elucidated by Paul W. Ewald of Amherst College. in

fection requiring direct contact will ordinarily drive a pathogen toward a state of lowered virulence, because the host must remain mobile enough to interact with others.But intermediaries that spread disease-causing agents, even from totally incapacitatedhosts, can cause a change toward more virulence. Behavioral choices, such as safersex, can also alter the makeup of the pathogen.

Evolution and the Origins of DiseaseCopyright 1998 Scientific American, Inc.

speed the pathogen’s transmission tonew hosts or both [see “The Evolutionof Virulence,” by Paul W. Ewald; SciENTIFIC AMERICAN, pril 1993]. Answers could come from studies examining whether blocking nasal secretionsshortens or prolongs illness, hut fewsuch studies have been done.

Humanity won huge battles in thewar against pathogens with the development of antibiotics and vaccines. Ourvictories were so rapid and seeminglycomplete that in 1969 U.S. SurgeonGeneral William H. Stewart said that itwas “time to close the book on infectiousdisease,” But the enemy, and the power

of natural selection, had been underestimated, The sober reality is that pathogens apparently can adapt to everychemical researchers develop. (“The warhas been won,” one scientist more recently quipped. “By the other side.”)

Antibiotic resistance is a classic demonstration of natural selection. Bacteriathat happen to have genes that allowthem to prosper despite the presence ofan antibiotic reproduce faster than others, and so the genes that confer resistance spread quickly. As shown by Nobel laureate Joshua Lederberg of theRockefeller University, they can evenjump to different species of bacteria,borne on bits of infectious DNA. Today

some strains of tuberculosis in NewYork City are resistant to all three mainantibiotic treatments; patients with thosestrains have no better chance of surviving than did TB patients a century ago.Stephen S. Morse of Columbia University notes that the multidrug-resistantstrain that has spread throughout theEast Coast may have originated in ahomeless shelter across the street fromColumbia-Presbyterian Medical Center.Such a phenomenon would indeed bepredicted in an environment wherefierce selection pressure quickly weedsout less hardy strains. The survivingbacilli have been bred for resistance.

Many people, including some physicians and scientists, still believe the outdated theory that pathogens necessarilybecome benign after long associationwith hosts. Superficially, this makessense. An organism that kills rapidlymay never get to a new host, so naturalselection would seem to favor lowervirulence. Syphilis, for instance, was ahighly virulent disease when it first arrived in Europe, hut as the centuriespassed it became steadily more mild.The virulence of a pathogen is, however, a life history trait that can increaseas well as decrease, depending on whichoption is more advantageous to its genes.

For agents of disease that are spread

directly from person to person, low virulence tends to be beneficial, as it allows the host to remain active and incontact with other potential hosts. Butsome diseases, like malaria, are transmitted just as well—or better—by the incapacitated. For such pathogens, whichusually rely on intermediate vectors likemosquitoes, high virulence can give aselective advantage. This principle hasdirect implications for infection controlin hospitals, where health care workers’hands can be vectors that lead to selection for more virulent strains.

In the case of cholera, public watersupplies play the mosquitoes’ role, Whenwater for drinking and bathing is contaminated by waste from immobilizedpatients, selection tends to increase virulence, because more diarrhea enhances the spread of the organism even if individual hosts quickly die. But, as Ewaldhas shown, when sanitation improves,selection acts against classical Vibriocholerae bacteria in favor of the morebenign El Tor biotype. Under these conditions, a dead host is a dead end. But aless ill and more mobile host, able to infect many others over a much longertime, is an effective vehicle for a pathogen of lower virulence. In another example, better sanitation leads to displacement of the aggressive Shigellaflexneri by the more benign S. sonnei.

Such considerations may he relevantfor public policy. Evolutionary theorypredicts that clean needles and the encouragement of safe sex will do morethan save numerous individuals fromHIV infection. If humanity’s behavioritself slows HIV transmission rates,strains that do not soon kill their hostshave the long-term survival advantageover the more virulent viruses that thendie with their hosts, denied the opportunity to spread. Our collective choicescan change the very nature of HIV.

Conflicts with other organisms arenot limited to pathogens. In times past,humans were at great risk from predators looking for a meal. Except in a fewplaces, large carnivores now pose nothreat to humans. People are in moredanger today from smaller organisms’defenses, such as the venoms of spidersand snakes. Ironically, our fears of smallcreatures, in the form of phobias, probably cause more harm than an interactions with those organisms do. Far moredangerous than predators or poisonersare other members of our own species.We attack each other not to get meatbut to get mates, territory and other re

90 SCIENTIFIC AMERICAN November 1998 Evolution and the Origins ofDisease

New Environments Bring New Health Threats

Common Threats to Health from 20,000 B.C.to Modern Times

Accidents

Starvation

Predation

Infectious diseases

Common Threats to Health Today (in Technologically Advanced Cultures)

Heart attack, stroke and other complications of atherosclerosis

Cancer

Other chronic diseases associated with lifestyle and longevity

Noninsulin-dependent diabetes

Obesity

New infectious diseases

Copyright 1998 Scientific American, Inc.

sources, Violent conflicts between mdividuals are overwhelmingly betweenyoung men in competition and give riseto organizations to advance these aims.Armies, again usually composed ofyoung men, serve smiIar objectives, athuge cost.

Even the most intimate human relationships give rise to conflicts havingmedical implications. The reproductiveinterests of a mother and her infant, forinstance, may seem congruent at firstbut soon diverge. As noted by biologistRobert L. Trivers in a now classic 1974paper, when her child is a few years old,the mother’s genetic interests may be bestserved by becoming pregnant again,whereas her offspring benefits from continuing to nurse, Even in the womb thereis contention. From the mother’s vantagepoint, the optimal size of a fetus is a bitsmaller than that which would bestserve the fetus and the father. This discord, according to David Haig of Harvard University, gives rise to an armsrace between fetus and mother over herlevels of blood pressure and blood sugar, sometimes resulting in hypertensionand diabetes during pregnancy.

Coping with Novelty

aking rounds in any modern hosI pital provides sad testimony tothe prevalence of diseases humanity hasbrought on itself. Heart attacks, for example, result mainly from atherosclerosis, a problem that became widespreadonly in this century and that remainsrare among hunter-gatherers. Epidemiological research furnishes the information that should help us prevent heartattacks: limit fat intake, eat lots of vegetables, and exercise hard each day. Buthamburger chains proliferate, diet foodslanguish on the shelves, and exercisemachines serve as expensive clothinghangers throughout the land. The proportion of overweight Americans is onethird and rising. We all know what isgood for us, Why do so many of us continue to make unhealthy choices?

Our poor decisions about diet andexercise are made by brains shaped tocope with an environment substantiallydifferent from the one our species nowinhabits. On the African savanna, wherethe modern human design was fine-tuned, fat, salt and sugar were scarceand precious. Individuals who had atendency to consume large amounts offat when given the rare opportunity hada selective advantage. They were more

likely to survive famines that killedtheir thinner companions. And we, theirdescendants, still carry those urges forfoodstuffs that today are anything butscarce. These evolved desires—inflamedby advertisements from competing foodcorporations that themselves survive by’selling us more of whatever we want tobuy—easily defeat our intellect andwillpower. How ironic that humanityworked for centuries to create environments that are almost literally flowingwith milk and honey, only to see oursuccess responsible for much moderndisease and untimely death,

Increasingly, people also have easyaccess to many kinds of drugs, especially alcohol and tobacco, that are responsible for a huge proportion of disease,health care costs and premature death.Although individuals have always usedpsychoactive substances, widespreadproblems materialized only followinganother environmental novelty: theready availability of concentrated drugsand new, direct routes of administration, especially’ injection. Most of thesesubstances, including nicotine, cocaineand opium, are products of natural selection that evolved to protect plantsfrom insects, Because humans share acommon evolutionary heritage with insects, many of these substances also affect our nervous system.

This perspective suggests that it is notjust defective individuals or disorderedsocieties that are vulnerable to the dangers of psychoactive drugs; all of us aresusceptible because drugs and our biochemistry have a long history of interaction. Understanding the details ofthat interaction, which is the focus ofmuch current research from both aproximate and evolutionary perspective, may well lead to better treatmentsfor addiction,

The relatively recent and rapid increase in breast cancer n’iust be the result in large part of changing environments and ways of life, with only a fewcases resulting solely from genetic abnormalities, Boyd Eaton and his colleagues at Emory University reportedthat the rate of breast cancer in today’s“nonmodern” societies is only a tinyfraction of that in the U.S. They hypothesize that the amount of time betweenmenarche and first pregnancy is a crucial risk factor, as is the related issue oftotal lifetime number of menstrual cvdes. In hunter-gatherers, menarche occurs at about age 15 or later, followedwithin a few years by pregnancy and

two or three years of nursing, then byanother pregnancy soon after. Only between the end of nursing and the nextpregnancy will the woman menstruateand thus experience the high levels ofhormones that may adversely affectbreast cells.

In modern societies, in contrast, menarche occurs at age ‘1.2 or 13—probablyat least in part because of a fat intakesufficient to allow an extremely youngwoman to nourish a fetus—and the firstpregnancy may be decades later or never. A female hunter-gatherer may have atotal of 1.50 menstrual cycles, whereasthe average woman in modern societieshas 400 or more, Although few wouldsuggest that women should becomepregnant in their teens to prevent breastcancer later, early administration of aburst of hormones to simulate pregnancy’ may reduce the risk. Trials to testthis idea are now under way at the University of California at San Diego.

Trade-offs and Constraints

C ompromise is inherent in every’adaptation. Arm bones three times

their current thickness would almostnever break, hut Hoino sapiens wouldbe lumbering creatures on a never-ending quest for calcium. More sensitiveears might sometimes be useful, but wewould be distracted by the noise of airmolecules banging into our eardrums,

Such trade-offs also exist at the genetic level, If a mutation offers a net reproductive advantage, it will tend to increase in frequency in a populationeven if it causes vulnerability’ to disease.People with two copies of the sickle cellgene, for example, suffer terrible painand die young. People with two copiesof the “normal” gene are at high risk ofdeath from. malaria, But individuals withone of each are protected from both malaria and sickle cell disease, Where malaria is prevalent, such people are fitter,in the Darwinian sense, than membersof either other group. So even thoughthe sickle cell gene causes disease, it isselected for where malaria persists.Which is the “healthy” allele in this environment? The question has no answer.There is no one normal human genome—there are only’ genes.

Many other genes that cause diseasemust also have offered benefits, at leastin some environments, or they’ wouldnot be so common. Because cystic fibrosis (CF) kills one out of 2,500 Caucasians, the responsible genes would ap

Et;olution and the Origins of Disease SCIENTIFIC AMERICAN November 1998 91Copyright 1998 Scientific American, Inc.

pear to be at great risk of being eliminated from the gene pool. And yet theyendure. For years, researchers mused thatthe CF gene, like the sickle cell gene,probably conferred some advantage.Recently a study by Gerald B. Pier ofHarvard Medical School and his colleagues gave substance to this informedspeculation: having one copy of the CFgene appears to decrease the chances ofthe bearer acquiring a typhoid fever infection, which once had a 15 percentmortality

Aging may be the ultimate exampleof a genetic trade-off, In 1957 one of us(Williams) suggested that genes thatcause aging and eventual death couldnonetheless he selected for if they hadother effects that gave an advantage inyouth, when the force of selection isstronger. For instance, a hypotheticalgene that governs calcium metabolismso that bones heal quickly but that alsohappens to cause the steady depositionof calcium in arterial walls might wellbe selected for even though it kills someolder people. The influence of suchpleiotropic genes (those having multipleeffects) has been seen in fruit flies andflour beetles, but no specific example hasyet been found in humans. Gout, however, is of particular interest, because itarises when a potent antioxidant, uricacid, forms crystals that precipitate outof fluid in joints. Antioxidants haveantiaging effects, and plasma levels ofuric acid in different species of primatesare closely correlated with averageadult life span. Perhaps high levels ofuric acid benefit most humans by slowing tissue aging, while a few pay theprice with gout.

Other examples are more likely to

APPENDIX is most likely hereto stay. Evolutionary pressureselects against the smaller appendix (a ve b cans fi iv

ci mon a id s veIl ng ca C Its kansing doo upp ak

ing infctio s nor lid di eatcuing. I argcr appcr diccs atIns a nally sd cird t

contribute to more rapid agmg. For in

stance, strong immune defenses protectus from infection but also inflict continuous, low-level tissue damage. It is alsopossible, of course, that most genes thatcause aging have no benefit at any age—they simply never decreased reproductive fitness enough in the natural environment to be selected against. Nevertheless, over the next decade researchwill surely identify specific genes thataccelerate senescence, and researcherswill soon thereafter gain the means tointerfere with their actions or evenchange them, Before we tinker, howevei we should determine whether theseactions have benefits early in life,

&cause evolution can take place onlyin the direction of time’s arrow, an organism’s design is constrained by structures already in place. As noted, thevertebrate eye is arranged backward.The squid eye, in contrast, is free fromthis defect, with vessels and nerves running on the outside, penetrating wherenecessary and pinning down the retinaso it cannot detach. The human eye’s

flaw results from simple bad luck; hundreds of millions of years ago, the layerof cells that happened to become sensitive to light in our ancestors was positioned differently from the corresponding layer in ancestors of squids. Thetwo designs evohed along separatetracks, and there is no going back,

Such path dependence also explainswhy the simple act of swallowing canbe life-threatening. Our respiratory andfood passages intersect because in anearly lungfish ancestor the air openingfor breathing at the surface was understandably located at the top of the snoutand led into a common space shared by

the food passageway. Because naturalselection cannot start from scratch, humans are stuck with the possibility thatfood will clog the opening to our lungs.

The path of natural selection can evenlead to a potentially fatal cul-de-sac, asin the case of the appendix, that vestigeof a cavity that our ancestors employedin digestion. Because it no longer performs that function, and as it can killwhen infected, the expectation mightbe that natural selection would haveeliminated it. The reality is more complex. Appendicitis results when inflammation causes swelling, which compresses the artery supplying blood to the appendix. Blood flow protects againstbacterial growth, so any reduction aidsinfection, which creates more swelling.If the blood supply is cut off completely, bacteria have free rein annl the appendix bursts, A slender appendix is especially susceptible to this chain ofevents, so appendicitis may, paradoxically, apply the selective pressure thatmaintains a large appendix. Far fromarguing that everything in the body isperfect, an evolutionary analysis revealsthat we liv e with some very unfortunatelegacies and that some vulnerabilitiesmay even be actively maintained by theforce of natural selection.

Evolution of Darwinian Medicine

D espite the power of the Darwinianparadigm, evolutionary biology is

just now being recognized as a basicscience essential for medicine. Most diseases decrease fitness, so it would seemthat natural selection could explain onlyhealth, not disease. A Darwinian approach makes sense only when the object of explanation is changed from diseases to the traits that make us vulnerable to diseases, The assumption thatnatural selection maximizes health alsois incorrect—selection maximizes thereproductive success of genes. Thosegenes that make bodies having superiorreproductive success will become morecommon, even if they compromise theindividual’s health in the end,

Finally, history and misunderstandinghave presented obstacles to the acceptance of Darwinian medicine, An evolutionary approach to functional analysis can appear akin to naive teleologyor vitalism, errors banished only recently, and with great effort, from medicalthinkmg. And, of course, whenever evolution and medicine are mentioned together, the specter of eugenics arises.

92 Scicxiicic AMERICAN November 1998Copyright 1998 Scientific American, Inc.

Evolution and the Origins of Disease

Selected Principles of Darwinian MedicineA Darwinian approach to medical practice leads to a shift in perspective. The following principles

provide a foundation for considering health and disease in an evolutionary context:

DEFENSES and DEFECTS are two fundamentally differentmanifestations of disease

BLOCKING defenses has costs as well as benefits

Because natural selection shapes defense regulationaccording to the SMOKE-DETECTOR PRINCIPLE, muchdefensive expression and associated suffering areunnecessary in the individual instance

Modern epidemics are most likely to arise from themismatch between PHYSIOLOGICAL DESIGN of ourbodies and NOVEL ASPECTS of our environment

Our DESIRES, shaped in the ancestral environment to leadus to actions that tended to maximize reproductivesuccess, now often lead us to disease and early death

The body is a bundle of COMPROMISES

There is no such thing as ‘the NORMAL body

There is no such thing as the NORMAL human genome”

Some GENES that cause disease may also have benefits,and others are quirks that cause disease only when theyinteract with novel environmental factors

GENETIC SELF-INTEREST will drive an individual’s actions,even at the expense of the health and longevity of theindividual created by those genes

VIRULENCE is a trait of the pathogen that can increase aswell as decrease

SYMPTOMS of infection can benefit the pathogen, thehost, both or neither

Disease is INEVITABLE because of the way that organismsare shaped by evolution

Each disease needs a PROXIMATE EXPLANATION of whysome people get it and others don’t, as well as anEVOLUTIONARY EXPLANATION of why members ofthe species are vulnerable to it

Diseases are not products of natural selection, but mostof the VULNERABILITIES that lead to disease are shapedby the process of natural selection

Aging is better viewed as a TRADE-OFF than a disease

Specific clinical recommendations must be based onCLINICAL STUDIES; clinical interventions based only ontheory are not scientifically grounded and may cause harm

Discoveries made through a Darwinianview of how all human bodies are alikein their vulnerability to disease will offer great benefits for individuals, butsuch insights do not imply that we canor should make any attempt to improvethe species. If anything, this approachcautions that apparent genetic defectsma have unrecognized adaptive significance, that a single “normal” genoine is nonexistent and that notions of“normality” tend to he simplistic.

The systematic application of evolu

tionary biology to medicine is a newenterprise. Like biochemistry at the beginning of this century, Darwinian medicine very likely will need to develop inseveral incubators before it can proveits power and utility. If it must progressonly from the work of scholars withoutfunding to gather data to test their ideas,it will take decades for the field to mature. Departments of evolutionary biology in medical schools would accelerate the process, but for the most partthey do not yet exist. If funding agen

cies had review panels with evolutionary expertise, research would developfaster, but such panels remain to be created. We expect that they will.

The evolutionary viewpoint providesa deep connection between the states ofdisease and normal functioning and canintegrate disparate avenues of medicalresearch as well as suggest fresh andimportant areas of inquirn Its unlirv andpower will ultimately lead to recognition of evolutionary biology as a basicmedical science.

The Authors

RANDOLPH M. NESSE and GEORGE C. WILLIAMS are the authors of the 1994hook Why We Gel Sick: The New Science or Da,winian Medicine. Nesse received hismedical degree from the Umversitv of Michigan Medical School in 194. He is now professor of psychiatry at that institution and is director of the Evolution and Human Adap—tation Program at the university’s Institute for Social Research. Williams received his doctorate in 1955 from the University of California, Los Angeles, and quickly became one ofthe world’s foremost evolutionary theorists. A member of the National Academy of Sciences, he is professor emeritus of ecology and evolution at the State University of NewYork at Stony Brook and edits the Quarterly Review of Biology.

Further Reading

EVOLUTION OF INiFeT1ous DISEASE. P. W.Ewald, Oxford University Press, 1994.

DARWINIAN PSYCHIATRY. M. T. McGuire andA. Troisi. Harvard University Press, 1998.

EvoIurIoN IN HEALTH AND DISEASE. Edited by S. Stearns. Oxford University Press,1998.

EVOLUTIONARY MEDICINE. W R. Trevathanci al. Oxford University Press (in press).

SCIENTIFIC AMERICAN November 1998 93Evolution and the Origins of DiseaseCopyright 1998 Scientific American, Inc.