W W W. S C I E N C E N E W S. O R G S E P T E M B E R 1 3 , 2 0 0 3 V O L . 1 6 4 1 6 5

phrenia and 15 people with bipolar disor-der, Bahn’s team reports in the Sept. 6Lancet. Several other proteins, which reg-ulate the genes that code for the myelin-making proteins, also exhibited low con-centrations in both groups of brains. Nosuch disturbances appeared in the pre-served brains of 15 people who had had nomental disorder.

“Our findings raise questions aboutmyelin’s role in these psychiatric illnesses,”Bahn says. She’s now directing an analysisof myelin-related proteins in 150 preservedbrains from people with schizophrenia,bipolar illness, or no mental disorder.

If the results hold, they will indicate thatdisrupted myelin production may set thestage for psychosis, a warping of one’s senseof reality that often occurs in schizophrenia,bipolar disorder, and brain ailments such asAlzheimer’s disease, Bahn suggests.

Signs of schizophrenia include apathy,disorganized behavior, and psychoticsymptoms such as hallucinations anddelusions. Bipolar disorder, sometimescalled manic depression, features swingsfrom severe depression to a type of agi-tated euphoria called mania. Psychoticdelusions, say of being invincible, are acommon element of mania.

Earlier studies linked schizophrenia todisturbances in myelin-producing genes inspecific brain regions. Further work needsto examine whether the same regionaleffects characterize bipolar disorder, sayKenneth L. Davis and Vahram Haroutun-ian, both psychiatrists at Mt. Sinai Schoolof Medicine in New York, in a commentarypublished with the new report.

Psychiatrist Elliot S. Gershon of the Uni-versity of Chicago calls Bahn’s report “veryinteresting and exciting news.” In the MayAmerican Journal of Human Genetics, Ger-shon and his coworkers reported that alter-ations of two other genes, found on chro-mosome 13, frequently occur in people witheither schizophrenia or bipolar disorder.The functions of those genes are poorlyunderstood.

Accumulating data also suggest that addi-tional genes, located on chromosome 8, con-tribute both to bipolar disorder and schiz-ophrenia (SN: 9/5/98, p. 151), Gershonadds.

However, any potential link between thesetwo mental disorders is controversial. Forinstance, a study led by Pekka Tienari ofFinland’s University of Oulu found that hav-ing a biological mother with schizophreniadidn’t increase rates of bipolar illness andother mood disorders in nearly 400 adultswho had been adopted as infants. Theseresults, published in the September Amer-ican Journal of Psychiatry, clash with thenotion that schizophrenia and bipolar dis-order share genetic influences.

The two disorders stem from variousinfluences, some shared and some unique,

Bahn contends. “It’s like inflammation,” shesays. “There can be many causes for thesame symptoms.” —B. BOWER

Sweet ReliefComfort food calms, with weighty effect

The sweet and fatty foods that people oftenturn to in times of stress might in factrelieve anxiety. That’s the good news inan innovative biological theory of people’sresponses to stress. The bad news is thatfor those with chronic stress, extra serv-ings of comfort food come with poten-tially dangerous baggage—extra fat aroundthe abdomen.

Chronic stress, such as financial worries,is less well understood than are intermittentbouts of acute stress. For example, scientistsknow that when a cat is suddenly attackedby a dog or a person prepares to give a speech,the adrenal gland pumps up production ofstress hormones, including those known asglucocorticoids. When present at high-enough concentrations, glucocorticoids pro-vide feedback to the stress-response system,eventually shutting it down.

However, it’s unclear how the stressresponse is controlled in animals that areanxious for days at a time. In an upcom-ing Proceedings of the National Academyof Sciences, physiologist Mary F. Dallmanof the University of California, San Fran-cisco and her colleagues aim to close thatknowledge gap.

Drawing on their rat studies and exper-iments done by others, the scientists pro-pose that glucocorticoids work differentlyin the long term than they do in the shortterm. When chronically present in the brainand body, the hormones maintain the stressresponse instead of shutting it down. At thesame time, they drive animals to seek outpleasurable foods and direct the added calo-ries to accumulate as abdominal fat.

However, there is a brake on the process,at least in animal experiments. That extrafat eventually checks the glucocorticoids’alarmist effects and tells the brain to takeit easy again.

Results from several experiments withrats support this view, the scientists say. Inone of them, Dallman and her colleaguessimulated chronic stress by increasing thebrain concentration of a rodent version ofthe glucocorticoid called cortisol. As corti-sol concentration rose, the rats respondedby drinking increasingly more sugar water,eating increasingly more lard, and gainingabdominal girth.

In another experiment, the researchersfound that rats with extra padding produceless-than-average concentrations of a brainchemical that triggers early molecular

events underlying the stress response. “If you put on some extra weight, there

seems to be some sort of signal that saysthings are better,” says Norman C. Pecoraroof UC-San Francisco, a coauthor of thepaper. While Dallman and her coworkers

Apromising cancer treatment hasbeen dealt a setback, as theauthors of a key paper havereluctantly retracted theirreport.

Three years ago, Alexander Kugler ofthe University of Göttingen in Germanyand his colleagues described limitedsuccess with a new cancer-vaccinestrategy in Nature Medicine (SN: 3/4/00,p. 149). The researchers had takentumor cells from people with kidneycancer, fused them to the patients’ ownimmune cells, and injected the combina-tion back into the patients. In several ofthe volunteers, this stimulated animmune response against secondarytumors and led to complete remission.

Criticism of the study’s quality andethics emerged, however, and the Uni-versity of Göttingen launched an investi-gation. In November 2002, officials thereconcluded that Kugler, but not his coau-thors, was guilty of negligence and thatthe paper didn’t meet standards of goodscientific practice. For example, the uni-versity’s ethics committee hadn’tapproved the human experimentation.

Kugler’s coauthors initially resistedretracting the paper because they felt theconclusions of their study remained valid,but the editors of Nature Medicine even-tually convinced the investigators thatthe large number of errors in the pub-lished data and ethical lapses by Kuglerwarranted the action. Other researchersstress that the retraction, published inthe September Nature Medicine, doesn’tnegate the potential of using fused tumorand immune cells to stimulate a patient’simmune system against cancer.

“There are plenty of papers . . . thatshow in animal studies that theapproach is valid,” says Richard Vile ofthe Mayo Clinic in Rochester, Minn.

Work on the strategy continues. Infact, in the journal issue that carries theretraction, Vile and his colleaguesdescribe a novel process for fusing can-cer and immune cells. —J. TRAVIS

Paper ChasedCancer-vaccine study is retracted

FOBS.9-13 9/10/03 14:39 Page 165

S C I E N C E N E W S1 6 6 S E P T E M B E R 1 3 , 2 0 0 3 V O L . 1 6 4

don’t know what signal the abdominal fatsends, they suspect it’s involved with theregulation of metabolism.

The model “puts a new and more mean-ingful slant on what we mean when we talkabout ‘comfort foods,’” says Bruce S.McEwen of Rockefeller University in NewYork. “These may actually calm down animportant brain system linked to anxiety.”

In a fast-paced society where food is easyto get, glucocorticoid action probably causeschronically stressed people to take in extracalories and to gain weight, says McEwen.

“People are somehow stressed, and theyare self-medicating because food is avail-able,” adds Pecoraro.

“We also eat sugar and fat because they aregood tasting and cheap,” notes AdamDrewnowski of the University of Washing-ton in Seattle. The stress response isn’t theonly brain pathway that controls consump-tion of sweet and high-fat foods, he adds.

Whatever accounts for the urge to eat abig helping of macaroni and cheese, it’sbest not to indulge every day. The abdom-inal weight gain that Dallman and her col-leagues have linked to glucocorticoidaction increases the risk of heart diseaseand diabetes. —K. RAMSAYER

River StatsTrickle InMajor floods may bewaning in Europe

In August 2002, parts of central Europeexperienced unprecedented flooding afterrecord rains fell upon saturated soils andbrimming reservoirs. Damages on the con-tinent added up to more than 25 billionEuros, and in Dresden, Germany, the ElbeRiver reached 9.4 meters above flood stage,a level not seen since the Middle Ages.

Despite the 2002 season, a new analy-sis by German researchers suggests thatextreme summer floods in the regionaren’t becoming more frequent. In fact,the scientists say, widespread inundationshave been on the wane for the past centuryor so.

For the study, team leader ManfredMudelsee of the University of Leipzig inGermany and his coworkers consideredregional floods along the central stretchesof the Elbe and Oder Rivers dating back atleast 700 years. The scientists culled data

from historical archives and modern instru-ments, and they analyzed summer and win-ter floods separately because they have dif-ferent causes.

Floods that occur from May throughOctober typically arise during or just afterlong periods of precipitation, saysMudelsee. The frequency of such inunda-tions hasn’t changed significantly on theElbe since 1820 or on the Oder since 1920.Other researchers have reported 10 majorsummer floods on the Elbe in the past 500years, and 4 of those high-water eventsoccurred between 1500 and 1550.

So-called winter floods in the region havedeclined, says Mudelsee. These cold-season floods are often caused by ice damsthat form when frozen rivers break up in thespring. Before 1850, 91 of 103 severe win-ter floods on the Elbe, and 28 of 34 of thoseon the Oder, were influenced by river ice.Between 1930 and 1970, however, just 2 of13 winter floods on the Elbe and 3 of 20 onthe Oder were affected by ice.

Mudelsee and his colleagues also esti-mated the ameliorating influence of reser-voirs and other flood-control measures andfound that they had little effect during largeinundations. The scientists report theirfindings in the Sept. 11 Nature.

“I’m not surprised that there’s not beena rise in floods,” says Phil Jones, a clima-tologist at the University of East Anglia inNorwich, England. Current amounts of pre-cipitation aren’t very different from thosemeasured during other wet periods of thepast 2 centuries. Modern floods may appearmore serious than past ones largely becausemore people live on floodplains now thanin earlier periods, he notes.

Results of many computerized climatemodels suggest that continuation of the cur-rent increase in average global temperature

will boost evaporation from the oceans. Anaccompanying rise in precipitation couldincrease the frequency and severity offloods, some scientists say.

“I’m convinced the climate is changing,but I’m not convinced that the frequency offloods has changed,” says Kenneth W. Pot-ter, an environmental engineer at the Uni-versity of Wisconsin–Madison. —S. PERKINS

SCIENCENEWSThis Week

AP

/WID

EWO

RLD

DEEP TROUBLE The same weather system that dumped record amounts of precipitation inGermany in August of 2002 caused massive floods in Melk, Austria.

Subscribe Today!

www.sciencenews.org1-800-552-4412

A02HR

FOBS.9-13 9/10/03 14:40 Page 166