Alzheimers Disease - Nature (07!14!11)

8/10/2019 Alzheimers Disease - Nature (07!14!11)

1/22

B Y J I M S C H N A B E L

In September 1984, a group of prominentresearchers from around the world met inScotland to discuss a disease that afflicted

sheep and goats.Scrapie, as they called it, was important

for more than agricultural reasons it wasalso the most easily studied example of anemerging class of diseases that destroyedthe brain. The illnesses jumped infectiouslyfrom animal to animal, yet yielded no trace ofa virus or other microorganism. One big cluewas that these diseases left behind insolubleclumps, or plaques, made from millions oftiny fibrils, each of which comprised hun-dreds or thousands of proteins. A striking

new hypothesis proposed that these fibrilsand their plaques marked the toxic passage ofinfectious proteinaceous particles, or prions.

On the first night of the conference, sev-eral researchers gathered for dinner. Amongthem were Colin Masters, a neuropathologistfrom the University of Western Australia, andKonrad Beyreuther, a protein-sequencingexpert from the University of Cologne inGermany. Masters began telling Beyreutherabout a human disease that featured plaqueslike those seen in scrapie and seemed to be verycommon. It was called Alzheimers disease.

Until then I had never heard of Alzheimers

disease, Beyreuther recalls.

It is easy to forget how recently Alzheimersdisease entered the public consciousness.For many decades after it first appeared in themedical literature, the term referred only toan obscure, early onset form of dementia.What we now know as common, late-onsetAlzheimers was then called senile dementia and it was so prevalent among the elderly thatit hardly seemed worth classifying as a disease(see A problem for our age, page S2).

THE MYSTERY PROTEIN

It is also easy to forget that at the dawn ofAlzheimers research, the disease was suspectedof being a prion disease we tend to think ofthe connection between prions and Alzheimersas being much more recent. In late 2010, for

example, a team led by Mathias Jucker at theUniversity of Tbingen, Germany, reportedthat they could, in essence, transmit Alzhei-mers-type brain pathology in a prion-likemanner by injecting Alzheimers brain matterinto the bodies of mice (Eisele, Y. S. et al. Sci-ence 330, 980982, 2010). Such findings havecontributed to a major rethink of the cause ofAlzheimers disease. But this rethink is partlya renaissance because, as the story of Mastersand Beyreuthers early interest in Alzheimersreminds us, the prion connection is not new.It was there at the beginning, says Masters.

As Masters knew in 1984, autopsies of

Alzheimers patients revealed brain plaques

resembling those seen in scrapie, often sur-rounded by dying neurons and their twistedaxons and dendrites. When doused with Congored, a standard pathology stain, and illumi-nated with polarized light, the Alzheimersplaques just like scrapie plaques displayedan apple-green shimmer, a prismatic sign of thehydrogen bonds that held their fibrils tightlytogether. Protein aggregates that had thispeculiar property were called amyloids.

Earlier that year George Glenner, an amy-loid researcher at the University of California,San Diego, reported isolating a small proteinfrom amyloid deposits in brain blood vessels inpeople with Alzheimers disease. Was the pro-tein embedded in Alzheimers brain plaques thesame as the one in Glenners vascular deposits?

Or was it more like the scrapie protein?Masters and Beyreuther, at their dinner in

Scotland, agreed to collaborate to find out, andtheir partnership probably did more than anyother to launch modern Alzheimers researchand its central idea: the amyloid hypothesis.

Masters had already painstakingly puri-fied a quantity of Alzheimers amyloid, ina process akin to bomb-grade uraniumenrichment. Whena sample arrived inGermany, Beyreutherand his colleagues brokeit down with formic

acid and sifted through

AM Y L O I D

Little proteins, big cluesAfter a quarter of a century, the amyloid hypothesis for Alzheimers disease isreconnecting to its roots in prion research.

NATURE.COM

research the roleof amyloid- inAlzheimers disease

go.nature.com/qpcyez

SIMONFRASER/JAMESKING-H

OLMES/SCIENCEPHOTOLIBRARY

At autopsy, the brains of Alzheimers patients (right) are filled with amyloid plaques, reminiscent of the plaques seen in the brains of animals with scrapie (left).

S 1 2 | N A T U R E | V O L 4 7 5 | 1 4 J U L Y 2 0 1 1

ALZHEIMERS DISEASE OUTLOOK


2/22

the debris to find the smallest stable protein.This turned out to be a tiny peptide of roughly40 amino acids, and Masters and Beyreuthercalled it A4. Sequencing A4 showed that it wasnot the scrapie protein, or indeed anything likeit, but was essentially the protein Glenner hadisolated from blood vessels.

Beyreuthers team quickly determined that

A4 is a fragment of a much larger neuronalprotein, amyloid precursor protein (APP).They found the gene that encodes APP onchromosome 21. This was a big clue, as peoplewith Downs syndrome, who have an extra copyof chromosome 21, were known to developAlzheimers-like brain plaques by 40 years ofage. The overproduction of APP and A4 wasnow revealed as the likely reason for the plaquesin Downs syndrome and probably in Alzhei-mers disease too.

TOO MUCH AGGREGATION

Other Alzheimers investigators readily

pursued the APP lead. But three other impor-tant clues from this initial burst of researchby Beyreuther and Masters would be almostentirely overlooked for most of the next decade.

The first was an observation by Beyreutherabout the forms of A4 in different solventmixes. He noted the presence of stable clusters,or oligomers, made of two, four or more copiesof A4. So strong was the peptides tendency toform these oligomers that in certain solutions,dimers made of two copies of A4 were moreprevalent than monomers.

The second clue was that full-length A4 isextremely prone to aggregate. After obtainingthe full A4 sequence, Beyreuther began to syn-thesize various lengths of it in his lab, includinga series that started at the 42nd (and terminal)amino acid of its longest variant and workedtowards the opposite end. When we cameclose to the end of the peptide and took it off theresin, we saw it getting aggregated, he remem-bers. I thought Mein Gott, its snowing! It wasaggregating so quickly. It was horrible.

The third clue came after Beyreuther andMasters raised the first antibodies to A4 andused them to detect amyloid deposits withunprecedented sensitivity in autopsied brains.The deposits were much more extensive thananyone had realized and were almost always

present in people older than 80 years of age. Inyounger brains the plaques tended to be sparserand more diffuse, but they were still detectablein about 20% of cognitively normal people whohad died in their fifties. The implication was thatAlzheimers disease is almost inevitable, withplaques beginning to form in the brain threedecades before symptoms develop. I thoughtthat was amazing, says Beyreuther.

THE AMYLOID HYPOTHESIS

By the end of the 1980s, Beyreuther and Mastershad largely completed their discovery work onA4. Other scientists, mostly from the United

States, took the lead on Alzheimers research,

and one of their first acts was to rename the A4protein amyloid-, where the referred to theclassic -sheet molecular structure of amyloids.They also put much less emphasis on theoriginal prion connection. Some of theseyoung guys who came after us didnt seem toknow what a prion was, says Masters.

Even so, they seemed to move swiftlytowards an understanding of how amyloid-causes Alzheimers disease. In the early andmid-1990s, in-vitro studies indicated thatamyloid- becomes toxic to neurons when itbegins to aggregate. Genetic studies of familieswith early onset Alzheimers disease detectedmutations within the gene that encodes APP,and analysis of one of these mutant APPgenes found that it causes a sevenfold over-production of amyloid- (see Finding risk fac-tors, page S20). Transgenic mice that overpro-duced human APP and amyloid- developedplaques resembling those seen in Alzheimersdisease, and their behaviour in standard testssuggested some cognitive deficits. The amyloidhypothesis seemed straightforward: when theamyloid- concentration in the brain becomestoo high, the protein aggregates into fibrils andplaques, and begins killing neurons.

It eventually became clear that the situationwas not quite that simple. Further genetic stud-ies showed that familial, early onset Alzheimersis usually caused not by the overproduction oftotal amyloid-, but by the relative overpro-duction of a less common variant of amyloid-known as A42, the full-length, 42-amino-acid variant whose extreme proneness toaggregation had so alarmed Beyreuther.

The A42 findings were still consistentwith the plaque hypothesis, particularly onceresearchers recognized in the mid-1990sthat the variant in most plaques is A42. Theproblem was that mouse models with an over-

dose of A42 like the first Alzheimers mouse

models that overexpressed APP lacked theheavy neuronal losses and cognitive decay asso-ciated with the human disease. These modelshave some cognitive decline, but its not as muchas a person with full-blown Alzheimers disease,by any stretch, says Harvard neurologist BruceYankner, a long-time Alzheimers researcher.

Some researchers suspected that mice, with

their small brains and short lives, cannot accu-rately model such a slow-burning, big-braindisease. But another possibility, which gainedcurrency in the late 1990s, is that amyloid-plaques are not the real drivers of dementia.Autopsy studies showed, for example, that theprogress of Alzheimers dementia does not cor-relate well with the development of plaques.As Beyreuther and Masters had initiallyobserved, the plaques become dense in the brainlong before any signs of cognitive decline.

Unfortunately, the major pharmaceuticalcompanies had already placed their bets onthe amyloid- plaque hypothesis, and numer-

ous drug-development programs would go onto fail in clinical trials. But in the meantime,a small group of researchers had begun todevelop a new hypothesis that encompassedAlzheimers and a variety of other amyloid-forming diseases.

OLIGOMERS REVISITED

The genetic evidence made it almost certain thatthe aggregation of amyloid- somehow leadsto Alzheimers disease. The fibrils in plaqueswere the most obvious type of aggregate, andtherefore the most obvious suspect. Onlyafter the plaque hypothesis began to fail didresearchers return to the other aggregates: theamyloid- oligomers first seen by Beyreutherand his colleagues in Cologne.

In the early and mid-1990s, Charles Glabeat the University of California, Irvine, andDennis Selkoe at Harvard University reportedfinding oligomers in experiments withamyloid-. They saw them as briefly existingintermediates on the way to disease-causingfibrils, rather than fully fledged drivers ofdisease. But in 1998, William Kleins lab atNorthwestern University in Evanston, Illi-nois, reported that oligomers could be thetrue culprits in Alzheimers disease. WhenKleins team added a chemical to a solution

of amyloid- to stop it forming fibrils, theamyloid- instead formed oligomers, whichthen began to kill nearby neurons. At leastsome of this toxicity seemed to be the resultof the oligomers weakening the synapses the junctions between neurons and impair-ing their ability to contribute to learning andmemory (see Two pathways of aggregation).Similar results soon followed from the Selkoeand Glabe labs, and in time mouse models alsodemonstrated oligomer toxicity.

In the 2000s, a new consensus began toemerge: that amyloid- fibrils are weakly toxicon their own, that they seem to provoke harmful

inflammation, and that they are prone, especially

Like prions (above), amyloid- might spread in an

infectious manner within tissues.

EYEO

FSCIENCE/SCIENCEPHOTOLIBRARY

1 4 J U L Y 2 0 1 1 | V O L 4 7 5 | N A T U R E | S 1 3



3/22

when their plaques become especially dense,to slough off soluble amyloid- that can thenreform into oligomers. But in this model,amyloid- oligomers are the more worrisomeneurotoxins. Indeed, the amyloid- fibrils areprobably protective to the extent that they trapaggregating amyloid- in a less harmful form.

Amyloid- oligomers are now thoughtto exert their harmful effects by bind-ing directly to the membranes of neurons,or to specific receptors the insulin andNMDA glutamate receptors are suspects needed for neuronal signalling. But ifamyloid- oligomers were merely toxic toneurons, they might never overwhelm theclearance mechanisms of the brain and causedisease. To do that, they seem to need anotherdeleterious property that is associated withprions: infectiousness.

PRIONS REVISITED

The idea that Alzheimers might be a priondisease was first suggested in 1984 by thefuture Nobel laureate Stanley Prusiner of theUniversity of California at San Francisco. Hisidea was widely dismissed after amyloid- wasfound to be different from scrapie protein. Butby the mid 2000s, it was clear that Prusiner hadbeen essentially correct. Both amyloid- andprion-disease proteins could fall into a statethat was both toxic and self-replicating.

Prusiner, who was also at the dinner inScotland with Masters and Beyreuther, wasapparently wrong about the replicationmechanism. He had initially proposed that an

infectious prion is a protein monomer with

a misfolded shape that can induce the samemisfolding in normal versions of the protein.

But as the chemist Peter Lansbury, then atMassachusetts Institute of Technology, showedin a series of in-vitroexperiments in themid-1990s, the key self-replicator in priondiseases and Alzheimers disease appears tobe an oligomer, not a monomer. Once formed,the oligomer becomes a template, or seed,that attracts new monomers, and aggregationaround that nucleus proceeds rapidly. Thisis one of those nonlinear phenomena inwhich small changes can have big effects,says Lansbury, now chief scientist at LinkMedicine, a biotechnology company inCambridge, Massachusetts.

One type of nucleus would serve as atemplate for new oligomers. Another wouldseed ever-lengthening fibrils. Lansburyshowed that this initial nucleation eventhappens faster with a particularly sticky

stretch of amino acids found on bothprion proteins and A42. Adding thisstretch from A42, or even adding full-length A42, can trigger the runawayaggregation of all the amyloid- in thevic ini ty. Bey re ut he rs sno w me ta ph or

was apt: a simi-lar nucleationphenomenon liesat the heart of iceand snow crys-tallization.

More recently,Jucker and oth-

ers have shown

that brain matter containing amyloid- fromAlzheimers patients can nucleate plaquesin mice. Amyloid- is less hardy than prionproteins and so is much less likely to jumpfrom one person to another, but it does seemto spread in an infection-like manner withintissues. I was away from amyloid- researchfor years, but Ive become interested againsince Jucker showed that the stuff is infec-tious, says Beyreuther, who is now atthe University of Heidelberg.

Similar infectious properties have beenobserved for aggregates of tau protein, whichappear in Alzheimers-affected cells late inthe disease, as well as for -synuclein proteinin Parkinsons disease. Researchers suspectthat numerous other amyloid-linked diseasesfeature the same toxic, oligomeric mechanismsand involve a slow spread of pathology start-ing in the regions of the brain most vulnerableto the disorder. We know, for example, that

people who present with Parkinsons motorsigns are almost always going to have Parkin-sons dementia 20 years later, says Lansbury.In contrast, Alzheimers disease affects mem-ory and cognition quite early on.

In principle, according to Beyreuther, therecould be protein structures in our food, airand water that get into the brain and promotedisease-causing spirals of protein aggregationlike the little bit of dust that seeds the icecrystals in the windows, he says. If thats true,then we are in trouble.

Jim Schnabel is a science writer based in

Miami, Florida.

Alzheimersdisease is almostinevitable, withplaques beginningto form in the threedecades before

symptoms develop.

S 1 4 | N A T U R E | V O L 4 7 5 | 1 4 J U L Y 2 0 1 1


TWO PATHWAYS OF AGGREGATION

Disaggregation

Plaques

Protofilaments

Amyloid-monomers

Fibrils

Fibrilseeds

Individual amyloid-peptides, which are produced normally by neurons, can assemble in at least two ways. One pathway leads

to insoluble, plaque-forming fibrils. The other pathway leads to soluble oligomers, which are small enough to enter synapses.These oligomers are suspected to be the main toxic species in Alzheimers disease.

Oligomers


4/221 4 J U L Y 2 0 1 1 | V O L 4 7 5 | N A T U R E | S 1 5


In 1907, Bavarian psychiatrist AloisAlzheimer published his description ofa delusional woman who had slowly lost

all cognitive function and died at 55 yearsof age. For decades thereafter, because of thepatients age, Alzheimers presenile dementiawas considered a rare disease of mid-life. In

the 1970s, neuropathologists realized thatsenile dementia was indistinguishablefrom the disease described by Alzheimer1.The clinical picture of progressive braindysfunction in association with the post-mortem brain pathology of extracellularamyloid deposits (plaques) and intraneu-ronal tangles was renamed Alzheimersdisease, and its definition was broadened toinclude the major cause of dementia that weknow today.

The accumulation of amyloid plaquesis the key distinguishing feature of Alzhei-mers disease, and research in the late 1980sidentified the major plaque protein as theamyloid- peptide. In the 1990s, scientistsconnected amyloid- pathology with genesencoding three proteins: the amyloid-precursor protein (APP), which is cleavedby -secretase to create amyloid-, and twoforms of presenilin (presenilin 1 and prese-nilin 2), which are involved in amyloid-generation2(see Finding risk factors, pageS20). Together, mutations in these genesaccount for about 3% of Alzheimers disease,and insertion of one of these mutationsinto the mouse genome created the firsttransgenic mice to form amyloid plaques.

But what about the other 97% of

Alzheimers disease? Parsimony woulddictate that amyloid- pathology might becentral to all forms of Alzheimers disease3.Less than five years after the plaque-formingmice were developed, the first amyloid--lowering drugs and vaccines were identified.However, initial human trials of the drugswere uninterpretable because they includedno methods for measuring how muchamyloid- is in the brain. That challenge wasovercome in 2004 using positron emissiontomography to identify amyloid plaques4.

And thats where the field stood for sixyears. Then, in 2010, researchers at the

University of Turku in Finland showed that

immunotherapy with an anti-amyloid-monoclonal antibody lowered plaqueburden by around 25%. Disappointingly,however, the treatment an infusion ofantibody every three weeks for 1.5 years brought no cognitive benefit for the patient5.

Why did this immunotherapy reduce brainplaques but fail to halt cognitive decline?Perhaps 1.5 years is not long enough orperhaps a 25% reduction in plaque burdenis insufficient. To allow for this possibil-ity, immunotherapy trials are continuingand results are expected in 2013. Another

possibility is that the monoclonal antibodyused might not recognize the most importantneurotoxic conformation(s) of amyloid-.Dozens of different monoclonal antibodies,as well as intravenous immunoglobulin, arenow in clinical trials, in the hope that one orseveral will recognize and neutralize the mostneurotoxic forms of amyloid-.

There is at least one more possible inter-pretation of the Turku study: that therapy tolower amyloid- levels will never succeed insymptomatic patients. Brain imaging datafrom presymptomatic individuals who carrypresenilin 1 mutations show that plaque

accumulation starts 1020 years before

clinical symptoms appear6. So if subjects entertrials at the first sign of cognitive impairment,they might already harbour substantialquantities of neurotoxic amyloid-.

The best hope for therapies aimed atamyloid- levels, therefore, is to doseprophylactically to stop it building up inthe first place. A diagnostic category of

presymptomatic Alzheimers disease wasrecently proposed for subjects with strongbiomarker-based evidence of disease butwho are cognitively intact7. Nevertheless, inthe absence of a perfect test for predictingwho will develop Alzheimers disease andwhen, prevention trials are highly dauntingwith regard to cohort size, trial duration andcost. The most obvious place to start is withcarriers of presenilin 1, presenilin 2 or APPmutations, where disease risk and timing ofonset are highly predictable. The DominantlyInherited Alzheimer Network has beenfounded to identify carriers of pathogenic

mutations worldwide and enrol them intoprevention trials with amyloid--loweringagents.

The amyloid- odyssey of the past 25 yearshas shown that conquering Alzheimersdisease is not a matter of removing amyloid-plaques from the brainpost hoc. But the role ofamyloid- must be resolved, and our quest foreffective interventions must be seen throughto a successful end. Alzheimers disease isalready a problem for the healthcare systemsof Western countries and is a growing threatto developing nations.

The best argument for sticking with strat-egies to lower amyloid- levels is that safe,effective compounds are within reach. Per-fecting the selection of subjects and the tim-ing of intervention could delay the onset ofAlzheimers disease substantially. A centuryof effort has brought us to a rational modelfor how Alzheimers disease might begin, andwe should not be discouraged by the prospectof another decade or two of work to settle theamyloid- issue and ultimately, we hope,defeat the illness. Prophylactic intervention tolower amyloid- levels is now the best hope.

Sam Gandy is neurology and psychiatryprofessor and associate director of the

Alzheimers Disease Research Center atMount Sinai School of Medicine and the JamesJ. Peters VA Medical Center in New York.e-mail: [email protected]

1. Terry, R. D. & Katzman, R.Ann. Neurol.14,497506 (1983).

2. Bertram, L., Lill, C. M. & Tanzi, R. E. Neuron68,270281 (2010).

3. Gandy, S.J. Clin. Invest.115, 11211129 (2005).4. Klunk, W. E. et al. Ann. Neurol.55, 306319 (2004).5. Rinne, J. O. et al. Lancet Neurol.9, 363372 (2010).6. Knight, W. D. et al. Brain134, 293300 (2011).7. Sperling, R. A.et al. Alzheimers Dement.7,

280292 (2011).

The author declares competing financial interests:

go.nature.com/hdiuds

PERSPECT IVE

Prevention is betterthan cureAttempts to reduce amyloid- in the brain have yet toshow clinical benefits. Starting treatment early is thebest hope, says Sam Gandy.

The process of Alzheimers can begin 20 years

before the mind shows signs of cognitive loss.


5/22

B Y S A R A H D E W E E R D T

Rhumba. Lindy hop. Cha-cha. Ballroomdancing may not be the first preventivetreatment for Alzheimers disease that

springs to mind, but it is an ideal prescriptionfor those concerned about their decliningmemory. In fact, says Perminder Sachdev, aneuropsychiatrist at the University of NewSouth Wales in Sydney, Australia, dancinghas a perfect blend of elements that help staveoff dementia. Theres cognitive activity,theres also physical activity, and theres socialinteraction as well.

A healthy Mediterranean-style diet is also

thought to be protective so that dance classcould be topped off with a big Greek salad anda glass of red wine.

Over the past decade, epidemiological stud-ies have shown that exercise, intellectual activ-ity, social relationships and a healthy diet alllead to a lower risk of dementia. Such findingshave to be interpreted with caution, however,because many researchers are sceptical about

the benefits, and becausewithdrawing from socialrelationships and otheractivities can be an earlysymptom of dementia,

not just a risk factor for it.

Even so, we have enough suggestive obser-vational data now from several studies toconclude that lifestyle factors are important inAlzheimers disease, much as they are in cardio-vascular disease, says Ronald Petersen, directorof the Alzheimers Disease Research Center atthe Mayo Clinic in Rochester, Minnesota.

FIT AND HEALTHY

The task now is to move from lifestyle factors tointerventions to find out how much exercise,what kind of intellectual activity and at whatstage each could influence the course of the dis-ease. We need to do more [clinical trials] wherewe actually intervene with cognitive activity,

training programmes and exercise, and with anappropriate control group, Petersen says.

Some of these trials are already under way.For example, in the Fitness for the Aging BrainStudy, researchers1 in Australia recruited170 people who were worried that theirmemory had deteriorated or who had mildcognitive impairment (MCI), a conditionthat is considered a precursor to Alzheimersdisease. The researchers assigned half ofthe participants to a six-month exerciseprogramme, either walking or doing otheraerobic exercise for 50 minutes, three timesa week. The other half, in the control group,

carried on with their usual level of activity.

After six months, those in the exercise groupslightly improved their scores on the cognitivesection of the Alzheimers Disease AssessmentScale (ADAS-Cog), a series of short memory,language and reasoning tests, whereas con-trol subjects declined at a rate consistent withnormal ageing. Whats more, the exercisehad lasting effects, leading to better scores12 months after the programme ended.

ADAS-Cog is commonly used in clini-cal trials of Alzheimers disease drugs, sothe researchers were able to compare theeffects of exercise with those of drugs calledacetylcholinesterase inhibitors, which reducethe breakdown of the neurotransmitter

acetylcholine. For people with MCI, regularexercise can help your brain more than takingthe medication that is currently available forAlzheimers disease, says one of the research-ers, Nicola Lautenschlager, who studies geriatricpsychiatry at the University of Melbourne.

PHYSICAL CHANGES

How does this connection between body andmind work? Studies in rodents have suggestedat least two different mechanisms2. First,exercise increases the activity of an enzymecalled neprilysin that metabolizes amyloid- the protein that makes up the characteristic

plaques of Alzheimers disease and might

ILLUSTRATION

BYGRACIALAM

S 1 6 | N A T U R E | V O L 4 7 5 | 1 4 J U L Y 2 0 1 1


P R E V E N T I O N

Activity is the best medicineCan exercise, social interaction and the Mediterranean diet really help to keep the cognitivedecline of Alzheimers disease at bay?

NATURE.COM

more research on thelinks between life-style and dementia

go.nature.com/ivcdp8


6/22

help clear it from the brain. Physical activityalso turns on the production of brain chemicalssuch as nerve growth factors, which promotethe formation of nerve cells and the connec-tions between them. This process is thought tomake the brain better able to cope despite thepathological changes of Alzheimers disease.

In the past few years, the development of

biomarkers(see Warning signs, page S5)that can indicate Alzheimers-related brainchanges in living people have allowedresearchers to explore more fully themechanisms of the mindbody connection.For example, one study this year of 120sedentary but healthy older adults withoutany memory problems assigned half theparticipants to a 3-days-a-week programmeof physical exercise3. After a year, researchersperformed magnetic resonance imaging(MRI) on several brain areas, including thehippocampus, the brain structure responsiblefor memory formation.

In older adults, the hippocampus typicallyshrinks by 12% each year, and this is whathappened in the control group. But in theexercise group, the volume of the hippocam-pus actually increased by 2%. Thats probablymillions of cells, says research team memberKirk Erickson, a psychologist at the Universityof Pittsburgh, Pennsylvania. With one year ofexercise, we are in essence rolling back theclock by one to two years.

BRAIN TRAINING

Another report4, also published this year,suggests that similar mechanisms are at workwhen people exercise their brains. Canadianresearchers used functional MRI to analysebrain activity in 15 people with MCI. After aone-week programme designed to teach theparticipants new memory strategies, there wasactivation in several additional brain regionsduring memory tests, suggesting that intactareas of the brain were able to take over fromdamaged areas. The participants also scoredbetter on the tests.

Many studies of cognitive stimulationand dementia make use of computer gamesdesigned to boost mental skills. Although suchbrain training interventions do not generallymake healthy people smarter, they produce

positive results in people with Alzheimersdisease and related conditions. One 2006 trialfunded by the US National Institutes of Healthshowed that brain training can counteractsome of the cognitive decline expected withageing5. In that study known as AdvancedCognitive Training for Independent and VitalElderly (ACTIVE) people over 65 years ofage who did a five- to six-week brain trainingprogramme focusing on memory, reason-ing or speed of processing skills were betterat these skills than control participants evenfive years later.

Computerized brain training programs

are popular among researchers because these

interventions are controllable and predict-able, especially compared with intellectualpursuits in the real world. But this doesntmean that people need to play computer gamesto stay mentally agile, says Sachdev. Instead, heargues, people are likely to benefit from anyintellectual pursuit that both requires effort(something where you challenge your brain)

and is enjoyable (so that you can sustain it).That could mean anything from taking up theclarinet to doing Sudoku puzzles.

FOOD FOR THOUGHT

Meanwhile, other lifestyle factors that canmodify the risk of Alzheimers disease arecontinuing to emerge through epidemiologi-cal research. These types of studies, involvingobservation of thousands of people and theirhabits, underpin our knowledge about theMediterranean diet, which includes a relativelyhigh consumption of fruits, vegetables, wholegrains and olive oil, relatively low consumption

of red meat and saturated fat, and a glass of redwine with dinner.Eating these foods has already been shown

to reduce the risk of cardiovascular disease,hypertension and diabetes. In the past fewyears, three independent epidemiologicalstudies conducted in New York6, Chicago7andBordeaux, France8, have shown that those whoeat mostly Greek peasant food also stay thesharpest mentally. There has been convergingevidence that adherence to such a diet is relatedto lower risk of cognitive decline or Alzheimersdisease, says Nikolaos Scarmeas, a neurologistat Columbia University in New York.

A team of Columbia University researchersincluding Scarmeas asked 1,880 New Yorkersdetailed questions about their eating habits,then studied them for an average of five-and-a-half years. They found that the people withthe most Mediterranean diet have up to a 40%lower risk of developing Alzheimers diseasethan those who eat less Mediterranean food6.Results like these are so promising that sev-eral groups around the world are planningrandomized trials of the Mediterranean diet asa way of preventing Alzheimers disease.

Evidence that social engagement helpsto prevent dementia also comes primarilyfrom observational studies. For example,

among more than 6,000 people aged 65 orolder in Chicago, those with the most exten-sive social networks and the highest levels ofsocial engagement have the lowest rates ofcognitive decline9.

It can be difficult to measure peoples levelof social engagement and it is even harderto design randomized trials to investigate it.Disentangling the effects of social engage-ment from those of other lifestyle elements isfar from straightforward. Still, social engage-ment is a form of intellectual engagement,argues Linda Teri, professor of psychosocialand community health at the University of

Washington in Seattle. Teri has designed

programmes to encourage physical activityand social connections in people with MCIand dementia. When we are with otherpeople, we are listening to the conversation,were tracking ideas, were forming our ownideas, she says. Were actually engaging inquite a bit of cognitive skills.

So people who exercise in groups may

benefit from both the social stimulation andthe physical activity. For example, considerErickson and colleagues research intoexercise and brain changes in healthy olderpeople3. Instead of aerobic exercise, the controlgroup met three times a week to do stretches.This did not increase the size of theirhippocampus, but it did improve their scoreson a simple computerized test of memory,similar to the improvements in the exercisegroup. Erickson suggests that this socialstimulation benefits other parts of the brainthat the study did not measure.

A LITTLE BIT BETTER

In some parts of the research community,the argument that lifestyle can help to delayAlzheimers disease is a tough sell. Last year,the US National Institutes of Health organizeda consensus panel on preventing Alzheimersdisease. It concluded that it is too soon to tellwhether lifestyle changes or any otherprevention strategy can affect the develop-ment or the course of Alzheimers disease.

Even those who are more bullish about theevidence say that lifestyle changes are likelyto have only a limited benefit. But becauseAlzheimers disease develops late in life, evensmall changes in risk or slight delays in thedevelopment of symptoms could greatlyreduce the burden of disease, as people wouldbe more likely to die from other causes beforebecoming mentally impaired.

As Erickson says: If we can at least preventsome of the normal age-related decline fromhappening, even if it doesnt eliminate therisk if it just reduces the risk of develop-ing Alzheimers disease or makes the qualityof life a little bit better I think weve gone along way.

Sarah DeWeerdtis a science writer based inSeattle, Washington.

1. Lautenschlager, N. T. et al. J. Am. Med. Assoc. 300,10271037 (2008).

2. Savica, R. & Petersen, R. C. Psychiatr. Clin. N. Am.34, 127145 (2011).

3. Erickson, K. I. et al. Proc. Natl Acad. Sci. USA108,30173022 (2011).

4. Belleville, S.et al.Brain134, 16231634 (2011).5. Willis, S. L. et al. J. Am. Med. Assoc. 296, 2805

2814 (2006).6. Scarmeas, N. et al. J. Am. Med. Assoc. 302,

627637 (2009).7. Tangney, C. C. et al. Am. J. Clin. Nutr.93, 601607

(2011).8. Fart, C. et al. J. Am. Med. Assoc.302, 638648

(2009).9. Barnes, L. L., Mendes de Leon, C. F., Wilson, R.

S., Bienias, J. L. & Evans, D. A. Neurology63,

23222326 (2004).

1 4 J U L Y 2 0 1 1 | V O L 4 7 4 | N A T U R E | S 1 7



7/22

B Y J I M S C H N A B E L

Using the formidable powers of theimmune system to attack one of thebodys own proteins seems like a risky

approach. But this is what nearly all vaccines,or immunotherapies, against Alzheimersdisease aim to do. Their target is amyloid-, atiny protein produced by neurons. Scientistsdo not know what function amyloid- evolvedto have in its ordinary, free-floating form.

But they do know that it is unusually proneto sticking to copies of itself, and that thisaggregation process seems to be the principaltrigger for Alzheimers disease.

The first vaccine against Alzheimersdisease Dublin-based Elan Pharmaceuti-cals AN-1792 was based on a particularlyaggregation-prone form of amyloid- knownas A42. In mice that had Alzheimers-likedeposits, or plaques, of amyloid- in theirbrains, it seemed enormously promising:it provoked a storm of anti-amyloid- anti-bodies that dissolved the plaques in older miceand stopped plaques from forming in youngerones. But in humans, AN-1792 was a disaster.Elan halted its first large clinical trial in 2002,after patients developed meningoencepha-litis, an inflammation of the brain and itsmembranes that was apparently caused byrogue immune cells1.

Most subsequent efforts have faredlittle better. Milder, second-generation activevaccines against amyloid- are still in clinicaltrials, but many researchers suspect thatthese will not be strong enough to provoke asufficient antibody response in elderlypatients with weak immune systems. Passive

vaccine infusions of lab-grown anti-amyloid-antibodies are meant to get round this

problem, but they havent performed well inclinical trials.

We in the field have had to look backand say, what did we do wrong? saysNorman Relkin, a neurologist at Weill CornellMedical College, part of Cornell Universityin New York.

But despite these disappointments, thereare hints of clinical success from a surpris-ing direction one that could lead to a

better understanding ofAlzheimers diseaseand to therapies andpreventives that really

work.

The vaccine that has raised some research-ers hopes is a mix of antibodies pooled fromdonated human blood. Known as intravenousimmunoglobulin (IVIg), it has long beenmarketed as a general booster for antibody-based immunity in people who lack it forgenetic reasons, and as a moderator of somerare autoimmune conditions.

The idea of using IVIg to treat Alzheimersdisease occurred to Relkin and his colleagueMarc Weksler after they found, in 2002, that

people with Alzheimers disease have lowerlevels of anti-amyloid- antibodies in theirblood than cognitively normal people of thesame age. They decided to set up a small,6-month study of IVIg in eight of Relkinspatients. The concept simply was to give backthese antibodies, since IVIg is derived fromthe plasma of young individuals who tend tohave higher levels, Relkin says.

The results were surprisingly good: sixpatients improved their cognitive scores,and a seventh stabilized. In a larger trial of24 patients, Relkin again found signs thatIVIg was working: the eight-person placebogroup worsened as expected, but nearly all the16 treated patients improved moderately onboth cognitive and quality-of-life measuresover the first 6 months (ref. 2). Their improve-ments were roughly equivalent to turningback the clock by 618 months. Whats more,they stayed at those levels for as long as thetreatment continued more than two yearsin some cases.

INJECTION OF REALISM

The results of small trials often fail to holdup in larger trials. But Relkins results haveinspired some optimism and some off-label prescribing of IVIg for Alzheimers

disease because the improved cognitiveand behavioural scores were dose dependentand have been backed up by changes inbiological markers, including lower levels ofamyloid- in cerebrospinal fluid and reducedbrain shrinkage. In fact, Relkin says, brainshrinkage is towards the normal range inindividuals who got the best dose, which is avery provocative finding.

The US National Institute on Aging, alongwith Baxter BioScience of Deerfield, Illinois,one of several producers of IVIg, is sponsor-ing a follow-up trial in 400 individuals withAlzheimers disease. The results could be ready

by the end of 2012. If the trial is successful,

it could lead to the first Alzheimers ther-apy approved by the US Food and DrugAdministration that modifies the disease,rather than just treats the symptoms.

But this would not be the end of the Alzhei-mers story, merely the end of the beginning.IVIg has several shortcomings. First, thosewho seemed to benefit from treatment hadonly modest gains. I have not seen anyonere-enrol in adult education classes, says Weksler.Second, there seems to be a limited window

of time when the therapy is effective. In thesmall trials carried out so far, the patients whostarted IVIg treatment later in the disease courseseemed more likely to keep worsening.

There are also problems of cost andavailability. IVIg is infused at high dosesevery two weeks in these studies, and patientsmight need them for the rest of their lives, at acost of thousands of US dollars per infusion.Worse still, the production capacity for bloodproducts from human donors is limited, anddemand for IVIg from Alzheimers patientsand their families would swiftly outstrip supply.We need next-generation products that areeasier to produce and are based on IVIgsmechanisms of action, says Relkin.

Unlike most other Alzheimers vaccines,IVIg has several plausible mechanisms.Although some of its antibodies may keepaggregates of amyloid- in check, othersmay counter brain inflammation and reduceaggregates of tau protein, which also contributeto dementia. Youre talking about a complexdisease that has many different pathologicalprocesses occurring either sequentially or inparallel, says Relkin. So IVIg in this respectis ideally suited.

By contrast, AN-1792 and other big pharmaAlzheimers vaccines have aimed squarely at

amyloid- in its natural, single-copy form,as well as in fibrils the long, insoluble,plaque-making aggregates that show up promi-nently in the brain and cerebral blood vessels ofAlzheimers patients. The lack of success withthese vaccines suggests that single-copy andfibril amyloid- might not be the best targetsin patients who already have dementia.

So far, for all these vaccines, there hasbeen only one published efficacy study: aphase II trial of bapineuzumab, Elans passiveanti-amyloid- antibody infusion. Thebeneficial effects of bapineuzumab seemedweak to non-existent and, even worse, at high

doses it caused brain swelling and associated

NATURE.COM

vaccines againstAlzheimers disease

go.nature.com/wattos

V AC C I N E S

Chasing the dreamAfter a decade of disappointments, hopes for a successful Alzheimers vaccine

that ameliorates symptoms and ultimately prevents the disease are rising again.

S 1 8 | N A T U R E | V O L 4 7 5 | 1 4 J U L Y 2 0 1 1



8/22

microbleeds in some patients with heavy vas-cular amyloid- deposits3. Autopsy and brainimaging studies of selected bapineuzumab andAN-1792 recipients suggest that these vaccinescan fail to slow the progress of dementia evenwhen they succeed in reducing plaques ofamyloid- in the brain4.

One reason for these disappointingresults may be that the vaccines address onlyamyloid- and do nothing to counteract braininflammation or tau aggregates. Anotherpossibility is that they are less effective atclearing the small, soluble clusters of amyloid-known as oligomers, which are now seen asfar more toxic than fibrils and which seem topromote the appearance of tau aggregates5(seeLittle proteins, big clues, page S12).

The short-term effects of IVIg could bedue to its ability to clear amyloid- oligomers,

Relkin says. Studies have suggested that youcan reverse signs of memory impairmentin mouse models within 24 hours of givinganti-oligomer antibodies, he says. Its won-derful that we have a potential therapeuticas well as something that is directing ustowards new avenues, new mechanisms, instudying the problem.

DREAM VACCINES

In the future, vaccines may also be used totreat people who have less advanced diseaseand so might get more benefit. Were allmoving towards the idea of treating patients

with very mild dementia or even before they

develop symptoms, says Dennis Selkoe, aneurologist at Harvard Medical School andlong-time Alzheimers researcher.

The ultimate dream is to be able to givepeople a vaccine when theyre still in their20s or 30s, to prevent the disease processfrom even starting, says Cynthia Lemere,a Harvard neurobiologist who tests activeanti-amyloid- vaccines in monkeys.

Lemere, Selkoe and others believe that untildementia sets in, amyloid- is the main driverof disease. Even the existing vaccine candi-dates might work well in this presymptomaticphase by keeping amyloid-, in all its forms,within manageable levels.

Other researchers favour a universal Alzhei-mers vaccine that leaves ordinary, single-copyamyloid- alone and instead targets structuresfound only on amyloid- aggregates, particu-

larly oligomers and incipient fibrils. Accordingto Relkin, the natural anti-amyloid- anti-bodies found in IVIg seem to target theseshapes, rather than single-copy amyloid-.

I see these as pathology-specific structures,so theyre ideal targets, says Charles Glabe,an Alzheimers vaccine researcher at theUniversity of California, Irvine. I think youdhave your best therapeutic effect this way, andthe fewest side effects.

To elicit antibodies against these targets,Glabe and others have vaccinated animalswith synthetic peptides that have the desiredshapes but contain non-human amino-acid

sequences, lowering the risk of autoimmune

reactions. These vaccines reduce brain pathol-ogy and improve memory-related behavioursin mouse models of Alzheimers disease, just asbroader anti-amyloid- vaccines do6. In prin-ciple, some of the aggregate-specific antibodiesevoked by these vaccines would bind to aggre-gates of other disease-linked proteins, such as-synuclein in Parkinsons disease or prionproteins in CreutzfeldtJakob disease (CJD),so the same approach could be used against allsuch diseases.

So far, none of these third-generationvaccines has had the corporate backing toreach clinical trials, but that could changequickly. If one of the existing vaccines showsa strong effectiveness profile in clinical trials,then I think interest will go way up, says Glabe.He would particularly welcome a success forIVIg, because it is widely believed to work on

the same principle as an oligomer vaccine.But investors tend to lump all immuno-therapies together, he says, so they rise andfall together even though they may have verydifferent targets.

Jim Schnabel is a science writer based in

Miami, Florida.

1. Gilman, S. et al. Neurology 64, 15531562 (2005).2. Hughes, R. A. C. et al. Clin. Exp. Immunol. 158

(Suppl 1), 3442 (2009).3. Salloway, S. et al. Neurology73, 20612070 (2009).4. Rinne, J. O.et al. Lancet Neurol.9, 363372 (2010).5. Jin, M. et al. Proc. Natl Acad. Sci. USA108,

58195824 (2011).

6. Goi, F. et al. PLoS One5, e13391 (2010).

1 4 J U L Y 2 0 1 1 | V O L 4 7 5 | N A T U R E | S 1 9



9/22

ALZHEIMERS DISEASE14 July 2011 / Vol 475 / Issue No. 7355

CONTENTS

EditorialHerb Brody,Michelle Grayson,Apoorva Mandavilli,Tony Scully,Nick Haines

Art & DesignAlisdair MacDonald,Barbara Izdebska

ProductionKarl Smart, EmiliaOrviss, LeonoraDawson-Bowling,Stephen Russell

SponsorshipGerard Preston, DavidBagshaw, Yvette Smith

MarketingElena Woodstock,Hannah Phipps

Project ManagersHelen Anthony,Claudia Deasy

Art DirectorKelly Buckheit Krause

Magazine EditorTim Appenzeller

Editor-in-Chief

Philip CampbellEditorial advisorMarie-Thrse Heemels

COVER ART: NIK SPENCER

The fog that envelops so many people as they age,severing them from their memories and thus fromtheir identity, used to be considered a normal part

of growing old along with sore joints, needing readingglasses and losing touch with popular music. However,to anyone who has seen a loved one slip behind the heavycurtains of what we now call Alzheimers disease, thedecline seems anything but natural. What kind of massivemalfunction in the brain can send an alert, robust, wittyperson into a state of persistent confusion?

The theory that plaques of amyloid- in the brain trigger thedisease has been called into question (page S12); Alzheimersdisease is a subtler foe. And without a handle on the diseasescause or genetic underpinnings (page S20), the developersof drugs (page S9) and vaccines (page S18) are working in afog of their own. Moreover, Alzheimers disease cannot bedefinitively diagnosed without an autopsy of the brain atwhich point the information is rather academic, at least forthat individual. So the search is intensifying for biomarkers clues that indicate reliably whether a person who is still aliveand healthy is destined for Alzheimers disease (pageS5).

The stakes are high. Alzheimers disease is a drain not only on

individuals and families, but also on societies, with the costs ofcare and lost productivity exceeding US$300 billion per year,which will only increase with rising incidence. More peoplethan ever are making it to old age, but dementia is the reward for6 out of every 100 individuals who get past 60 years (page S2).

We can take some encouragement from the findings thatthere may be non-medical steps that people can take to wardoff the disease (page S16) and that the prescribed activities,such as dancing and playing games, are pleasant enough intheir own right.

We thank Eli Lilly for the financial support that hasmade this Outlookpossible. As always, Natureretains sole

responsibility for all editorial content.Herb BrodySupplements Editor, Nature Outlook.

S2 DEMENTIA

A problem for our ageAn ageing global population meansdementia is a rapidly growing disease

S5 BIOMARKERS Warning signs

New ways to track cognitive decline, frombrain imaging to simple patient tests

S8 PERSPECTIVE In search of biomarkers

Neil S. Buckholtz

S9 DRUGS A tangled web of targets

Obscure pathology is keeping drugdevelopment in the dark

S2 AMYLOID Little proteins, big clues

Fibril clumps and tau tangles are theusual suspects, but are they guilty?

S5 PERSPECTIVE Prevention is better than cure Sam Gandy

S6 PREVENTION

Activity is the best medicine A healthy lifestyle might ward off thesymptoms of Alzheimers

S8 VACCINES Chasing the dream

Setbacks and successes in the pursuit ofdisease-modifying treatments

S20 GENETICS Finding risk factors

A diverse range of genes are implicatedin the disease, complicating the picture

1 4 J U L Y 2 0 1 1 | V O L 4 7 5 | N A T U R E | S 1

OUTLOOK

Nature Outlooksare sponsored supplements that aim to stimulateinterest and debate around a subject of interest to the sponsor, whilesatisfying the editorial values of Natureand our readers expecta-tions. The boundaries of sponsor involvement are clearly delineatedin the Nature Outlook Editorial guidelines available at http://www.nature.com/advertising/resources/pdf/outlook_guidelines.pdf

CITING THE OUTLOOK

Cite as a supplement to Nature, for example, Nature Vol XXX,No. XXXX Suppl, SxxSxx (2011). To cite previously publishedarticles from the collection, please use the original citation, whichcan be found at the start of each article.

VISIT THE OUTLOOK ONLINEThe Nature OutlookAlzheimers disease supplement can be found

at http://www.nature.com/nature/outlook/Alzheimers/

All featured articles will be freely available for 6 months.

SUBSCRIPTIONS AND CUSTOMER SERVICES

For UK/Europe (excluding Japan):Nature Publishing Group,Subscriptions, Brunel Road, Basingstoke, Hants, RG21 6XS, UK.Tel: +44 (0) 1256 329242. Subscriptions and customer services forAmericas including Canada, Latin America and the Caribbean:Nature Publishing Group, 75 Varick St, 9th floor, New York, NY10013-1917, USA. Tel: +1 866 363 7860 (US/Canada) or +1 212 7269223 (outside US/Canada). Japan/China/Korea:Nature PublishingGroup Asia-Pacific, Chiyoda Building 5-6th Floor, 2-37 IchigayaTamachi, Shinjuku-ku, Tokyo, 162-0843, Japan. Tel: +81 3 3267 8751.

CUSTOMER [email protected]

Copyright 2011 Nature Publishing Group

COLLECTION

S23 Amyloid- and tau a toxic pas de deux

in Alzheimers diseaseL. M. Ittner & J. Gtz

S29 The ethics of informed consent inAlzheimer disease researchScott Y. H. Kim

S34 Recollection of lost memoriesR. C . Malenka & R. Malinow

S35 Gamma-secretase activating proteinis a therapeutic target for AlzheimersdiseaseHe et al.

S39 Clinical trials of disease-modifyingtherapies for neurodegenerativediseases: the challenges and the future

Anthony E. Lang


10/22

B Y M I C H A E L E I S E N S T E I N

Headlines trumpet the discovery ofgenes associated with Alzheimersdisease so often that one might

think the genetic foundations of the diseasemust surely be mapped out in their entirety.Certainly for those who develop the earlyonset, or familial, form of the disease in latemiddle age, the lions share of the blame canbe attributed to three genes:APP,PSEN1and PSEN2. Each of these genes plays a rolein producing amyloid-, the accumulationof which is widely thought to trigger thedisorders characteristic neurodegeneration.

However, more than 90% of Alzheimers

cases are of the late-onset form, which

typically manifests in people older than65 years and seems to have a separate poolof genetic risk factors. Efforts to identifyfactors directly involved in the processing andaccumulation of amyloid- have yielded atleast a dozen candidate genes implicatedin this form of the disease, but their rolesare still unclear and their total contributioncannot account for the estimated 6080%hereditary risk of late-onset disease1.

One factor a common variant of the geneencoding apolipoprotein E (ApoE) hascome to dominate the Alzheimers landscape2.Just one copy of this variant, calledAPOE4,

increases disease risk fourfold; two copies

raise the risk about tenfold. If youre going totry to predict whos going to get Alzheimers,APOE is probably equivalent to the rest of thegenes combined, says Gerard Schellenberg,director of the US-based Alzheimers DiseaseGenetics Consortium.

Although APOEplays a leading role inthe Alzheimers story, it relies on a large

supporting cast. Discovery of these othergenetic players gained momentum with the riseof genome-wide association studies (GWAS).In this approach, researchers analyse millionsof single nucleotide polymorphisms (SNPs) variations scattered throughout the genome in tens of thousands of affected and healthyindividuals. By finding genomic changesthat correlate with disease, they can uncovercandidate genes or harmful mutations.

STATISTICAL POWER

Well over a dozen GWAS studies on Alzheimersdisease have been published, most of them

from large consortia in Europe and the UnitedStates. Studies of this sort are often criticizedfor finding false positive associations, whichcannot be replicated by other studies, and theearly Alzheimers studies were no exception.But later efforts analysed many more SNPs inthe genomes of large populations of people withlittle overall genetic variability between them,increasing the statistical power and allowingscientists to identify variants in more than tengenes associated with increased risk35.

At a 2009 meeting, for example, PhilippeAmouyel, chair of the EU Joint ProgrammingInitiative on Neurodegenerative Diseases,compared data with Cardiff University geneti-cist Julie Williams, a long-time colleague. Wehad found exactly the same genes, Amouyelrecalls. This was really important because itreinforces the fact that these genes were notjust appearing through statistical bias.

The results have been further bolstered byvalidation in independent study groups, aswell as by meta-analyses, which collectivelyexamine multiple studies and assess theirstatistical power. When people criticizeGWAS, the best answer is that when we do alarge, completely independent study, we getthe same result, says Schellenberg.

The candidate genes also make biological

sense, as most are involved with the inflam-matory damage and metabolic disruptionsthat scientists have long associated with thedisease ( see Genetic risk factors for Alzhei-mers disease). Its an assortment of genesthat seem to be associated with lipidmetabolism and immune response, saysRichard Mayeux, co-director of Colum-bia Universitys Taub Institute for Researchon Alzheimers Disease and the AgingBrain in New York. This was sort ofpredictable, but we didnt have the data tosupport it until now. Importantly, many ofthe genes also interact with the amyloid-

pathway, which is still widely seen as the

BSIP,

MENDIL/SPL

G E N E T I C S

Finding riskfactorsUncovering genes that are linked with Alzheimers diseasecan help researchers understand what causes the disease.But its not easy.

More than 90% of Alzheimers cases manifest in people over 65 years of age.

S 2 0 | N A T U R E | V O L 4 7 5 | 1 4 J U L Y 2 0 1 1



11/22

initiating trigger for the disease (seeLittle proteins, big clues, page S12).

But these newly discovered genes donot resolve any debates about the origin ofthe disease if anything, they potentiallyprovide support for many different modelsof Alzheimers pathogenesis. Those whohave been working on amyloid-independent

pathways will say that genetics is provingit, while those working on amyloid will say,See, its as weve said, says Christine VanBroeckhoven, a molecular geneticist affiliatedwith Belgiums University of Antwerp.

DELIVERY TRUCK

Several of the candidate genes tie intomultiple pathways, further complicatingthe picture. For example, clusterin (encodedby the CLU gene), which is one of the newrisk factors most strongly associated withAlzheimers disease, is thought to be involvedin both amyloid- aggregation and clear-

ance. It is also known as apolipoprotein-J,and is best known for helping ApoE facili-tate cholesterol trafficking in the centralnervous system. Another risk factor, com-plement receptor 1 (CR1), is an importantcomponent of the innate immune responseagainst infection, but is also linked to theclearance of circulating amyloid-. Butvar iants in genes such as CLUand CR1make relatively small contributions tothe overall risk, increasing it by roughly15%, so they have much less effect on therisk thanAPOE.

Exactly how ApoE might cause Alzheimersdisease is a matter of debate. As well as beingthe main transporter of cholesterol and otherlipids and lipid-soluble molecules into thecentral nervous system, it is also thought tohelp remove amyloid- from the brain,although the mechanism is not yet clear. Thereare three major variants of the gene for ApoE.The protein produced by the high-riskAPOE4variant is the least stable, significantly impair-ing the movement of cholesterol and amyloid-within the brain, whereasAPOE2encodesa protein that is more abundant and actuallyconfers protection against Alzheimers diseaserelative to the commonAPOE3allele.

ApoE also modulates the inflammatory

response to cellular damage in the brain,points out Thomas Montine, director ofneuropathology at the University ofWashington in Seattle. This reaction, mediatedby the bodys innate immune system, could be

triggered by amyloid--induced cell death,but it might also bea response to otherneurological trauma,such as stroke. Ineither case, a pro-longed inflammatoryresponse can result in

the gradual build-up

of toxic chemical by-products that furtheraccelerate the death of neurons. Similardamage is seen in other neurodegenera-tive conditions, such as Parkinsons disease.Almost all of the hypotheses are covered byAPOE4, says Amouyel.

Several researchers are convinced thatApoEs role in cholesterol transport is the key

to its importance in Alzheimers disease. Thebrain has 25% of the bodys cholesterol content,even though it only makes up 2% of the bodyweight, says Judes Poirier, a neurobiologistat McGill University in Montreal, Canada.The brains capacity to rewire itself, a propertyknown as plasticity, depends on the ability tobuild and stabilize new synaptic connections.This in turn requires cholesterol, and mice thatlack ApoE or express theAPOE4variant exhibitdramatic problems in the repair of synapticdamage. ApoE is your ultimate delivery truckwhen you need lipids to maintain or restoreneural plasticity, Poirier says.

MULTIPLE ROLES

This central role for ApoE is supportedby evidence that variants in several othercholesterol-linked genes also increase therisk of Alzheimers disease. One such gene isPICALM, which encodes a protein that assistsApoE in lipid traffic; another isABCA7, whichis also involved in cholesterol transport. Werenow talking about six or seven new, stronglyreplicated genetic factors, all associated withlipid homeostasis in the brain, says Poirier.

ApoE also seems to be a bridge betweenAlzheimers disease and other physiologi-cal disorders. The associations with cardio-

vascular disease and diabetes are strong youvery seldom find a study that doesnt show thisassociation, says Mayeux. The problem is, astroke alone or the presence of diabetes alonedoesnt cause the disease. But those who carryAPOE4and have diabetes are twice as likely asnon-diabetics with this variant to eventuallydevelop Alzheimers disease6.

Another piece of theAPOE4puzzle is itslink to a higher risk of heart attack and stroke.That alone should be telling us that maybe itsrole here is actually lipid metabolism insteadof some exotic amyloid--interacting scheme,says Schellenberg. Accordingly, there is some

evidence that taking statins, which lowercholesterol levels, may delay or prevent theonset of the cognitive decline associated withAlzheimers disease, although clinical trials ofstatin use have yielded inconclusive results.

The available data fail to tie these variousthreads together satisfactorily, but severalambitious projects that are underwaymight help. For example, four of the largestAlzheimers GWAS groups have joinedforces, forming a mega-consortium knownas the International Genomics of AlzheimersProject. The project will draw on data froma total of 40,000 people with Alzheimers

disease and unaffected controls, and will

attempt a mega-meta-analysis, delvingdeeper in search of previously overlooked riskfactors. Were working with more than 10million SNPs, says Amouyel. That is verydense coverage of the genomic map.

The project also aims to identify whichpathological features relate to specific genes.But differences in sample collection andstorage across different groups are likelyto complicate that goal. Van Broeckhovenpoints out that for many GWAS cohorts,researchers do not have access to a detailedmedical history or post-mortem tissuecollected using standardized autopsyprotocols. This led to a lot of valuable diseasedata being lost before the study even began.Knowing what we know today, we have

to say that we have missed lots of opportu-nities in our sampling procedures, saysVan Broeckhoven.

EXPLORING EXOMES

The GWAS studies are inherently limitedby the distribution of known SNPs withinthe genome, leaving gaps that might concealvariants affecting the risk of disease. Becauseof the challenges of deriving statisticallyrobust data for rare variants, these studiesalso typically ignore SNPs that are estimatedto occur in less than 5% of the population.

However, the falling costs and increasing

speed of DNA sequencing have made it easier

CAROL&MIKEWERNER,

VISUALSUNLIMITED/SPL

Genes regulating cholesterol are mutated in

Alzheimers; could statins be a treatment?

Were nowtalking aboutseven stronglyreplicatedgenetic factors,all associatedwith lipid

homeostasis.

1 4 J U L Y 2 0 1 1 | V O L 4 7 5 | N A T U R E | S 2 1



12/22

for scientists to comb through entire genomes,and Schellenberg and colleagues are planningto use this approach to fill in the blanks. Tosave time and money, his team plans to focusinitially on the exome the subset of thegenome that contains all the genes that areexpressed in the search for causal mutations.Id rather have 2,000 exomes sequenced than100 genomes, says Schellenberg, because ifyoure looking for something rare you need tohave a big sample.

Old-fashioned approaches to finding geneshavent died out either, and several research-ers are continuing to examine factors thatwere identified based on a hypothetical asso-

ciation with Alzheimers disease. For example,Mayeuxs group has identified several disease-associated SNPs within the SORL1gene,which encodes a protein that participates inthe cellular uptake of APP. There were a lot ofdoubters because it was a candidate gene, butit holds up in the latest GWAS, says Mayeux.The role of SORL1 is also supported byfunctional evidence: mice that producelower levels of its protein accumulate moreamyloid- in the brain7.

Montines group identified anothercandidate while searching for physiologicalindicators in the blood or cerebrospinal fluid

that might indicate the onset of Alzheimers

disease8. Brain-derived neurotrophic fac-tor is linked to several other neurologicalconditions, and levels of this protein provedto be a powerful predictor of Alzheimersdisease. However, there is no clear evidence ofa causative role for variations in this gene. Welooked and couldnt find an association, butwe also havent sequenced the whole gene yet,says Montine.

A LIFETIME OF DAMAGE

A final component of risk is likely to emergefrom the interface between genetic predispo-sition and physiological insults accumulatedover the course of a lifetime. In a disease

thats so strongly related to ageing, what we doand what weve been exposed to throughoutour lives are likely to figure very importantly,says Montine.

For example, diabetes and stroke canlead to the production of highly reactivecompounds known as free radicals, whichinduce toxic chemical modifications in fats,proteins and nucleic acids. This sort of oxida-tive stress seems to be a general feature in thebrains of people with Alzheimers disease, andcould damage or kill neurons. Its a normalcomponent of ageing, but theres even morefree-radical injury that occurs in people with

Alzheimers, says Montine. Mitochondria,

the energy centres of the cell, normallykeep oxidative stress in check, and severalstudies are underway to assess whethermitochondrial DNA also contains risk factorsfor Alzheimers disease.

Attempts to understand the environ-mental aspect face the same problems thatconfront the geneticists: it is time consumingand expensive to acquire data, analyse it andthen construct hypotheses that might provemeaningful for diagnosis, prognosis andtreatment. The genetics defines relevancebut not mechanism, says Montine, and nowits up to experimentalists to try to figure outhow things work.

Michael Eisensteinis a science writer basedin Philadelphia, Pennsylvania.

1. Gatz, M.et al.Arch. Gen. Psychiatry63, 168174(2006).

2. Strittmatter, W. J. et al. Proc. Natl Acad. Sci. USA90, 19771981 (1993).

3. Bertram, L., Lill, C. M. & Tanzi, R. E. Neuron 68,270281 (2010).

4. Hollingworth, P. et al. Nature Genet. 43, 429435(2011).

5. Naj, A. C. et al. Nature Genet.43, 436441 (2011).6. Peila, R. et al. Diabetes 51, 12561262 (2002).7. Andersen, O. M. et al. Proc. Natl Acad. Sci. USA

102, 1346113466 (2005).8. Zhang, J. et al. Am. J. Clin. Pathol.129,

526529 (2008).

Amyloidprecursorprotein

Amyloid- accumulation

Endocyticrecycling

Immune response

Lipidtransport

Cholesterolmetabolismand trafficking

Nucleus

Inflammation

APOE

GENETIC RISK FACTORS FOR ALZHEIMER'S DISEASE

Several genes implicated in Alzheimers pathogenesis are involved in multiplecellular pathways, which illustrates the complexity of the disease.

A

SORL1

PICALM

APP

PSEN2

PSEN1

Synapticmaintenance

APOE

PICALM

CLU

ABCA7

APOE

Clearance of amyloid-

at blood brain barrier

APOE

CLU

APOE

CR1


S 2 2 | N A T U R E | V O L 4 7 5 | 1 4 J U L Y 2 0 1 1


13/22

B Y A L I S O N A B B O T T

The world is getting richer. But wealthbrings its own burdens. Prosperouspeople live longer and old age carries a

high risk of dementia a condition that is sofar neither preventable nor curable.

In 2000, for example, 4.5% of the populationof the United States was older than 65 years,and there were 411,000 new cases of Alzhei-mers disease. Ten years on, those numbers hadrisen to 5.1% of the US population and 454,000cases, according to the Alzheimers Associationin the United States.

This same trend is happening across theworld. In fact, when Alzheimers disease isconflated with other dementias with similarclinical profiles, it covers an estimated 35.6million people around 0.5% of the globalpopulation. And these figures are about to getworse: the number of people with dementia

is set to double in the next 20 years, according

to the World Alzheimer Report 2010, a globalassessment of the economic impact of dementia.

Commissioned by Alzheimers Disease Inter-national (ADI) a federation of Alzheimersassociations around the world the reportgathered numbers on a range of Alzheimers-like dementia. Dozens of teams are working

to find ways to predict, prevent, diagnose andtreat the condition, but so far their efforts haveachieved only limited success. As a result, theeconomic costs of dementias will likely becrippling, the report says.

In 2010, the global economic impact ofdementias was US$604 billion. This f iguredwarfs the costs of cancer or heart disease.Based on demographics, the ADI reportforesees an 85% increase in cost by 2030, withdeveloping countries bearing an increasingshare of the economic burden.

We are seeing a linear increase in prevalencein rich countries, but an exponential increase

in low-income countries, says report co-author

Anders Wimo, an epidemiologist at theKarolinska Institute in Stockholm. The needfor solutions is urgent.

The ADI report used the best available datato determine the direct medical and socialcare costs, as well as the indirect costs, whichmostly relate to family care and reduced

productivity. Nearly 90% of the global costsin 2010, it says, are borne by rich countries about 70% in Western Europe and NorthAmerica and less than 1% by low-incomecountries, where there is greater relianceon unpaid home care (see Global costs ofdementia). There is a fiftyfold difference inthe cost of care per person between the richestcountries and the poorest.

AGEING IN ASIA

Just under half of people with dementia live inhigh-income countries, 39% live in middle-income countries, and only 14% live in low-

income countries, the report says. But these

D E M E N T I A

A problem for our ageAs the number of Alzheimers cases rises rapidly in an ageing global population,the need to understand this puzzling disease is growing.

S 2 | N A T U R E | V O L 4 7 5 | 1 4 J U L Y 2 0 1 1


JUSTINJIN/PANOS

Rising life expectancy in developing countries such as China will bring with it an increase in the number of people with dementias.


14/22

proportions are forecast to change dramaticallyin the coming decades, particularly in rapidlydeveloping countries such as China and India,for two important reasons.

The first reason is demographic. In com-piling the ADI report, Wimo and co-authorMartin Prince of the Institute of Psychiatry atKings College London reviewed the availableepidemiological studies. They found that theprevalence of dementias in people aged over60 is fairly uniform across the world between5% and 7%.

As living standards increase in countries suchas India and China, this will lead to increasedlife expectancy. Given that the biggest riskfactor for dementia is age, a longer-living globalpopulation means there will be more peoplewith dementia. The report predicts that thenumber of people with dementia will roughly

double every 20 years, to 65.7 million in 2030and 115.4 million in 2050 (see Estimatedgrowth of dementia). Most of this increase willbe in developing countries.

Second, as wages rise, demand for morecostly professional care will also increase at least, that is what happened in wealthiercountries where the Alzheimers epidemic hitearlier. China has particular reason to worry:its one-child policy took effect in 1978, mean-ing that parents who reach old age in the next20 years may not be able to rely on home care.

There are no comparable detailed globalanalyses for other chronic diseases. But

Dementia 2010, a report commissioned by

the UK Alzheimers Research Trust, estimatedthat the annual national cost of dementias was23 billion (US$38 billion), nearly twice thatof cancer (12 billion) and far more than thecosts of heart disease (8 billion) and stroke(5 billion) (see Comparing costs).

The allocation of public research funds tothese diseases does not reflect this hierarchy,however. In 2008, UK public spending oncancer research was 12 times higher than ondementia (see Comparing Investment). Inthe United States, the National Institutes ofHealth spends 13 times more on cancer thanon Alzheimers-like dementias. We cant fundall the good ideas we have in grant applications,says Neil Buckholtz, chief of dementia researchat the US National Institute on Aging (NIA) inBethesda, Maryland.

TACKLING THE DISEASE

As the scale of the threat looms large, somecountries are launching programmes to tackledementia on several fronts. For example, in2009, Germany opened the German Centre forNeurodegenerative Diseases (DZNE) in Bonnat a cost of 66 million (US$95 million) per year.Developing treatment and preventive strategieswill depend on clearly defining the disease andlearning more about its clinical manifestations,says DZNE director Pierluigi Nicotera.

But these researchers will be aiming at amaddeningly elusive target. Fundamentalquestions about the disease such as what

its main cause is, and even what pathologies

define it remain unanswered (see Com-mon types of dementia). The label Alzheimersdisease was not widely used to describedementia until 1976, when Robert Butler,the founding director of the NIA, coined theterm, partly to make it easier to attract researchfunds to study the condition. At the time,the syndrome wherein some elderly peoplebecame forgetful and child-like was knownas senile dementia. This was not viewed as adisease that might be prevented or cured, butas an intrinsic part of getting old.

Alzheimers disease is widely thought tobe driven by amyloid pathology, in whichpeptides of amyloid- are generated in thebrain and clump together into plaques. Theplaques release toxic fragments of amyloid-,which wreak havoc by a mechanism that is notyet completely understood (see Little proteins,

big clues, page S12).Another form of dementia with similar

symptoms is driven by vascular pathology.Leaking blood vessels deprive small areas ofthe brain of blood and oxygen, and thesemicrostrokes damage brain tissue andeventually result in cognitive defects.Scientists are still arguing about what propor-tions of dementias are driven by plaques andby vascular pathology. Post-mortem analysesof brains from people with dementia suggestthat there is no simple answer: Alzheimers-type pathology is more common, but it nearlyalways coexists with vascular pathology.

A 2011 investigation of more than 450 brains

1 4 J U L Y 2 0 1 1 | V O L 4 7 5 | N A T U R E | S 3


NORTHAMERICA

63.7 M

CENTRAL& SOUTHAMERICA

57 M

EUROPE

60.2 M

AFRICA

72 M

AUSTRALASIA& OCEANIA

5.3 M200 3.4 M

2030 7.7 M

2050 16.1 M

SOUTH &SOUTHEAST ASIA

75.8 M

200 7.0 M

2030 4.6 M

2050 29.2 M

200 0.3 M

2030 0.6 M

2050 0.9 M

ESTIMATED GROWTH OF DEMENTIAThe number of people with dementia will roughly double every 20 years, with the biggest increases in developing countries.

200 4.4 M

2030 7. M

2050 .0 M

200 0.0 M

2030 4.0 M

2050 8.7 M

200 .9 M

2030 3.9 M

2050 8.7 M

EAST ASIA

225.4 M

200 8.7 M

2030 17.9 M

2050 30.8 M

Populationaged >60years in 2010

People withdementia

Source:WorldAlzheimerReport2009,

AlzheimersDiseaseInternational


15/22

from the UK Cognitive Function and AgeingStudies identified vascular damage in four-fifths of brains from individuals with demen-tia, and found plaques in nearly all of them(Wharton, S. B. et al. J. Alzheimers Disease, inpress). Scientists suspect that vascular pathol-ogy usually accelerates the damage driven byamyloid pathology. But the same study foundthat three-quarters of brains from individualswithout dementia also had vascular pathology,and some of those from the oldest individualsshowed a significant burden of plaques.

AIMING AT AMYLOID

Amid this confusion, companies interestedin developing therapies have primarily beentargeting amyloid pathology, encouraged bythe fact that the heritable, early onset formof Alzheimers disease is mostly caused bymutations in the genes responsible for theproduction and metabolism of amyloid-.These familial cases account for fewer than5% of total dementia (see Finding risk factors,page S20), but the companies hope that a sig-nificant proportion of later-onset dementia willbe, one way or another, driven by amyloid-.

There is a level of wishful thinking in this, says

Nicotera. But so far none of the amyloid-basedstrategies has been successful (see A tangledweb of targets, page S9). Yet drug develop-ers have not given up on the concept. Morereliable biomarkers of Alzheimers disease arebeing developed (see Warning signs, page S5),potentially making it possible to carry out trialson patients before symptoms, and irreversibledamage, set in.

Some scientists are also wondering whetherit might be valuable to target vascular pathol-ogy as well. In fact, drugs such as statins,which lower cholesterol levels in the blood,and drugs to reduce blood pressure are nowroutinely given long term to patients at highrisk of heart attack or stroke. If vascularpathology drives a significant proportion ofdementias, those who have benefited fromthe long-term cardiovascular treatment intro-duced in the past two or three decades mightbe protected from dementias as well.

Few epidemiological surveys have so farbacked this up, but the authors of the mostrigorous survey to date, the Rotterdam Study,announced at the Alzheimers Disease Inter-national conference in Toronto in March

2011 that they have observed a slowing in

the number of people being diagnosed withdementia.

Launched in 1990, the Rotterdam Studyis considered to be a model for epidemiol-ogy trials. Intended to pinpoint the factorsthat contribute to various diseases, includingdementia, in the elderly, it has recruited nearly15,000 middle-aged individuals from a localpopulation in three cohorts in 1990, 2000and 2006 and is following their progress.Preliminary results have shown a smalldecrease in the age-specific incidence ofdementias, and fewer plaques and less vasculardamage among undiagnosed individuals, saysepidemiologist Monique Breteler, head of theneurological and imaging part of the survey.

If dementias were ever to come under con-trol, other medical problems of the elderlywould become more prominent, notes RudiWestendorp, who studies healthy ageing atLeiden University Medical Centre in theNetherlands. Because people with dementiaare either less aware of pain or are unable toexpress their distress, painful illnesses likeherpes zoster [shingles] are probably beingmasked by dementia, he says. Sight and

hearing fail distressingly when we get old we need to invest more heavily in researchaimed at circumventing this, like developingneural implants to bypass damaged retinas.

Westendorp is an optimist who believesthat solutions will be found to these problems,including the dementias, in the foreseeablefuture if countries invest in research now. Mostof the problems that come with old age, he says,will have a medical solution so living to agrand old age need not carry such a social andeconomic burden.

Alison Abbott is Natures senior European

correspondent.

COMMON TYPES OF DEMENTIAThere is a great deal of overlap between the symptoms of various dementias.

Dementia type Symptoms Neuropathology Proportion ofdementia cases

Alzheimer'sdisease

Impaired memory, depression,poor judgement and confusion

Amyloid plaques andneurofibrillary tangles

5080%

Vascular dementia Similar to Alzheimers disease,but memory less affected

Decreased blood flow to the brainowing to a series of small strokes

2030%

Frontotemporaldementia

Changes in personality andmood, and difficulties withlanguage

Damage limited to frontal andtemporal lobes

510%

Dementia with

Lewy bodies

Similar to Alzheimers disease,

also hallucinations, tremors

Cortical Lewy bodies (of the protein

a-synuclein) inside neurons


16/22

B Y R U T H W I L L I A M S

Each week for the past six years, boxafter delivery box of blood, cerebro-spinal fluid (CSF) and urine samples

have arrived at a lab in the University ofPennsylvania in Philadelphia. Researchersthere have documented, divided, labelledand stored the samples, row after row, inseven enormous freezers.

Some 14,000 samples have been divided

into 160,000 tubes and each one isprecious. We have back-up freezers andalarm systems in case of electrical failures,says John Trojanowski, director of theAlzheimers Disease Center at the Universityof Pennsylvania.

Theres good reason for these precautions.The specimens, accompanied by detailedmedical histories, cognitive and clinicalmeasures, and high-resolution brain images,are among the most highly annotatedbiological samples in the entire history ofAlzheimers disease research at least,thats the claim of the Alzheimers DiseaseNeuroimaging Initiative (ADNI). Trojanowskiis co-leader of ADNIs biomarker division.

At the moment, definitive diagnosis ofAlzheimers disease requires post-mortemanalysis of the brain. While someone is stillalive, the best bet is to assess their behaviourand memory, and rule out other disorders.Doctors are desperate for a marker that canreliably tell them who will get Alzhei-mers disease, and what stage of thedisease someone is going through.

A marker like that would, of course,be useful in the clinic, but it would alsohelp researchers test drugs designedto slow the decline. The prevailing

hypothesis in Alzheimers diseaseis that deposition of the amyloid-protein leads to the formation ofinsoluble amyloid plaques betweenbrain cells, and that these plaquesare implicated in the dysfunction anddeath of brain cells (see Little proteins,big clues, page S12).

Pharmaceutical companies weremaking drugs aimed at pulling theamyloid out, or reducing the amyloid,and they needed measures to monitorthe effects of these treatments, saysMichael Weiner, professor of medi-

cine, radiology and psychiatry at the

University of California, San Francisco, andADNIs principal investigator. Obviously,imaging and biomarkers were going to beimportant tools in all of this.

WORLDWIDE NETWORK

Launched in 2004, ADNI is one of the largestand longest-running studies of Alzheimersdisease1. Its goal is to find biological markersthat can help determine how advanced some-ones disease is and predict how well they will

respond to treatment. The effort has alreadyvalidated a few sensitive markers found bysmaller studies.

This US$160-million project is fundedjointly by the US National Institutes of Health(NIH), 20 of the biggest pharma-ceuticalcompanies in the world, including Merck,AstraZeneca, Pfizer and GSK, and two non-profit partners, the Alzheimers Association andthe Alzheimers Drug Discovery Foundation.It is the largest publicprivate partnership thatthe NIH has, says Weiner.

So far, ADNI has recruited 1,000 volunteersat 59 centres across the United States andCanada. Collaborative centres have also beenset up in Europe, Japan, Australia and elsewhere.What we are trying to do is establish aworldwide network of sites that are all usingsimilar methods and sharing data, says Weiner.This makes it much easier to do internationaltreatment trials and also allows us to look atdifferences between countries.

ADNI is the best-funded effort in the huntfor Alzheimers biomarkers, but it is by nomeans the only one (see Finding risk factors,page S20). Dozens of research teams areanalysing brain images, DNA sequencevariations and patterns in the expression ofgenes, proteins and immune molecules. Ineach case the aim is to identify measurabledifferences that either aid the diagnosis ofAlzheimers disease or reflect its progression.

A CLEAR PICTURE

Weiner says he wanted to do a multi-sitestudy to compare different brain imagingtechniques, such as magnetic resonanceimaging (MRI) and positron emissiontomography (PET), which could be used todetect changes in brain structure and metab-olism associated with Alzheimers disease.He approached several pharmaceuticalcompanies, but the project was too expensivefor any company to do it alone.

He then contacted Neil Buckholtz, chiefof the Dementias of Aging Branch at theUS National Institute on Aging (NIA).Buckholtz had been pondering a similar idea,so they began a series of discussions that ledto the launch of ADNI a year later.

Of the 800 volunteers originally recruited,200 had Alzheimers disease, 400 had mildcognitive impairment (MCI) a conditionwith high risk for progression to Alzheimersdisease and 200 were healthy age-matched

controls (including Weiner himself).After spending about a year stand-ardizing operations and techniques,the team began using PET with aradioisotope of glucose known asFDG to measure brain metabolicactivity, and using MRI to measure

the volume of specific brain regions.They also recorded levels in blood

and CSF of various chemicals, includ-ing amyloid-, tau protein, sul-phatides (components of nerve cellmembranes), isoprostanes (markersof oxidative stress) and homocysteine(an amino acid), all of which hadbeen shown to be altered in Alzhei-mers disease. ADNIs main goalhas been to validate discoveries thatwere made in other smaller studies,says Weiner, and to show that theseresults really are replicable and

clinically useful.

B I O M A R K E R S

Warning signsThe hunt is on for biomarkers that signal the descent into

Alzheimers disease. One initiative is leading the pack.

AD Control

Max

MR MR[C-11]PIB PET [C-11]PIB PET

Min

Pittsburgh compound B (PiB) lights up amyloid plaques in positron

emission tomography (PET) images of the human brain.

1 4 J U L Y 2 0 1 1 | V O L 4 7 5 | N A T U R E | S 5

8/10/2019 Alzheimers Disease - Nature (07

Documents

Alzheimers Disease - Nature (07!14!11)