Is­rael Rosenfield and Edward Ziff

The New York Review of Books - - Contents - Is­rael Rosenfield and Edward Ziff

Epi­ge­net­ics: The Evo­lu­tion Revo­lu­tion

At the end of the eigh­teenth cen­tury, the French nat­u­ral­ist Jean-Bap­tiste La­marck noted that life on earth had evolved over long pe­ri­ods of time into a strik­ing va­ri­ety of or­gan­isms. He sought to ex­plain how they had be­come more and more com­plex. Liv­ing or­gan­isms not only evolved, La­marck ar­gued; they did so very slowly, “lit­tle by lit­tle and suc­ces­sively.” In La­mar­ck­ian the­ory, an­i­mals be­came more di­verse as each crea­ture strove to­ward its own “per­fec­tion,” hence the enor­mous va­ri­ety of liv­ing things on earth. Man is the most com­plex life form, there­fore the most per­fect, and is even now evolv­ing. In La­marck’s view, the evo­lu­tion of life de­pends on vari­a­tion and the ac­cu­mu­la­tion of small, grad­ual changes. These are also at the cen­ter of Dar­win’s the­ory of evo­lu­tion, yet Dar­win wrote that La­marck’s ideas were “ver­i­ta­ble rub­bish.” Dar­winian evo­lu­tion is driven by ge­netic vari­a­tion com­bined with nat­u­ral se­lec­tion—the process whereby some vari­a­tions give their bear­ers bet­ter re­pro­duc­tive suc­cess in a given en­vi­ron­ment than other or­gan­isms have.1 La­mar­ck­ian evo­lu­tion, on the other hand, de­pends on the in­her­i­tance of ac­quired char­ac­ter­is­tics. Gi­raffes, for ex­am­ple, got their long necks by stretch­ing to eat leaves from tall trees, and stretched necks were in­her­ited by their off­spring, though La­marck did not ex­plain how this might be pos­si­ble.

When the molec­u­lar struc­ture of DNA was dis­cov­ered in 1953, it be­came dogma in the teach­ing of bi­ol­ogy that DNA and its coded in­for­ma­tion could not be al­tered in any way by the en­vi­ron­ment or a per­son’s way of life. The en­vi­ron­ment, it was known, could stim­u­late the ex­pres­sion of a gene. Hav­ing a light shone in one’s eyes or suf­fer­ing pain, for in­stance, stim­u­lates the ac­tiv­ity of neu­rons and in do­ing so changes the ac­tiv­ity of genes those neu­rons con­tain, pro­duc­ing in­struc­tions for mak­ing pro­teins or other mol­e­cules that play a central part in our bod­ies.

The struc­ture of the DNA neigh­bor­ing the gene pro­vides a list of in­struc­tions— a gene pro­gram—that de­ter­mines un­der what cir­cum­stances the gene is ex­pressed. And it was held that these in­struc­tions could not be al­tered by the en­vi­ron­ment. Only mu­ta­tions, which are er­rors in­tro­duced at ran­dom, could change the in­struc­tions or the in­for­ma­tion en­coded in the gene it­self and drive evo­lu­tion through nat­u­ral se­lec­tion. Sci­en­tists dis­cred­ited any La­mar­ck­ian claims that the en­vi­ron­ment can make last­ing, per­haps her­i­ta­ble al­ter­ations in gene struc­ture or func­tion.

But new ideas closely re­lated to La­marck’s eigh­teenth-cen­tury views have be­come central to our un­der­stand­ing of ge­net­ics. In the past fif­teen years these ideas—which be­long to a de­vel­op­ing field of study called epi­ge­net­ics—have been dis­cussed in nu­mer­ous ar­ti­cles and sev­eral books, in­clud­ing Nessa Carey’s 2012 study The Epi­ge­netic Revo­lu­tion2 and The Deep­est Well, a re­cent

1See our es­say “Evolv­ing Evo­lu­tion” in these pages, May 11, 2006.

2The Epi­ge­netic Revo­lu­tion: How Mod­ern Bi­ol­ogy is Rewrit­ing Our work on child­hood trauma by the physi­cian Na­dine Burke Har­ris.3

The de­vel­op­ing lit­er­a­ture sur­round­ing epi­ge­net­ics has forced bi­ol­o­gists to con­sider the pos­si­bil­ity that gene ex­pres­sion could be in­flu­enced by some her­i­ta­ble en­vi­ron­men­tal fac­tors pre­vi­ously be­lieved to have had no ef­fect over it, like stress or de­pri­va­tion. “The DNA blue­print,” Carey writes,

isn’t a suf­fi­cient ex­pla­na­tion for all the some­times wonderful, some­times aw­ful, com­plex­ity of life. If the DNA se­quence was all that mat­tered, iden­ti­cal twins would al­ways be ab­so­lutely iden­ti­cal in ev­ery way. Ba­bies born to mal­nour­ished moth­ers would gain weight as eas­ily as other ba­bies who had a health­ier start in life.

That might seem a com­mon­sen­si­cal view. But it runs counter to decades of sci­en­tific thought about the in­de­pen­dence of the ge­netic pro­gram from en­vi­ron­men­tal in­flu­ence. What find­ings have made it pos­si­ble?


1975, two English bi­ol­o­gists, Robin Hol­l­i­day and John Pugh, and an Amer­i­can bi­ol­o­gist, Arthur Riggs, in­de­pen­dently sug­gested that methy­la­tion, a chem­i­cal mod­i­fi­ca­tion of DNA that is her­i­ta­ble and can be in­duced by en­vi­ron­men­tal in­flu­ences, had an im­por­tant part in con­trol­ling gene ex­pres­sion. How it did this was not un­der­stood, but the idea that through methy­la­tion the en­vi­ron­ment could, in

Un­der­stand­ing of Ge­net­ics, Dis­ease, and In­her­i­tance (Columbia Univer­sity Press, 2012).

3The Deep­est Well: Heal­ing the LongTerm Ef­fects of Child­hood Ad­ver­sity (Houghton Mif­flin Har­court, 2018). fact, al­ter not only gene ex­pres­sion but also the ge­netic pro­gram rapidly took root in the sci­en­tific com­mu­nity. As sci­en­tists came to bet­ter un­der­stand the func­tion of methy­la­tion in al­ter­ing gene ex­pres­sion, they re­al­ized that ex­treme en­vi­ron­men­tal stress—the re­sults of which had ear­lier seemed self­ex­plana­tory—could have ad­di­tional bi­o­log­i­cal ef­fects on the or­gan­isms that suf­fered it. Ex­per­i­ments with lab­o­ra­tory an­i­mals have now shown that these out­comes are based on the trans­mis­sion of ac­quired changes in ge­netic func­tion. Child­hood abuse, trauma, famine, and eth­nic prej­u­dice may, it turns out, have long-term con­se­quences for the func­tion­ing of our genes.

These ef­fects arise from a newly rec­og­nized ge­netic mech­a­nism called epi­ge­n­e­sis, which en­ables the en­vi­ron­ment to make long-last­ing changes in the way genes are ex­pressed. Epi­ge­n­e­sis does not change the in­for­ma­tion coded in the genes or a per­son’s ge­netic makeup—the genes them­selves are not af­fected—but in­stead al­ters the man­ner in which they are “read” by block­ing ac­cess to cer­tain genes and pre­vent­ing their ex­pres­sion. This mech­a­nism can be the hid­den cause of our feel­ings of de­pres­sion, anx­i­ety, or para­noia. What is per­haps most sur­pris­ing of all, this al­ter­ation could, in some cases, be passed on to fu­ture gen­er­a­tions who have never di­rectly ex­pe­ri­enced the stresses that caused their fore­bears’ de­pres­sion or ill health. Nu­mer­ous clin­i­cal stud­ies have shown that child­hood trauma—aris­ing from parental death or di­vorce, ne­glect, vi­o­lence, abuse, lack of nu­tri­tion or shel­ter, or other stress­ful cir­cum­stances—can give rise to a va­ri­ety of health prob­lems in adults: heart dis­ease, cancer, mood and di­etary dis­or­ders, al­co­hol and drug abuse, in­fer­til­ity, sui­ci­dal be­hav­ior, learn­ing deficits, and sleep dis­or­ders. Since the pub­li­ca­tion in 2003 of an in­flu­en­tial pa­per by Ru­dolf Jaenisch and Adrian Bird, we have started to un­der­stand the ge­netic mech­a­nisms that ex­plain why this is the case. The body and the brain nor­mally re­spond to dan­ger and fright­en­ing ex­pe­ri­ences by re­leas­ing a hor­mone—a glu­co­cor­ti­coid—that con­trols stress. This hor­mone pre­pares us for var­i­ous chal­lenges by ad­just­ing heart rate, en­ergy pro­duc­tion, and brain func­tion; it binds to a pro­tein called the glu­co­cor­ti­coid re­cep­tor in nerve cells of the brain.

Nor­mally, this bind­ing shuts off fur­ther glu­co­cor­ti­coid pro­duc­tion, so that when one no longer per­ceives a dan­ger, the stress re­sponse abates. How­ever, as Gus­tavo Turecki and Michael Meaney note in a 2016 pa­per sur­vey­ing more than a decade’s worth of find­ings about epi­ge­net­ics, the gene for the re­cep­tor is in­ac­tive in peo­ple who have ex­pe­ri­enced child­hood stress; as a re­sult, they pro­duce few re­cep­tors. With­out re­cep­tors to bind to, glu­co­cor­ti­coids can­not shut off their own pro­duc­tion, so the hor­mone keeps be­ing re­leased and the stress re­sponse con­tin­ues, even af­ter the threat has sub­sided. “The term for this is dis­rup­tion of feed­back in­hi­bi­tion,” Har­ris writes. It is as if “the body’s stress ther­mo­stat is bro­ken. In­stead of shut­ting off this sup­ply of ‘heat’ when a cer­tain point is reached, it just keeps on blast­ing cor­ti­sol through your sys­tem.”

It is now known that child­hood stress can de­ac­ti­vate the re­cep­tor gene by an epi­ge­netic mech­a­nism—namely, by cre­at­ing a phys­i­cal bar­rier to the in­for­ma­tion for which the gene codes. What cre­ates this bar­rier is DNA methy­la­tion, by which methyl groups known as methyl marks (com­posed of one car­bon and three hy­dro­gen atoms) are added to DNA. DNA methy­la­tion is long-last­ing and keeps chro­matin— the DNA-pro­tein com­plex that makes up the chro­mo­somes con­tain­ing the genes—in a highly folded struc­ture that blocks ac­cess to se­lect genes by the gene ex­pres­sion ma­chin­ery, ef­fec­tively shut­ting the genes down. The long-term con­se­quences are chronic in­flam­ma­tion, di­a­betes, heart dis­ease, obe­sity, schizophre­nia, and ma­jor de­pres­sive dis­or­der.

Such epi­ge­netic ef­fects have been demon­strated in ex­per­i­ments with lab­o­ra­tory an­i­mals. In a typ­i­cal ex­per­i­ment, rat or mouse pups are sub­jected to early-life stress, such as re­peated ma­ter­nal sep­a­ra­tion. Their be­hav­ior as adults is then ex­am­ined for ev­i­dence of de­pres­sion, and their genomes are an­a­lyzed for epi­ge­netic mod­i­fi­ca­tions. Like­wise, preg­nant rats or mice can be ex­posed to stress or nu­tri­tional de­pri­va­tion, and their off­spring ex­am­ined for be­hav­ioral and epi­ge­netic con­se­quences.

Ex­per­i­ments like these have shown that even an­i­mals not di­rectly ex­posed to trau­matic cir­cum­stances—those still in the womb when their par­ents were put un­der stress—can have blocked re­cep­tor genes. It is prob­a­bly the trans­mis­sion of glu­co­cor­ti­coids from mother to fe­tus via the pla­centa that al­ters the fe­tus in this way. In hu­mans, pre­na­tal stress af­fects each stage of the child’s mat­u­ra­tion: for the fe­tus, a greater risk of preterm de­liv­ery, de­creased birth

weight, and mis­car­riage; in in­fancy, prob­lems of tem­per­a­ment, at­ten­tion, and men­tal devel­op­ment; in child­hood, hy­per­ac­tiv­ity and emo­tional prob­lems; and in adult­hood, ill­nesses such as schizophre­nia and de­pres­sion.

What is the sig­nif­i­cance of these find­ings? Un­til the mid-1970s, no one sus­pected that the way in which the DNA was “read” could be al­tered by en­vi­ron­men­tal fac­tors, or that the ner­vous sys­tems of peo­ple who grew up in stress-free en­vi­ron­ments would de­velop dif­fer­ently from those of peo­ple who did not. One’s devel­op­ment, it was thought, was guided only by one’s ge­netic makeup. As a re­sult of epi­ge­n­e­sis, a child de­prived of nour­ish­ment may con­tinue to crave and con­sume large amounts of food as an adult, even when he or she is be­ing prop­erly nour­ished, lead­ing to obe­sity and di­a­betes. A child who loses a par­ent or is ne­glected or abused may have a ge­netic ba­sis for ex­pe­ri­enc­ing anx­i­ety and de­pres­sion and pos­si­bly schizophre­nia. For­merly, it had been widely be­lieved that Dar­winian evo­lu­tion­ary mech­a­nisms— vari­a­tion and nat­u­ral se­lec­tion—were the only means for in­tro­duc­ing such long-last­ing changes in brain func­tion, a process that took place over gen­er­a­tions. We now know that epi­ge­netic mech­a­nisms can do so as well, within the life­time of a sin­gle per­son.

It is by now well es­tab­lished that peo­ple who suf­fer trauma di­rectly dur­ing child­hood or who ex­pe­ri­ence their mother’s trauma in­di­rectly as a fe­tus may have epi­ge­net­i­cally based ill­nesses as adults. More con­tro­ver­sial is whether epi­ge­netic changes can be passed on from par­ent to child. Methyl marks are sta­ble when DNA is not repli­cat­ing, but when it repli­cates, the methyl marks must be in­tro­duced into the newly repli­cated DNA strands to be pre­served in the new cells. Re­searchers agree that this takes place when cells of the body di­vide, a process called mi­to­sis, but it is not yet fully es­tab­lished un­der which cir­cum­stances marks are pre­served when cell di­vi­sion yields sperm and egg—a process called meio­sis—or when mi­totic di­vi­sions of the fer­til­ized egg form the em­bryo. Trans­mis­sion at these two lat­ter steps would be nec­es­sary for epi­ge­netic changes to be trans­mit­ted in full across gen­er­a­tions.

The most re­veal­ing in­stances for stud­ies of in­ter­gen­er­a­tional trans­mis­sion have been nat­u­ral dis­as­ters, famines, and atroc­i­ties of war, dur­ing which large groups have un­der­gone trauma at the same time. These stud­ies have shown that when women are ex­posed to stress in the early stages of preg­nancy, they give birth to chil­dren whose stress-re­sponse sys­tems mal­func­tion. Among the most widely stud­ied of such trau­matic events is the Dutch Hunger Win­ter. In 1944 the Ger­mans pre­vented any food from en­ter­ing the parts of Hol­land that were still oc­cu­pied. The Dutch re­sorted to eat­ing tulip bulbs to over­come their stom­ach pains. Women who were preg­nant dur­ing this pe­riod, Carey notes, gave birth to a higher pro­por­tion of obese and schiz­o­phrenic chil­dren than one would nor­mally ex­pect. These chil­dren also ex­hib­ited epi­ge­netic changes not ob­served in sim­i­lar chil­dren, such as sib­lings, who had not ex­pe­ri­enced famine at the pre­na­tal stage.

Dur­ing the Great Chi­nese Famine (1958–1961), mil­lions of peo­ple died, and chil­dren born to young women who ex­pe­ri­enced the famine were more likely to be­come schiz­o­phrenic, to have im­paired cog­ni­tive func­tion, and to suf­fer from di­a­betes and hy­per­ten­sion as adults. Sim­i­lar stud­ies of the 1932– 1933 Ukrainian famine, in which many mil­lions died, re­vealed an el­e­vated risk of type II di­a­betes in peo­ple who were in the pre­na­tal stage of devel­op­ment at the time. Al­though pre­na­tal and ear­ly­child­hood stress both in­duce epi­ge­netic ef­fects and adult ill­nesses, it is not known if the mech­a­nism is the same in both cases.

Whether epi­ge­netic ef­fects of stress can be trans­mit­ted over gen­er­a­tions needs more re­search, both in hu­mans and in lab­o­ra­tory an­i­mals. But re­cent com­pre­hen­sive stud­ies by sev­eral groups us­ing ad­vanced ge­netic tech­niques have in­di­cated that epi­ge­netic mod­i­fi­ca­tions are not re­stricted to the glu­co­cor­ti­coid re­cep­tor gene. They are much more ex­ten­sive than had been re­al­ized, and their con­se­quences for our devel­op­ment, health, and be­hav­ior may also be great. It is as though na­ture em­ploys epi­ge­n­e­sis to make long-last­ing ad­just­ments to an in­di­vid­ual’s ge­netic pro­gram to suit his or her per­sonal cir­cum­stances, much as in La­marck’s no­tion of “striv­ing for per­fec­tion.” In this view, the ill health aris­ing from famine or other forms of chronic, ex­treme stress would con­sti­tute an epi­ge­netic mis­cal­cu­la­tion on the part of the ner­vous sys­tem. Be­cause the brain pre­pares us for adult ad­ver­sity that matches the level of stress we suf­fer in early life, psy­cho­log­i­cal dis­ease and ill health per­sist even when we move to an en­vi­ron­ment with a lower stress level.

Once we rec­og­nize that there is an epi­ge­netic ba­sis for dis­eases caused by famine, eco­nomic de­pri­va­tion, war-re­lated trauma, and other forms of stress, it might be pos­si­ble to treat some of them by re­vers­ing those epi­ge­netic changes. “When we un­der­stand that the source of so many of our so­ci­ety’s prob­lems is ex­po­sure to child­hood ad­ver­sity,” Har­ris writes,

the so­lu­tions are as sim­ple as re­duc­ing the dose of ad­ver­sity for kids and en­hanc­ing the abil­ity of care­givers to be buf­fers. From there, we keep work­ing our way up, trans­lat­ing that un­der­stand­ing into the cre­ation of things like more ef­fec­tive ed­u­ca­tional cur­ric­ula and the devel­op­ment of blood tests that iden­tify biomark­ers for toxic stress—things that will lead to a wide range of so­lu­tions and in­no­va­tions, re­duc­ing harm bit by bit, and then leap by leap.

Epi­ge­net­ics has also made clear that the stress caused by war, prej­u­dice, poverty, and other forms of child­hood ad­ver­sity may have con­se­quences both for the per­sons af­fected and for their fu­ture—un­born—chil­dren, not only for so­cial and eco­nomic rea­sons but also for bi­o­log­i­cal ones.

Chil­dren in Am­s­ter­dam dur­ing the Dutch Hunger Win­ter, 1944–1945

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.