sageke.sciencemag.org/cgi/content/full/2003/50/ns8 Page 1

Mutations that cripple a molecular relay system simi-lar to the mammalian insulin and insulin-like growthfactor 1 signaling pathways enhance longevity inworms and flies. The pathways appear to influencemammalian life span as well, but that idea is far fromproven. Now, rare genetic perturbations in those sig-nals are opening a window onto how the pathwaysinfluence human aging. And new studies on rodentswith similar genetic snafus might help scientists inkin the connections.

When it comes to research on aging, lab worms strut their stuff. Butwhat everyone wants to know is whether the secrets worms divulgeapply to people. Studies on humans with rare mutations might pro-vide the answer. The idea is tantalizing, but investigating aging inpeople—especially small numbers of them—is arduous. Ongoingwork on mainstay lab mammals might help bolster the enterprise.

Research during the past decade has revealed a signaling path-way that influences aging in several organisms. Mutations thatcripple this pathway can extend the longevity of nematodes andfruit flies twofold or more. The single invertebrate pathwayresembles two overlapping pathways in mammals, onecontrolled by insulin and the other by insulin-like growthfactor 1 (IGF-1) (see “Growing Old Together”*). Becausethe machinery is similar, researchers have hoped that, as inworms and flies, tweaking one or both pathways could ex-tend human longevity or improve health in the elderly.Worm and fly biologists have elucidated many details ofhow aging is influenced by these signals, which controlgrowth and metabolism by turning genes on and off. But akey problem is proving that the same mechanisms operatein people. The answer could be obtainable: Humans withrare mutations that disrupt IGF-1 signaling are the geneticcounterparts to certain lines of long-lived rodents. Discern-ing whether these individuals also show exceptionallongevity is a daunting task, especially because quantifyinghuman aging is difficult (see Miller Perspective† and“Magic Markers”‡). But where studies of these humanslose steam, continuing work on rodents might help fill inlinks from worms and flies to people.

A Human AngleCertain dwarf rodents provided the first hints that the wormand fly signals modulate aging in mammals. These diminu-tive animals—such as the Ames dwarf§ and Snell dwarf||

mice—outlive normal creatures by about 50%. The rodents

lack proteins important for pituitary development; normally, thepituitary gland disperses growth hormone (GH) into the blood-stream and tissues crank out IGF-1 in response. Because thesedwarfs don’t produce GH, they carry abnormally small IGF-1concentrations, connecting a dearth of IGF-1 with slowed aging(see Bartke Viewpoint¶).

Technical obstacles and ethical concerns preclude genetic ma-nipulations to test the hypothesis in people. But nature might pro-vide a viable avenue by which to explore the question: Thesedwarf rodents already have human counterparts. For instance, sev-eral human populations carry mutations in Prop1#, the same genethat falters in Ames dwarf mice. These individuals could help re-searchers decipher how insulin and IGF-1 signaling affect aging inpeople. The prospects are tremendous, but bringing the idea tofruition will be laborious: Because these mutations are extremelyrare—individuals with any given mutation typically number farless than 100—researchers can’t get the volume of data necessary

Power to the PeopleExperimental animals have taken center stage as researchers have connected insulin-related signaling to

aging—and continuing studies should fill in the links. People with glitches in the mammalian versions ofthese pathways might also help researchers understand how the signals influence human aging

R. John Davenport(Published 17 December 2003)

N E W S S Y N T H E S I S

Pituitary paradox. A family from the Dominican Republic includes indi-viduals with inherited pituitary deficiencies generated by a defect inProp1. An affected 18-year-old girl (bottom row, right) appears unusuallyyouthful compared with her unaffected 22-year-old sister (bottom row,left), but 40- and 38-year-old siblings (second row from the top, left andcenter) have already developed wrinkled skin and appear aged com-pared with their 35-year-old unaffected brother (second row, far right).The parents are in the top row, and three pituitary-deficient family mem-bers are in the third row (aged 33, 29, and 27, left to right).

SAGE KESCIENCE OF AGING KNOWLEDGE ENVIRONMENT

CR

ED

IT:A

.L.R

OS

EN

BLO

OM

ET

AL.

, J.C

LIN

.EN

DO

.ME

TAB

.84,

50–

57 (

1999

)

* sageke.sciencemag.org/cgi/content/abstract/2001/2/or14† sageke.sciencemag.org/cgi/content/full/2001/1/pe2‡ www.sagecrossroads.net/news_102003.cfm§ sageke.sciencemag.org/cgi/content/full/2001/1/tg11|| sageke.sciencemag.org/cgi/content/full/2001/3/tg13

¶ sageke.sciencemag.org/cgi/content/full/2002/16/vp4# sageke.sciencemag.org/cgi/genedata/sagekeGdbGene;116

sageke.sciencemag.org/cgi/content/full/2003/50/ns8 Page 2

to make accurate assessments of longevity, says endocrinologistJohn Parks of Emory University School of Medicine in Atlanta,Georgia. Still, the promise these groups provide is too great to ig-nore. “It’s imperative to look at the evidence for and against ef-fects on life span in these humans,” says Parks.

One group of humans with inherited pituitary flaws lives onthe island of Krk, off the coast of Croatia. Records provide agesof death for four individuals in this group, who lived to 68, 77,83, and 91 years. “It’s clear that they were older than average,”says Parks—but if these folks are typical, the Krk inhabitantsaren’t as hardy as their rodent cousins are. “These were not peo-ple who were living to 120.” Why humans might show life spanchanges that are different from those of similarly modified ro-dents is unknown. Detrimental effects of hormone deprivation,which stunts growth, might be more pronounced in an organismthat lives nearly 100—instead of about 2—years, or they mightbe minimized in the sheltered environment in which lab animalsexist (see “Get Wild”**). Researchers need to gather statistics onthe 20 other known, affected individuals, but even with thosenumbers, getting robust longevity data will be difficult, saysParks: “The bottom line is, we’re not sure exactly what’s goingon in these people [in terms of longevity].” Complicating mat-ters further, many individuals with these pituitary defects nowreceive hormone therapy; the therapy helps patients maturemore normally, but “it kills the experiment,” says physiologistAndrzej Bartke of Southern Illinois University School ofMedicine in Springfield.

Despite a paucity of longevity data, these populations arehinting at how altering hormone concentrations with the goal ofextending life might impact health. Investigations of other hu-mans with malfunctioning pituitaries suggest that even if suchchanges confer long life, they come with a downside. For in-stance, members of a family in the Dominican Republic withmutations in the Prop1 gene are shorter than average and, evenin their 20s, appear adolescent and don’t show signs of puber-ty—so even if such individuals lived long, they’d be saddledwith unusually small stature and fertility problems. They never-theless exhibit characteristics of aging, such as wrinkled skin.Other individuals acquire age-related diseases unusually early.Members of the Hutterite Brethren in North America—who al-so carry a Prop1 mutation—develop osteoporosis while stillyoung, for example.

In a Brazilian population with a genetic variation analogousto that carried by another long-lived dwarf rodent, the Littlemouse,†† several people are in their 60s and 70s, establishingthat they have at least normal longevity. However, many of themhave increased blood pressure, cholesterol concentrations, andfat deposits. In addition, people in an Ecuadorian clan withLaron‡‡ syndrome—caused by mutations in the gene that en-codes the GH receptor protein—have relatively normal mortal-ity. However, like the Brazilian population, they display riskfactors for cardiovascular disease. Why these people have ap-parently normal life spans isn’t clear. Such findings at firstmight disappoint those interested in manipulating hormone sig-naling to improve health—but these apparent contradictions re-semble those seen in the long-lived rodents, bolstering the no-tion that the animals will yield insights into human physiology.

Dwarf rodents also show signs of rapid aging, says geneticistKevin Flurkey of the Jackson Laboratory in Bar Harbor, Maine.Despite having long lives ahead of them, young animals haveabnormally low blood cell counts and their wounds heal slowly.On the other hand, the mice experience a slower than normaldecline in brain function, says Bartke. “Characteristics of accel-erated aging and delayed aging are [both] expressed in thesedwarf mice,” says Flurkey, “as if one aging clock is set ahead,and another is set back.”

Despite the difficulties in distilling sound numbers from hu-man populations, Parks thinks that these groups harbor importantinformation. He wants to identify physiological parameters thatprovide an indicator of aging and measure them in these popula-tions. Such studies could provide a more practical method ofgauging longevity than life-span records. They could also revealhow quickly different tissues deteriorate, despite normal life span.He’d like to use “a battery of tests to give an inference of how faralong these people are in the aging process.” No such study is inthe works as of yet, and no one has uncovered a reliable biomarkerfor aging, but Parks says, “I need a really good aging associate:someone who knows all of the ins and outs of human aging andthe limitations of the assays.”

Data from humans with other problems might help bolster theendeavor. Cancer stems pituitary function far more frequentlythan mutations do, Parks notes. Individuals with pituitary cancerhave shortened life spans, but scientists don’t know whether the

N E W S S Y N T H E S I S

SAGE KESCIENCE OF AGING KNOWLEDGE ENVIRONMENT

Delayed development … delayed longevity? Members of theHutterite Brethren with Prop1 mutations (front, aged 55, 68, and63, left to right) are short in stature and show some signs of ac-celerated aging, such as early osteoporosis, but they attain atleast normal longevity. Such populations might help researchersunderstand whether blocking insulin/IGF-1 signaling extends lifespan in humans.

CR

ED

IT:B

ER

NA

RD

CO

RE

NB

LUM

, UN

IVE

RS

ITY

OF

CA

LGA

RY

** sageke.sciencemag.org/cgi/content/full/2003/8/ns2†† sageke.sciencemag.org/cgi/content/full/2001/4/tg14‡‡ sageke.sciencemag.org/cgi/content/full/2002/8/tg1

sageke.sciencemag.org/cgi/content/full/2003/50/ns8 Page 3

hormone deficiency or the cancer is the fatal flaw. Parks says thatif researchers engineer a mouse model of this so-called acquiredhypopituitarism—triggered in some way other than cancer—itcould help reveal whether leaving pituitary function intact duringearly life but negating it later could allow animals to develop nor-mally yet live extra-long. Perhaps the animals would also provideinsight into whether extended life span is separable from smallbody size, an ongoing controversy (see below).

Mousing for LongevityResearchers will continue to strive for the ulti-mate prize: discerning how insulin-related sig-naling influences human life span. But scien-tists won’t be able to uncover all the details theywant from humans, so future work on rodents—dwarf and otherwise—will continue to con-tribute. Previous studies on the Ames and Snellanimals suggested a connection between de-creased IGF-1 amounts and longevity, but themice also harbor reduced amounts of thyroid-stimulating hormone and prolactin, creating un-certainty about which alteration is responsiblefor life-span extension. Other dwarf mice impli-cate IGF-1 more directly, because they carrymutations that dampen only GH signaling. ButGH can exert its influence on cells and tissuesby routes other than those that involve IGF-1, soresearchers have begun tweaking the IGF-1 sys-tem itself, rather than manipulating it throughGH. For instance, scientists have removed themouse gene that encodes the IGF-1 receptor(IGF-1R), a cell surface protein that gloms ontoIGF-1. Animals that lack IGF-1R die soon afterbirth, but mice with a half-dose of IGF-1R sur-vive—and live about 25% longer than normalmice do, according to work published in Jan-uary 2003 in Nature (see “One for All”§§).

To pinpoint how the hormone signals brakeaging, future studies should tweak the pathwayin more subtle ways, says neuroendocrinologistWilliam Sonntag of Wake Forest UniversitySchool of Medicine in Winston-Salem, NorthCarolina. For instance, previous work revealedthat curtailing insulin-related signaling in wormsduring adulthood, but not during early life devel-opment, extends life span (see Sonntag andRamsey Perspective||||). To find out whether thesame holds true in mammals, “what we need areanimal models where we can precisely control IGF-1 activityin adults or during development,” he says. Bartke agrees: “Weneed to be more selective. We need to turn [IGF-1] off duringdifferent stages of life or turn it down or eliminate it in differ-ent types of cells.”

Researchers are already beginning to conduct such studies.For instance, work published last year in Aging Cell by physiol-ogist Richard Miller of the University of Michigan, Ann Arbor,and colleagues suggests that in mice, reduced IGF-1 concentra-tions early in life—rather than late, as in worms—is a key to

longevity (see “The Shrimps Shall Inherit the Earth”¶¶). Thatfinding doesn’t mesh with the way hormone therapy is adminis-tered. “The [U.S. Food and Drug Administration] recently madeit possible to administer IGF-1 and GH deliberately to childrenwho have no medical complaint other than they are short,” saysMiller. “That’s remarkably shortsighted. All the evidence [sup-porting the notion that reducing IGF-1 extends life span] sug-gests that [this treatment] may be robbing children of life.”

Insulin InsightsOther studies hint that combinations of signals might diallongevity up and down in mammals. In these creatures, the sin-gle worm and fly insulin-like pathway splits in two; insulinsparks one signal, and IGF-1 ignites the other. Insulin regulatesmetabolism, controlling glucose uptake from the blood intocells, whereas IGF-1 spurs cell growth and prods animals to de-velop into adults. Each pathway operates in many different tis-sues, and some tissues carry both systems. Moreover, both path-ways harness many of the same components, and insulin andIGF-1 can each, when presenting high concentrations, activate

N E W S S Y N T H E S I S

SAGE KESCIENCE OF AGING KNOWLEDGE ENVIRONMENT

Pituitary glandBlood sugar

IGF-1

IGF-1

IRS-1,2

PI3K

RAS

Gene activation

or suppression

Increased growth

and cell division

Accelerated aging?

Glucose uptake

Accelerated aging?

Cell in target tissue

MAPK

cascade

PancreasInsulin

Insulin

GH

IGF-1 receptor Insulin receptor

Division of labor. IGF-1 and insulin utilize similar machinery to control separatecellular events. Insulin regulates primarily glucose uptake and storage, whereasIGF-1 spurs cellular reproduction and growth. How each of the pathways con-tributes to aging in mammals remains unclear, but rare human mutations and en-gineered rodents might help clarify the issue.

SO

UR

CE

:V.D

.LO

NG

O A

ND

C.E

.FIN

CH

, SC

IEN

CE

299,

134

2–13

46 (

2003

), S

CIE

NC

E’S

ST

KE

(S

TK

E.S

CIE

NC

EM

AG

.OR

G/C

GI/C

M/S

TK

EC

M;C

MP

_120

69),

BIO

CA

RTA

(WW

W.B

IOC

AR

TA.C

OM

/GE

NE

S/IN

DE

X.A

SP

);IL

LUS

TR

ATIO

N:J

ULI

E W

HIT

E

§§ sageke.sciencemag.org/cgi/content/full/2002/49/nf15|||| sageke.sciencemag.org/cgi/content/full/2002/43/pe17 ¶¶ sageke.sciencemag.org/cgi/content/full/2002/37/nf10

sageke.sciencemag.org/cgi/content/full/2003/50/ns8 Page 4

the other hormone’s receptor.Recent studies suggest that stifling insulin signaling delays

aging. In work published last February in Science, diabetes re-searcher C. Ronald Kahn of the Joslin Diabetes Center inBoston and colleagues engineered mice so that they don’t makethe insulin-receptor protein in their fat cells; the insulin receptornormally prods cells to store sugar as fat upon binding insulin.These lean animals carry only a small amount of fat and don’tdevelop signs of diabetes, despite eating as much as or morethan normal mice do. In addition, the animals live 20% longerthan controls (see “Lasting Without Fasting”##). The study im-plies that defects in insulin signaling stall aging, presumably inaddition to the ones in the IGF-1 pathway that have been uncov-ered by the various dwarf mutation studies.

Because of the similarities between the insulin and IGF-1pathways, assessing which is most important in aging will bedifficult, says molecular biologist John Kopchick of Ohio Uni-versity College of Osteopathic Medicine in Athens. But he’swilling to speculate: “I think it’s insulin,” he says. Recent stud-ies from his research team compared longevity of mice missingthe GH receptor (Laron mice) and mice that produce a crippledGH variant that prevents normal signaling through the receptor.Mice without the GH receptor lived longer than normal, butmice with the altered GH molecule lived a normal life span. Al-though IGF-1 quantities are small in both animals, Kopchicksays, insulin amounts are small only in the Laron mice. The re-sult, he says, suggests that insulin is predominantly responsiblefor extra lifetime, although he doesn’t exclude a role for IGF-1.

Kopchick’s research might also help resolve another contro-versy: whether reducing IGF-1 amounts extends life because itdecreases body size. Much research points to a trend: Within aspecies—dogs or humans or mice—petite individuals tend tolive longer. Researchers are currently trying to separate thebody-shrinking and life-extending effects of reduced amountsof GH and IGF-1, says Bartke. For instance, Kopchick’s studysuggests that body size might not confer a longevity advantage.Both types of mice—those without the GH receptor and thosewith the GH blocker—were small, but only one was long-lived.If researchers can tease these qualities apart, they might eventu-ally be able to extend life while preserving normal stature.

Although lab results suggest that limiting IGF-1 and GH ex-tends longevity, consumers shell out untold dollars on GH-boosting injections, supplements, and sprays. Thanks to craftysales pitches based on shaky claims, some adults believe that aburst of hormone—rather than a dearth of it—will counteractaging’s toll, keeping them wrinkle-free and virile. A limitednumber of scientifically sound studies suggest that GH makesbodies leaner and more muscular but could also increase therisk of cancer and diabetes (see “Lean, Yes—But Mean?”***).Most researchers who study basic science say that the dataaren’t convincing enough to justify the risks of GH therapy—despite its allure. But arguing this point in the face of flashysales pitches and inspiring testimonials is difficult. “Some peo-ple in clinical medicine think that this is a very effective therapy[against some of the effects of aging],” says Bartke. “Againstthis background, the burden of proof that IGF-1 actually accel-erates aging is very heavy.” But future studies on rodents andunusual human populations might solidify the idea that shuttingdown IGF-1 is the path for people who want to worm their wayout of aging.

*** sageke.sciencemag.org/cgi/content/abstract/2002/47/nw158

R. John Davenport is an associate editor of SAGE KE. He appreci-ates finding new pathways—especially in the mountains.

References

1. J. A. S. Barreto-Filho et al., Familial isolated growth hormone deficiency isassociated with increased systolic blood pressure, central obesity, and dys-lipidemia. J. Clin. Endocrinol. Metab. 87, 2018-2023 (2002).

2. M. Blüher, B. B. Kahn, C. R. Kahn, Extended longevity in mice lacking theinsulin receptor in adipose tissue. Science 299, 572-574 (2003).

3. K. T. Coschigano et al., Deletion, but not antagonism, of the mouse growthhormone receptor results in severely decreased body weights, insulin andIGF-I levels and increased lifespan. Endocrinology 144, 3799-3810 (2003).

4. M. Holzenberger et al., IGF-1 receptor regulates lifespan and resistance tooxidative stress in mice. Nature 421, 182-187 (2003).

5. C. Krzisnik et al., J. Endocr. Genet. 1, 9-19 (1999).6. R. G. McArthur, K. Morgan, J. A. Phillips, M. Bala, J. Klassen, The natural

history of familial hypopituitarism. Am. J. Med. Genet. 22, 553-566 (1985).7. A. L. Rosenbloom, A. S. Almonte, M. R. Brown, D. A. Fisher, L. Baumbach,

J. S. Parks, Clinical and biochemical phenotype of familial anterior hypo-pituitarism from mutation of the PROP1 gene. J. Clin. Endocrinol. Metab.84, 50-57 (1999).

8. Z. Laron, Growth hormone insensitivity (Laron syndrome). Rev. Endocr.Metab. Disord. 3, 347-355 (2002).

N E W S S Y N T H E S I S

SAGE KESCIENCE OF AGING KNOWLEDGE ENVIRONMENT

## sageke.sciencemag.org/cgi/content/full/2003/5/nw21