Identification of New Genes and Pathways
that Act to Delay C. elegans Aging
by
Alina Berdichevsky
B. Sc.
Tel-Aviv University, 1999
M. Sc.
Weitzmann Institute of Science, 2001
Submitted to the Department of Biology in partial fulfillment of the
requirements for the degree of
DOCTOR OF PHILOSOPHY
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
Septemb r 2007
@ 2007 Alina Berdichevsky. All rights reserved.
The author hereby grants MIT permission to reproduce and to distribute
paper and electronic copies of this thesis document in whole or in part.
Signature of Author:
eep6artment of Biology, eptembf 2007
Certified by:
)
H. Robert Horvi)
Professor of Biology
Thesis Supervisor
Leonard Guarinte
Professor of Biology
Thesis supervisor
Accepted by:
Steven Bell
Steven Bell
Cnair of the Giraduate Committee
Department of Biology
i
OF TEOHNOLOGY
ARCH-VE•S
FEB 2 2008
t
i8zm
LIBRARIES
I
1
Identification of New Genes and Pathways
that Act to Delay C. elegans Aging
Alina Berdichevsky
Abstract
Aging of an organism is determined by both stochastic and genetic
components. The importance of genes is illustrated by the discovery of
single gene mutations that alter lifespans of species ranging from
invertebrates C. elegans and D. melanogasterto mice. To better understand
the mechanisms underlying genetic regulation of the rate of aging, we have
studied aging and lifespan determination in C. elegans.
We have investigated the mechanism of action of the sir-2.1 gene,
which extends C. elegans lifespan when overexpressed. We discovered that
sir-2.1 acts to promote longevity in parallel to low insulin signaling, in a
stress-dependent pathway that converges with the insulin-like pathway on a
forkhead transcription factor DAF-16. We discovered that 14-3-3 proteins
play a role in C. elegans lifespan determination. 14-3-3 proteins interact
with the SIR-2.1 protein, and the 14-3-3 genes par-5 andftt-2 are required
for the longevity mediated by sir-2.I1overexpression, indicating that 14-3-3
proteins act to delay C. elegans aging in the sir-2.I1-dependent longevity
pathway.
To identify new genes that act to delay C. elegans aging we
developed a genetic screen for mutants that prematurely accumulate the
age-related fluorescent pigment lipofuscin. In this screen, we isolated lossof-function mutations in the gene kat-1. These mutations confer increased
lipofuscin accumulation, short lifespan, and other abnormalities
characteristic of premature aging. kat-1 encodes a ketoacyl thiolase
involved in fatty acid beta-oxidation, suggesting that defects in fat
metabolism can affect the regulation of aging.
Thesis Supervisors: H. Robert Horvitz
Professor of Biology
Leonard Guarente
Professor of Biology
Acknowledgments
I would like to thank my advisors Lenny Guarente and Bob Horvitz
for their guidance, their constant encouragement and support, and for giving
me the freedom of choice. Thanks for all members of my graduate
committee during these years: Marcia Haigis, Dennis Kim, Terry OrrWeaver, and especially Chris Kaiser and Richard Hynes for their input and
ideas and for reading this thesis.
I'd like to thank many members of the Guarente and the Horvitz labs
for their advice. Special thanks to Nick Bishop, Gil Blander, Brendan
Galvin, Niels Ringstad and Hillel Schwartz for ideas, arguments,
discussions, and critical reading of my manuscripts. Thanks to Beth Castor
and Na An for making the lab feel more like home, and for all the help with
strains, sequencing, and deletion hunt. I would like to thank Peter Reddien
for the great time during my rotation and for recruiting me to the Horvitz
lab.
I'd like to thank all friends of mine both in US and abroad who
shared a lot of happy and distressing moments. Brendan Galvin was my
sarcastic roommate all these years, and I would like to thank him for always
being in the right place at the right time for me and for all the balloons
blown on weekends for Ron. Takashi Hirose was a great bay-mate for the
last two years, and a good friend.
I would like to thank my parents who managed to survive these six
years so far from their daughter and grandchildren, and who have always
encouraged me to go on.
Igor, Ron, and Dana filled my life and I would not be able to carry on
without them.
Table of contents
A bstract..........................................................................
2
Acknowledgments................................................................. 3
Table of C ontents ..................................................................
.4
Chapter 1. Introduction: Genetic and Molecular Pathways that
Control A ging...................
................................................ 11
O verview .......................................................................... 12
The conserved features of aging ........................................... 13
Isolation of first mutants with abnormal aging rates...
C. elegans Aging ......................
..............
15
..................................... 17
An insulin-like pathway controls the rate of aging in C. elegans
And other organisms......................................................... 18
Dauer pathway .................................................. 18
Some but not all dauer pathway mutants age abnormally....19
Molecular cloning of dauer pathway genes reveals a
genetic pathway that controls aging.........................20
The role of insulin-like pathway in aging isconserved........21
Aging, DNA damage, and the stress response ............................. 22
Sirtuins and lifespan determination......................................23
Mitochondria and the metabolic control of aging......................25
The "rate of living" hypothesis....
...............
.......
25
C. elegans mitochondrial mutants are long-lived ............ 25
Metabolic activity during aging and calorie restriction.......26
Other genes and pathways that control C. elegans aging .............. 27
Mammalian genes that regulate aging .................................... 28
Conclusions .................................................................... 30
R eferences.........................
.........................................31
Figure 1. Current theories proposed to explain the cause of aging .... 45
Figure 2. Genes from one of the two arms of the dauer pathway
regulate C. elegans longevity......................................... 47
Figure 3. Insulin/IGF pathway regulates C. elegans longevity and
stress-resistance.......................
...................... 49
Chapter 2. C. elegans SIR-2.1 interacts with 14-3-3 proteins to activate
DAF-16 and extend lifespan..........................................
Summary ...................................................
Introduction ..............................
R esults.............................
....... 51
.............
52
..................................... 53
............................................ 56
C. elegans SIR-2.1 interacts with 14-3-3 proteins .........................
56
C. elegans 14-3-3 proteins can interact with DAF-16 and act
to retain DAF-16::GFP in the cytoplasm.............................57
sir-2.1 overexpression slows DAF-16::GFP exit from the nucleus
during recovery from stress .................................. .............. 60
The 14-3-3 genesftt-2 and par-5 are required for
sir-2.I1-mediated lifespan extension....................................61
sir-2.1, par-5 andftt-2 are not required for lifespan extension in
animals with reduced insulin/IGF signaling ........................... 63
sir-2.I1can promote resistance to stress in anftt-2-dependentway.... 63
sir-2.1 can promote daf-16-dependent transcription in a manner
dependent onftt-2.........................................................64
Following stress, SIR-2.1 interacts with DAF-16 in a
14-3-3-dependent fashion.................................. .................. 65
Discussion...........................
............................................ 67
C. elegans SIR-2.1 and 14-3-3 proteins can function in
a stress-response pathway of longevity...............................67
SIR-2.1 and 14-3-3 may affect longevity by regulating DAF-16
Forkhead in parallel to the insulin-like pathway.....................69
A model for the roles of SIR-2.1 and 14-3-3 proteins in DAF-16
regulation and longevity .................................................... 70
Experimental procedures....................................................71
Antibody production, immunofluorescence, and cell fractionations ..71
Immunoprecipitation and western blots................................72
Lifespan, stress-resistance and RNAi analyses ......................... 73
Strains...........................................................................74
Molecular biology..............................................................77
Acknowledgments .............................................................. 78
R eferences ........................................................................... 79
T ables............................................................................... 86
Table 1. Reducing par-5 andftt-2 by RNAi results in nuclear
accumulation of DAF-16::GFP........................................................ 86
Table 2. Analyses of lifespan experiments ............................... 87
Table 3. daf-16 targets upregulated in a sir-2.I1overexpresser ........
Figures ......................... ........................................
88
89
Figure 1. C. elegans SIR-2.1 interacts with 14-3-3 proteins........89
Figure 2. C. elegans 14-3-3 proteins interact specifically
with SIR-2.1 and DAF-16 ............................................. 92
Figure 3. 14-3-3 proteins affect DAF-16 localization.....
......... 94
Figure 4. Overexpression of sir-2.1 delays DAF-16 exit
from the nucleus during recovery from stress .......................96
Figure 5. Overexpression of sir-2.1 is required for longevity
and stress resistance of the pklsl642 transgenic animals.............98
Figure 6. ftt-2 and par-5 are required for sir-2.I1-mediated
lifespan extension.....................
................................. 100
Figure 7. Overexpression of sir-2.1 leads to stress-resistance
and activation of the daf-16 target gene sod-3 in an
ftt-2-dependent manner..............................
102
Figure 8. DAF-16 and SIR-2.1 interact in worms following
stress, and this interaction depends on 14-3-3....................104
Figure 9. daf-2 RNAi results in low insulin-like signaling........ 106
Figure 10. A model for the roles of SIR-2.1 and 14-3-3
in DAF-16 regulation of stress resistance and lifespan ............108
Chapter 3. 3-Ketoacyl thiolase controls the rate of aging in C. elegans.. 110
Abstract ...................................................................... 111
Introduction .........................................................
R esults..........................
...... 112
......................................... 115
Aging pigment that accumulates in old worms fluoresces
with a spectrum similar to that of mammalian lipofuscin ....... 115
Mutants with abnormal aging rates show abnormal
fluorescence accumulation ..............................
116
Screen for mutants with increased lipofuscin..................116
Mutants isolated in the screen correspond to at least
three complementation groups ........................................ 117
Three isolates premature display multiple attributes of aging... 118
Mapping and cloning of the gene identified by n4430........... 120
T02G5.8 encodes a conserved thiolase involved in fatty
acid metabolism.............................................
121
kat-1 delays aging independently of most known longevity
pathways ........................................................
122
kat-1 mutants are sensitive to starvation .................................. 123
Other fatty-acid beta-oxidation genes affect aging ........................ 123
Discussion..........................................................................
126
KAT-1 thiolase regulates C. elegans aging..............................126
Fatty acid oxidation in aging and disease ................................. 127
Does impaired fatty acid beta-oxidation cause aging? ...................... 128
Experimental procedures..........................................................130
Strains ........................................................................ 130
Isolation of mutants with high lipofuscin-like fluorescence ........ 130
Linkage group assessment ........................................
131
Polymorphism mapping ................................. ................ 131
Transgenic animals ........................................
.......131
RNAi analyses ............................................
.......132
Allele sequences ..........................................
...... 132
Analyses of lifespan and stress-resistance ......................... 132
Behavioral assays ........................................
Electron Microscopy .........................................
133
...... 134
Acknowledgments .................................................... 135
References .......................................................... 135
Figure and tables ................................................................... 143
Table 1. Characterization of screen isolates............................ 143
Table 2. Epistasis Analyses of kat-1 with other aging pathways..... 144
Table 3. kat-1 mutants tissue aging...................................... 145
Table 4. Analyses of lipofuscin accumulation and lifespan in
animals subjected to RNAi knock down of fatty acid beta-
oxidation genes.......................
..................................... 146
Figure 1. Old worms accumulate autofluorescent pigment, and
amount of the pigment correlates with lifespan in aging mutants.......147
Figure 2. A screen for mutants with high accumulation of
lipofuscin-like fluorescent pigment....................................149
Figure 3. Mutations in gene specified by n4430, n4434, and n4435
result in increased fluorescence and short lifespan........
............. 151
Figure 4. Behavior abnormalities of n4430, n4434,
and n4435 mutants .........................................................
153
Figure 5. kat-1 tissue aging...............................................155
Figure 6. Mapping and cloning of kat-1 ... .........
......... ..
. 157
Figure 7. kat-1 controls fluorescence accumulation and lifespan.......159
Figure 8. kat-1 encodes a conserved ketoacyl thiolase................161
Figure 9. kat-1 stress resistance ........................................... 163
Figure 10. 3-ketoacyl thiolase catalyzes the last step in fatty acid
beta-oxidation pathway ..................................................... 165
Figure 11. Reducing fatty acid oxidation genes causes
increased lipofuscin accumulation in C. elegans.......................
.. 167
Addendum to Chapter 3. FLU-1 kynurenine monooxigenase controls
intestinal fluorescence accumulation in C. elegans..............................169
A bstract... ..........................
............................................. 170
Results .............................................................................
171
Caracterization of LGV isolates......................................171
Mapping and cloning offlu-1 ........................................ 171
flu-i encodes a conserved metabolic enzyme, kynurenine 3monooxygenase ..................................................... 173
D iscussion.........................
..........................
.................. 175
Conclusion and future directions............................................177
R eferences..........................
............................................ 178
Tables .................... .............................
..............
179
Table 1. Three of the seven screen isolates with mutations that
map to chromosome V have coding mutations in the gene
encoding FLU-1 R07B7.5 kinurenine 3-monooxygenase............179
Figures.............................
............................................. 180
Figure 1. Chromosome V mutants accumulate high amounts
of autofluorescence.....................................................
. 180
Figure 2. Mapping and cloning offlu-1 ... ... ... ... .................. 182
Figure 3. flu-1 encodes a 3-kynurenine mooxygenase,
a conserved metabolic enzyme....................................
184
Figure 4. Kynurenine metabolic pathway in tryptophan
m etabolism .....................................................
.......
186
Appendix 1. A Stress Response Pathway Involving Sirtuins, Forkheads,
and 14-3-3 Proteins .....................................................................
188
Introduction: Sir2 and mechanisms of life span determination .... 189
14-3-3 proteins: new players in life span determination ........... 190
Mechanism of SIR-2.1 effect on life span..........................192
SIR-2.1 pathway for life span extension..............................193
Emerging group of DAF-16 co-factors .....
References...............................
Figures....
.......................
...............................
193
196
............. .............. . ..............
200
Figure 1. SIR-2.1 mechanism of life span extension. ............. 200
Figure 2. DAF-16 can be activated by different cofactors..........202
Chapter 1
Introduction
Genetic and Molecular Pathways that Control Aging
Overview
Aging of an organism is generally accompanied by a progressive
dysfunction of organs and tissues causing physiological decline and
eventually death. Several theories have been proposed to explain the causes
of aging, including oxidative stress leading to accumulation of damage to
proteins and DNA, mitochondrial dysfunction, and changes in metabolic rate
(Figure 1). Free radicals, or reactive oxygen species, are by-products of
cellular metabolism and accumulate during aging, causing oxidative stress
and posing a threat to cells and tissues. A "rate of living" hypothesis
suggests that the rate of aging is inversely proportional to metabolic rate, i.e.
higher metabolic rate accelerates aging and limits lifespan. Mitochondria
provide energy to cells, and mitochondrial dysfunction caused by an
accumulation of oxidative damage has also been suggested as a cause of
aging.
Over the last three decades, experiments involving C. elegans and other
organisms have provided evidence supporting many of these theories and
have demonstrated that in addition to the stochastic component of the aging
process certain genes regulate the timing of aging. Interfering with the
function of such genes and the pathways they regulate can significantly
change the aging rate of an organism, resulting in delayed aging and
increased longevity or premature aging (progeria). At least some of the
mechanisms that control lifespan are conserved from nematodes to
mammals, as are certain metabolic, cellular and behavioral features of aging
organisms. Studies of short-lived and long-lived mutants of simple
organisms might advance our understanding of the genetic regulation of
human aging and identify potential therapeutic targets that can be
manipulated to delay aging and aging-related disease.
The conserved features of aging
Aging can be described as an increased probability of death with
time, increased susceptibility to various diseases and decreased resistance to
injury and infection (Knight 1995; Kirkwood and Austad 2000). While
lifespans of individuals of the same species might differ substantially, a
population shows a characteristic lifespan distribution, with a predictable
mean and maximum lifespan (Piantanelli et al. 1992). During the first two
thirds of an organism's lifespan, the mortality is very low, and it increases
exponentially during the third phase of the lifespan (Witten and Eakin 1997).
This mortality pattern is conserved in evolution, with species as different as
nematodes and humans showing similar shapes of lifespan curves, although
the absolute values for mean and maximum lifespan vary greatly (Finch
1990; Piantanelli et al. 1992). Historically, longevity has often been used as
the ultimate measure of aging and the terms longevity and aging have been
used interchangeably. However, a distinction has to be made between
longevity, which is an interval of time between the birth and the death of an
organism, and the rate of aging, which reflects the physiological decline of
an organism during the second half of its lifespan, which precedes death. As
aging is a complex phenomenon and lifespan can be limited by different
factors, alteration of the rate of aging does not necessarily imply an effect on
longevity (Kirkwood 1997). Also, numerous diseases limit longevity without
affecting the rate of aging, complicating the distinction between short
lifespans caused by premature aging and unrelated disease.
Comprehensive studies of the rate of aging in simple organisms were
limited because of the lack of reliable markers of aging in species other than
mammals. Therefore, descriptive studies of aging in simple organisms are
valuable, because they facilitate the distinction between old and sick
animals, and allow one to follow the rate of aging. Aging-related diseases,
i.e. diseases with time of onset in the late adulthood and exponential increase
in prevalence with age, might differ in diverse species, but some basal
aspects of organismal changes with age seem to be conserved (see below).
Damage to cells, both from external sources (UV light, toxic
chemicals) and internal metabolism products (reactive oxygen species)
contributes to aging (Finch 1990; Shigenaga et al. 1994; Kirkwood and
Austad 2000; Stevnsner et al. 2002; Kujoth et al. 2005). At the cellular level,
many cell types stop dividing after a limited number of replications
(Hayflick 1979). Such postreplicative cells accumulate damage that
eventually causes malfunction and/or death of the cell. Old cells from
different tissues show structural defects, such as the appearance of protein
aggregates, necrosis-like membrane structures, and accumulation of
fluorescent substances, often called lipofuscin ("fluorescent lipid") as
oxidized lipids are major components of this material (Chen and Hillman
1999; Hinault et al. 2006; Widmer et al. 2006). These features are observed
in many multicellular organisms, including worms, flies, and mammals
(Miquel et al. 1974; Klass 1977; Clokey and Jacobson 1986; Hosokawa et
al. 1994; Brunk and Terman 2002; Herndon et al. 2002; Gerstbrein et al.
2005).
On a tissue level, reproductive organs, digestive organs and muscles
seem to be universally affected in old age (Thomson and Keelan 1986;
Herndon et al. 2002; Di Iorio et al. 2006). The nervous system, on the other
hand, greatly suffers from aging in mammals, but might be relatively
unaffected in nematodes (Garigan et al. 2002; Herndon et al. 2002; Miller
and O'Callaghan 2003; Well et al. 2007). Conserved behavioral aspects of
old age include progressive slowing of locomotion (which might be
attributed to muscle failure), decrease in food intake, and cessation of
reproduction (Klass 1977; Glenn et al. 2004; Huang et al. 2004).
Finally, old age is associated with metabolic changes, which might
result from defects in mitochondrial metabolic pathways controlling energy
production. Mitochondrial DNA accumulates numerous mutations with age,
leading to mitochondrial dysfunction, which results in reduction in oxidative
phosphorylation efficiency, fat metabolism abnormalities, and promotes
hyperglycemia and insulin resistance (Shigenaga et al. 1994; Wallace 2005;
Reddy and Rao 2006).
Isolation of first mutants with abnormal aging rates
Until recently, it was broadly thought that aging stems from multiple
stochastic events leading to multi-system failure, and therefore represents a
passive deterioration process that is not controlled by individual genes or
cellular functions. During the last three decades, scientists began questioning
the idea that aging is entirely stochastic, because similar organisms with
very comparable metabolic rates often had lifespans that differed two to five
fold. The examples include mice and bats, rats and chipmunks, and even
inbred and outbred mice (Austad and Fischer 1991).
Several groups began searching for mutants in simple organisms with
abnormal aging rates, and such mutants were indeed found. In 1983 Michael
Klass screened for aging genes in C. elegans and identified several mutants
with long lifespans (Klass 1983). Mutants in age-i, one of the genes isolated
in Klass's screen, have long lifespans, increased resistance to stress, and
impaired reproduction (Friedman and Johnson 1988). Later, age-i and other
C. elegans mutants with abnormal aging were key for the identification of
conserved genetic pathways that control aging and longevity. These
pathways will be described below. Another exciting story was the discovery
of methuselah gene in Drosophila,mutations in which resulted in 25%
lifespan extension (Lin et al. 1998). Mutations in several Drosophilagenes
were isolated in longevity screens and other genetic experiments. Many of
these mutants exhibited increased resistance to stress in addition to extended
lifespan (Seong et al. 2001; Helfand and Rogina 2003).
Several groups have studied the genetic control of aging in budding
yeast. In S. cerevisiae,a yeast mother cells can give rise to a limited number
of daughter cells and subsequently stops dividing. This replicative aging can
be followed and modulated genetically (Kaeberlein et al. 2001). Studies of
yeast replicative aging identified a new class of aging genes: silent
information regulators, or sirs (Kaeberlein et al. 1999). One gene from this
group, sir2, was shown to also regulate longevity of fruit flies and
nematodes (Tissenbaum and Guarente 2001; Rogina and Helfand 2004),
suggesting that the sir2 role in the regulation of aging is conserved.
In the last decade, several mouse mutants with abnormal aging rates
were also identified, and targeted genetic manipulations were shown to
cause longer lifespans or progeria in mice (Hekimi 2006). In particular,
mitochondrial function and DNA repair seem to be key processes that
determine the rate of aging in mice. Studies of aging and longevity of
rodents have always been hampered by the fact that their lifespans are
relatively long, ranging from three to five years, which makes genetic
analysis of mutants very time-consuming. This difficulty might explain in
part why the majority of current knowledge about the genetic control of
aging and longevity came from the study of simple organisms, primarily the
nematode C. elegans.
C. elegans Aging
The first experiments describing aging of the nematode C. elegans
were conducted in the 1970s by Michael Klass (Klass 1977). C. elegans has
a short lifespan, on average around three weeks, and, as noted above, shows
a sharp increase in mortality rate in the last third of its lifespan, similar to
that observed in mammals. Behavioral analysis of aging C. elegans
nematodes showed that worms gradually slow the rates of locomotion,
feeding, and defecation with age (Klass 1977; Huang et al. 2004). Other
groups studied accumulation of the fluorescent pigment lipofuscin, which is
known to accumulate with age in postreplicative cells in many organisms,
including humans. The autofluorescent pigment accumulation in old
nematodes was shown to localize to the secondary lysosomes, much like
human lipofuscin, and the rate of lipofuscin accumulation was altered in
aging mutants (Clokey and Jacobson 1986; Gerstbrein et al. 2005; Chapter
3 of this thesis).
Excellent genetic tools developed for studies of C. elegans biology
helped isolate numerous mutants with abnormal lifespans and significantly
promoted our understanding of genetic and molecular pathways involved in
aging.
An insulin-like pathway controls the rate of aging in C. elegans and
other organisms
Dauer pathway
Studies of C. elegans development identified mutants that
subsequently defined the first genetic pathway involved in lifespan
regulation in multiple species. Nematodes develop from a fertilized egg to
adulthood through four larval stages L1 through L4, separated by moltings.
Both the timing of development and developmental decisions in C. elegans
are tightly regulated by temperature and food availability. When the
environment becomes unfavorable, either because the temperature is too
high, food is scarce, or the worms are too crowded, an alternative
developmental pathway is activated, which is called the dauer pathway.
Dauer is an alternative third larval stage, and the decision to become dauer
has to be made before the transition between the L2 and L3 larval states.
Dauers are structurally different from L3 larvae. They are extremely
resistant to stress, have low metabolic rate, do not require feeding, and
therefore are able to survive for extended periods of time in harsh
environmental conditions (Riddle 1997). Genetic screens for mutants with
abnormal dauer formation resulted in the identification of mutants that
entered the dauer state even when conditions were favorable (Daf-c, dauerconstitutive mutants), and mutants that failed to initiate the dauer program
even when conditions were unfavorable (Daf-d, dauer-defective mutants)
(Albert and Riddle 1988). Epistasis experiments helped order the mutants
into a dauer pathway, which consisted of two groups of genes acting in
parallel to promote dauer formation (Larsen et al. 1995).
Some but not all dauer pathway mutants age abnormally
In 1993, Cynthia Kenyon made a historic observation that one of the dauer
pathway mutants, a constitutive-dauer mutant daf-2 has an extremely long
lifespan, more than twice as long as than that of the wild type (Kenyon et al.
1993). Both longevity and constitutive dauer formation of daf-2 mutants
require a functional daf-16 gene, mutations in which result in a dauerdefective phenotype. Later it was shown that the longevity of age-i mutants
also depends on daf-16, and that age-i animals are dauer-constitutive,
similar to the daf-2 mutants, suggesting that the dauer pathway controls
longevity in C. elegans (Larsen et al. 1995). Moreover, the dauerconstitutive mutants age-1 and daf-2 not only had longer lifespans but also
maintained the appearance and behavior of young animals much longer
than the wild type, arguing that these mutations indeed change the aging
rate of mutant organisms (Kenyon et al. 1993; Herndon et al. 2002).
Importantly, not all the mutants of the dauer pathway had lifespan
abnormalities. In fact, only mutants in one of the two arms of the dauer
pathway had abnormal lifespans (Figure 2), suggesting that aging and entry
into dauer state are differentially controlled (Larsen et al. 1995). The timing
of aging regulation was also different from that of dauer entry: even though
the same genes were shown to control dauer formation and lifespan, dauer
formation depended on signaling during embryogenesis and early larval
development, whereas longevity effects required signaling during adulthood
(Dillin et al. 2002a).
19
Molecular cloning of dauer pathway genes reveals a genetic pathway
that controls aging
It took several years before the molecular identities of the above
longevity genes were uncovered. Studies primarily in the Ruvkun laboratory
led to cloning of genes responsible for mutant phenotypes in daf-2, age-i,
and daf-16 animals (Kimura et al. 1997; Lin et al. 1997; Paradis and Ruvkun
1998; Wolkow et al. 2002). It was shown that daf-2 encodes the C. elegans
homolog of insulin/IGF receptor, age-1 encodes a PI3 kinase, and daf-16
encodes a forkhead transcription factor homologous to mammalian Foxo.
These proteins are members of the insulin signaling pathway that has been
studied extensively in mammalian cells (Cantley 2002). Subsequent genetic
and molecular studies identified multiple additional members of the
insulin/IGF pathway in C. elegans and in mammals, including akt-1 and akt2 genes encoding kinases homologous to the mammalian AKT/PKB, sgk-1
that encodes SGK kinase, PTEN phosphatase gene daf-18, and others
(Braeckman et al. 2001). The biochemical events involved in signal
transduction along the kinase cascade upon activation of the insulin pathway
were elucidated primarily in mammalian cells (Saltiel and Kahn 2001), but
experiments in C. elegans provided important information on the genetic
interactions among insulin-like pathway genes, and studies in nematodes and
mammals were complementary.
The insulin/IGF-1 signaling (Figure 2) originates by binding of an
insulin-like molecule to the DAF-2 transmembrane receptor, which activates
a cascade of intracellular kinases including PI3 kinase AGE-1, AKT/PKB
and the DAF-16 forkhead family transcription factor (Guarente and Kenyon
2000; Finch and Ruvkun 2001; Kenyon 2005). Mutations that decrease
signaling by the DAF-2 receptor pathway cause an extension of the lifespan
20
of adults (and more frequent entry into the dauer developmental state by
larvae) by triggering the nuclear localization of DAF-16 (Henderson and
Johnson 2001; Lee et al. 2001; Lin et al. 2001). Such mutations also increase
the resistance of animals to oxidative and genotoxic stress. The longevity
and stress resistance engendered by a reduction in insulin/IGF signaling is
abolished in mutants defective in daf-16 (Murakami and Johnson 1996; Lin
et al. 1997; Paradis and Ruvkun 1998; Honda and Honda 1999). The
forkhead transcription factor DAF- 16 can activate a plethora of target genes
that promote longevity and stress-resistance (Cahill et al., 2001; Honda and
Honda, 1999; Lee et al., 2003; McElwee et al., 2003).
The role of the insulin-like pathway in aging is conserved
A major impact of the studies of the C. elegans insulin-like pathway in
worm lifespan determination was that the role of this pathway in aging
regulation is evolutionary conserved. Studies of Drosophilashowed that
mutants in the Drosophilainsulin receptor and insulin-receptor substrate
were extremely long-lived and stress-resistant, much like C. elegans mutants
(Helfand and Rogina 2003). Mutations in the insulin pathway in mammals
cause insulin resistance, hyperglycemia and diabetes mellitus, and mutations
in the IGF somatotrophic axis cause dwarfism. Recent findings from studies
of mice, however, indicate that dwarf mice with mutations in growth
hormone (GH), that have low levels of circulating IGF, and IGFR (IGF
receptor) mutants have 25% to 60% longer lifespans than the wild type
(Flurkey et al. 2001). Also, tissue specific mutations in the insulin receptor
(IR) gene cause an increase in mean and maximum lifespan (Bluher et al.
2002). An important observation is that effects of insulin/IGF mutations on
murine lifespan are often gender-specific. This phenomenon does not have
an explanation at this point but suggests that hormonal signal circuitries and
other gender-specific biological processes have to be further studied to
understand the differential effects on lifespan of insulin pathway mutations
in males and females.
Aging, DNA damage, and the stress response
Accumulation of damage to DNA and proteins resulting from
oxidative stress as a consequence of normal metabolism is now a widely
accepted explanation of the aging process. This hypothesis emerged from the
ideas of Harman (1956) in the middle of the last century and was supported
by a clear association between aging and the presence of oxidative stress and
oxidative damage. When the level of damage reaches a certain threshold,
cellular dysfunction occurs, eventually leading to organ dysfunction
characteristic to aging. Therefore, the ability of the organism to control the
rate of damage accumulation and to repair the damage is likely important for
longevity and the rate of aging. In accordance with this idea, many human
disorders caused by mutations in genes that control DNA maintenance and
repair of DNA damage result in a phenotype that resembles premature aging
and are referred to as progeroid disorders. Examples include Xeroderma
Pigmentosum, which results from Nucleotide Excision Repair defects;
Werner Syndrome, caused by mutations in a DNA helicase; and Ataxia
Telangiectasia, caused by mutations in the ATM kinase required for repair
of double-stranded DNA breaks (Shiloh and Rotman 1996; Lehmann 2003;
Niedernhofer et al. 2006; Orren 2006). Importantly, patients with any of
these disorders show many but not all attributes of aging and hence they are
referred to as "segmental" progeroid syndromes.
One common characteristic of these progeroid diseases is greatly
increased cellular sensitivity to stress. In simple organisms, mutations that
cause stress sensitivity often also cause short lifespan and other attributes of
premature aging. Conversely, increased resistance to stress is a characteristic
of long-lived mutants of the insulin-like pathway in nematodes and fruit flies
(Honda and Honda 1999; Guarente and Kenyon 2000; Braeckman et al.
2001; Honda and Honda 2002). However, resistance to stress does not
always parallel longevity. In fact, some long-lived mutants are less resistant
to certain stressors, while increased resistance to stress is not always
associated with longevity (Munoz and Riddle 2003). These results imply that
variables other than stress response contribute to the determination of
longevity and aging rate.
Sirtuins and lifespan determination
sir2 (silent information regulator 2) encodes a conserved protein
deacetylase, which controls stress-resistance and lifespan determination in
different organisms. The role of sir2 in aging was first discovered in S.
cerevisiae,where it was identified in a screen for genes that control yeast
replicative aging. In S. cerevisiae, extra copies of sir2 cause significant
extension of replicative lifespan, while deletion of sir2 results in lifespan
shortening (Kaeberlein et al. 1999). Furthermore, sir2 is required for the
lifespan extension caused by calorie restriction in S. cerevisiae (Lin et al.
2002). Many organisms, including C. elegans, Drosophila,mice and
23
humans, have multiple sir2 homologs (sirtuins). The nematode genome
encodes four sirtuins, sir-2.I1- sir-2.4, and humans have seven sirtuins,
SIRT1-SIRT7, many of which have important roles in metabolism or
genomic integrity (Frye 2000; Haigis and Guarente 2006).
Some sirtuins, including the yeast ortholog Sir2 and the worm SIR2.1 possess NAD-dependent deacetylase activity (Imai et al. 2000;
Viswanathan M., personal communication). Studies of the yeast and
mammalian sirtuins indicated that they can bind and deacetylate histones as
well as several other proteins with important roles in metabolism and stress
response. Some examples of sirtuin-binding proteins are p53, forkhead
transcription factors, and PGC-1 (Luo et al. 2001; Brunet et al. 2004;
Daitoku et al. 2004; Rodgers et al. 2005; Gerhart-Hines et al. 2007).
Enzymatic activity of Sir2 is tightly controlled by the NAD/NADH ratio,
and it was proposed that Sir2 might act as a sensor of the metabolic state of
the cell and transfer this information to downstream effectors (Guarente and
Kenyon 2000).
Introducing extra copies of the closest C. elegans homologue of the
yeast sir2, sir-2.I1into C. elegans genome results in increased resistance to
stress and 30-50% extension of the worm lifespan (Tissenbaum and
Guarente 2001; Berdichevsky et al. 2006). Furthermore, deleting sir-2.I1
causes 15% shortening of lifespan (Viswanathan et al. 2005). Experiments
by us and others indicate that C. elegans sir-2.I1acts to promote longevity
via a stress-dependent pathway that converges with the insulin-like pathway
at the level of transcription factor DAF-16 (Berdichevsky et al. 2006; Wang
et al. 2006). Overexpressing the fly sir2 homologue also results in lifespan
extension, and the mechanism underlying this effect in flies may be similar
to calorie restriction (Rogina and Helfand 2004).
Mitochondria and the metabolic control of aging
The "rate of living" hypothesis
Mitochondrial DNA is especially prone to the accumulation of
damage and mutations caused by reactive oxygen species (ROS), byproducts of oxidative phosphorylation. Since ROS production is
proportional to metabolic rate, it has been suggested that metabolic rate is
coupled to lifespan (Pearl 1928). This "rate of living" hypothesis proposes
that duration of lifespan inversely depends on the metabolic rate of an
organism.
C. elegans mitochondrial mutants are long-lived
Recent studies of nematodes support the idea of an inverse
relationship between metabolic rate and aging: mutants in many C. elegans
mitochondrial genes that affect components of the oxidative phosphorylation
chain and therefore inhibit energy production develop slowly, show
decreased rates of locomotion, decreased brood sizes, and have up to 2-fold
longer lifespans (Felkai et al. 1999; Feng et al. 2001; Dillin et al. 2002b).
The mitochondrial gene clk-i encodes a member of the ubiquinone pathway,
a component of ETC complex IV, and clk-i mutants are long-lived (Felkai et
al. 1999). This effect is additive to that of loss-of-function of daf-2 insulinlike receptor gene, and double mutants between daf-2 and clk-i live much
longer than any single mutant, indicating that mitochondrial function affects
lifespan in parallel to the insulin-like pathway. Other C. elegans
mitochondrial genes that when mutated cause long lifespan include isp-1, a
complex IV protein; ctb-1, another complex IV protein encoded by
mitochondrial genome (Feng et al. 2001); and lrs-2, a mitochondrial leucyl
tRNA-synthetase (Lee et al. 2003). RNAi knockdown of several genes
encoding mitochondrial ETC subunits extend the worm's lifespan. In a
genome-wide RNAi screen for genes causing lifespan extension of C.
elegans, approximately 20% of genes the knockdown of which promotes
longevity encode proteins involved in mitochondrial electron transport chain
(ETC), suggesting a pivotal role for the mitochondria and therefore
metabolic rate in the regulation of longevity (Lee et al. 2003; Aguilaniu et al.
2005). Moreover, mice heterozygous for the mClkl, the homolog of C.
elegans clk-I gene were reported to have extended lifespans (Liu et al.
2005).
Metabolic activity during aging and calorie restriction
While in some mutants reduced metabolic activity is associated with
increased longevity, aging organisms also show reduced metabolism. A
comprehensive study of C. elegans metabolic activity showed a linear
reduction in oxygen consumption and ATP content with age (Braeckman et
al. 2002). Similar results were obtained from studies of mouse, rat, and
human tissues (Meacci et al. 1993; Imre et al. 2000; Martin et al. 2007).
Metabolic defects associated with mammalian aging lead to several
disorders, such as diabetes mellitus and metabolic syndrome (Shigenaga et
al. 1994; Toth and Tchernof 2000; Saltiel and Kahn 2001).
The most well-known means to extend lifespan is to limit food
consumption. Calorie restriction (CR) promotes longevity and delays agingrelated disorders in all organisms tested, including worms, flies, mice and
primates (Koubova and Guarente 2003). It had been suggested in the past
that CR causes lifespan extension by reducing metabolic rate, much like
mitochondrial mutants, therefore limiting damage to cells and delaying
tissue and organismal aging. Recent results, however, do not support this
hypothesis. It appears that at least in some species, calorie-restriction does
not lead to reduced metabolism. Studies of yeast and worms show that CR
actually causes increased respiration and oxygen consumption (Houthoofd et
al. 2002a; Houthoofd et al. 2002b; Lin et al. 2002; Bishop and Guarente
2007). Additionally, mammals subjected to CR are more active than agematched controls and do not display differences in body temperature or ATP
production (McCarter RJM 1997; Weed et al. 1997). These findings argue
that reduced metabolic rate is not always associated with delayed aging and
longer lifespan. The molecular mechanisms underlying the CR effect on
aging are largely unknown. CR extends lifespan independently of the
insulin-like signaling, as insulin-like pathway mutants respond normally to
CR (Houthoofd et al. 2003). sir2 homologs are required for CR-induced
increase in respiration and lifespan in yeast, and possibly in flies (Lin et al.
2000; Rogina and Helfand 2004). Studies are underway to identify genes
and pathways that play a role in the beneficial effects of calorie restriction
on aging and aging-related disease (Bishop and Guarente 2007; Panowski et
al. 2007).
Other genes and pathways that control C. elegans aging
Directly reducing reproduction can also have a positive effect on C.
elegans lifespan, as ablation of worm germline cell precursors by laser or
mutation results in lifespan extension (Hsin and Kenyon 1999), suggesting
that signals from the germline limit lifespan in worms. The germline
signaling effect on lifespan requires functional DAF-16 Foxo, but is likely
independent from insulin/IGF receptor DAF-2, since reducing insulin-like
pathway activity in the germline signaling mutants causes synergistic effect
on lifespan (Arantes-Oliveira et al. 2003).
Nematodes can sense aspects of their environment, such as
temperature and food availability, and modulate their behavior accordingly.
C. elegans mutants with defects in sensory perception are often long-lived
(Apfeld and Kenyon 1999; Alcedo and Kenyon 2004). This effect is
unlikely to be caused by food restriction, because CR can extend lifespans of
the C. elegans sensory mutants to the same extent as of the wild type
(Bishop N., personal communication).
Recently, several reports indicated that inhibiting translation in adult
nematodes results in increased lifespan (Hansen et al. 2007; Syntichaki et al.
2007). Also, autophagy has been shown to play a role in lifespan
determination, as longevity of insulin-like pathway mutants requires the becI gene, which is involved in autophagy, and overexressing autophagy genes
extends worm lifespan (Melendez et al. 2003). Taken together, these
findings suggest that deregulation of protein synthesis and/or accumulation
of misfolded proteins in aging nematodes contribute to aging, and preventing
protein accumulation by either inducing autophagy or attenuating protein
synthesis can delay aging.
Mammalian genes that regulate aging.
The relatively long lifespans of mice and rats have limited the genetic
analysis of aging pathways in mammals, and only a small number of genes
have been shown so far to affect mammalian aging. On the other hand, the
interpretation of progeroid mutants is easier in rodents than in invertebrates,
because old rodents display more markers characteristic of human aging.
Many human progeroid disorders caused primarily by defects in DNA repair
have been recapitulated in mice, including Werner syndrome, Ataxia
Telangiectasia, Xeroderma Pigmentosum (XP), and Cockayne syndrome
(van de Ven et al. 2007). For XP mice, analysis of the progeroid phenotype
of mutants was extended beyond anatomical descriptions, and transcriptional
profiling of mutant animals showed remarkable similarity between profiles
of progeroid mutants and old animals (Niedernhofer et al. 2006; van der
Pluijm et al. 2006). A gain-of-function mutation in the p53 protein, which is
involved in maintaining genomic stability, also causes a progeroid
phenotype in mice, possibly because of increased programmed cell death
leading to degeneration of various organs and tissues (Tyner et al. 2002).
A mitochondrial link to aging was also found in mice: overexpression
of a mutator DNA polymerase in mitochondria leads to an increased number
of mutations in mitochondrial DNA, and premature aging (Kujoth et al.
2005). By contrast, superoxide dismutase mimetics and overexpression of
peroxisomal catalase in mitochondria extend murine lifespan, presumably by
reducing the rate of reactive oxygen species accumulation (Melov et al.
2000; Schriner et al. 2005). For some human and mouse progeroid
syndromes, mutations responsible for the phenotype may be identified, but
the mechanisms by which they act or pathways that they regulate are
unknown. Examples include human Hutchinson-Gilford progeroid syndrome
resulting from mutations in the LaminA gene (Eriksson et al. 2003), and the
apparently causal role of the Klotho protein in mouse aging (Kurosu et al.
2005; Masuda et al. 2005).
29
Conclusions
This is a very exciting time for studies of the genetic regulation of
aging. Numerous pathways that affect longevity have recently been
identified. It is clear that genes involved in responses to DNA damage,
mitochondrial function and metabolic pathways contribute to the
determination of lifespan. Studies of simple organisms like the nematode C.
elegans are extremely valuable for the identification of new pathways that
regulate longevity and for understanding the genetic interactions among
different longevity pathways. It is tempting to speculate that one or more of
these pathways act in humans and can be manipulated to delay aging or
aging-related disease.
One limitation of the field is its focus on genes that limit lifespan,
mutations in which cause increased longevity. There are few comprehensive
studies of genes that act to slow the rate of aging, primarily caused by the
complication of distinguishing progeria from sickness in simple organisms.
Development of reliable biomarkers of aging should solve this problem and
facilitate studies of genes that promote longevity. This thesis describes three
such genes, C. elegans sir-2.1,ftt-2, and kat-1, which act in new pathways of
aging regulation.
References
Aguilaniu, H., Durieux, J., and Dillin, A. 2005. Metabolism, ubiquinone
synthesis, and longevity. Genes Dev 19(20): 2399-2406.
Albert, P.S. and Riddle, D.L. 1988. Mutants of Caenorhabditiselegans that
form dauer-like larvae. Dev Biol 126(2): 270-293.
Alcedo, J. and Kenyon, C. 2004. Regulation of C. elegans longevity by
specific gustatory and olfactory neurons. Neuron 41(1): 45-55.
Apfeld, J. and Kenyon, C. 1999. Regulation of lifespan by sensory
perception in Caenorhabditiselegans. Nature 402(6763): 804-809.
Arantes-Oliveira, N., Berman, J.R., and Kenyon, C. 2003. Healthy animals
with extreme longevity. Science 302(5645): 611.
Austad, S.N. and Fischer, K.E. 1991. Mammalian aging, metabolism, and
ecology: evidence from the bats and marsupials. J Gerontol 46(2):
B47-53.
Berdichevsky, A., Viswanathan, M., Horvitz, H.R., and Guarente, L. 2006.
C. elegans SIR-2.1 interacts with 14-3-3 proteins to activate DAF-16
and extend life span. Cell 125(6): 1165-1177.
Bishop, N.A. and Guarente, L. 2007. Two neurons mediate diet-restrictioninduced longevity in C. elegans. Nature 447(7144): 545-549.
Bluher, M., Michael, M.D., Peroni, O.D., Ueki, K., Carter, N., Kahn, B.B.,
and Kahn, C.R. 2002. Adipose tissue selective insulin receptor
knockout protects against obesity and obesity-related glucose
intolerance. Dev Cell 3(1): 25-38.
Braeckman, B.P., Houthoofd, K., and Vanfleteren, J.R. 2001. Insulin-like
signaling, metabolism, stress resistance and aging in Caenorhabditis
elegans. Mech Ageing Dev 122(7): 673-693.
-. 2002. Assessing metabolic activity in aging Caenorhabditiselegans:
concepts and controversies. Aging Cell 1(2): 82-88; discussion 102103.
Brunet, A., Sweeney, L.B., Sturgill, J.F., Chua, K.F., Greer, P.L., Lin, Y.,
Tran, H., Ross, S.E., Mostoslavsky, R., Cohen, H.Y., Hu, L.S.,
Cheng, H.L., Jedrychowski, M.P., Gygi, S.P., Sinclair, D.A., Alt,
F.W., and Greenberg, M.E. 2004. Stress-dependent regulation of
FOXO transcription factors by the SIRT1 deacetylase. Science
303(5666): 2011-2015.
Brunk, U.T. and Terman, A. 2002. Lipofuscin: mechanisms of age-related
accumulation and influence on cell function. FreeRadic Biol Med
33(5): 611-619.
Cantley, L.C. 2002. The phosphoinositide 3-kinase pathway. Science
296(5573): 1655-1657.
Chen, S. and Hillman, D.E. 1999. Dying-back of Purkinje cell dendrites
with synapse loss in aging rats. JNeurocytol 28(3): 187-196.
Clokey, G.V. and Jacobson, L.A. 1986. The autofluorescent "lipofuscin
granules" in the intestinal cells of Caenorhabditiselegans are
secondary lysosomes. Mech Ageing Dev 35(1): 79-94.
Daitoku, H., Hatta, M., Matsuzaki, H., Aratani, S., Ohshima, T., Miyagishi,
M., Nakajima, T., and Fukamizu, A. 2004. Silent information
regulator 2 potentiates Foxo 1-mediated transcription through its
deacetylase activity. Proc Natl Acad Sci USA 101(27): 1004210047.
Di lorio, A., Abate, M., Di Renzo, D., Russolillo, A., Battaglini, C., Ripari,
P., Saggini, R., Paganelli, R., and Abate, G. 2006. Sarcopenia: age-
32
related skeletal muscle changes from determinants to physical
disability. Int JlmmunopatholPharmacol19(4): 703-719.
Dillin, A., Crawford, D.K., and Kenyon, C. 2002a. Timing requirements for
insulin/IGF-1 signaling in C. elegans. Science 298(5594): 830-834.
Dillin, A., Hsu, A.L., Arantes-Oliveira, N., Lehrer-Graiwer, J., Hsin, H.,
Fraser, A.G., Kamath, R.S., Ahringer, J., and Kenyon, C. 2002b.
Rates of behavior and aging specified by mitochondrial function
during development. Science 298(5602): 2398-2401.
Eriksson, M., Brown, W.T., Gordon, L.B., Glynn, M.W., Singer, J., Scott,
L., Erdos, M.R., Robbins, C.M., Moses, T.Y., Berglund, P., Dutra,
A., Pak, E., Durkin, S., Csoka, A.B., Boehnke, M., Glover, T.W., and
Collins, F.S. 2003. Recurrent de novo point mutations in lamin A
cause Hutchinson-Gilford progeria syndrome. Nature 423(6937):
293-298.
Felkai, S., Ewbank, J.J., Lemieux, J., Labbe, J.C., Brown, G.G., and
Hekimi, S. 1999. CLK-1 controls respiration, behavior and aging in
the nematode Caenorhabditiselegans. Embo J 18(7): 1783-1792.
Feng, J., Bussiere, F., and Hekimi, S. 2001. Mitochondrial electron
transport is a key determinant of life span in Caenorhabditiselegans.
Dev Cell 1(5): 633-644.
Finch, C. 1990. Longevity, Senescence, and the Genome. . Chicago,IL,
USA: Chicago UniversityPress.
Finch, C.E. and Ruvkun, G. 2001. The genetics of aging. Annu Rev
Genomics Hum Genet 2: 435-462.
Flurkey, K., Papaconstantinou, J., Miller, R.A., and Harrison, D.E. 2001.
Lifespan extension and delayed immune and collagen aging in
mutant mice with defects in growth hormone production. Proc Natl
Acad Sci USA 98(12): 6736-6741.
Friedman, D.B. and Johnson, T.E. 1988. A mutation in the age-i gene in
Caenorhabditiselegans lengthens life and reduces hermaphrodite
fertility. Genetics 118(1): 75-86.
Frye, R.A. 2000. Phylogenetic classification of prokaryotic and eukaryotic
Sir2-like proteins. Biochem Biophys Res Commun 273(2): 793-798.
Garigan, D., Hsu, A.L., Fraser, A.G., Kamath, R.S., Ahringer, J., and
Kenyon, C. 2002. Genetic analysis of tissue aging in Caenorhabditis
elegans: a role for heat-shock factor and bacterial proliferation.
Genetics 161(3): 1101-1112.
Gerhart-Hines, Z., Rodgers, J.T., Bare, O., Lerin, C., Kim, S.H.,
Mostoslavsky, R., Alt, F.W., Wu, Z., and Puigserver, P. 2007.
Metabolic control of muscle mitochondrial function and fatty acid
oxidation through SIRT1/PGC-lalpha. Embo J26(7): 1913-1923.
Gerstbrein, B., Stamatas, G., Kollias, N., and Driscoll, M. 2005. In vivo
spectrofluorimetry reveals endogenous biomarkers that report
healthspan and dietary restriction in Caenorhabditiselegans. Aging
Cell 4(3): 127-137.
Glenn, C.F., Chow, D.K., David, L., Cooke, C.A., Gami, M.S., Iser, W.B.,
Hanselman, K.B., Goldberg, I.G., and Wolkow, C.A. 2004.
Behavioral deficits during early stages of aging in Caenorhabditis
elegans result from locomotory deficits possibly linked to muscle
frailty. J GerontolA Biol Sci Med Sci 59(12): 1251-1260.
Guarente, L. and Kenyon, C. 2000. Genetic pathways that regulate ageing
in model organisms. Nature 408(6809): 255-262.
34
Haigis, M.C. and Guarente, L.P. 2006. Mammalian sirtuins--emerging roles
in physiology, aging, and calorie restriction. Genes Dev 20(21):
2913-2921.
Hansen, M., Taubert, S., Crawford, D., Libina, N., Lee, S.J., and Kenyon,
C. 2007. Lifespan extension by conditions that inhibit translation in
Caenorhabditiselegans. Aging Cell 6(1): 95-110.
Harman, D. 1956. Aging: a theory based on free radical and radiation
chemistry. J Gerontol 11(3): 298-300.
Hayflick, L. 1979. Progress in cytogerontology. Mech Ageing Dev 9(5-6):
393-408.
Hekimi, S. 2006. How genetic analysis tests theories of animal aging. Nat
Genet 38(9): 985-991.
Helfand, S.L. and Rogina, B. 2003. Genetics of aging in the fruit fly,
Drosophilamelanogaster.Annu Rev Genet 37: 329-348.
Henderson, S.T. and Johnson, T.E. 2001. daf-16 integrates developmental
and environmental inputs to mediate aging in the nematode
Caenorhabditiselegans. Curr Biol 11(24): 1975-1980.
Hemdon, L.A., Schmeissner, P.J., Dudaronek, J.M., Brown, P.A., Listner,
K.M., Sakano, Y., Paupard, M.C., Hall, D.H., and Driscoll, M. 2002.
Stochastic and genetic factors influence tissue-specific decline in
ageing C. elegans. Nature 419(6909): 808-814.
Hinault, M.P., Ben-Zvi, A., and Goloubinoff, P. 2006. Chaperones and
proteases: cellular fold-controlling factors of proteins in
neurodegenerative diseases and aging. JMol Neurosci 30(3): 249265.
Honda, Y. and Honda, S. 1999. The daf-2 gene network for longevity
regulates oxidative stress resistance and Mn-superoxide dismutase
gene expression in Caenorhabditiselegans. FasebJ 13(11): 13851393.
-. 2002. Oxidative stress and life span determination in the nematode
Caenorhabditiselegans. Ann N YAcad Sci 959: 466-474.
Hosokawa, H., Ishii, N., Ishida, H., Ichimori, K., Nakazawa, H., and
Suzuki, K. 1994. Rapid accumulation of fluorescent material with
aging in an oxygen-sensitive mutant mev-1 of Caenorhabditis
elegans. Mech Ageing Dev 74(3): 161-170.
Houthoofd, K., Braeckman, B.P., Johnson, T.E., and Vanfleteren, J.R.
2003. Life extension via dietary restriction is independent of the
Ins/IGF-1 signalling pathway in Caenorhabditis elegans. Exp
Gerontol 38(9): 947-954.
Houthoofd, K., Braeckman, B.P., Lenaerts, I., Brys, K., De Vreese, A., Van
Eygen, S., and Vanfleteren, J.R. 2002a. Axenic growth up-regulates
mass-specific metabolic rate, stress resistance, and extends life span
in Caenorhabditiselegans. Exp Gerontol 37(12): 1371-1378.
-. 2002b. No reduction of metabolic rate in food restricted Caenorhabditis
elegans. Exp Gerontol 37(12): 1359-1369.
Hsin, H. and Kenyon, C. 1999. Signals from the reproductive system
regulate the lifespan of C. elegans. Nature 399(6734): 362-366.
Huang, C., Xiong, C., and Kornfeld, K. 2004. Measurements of age-related
changes of physiological processes that predict lifespan of
Caenorhabditiselegans. Proc Natl Acad Sci USA 101(21): 80848089.
Imai, S., Armstrong, C.M., Kaeberlein, M., and Guarente, L. 2000.
Transcriptional silencing and longevity protein Sir2 is an NADdependent histone deacetylase. Nature 403(6771): 795-800.
36
Imre, S., Firbas, J.H., and Noble, R.C. 2000. Reduced lipid peroxidation
capacity and desaturation as biochemical markers of aging. Arch
Gerontol Geriatr31(1): 5-12.
Kaeberlein, M., McVey, M., and Guarente, L. 1999. The SIR2/3/4 complex
and SIR2 alone promote longevity in Saccharomyces cerevisiae by
two different mechanisms. Genes Dev 13(19): 2570-2580.
-. 2001. Using yeast to discover the fountain of youth. Sci Aging Knowledge
Environ 2001(1): pel.
Kenyon, C. 2005. The plasticity of aging: insights from long-lived mutants.
Cell 120(4): 449-460.
Kenyon, C., Chang, J., Gensch, E., Rudner, A., and Tabtiang, R. 1993. A C.
elegans mutant that lives twice as long as wild type. Nature
366(6454): 461-464.
Kimura, K.D., Tissenbaum, H.A., Liu, Y., and Ruvkun, G. 1997. daf-2, an
insulin receptor-like gene that regulates longevity and diapause in
Caenorhabditiselegans. Science 277(5328): 942-946.
Kirkwood, T.B. 1997. The origins of human ageing. Philos Trans R Soc
Lond B Biol Sci 352(1363): 1765-1772.
Kirkwood, T.B. and Austad, S.N. 2000. Why do we age? Nature
408(6809): 233-238.
Klass, M.R. 1977. Aging in the nematode Caenorhabditiselegans: major
biological and environmental factors influencing life span. Mech
Ageing Dev 6(6): 413-429.
-. 1983. A method for the isolation of longevity mutants in the nematode
Caenorhabditiselegans and initial results. Mech Ageing Dev 22(3-4):
279-286.
37
Knight, J.A. 1995. The process and theories of aging. Ann Clin Lab Sci
25(1): 1-12.
Koubova, J. and Guarente, L. 2003. How does calorie restriction work?
Genes Dev 17(3): 313-321.
Kujoth, G.C., Hiona, A., Pugh, T.D., Someya, S., Panzer, K., Wohlgemuth,
S.E., Hofer, T., Seo, A.Y., Sullivan, R., Jobling, W.A., Morrow, J.D.,
Van Remmen, H., Sedivy, J.M., Yamasoba, T., Tanokura, M.,
Weindruch, R., Leeuwenburgh, C., and Prolla, T.A. 2005.
Mitochondrial DNA mutations, oxidative stress, and apoptosis in
mammalian aging. Science 309(5733): 481-484.
Kurosu, H., Yamamoto, M., Clark, J.D., Pastor, J.V., Nandi, A., Gurnani,
P., McGuinness, O.P., Chikuda, H., Yamaguchi, M., Kawaguchi, H.,
Shimomura, I., Takayama, Y., Herz, J., Kahn, C.R., Rosenblatt, K.P.,
and Kuro-o, M. 2005. Suppression of aging in mice by the hormone
Klotho. Science 309(5742): 1829-1833.
Larsen, P.L., Albert, P.S., and Riddle, D.L. 1995. Genes that regulate both
development and longevity in Caenorhabditiselegans. Genetics
139(4): 1567-1583.
Lee, R.Y., Hench, J., and Ruvkun, G. 2001. Regulation of C. elegans DAF16 and its human ortholog FKHRL1 by the daf-2 insulin-like
signaling pathway. Curr Biol 11(24): 1950-1957.
Lee, S.S., Lee, R.Y., Fraser, A.G., Kamath, R.S., Ahringer, J., and Ruvkun,
G. 2003. A systematic RNAi screen identifies a critical role for
mitochondria in C. elegans longevity. Nat Genet 33(1): 40-48.
Lehmann, A.R. 2003. DNA repair-deficient diseases, xeroderma
pigmentosum, Cockayne syndrome and trichothiodystrophy.
Biochimie 85(11): 1101-1111.
38
Lin, K., Dorman, J.B., Rodan, A., and Kenyon, C. 1997. daf-16: An HNF3/forkhead family member that can function to double the life-span
of Caenorhabditiselegans. Science 278(5341): 1319-1322.
Lin, K., Hsin, H., Libina, N., and Kenyon, C. 2001. Regulation of the
Caenorhabditiselegans longevity protein DAF- 16 by insulin/IGF- 1
and germline signaling. Nat Genet 28(2): 139-145.
Lin, S.J., Kaeberlein, M., Andalis, A.A., Sturtz, L.A., Defossez, P.A.,
Culotta, V.C., Fink, G.R., and Guarente, L. 2002. Calorie restriction
extends Saccharomyces cerevisiae lifespan by increasing respiration.
Nature 418(6895): 344-348.
Lin, Y.J., Seroude, L., and Benzer, S. 1998. Extended life-span and stress
resistance in the Drosophilamutant methuselah. Science 282(5390):
943-946.
Liu, X., Jiang, N., Hughes, B., Bigras, E., Shoubridge, E., and Hekimi, S.
2005. Evolutionary conservation of the clk-l-dependent mechanism
of longevity: loss of mclkl increases cellular fitness and lifespan in
mice. Genes Dev 19(20): 2424-2434.
Luo, J., Nikolaev, A.Y., Imai, S., Chen, D., Su, F., Shiloh, A., Guarente, L.,
and Gu, W. 2001. Negative control of p53 by Sir2alpha promotes cell
survival under stress. Cell 107(2): 137-148.
Martin, C., Dubouchaud, H., Mosoni, L., Chardigny, J.M., Oudot, A.,
Fontaine, E., Vergely, C., Keriel, C., Rochette, L., Leverve, X., and
Demaison, L. 2007. Abnormalities of mitochondrial functioning can
partly explain the metabolic disorders encountered in sarcopenic
gastrocnemius. Aging Cell 6(2): 165-177.
Masuda, H., Chikuda, H., Suga, T., Kawaguchi, H., and Kuro-o, M. 2005.
Regulation of multiple ageing-like phenotypes by inducible klotho
39
gene expression in klotho mutant mice. Mech Ageing Dev 126(12):
1274-1283.
McCarter RJM, S.I., Ikeno Y, Higami Y, Hubbard GB, Yu BP, McMahan
CA. 1997. Physical activity as a factor in the action of dietary
restriction on aging: Effects in Fischer 344 rats. AGING-CLINICAL
AND EXPERIMENTAL RESEARCH 9((1-2)): 73-79.
Meacci, E., Vannini, F., Vasta, V., Farnararo, M., and Bruni, P. 1993.
Effect of aging on insulin regulation of fructose 2,6-bisphosphate
metabolism in human fibroblasts. Biochem Mol Biol Int 30(1): 91-98.
Melendez, A., Talloczy, Z., Seaman, M., Eskelinen, E.L., Hall, D.H., and
Levine, B. 2003. Autophagy genes are essential for dauer
development and life-span extension in C. elegans. Science
301(5638): 1387-1391.
Melov, S., Ravenscroft, J., Malik, S., Gill, M.S., Walker, D.W., Clayton,
P.E., Wallace, D.C., Malfroy, B., Doctrow, S.R., and Lithgow, G.J.
2000. Extension of life-span with superoxide dismutase/catalase
mimetics. Science 289(5484): 1567-1569.
Miller, D.B. and O'Callaghan, J.P. 2003. Effects of aging and stress on
hippocampal structure and function. Metabolism 52(10 Suppl 2): 1721.
Miquel, J., Tappel, A.L., Dillard, C.J., Herman, M.M., and Bensch, K.G.
1974. Fluorescent products and lysosomal components in aging
Drosophilamelanogaster.J Gerontol 29(6): 622-637.
Munoz, M.J. and Riddle, D.L. 2003. Positive selection of Caenorhabditis
elegans mutants with increased stress resistance and longevity.
Genetics 163(1): 171-180.
Murakami, S. and Johnson, T.E. 1996. A genetic pathway conferring life
extension and resistance to UV stress in Caenorhabditiselegans.
Genetics 143(3): 1207-1218.
Niedernhofer, L.J., Garinis, G.A., Raams, A., Lalai, A.S., Robinson, A.R.,
Appeldoorn, E., Odijk, H., Oostendorp, R., Ahmad, A., van
Leeuwen, W., Theil, A.F., Vermeulen, W., van der Horst, G.T.,
Meinecke, P., Kleijer, W.J., Vijg, J., Jaspers, N.G., and Hoeijmakers,
J.H. 2006. A new progeroid syndrome reveals that genotoxic stress
suppresses the somatotroph axis. Nature 444(7122): 1038-1043.
Orren, D.K. 2006. Werner syndrome: molecular insights into the
relationships between defective DNA metabolism, genomic
instability, cancer and aging. FrontBiosci 11: 2657-2671.
Paradis, S. and Ruvkun, G. 1998. Caenorhabditiselegans Akt/PKB
transduces insulin receptor-like signals from AGE-1 PI3 kinase to the
DAF-16 transcription factor. Genes Dev 12(16): 2488-2498.
Pearl, R. 1928. The rate of living: Being an account of some experimental
studies on the biology of life duration. . AA Knopf New York.
Piantanelli, L., Rossolini, G., and Nisbet, R. 1992. Modelling survivorship
kinetics: a two-parameter model. Gerontology 38(1-2): 30-40.
Reddy, J.K. and Rao, M.S. 2006. Lipid metabolism and liver inflammation.
II. Fatty liver disease and fatty acid oxidation. Am JPhysiol
GastrointestLiver Physiol 290(5): G852-858.
Riddle, D., Blumenthal, T, Meyer, BJ, Priess, J. 1997. C. elegans II. CSHL
Press, Cold Spring Harbor,New York.
Rodgers, J.T., Lerin, C., Haas, W., Gygi, S.P., Spiegelman, B.M., and
Puigserver, P. 2005. Nutrient control of glucose homeostasis through
a complex of PGC-lalpha and SIRT1. Nature 434(7029): 113-118.
Rogina, B. and Helfand, S.L. 2004. Sir2 mediates longevity in the fly
through a pathway related to calorie restriction. Proc Natl Acad Sci U
SA 101(45): 15998-16003.
Saltiel, A.R. and Kahn, C.R. 2001. Insulin signalling and the regulation of
glucose and lipid metabolism. Nature 414(6865): 799-806.
Schriner, S.E., Linford, N.J., Martin, G.M., Treuting, P., Ogburn, C.E.,
Emond, M., Coskun, P.E., Ladiges, W., Wolf, N., Van Remmen, H.,
Wallace, D.C., and Rabinovitch, P.S. 2005. Extension of murine life
span by overexpression of catalase targeted to mitochondria. Science
308(5730): 1909-1911.
Seong, K.H., Ogashiwa, T., Matsuo, T., Fuyama, Y., and Aigaki, T. 2001.
Application of the gene search system to screen for longevity genes
in Drosophila.Biogerontology2(3): 209-217.
Shigenaga, M.K., Hagen, T.M., and Ames, B.N. 1994. Oxidative damage
and mitochondrial decay in aging. ProcNatl Acad Sci US A 91(23):
10771-10778.
Shiloh, Y. and Rotman, G. 1996. Ataxia-telangiectasia and the ATM gene:
linking neurodegeneration, immunodeficiency, and cancer to cell
cycle checkpoints. J Clin Immunol 16(5): 254-260.
Stevnsner, T., Thorslund, T., de Souza-Pinto, N.C., and Bohr, V.A. 2002.
Mitochondrial repair of 8-oxoguanine and changes with aging. Exp
Gerontol 37(10-11): 1189-1196.
Syntichaki, P., Troulinaki, K., and Tavernarakis, N. 2007. eIF4E function in
somatic cells modulates ageing in Caenorhabditiselegans. Nature
445(7130): 922-926.
Thomson, A.B. and Keelan, M. 1986. The aging gut. Can JPhysiol
Pharmacol64(1): 30-38.
Tissenbaum, H.A. and Guarente, L. 2001. Increased dosage of a sir-2 gene
extends lifespan in Caenorhabditiselegans. Nature 410(6825): 227230.
Toth, M.J. and Tchernof, A. 2000. Lipid metabolism in the elderly. Eur J
Clin Nutr 54 Suppl 3: S121-125.
Tyner, S.D., Venkatachalam, S., Choi, J., Jones, S., Ghebranious, N.,
Igelmann, H., Lu, X., Soron, G., Cooper, B., Brayton, C., Hee Park,
S., Thompson, T., Karsenty, G., Bradley, A., and Donehower, L.A.
2002. p53 mutant mice that display early ageing-associated
phenotypes. Nature 415(6867): 45-53.
van de Ven, M., Andressoo, J.O., Holcomb, V.B., Hasty, P., Suh, Y., van
Steeg, H., Garinis, G.A., Hoeijmakers, J.H., and Mitchell, J.R. 2007.
Extended longevity mechanisms in short-lived progeroid mice:
identification of a preservative stress response associated with
successful aging. Mech Ageing Dev 128(1): 58-63.
van der Pluijm, I., Garinis, G.A., Brandt, R.M., Gorgels, T.G., Wijnhoven,
S.W., Diderich, K.E., de Wit, J., Mitchell, J.R., van Oostrom, C.,
Beems, R., Niedernhofer, L.J., Velasco, S., Friedberg, E.C., Tanaka,
K., van Steeg, H., Hoeijmakers, J.H., and van der Horst, G.T. 2006.
Impaired genome maintenance suppresses the growth hormone-insulin-like growth factor 1 axis in mice with Cockayne syndrome.
PLoS Biol 5(1): e2.
Viswanathan, M., Kim, S.K., Berdichevsky, A., and Guarente, L. 2005. A
role for SIR-2.1 regulation of ER stress response genes in
determining C. elegans life span. Dev Cell 9(5): 605-615.
Wallace, D.C. 2005. A mitochondrial paradigm of metabolic and
degenerative diseases, aging, and cancer: a dawn for evolutionary
medicine. Annu Rev Genet 39: 359-407.
Wang, Y., Oh, S.W., Deplancke, B., Luo, J., Walhout, A.J., and
Tissenbaum, H.A. 2006. C. elegans 14-3-3 proteins regulate life span
and interact with SIR-2.1 and DAF-16/FOXO. Mech Ageing Dev
127(9): 741-747.
Weed, J.L., Lane, M.A., Roth, G.S., Speer, D.L., and Ingram, D.K. 1997.
Activity measures in rhesus monkeys on long-term calorie restriction.
Physiol Behav 62(1): 97-103.
Well, D., Yang, H., Houseni, M., Iruvuri, S., Alzeair, S., Sansovini, M.,
Wintering, N., Alavi, A., and Torigian, D.A. 2007. Age-related
structural and metabolic changes in the pelvic reproductive end
organs. Semin Nucl Med 37(3): 173-184.
Widmer, R., Ziaja, I., and Grune, T. 2006. Protein oxidation and
degradation during aging: role in skin aging and neurodegeneration.
Free Radic Res 40(12): 1259-1268.
Witten, T.M. and Eakin, T. 1997. Multiphasic models of survival: analysis
of mortality rate change regions and the issue of finite species
lifespan. Exp Gerontol 32(3): 259-285.
Wolkow, C.A., Munoz, M.J., Riddle, D.L., and Ruvkun, G. 2002. Insulin
receptor substrate and p55 orthologous adaptor proteins function in
the Caenorhabditiselegans daf-2/insulin-like signaling pathway. J
Biol Chem 277(51): 49591-49597.
Figures
Figure 1. Current theories proposed to explain the cause of aging. It is
generally accepted that oxidative stress plays a central role in aging. Free
radicals, or reactive oxygen species, are by-products of cellular metabolism
and accumulate during aging, causing oxidative stress. The result is damage
to proteins and DNA, which in turn can lead to mitochondrial dysfunction,
energy deficit and metabolic abnormalities associated with aging. A "rate of
living" hypothesis suggests that the rate of aging is inversely proportional to
metabolic rate, i.e. higher metabolic rate accelerates aging and limits
lifespan.
45
Figure 1
Oxidative stress
Damage to DNA
and proteins
I
AGING
Mitochondrial
dysfunction
Metabolic
Metabolic
abnormalities
46
Figure 2. Genes from one of the two arms of the dauer pathway
regulate C. elegans longevity (edited from Larsen 1995). daf-2 and daf-7
mutants are Daf-c (form dauers constitutively), indicating that these genes
have a role in promoting continuous, non-Dauer development. Four other
Daf-c genes, daf-1, daf-4, daf-8, and daf-14, have genetic interactions
similar to those of mutations in daf-7, suggesting that they function with
daf-7. Genetic epistasis analysis indicated that Daf-d (dauer-defective)
genes daf-3 and daf-5 act downstream of daf-7 pathway, because mutations
in any of these genes suppress the effects of mutations in daf-7, daf-1, daf4, daf-8 or daf-14. daf-3 and daf-5 seem to function specifically in this
pathway since they do not strongly suppress mutations in Daf-c genes daf-2
and age-1/daf-23. By contrast, dauer-defective gene daf-16 suppresses the
effects of mutations in daf-2 and age-1/daf-23, but not of mutations in daf7. Mutations in daf-2 and age-1/daf-23 genes confers longevity to adults,
while mutations in daf-7, daf-1, daf-4, daf-8 or daf-14 do not.
47
Figure 2.
daf-7
d
Sdaf-4
daf-8
" daf-14
daf-3
daf-5
DAUER ENTRY
daf-2 -*
age-1/daf-23
Hdaf-16
ADULT LONGEVITY
Figure 3. Insulin/IGF pathway regulates C. elegans longevity and
stress-resistance. The signaling cascade is initiated by binding of an
insulin-like substrate molecule to the receptor encoded by the daf-2 gene.
This binding event leads to activation of age-i-encoded PI3-kinase, which
generates PtdIns-3,4,5-P3, whereas the DAF-18 PTEN phosphatase homolog
degrades this second messenger. PtdIns-3,4,5-P 3 and its derivative PtdIns-3,4P2 lead to the activation of PKB-family kinases, encoded by akt-1, akt-2, and
sgk-1 . Phosphorylation by PKB blocks the activity of the daf-16 gene
product (forkhead / winged helix transcription factor) by phosphorylation,
which leads to DAF-16 exclusion from the nucleus. This translocation event
prevents activation of the daf-16 target genes, some of which function in
stress-resistance and lifespan determination.
49
Figure 3.
INSULIN / IGF
DAF-2
Insulin-like Receptor
AGE-1
PI3K
I
PI(3,4,5)P 3
AKT- 1/AKT-2 /SGK- 1
PKB/SGK
Longevity
Stress Resistance
50
DAF-18
PTEN
Chapter 2.
C. elegans SIR-2.1 interacts with 14-3-3 proteins to activate DAF-16
and extend lifespan
Ala Berdichevsky, Mohan Viswanathan,
H. Robert Horvitz and Leonard Guarente
This chapter is modified from the following manuscript:
Berdichevsky, A., Viswanathan, M., Horvitz, H.R., and Guarente, L. 2006.
C. elegans SIR-2.1 interacts with 14-3-3 proteins to activate DAF-16 and
extend life span. Cell 125(6): 1165-1177
Mohan Viswanathan contributed anti-SIR-2.1 polyclonal serum that I
affinity purified, and made and initially characterized the sir-2.1overexpressing strain NL3909. We analyzed the NL3909 transcriptional
profiles together.
SUMMARY
The longevity of Caenorhabditiselegans is promoted by extra copies of
the sir-2.1 gene in a manner dependent on the forkhead transcription
factor DAF-16. We identify two C. elegans 14-3-3 proteins as SIR-2.1
binding partners and show that 14-3-3 genes are required for the
lifespan extension conferred by extra copies of sir-2.1. 14-3-3 proteins
also are required for SIR-2.1-induced transcriptional activation of
DAF-16 and stress resistance. Following heat stress, SIR-2.1 can bind
DAF-16 in a 14-3-3-dependent manner. By contrast, low insulin-like
signaling does not promote SIR-2.1 / DAF-16 interaction, and sir-2.1
and the 14-3-3 genes are not required for the regulation of lifespan by
the insulin-like signaling pathway. We propose the existence of a stressdependent pathway in which SIR-2.1 and 14-3-3 act in parallel to the
insulin-like pathway to activate DAF-16 and extend lifespan.
52
INTRODUCTION
C. elegans lifespan is controlled by an insulin/IGF signaling
pathway, which includes the DAF-2 transmembrane receptor, a series of
intracellular kinases, and the DAF-16 forkhead family transcription factor
(Guarente and Kenyon, 2000; Finch and Ruvkun, 2001; Kenyon, 2005).
This insulin/IGF pathway regulates the generation of dauer larvae, which
are long-lived developmental variants that arise from second-stage (L2)
larvae when conditions are harsh (Larsen et al., 1995; Vanfleteren and
Braeckman, 1999). Mutations that decrease signaling by the DAF-2
receptor pathway cause an extension of the lifespan of adults (and more
frequent entry into the dauer developmental state by larvae) by triggering
the nuclear localization of DAF-16 (Henderson and Johnson, 2001; Lee et
al., 2001; Lin et al., 2001). Such mutations also increase the resistance of
animals to oxidative and genotoxic stress. The longevity and stress
resistance engendered by a reduction in insulin/IGF signaling is abolished
in mutants defective in daf-16 (Murakami and Johnson, 1996; Lin et al.,
1997; Paradis and Ruvkun, 1998; Honda and Honda, 1999). The forkhead
family of transcription factors includes mammalian FOXO 1, 3, and 4. In
mammals, activation of the insulin/IGF pathway causes phosphorylation of
forkhead proteins and their retention in the cytoplasm (Brunet et al., 1999;
Nakae et al., 1999; Tang et al., 1999). Nuclear localization of forkhead
proteins can be elicited by depletion of insulin from serum or by stress
(Tran et al., 2003; Brunet et al., 2004; Essers et al., 2004; Van Der Heide et
al., 2004; van der Horst et al., 2004).
C. elegans sir-2.1 and its orthologs in Sacharomyces cerevisiae and
Drosophilamelanogastercan regulate lifespan (Kaeberlein et al., 1999;
Tissenbaum and Guarente, 2001; Rogina and Helfand, 2004). Transgenic
worms with extra copies of sir-2.1 live longer than the wild type. Because
this longevity requires daf-16, it was proposed that sir-2.1 works by downregulating insulin signaling (Tissenbaum and Guarente, 2001). This genetic
inference was bolstered by the finding that transgenes overexpressing sir2.1 conferred no further extension in lifespan to daf-2 mutants (Tissenbaum
and Guarente, 2001). Nevertheless, the mechanistic link between sir-2.I1
and the insulin/IGF pathway and its target daf-16 is unknown.
The proteins encoded by yeast SIR2 and its orthologs are NADdependent deacetylases (Blander and Guarente, 2004). The mammalian
homolog Sirtl interacts with many proteins, including transcription factors,
such as p53, PPARO, and NF- OB, and transcriptional cofactors, such as
p300 and CBP (Luo et al., 2001; Langley et al., 2002; Takata and Ishikawa,
2003; Picard et al., 2004; van der Horst et al., 2004; Vaquero et al., 2004;
Yeung et al., 2004; Bouras et al., 2005). In mammals, Sirtl can bind to
and deacetylate forkhead proteins. Deacetylation of forkhead transcription
factors by Sirt 1 can result in either repression or activation of the
transcription of their target genes (Brunet et al., 2004; Motta et al., 2004;
Bouras et al., 2005) and appears to increase the resistance of mammalian
cells to DNA damage-induced apoptosis (Luo et al., 2001; Brunet et al.,
2004; Cohen et al., 2004; Motta et al., 2004).
14-3-3 proteins are highly conserved small acidic proteins that bind
phosphoserine and phosphothreonine residues in a context-specific manner
54
(Durocher et al., 2000). Through interactions with their partners, 14-3-3
proteins regulate key biological processes, such as the cell cycle, apoptosis,
and transcription (Tzivion et al., 2001). 14-3-3 proteins affect their targets
by multiple mechanisms, including by activating or inhibiting intrinsic
protein activities, by altering subcellular protein localization, and by
scaffolding interactions between two binding partners. In mammalian cells,
14-3-3 proteins bind to phosphorylated forkhead (Brunet et al., 1999;
Durocher et al., 2000; Obsil et al., 2003), and this binding leads to retention
of forkhead in the cytoplasm, rendering it inactive. Mammalian 14-3-3
proteins can also bind the C. elegans forkhead protein DAF-16 when it is
phosphorylated (Cahill et al., 2001). In C. elegans, two genes, par-5 andftt2, encode 14-3-3-like proteins, both of which are enriched in the embryo
and in the gonad (Wang and Shakes, 1997; Morton et al., 2002). Mutations
in par-5 affect the asymmetry of early embryonic cell divisions and cause
maternal-effect lethality (Morton et al., 2002). ftt-2 is expressed in the soma
(Wang and Shakes, 1997); the effect of the loss offtt-2 function has not
been described.
In this report, we show that SIR-2.1 acts with 14-3-3 proteins to
affect DAF-16 activity and lifespan. Our findings suggest that SIR-2.1 and
14-3-3 may act in parallel to the insulin-like pathway to regulate DAF-16
and extend lifespan.
RESULTS
C. elegans SIR-2.1 interacts with 14-3-3 proteins.
To identify SIR-2.1-interacting proteins in C. elegans, we raised a
polyclonal antibody that recognizes worm SIR-2. 1. Immunoflurescence
studies using this antibody showed nuclear staining in wild-type animals
(Figure 1A, left), consistent with the reported nuclear localization of the
mammalian homolog Sirtl (Langley et al., 2002). No staining was seen in
sir-2.1(ok434) mutants, which are deleted for the sir-2.I1gene (Figure 1A,
right), indicating that this antibody is specific to SIR-2.1. We isolated SIR2.1 protein complexes by co-precipitation with the anti-SIR-2.1 antibody
and analyzed the complexes by mass spectrometry. We assayed lysates of
three different strains: the wild type (N2), sir-2.I (ok434A) (which does not
produce full length SIR-2. 1 protein), and a transgenic sir-2.I1overexpresser
(O/E). We detected SIR-2.1 and one additional protein in the sir-2.I (O/E)
lysate (Figure 1B). This protein was subjected to mass spectrometry, which
revealed that it contained both C. elegans 14-3-3-like proteins, PAR-5 and
FTT-2, and no other proteins (Figure IC). PAR-5 and FTT-2 share 88%
sequence identity and are 78% (PAR-5) and 82% (FTT-2) identical with
human 14-3-3ý.
To address whether SIR-2.1/14-3-3 interaction also occurs in wildtype animals with endogenous levels of SIR-2. 1, we repeated the anti-SIR2.1 immuno-precipitations and analyzed the precipitates by western blotting
using antibody raised against mouse 14-3-3 recognizing both C. elegans 143-3 proteins (Figure ID). We detected 14-3-3 immunoreactivity in SIR-2.1
56
precipitates isolated from both the wild type and the SIR-2.1overexpressing strain. The precipitation of another C. elegans deacetylase
protein, HDA-1, from wild-type worm lysates did not contain 14-3-3,
indicating that the interaction is specific to SIR-2.1 (see Figure 2A).
14-3-3 proteins in C. elegans are mostly localized to the cytoplasm
(Wang and Shakes, 1997). In mammals, 14-3-3 proteins are known to
shuttle between the cytoplasm and the nucleus (Brunet et al., 2002; Van Der
Heide et al., 2004). To see whether 14-3-3 proteins are present in C. elegans
nuclei, we performed cell fractionation experiments. While we detected
SIR-2.1 protein in the nuclear fraction but not in the cytosolic fraction, 143-3 proteins were present in both nuclear and cytosolic fractions (Figure
lE). These observations are consistent with the hypothesis that the
interaction between SIR-2.1 and 14-3-3 proteins occurs in the nucleus.
C. elegans 14-3-3 proteins can interact with DAF-16 and act to retain
DAF-16::GFP in the cytoplasm.
Since mammalian 14-3-3 proteins bind phosphorylated FOXO
proteins (the homologs of DAF-16) and control their subcellular
localization (Brunet et al., 1999; Brunet et al., 2002; Cahill et al., 2001;
Obsil et al., 2003), we investigated whether C. elegans 14-3-3 proteins can
bind DAF- 16 and affect the localization of DAF-16.
We immunoprecipitated a functional DAF-16::GFP translational
fusion protein (Lee et al., 2001) from worm lysates using an anti-GFP
antibody and observed that 14-3-3 proteins co-immunoprecipitated with
DAF-16::GFP. The interaction between 14-3-3 proteins and DAF-16::GFP
57
was likely specific to DAF-16 since 14-3-3 proteins did not co-precipitate
with a LIN-53::GFP translational fusion protein in a similar experiment
(Figure 2B). Thus 14-3-3 proteins can physically interact with DAF- 16,
similar to the interaction between mammalian 14-3-3 and the DAF- 16
homologue FOXO.
We next investigated whether low 14-3-3 levels will affect the
subcellular localization of DAF-16. In this experiment, we used RNAmediated gene interference (RNAi) to reduce the expression of the 14-3-3
genes. We constructed four non-overlapping RNAi clones, two targeting
each of the two 14-3-3 genes, par-5 andftt-2 (see Experimental
Procedures). Progeny of animals fed with bacteria expressing par-5 RNAi
expressed a maternal-effect lethal phenotype similar to that previously
described forpar-5 loss-of-function (If) mutants (Morton et al., 2002). The
progeny of animals fed withftt-2 RNAi frequently retained eggs and died of
internally hatched progeny, a phenotype similar to that offtt-2(lf) mutants
(see below). Feedingpar-5 orftt-2 RNAi did not cause any apparent
abnormalities in animals that were transferred to RNAi plates as L4 larvae
or as adults. At the protein level, par-5(RNAi) had a minimal effect on the
level of total 14-3-3 proteins (data not shown), whereas ftt-2(RNAi) caused
substantial reduction in 14-3-3 protein level (see Figure 5B below). It is
important to note that the two C. elegans 14-3-3 genes are very similar, so it
is possible that RNAi of one gene would result in reduction of function of
the other. Nevertheless, we believe that at least some of the RNAi effects
are gene-specific, since par-5(RNAi) but notftt-2(RNAi) resulted in a
maternal-effect lethal phenotype similar to that caused by mutation in par5, andftt-2(RNAi) but not par-5(RNAi) resulted in a lower level of 14-3-3
proteins.
58
Under normal conditions, DAF-16::GFP has a diffuse mostly
cytoplasmic localization. Reducing the expression ofpar-5 orftt-2 or both
by RNAi caused a pronounced nuclear accumulation of DAF-16::GFP
(Figure 3A). Nuclear DAF-16::GFP localization was most prominent in the
intestinal cells but was also apparent in muscle and hypodermis (we scored
animals as positive for nuclear DAF-16::GFP if most or all intestinal cells
showed nuclear GFP). We did not see GFP accumulation in neuronal
nuclei, perhaps because of the previously described insensitivity of this
tissue to RNAi (Timmons et al., 2001). RNAi targeting offtt-2 led to
nuclear DAF-16::GFP in 68% or 100% of animals, depending on the RNAi
clone used, while par-5 RNAi caused nuclear DAF-16::GFP in about 50%
of animals (Table 1). The par-5 RNAi result suggests that this biological
assay of DAF-16::GFP localization is likely more sensitive than western
blots, in which we could not detect a significant reduction in 14-3-3 level
following par-5(RNAi) feeding. In addition to altering the localization of
DAF-16::GFP, ftt-2(RNAi) but not par-5(RNAi) caused arrest at the L2-L3
larval stages of the progeny of daf-16::gfp animals. Larval arrest was not
observed when wild-type animals were subjected toftt-2(RNAi) and may
have been a consequence of the increased nuclear localization of overexpressed DAF-16::GFP.
To confirm that loss offtt-2 function can alter DAF-16 localization,
we generated anftt-2 mutant strain by screening a library of chemically
mutagenized animals using PCR (Jansen et al., 1997; Liu et al., 1999).
ftt-2(n4426A) worms bear a 668 bp deletion that removes part of the
promoter and the start codon of the predictedftt-2 gene and therefore is a
good candidate for being a null allele offtt-2. These mutants develop
normally but die in early adulthood, because of bursting at the vulva or
internal hatching of progeny (data not shown). We introduced the daf16::gfp transgene into theftt-2(n4426A) strain. Larvae homozygous forftt2(A) had nuclear DAF- 16::GFP, similar to animals treated withftt-2(RNAi)
(Figure 3B), indicating thatftt-2 function may be required for the proper
subcellular localization of DAF-16. These experiments suggest that, as
proposed for the homologous proteins in mammals, C. elegans 14-3-3
proteins can act to retain the forkhead protein DAF-16 in the cytoplasm.
sir-2.1 overexpression slows DAF-16::GFP exit from the nucleus during
recovery from stress.
We next asked whether sir-2.1 levels could also affect the cellular
localization of DAF- 16. We crossed the daf-1 6::gfp transgene into two
different long-lived strains over-expressing sir-2.1, geln3, and pkIsl 642.
The DAF-16::GFP localization in these strains was cytoplasmic, as in the
control strain with wild-type levels of sir-2.1 (data not shown). Therefore
we concluded that under normal conditions SIR-2.1 overexpression does
not affect DAF-16 localization.
DAF-16 is known to translocate into the nucleus after heat shock. To
investigate a possible more subtle effect of SIR-2.1 on DAF-16 localization,
we studied the kinetics of stress-induced nuclear localization in the strain
with extra copies of sir-2.1. We found that in a wild type background and in
a strain over-expressing SIR-2. 1, DAF-16::GFP entered the nucleus with
similar kinetics, such that within 30 min. of a shift to 370 C over 80% of
animals in both strains had nuclear DAF-16::GFP (see Figure 4). The
kinetics of relocalization of DAF-16::GFP to the cytoplasm, however,
differed: loss of nuclear localization occurred significantly more slowly in
60
the strain with the higher level of SIR-2. 1. For example, after 30 min. of
recovery, the over-expressing strain showed more than double the number
of worms that had cells with nuclear DAF- 16 (46% as compared to 18%;
Figure 4B). These experiments suggest that following stress increased SIR2.1 function can cause DAF-16 to remain in the nucleus, and therefore
promote the function of DAF-16 to mediate a cellular stress response.
The 14-3-3 genesftt-2 and par-5 are required for sir-2.1-mediated
lifespan extension.
As FTT-2 and PAR-5 interact with SIR-2.1 and affect DAF-16
localization, we asked whether 14-3-3 genes regulate C. elegans lifespan.
We used RNAi to reduce the levels of par-5 andftt-2 in adults and assayed
their lifespans. As controls, we measured lifespans of animals fed bacteria
expressing a control RNAi vector and animals fed daf-16 RNAi. RNAi
reduction ofpar-5,ftt-2 and daf-16 had similar effects on the lifespans of
worms with wild-type levels of sir-2.1: they resulted in a shortening of
lifespan by approximately 20% compared to the vector control (see Figure
6A).
Next we asked whether downregulation of 14-3-3 genes would affect
lifespan in long-lived animals with extra copies of sir-2.1. The sir-2.I1gene
R1 1A8.4 is predicted to be a downstream gene of a two-gene operon
(CEOP4372). To overexpress sir-2.1 with its regulatory regions we made a
new transgenic strain NL3909 pkIs1642 by microparticle bombardment.
The construct in the NL3909 strain contains the entire operon, including the
sir-2.1 coding sequence plus 2.5 kb of sequence upstream of sir-2.1
containing the gene R11A8.5, predicted to encode a glutathione S-
transferase-related protein, as well as 600 bases further upstream. Animals
carrying pkls 1642 transgene showed marked extension of lifespan relative
to the wild type (Figure 5A). Control experiments indicated that
overexpression of sir-2.1 but not of R11A8.5 is required for lifespan
extension ofpklsl642 transgenic animals (see Experimental Procedures and
Figure 5A). Overexpression of sir-2.1 inpkls1642 strain also promoted
increased resistance to heat shock (Figure 5B). Reduction of 14-3-3 levels
in animals with extra copies of sir-2.1 fully suppressed the lifespan
extension observed in this strain (Figure 6B). Whileftt-2 RNAi resulted in a
shortened lifespan indistinguishable from the effect of daf-16 RNAi, par-5
RNAi suppressed the lifespan of the sir-2.1 overexpresser to a wild-type
duration. Table 2 details the data for these and all other lifespan
experiments presented in Figure 6.
To address the possibility that the worms fed with the par-5 orftt-2
RNAi bacteria have non-specifically shortened lifespans because they are
sick, we tested the effects ofpar-5 andftt-2 RNAi on the lifespans of sir2.1 (ok434A) and daf-1 6(lf) mutants, both of which have slightly shorter
lifespans than the wild type. In these strains reducing 14-3-3 gene
expression caused little or no reduction in lifespan (see Figures 6C and 6D).
More strikingly, reducing 14-3-3 gene expression in a daf-2 mutant, which
like the sir-2.I1-overexpressing strain displays an increased lifespan, did not
shorten lifespan (see below). Thus, the effects ofpar-5 and ftt-2 RNAi on
lifespan probably are not caused by non-specific sickness. Rather, par-5
andftt-2 appear to be specifically required for the sir-2.1-dependent
lifespan extension.
Our findings indicate that reduction of eitherpar-5 orftt-2 triggered
nuclear localization of DAF-16 but did not promote longevity. This
62
observation suggests that nuclear localization of DAF- 16 may not be
sufficient to activate target genes and extend lifespan.
sir-2.1, par-5 and ftt-2 are not required for lifespan extension in animals
with reduced insulin/IGF signaling.
C. elegans sir-2.1 has been suggested to affect lifespan via the
insulin-like pathway (Tissenbaum and Guarente, 2001). To analyze the
interactions of the 14-3-3 genes with the insulin-like pathway, we tested the
effects of reducing par-5 andftt-2 in a daf-2 mutant, which has extended
lifespan. In this experiment, the lifespan extension of daf-2(el370) mutants
was fully suppressed by daf-16 RNAi, as expected. By contrast, par-5
RNAi andftt-2 RNAi had no effect (Figure 6E). Likewise, the loss of sir2.1 did not reduce the long lifespan of the daf-2 mutant and may have
lengthened it slightly (Figure 6F). These findings indicate that sir-2.1, par-5
andftt-2 do not function in the insulin-like pathway downstream of daf-2.
Instead, sir-2.1 and the 14-3-3 genes may act upstream of the DAF-2
insulin-like receptor (for instance, by controlling production of its ligands)
or in a pathway of lifespan determination parallel to that of insulin
signaling. Since daf-16 is required for the lifespan extension by extra copies
of sir-2.1, any parallel pathway must converge on the DAF- 16 transcription
factor.
sir-2.1 can promote resistance to stress in anftt-2-dependent way.
DAF- 16 function is important for the stress response (Hsu et al.,
2003; Lamitina and Strange, 2005; Lin et al., 1997). As SIR-2.1
63
overexpression led to extension of lifespan in a daf-16- andftt-2-dependent
manner, we asked whetherftt-2 might also be involved in stress resistance
caused by extra copies of sir-2.1. We examined the effects of sir-2.I1andftt2 on resistance to heat stress, by determining the survival of animals
subjected to 320 C heat shock (Figure 7A), and to oxidative stress, by
assaying development in 0.25 mM paraquat (Figure 7B). In both cases, the
sir-2.1 overexpressing strain was stress-resistant compared to the wild type,
and the sir-2.1 deletion strain was stress-sensitive (Figure 7A and B).
Moreover, anftt-2 loss-of-function mutation also caused sensitivity to stress
and completely abolished the stress resistance of the sir-2.1-overexpressing
strain (Figure 7A and B). These findings show that sir-2.1 and 14-3-3
function together in determining stress resistance.
sir-2.1 can promote daf-16-dependent transcription in anftt-2dependent fashion.
The genetic interactions between sir-2.I1and daf-16 led us to
speculate that sir-2.I1may activate daf-16. To test this hypothesis we used a
GFP reporter for a known daf-16 target, sod-3 (Cahill et al., 2001; Honda
and Honda, 1999; Lee et al., 2003; McElwee et al., 2003). This sod-3::gfp
reporter, which has been used to assay daf-16 activity (Libina et al., 2003),
was crossed into strains with high-copy-number (geln3) or low-copynumber (pkls1642) transgenes carrying extra copies of sir-2.I1,and the sod3::gfp signal in these strains was visualized by fluorescence microscopy
(Figure 7C-E). In both SIR-2.1 overexpressing strains, the GFP signal was
much stronger than that in the wild type, indicating that transcription from
the sod-3 promoter was upregulated. A loss-of-function mutation in daf-16
64
eliminated this upregulation of sod-3 (Figure 7G, H), suggesting that it
results from the activation of DAF- 16 by the elevated levels of SIR-2. 1.
Transcriptional array data also indicated an elevated expression of sod-3
and other DAF-16 targets in animals overexpressing SIR-2.1 (Viswanathan
et al., 2005 and Table 3).
Next we tested whether this activation of DAF-16 by SIR-2. 1 would
be influenced by reducing the levels of 14-3-3. Animals bearing a deletion
inftt-2 did not show upregulation of sod-3::gfp expression (Figure 7J),
although these mutants displayed a nuclear localization of DAF-16::GFP
(Figure 3B). However, the upregulation of sod-3 by the SIR-2.1overexpressing transgene was abolished by deletion offtt-2 (Figure 7K).
Moreover, the requirement forftt-2 was specific for the activation of DAF16 by SIR-2.1 overexpression, as the ftt-2 deletion did not reduce the high
sod-3.:.gfp expression seen in the daf-2(e3 70) mutant (see Figure 7F and
I). In summary, overexpression of SIR-2.1 appears to promote DAF-16
transcriptional activity and requires 14-3-3 to do so.
Following stress, SIR-2.1 interacts with DAF-16 in a 14-3-3-dependent
manner.
In mammals, both 14-3-3 proteins and a SIR-2.1 homolog Sirt1 can
physically interact with mammalian FOXO proteins (Brunet et al., 1999;
Brunet et al., 2002; Obsil et al., 2003; Brunet et al., 2004; Motta et al.,
2004; van der Horst et al., 2004). To ask whether comparable interactions
occur among the related proteins of C. elegans, we used a functional DAF16::GFP (Lee et al., 2001) and antisera to GFP and SIR-2.1 for coimmunoprecipitation experiments and analyzed the precipitates using
western blots (Figure 8A). Since SIR-2.1 is a nuclear protein and DAF-16
is normally cytoplasmic (Henderson and Johnson, 2001; Lee et al., 2001;
Lin et al., 2001), we reasoned that any interaction would require nuclear
entry of DAF-16. Heat stress causes nuclear accumulation of DAF-16
(Henderson and Johnson, 2001; Lin et al., 2001). Thus, we
immunoprecipitated DAF-16::GFP or SIR-2.1 from untreated animals and
from animals that had been heat-shocked and assayed for coimmunoprecipitation of SIR-2.1 and DAF-16. As expected, no SIR2. 1/DAF- 16 complex was observed in the absence of heat shock. After heat
shock, SIR-2.1 co-immunoprecipitated with DAF- 16 (Figure 8A). Thus,
following stress DAF-16 can physically interact with SIR-2.1 proteins in C.
elegans. Notably, 14-3-3 was found in the DAF-16::GFP
immunoprecipitate before and after heat shock and not in that of the control
animals lacking DAF-16::GFP (Figure 8A and 2B).
One of the known biological roles of 14-3-3 proteins is to act as a
scaffold, promoting interactions between two different proteins (Tzivion et
al., 2001). We asked whether 14-3-3 proteins were important for the
interaction between SIR-2.1 and DAF-16.
We immunoprecipitated DAF-
16::GFP protein from lysates of animals grown onftt-2(RNAi) or vector
RNAi and subjected to heat shock and assessed the levels of SIR-2.1 in the
precipitates. ftt-2(RNAi) markedly reduced the amount of SIR-2.1 that coprecipitated with DAF-16 (Figure 8B). par-5(RNAi) had no effect on the
level of 14-3-3 proteins in worms or the co-immunoprecipitation of SIR-2.1
with DAF-16 (data not shown). We conclude that following heat stress SIR2.1 interacts with DAF- 16 in a 14-3-3-dependent manner.
We tested whether reducing insulin-like signaling, which should
cause accumulation of dephosphorylated nuclear DAF-16, will affect
interactions between DAF-16, 14-3-3 and SIR-2.1 proteins. RNAi reduction
of daf-2 results in decreased insulin-like signaling leading to lifespan
extension and accumulation of DAF-16::GFP in the nucleus ((Henderson
and Johnson, 2001; Murphy et al., 2003) and Figure 9). However, unlike
heat shock, daf-2(RNAi) did not promote the interaction between SIR-2.1
and DAF-16 (Fig. 8B). After heat shock of the daf-2 RNAi treated worms, a
low level of DAF-16/SIR-2.1 interaction was observed, which may be due
to a fraction of DAF-16 that was not affected by daf-2 RNAi. These
findings suggest that DAF-16 phosphorylation and interaction with 14-3-3
may be important for the formation of SIR-2.1/ DAF-16 complex.
DISCUSSION
C. elegans SIR-2.1 and 14-3-3 proteins can function in a stress-response
pathway of longevity.
The longevity of C. elegans is increased by extra copies of the sir-2.I1
gene, and this effect is dependent on daf-16 (Tissenbaum and Guarente,
2001). Here we provide evidence for a mechanistic basis of this genetic
interaction: SIR-2. 1 and DAF-16 proteins can physically interact. We found
that SIR-2.1 functions in stress responses, as worms overexpressing SIR-2.1
were resistant to heat and oxidative stresses and sir-2.I1loss of function
mutants were stress-sensitive. We also found that SIR-2.1 can activate
DAF-16-dependent transcription, as overexpression of SIR-2.1 increased
transcription of the DAF-16 target gene sod-3 in a daf-16-dependent way.
Moreover, we found that 14-3-3 proteins are required for the lifespan
extension, stress resistance and sod-3 activation conferred by extra copies
67
of sir-2.1. Deletion offtt-2 prevented stress resistance and sod-3 activation
by SIR-2.1 overexpression, and RNAi of either par-5 orftt-2 suppressed the
SIR-2.1-mediated lifespan extension. We are not certain that par-5
functions in the lifespan extension caused by SIR-2.1 overexpression,
because RNAi ofpar-5 may also targetftt-2.
Mammalian 14-3-3 proteins can interact with phosphorylated FOXO
proteins and cause their retention in the cytoplasm (Brunet et al., 1999;
Durocher et al., 2000; Obsil et al., 2003). Insulin signaling promotes this
retention by causing phosphorylation of FOXO proteins. Here we
demonstrate that 14-3-3 proteins appear to be important for the cytoplasmic
localization of DAF-16, since reducing the expression levels of par-5 orftt2 results in nuclear accumulation of DAF-16::GFP. Thus, C. elegans 14-3-3
proteins can act to retain the C. elegans forkhead protein DAF-16 in the
cytoplasm, similar to their mammalian homologs.
Our data thus suggest that two different 14-3-3 protein complexes are
present in C. elegans: a DAF-16/14-3-3 complex in the cytoplasm and a
SIR-2.1/14-3-3 complex in the nucleus. Consistent with this hypothesis, we
could not detect an interaction between DAF-16 and SIR-2. 1 under normal
conditions. However, heat shock, which is known to promote DAF-16
localization to the nucleus (Henderson and Johnson, 2001; Lin et al., 2001),
triggered the formation of a SIR-2. 1/DAF-16 complex. Both
phosphorylation of DAF- 16 at the AKT sites and FTT-2 appear to be
required for this interaction between DAF- 16 and SIR-2. 1, since RNAi of
daf-2 orftt-2 reduced their association.
In summary, 14-3-3 proteins are required for the daf-16-dependent
biological effects of SIR-2.1 as well as for the physical association of SIR2.1 and DAF-16. Our results suggest that 14-3-3 genes and sir-2.1 act
together in a stress-mediated pathway of longevity. Even under normal
conditions animals may experience a basal level of stress, which may
explain the effects of SIR-2.1 and 14-3-3 in an absence of a specific stress
treatment. Our data indicate that 14-3-3 proteins play two antagonistic roles
in DAF-16 regulation. First, as previously described in mammals, 14-3-3
proteins bind to phosphorylated DAF-16 and promote its retention in the
cytoplasm. Second, 14-3-3 proteins function in a previously unknown
pathway to facilitate the association of DAF- 16 with SIR-2.1 in the nucleus
to elicit activation of sod-3, stress resistance and extension of lifespan. A
recent report suggests that the C. elegans P3-Catenin ortholog bar-i can
activate DAF-16 following stress (Essers et al., 2005). It will be interesting
to see if bar-i participates in the 14-3-3/SIR-2. 1-mediated mechanism of
DAF- 16 regulation.
SIR-2.1 and 14-3-3 may affect longevity by regulating DAF-16
Forkhead in parallel to the insulin-like pathway.
It has been suggested that sir-2.I1promotes longevity by
downregulating the insulin-like signaling pathway (Tissenbaum and
Guarente, 2001). We observed that sir-2.1, par-5 andftt-2 were not required
for lifespan extension in a daf-2 mutant. Furthermore,ftt-2 was not required
for the activation of sod-3 by a reduction in insulin-like signaling, even
thoughftt-2 was required for activation of sod-3 by SIR-2.1 overexpression.
Our findings therefore indicate that sir-2.1 andftt-2 do not act in the
insulin/IGF pathway downstream of DAF-2. We cannot rule out the
possibility that SIR-2.1 functions upstream of the DAF-2 receptor, for
example, by regulating insulin production. However, the stress-inducible
69
14-3-3-dependent physical interaction between SIR-2.1 and DAF-16
suggests that SIR-2.1 and 14-3-3 act in a pathway parallel to insulin-like
signaling that like the insulin-like pathway converges on the DAF-16
transcription factor.
A model for the roles of SIR-2.1 and 14-3-3 in DAF-16 activation and
longevity.
Our results suggest the following model for the role of SIR-2.1 in
DAF-16 activation (Figure 10). Under normal conditions, DAF-16 is mostly
inactive, present in the cytoplasm and bound by 14-3-3 proteins. A
reduction in insulin/IGF signaling, for example in daf-2 mutants, renders
DAF-16 nuclear by reducing its state of phosphorylation by AKT kinase,
thereby dissociating it from 14-3-3. Nuclear localization of DAF-16 is not
sufficient for activation of its target genes, since reducing expression offtt2 rendered DAF-16 nuclear but did not activate the DAF-16 target sod-3 or
extend lifespan. That nuclear DAF-16 is not necessarily active is consistent
with the finding that overexpression of DAF-16 mutated at all known AKT
sites leads to constitutively nuclear DAF-16 localization but does not
extend lifespan (Lin et al., 2001). It is possible that DAF-16 requires an
additional cofactor that acts with DAF- 16 to regulate transcription when
insulin/IGF signaling is reduced. Alternatively, the nuclear DAF-16 that
accumulates in worms with lowftt-2 levels may be non-functional because
DAF-16 is insufficiently dephosphorylated.
In response to stress, forkhead proteins translocate to the nucleus
(Brunet et al., 2004; Henderson and Johnson, 2001; Lin et al., 2001; Tran et
al., 2003) in a process that may involve the JNK kinase signaling pathway
(Oh et al., 2005). We propose that SIR-2.1 binds the nuclear DAF-16
produced by stress, but not the nuclear DAF-16 produced by low insulinlike signaling. The resulting SIR-2.1/DAF-16 complex promotes DAF-16dependent transcription, stress resistance and longevity, in a manner
dependent on 14-3-3 proteins. One possibility is that 14-3-3 scaffolds the
interaction between SIR-2.1 and DAF-16 and a ternary complex among
SIR-2.1, DAF-16 and 14-3-3 participates in the transcriptional activation of
DAF-16 target genes. Another possibility is that 14-3-3 mediates a
modification of DAF-16 and/or SIR-2.1 following stress, triggering the
association of a binary complex of DAF-16 and SIR-2.1. This alternative
model requires two additional steps involving an unknown 14-3-3-mediated
modification of either SIR-2.1 or DAF-16 and dissociation of 14-3-3 from
both SIR-2.1 and DAF- 16.
Our proposed activation of FOXO proteins by a pathway involving
SIR-2.1 and 14-3-3 may be a general molecular mechanism for the
regulation of longevity by Sir2 orthologs in metazoa.
EXPERIMENTAL PROCEDURES
Antibody production, immunofluorescence, and cell fractionations
Rabbit polyclonal anti-SIR-2.1 antiserum was raised against purified fulllength His-tagged SIR-2.1 protein and affinity-purified using His-tagged
SIR-2.1 bound to nitrocellulose membranes. The mouse monoclonal
antibody to GFP was anti-AFP mAb 3E6 (QbioGene). The anti-DAF-16
rabbit polyclonal antibody was a gift from G. Ruvkun. The mouse
monoclonal anti-14-3-3 1(H8) antibody was purchased from Santa Cruz
Biotechnology. To control for total protein content in loading we used the
anti-aTubulin mouse monoclonal antibody Dmla (Sigma). HDA-1 rabbit
polyclonal antibody was purchased from Santa Cruz.
Whole-mount immunofluorescence of larvae and adult animals was
performed as described (Finney and Ruvkun, 1990) with purified anti-SIR2.1 antibody and visualized by Nomarski fluorescence microscopy.
Cell fractionations were performed as described (Chen et al., 2000). Celamin antibody was from Y. Gruenbaum (Liu et al., 2000). SQV-4 antibody
was obtained from Ho-Yon Hwang (Hwang and Horvitz, 2002).
Immunoprecipitation and western blots
Worm lysates for immunoprecipitation (IP) were made by sonication of
pelleted worms in 5 volumes of IP buffer (10 mM Hepes-KOH, 250 mM
NaC1, 0.5% NP-40, 10% glycerol, 5 mM EDTA, 1 mM DTT, Protease
Inhibitors: one Complete Mini tablet (Roche Biochemical) per 10 ml IP
buffer), for three 10 sec bursts on ice, and the lysates were centrifuged at
10,000 rpm for 20 min at 40 C. The supernatant was saved, and total protein
concentration was determined using a BioRad Dc Protein Assay kit. For
mass spectrometry analysis, 10 mg of total protein was used per IP. For
western blot analysis, 1-3 mg of total protein was used per IP. Input lanes
contained 20-50 [tg of total protein.
Immunoprecipitations were performed by first binding the antibody to
Amersham fast-flow Protein A beads for 2 hr at 40 C, followed by washes
with IP buffer, and incubation with worm lysates at 40 C. Precipitated
proteins were eluted by boiling in SDS protein sample buffer. The samples
were then centrifuged 5 min at 13,000 rpm to separate Protein A beads, and
the supernatants were loaded onto 4-15% gradient polyacrylamide gels
72
(BioRad). For mass spectrometry analysis, gels were stained with
Coomassie blue, and bands were excised and sent to the Taplin Biological
Mass Spectrometry Unit at Harvard Medical School for analysis.
For western blot analysis, proteins were transferred to nitrocellulose
membranes. Membranes were blocked with 5% milk in PBST (20 mM
sodium phosphate pH 7.2, 150 mM NaCl, 0.01% Tween-20) for 30 min at
room temperature, and then incubated with the indicated primary antibodies
in PBST for 60 min at room temperature or overnight at 4VC. The
membranes were then incubated with anti-mouse and anti-rabbit
horseradish peroxidase-conjugated secondary antibodies that were
purchased from Bio-Rad, and the signal was detected using Western
Lightning Chemiluminescence reagent (Perkin Elmer).
Lifespan, stress-resistance and RNAi analyses
Unless stated otherwise, lifespan assays were performed at 200 C and
initiated by transferring L4 larvae of the indicated genotypes to plates
containing 10 [tM fluorodeoxyuridine (FUDR). Unless indicated otherwise,
all RNAi and lifespan experiments were performed using N-terminal par-5
andftt-2 RNAi clones. In many cases, the plates contained Carbenicillin,
Tetracyclin, and IPTG, which are required for the standard RNAi feeding
protocol (Kamath et al., 2003). We used Prism 4 software to perform
statistical analyses of survival curves. Animals that crawled off the plate,
burst at the vulva or died from internal hatching of progeny were excluded
from the analyses. In all lifespan assays reported here, between 90% and
95% of animals died naturally and were scored.
For RNAi analysis of DAF-16::GFP localization, L4 animals were
transferred to RNAi plates containing bacteria induced to express an RNAi
clone or carrying empty RNAi feeding vector as a control (Kamath et al.,
2003). The phenotype of progeny was scored after 48-72 hr at 20 0 C.
Heat-shock assays were performed at 320 C using one-day-old adults.
For each genotype, 50-70 animals were transferred to 5 cm plates, and
viability was scored every two to eight hrs after shift to 32TC. Statistical
analysis was performed using Prism 4 software, as for the lifespan assays.
Assays of paraquat sensitivity were performed at 200 C by transferring
arrested Li larvae to plates containing 0.25 mM paraquat. Development
was assessed 2, 3, and 4 days later, and the percent of animals that
developed to adulthood was scored. All strains assayed contained
muIs84[sod-3::gfp]in the background; expression of this transcriptional
reporter did not alter stress resistance (data not shown). The student t-test
was used to evaluate the significance of differences in the means.
Strains
Standard nematode growth medium (NGM) (Brenner, 1974) was
used for C. elegans growth and maintenance at 20 0 C. Unless otherwise
stated, plates were seeded with E. coli OP50 bacteria (Brenner, 1974).
NL3909 pkIsl642 [unc-119 sir-2.1] is a low-copy sir-2.I1transgenic
overexpresser strain produced by microparticle bombardment. The
transgene inpkIs1642 includes sir-2.1 coding sequence plus 2.5 kb of
upstream sequence, including the predicted gene R11A8.5, subcloned into a
pRP2510 vector (a gift of E. Ivanov). The construct was introduced into
unc-119(ed3) worms, and transgenic animals were selected based on the
unc-119(+) phenotype (coordinated locomotion and ability to form dauer
larvae). The presence of the transgene was verified by PCR. As a control,
the vector pRP2510, which contains the unc-119 gene alone, was
transformed into unc-119(ed3) animals, generating strain NL3908 pkls1641
[unc-119]. To test whether the effects on stress resistance and lifespan
observed in the NL3909 strain are indeed attributable to sir-2.1, we used
RNAi to reduce expression of each of sir-2.1 and R11A8.5 genes in the
pkls1642 background. Northern blot analysis confirmed that RNAi of both
genes is specific: that is, sir-2.I1RNAi reduced the level of the sir-2.I1
transcript but did not affect the level of R11A8.5 transcript; similarly,
R11A8.5 RNAi reduced the level of the R11A8.5 transcript but did not
affect sir-2.1 level. Then we performed lifespan and heat-stress assays of
pkls1642 transgenic animals fed control RNAi, sir-2.1 RNAi or R 11A8.5
RNAi bacteria. Both lifespan extension and stress resistance ofpkIsl642
animals were suppressed by sir-2.1 RNAi but were not significantly
affected by R11A8.5 RNAi (See Figure 4).
Strain VC199 sir-2.1(ok434) was obtained from the C. elegans Gene
Knockout Consortium via the Caenorhabditiselegans Genetics Center and
was outcrossed five times to the wild type prior to use.
ftt-2(n4426) was obtained by screening a deletion library. This
mutant has a 668 bp deletion of the promoter and most of the first exon of
the predictedftt-2 gene, from 22920 to 23598 on cosmid F52D10. The
strain was outcrossed six times to the wild type. Presence of the deletion
was verified by PCR.
The muls84[sod-3::gfp]transgene from strain CF 1553 was from the
CGC. The muls84[sod-3::gfp]transgene was crossed into the following
genetic backgrounds by standard genetic methods: geln3[sir-2.1 rol75
6(sul 006)], pkls1642[sir-2.1 unc-119], daf-16(mgDf50); pks1l642,
daf-16(mgDf50); geln3,ftt-2(n4426), pklsl642; ftt-2(n4426), daf-2(e1370);
ftt-2(n4426).
The daf-16::gfp transgene xrls87 [daf-16a::GFP::DAF-I6brol-
6(su1006)] was a gift from R. Lee and G. Ruvkun (Lee et al., 2001).
The daf-16::gfp transgene xrls87 [daf-]6a::GFP::DAF-16b rol-6(sul 006)]
was crossed into N2, geln3, and pkls1642 backgrounds by standard genetic
methods, and the presence of geln3 was verified by anti-SIR-2.1 western
blot. To generate daf-16::gfp;ftt-2(n4426) animals we crossed daf-16::gfp
transgenic males withftt-2(n4426) hermaphrodites, generatingftt-2 (+/A)
heterozygotes, and examined the progeny of theftt-2 (+/A) worms for DAF16::GFP localization. Two classes of GFP-expressing progeny were
observed: the first, constituting more than 85% of GFP-expressing progeny,
had cytoplasmic DAF-16::GFP and at least one wild-type copy offtt-2; the
second class had nuclear DAF-16::GFP in all cells. Worms of this second
class arrested as L1 or L2 larvae and were shown by single-animal PCR to
be homozygous for the ftt-2 deletion.
lin-53(n833) unc-76(e911); lin-15A(n767); nIs119[LIN-53::GFP]
was a gift from M. Harrison (personal communication).
Molecular biology
To constructpar-5 andftt-2 RNAi clones, the cDNAs of these genes were
amplified by RT-PCR from N2 RNA purified using Trizol (Invitrogen). To
construct the N-terminal RNAi clones for par-5 andftt-2, we used the
primers AAAATGGAAGATGGGGGTTC and
GGAATACCGCGTCAAGGTAA for par-5 and
AGAGTCAAGTTGTCACGGAGAAG and
76
CGATGTTTTCGTCCTATTTTCTG primers forftt-2. The cDNA
fragments were then digested with SacI endonuclease (New England
Biolabs), resulting in a 168-611 fragment for par-5 and a 275-614 fragment
forftt-2, and cloned into the SacI site of the pPD 129.36 RNAi feeding
vector. In addition, AGA GCT CGG ACT CGA CCT TGA TCA TGC and
AGA GCT CTT AGT TTC CAG CTT CCT GG, and AGA GCT CTG
ACG CCA TCG CTG AGC TCG and AGA GCT CAG TTG CCT CCC
TCT GTC TCG TTG GC primers were used to amplify a C-terminal
portion of the par-5 andftt-2 genes, respectively, from N2 genomic DNA
and to introduce a SacI site at the 5' and 3' ends of the amplified fragment.
SacI cleavage resulted in new non-overlapping fragments (639-747 for par5 and 614-745 forftt-2) that were cloned into a SacI site of pPD 129.36 to
obtain non-overlapping C-terminal RNAi clones for par-5 andftt-2. The
feeding RNAi strain for daf-16 was a gift of S. Lee and G. Ruvkun. The
RNAi clone for daf-2 was made by subcloning cDNA corresponding to
exon 9 of daf-2 into the pPD 129.36 RNAi feeding vector.
Acknowledgments
We thank the Caenorhabditiselegans Genetics Center for providing strains,
Gary Ruvkun for the daf-16:.:gfp strain, daf-16 RNAi clone and DAF-16
antiserum, Ronald Plasterk for help with constructing the low-copy sir-2.I1
overexpresser, Andrew Hellman for PCR-screening the deletion library to
identify theftt-2(n4426A) strain, Na An and Beth Castor for strain
maintenance and determining DNA sequences, respectively, members of
the Guarente and Horvitz laboratories for suggestions, and Nick Bishop, Gil
Blander, Brendan Galvin, Mike Hurwitz, Niels Ringstad and Hillel
Schwartz for critically reading the manuscript.
A.B. was supported by a graduate fellowship from the American Federation
for Aging Research. A.B. and H.R.H. were supported by the Ellison
Medical Foundation. H.R.H. is an Investigator of the Howard Hughes
Medical Institute. M.V. was supported by a fellowship from the Jane Coffin
Childs Memorial Fund for Medical Research. L.G. was supported by grants
from the NIH.
References
Blander, G., and Guarente, L. (2004). The Sir2 family of protein
deacetylases. Annu Rev Biochem 73, 417-435.
Bouras, T., Fu, M., Sauve, A. A., Wang, F., Quong, A. A., Perkins, N. D.,
Hay, R. T., Gu, W., and Pestell, R. G. (2005). SIRT1 deacetylation and
repression of P300 involves lysine residues 1020/1024 within the cell-cycle
regulatory domain 1. J Biol Chem. 280, 10264-10276
Brenner, S. (1974). The genetics of Caenorhabditiselegans. Genetics 77,
71-94.
Brunet, A., Bonni, A., Zigmond, M. J., Lin, M. Z., Juo, P., Hu, L. S.,
Anderson, M. J., Arden, K. C., Blenis, J., and Greenberg, M. E. (1999). Akt
promotes cell survival by phosphorylating and inhibiting a Forkhead
transcription factor. Cell 96, 857-868.
Brunet, A., Kanai, F., Stehn, J., Xu, J., Sarbassova, D., Frangioni, J. V.,
Dalal, S. N., DeCaprio, J. A., Greenberg, M. E., and Yaffe, M. B. (2002).
14-3-3 transits to the nucleus and participates in dynamic
nucleocytoplasmic transport. J Cell Biol 156, 817-828.
Brunet, A., Sweeney, L. B., Sturgill, J. F., Chua, K. F., Greer, P. L., Lin, Y.,
Tran, H., Ross, S. E., Mostoslavsky, R., Cohen, H. Y., et al. (2004). Stressdependent regulation of FOXO transcription factors by the SIRT1
deacetylase. Science 303, 2011-2015.
Cahill, C. M., Tzivion, G., Nasrin, N., Ogg, S., Dore, J., Ruvkun, G., and
Alexander-Bridges, M. (2001). Phosphatidylinositol 3-kinase signaling
inhibits DAF-16 DNA binding and function via 14-3-3-dependent and 14-33-independent pathways. J Biol Chem 276, 13402-13410.
Chen, F., Hersh, B. M., Conradt, B., Zhou, Z., Riemer, D., Gruenbaum, Y.,
and Horvitz, H. R. (2000). Translocation of C. elegans CED-4 to nuclear
membranes during programmed cell death. Science 287, 1485-1489.
Cohen, H. Y., Miller, C., Bitterman, K. J., Wall, N. R., Hekking, B.,
Kessler, B., Howitz, K. T., Gorospe, M., de Cabo, R., and Sinclair, D. A.
(2004). Calorie restriction promotes mammalian cell survival by inducing
the SIRT1 deacetylase. Science 305, 390-392.
Durocher, D., Taylor, I. A., Sarbassova, D., Haire, L. F., Westcott, S. L.,
Jackson, S. P., Smerdon, S. J., and Yaffe, M. B. (2000). The molecular
basis of FHA domain:phosphopeptide binding specificity and implications
for phospho-dependent signaling mechanisms. Mol Cell 6, 1169-1182.
Essers, M. A., de Vries-Smits, L. M., Barker, N., Polderman, P. E.,
Burgering, B. M., and Korswagen, H. C. (2005). Functional interaction
between beta-catenin and FOXO in oxidative stress signaling. Science 308,
1181-1184.
Essers, M. A., Weijzen, S., de Vries-Smits, A. M., Saarloos, I., de Ruiter,
N. D., Bos, J. L., and Burgering, B. M. (2004). FOXO transcription factor
activation by oxidative stress mediated by the small GTPase Ral and JNK.
Embo J 23, 4802-4812.
Finch, C. E., and Ruvkun, G. (2001). The genetics of aging. Annu Rev
Genomics Hum Genet 2, 435-462.
Finney, M., and Ruvkun, G. (1990). The unc-86 gene product couples cell
lineage and cell identity in C. elegans. Cell 63, 895-905.
Guarente, L., and Kenyon, C. (2000). Genetic pathways that regulate ageing
in model organisms. Nature 408, 255-262.
Henderson, S. T., and Johnson, T. E. (2001). daf-16 integrates
developmental and environmental inputs to mediate aging in the nematode
Caenorhabditiselegans. Curr Biol 11, 1975-1980.
Honda, Y., and Honda, S. (1999). The daf-2 gene network for longevity
regulates oxidative stress resistance and Mn-superoxide dismutase gene
expression in Caenorhabditiselegans. Faseb J 13, 1385-1393.
Hsu, A. L., Murphy, C. T., and Kenyon, C. (2003). Regulation of aging and
age-related disease by DAF-16 and heat-shock factor. Science 300, 11421145.
80
Hwang, H. Y., and Horvitz, H. R. (2002). The Caenorhabditiselegans
vulval morphogenesis gene sqv-4 encodes a UDP-glucose dehydrogenase
that is temporally and spatially regulated. Proc Natl Acad Sci U S A 99,
14224-14229.
Jansen, G., Hazendonk, E., Thijssen, K. L., and Plasterk, R. H. (1997).
Reverse genetics by chemical mutagenesis in Caenorhabditiselegans. Nat
Genet 17, 119-121.
Kaeberlein, M., McVey, M., and Guarente, L. (1999). The SIR2/3/4
complex and SIR2 alone promote longevity in Saccharomyces cerevisiae by
two different mechanisms. Genes Dev 13, 2570-2580.
Kamath, R. S., Fraser, A. G., Dong, Y., Poulin, G., Durbin, R., Gotta, M.,
Kanapin, A., Le Bot, N., Moreno, S., Sohrmann, M., et al. (2003).
Systematic functional analysis of the Caenorhabditiselegans genome using
RNAi. Nature 421, 231-237.
Kenyon, C. (2005). The plasticity of aging: insights from long-lived
mutants. Cell 120, 449-460.
Lamitina, S. T., and Strange, K. (2005). Transcriptional targets of DAF-16
insulin signaling pathway protect C. elegans from extreme hypertonic
stress. Am J Physiol Cell Physiol 288, C467-474.
Langley, E., Pearson, M., Faretta, M., Bauer, U. M., Frye, R. A., Minucci,
S., Pelicci, P. G., and Kouzarides, T. (2002). Human SIR2 deacetylates p53
and antagonizes PML/p53-induced cellular senescence. Embo J 21, 23832396.
Larsen, P. L., Albert, P. S., and Riddle, D. L. (1995). Genes that regulate
both development and longevity in Caenorhabditiselegans. Genetics 139,
1567-1583.
Lee, R. Y., Hench, J., and Ruvkun, G. (2001). Regulation of C. elegans
DAF-16 and its human ortholog FKHRL 1 by the daf-2 insulin-like
signaling pathway. Curr Biol 11, 1950-1957.
Lee, S. S., Kennedy, S., Tolonen, A. C., and Ruvkun, G. (2003). DAF-16
target genes that control C. elegans life-span and metabolism. Science 300,
644-647.
Libina, N., Berman, J. R., and Kenyon, C. (2003). Tissue-specific activities
of C. elegans DAF-16 in the regulation of lifespan. Cell 115, 489-502.
Lin, K., Dorman, J. B., Rodan, A., and Kenyon, C. (1997). daf-16: An
HNF-3/forkhead family member that can function to double the life-span of
Caenorhabditiselegans. Science 278, 1319-1322.
Lin, K., Hsin, H., Libina, N., and Kenyon, C. (2001). Regulation of the
Caenorhabditiselegans longevity protein DAF-16 by insulin/IGF-1 and
germline signaling. Nat Genet 28, 139-145.
Liu, J., Rolef Ben-Shahar, T., Riemer, D., Treinin, M., Spann, P., Weber,
K., Fire, A., and Gruenbaum, Y. (2000). Essential roles for Caenorhabditis
elegans lamin gene in nuclear organization, cell cycle progression, and
spatial organization of nuclear pore complexes. Mol Biol Cell 11, 39373947.
Liu, L. X., Spoerke, J. M., Mulligan, E. L., Chen, J., Reardon, B.,
Westlund, B., Sun, L., Abel, K., Armstrong, B., Hardiman, G., et al. (1999).
High-throughput isolation of Caenorhabditiselegans deletion mutants.
Genome Res 9, 859-867.
Luo, J., Nikolaev, A. Y., Imai, S., Chen, D., Su, F., Shiloh, A., Guarente,
L., and Gu, W. (2001). Negative control of p53 by Sir2alpha promotes cell
survival under stress. Cell 107, 137-148.
McElwee, J., Bubb, K., and Thomas, J. H. (2003). Transcriptional outputs
of the Caenorhabditiselegans forkhead protein DAF-16. Aging Cell 2,
111-121.
Morton, D. G., Shakes, D. C., Nugent, S., Dichoso, D., Wang, W., Golden,
A., and Kemphues, K. J. (2002). The Caenorhabditiseleganspar-5 gene
encodes a 14-3-3 protein required for cellular asymmetry in the early
embryo. Dev Biol 241, 47-58.
Motta, M. C., Divecha, N., Lemieux, M., Kamel, C., Chen, D., Gu, W.,
Bultsma, Y., McBumey, M., and Guarente, L. (2004). Mammalian SIRT1
represses forkhead transcription factors. Cell 116, 551-563.
82
Murakami, S., and Johnson, T. E. (1996). A genetic pathway conferring life
extension and resistance to UV stress in Caenorhabditiselegans. Genetics
143, 1207-1218.
Murphy, C. T., McCarroll, S. A., Bargmann, C. I., Fraser, A., Kamath, R.
S., Ahringer, J., Li, H., and Kenyon, C. (2003). Genes that act downstream
of DAF- 16 to influence the lifespan of Caenorhabditiselegans. Nature 424,
277-283.
Nakae, J., Park, B. C., and Accili, D. (1999). Insulin stimulates
phosphorylation of the forkhead transcription factor FKHR on serine 253
through a Wortmannin-sensitive pathway. J Biol Chem 274, 15982-15985.
Obsil, T., Ghirlando, R., Anderson, D. E., Hickman, A. B., and Dyda, F.
(2003). Two 14-3-3 binding motifs are required for stable association of
Forkhead transcription factor FOXO4 with 14-3-3 proteins and inhibition of
DNA binding. Biochemistry 42, 15264-15272.
Oh, S. W., Mukhopadhyay, A., Svrzikapa, N., Jiang, F., Davis, R. J., and
Tissenbaum, H. A. (2005). JNK regulates lifespan in Caenorhabditis
elegans by modulating nuclear translocation of forkhead transcription
factor/DAF-16. Proc Natl Acad Sci U S A 102, 4494-4499.
Paradis, S., and Ruvkun, G. (1998). Caenorhabditis elegans Akt/PKB
transduces insulin receptor-like signals from AGE-i PI3 kinase to the DAF16 transcription factor. Genes Dev 12, 2488-2498.
Picard, F., Kurtev, M., Chung, N., Topark-Ngarm, A., Senawong, T.,
Machado De Oliveira, R., Leid, M., McBurney, M. W., and Guarente, L.
(2004). Sirtl promotes fat mobilization in white adipocytes by repressing
PPAR-gamma. Nature 429, 771-776.
Rogina, B., and Helfand, S. L. (2004). Sir2 mediates longevity in the fly
through a pathway related to calorie restriction. Proc Natl Acad Sci U S A
101, 15998-16003.
Takata, T., and Ishikawa, F. (2003). Human Sir2-related protein SIRTI
associates with the bHLH repressors HES 1 and HEY2 and is involved in
HES 1- and HEY2-mediated transcriptional repression. Biochem Biophys
Res Commun 301, 250-257.
Tang, E. D., Nunez, G., Barr, F. G., and Guan, K. L. (1999). Negative
regulation of the forkhead transcription factor FKHR by Akt. J Biol Chem
274, 16741-16746.
Timmons, L., Court, D. L., and Fire, A. (2001). Ingestion of bacterially
expressed dsRNAs can produce specific and potent genetic interference in
Caenorhabditiselegans. Gene 263, 103-112.
Tissenbaum, H. A., and Guarente, L. (2001). Increased dosage of a sir-2
gene extends lifespan in Caenorhabditiselegans. Nature 410, 227-230.
Tran, H., Brunet, A., Griffith, E. C., and Greenberg, M. E. (2003). The
many forks in FOXO's road. Sci STKE 2003, RE5.
Tzivion, G., Shen, Y. H., and Zhu, J. (2001). 14-3-3 proteins; bringing new
definitions to scaffolding. Oncogene 20, 6331-6338.
Van Der Heide, L. P., Hoekman, M. F., and Smidt, M. P. (2004). The ins
and outs of FoxO shuttling: mechanisms of FoxO translocation and
transcriptional regulation. Biochem J 380, 297-309.
van der Horst, A., Tertoolen, L. G., de Vries-Smits, L. M., Frye, R. A.,
Medema, R. H., and Burgering, B. M. (2004). FOXO4 is acetylated upon
peroxide stress and deacetylated by the longevity protein hSir2(SIRT1). J
Biol Chem 279, 28873-28879.
Vanfleteren, J. R., and Braeckman, B. P. (1999). Mechanisms of life span
determination in Caenorhabditiselegans. Neurobiol Aging 20, 487-502.
Vaquero, A., Scher, M., Lee, D., Erdjument-Bromage, H., Tempst, P., and
Reinberg, D. (2004). Human SirTI interacts with histone H1 and promotes
formation of facultative heterochromatin. Mol Cell 16, 93-105.
Viswanathan, M., Kim, S. K., Berdichevsky A., and Guarente L. (2005). A
role for sir-2.I1regulation of ER stress response genes in determining C.
elegans lifespan. Dev Cell 9, 605-15.
Wang, W., and Shakes, D. C. (1997). Expression patterns and transcript
processing offtt-1 andftt-2, two C. elegans 14-3-3 homologues. J Mol Biol
268, 619-630.
Yeung, F., Hoberg, J. E., Ramsey, C. S., Keller, M. D., Jones, D. R., Frye,
R. A., and Mayo, M. W. (2004). Modulation of NF-kappaB-dependent
transcription and cell survival by the SIRT1 deacetylase. Embo J 23, 23692380.
TABLES
Table 1: Reducingpar-5 andftt-2 by RNAi results in nuclear
accumulation of DAF-16::GFP
% nuclear DAF-16::GFPa
% arrested
L3 larvae
(n)
Adults (n)
larvae
RNAi vector
0 (123)
0 (>200)
0
par-5 RNAi (N-terminal)
18 (82)
50 (80)
0
ftt-2 RNAi (N-terminal)
100 (91)
NA
100
par-5+ftt-2 RNAi
100 (66)
NA
96
ND
48 (219)
0
68 (520)
NA
80
(N-terminal)
par-5 RNAi (C-terminal)
ftt-2 RNAi (C-terminal)
GFP localization was scored using fluorescence microscopy to
observe the progeny of young adults fed the indicated RNAi
bacteria.
a All animals with any cells containing nuclear GFP were scored as
positive for nuclear DAF-16::GFP.
n = total number of animals examined, ND = not determined, NA =
non-applicable because of larval arrest.
Table 2: Analyses of lifespan experiments
sir-2.1
genotype
RNAi
WT
WT
WT
WT
WT
WT
(+)
(+)
(+)
(+)
(+)
(+)
Vector
par-5
ftt-2
Vector
par-5
ftt-2
20
17
17
14
14
14
70(1)
56(1)
61(1)
69(1)
65(1)
71(1)
P=0.0407
P=0.0007
P<0.0001b
P=0.0840
P=0.0844
pkIsl641
WT (+)
Vector
20
101(2)
P= 0 .4 14 9 b
[unc-119]
WT (+)
WT (+)
WT (+)
par-5
ftt-2
daf-16
16
16
14
104(2)
100(2)
35(1)
P<0.0001
P<0.0001
P<0.0001
sir-2.1(ok434)
If(-)
If(-)
Vector
par-5
16
15
94(2)
75(2)
P< 0 .0 00 1b
P=0.0412
If(-)
If(-)
ftt-2
daf-16
14
14
68(2)
43(1)
P=0.0006
P<0.0001
O/E (++)
Vector
30
122(2)
P< 0 .0 0 01 b
[sir-2.1, unc-119] O/E (++)
par-5
19
138(2)
P<0.0001
daf-2(e1370)
O/E (++)
O/E (++)
WT (+)
ftt-2
daf-16
Vector
15
15
41
169(2)
94(2)
100(2)
P<0.0001
P<0.0001
P< 0 .0 00 1b
par-5
ftt-2
daf-16
None
38
38
18
39
54(1)
106(2)
111(2)
122(2)
P=0.7011
P=0.1219
P<0.0001
daf-2(e1370)
WT
WT
WT
WT
sir-2.1(ok434)
If (-)
None
10
91(2)
daf-2(e13 70);
sir-2.1(ok434)
If (-)
None
43
143(2)
Genotype
N2
daf-16(mgDf50)
pkIsl642
(+)
(+)
(+)
(+)
Median # animals P value vs.
Shown
lifespan, that died controla in figure:
(# trials)
days
6A--D
6D
6A
6C
6B
6E
6F
P<0.0001 c
All statistical analyses were performed with Prism4 software, which uses KaplanMeier survival analysis, and the log-rank test to analyze survival curves.
If, loss-of-function; O/E, overexpression.
aUnless stated otherwise, p-values compare the lifespans of worms grown on bacteria
expressing a particular RNAi clone to the lifespans of worms of the same genotype
grown on bacteria expressing the RNAi vector control.
bp-value vs. the lifespans of wild-type (N2) worms grown on an RNAi vector
control.
Cp-value compares the lifespans of sir-2.1(ok434) animals to daf-2(e 13 70); sir2(ok434) double mutants.
87
Table 3: daf-16 targets upregulated in a sir-2.1 overexpresser.
Reported as
daf-16 target
(ref)
(3)
Elevation fold
in sir-2.1 o/e#
Description
2.8
lys-7 protein involved in the response
to pathogenic bacteria
(3)
2.3
C05E4.9
gei-7 member of the malate synthase and the
isocitrate lyase family
(1, 2, 3)
2.7
sod-3 superoxide dismutase
C08A9.1
(3)
2.3
dod-12 acyl-coA dehydrogenase
C55B7.4
(3)
2.2
Protein of unknown function
E01G4.3
(3)
3.5
Protein of unknown function
F15E6.4
(3)
2.4
D-aspartate oxidase
F18E3.7
(3)
2.5
F21C10.10 Protein of unknown function
(3)
2.3
Member of a protease protein family
F28A12.4
(3)
6.3
acs-2 fatty-acid acyl-coA ligase
F28F8.2
(3)
2.2
Protein of unknown function
F43H9.4
(3)
3.1
F45D3.4
Protein of unknown function
(3)
2.0
gst-4 glutathione S-transferase
K08F4.7
(3)
2.3
Member of a transporter/facilitator family
K09C4.5
(3)
2.1
Protein similar to asparagine synthase
M02D8.4
(3)
2.3
R12A1.4
ges-1 carboxylesterase expressed in gut cells
(3)
2.8
Member of the carboxylesterase protein family
T02B5.1
(3)
2.5
T10B9.1
cyp-13A4/dod-1 cytochrome P450 family
3.9
(3)
hsp-16.1 heat shock protein
T27E4.8
(3)
2.0
W01B11.6 Thioredoxin
4.0
(3)
W06D12.3 fat-5 delta-9 fatty acid desaturase
ZC410.5
Protein of unknown function
2.0
(3)
"For a full list of genes that change expression levels in animals with extra copies of sir-2.1,
see ref.4: Viswanathan et al, 2005.
Gene
C02A12.4
References
1. Libina, N., Berman, J. R., and Kenyon, C. (2003). Tissue-specific activities of C. elegans
DAF-16 in the regulation of lifespan. Cell 115, 489-502.
2. McElwee, J., Bubb, K., and Thomas, J. H. (2003). Transcriptional outputs of the
Caenorhabditiselegans forkhead protein DAF-16. Aging Cell 2, 111-121.
3. Murphy, C. T., McCarroll, S. A., Bargmann, C.I., Fraser, A., Kamath, R. S., Ahringer, J.,
Li, H., and Kenyon, C. (2003). Genes that act downstream of DAF-16 to influence the
lifespan of Caenorhabditiselegans. Nature 424, 277-283.
4. Viswanathan, M., Kim, S. K., Berdichevsky A., and Guarente L. (2005). A role for sir-2.1
regulation of ER stress response genes in determining C. elegans lifespan. Dev Cell 9, 605-15
88
FIGURES
Figure 1: C. elegans SIR-2.1 interacts with 14-3-3 proteins.
(A) SIR-2.1 is a nuclear protein. Whole-mount larvae and adults were
incubated with anti-SIR-2.1 antibody followed by washes and incubation
with FITC-conjugated secondary anti-rabbit antibody. Fluorescence
microscopy revealed nuclear staining in wild-type adult hermaphrodites
(WT, left) but not in a sir-2.I1deletion strain sir-2.1(ok434) mutants (sir2.1(A), right). Black arrows point at intestinal nuclei, and white arrow
points at germline nuclei. (*) marks non-specific staining of the pharynx
muscle. (B) 14-3-3 proteins coimmunoprecipitate with SIR-2.1. Sonicates
were prepared from synchronized 1-day-old adults of three strains with
different levels of sir-2.1: sir-2.1(ok434), (sir-2.I (A)), the N2 wild-type
strain (WT), and the sir-2.1 high-copy transgenic overexpresser geln3 (sir2.1(0/E)). All samples contained 10 mg of total protein. The SIR-2.1 band
and a 30 kDa co-precipitating protein band are indicated by arrows. H,
heavy chain; L, light chain. (C) Peptides identified by mass spectrometry of
the 30 kDa band. The numbers represent the first and the last amino acids of
each recovered peptide. (D) SIR-2.1 and 14-3-3 interact in wild-type
animals. SIR-2.1 was precipitated from lysates of wild type, sir-2.1(A) and
SIR-2.1 (O/E) animals, and precipitates were analyzed using western blots
with the indicated antibodies. Input lanes contained 50 [tg of lysate
proteins. Mock IP, to control for non-specific antibody interactions we
substituted the pre-immune serum for the SIR-2.1 antibody. (E) Both SIR2.1 and 14-3-3 are present in worm nuclear fractions. Western blot of
subcellular fractionations of wild-type embryos, separated into nuclear and
89
cytosolic fractions. First lane contains crude extract prior to separation. N,
nuclear extract, C, cytosolic extract. Ce-lamin, a nuclear fraction marker,
SQV-4, C. elegans UDP glucose dehydrogenase, a cytosolic marker.
90
Figure 1.
/ý(p
--
kDa
90-
-SIR-2.1
60"
403020-
WT. DAPI
--ý6
si-.£
&
-H
-14 -3-3]
A
C. Protein
(molecular
mass, Da)
FFT-2
WT, SIR-2,1 IF
(28,067)
iOK___
Input
MOCK
Input
IP
SIR-2.1 IP
SIR-2.1: WT A OIE OIE WT A OlE
PAR-6
(28,163)
Tubulin -
I
14-3-3
NICN C
SQV-4 [1Ce-lamin I
SIR-2.1
14-3-3
VOEDUIM
KHELGACELSN
VTEAEL•sNE•
,LAEVASGD
YLWMEA ODRNSV•E
SQ1QSYBFIAIK
AtiE
SIR-2.1 i- -I
E.
Peptides identified
Pw
1
21-25
3031-43
130-140
131-146
14-148
225-24
3-435
AnQamuwsnum
ONKWISROAba•
13013
1--1M
22D~2.
Figure 2. C. elegans 14-3-3 proteins interact specifically with SIR-2.1
and DAF-16. (A): 14-3-3 co-precipitates with SIR-2.1 but not with HDA-1
proteins. Immunoprecipitations were performed from wild-type animal
lysates with indicated antibodies, and analyzed by western blot with the
same antibodies and with 14-3-3 antibodies. (B): C. elegans 14-3-3 proteins
can bind DAF-16. Anti-GFP precipitations were performed from strains
expressing DAF-16::GFP and LIN-53::GFP fusion proteins, respectively.
14-3-3 co-precipitated with DAF-16::GFP translational fusion protein but
not with a LIN-53::GFP translational fusion protein. LC, light chain. Input
lanes contained 50 Rg of lysate proteins. See Figure 1 and Experimental
Procedures for details.
Figure 2.
A.
antiHDA-1
IP
antiSIR-2.1
In
IP
In
Western blot with
precipitating Ab
I--
I !!!
€
Western blot with
anti-14-3-3 Ab
B.
Strain:
WB:
LIN-53::GFP DAF-16::GFP
anti-GFP
anti-DAF-16
IP
In TP
TuI
IP with
iti-GFP Ab
14-3-3
LC
estern blot with
mnti-14-3-3 Ab
Figure 3: 14-3-3 proteins affect DAF-16 localization. (A) Reducing par5,ftt-2, or both by RNAi results in nuclear accumulation of DAF-16::GFP.
Top left panel shows control animals fed bacteria carrying the RNAi vector.
Top right panel shows DAF-16::GFP accumulation in the nuclei of
intestinal cells in worms treated with par-5 RNAi. Bottom panels show
nuclear accumulation of DAF- 16::GFP in the intestinal and hypodermal
cells offtt-2 RNAi (left) or double par-5 + ftt-2 RNAi animals (right).
White arrows mark GFP accumulation in hypodermal nuclei, and white
arrowheads mark nuclei of intestinal cells. Images of L3 progeny of animals
fed with bacteria expressing the RNAi clones indicated were taken at 400X
magnification using Nomarski fluorescence microscopy, 2,000 ms exposure
time.
(B)ftt-2 deletion results in nuclear DAF-16::GFP. Left, DAF-16::GFP in
animals with wild-typeftt-2. Right, nuclear accumulation of DAF-16::GFP
inftt-2(n4426A) animals. Images of L1 larvae were taken at 1000X
magnification using fluorescence microscopy, 1,000 ms exposure time.
94
Figure 3.
A.
vecto RNi
B.
par-
RNm
Figure 4. Overexpression of sir-2.1 delays DAF-16 exit from the nucleus
during recovery from stress. (A and B) DAF-16::GFP localization in
animals with endogenous SIR-2.1 levels and in animals overexpressing
SIR-2.1 30 minutes after heat pulse. Worms from indicated genotypes were
subjected to heat pulse of 30 min at 370 C. DAF-16::GFP localizations was
monitored before the heat pulse, and every 10 min thereafter. For recovery
worms were transferred to 20 0 C, and DAF-16::GFP localization was again
monitored every 10 min. (B) Kinetics of DAF-16::GFP subcellular
relocalization during recovery from heat stress.
96
Figure 4.
sir-d(I
B.
100
90
7O
30
20
10
0
I
Ti mfn
f5
20 mihm 3 mE sW Eu
k
Time of recovery after heat shock
97
121 min
Figure 5. Overexpression of sir-2.1 is required for lifespan extension
and stress resistance of the pklsl642 transgenic animals. (A): sir-2.1
RNAi but not R11A8.5 RNAi suppresses lifespan extension ofpklsl642
animals. Median lifespans are: pkls]641 + vector RNAi: 24 days (n=l 12);
pkIs1642 + vector RNAi: 32 days (n=l 13); pkls]642 + R11A8.5 RNAi: 34
days (n= 116); pkls1642 + sir-2.1 RNAi: 24 days (n=115). (B): pkIs1642
animals fed sir-2.1 RNAi show reduced resistance to heat stress relative to
animals fed control RNAi vector or R11A8.5 RNAi bacteria. Median
survivals at 32TC are: N2, 25 hours (n=64); pkls]642 + vector RNAi: 48
hours (n=68); R11A8.5 RNAi: 40 hours (n=78), p>0.05, ns; sir-2.1 RNAi:
25 hours (n=66), p<0.001.
Figure 5.
A.
!641 + vector RNAi
642 + vector RNAi
'642 + R11A8.5 RNAi
642 + sir-2.1 RNAi
0
days
i
i •'l
r-2.1 RNAi
dtor RNAi
1A8.5 RNAi
hours
Figure 6:ftt-2 and par-5 are required for sir-2.1-mediated lifespan
extension. Lifespan analyses were performed using RNAi plates as
described in Experimental Procedures. A-E: The black dashed line
represents the lifespans of wild-type (N2) control animals fed bacteria
carrying control RNAi vector, and solid lines represent lifespans of animals
of indicated genotypes fed with bacteria carrying control vector (black),
par-5 RNAi (magenta),ftt-2 RNAi (blue), or daf-16 RNAi (red). The sir2.1(wt) animals in (A) and pk1642[sir-2.10/EJanimals carried the unc-119
coinjection marker. Experiments were performed at 20 0 C. (A) par-5 andftt2 RNAi shorten the lifespans of animals with endogenous sir-2.I1levels by
-25%. (B) RNAi targeting par-5 orftt-2 prevents the lifespan extension by
extra copies of sir-2.1. Worms fedftt-2 RNAi had lifespans similar to those
of animals fed daf-16 RNAi (-50% reduction compared to the vector
control), and par-5 RNAi had a weaker effect, as lifespans were shortened
to the level of the wild type (-30% reduction relative to the vector control).
(C) Effects ofpar-5,ftt-2, and daf-16 RNAi on the short lifespan of sir-2.1
loss-of-function animals. Reductions in the expression levels of any of the
three genes resulted in a similar slight (<10%) shortening of the life span of
sir-2.1(ok434) animals. (D) par-5 andftt-2 RNAi do not significantly
shorten the life span of daf-16(mgDf50) animals. (E)ftt-2 RNAi do not
significantly shorten the long life span of daf-2(e1370) animals. (F) Loss of
sir-2.1 function does not shorten the life span of daf-2(e1370) animals.
Epistasis analysis of sir-2.1 and daf-2 genes showed that double mutants
bearing loss-of-function mutations in both genes had long life spans, similar
to those of daf-2(l) animals. The experiment in (F) was performed at 250 C.
100
Figure 6.
.
--
--
--
.--
oW rNAi
0lbA IA.
fI5u
j
,
edaer
Niw
r-5 R"4NA
-2 RNAi
Q1(OE)+par-5 RNAi
U1(cO+#I-2 RNMi
1(EeW+daf-s RNAi
Led
vekr RNAi
D.
n
4
0).VK RNA
O*)par-5RNA
O+it-2RNAi
ul
rjWr-5 RN4Ai
lftt-2 RNAi
H+da-16 RNAi
LAi
or RNAi
!
dWys
E.
---
-a
i
12WalVMayrr;SSRW
apsect
Jfeg
1370); sir-2.1(A)
4e.137r0W R*M
1370)
-2*137)%peRNMi
#Vf3r+R RNM
I
duy
do,
101
Figure 7. Overexpression of sir-2.1 leads to stress-resistance and
activation of the daf-16 target gene sod-3 in anftt-2-dependent manner.
(A-B): sir-2.1 andftt-2 promote stress resistance. WT, wild type; sir-2.1(A),
sir-2.1(ok434); ftt-2(A), ftt-2(n4426); sir-2.I (o/e), pkIs1642. A. Young
adults of the indicated genotypes were subjected to heat shock at 320 C, and
their viability was scored over the next 40 hr at the indicated time points. B.
Synchronized L larvae were transferred to plates containing 0.25 mM
paraquat. The number of animals that developed past the L4 stage was
assessed 96 hr later. Error bars represent standard deviations from triplicate
experiments. The p values vs. the wild type are: sir-2.1(A), p=0.012 ; sir-
2.1(o/e), p=0.001;ftt-2(A), p=0.004;sir-2.1(o/e);ftt-2(A) relative to sir2.1(o/e), p=0.001. (C-K) SIR-2.1 overexpression promotes activation of
sod-3 transcription, in daf-16 andftt-2-dependent manner. The psod3::gfp(muls84) reporter was crossed into the indicated genetic backgrounds,
and the expression of the reporter was observed using fluorescence
microscopy. Note higher levels of sod-3::gfp expression in both strains
overexpressing sir-2.1, the high-copy overexpresser geln3 (D), and the lowcopy overexpresser pkIsl642 (E), relative to that in the wild type (C). In
daf-16(mgDf5O) animals sod-3::gfp expression was not increased by
transgenes carrying extra copies of sir-2.I1(compare D to G and E to H).
Theftt-2 deletion did not have a pronounced effect on sod-3::gfP expression
in the wild-type background (J) but totally prevented induction of sod3::gfp expression in worms with extra copies of sir-2.1 (compare K to E).
This effect offitt-2(A) was specific for sir-2.1, since anftt-2 deletion did not
reduce elevated sod-3::gfp expression in the daf-2(lf) strain (compare I to
102
F). All images were taken at 50X magnification using Nomarski
fluorescence microscopy, with an exposure time of 300 ms.
Figure 7.
---
*-21
L. PI "/,,--,•| '
+
(
tit)
2-
2.f +);ft-2(&
}(
.2f(OQE);9f-2(+)
2.f(oE);fft-2(2Q
0
4
8
i
S
12 1i 20 24 28 32 3X 40
sir-2.1 + OIE A + O/E
Mft-2+ + + A A
hr
D
E
C=,
si-.IOE pi
wl typ e-
Gl
s ir,O
rt.H++
F
103
62
Figure 8. DAF-16 and SIR-2.1 interact in worms following stress, and
this interaction depends on 14-3-3. (A) Coimmunoprecipitation of SIR2.1 with DAF-16::GFP and 14-3-3 proteins following heat shock. Worm
lysates were prepared from geln3 worms (lacking DAF-16::GFP) or
animals overexpressing DAF-16::GFP fusion protein. SIR-2.1 or DAF16::GFP were precipitated from worm lysates using SIR-2.1 antibody or an
anti-GFP monoclonal antibody. Precipitates were separated by SDS PAGE
and analyzed using western blots with antibodies to DAF-16 (polyclonal
serum), SIR-2.1, and 14-3-3. +, animals were heat-shocked for 45 min at
37TC prior to lysis (B) Coimmunoprecipitation of SIR-2.1 and DAF16::GFP in animals with low insulin signaling and reduced levels of 14-3-3
proteins. Lysates were made as in (A). Worms were grown on RNAi plates
containing L4440 vector control bacteria (vector), daf-2 RNAi bacteria, or
ftt-2 RNAi bacteria. The lysates were analyzed as in (A). (C) A summary of
immunoprecipitation data above: associations between SIR-2.1, 14-3-3, and
DAF-16 under normal conditions, following stress, and upon low insulinlike signaling.
104
Figure 8.
input
IP antibody
Pre SIR-2.1
DAF-16::GFP - + +
Heatshock +4
+
DAF-16
GFP
4-444-
---
-
-
SIR-2.1
N
i
I
---
.......
- -
...
*LC
--
Tubulin
IP anti-GFP
vector def-2 ftt-2
-I- - +
input
vector daf-2 fW-2
-.Heat shock -+
RNAi
-
+
DAF-16
DAF-16 L
SIR-2.1I
i SIR-2.1
_
4
Tubulin
RNAi
114-3-3
:___..__._.._.______....__.__.~_~~_~,_,
rl~~~~~l~~~--~~---~----1
input
vector def-2
Heatshock
IPanti-SIR-2.1
ftt-2 vector def-2 Wf-2
- +
-
DAF-16
DAF-16 I ---- ...
SIR-2.1
.i 4
14-3-3
Complexes
I
-- ------
-----
m
1
Normal
i
Stress
conditions
SIR-2.1
eee
2
14-3-3
Low insulin
signaling
Low insulin
+ stress
SIR-2.1/14-3-3
+
+
+
+
DAF-16/14-3-3
+
+
-/-
+1-
SIR-2_1/DAF-16
.
+
-
105
+1-
Figure 9. daf-2 RNAi results in low insulin-like signaling. (A): DAF16::GFP accumulates in nuclei of animals subjected to daf-2 RNAi. daf16::gfp animals were fed daf-2 RNAi bacteria or control RNAi vector
bacteria for two generations. The picture shows young adult progeny of
animals transferred to RNAi plates at L1 larval stage. Arrows point at
intestinal nuclei, arrowheads point at hypodermal nuclei. (B): Animals fed
daf-2 RNAi show significant lifespan extension relative to animals fed
control RNAi vector. Median lifespans are: vector RNAi: 20 days (n=64),
daf-2 RNAi: 32 days (n=52), p<0.001.
106
Figure 9.
A.
veto
U
-
B.
- vedor RNA
-- daf-2 RNAi
I
I'
._
0 4 8 12 16 20 24 28 32 36 40 44 48
days
107
Figure 10. A model for the roles of SIR-2.1 and 14-3-3 in DAF-16
regulation of stress resistance and lifespan. We propose that following
stress SIR-2.1 binds DAF-16 in the nucleus in a 14-3-3-dependent manner
and the resulting complex participates in transcriptional activation of DAF16 target genes. 14-3-3 may promote the interaction between SIR-2.1 and
DAF- 16 either by scaffolding the complex or through a modification of
DAF-16 or SIR-2.1 following stress. Under low insulin-like signaling
conditions DAF-16 is not phosphorylated at the Akt sites, becomes
dissociated from 14-3-3 and accumulates in the nucleus. Nuclear DAF-16
produced by low insulin-like signaling does not bind SIR-2.1 and does not
require sir-2.1 and 14-3-3 function for activation. It is possible that another
nuclear cofactor promotes DAF-16 activity under low insulin conditions.
108
Figure 10.
Iss
eir
I
(d
so
-ff
LITO RPSfl T
"~>
fie span
J
StresAR
/
f
Reduced irisulin/IGF
signaling
cytoplasm
nucleus
109
Chapter 3
3-Ketoacyl thiolase controls the rate of aging in C. elegans
Ala Berdichevsky, Erika Hartweig, Leonard Guarente,
and H. Robert Horvitz
Erika Hartweig performed the electron microscopy experiments in this
chapter.
110
ABSTRACT
Studies of long-lived C. elegans mutants identified several genes that
function to limit the nematode lifespan, i.e. mutate to extend lifespan. By
contrast, studies of genes that normally function to delay aging were limited.
We identified a keto-acyl thiolase gene kat-1 in a genetic screen for C.
elegans mutants that age prematurely. Loss of kat-1 function results in
marked accumulation of aging pigment lipofuscin, short lifespan, early
behavioral decline, and other phenotypes characteristic of premature aging.
kat-1 encodes a conserved metabolic enzyme involved in fatty acid
oxidation. Our findings indicate that kat-1 acts to delay C. elegans aging
independently of known pathways for longevity determination, and suggest
that fatty acid oxidation may be important for successful aging of C.
elegans.
111
INTRODUCTION
Over the last few decades, several lines of experiments using C. elegans
and other organisms have resulted in progress towards understanding
molecular mechanisms of aging (Russell and Seppa 1987; Guarente and
Kenyon 2000; Braeckman et al. 2001; Gershon and Gershon 2002). C.
elegans aging, like aging of many organisms, has both predetermined and
stochastic components: animals of identical genotype will not have identical
lifespans, but a population under controlled conditions will show a
predictable distribution (Klass 1977). Loss-of-function mutations in single
genes in C. elegans can substantially increase lifespan (Kenyon et al. 1993;
Tissenbaum and Ruvkun 1998; Hsin and Kenyon 1999; Arantes-Oliveira et
al. 2003), indicating that lifespan determination in nematodes has a major
genetic component. Research in the field has largely focused on genes that
normally function to reduce longevity, that is, genes that mutate to extend
lifespan. This focus has been primarily a consequence of the difficulties in
distinguishing accelerated aging from sickness in short-lived mutants.
While longevity remains the most commonly used measure of the
nematode aging rate, additional biomarkers of C. elegans aging have been
characterized (Klass 1977; Garigan et al. 2002; Herndon et al. 2002;
Gerstbrein et al. 2005). These biomarkers include behavioral decline and
tissue aging. Strikingly, old nematodes display several features characteristic
of mammalian aging. First, aging of C. elegans is characterized by a
progressive decline in locomotion, both in velocity and in coordination:
aging worms move more slowly, become uncoordinated, and eventually
cease movement and become paralyzed (Klass 1977; Herndon et al. 2002;
Huang et al. 2004). Aging animals also show decreased feeding and
112
defecation rates (Klass 1977; Huang et al. 2004). Food consumption by C.
elegans is facilitated by contractions of the pharyngeal muscle that grinds
the bacteria and pumps it into the intestine (pharyngeal pumping). Old
worms show a gradual decrease of pharyngeal pumping rate, followed by a
cessation of pumping around day 16 of adulthood. Lack of visible pumping
does not lead to immediate death, and animals can live for more than a week
without feeding (Huang et al. 2004).
Old nematodes, like old humans, show progressive tissue and organ
deterioration. Muscles degenerate in aged worms, and gonad, intestine, and
connective tissues deteriorate (Herndon et al. 2002). Herndon et al. (2002)
observed that senescence in C. elegans is tissue-specific. While the nervous
system of C. elegans maintains normal anatomy and expresses appropriate
markers until a very old age, the muscle degenerates extensively: body-wall
muscle nuclei change shape and disintegrate, and sarcomeres become
disrupted. Old worms also seem to undergo misregulation of biosynthetic
processes and display thickening of their cuticles (normally, the cuticle is
synthesised during the larval stages prior to molting and not in adulthood)
and a production and diffusion throughout the body cavity of yolk proteins,
even though the oocytes that normally take up yolk proteins are no longer
produced (Garigan et al. 2002; Herndon et al. 2002).
The intestinal cells of C. elegans accumulate autofluorescent material in
the cytoplasm, just as post-mitotic mammalian cells accumulate the aging
pigment lipofuscin (Clokey and Jacobson 1986; Gerstbrein et al. 2005).
Lipofuscin consists of variable lipid peroxidation products, oxidized
proteins, and iron, and resists proteolytic degradation (Yin 1996). This
material accumulates in the secondary lysosomes, the cell's waste
compartment (Koenig 1964; Clokey and Jacobson 1986; Brunk and Terman
113
2002). The amount of lipofuscin gradually increases throughout adulthood,
and C. elegans autofluorescent granules can be detected by fluorescence
microscopy.
Tissue degeneration, the accumulation of lipofuscin, and behavioral
decline provide criteria to distinguish between aged and sick worms and
therefore can be used to facilitate the study of the aging process in C.
elegans. Specifically, these features provide biomarkers to identify genes
involved in the regulation of aging rate and for characterizing mutants that
show premature aging, or progeria, and may reveal a new class of aging
genes in C. elegans and other organisms.
Here we report the identification of a gene, kat-1, that controls the rate
of aging in C. elegans. We isolated kat-1 in a genetic screen for mutants
with increased accumulation of the aging pigment lipofuscin, and animals
bearing loss-of-function mutations in kat-1 age and die prematurely. kat-1
encodes a conserved enzyme involved in fatty acid metabolism and was
previously implicated in the control of fat accumulation in nematodes. Our
results indicate that C. elegans kat-1 acts in a new pathway that controls
longevity and the rate of aging.
114
RESULTS
Aging pigment that accumulates in old worms fluoresces with a
spectrum similar to that of mammalian lipofuscin
Of the biomarkers of aging we could use to screen for C. elegans
mutants that age prematurely, we chose to focus on the age-related
autofluorescence accumulation. This autofluorescence was readily
detectable in living old animals under fluorescence-equipped dissecting
microscope, whereas young animals accumulated negligible amounts of
autofluorescence (Figure lA). We determined the spectrum of the
autofluorescent pigment that accumulates in old C. elegans hermaphrodites
and compared it with the spectrum of mammalian lipofuscin. We prepared
lysates of wild-type worms of different ages and measured the fluorescence
in these extracts by fluorimeter using different excitation/emission
combinations (data not shown). The best signal-to-noise ratio was
observed using the excitation wavelength of 350 nm. At this excitation, the
emitted fluorescence peaked at approximately 430 nm, and the magnitude
of the peak fluorescence correlated with the age of the animals (Figure
IB). Notably, mammalian lipofuscin fluoresces with a very similar
spectrum: its excitation and emission maxima are 345 nm and 430 nm,
respectively (Csallany and Ayaz 1976; Brunk and Terman 2002). These
experiments show that worm aging pigment is similar to that accumulating
in old mammalian cells.
The characterization of the fluorescence signal by fluorimetry helped
us define a custom filter set for monitoring intestinal autofluorescence
(excitation 365 nm, emission 420-440 nm). Subsequent experiments were
performed with this new filter set, which allowed detection of increased
115
fluorescence signal in younger worms, facilitating screening and recovery
of the isolates.
Mutants with abnormal aging rates show abnormal fluorescence
accumulation
Next we wanted to determine whether mutants with abnormal lifespans
will show differences in accumulation of the lipofuscin-like aging pigment.
Mutations in genes involved in the insulin-like pathway significantly alter
the C. elegans lifespan and rate of aging (Finch and Ruvkun 2001; Kenyon
2005). We examined a long-lived mutant with a loss-of-function mutation
in the insulin-like receptor daf-2 gene and a short-lived mutant with a lossof-function mutation in the forkhead transcription factor gene daf-16. We
also tested autofluorescence accumulation in another short-lived mutant
defective in the gene sir-2.1, which acts in a different aging pathway
(Berdichevsky et al. 2006). Both short-lived mutants accumulated more
lipofuscin-like fluorescence than the wild-type animals, at all ages tested.
The long-lived daf-2 mutants accumulated less autofluorescence than did
the wild type (Figure IC). Similar observations have been reported by
Gerstbrein et al. (2005). These results indicate that lipofuscin accumulation
is a reliable marker of aging, and that this biomarker is suitable for
screening for mutants that age prematurely.
Screen for mutants with increased lipofuscin
To identify genes that when disrupted cause premature aging, we
performed a screen for mutants that show premature accumulation of
lipofuscin autofluorescence (Figure 2). We mutagenized L4
hermaphrodites with EMS and isolated synchronized F2 progeny. We then
116
treated the L4 F2 worms with 5' fluoro-deoxyuridine (FUdR), which
blocks DNA replication and prevents worm eggs from hatching. This
treatment allows observation of aged F2 worms without mixing these
animals with the next generation. The F2 animals were screened for
increased lipofuscin accumulation when they reached the second or third
day of adulthood. Wild-type animals of this age accumulate a negligible
amount of intestinal autofluorescense (Figure 1). Isolates with high levels
of lipofuscin were moved to plates without FUdR for recovery. The
phenotype of high autofluorescence was retested for in the progeny of the
original isolates. We screened approximately 20,000 haploid genomes and
picked 51 candidate mutants, 16 of which did not produce viable progeny.
Of the 35 fertile isolates, 24 failed to retest. We obtained 11 mutant strains
that reliably exhibited the premature accumulation of intestinal
autofluorescence.
Mutants isolated in the screen correspond to at least three
complementation groups
We outcrossed the isolates from the screen six times to the wild type
before characterization. Three of the isolates, n4430, n4434, and n4435,
exhibited a weakly semidominant phenotype: heterozygous animals
showed weak accumulation of autofluorescence and were distinguishable
from both the wild type and the mutant. Among the progeny of these
heterozygous animals approximately one quarter showed strong
fluorescence accumulation, one quarter had no detectable fluorescence and
one half showed weak accumulation of autofluorescence. The phenotype of
the n4501 isolate was recessive, whereas the phenotypes of the other seven
117
mutants, n4427, n4428, n4429, n4432, n4433, and n4436, were strongly
semidominant: heterozygous animals showed high levels of
autofluorescence but were still distinguishable from the homozygous
mutants.
We mapped n4430, n4434, n4435, and n4501 to linkage group II
(LGII), and n4427, n4428, n4429, n4432, n4433, and n4436 mapped to
LGV (data not shown). n4430, n4434, and n4435 failed to complement,
while n4501 complemented n4430 for the increased lipofuscin
accumulation, and therefore represents another complementation group.
The progeny of crosses between different chromosome V mutants all
showed the mutant phenotype. However the interpretation of these
complementation tests is complicated by the semidominant phenotype
caused by these mutations. In short, the 11 isolates with high
autofluorescence represent at least three complementation groups, two on
LGII and one or more on LGV (Table 1), suggesting that at least three
genes control autofluorescence accumulation in C. elegans.
Three isolates premature display multiple attributes of aging
To assess whether our mutants age prematurely, we tested them for
other aging-related characteristics. Progeroid mutants should have shorter
lifespans than wild-type animals. Therefore, we performed lifespan
analysis of our mutants (see Table 1 and Figure 3). Seven had 25-40%
shorter lifespans than the wild type, whereas four had normal or almost
normal lifespans (Table 1). Notably, all three allelic LGII mutants had
short lifespans (Figure 3). The other LGII mutant, n4501, had only a slight
(15%) shortening of lifespan, and of LGV mutants, four had shorter
lifespans, while others had nearly wild-type lifespans.
118
We also assessed the behavioral declines with age of our mutants.
The most obvious behavioral phenotype of aged worms is a progressive
slowing of the rates of locomotion and pharyngeal pumping with time.
Young worms produce robust sinusoidal movements with about 50 bodybends per minute. After about 9 days of adulthood worms slow down their
locomotory rate by 50%, and by day 18 most have stopped crawling
completely (Hemdon et al. 2002; Huang et al. 2004). A similar slowing of
the pharyngeal pumping occurs (Huang et al. 2004). We have characterized
the behavioral decline with age of our screen isolates. The three allelic LGII
mutants n4430, n4434, and n4435 initially had wild-type rates of
locomotion and pumping, but from the mid-point in their lives (day 9)
showed a premature decline in both locomotion and pumping (Figure 4A,
B). The LGII mutants n4430, n4434, and n4435 also showed premature
onset of uncoordinated locomotion, another biomarker of aging in worms
(Herndon et al. 2002), similar to that observed in daf-16(lf) mutants that age
prematurely (Figure 4C). None of the other mutants, including those with
short lifespans, developed behavioral abnormalities with age (see Table 1
and Figure 4A, B).
To further analyze the phenotype of allelic LGII mutants we
performed electron microscopy analysis to look at tissue integrity in aging
wild type and in n4430 mutant animals. Old wild-type animals show
profound decline in the integrity of many organs, including intestine, gonad,
and muscle, and accumulation of electron-dense material (Herndon et al.
2002). The intestinal lumen of old worms is often bloated with bacteria, and
the intestinal cells show membrane swirls typical of necrosis (Garigan et al.
2002). n4430 animals showed early decline of intestine and gonad
119
compared to the wild type, greatly increased accumulation of electrondense material, and early appearance of necrosis-like membrane swirls
(Table 3 and Figure 5).
Based on these results, we conclude that the gene mutated in n4430,
n4434, and n4435 is likely to control normal onset and progression of aging
in C. elegans.
Mapping and cloning of the gene identified by n4430
We cloned the gene mutated in n4430 animals using a combination of
positional cloning, polymorphism mapping and RNAi. We genetically
mapped n4430 on LGII to the region between dpy-10 and unc-4 and close to
unc-104 (Figure 6A). We then used polymorphism mapping and bli-2 n4430
unc-4/CB4856 recombinants to narrow the region to approximately 155 kb,
spanned by three cosmids, C15F1, C52E12 and C06B7. We injected of a
pool of these three cosmids and observed rescue of the premature
autofluorescence observed in n4430 animals (see Figure 6B). When injected
individually, only cosmid C06B7, which contains 16 predicted genes,
rescued the fluorescence accumulation of n4430 animals. By contrast,
cosmid ZK1127, which contains 12 of the 16 C06B7 ORFs, did not rescue
(Figure 6). In parallel, we knocked down each of the genes contained in
C06B7 by feeding dsRNAs, and found that reducing one of the 16 genes on
C06B7, T02G5.8, results in high autofluorescence. We injected a PCR
product containing the T02G5.8 gene into n4430 animals and observed that
TO2G5.8 can rescue fluorescence accumulation abnormalities in the n4430
mutants (Figure 7A). Importantly, the same fragment also restored normal
lifespan to the short-lived n4430 mutants (Figure 7B). We determined the
120
sequence of TO2G5.8 in n4430 mutants and found a missense mutation
(G1550A) in the last exon of this gene (Figure 6C). We determined the
sequences of TO2G5.8 in the two allelic mutants, n4434 and n4435, and
found lesions in both (n4434: G504A, n4435: G1402A) (Figure 6). C.
elegans mutants with increased intestinal autofluorescence were previously
isolated by Babu (1974). Mutations in one of the genes he described, flu-3,
resulted in an increase in gut autofluorescence similar to that observed in the
n4430 mutants, and mapped close to dpy-10 on LGII (Babu 1974).flu3(el001) failed to complement n4430 for intestinal fluorescence
accumulation, suggesting that they are allelic. We sequenced T02G5.8 influ3(el001) and found that T02G5.8 is mutated in this strain (Ala271Thr,
Ala279Val), confirming that T02G5.8 isflu-3.
T02G5.8 encodes a conserved thiolase involved in fatty acid
metabolism.
The gene T02G5.8 encodes a protein with 52% identity and 70%
similarity to the mammalian enzyme acetoacetyl CoA thiolase (EC 2.3.1.9),
which is involved in the beta-oxidation of fatty acids in mitochondria and
peroxisomes (Kegg pathway database,
http://www.genome.ad.jp/kegg/pathway.html). T02G5.8 was named kat-1
(for keto-acyl thiolase-1) by Mak et al (2006), who isolated this gene in a
screen for C. elegans mutants with abnormal fat metabolism (see
Discussion). We will use this name, instead offlu-3, below. kat-1 is well
conserved from worms to humans, and mutations identified in our screen
result in substitutions of conserved amino acids (Figure 8). To analyze the
null phenotype of kat-1, we obtained a deletion allele of kat-1, tm1037, from
the Japanese knock-out consortium (National Bioresource Project for the
121
Experimental Animal Nematode C. elegans). kat-1(tml037A) showed
increased lipofuscin fluorescence, a shortened lifespan, and early behavioral
decline comparable to those observed for the kat-1 mutants n4430, n4434,
and n4435 (Figure 4C and data not shown).
kat-1 delays aging independently of most known longevity pathways
Several pathways are known that control C. elegans aging (Kenyon
2005). To see whether kat-1 acts to delay aging via one of these known
longevity pathways, we crossed the kat-1 mutation into backgrounds of
long-lived C. elegans mutants and tested whether kat-1 is required for the
lifespan extension of these mutants. In particular we made kat-1 double
mutants with mutants defective in the insulin-like receptor gene daf-2
(Kenyon et al. 1993), mitochondrial mutant isp-1 (Feng et al. 2001),
germline specification mutant mes-1 (Berkowitz and Strome 2000), and
chemosensory mutants osm-5 and che-3 (Apfeld and Kenyon 1999). In
most cases we saw that kat-1 is not required for the longevity induced by
mutations in these genes (see Table 2). The only exception was the
germline signaling gene mes-1. Lack of a germline significantly increases
the lifespan of C. elegans (Hsin and Kenyon 1999), and mes-1 temperature
sensitive mutants that lacked a germline had a 63% longer lifespan than
mes-1 animals that had a germline (Table 2). Although kat-1; mes-1 double
mutants showed a lifespan extension relative to kat-1 alone, this extension
was only 26% (Table 2). These results suggest that kat-1 acts to delay aging
via an unknown pathway, which may be partially required for lifespan
regulation by germline signaling.
122
kat-1 mutants are sensitive to starvation
As many aging mutants show an altered resistance to stress, we tested
stress responses of kat-1 mutants. We assayed three types of stresses: heat
stress, oxidative stress, and starvation. In contrast to daf-1 6 progeroid
animals, which are more sensitive to both heat and oxidative stress, kat-1
mutants showed wild-type sensitivity to these stressors (Figure 9A and B).
kat-1 was not needed for the increased resistance to heat shock of longlived animals bearing mutations in the daf-2 insulin-like receptor gene
(Figure 9A), suggesting that it is not involved in cellular responses to heat
and oxidative stress.
By contrast, kat-1 was important for normal response to starvation
stress, because kat-1 mutants show increased sensitivity to starvation
compared to wild type (Figure 8C). For example, after nine days without
food, over 90% of kat-1 mutants had died, while two thirds of the wild-type
animals were still alive. Notably, kat-1 mutants also were less responsive to
calorie restriction (CR) relative to the wild type, as they showed a less
significant lifespan extension upon CR (35% extension for the wild type vs.
24% extension for the kat-1 mutants, see Table 2), consistent with the idea
that kat-1 function is required for coping with energy deficit.
Other fatty-acid beta-oxidation genes affect aging
Mutations in the gene kat-1 result in abnormalities in fat metabolism
(Mak et al. 2006). The enzyme encoded by kat-1 likely catalyzes the last
step of fatty acid breakdown, and kat-1 loss-of-function might result in a
decreased activity of the entire lipid beta-oxidation pathway. Beta-oxidation
of fatty acids involves binding of the camitine molecule, which is required
for the transport of fatty acids into the mitochondria. This reaction is
123
catalyzed by carnitine transferase and it is a rate-limiting step of fatty acid
beta-oxidation (Shepherd et al. 1966). Other enzymatic steps include CoA
binding and subsequent oxidation reactions that lead to the production of 3ketoacyl CoA, which is cleaved into acetyl CoA and acyl CoA chain, now
shortened by two carbons (Figure 10). C. elegans genome contains multiple
genes encoding homologs of enzymes that catalyze different steps of betaoxidation pathway (see Table 4). We hypothesized that the progeroid
phenotype of kat-1 mutants is the result of an impaired fat oxidation by
these animals. We therefore asked whether other genes involved in fatty
acid beta-oxidation also affect the rate of worm aging. Specifically, we used
RNAi to reduce expression of the C. elegans enzymes predicted to catalyze
different steps of the fatty acid oxidation pathway and assessed lifespan and
fluorescence accumulation in these animals. The results of these
experiments are summarized in Table 4. While RNAi of many of the genes
tested did not cause aging abnormalities, reducing several fatty-acid
oxidation genes, including cpt-1 carnitine palmitoyl transferase 1, ech-2
enoyl dehydrogenase 2, ard-1 dehydrogenase 1, and another ketoacyl
thiolase T02G5.4, resulted in a significant shortening of lifespan and/or
premature accumulation of lipofuscin fluorescence (see Figure 11 and Table
4). RNAi reduction of other putative beta-oxidation enzymes did not lead to
any visible abnormalities, suggesting that they are either not involved in the
regulation of aging, or that they act redundantly with other genes encoding
members of the same category of enzymes. Alternatively, the RNAi may
not have caused knock down of these genes to the extent required to see the
abnormalities.
124
These results are consistent with the hypothesis that the premature
aging phenotype of kat-1 mutants is not kat-l-specific but rather results
from down-regulation of the fatty acid beta-oxidation pathway.
125
DISCUSSION
KAT-1 thiolase regulates C. elegans aging.
We report the isolation of C. elegans loss-of-function mutations in the
ketoacyl thiolase gene kat-1 in a genetic screen for animals that age
prematurely. Here we show that kat-1 mutants have multiple defects
characteristic of accelerated aging: they precociously accumulate high
levels of aging pigment lipofuscin, show premature tissue aging, and have
short lifespans. These animals also show early behavioral decline typical of
old age. This finding indicates that kat-1 functions to promote longevity.
The phenotype of kat-1 loss-of-function mutants is unlikely to be
caused by non-specific sickness, because these mutants are not more
sensitive to heat and oxidative stress than the wild type. In addition, young
kat-l(lf) animals do not show any behavioral or tissue abnormalities apart
from increased lipofuscin. Therefore, we believe that kat-1 generally
functions to maintain normal rate of aging in worms, and its loss-offunction accelerates the aging process.
Mutations in kat-1 were previously identified in a screen for mutants
displaying synthetic fat accumulation with tub-1 loss-of-function. tub-i
encodes a homologue of mammalian Tubby, which is involved in glucose
and fat metabolism (Wang et al. 2006). kat-1 acts in parallel to tub-i in the
mitochondria of intestinal and muscle cells to control triglyceride
accumulation in C. elegans (Mak et al. 2006). Defects in the mitochondrial
ketoacyl-CoA thiolase activity are found in patients with a rare metabolic
disease characterized by impaired ketone body metabolism, lactic acid
build-up, and inability to sustain starvation (Fukao et al. 2001). We found
126
that C. elegans KAT-1 also results in sensitivity to starvation, suggesting
that the two proteins may play similar roles in energy homeostasis.
Fatty acid oxidation in aging and disease
Fatty acid beta-oxidation is a crucial process for acquiring energy from
dietary fats and fat stores (Ockner 2004). Fat is the main energy store in the
cell, and fatty acid oxidation is pivotal at times of food deprivation, such as
starvation or fasting (Finn and Dice 2006), and in organs that heavily rely
on energy produced from fat, such as liver, skeletal muscle, white adipose
tissue, and the heart (Dhalla et al. 1992; Wolfe 1998; Ockner 2004). Fatty
acid oxidation is carried out in mitochondria and peroxisomes, and integrity
of these organelles is crucial for effective energy production from fat fuels
(Reddy and Mannaerts 1994; Shigenaga et al. 1994; McGarry 1998; Reddy
and Rao 2006). Consistent with the crucial role of beta-oxidation in energy
homeostasis, mutations in genes encoding metabolic enzymes and upstream
regulators of fatty acid oxidation have been implicated in a variety of
human disorders. For example, mutations in the mitochondrial trifunctional
protein that possesses fatty acid dehydrogenase, enoyl-coA hydratase and
hydroxyacyl-coA dehydrogenase activities result in severe metabolic
disease characterized by inability to oxidize fats, liver abnormalities, and
inability to sustain starvation (Angdisen et al. 2005). Accordingly,
mutations in several fatty acid metabolism enzymes in rodents also cause
metabolic defects and early death (Ibdah et al. 2005; Tolwani et al. 2005;
Huyghe et al. 2006; Kao et al. 2006).
Normal mitochondrial energy production via oxidative phosphorylation
results in an accumulation of reactive oxygen species that ultimately cause
damage to mitochondria and mitochondrial dysfunction (Singh 2006).
127
Mitochondrial DNA mutations and non-functional mitochondria
accumulate during aging (Wallace 2005; Kujoth et al. 2006). Numerous
studies over the past 30 years showed an association between aging and
decreased fatty acid oxidation. Fatty acid oxidation rates are low in a
variety of tissues in old animals, such as skeletal muscles, liver, adipose
tissue and myocardial muscle (Mori et al. 1988; Paradies et al. 1995; Guo et
al. 2007). Tissues from old animals show decreased activities of multiple
enzymes implicated in fatty acid metabolism (a few examples include
Acetyl-coA dehydrogenase, AMPK, PGC-lbeta, and camitine acetyltransferase (Mori et al. 1988; Nogalska et al. 2003; Gonzalez et al. 2004)).
Accumulation of triglycerides and various products of incomplete lipid
oxidation also occurs at old age, leading to so-called lipotoxicity, and is
associated with age-related disorders like type 2 diabetes, metabolic
syndrome and Alzheimer's disease (Montine and Morrow 2005; Reddy and
Rao 2006; Weinberg 2006).
Does impaired fatty acid beta-oxidation cause aging?
Despite the apparent association between decline in fatty acid
oxidation and aging, it is unclear whether aging causes metabolic
abnormalities, or metabolic abnormalities contribute to the aging process,
and both ideas are discussed in the field (Toth and Tchernof 2000; Slawik
and Vidal-Puig 2006). That loss of function of KAT-1, an enzyme involved
in fatty acid breakdown, causes progeria raises the possibility that fatty acid
beta-oxidation pathway is important for the control of aging and lifespan.
Reducing expression of several beta-oxidation genes caused effects similar
to that of kat-1, namely increased lipofuscin and short lifespan. These
results are consistent with the hypothesis that beta-oxidation pathway is
128
involved in the regulation of C. elegans aging. Notably, overexpressing katI did not result in lifespan extension, perhaps due to the fact that other steps
of the pathway were limiting. Indeed, the reaction catalyzed by kat-1 is not
the rate-limiting reaction in fatty acid oxidation (Shepherd et al. 1966).
What causes the aging abnormalities of kat-1 mutants and of the
animals with a presumptive reduction of fatty acid oxidation? One
possibility is that inability to efficiently process fatty acids may result in
insufficient cellular energy, leading to cellular and organismal death.
Alternatively, defects in fatty acid oxidation might lead to the accumulation
of toxic substances originating from intermediates of lipid peroxidation;
such toxic substances might promote cellular damage and aging. Further
experiments to identify compounds that accumulate in kat-1 mutants as well
as studies of the energy utilization throughout the C. elegans lifespan might
shed light on the mechanistic basis of the effect of kat-1 and of fatty acid
oxidation on aging.
129
EXPERIMENTAL PROCEDURES
Strains
Strains were cultured as described by Brenner (Brenner 1974) and
maintained at 20 0 C unless specified otherwise. The wild-type strain
described in this study was the Bristol strain N2 (Brenner 1974). The
following mutations were used (described by Riddle (1997) unless
otherwise noted):
LGI: dpy-5(e61), daf-16(mgDf50) (Ogg et al. 1997), che-3(e1124)
LGII: unc-85(n319), bli-2(e768), dpy-10(e128), unc-104(e1265), unc4(e120), bli-1(e769), kat-1(tm1037, n4430, n4434, m4435, e1001) (this
study).
LGIII: unc-32(e189), daf-2(e1370)
LGIV: unc-5(e53), sir-2.1(ok434), isp-l(qm150)
LGV: dpy-11(e224), unc-76(e911),
n4427, n4428, n4429, n4431, n4432, n4433, n4436 (this study)
LGX: lon-2(e678), osm-5(p813), mes-l(bn74ts)
Isolation of mutants with high lipofuscin-like fluorescence
We mutagenized N2 hermaphrodites at late L4 stage with ethyl
methanesulphonate (EMS) as described by Brenner (1974). After recovery
on food for 1 hour, individual Po animals were transferred to 50 mm Petri
plates. After 5 to 6 days the Fl animals were dissolved by bleaching in
NaOH and bleach solution. Bleaching dissolves larvae and adults, but not
embryos, and this procedure results in a synchronized culture, which is
important when studying aging. The F2 animals were transferred to 50 mm
Petri plates, 100 animals per plate. At the L4 stage, fluorodeoxyuridine
130
(FUdR) was added to plates to final concentration of 25 uM. Animals were
screened for high intestinal autofluorescence at day 2 or 3 of adulthood. We
screened animals equivalent to approximately 20,000 mutagenized haploid
genomes using this procedure.
Linkage group assessment
We used the following markers to determine the chromosomal linkage
of newly isolated mutations that caused high autofluorescence: dpy-5 I, bli2 II, unc-32 III, unc-5 IV, dpy-11 V lon-2 X We generated animals
heterozygous for the new mutation and for three of these markers. We
picked progeny with high autofluorescence and determined if these progeny
segregated the above markers.
Polymorphism mapping
We used CB4856 and bli-2 n4430 unc-4 strains for single nucleotide
polymorphisms (SNP) mapping, similarly to the method described by
(Wicks et al. 2001). We crossed CB4856 males with bli-2 n4430 unc-4
hermaphrodites, picked F2 animals recombinant between bli-2 and unc-4
(showing only one marker) and obtained homozygotes for the recombinant
chromosome. We tested animals in next generation for the presence of the
n4430 mutation and identified the recombination site by following sequence
polymorphisms specific to N2 or CB4856 genome.
Transgenic animals
Germline transformation was performed as described by Mello et al.
(1991) by injecting cosmid DNA (5-20 ng/ul) or PCR products (5-50 ng/ul)
into kat-1 (n4430); unc- 76(e911) mutants. An unc-76 rescuing construct
131
p76-16B (Bloom and Horvitz 1997) was used as co-injection marker. NonUnc transgenic lines were established and scored.
RNAi analyses
We used the whole-genome C. elegans RNAi library (Kamath et al.
2003) as a source for dsRNA-expressing bacterial strains we used in our
study. Feeding RNAi was performed as described by Fire et al. (1998), and
progeny of animals subjected to RNAi were scored at day 3 of adulthood
for autofluorescence accumulation. As RNAi negative controls, animals
were fed HT115 bacteria containing pL4440, a backbone vector in which
library constructs were subcloned.
Allele sequences
We used PCR-amplified regions of genomic DNA as templates in
determining gene sequences. For each gene investigated we determined the
sequences of the entire gene plus 250 bases of the upstream promoter
sequence. At least two independent PCR products were used to confirm an
observed mutation. All sequences were determined using an automated ABI
373 DNA sequencer (Applied Biosystems).
Analyses of lifespan and stress-resistance
Unless stated otherwise, lifespan assays were performed at 20 0 C and
initiated by transferring L4 larvae of the indicated genotypes to plates
containing 10 [tM fluorodeoxyuridine (FUdR). We used Prism 4 software
to perform statistical analyses of survival curves. Animals that crawled off
the plate, burst at the vulva or died from internal hatching of progeny were
132
excluded from the analyses. In all lifespan assays reported here, between
90% and 95% of animals died naturally and were scored.
Heat-shock assays were performed at 320C using one-day-old adults.
For each genotype, 50-70 animals were transferred to 5 cm plates, and
viability was scored every 2 - 8 hrs after shift to 320 C. Statistical analysis
was performed using Prism 4 software, as for the lifespan assays.
Assays of paraquat sensitivity were performed at 20 0 C by transferring
arrested L1 larvae to plates containing 0.25 mM paraquat. Development
was assessed 2, 3, and 4 days later, and the percentage of animals that
developed to adulthood was scored. Starvation survival analyses were
performed with synchronized L larvae in M9 with or without food.
Survival was assayed every 2 - 4 days by movement of larvae in liquid
droplets. The student t-test was used to evaluate the significance of
differences in the means.
Behavioral assays
Animals were grown on regular NGM plates without FUdR and
transferred every other day to separate them from their progeny. For
behavioral assays, animals were transferred to fresh NGM plates (seeded
with bacteria the night before) and maintained for 30 min. before
locomotion and/or pumping rates were counted. Locomotion rate was
determined by counting body bends per minute of moving animals in the
center of a fresh bacterial loan. Pumping rate was assayed by counting the
number of movements of the pharyngeal muscle per 30 seconds in animals
present in the bacterial loan. At least 15 animals per genotype per time
point were assayed. Classification of aging animals into behavioral classes
133
was done as described by Hemdon et al. (2002). At least 80 animals per
genotype were assayed.
Electron Microscopy
Adult hermaphrodites of various ages were fixed in a mixture of 0.8%
gluteraldehyde and 0.8% OsO4 in 0.1 M cacodylate buffer (pH 7.2) for one
hr at 4°C in the dark. Worms were cut in this fixative with a razor blade, the
head regions discarded, and postfixed in 2% OsO4 in 0.1 M cacodylate
buffer overnight at 40 C. Approximately four posterior regions were then
aligned in an agar block, and the agar blocks were dehydrated and
infiltrated into an Epon-Araldite mixture. Sections of the midbody region
(-30 nm) were collected on Pioloform-coated slotted grids. Sections were
post-stained with 1% uranyl acetate and Reynolds lead citrate. Sections
were viewed using a JEOL 1200EX electrone microscope at 80 KV.
Sections from at least ten animals per genotype were examined and
analyzed.
134
Acknowledgments
We thank the Caenorhabditiselegans Genetics Center for providing strains,
Dr. Shohei Mitani and the Japanese knock-out consortium for the kat1(tm1037) strain, Na An and Beth Castor for strain maintenance and
determining DNA sequences, respectively, Brendan Galvin, Niels Ringstad,
Hillel Schwartz, and other members of the Guarente and Horvitz
laboratories for suggestions.
References
Angdisen, J., Moore, V.D., Cline, J.M., Payne, R.M., and Ibdah, J.A. 2005.
Mitochondrial trifunctional protein defects: molecular basis and
novel therapeutic approaches. Curr Drug Targets Immune Endocr
Metabol Disord5(1): 27-40.
Apfeld, J. and Kenyon, C. 1999. Regulation of lifespan by sensory
perception in Caenorhabditiselegans. Nature 402(6763): 804-809.
Arantes-Oliveira, N., Berman, J.R., and Kenyon, C. 2003. Healthy animals
with extreme longevity. Science 302(5645): 611.
Babu, P. 1974. Biochemical genetics of Caenorhabditiselegans. Molec Gen
Genet 135: : 39-44.
Berdichevsky, A., Viswanathan, M., Horvitz, H.R., and Guarente, L. 2006.
C. elegans SIR-2.1 interacts with 14-3-3 proteins to activate DAF-16
and extend life span. Cell 125(6): 1165-1177.
Berkowitz, L.A. and Strome, S. 2000. MES-1, a protein required for
unequal divisions of the germline in early C. elegans embryos,
resembles receptor tyrosine kinases and is localized to the boundary
135
between the germline and gut cells. Development 127(20): 44194431.
Bloom, L. and Horvitz, H.R. 1997. The Caenorhabditiselegans gene unc76 and its human homologs define a new gene family involved in
axonal outgrowth and fasciculation. Proc Natl Acad Sci US A 94(7):
3414-3419.
Braeckman, B.P., Houthoofd, K., and Vanfleteren, J.R. 2001. Insulin-like
signaling, metabolism, stress resistance and aging in Caenorhabditis
elegans. Mech Ageing Dev 122(7): 673-693.
Brenner, S. 1974. The genetics of Caenorhabditiselegans. Genetics 77(1):
71-94.
Brunk, U.T. and Terman, A. 2002. Lipofuscin: mechanisms of age-related
accumulation and influence on cell function. Free Radic Biol Med
33(5): 611-619.
Clokey, G.V. and Jacobson, L.A. 1986. The autofluorescent "lipofuscin
granules" in the intestinal cells of Caenorhabditiselegans are
secondary lysosomes. Mech Ageing Dev 35(1): 79-94.
Csallany, A.S. and Ayaz, K.L. 1976. Quantitative determination of organic
solvent soluble lipofuscin pigments in tissues. Lipids 11(5): 412-417.
Dhalla, N.S., Elimban, V., and Rupp, H. 1992. Paradoxical role of lipid
metabolism in heart function and dysfunction. Mol Cell Biochem
116(1-2): 3-9.
Feng, J., Bussiere, F., and Hekimi, S. 2001. Mitochondrial electron
transport is a key determinant of life span in Caenorhabditiselegans.
Dev Cell 1(5): 633-644.
Finch, C.E. and Ruvkun, G. 2001. The genetics of aging. Annu Rev
Genomics Hum Genet 2: 435-462.
136
Finn, P.F. and Dice, J.F. 2006. Proteolytic and lipolytic responses to
starvation. Nutrition 22(7-8): 830-844.
Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., and Mello,
C.C. 1998. Potent and specific genetic interference by doublestranded RNA in Caenorhabditiselegans. Nature 391(6669): 806811.
Fukao, T., Scriver, C.R., and Kondo, N. 2001. The clinical phenotype and
outcome of mitochondrial acetoacetyl-CoA thiolase deficiency (betaketothiolase or T2 deficiency) in 26 enzymatically proved and
mutation-defined patients. Mol Genet Metab 72(2): 109-114.
Garigan, D., Hsu, A.L., Fraser, A.G., Kamath, R.S., Ahringer, J., and
Kenyon, C. 2002. Genetic analysis of tissue aging in Caenorhabditis
elegans: a role for heat-shock factor and bacterial proliferation.
Genetics 161(3): 1101-1112.
Gershon, H. and Gershon, D. 2002. Caenorhabditiselegans--a paradigm for
aging research: advantages and limitations. Mech Ageing Dev 123(4):
261-274.
Gerstbrein, B., Stamatas, G., Kollias, N., and Driscoll, M. 2005. In vivo
spectrofluorimetry reveals endogenous biomarkers that report
healthspan and dietary restriction in Caenorhabditiselegans. Aging
Cell 4(3): 127-137.
Gonzalez, A.A., Kumar, R., Mulligan, J.D., Davis, A.J., and Saupe, K.W.
2004. Effects of aging on cardiac and skeletal muscle AMPK
activity: basal activity, allosteric activation, and response to in vivo
hypoxemia in mice. Am JPhysiolRegul Integr Comp Physiol 287(5):
R1270-1275.
137
Guarente, L. and Kenyon, C. 2000. Genetic pathways that regulate ageing
in model organisms. Nature 408(6809): 255-262.
Guo, W., Pirtskhalava, T., Tchkonia, T., Xie, W., Thomou, T., Han, J.,
Wang, T., Wong, S., Cartwright, A., Hegardt, F.G., Corkey, B.E.,
and Kirkland, J.L. 2007. Aging results in paradoxical susceptibility
of fat cell progenitors to lipotoxicity. Am JPhysiolEndocrinolMetab
292(4): E1041-1051.
Herndon, L.A., Schmeissner, P.J., Dudaronek, J.M., Brown, P.A., Listner,
K.M., Sakano, Y., Paupard, M.C., Hall, D.H., and Driscoll, M. 2002.
Stochastic and genetic factors influence tissue-specific decline in
ageing C. elegans. Nature 419(6909): 808-814.
Hsin, H. and Kenyon, C. 1999. Signals from the reproductive system
regulate the lifespan of C. elegans. Nature 399(6734): 362-366.
Huang, C., Xiong, C., and Kornfeld, K. 2004. Measurements of age-related
changes of physiological processes that predict lifespan of
Caenorhabditiselegans. ProcNatl Acad Sci USA 101(21): 80848089.
Huyghe, S., Mannaerts, G.P., Baes, M., and Van Veldhoven, P.P. 2006.
Peroxisomal multifunctional protein-2: the enzyme, the patients and
the knockout mouse model. Biochim Biophys Acta 1761(9): 973-994.
Ibdah, J.A., Perlegas, P., Zhao, Y., Angdisen, J., Borgerink, H., Shadoan,
M.K., Wagner, J.D., Matem, D., Rinaldo, P., and Cline, J.M. 2005.
Mice heterozygous for a defect in mitochondrial trifunctional protein
develop hepatic steatosis and insulin resistance. Gastroenterology
128(5): 1381-1390.
Kamath, R.S., Fraser, A.G., Dong, Y., Poulin, G., Durbin, R., Gotta, M.,
Kanapin, A., Le Bot, N., Moreno, S., Sohrmann, M., Welchman,
138
D.P., Zipperlen, P., and Ahringer, J. 2003. Systematic functional
analysis of the Caenorhabditiselegans genome using RNAi. Nature
421(6920): 231-237.
Kao, H.J., Cheng, C.F., Chen, Y.H., Hung, S.I., Huang, C.C., Millington,
D., Kikuchi, T., Wu, J.Y., and Chen, Y.T. 2006. ENU mutagenesis
identifies mice with cardiac fibrosis and hepatic steatosis caused by a
mutation in the mitochondrial trifunctional protein beta-subunit. Hum
Mol Genet 15(24): 3569-3577.
Kenyon, C. 2005. The plasticity of aging: insights from long-lived mutants.
Cell 120(4): 449-460.
Kenyon, C., Chang, J., Gensch, E., Rudner, A., and Tabtiang, R. 1993. A C.
elegans mutant that lives twice as long as wild type. Nature
366(6454): 461-464.
Klass, M.R. 1977. Aging in the nematode Caenorhabditiselegans: major
biological and environmental factors influencing life span. Mech
Ageing Dev 6(6): 413-429.
Koenig, H. 1964. Neuronal lipofuscin in disease. Its relation to lysosomes.
Trans Am Neurol Assoc 89: 212-213.
Kujoth, G.C., Leeuwenburgh, C., and Prolla, T.A. 2006. Mitochondrial
DNA mutations and apoptosis in mammalian aging. Cancer Res
66(15): 7386-7389.
Mak, H.Y., Nelson, L.S., Basson, M., Johnson, C.D., and Ruvkun, G. 2006.
Polygenic control of Caenorhabditiselegans fat storage. Nat Genet
38(3): 363-368.
McGarry, J.D. 1998. Glucose-fatty acid interactions in health and disease.
Am J Clin Nutr 67(3 Suppl): 500S-504S.
139
Mello, C.C., Kramer, J.M., Stinchcomb, D., and Ambros, V. 1991. Efficient
gene transfer in C.elegans: extrachromosomal maintenance and
integration of transforming sequences. Embo J 10(12): 3959-3970.
Montine, T.J. and Morrow, J.D. 2005. Fatty acid oxidation in the
pathogenesis of Alzheimer's disease. Am JPathol166(5): 1283-1289.
Mori, S., Morisaki, N., Saito, Y., and Yoshida, S. 1988. Age-related
changes of long-chain fatty acid metabolism in rat liver. Mech Ageing
Dev 44(2): 175-183.
Nogalska, A., Pankiewicz, A., Goyke, E., and Swierczynski, J. 2003. The
age-related inverse relationship between ob and lipogenic enzymes
genes expression in rat white adipose tissue. Exp Gerontol 38(4):
415-422.
Ockner, R.K. 2004. Integration of Metabolism, Energetics, and Signal
Transduction
Ogg, S., Paradis, S., Gottlieb, S., Patterson, G.I., Lee, L., Tissenbaum, H.A.,
and Ruvkun, G. 1997. The Fork head transcription factor DAF-16
transduces insulin-like metabolic and longevity signals in C. elegans.
Nature 389(6654): 994-999.
Paradies, G., Ruggiero, F.M., Petrosillo, G., Gadaleta, M.N., and
Quagliariello, E. 1995. Carnitine-acylcarnitine translocase activity in
cardiac mitochondria from aged rats: the effect of acetyl-L-camitine.
Mech Ageing Dev 84(2): 103-112.
Reddy, J.K. and Mannaerts, G.P. 1994. Peroxisomal lipid metabolism. Annu
Rev Nutr 14: 343-370.
Reddy, J.K. and Rao, M.S. 2006. Lipid metabolism and liver inflammation.
II. Fatty liver disease and fatty acid oxidation. Am JPhysiol
GastrointestLiver Physiol 290(5): G852-858.
140
Russell, R.L. and Seppa, R.I. 1987. Genetic and environmental
manipulation of aging in Caenorhabditiselegans. Basic Life Sci 42:
35-48.
Shepherd, D., Yates, D.W., and Garland, P.B. 1966. The rate-limiting step
in the oxidation of palmitate or palmitoyl-coenzyme A by rat-liver
mitochondria. Biochem J 98(1): 3C-4C.
Shigenaga, M.K., Hagen, T.M., and Ames, B.N. 1994. Oxidative damage
and mitochondrial decay in aging. ProcNatl Acad Sci US A 91(23):
10771-10778.
Singh, K.K. 2006. Mitochondria damage checkpoint, aging, and cancer.
Ann N YAcad Sci 1067: 182-190.
Slawik, M. and Vidal-Puig, A.J. 2006. Lipotoxicity, ovemutrition and
energy metabolism in aging. Ageing Res Rev 5(2): 144-164.
Tissenbaum, H.A. and Ruvkun, G. 1998. An insulin-like signaling pathway
affects both longevity and reproduction in Caenorhabditiselegans.
Genetics 148(2): 703-717.
Tolwani, R.J., Hamm, D.A., Tian, L., Sharer, J.D., Vockley, J., Rinaldo, P.,
Matern, D., Schoeb, T.R., and Wood, P.A. 2005. Medium-chain acylCoA dehydrogenase deficiency in gene-targeted mice. PLoS Genet
1(2): e23.
Toth, M.J. and Tchernof, A. 2000. Lipid metabolism in the elderly. Eur J
Clin Nutr 54 Suppl 3: S121-125.
Wallace, D.C. 2005. The mitochondrial genome in human adaptive
radiation and disease: on the road to therapeutics and performance
enhancement. Gene 354: 169-180.
Wang, Y., Sebum, K., Bechtel, L., Lee, B.Y., Szatkiewicz, J.P., Nishina,
P.M., and Naggert, J.K. 2006. Defective carbohydrate metabolism in
141
mice homozygous for the tubby mutation. Physiol Genomics 27(2):
131-140.
Weinberg, J.M. 2006. Lipotoxicity. Kidney Int 70(9): 1560-1566.
Wicks, S.R., Yeh, R.T., Gish, W.R., Waterston, R.H., and Plasterk, R.H.
2001. Rapid gene mapping in Caenorhabditiselegans using a high
density polymorphism map. Nat Genet 28(2): 160-164.
Wolfe, R.R. 1998. Metabolic interactions between glucose and fatty acids
in humans. Am J Clin Nutr 67(3 Suppl): 519S-526S.
Yin, D. 1996. Biochemical basis of lipofuscin, ceroid, and age pigment-like
fluorophores. Free Radic Biol Med 21(6): 871-888.
142
FIGURES AND TABLES
Table 1. Characterization of screen isolates
Mutation Phenotype
wild-type
Chromosome Complementation Locomotory
decline
group
Pumping
decline
Mean lifespan, da3
200C(250 C)
-
-
normal
normal
21(12)
n4430
Weakly
semidominant
II
1
premature
premature
14(11)
n4434
Weakly
semidominant
II
1
premature
premature
14(9)
n4435
Weakly
semidominant
II
1
premature
premature
14(10)
n4501
Recessive
II
2
normal
normal
18
n4427 Semidominant
V
3*
normal
normal
16(11)
n4428 Semidominant
V
3*
normal
normal
19(11)
n4429 Semidominant
V
3*
normal
normal
19(11)
n4431 Semidominant
V
3*
normal
normal
19(11)
n4432 Semidominant
V
3*
normal
normal
12(10)
n4433 Semidominant
V
3*
normal
normal
19(11)
n4436 Semidominant
V
3*
normal
normal
16(11)
* these mutants might or might not be allelic, as they cause a semidominant
phenotype.
143
Table 2. Epistasis studies of kat-1 with genes in other aging pathways
Pathway
Insulin-like
pathway
Mitochondrial
respiration
Sensation
Gemiline
signaling
Genotype
Median
life span
N/
# of trials
wild-type (20C125c*)
20/12
>500/>10
kat-1(n4430) (20c25c*)
17/10
>500/>10
daf-2(e1370)*
35
76/2
kat-1(n4430);daf-2(e 370) *
32
92/2
daf-16(mgDf50)
12
157/2
daf-16(m gDf50);kat-1(n4430)
11
261/3
isp-I(qml50)*
14
42/1
isp-1(qml50);)kat-1(n4430)*
12
44/1
osm-5(pr813)
41
50/2
kat-l (n4430);osm-5(pr813)
38
43/2
che-3(el 24)
26
50/2
che-3(e1124); kat-1 (n4430)
25
47/2
mes-l(bn74)
gonad +
19
44/2
gonad -
31
54/2
gonad +
18
53/2
gonad -
24
36/2
kat-1(n4430);
mes-1(bn74)
CR
wild-type, CR
49/1
kat-1(n4430), CR
44/1
*, experiments performed at 25 0 C; CR, calorie restriction.
144
Table 3. n4430 mutants display premature tissue aging
Gonad
Electron-dense
deposits
I
1
1
wild type day 8 (15)
1.5C
1.6
2.1
wild type day 13 (5)
1.8
3.2
2.8
type day 18 (5)
Iwild
wild type day 23 (4)
2.4
3.4
3.2
3.25
4
3.75
n4430 day 2 (5)
1i
1
1.8
n4430 day 8 (14)
2.7
3.6
3.2
Worms analyzed (n)
Gut
wild type day 2 (5)
Key:
1 = no decline, i.e. youthful-looking tissue/ no deposits
2 = tissues showing early markers of degeneration/ few deposits
3 = marked degeneration of tissues/ deposits cover up to 50% of section
4 = total degeneration, i.e. organ is indistinguishable/deposits cover over 50% of section
Electron microscopy images were analyzed for the accumulation of electrondense deposits and abnormalities in gonad and intestine. Each animal was
given a score for gonad, intestine, and electron-dense deposits. The numbers
represent mean scores of all animals in the group. For example, the "Gut" score
of 3.25 for 23-day-old wild-type adults indicates that 3 out of 4 animals had
completely degenerated intestines scored as 4, and 1 animal had markedly
degenerated yet distinguishable intestine, scored as 3.
145
Table 4. Analyses of lipofuscin accumulation and lifespan in animals
subjected to RNAi knock down of fatty acid beta-oxidation genes.
cpt-1 Y46G5A.17
nd
Lifespan
(%control)
nd
cpt-2 R07H5.2
++
short (75)
F37C12.7
++
dead
R09E10.3
++
short (75)
acs-1 7 C46F4.2, F47G6.2,
R07C3.4, R09E10.4
n
n
Y76A2B.3
C48B4.1, F25C8.1
n
n
short (72)
short (70-85)
F08A8.3, F08A8.4, T10E9.9,
F58F9.7, F28A10.6, K05F1.3,
n
n
ech-2 F38H4.8
++
short (75)
ech-9 FOIG10.3
++
short (85)
Lipofuscin
Step in beta-oxidation (Enzyme)
FA transport into mitochondria
(Carnitine O-palmytoyl transferase)
Gene(s)
CoA binding
Fatty Acid acyl-coA ligase/ Acyl coA
synthetase
FA breakdown
Fatty Acid acyl-coA oxidase/
Fatty Acid acyl co-A dehydrogenase/
Enoyl CoA hydratase
K06A5.6, T25G12.5
ard-1 FOIG4.2
Iva (na)
ech-1 C29F3.1, ech-8
FOIG10.2, ech-3 F43H9.1
n
n
ech-7 Y105E8A.4, B0272.3,
hacd-1 R09B5.6
n
short (75-85)
TO2G5. 7
n
short (85)
n
kat-1 T02G5.8
+++-
short (85)
Acetyl coA release
Acetoacetyl co-A thiolase
T02G5.4
nd, not done; n, normal; Iva, larval arrest; na, not applicable; ++/+++ indicate increased
lipofuscin accumulation.
146
Figure 1. Old worms accumulate autofluorescent pigment, and amount of the
pigment correlates with lifespan in aging mutants. A. Intestinal autofluorescence of
1-day-old (top) and 12-days-old (bottom) wild-type adult hermaphrodites was captured
using a DAPI filter (excitation 365-380 nm, emission 450 nm), with an exposure time of
1000 ms., magnification 50x. Here and elsewhere ages of animals are indicated as days
post-development, i.e. day 1 is first day of adulthood. B. Emission spectrum of
fluorescence of worm extracts prepared from wild-type animals of indicated ages, at 350
nm excitation. The fluorescence was measured using by spectroflurimetry. At least 100
animals were used for each extract (see Experimental Procedures for details). C, D.
Emission spectrum of fluorescence of worm extracts prepared from 2-days-old (C) and
8-days-old (D) wild-type adults (wt, blue curves), long-lived daf-2(e1370) mutants
animals (purple curves), and short-lived daf-16(mgDf50) (cyan curves) and sir2.1 (ok434) (yellow curves).
147
Figure 1.
B.
A.
14000
12000
I-day-old
10000
wild type
8000
OOF,-
W
-
--
--
-
--
6000
4000
2000
12-day-old
wild type
77T7- I
0
;50
390
430
470
510
550
Emissio,
-day
C-
2
-day
5
-day
8
-day
590
nam
12
D.
I0R00
12000
10000
10000
BA~ill
4000
6000
4000
4000
2000
2000
S
350
t
350
390
430
470
510
550
o90
3
430
470
510
550
590
Emissbi , sm
ir-21 d1iy
Am
daf4My
Emissiona m
-
wtday2 -duf-2di2
wir-2•dry2
d-16day2 I
148
L-
wtday8 -d-•lqAv
Figure 2. A screen for mutants with high accumulation of
lipofuscin-like fluorescent pigment. We screened synchronized F2
progeny of EMS-mutagenized hermaphrodites at day 3 of adulthood for
high lipofuscin-like autofluorescence. Initial isolates were retested in the
next generation.
149
Figure 2.
EMS/
Fl adults
I
Bleach to synchronize progeny
Screen young adult progeny for high autofluorescence
-"
Retest candidates for high autofluorescence of young adults
150
Figure 3. Mutations in gene specified by n4430, n4434, and n4435 result
in increased fluorescence and short lifespan
A. Intestinal lipofuscin of 3-day-old wild-type and n4430 adults. B. n4430,
n4434, and n4435 animals have shortened lifespans. Median lifespans were
as follows: N2, 18 days, n=79; n4430, 13 days, n=105; n4434, 16 days,
n=95; n4435, 13 days, n=122.
151
Figure 3.
B.
A.
n4430
wild type
Lipofuscin
accumulation
• .
L,
80
~~Ctl;~7
60
40
= 20
wild type
C.
days
-
wild type
T kat-1 (n4430)
Skat-1(n4434)
- kat-1(n4435)
152
n4430
7-
Figure 4. Behavior abnormalities of n4430, n4434, and n4435 mutants.
A, B. Decline in locomotion rate (A) and in pharyngeal pumping rate (B)
with age was assessed in wild-type animals (grey bars), three LGII mutants
n4430, n4434, n4435 (blue and light blue bars), and two chromosome V
mutants with high intestinal fluorescence, n4427 (scarlet bars) and n4433
(pink bars). Three LGII mutants behave normally as young adults, but show
premature decline in locomotion and pumping evident from day 9 of their
lifespan. C. Aging animals were classified into four groups based on their
locomotion: A (green bars), animals with robust coordinated sinusoidal
locomotion; B (yellow bars), uncoordinated and/or sluggish animals; C (red
bars), animals that do not move forward or backward (but do move their
heads or shudder upon prodding); D (black bars), paralyzed/dead animals.
Four allelic mutants were tested: n4430 (1), n4434 (2), n4435 (3), and
tm1037 (see below) (4). Wild-type animals (N2) and progeroid daf16(mgDf50) mutants were used as controls.
153
Figure 4.
A. ·.E
40
SrN2
35
I,1,
I w4427
Ili
I
il
IT1435
II
0
day 1
day 5
day 9
day 13
day 16
SN2
B. .1
In• ,••.t
Pi
a
II
,
PI
III
b6
a
E3
I
k
class
FB
U
I1
-"
day 1
Cc
100%
llI
day 5
day 5
day 9
day 9
--
day 13
IN
-
20%
1234
z
day 21
day 27
11
-1111111111 111111 IF
40%
11234
day 16
day 13
60%
0%
day 16
1234
Iz
1234
MS --F
154
1 1234
1234
ze4
t ýZzu it~
Figure 5. kat-1 tissue aging. Electron microscopy analyses were
performed to look at tissue aging of kat-] mutants. All animals of a given
age were grown identically and fixed on the same day. Figures show midbody cross-sections of wild-type worms and n4430 mutants of indicated
ages. Eight days old n4430 mutants show degeneration of the gonad and the
intestine, with evident packing of bacteria in the lumen and increased
necrotic membrane swirls (red arrows), which are present in the wild type
only at later time points, and accumulation of electron-dense material (red
arrowheads). i, intestine; g, gonad.
155
Figure 5.
I
I
I
.
I
wt, dayI 2
wt day 8
Ibo
Sc.
wt. day 131
=me.,-,.q
ma
r
L
j,
•.
vt
---
day
13
e.
f.
n4430. davy 8 1
J.
wk
156
day
18
re
R.
4430, day 8
n4430 dav8 I
Figure 6. Mapping and cloning of kat-1. A. Three-point mapping placed
the n4430 mutation right of or close to unc-104 on chromosome II. B. The
155 kb n4430-containingregion identified by polymorphism mapping is
covered by three cosmids, one of which (C06B7) rescued the fluorescence
accumulation by n4430 animals. C. Mutation sites and amino acid changes
of the three alleles of T02G5.8 identified in our screen and the e1001 allele.
157
Figure 6.
A.
-3.22
-0.98
I--I_
1.75
0 0.21
2.85
m.u.
B.
C15F1
T02G5
C52E12
ZK1127
C06B7.•
...
,,
i
Lines rescued/total:
Cosmid:
/ /·
.•
C15F1]
/
C52E12 .............................. 11/11
C06B7
i
C SF1 ................................... 0/3
C52E12 .................................. 0/4
C06B7..................................7/7
T02G5.9
TO2GS.Sl
ZK1127 ................................... 0/3
'-,
,,_
T02GS.6
msh-4
hat-1
krs-1
TO2GS.7
A
A
T02G5.8
II
A4434
G134E
1
II
I I
n4435 4430
elOOl4
C393Y
G368E
A279T
A273V,
tm103 7 deletion
0.1 kb
158
Figure 7. kat-1 controls fluorescence accumulation and lifespan. A.
Intestinal fluorescence accumulation in young (1-day-old), middle-aged (5days-old) and old (12-days-old) n4430 animals compared with wild-type
nematodes (N2), progeroid daf-16(l animals, and n4430 animals injected
with genomic fragment containing T02G5.8 ORF and 2.5 kb of upstream
promoter sequence. Images were taken as in Figure 1. B. Lifespan of n4430
animals compared with wild-type (N2) lifespan and n4430 animals injected
with the rescuing genomic fragment (as in A). Median lifespans were as
follows: N2, 18 days, n=109, kat-](n4430), 15 days, n=95, kat-](n4430);
unc-76(e911); nls]234 [unc-76, T02G5.8), 18 days, n=83.
159
Figure 7.
A.
Wild type
kat1(n4430)
day I
day 5
day 12
B.
days
-l- Wild type
* n4430
n4430 + T02G5.8
160
kat-1(n4430);
alerrnAr;
RI
Figure 8. kat-1 encodes a conserved ketoacyl thiolase. Alignments of
KAT-1 amino acid sequence with sequences of human, mouse, and
drosophila homologous proteins. Dark blue, amino acids identical in all
four sequences. Light blue, sequences identical in 2-3 of the four sequences.
Arrows indicate amino acids mutated in the screen isolates.
161
Figure &
IML-SSGHRI
---------------
FlyACA
IEV CLERSVRSKPTL
-- MRALVYALHOVVRRPLLRO
H]• Aca1 MIVLRALLRSGARSRSPLLRR
-
ki::
I
YAT-1T
fly ACAT1
iS
RRITTSRRLSH
I YVERSYVSKPTL
SQ
Liv
TNY
10
I T
UI, AA
K TA
I
PL
10
U
SV
E
S
GSLL
j
L-
ssv
BID 0
IS-+
KI,
KAT-I
RyACAIf
MBm Acail
Hýý Acai
MG TK
KAT-1
Fly ACATI
Mamse ACaml
I~m •Acal
'"-
-- I
7
411P
LE Q
I
OME
S
WON
VIKS
A
K
T
Aý'
T
.IENI
TISl
I
I
S
Mous Acatl
A=ctI
Hmusu
HaT
allfl] A271T--
0vDV
0I
1
MSeI
A279V
EQSL
KAGI
RLN%
I
R LM \
I
Fy ACATI
Huma Acld
Marn Acatl
[man Arui
D
AnVTV
Lu
Fly ACATi
EAT-I
V
-
A H oo a KIQ00Q
i
L
(
SM
anM
LTOF I P
LI
I. V
· Y
Sil V OVv
I
1
v
3
U
_ ,&r.c
f
7i im IVi
-••"S
A"ILLSM
L
JE
nm
-
U.-
niI
Figure 9. kat-1 stress resistance. A. kat-1 mutants do not show
increased sensitivity to heat shock. Animals were subjected to 32C heat
shock at first day of adulthood, and survival was measured during the next
64 hours. At least 60 animals are presented in each curve. B. kat-1 mutants
do not show increased sensitivity to oxidative stress. Wild-type and kat1 (n4430) animals were allowed to lay eggs on plates containing 25 mM and
50 mM paraquat. Fraction of animals developing to adulthood was
measured after 4 days and 6 days respectively. Development of least 300
eggs per PQ concentration per genotype was monitored. A representative
experiment of three independent experiments is shown. C. Survival of
arrested Ll larvae from wild type and kat-1 mutants in M9 buffer lacking
bacteria was monitored every 2 to 3 days. At every time point, samples
containing approximately 100 animals were observed in a drop of M9, and
percent of swimming animals was determined. The numbers shown are the
means of three samples per time point per genotype.
163
Figure 9.
A.i
It-
-
9..
777
so
. * kat-f (n4430)
rr
dF-2(el370)
--& At-I (n4430); da-2(ef370)
a
Be74
am
,,
I
0
Pua
B.
estanc
Paraquat resistance
10t0
so
70
60
50
Swildtyp-qe
g W-l·n~f~q
40
30
20
10
0
0.5s
at
P&,IULM
149 starvation toQle
C
nce
60
mwi type
pa
k4sii4
40
. a-·tmes
20
I
9
13
days
3
Figure 10. 3-ketoacyl thiolase catalyzes the last step in fatty acid beta-oxidation
pathway. Fatty acid oxidation starts by binding of a CoA group to fatty acids,
catalyzed by acyl-CoA synthases; the resulting acyl-CoAs bind camitine (in the
reaction catalyzed by camitine palmitoyltransferase), which facilitates entry into
mitochondrial matrix. Inside mitochondria, acyl-coAs are converted to enoyl-CoAs by
acyl-CoA oxidases and acyl-CoA dehydrogenases, enoyl CoA is reduced by enoylCoA hydratase into 3-hydroxyacyl-CoA. 3-hydroxyacyl-CoA dehydrogenase then
catalyzes production of 3-Ketoacyl-CoA. Finally, the last step of the fatty acid
oxidation is catalyzed by a 3-ketoacyl thiolase (e.g. KAT-1 in C. elegans) and it
involves binding of another CoA group and cleavage of a thiol bond that produces
Acetyl-CoA and Acyl-CoA chain shortened by two carbons (see box). This Acyl-CoA
molecule can enter the cycle again to be processed further.
165
Figure 10.
Fatty acid + CoA
Acyl-CoA synthetase
Carntine palmitoyltransferase
Acyl-CoA
-
*I.-o
Acyl-CoA oxidase
Acyl-CoA dehydrogenase
Enoyl-CoA
Enoyl-CoA hydratase
3-Hydroxyacyl-CoA
3-Hydroxyacyl-CoA dehydrogenase
3-Ketoacyl-CoA
3-Ketoacyl-CoA thiolase
CoA
0
0
R-CH -CH -C-C-CoA
2
52
COA-SH
o
R-CH 2 -C -CoA
~2t
Acetyl-CoA
0
+
CHl-C-CoA
3
166
Acyl-CoA
Figure 11. Reducing fatty acid oxidation genes causes increased
lipofuscin accumulation in C. elegans. Intestinal autofluorescence of 3days-old wild-type animals grown on bacteria expressing indicated RNAi
clones was photographed as described in Figure 1. Note that RNAi
knockdown of ard-1 (acyl-CoA dehydrogenase) resulted in larval arrest at
L2-L4 stages. Rare escapers that reached adulthood were small and sterile.
The image below shows ard-1 (RNAi) adult escapers.
167
Figure 11.
kat-1 (RNAi)
Vector (RNAi)
TO2GS.4 thiolase (RNAi)
168
ard-1 (RNAi)
Addendum to Chapter 3.
FLU-1 kynurenine monooxigenase controls
intestinal fluorescence accumulation in C. elegans.
169
ABSTRACT
Characterization of mutants isolated in genetic screens for increased
accumulation of autofluorescence led to identification offlu-1 gene, which
controls intestinal fluorescence levels in C. elegans. In our screen, we
isolated three alleles offlu-1, all of which confer high levels of intestinal
autofluorescence. Some but not all mutations influ-1 were associated with
reduced longevity, and none resulted in premature behavioral decline,
indicating that whileflu-1 may control some aspects of aging, it is unlikely
to be involved in the general regulation of aging process. We cloned the
gene mutated influ-1 and found thatflu-1 encodes a kynurenine
monooxygenase, an enzyme involved in tryptophan metabolism.
170
RESULTS
Characterization of LGV isolates.
In our screen for C. elegans mutants with premature accumulation of
intestinal autofluorescence, we isolated seven mutants that mapped to
chromosome V (see Chapter 3 Table 1). Mutations in these isolates semidominantly cause increased accumulation of autofluorescence. Lifespan
analysis of the seven isolates showed that four mutants had normal
lifespans, whereas three mutants, n4427, n4432, and n4436 had shorter
lifespans than the wild type (Table 1). Quantitation of the autofluorescence
in n4427 and n4432 showed that these mutations resulted in strong
autofluorescence accumulation, much stronger than that seen in a daf-16
progeroid mutant used as a control (Figure 1). The semi-dominant
phenotype of these mutants complicated the interpretation of
complementation analysis. Genetic mapping indicated that all seven
mutations were close to each other on chromosome V: all showed similar
linkage to dpy-11 on LGV (data not shown), raising the possibility that they
were alleles of the same gene. We set to map and clone the gene mutated in
one of the shorter-lived mutants, n4427.
Mapping and cloning offlu-1.
We used a combination of positional cloning, polymorphism cloning
and RNAi to map and clone the gene mutated in n4427. We first located it
to the right of or close to dpy-11 on V (see Figure 2A): in the cross between
n4427 and unc-46 dpy-11, 20/20 Unc-non-Dpy recombinants showed high
autofluorescence, and 0/7 Dpy-non-Unc recombinants showed high
autofluorescence. We then mapped it to the right of or close to unc-23: in
171
the cross between n4427 and dpy-11 unc-23, 0/16 Unc-non-Dpy
recombinants showed high autofluorescence, and 3/3 Dpy-non-Unc
recombinants showed high autofluorescence. Next- we mapped the
mutation to the left of unc-76: in the cross between n4427 and dpy-11 unc76, 18/34 Unc-non-Dpy recombinants showed high autofluorescence, and
13/27 Dpy-non-Unc recombinants showed high autofluorescence. Finally,
we mapped the mutation to the right of or close to lon-3 and left of or close
to rol-4-(Figure2A): in the cross between n4427 and unc-42 lon-3, 5/5
Unc-non-Lon recombinants showed high autofluorescence, and 0/15 Lonnon-Unc recombinants showed high autofluorescence. Similarly, in the
cross between n4427 and rol-4 dpy-21, 0/26 Rol-non-Dpy recombinants
showed high autofluorescence, and 4/4 Dpy-non-Rol recombinants showed
high autofluorescence.
We then double-cis-marked n4427 with dpy-11 and unc-76 to use the
resulting strain in single nucleotide polymorphism (SNP) mapping. We
performed SNP mapping as described (Wicks et al. 2001), and were able to
narrow down the region to approximately 450 kbp interval on chromosome
V. The above region was covered by 11 cosmids (Figure 2B). Eight of these
eleven cosmids, when injected in pooled combinations, did not rescue
fluorescence accumulation in n4427. We then used RNAi to knock down
the candidate genes in the region determined by SNP mapping, including
genes not covered by the injected cosmids (see Figure 2C). We performed
the RNAi experiments in an eri-I genetic background, which is
hypersensitive to RNAi (see Chapter 3 Experimental Procedures for
details). Of the 48 RNAi clones tested, one RNAi clone, RO7B7.5, resulted
in a striking accumulation of autofluorescence in eri-1 animals. We
sequenced the coding regions of RO7B7.5 in n4427 and found a mutation in
172
the fourth exon of this gene. DNA sequencing of RO7B7.5 in all other LGV
isolates revealed that two other isolates n4432 and n4433 also carried
mutations in this gene. Notably, n4433 animals, which carried a nonsense
mutation in RO7B7.5 did not show lifespan abnormalities despite high
accumulation of fluorescence (see Discussion). In the other chromosome Vlinked isolates, n4428, n4429, n4431, and n4436, we did not find mutations
in the coding regions of RO7B7.5 (Table 1). These isolates may represent
alleles of other gene(s), or they can carry mutations in the non-coding
regulatory regions of RO7B7.5. Sequencing of introns and promoter regions
of RO7B7.5 in these mutants should answer this question.
Babu (1974) performed a genetic screen for mutants with
abnormal autofluorescence. One of the genes he described,flu-1, mapped to
chromosome V and exhibited similar effects on gut autofluorescence to the
RO7B7.5 mutants. The molecular identity of theflu-1 gene was unknown,
but extracts offlu-1 mutant animals showed decreased kynurenine
monooxygenase activity (Siddiqui and Babu 1980). We sequenced RO7B7.5
in theflu-1(e1002) allele and identified a mutation in the coding region of
the gene. Interestingly, the mutation in e1002 changed the same amino acid
as the mutation in n4427 (see Figure 2D). Therefore, RO7B7.5 isflu-1, and
we will use this name below.
flu-i encodes a conserved metabolic enzyme, kynurenine 3monooxygenase.
R07B7.5 flu-i encodes a protein with high similarity to metabolic
enzyme involved in the oxidative metabolism of tryptophan, kynurenine 3monooxygenase (EC1.14.13.9). The C. elegans FLU-1 protein is 45%
identical and 68% similar to human Kmo-1 (Figure 3). This enzyme uses
173
oxygen, NADPH, and FAD for the catalytic oxidation of L-kynurenine to 3hydroxykynurenine and water (Figure 4A). Kynurenine is a key compound
in the kynurenine pathway, a major pathway for tryptophan metabolism in
mammals, which is involved in synthesis of nicotinic acid (Fig 4B).
Previous studies by Babu et al showed thatflu-1 (el 002) animals
accumulate high amounts of anthranilic acid and have low concentrations of
3-hydroxy-kynurenine caused by the block in this step of the pathway
(Siddiqui and Babu 1980 and Figure 4B). That the nonsense mutation in
n4433 and RNAi knock down of RO7B7.5 result in high accumulation of
gut autofluorescence suggests that this phenotype is caused by loss of
function of kynurenine monooxygenase. In accordance with this, extracts
fromflu-i(e1002) mutants had less than 10% of the wild-type kynurenine
hydroxylase specific activity (Siddiqui and Babu 1980). The semidominant
nature of theflu-1 mutations implies haplo-insufficiency, suggesting that
the kynurenine pathway in C. elegans is dose-sensitive.
174
DISCUSSION
The essential amino acid tryptophan is metabolized via several
metabolic pathways, which result in the generation of several biologically
important molecules. Tryptophan metabolites include serotonin, the 'sleep'
hormone melatonin, an odor molecule indole, and the neurotransmitter
tryptamine (Moroni 1999). The main pathway of the tryptophan metabolism
is the kynurenine pathway involved in the biosynthesis of quinolinic acid,
the precursor for nicotinic acid and NAD. Tryptophan metabolism and
kynurenine pathway in particular are important for normal brain function,
because different intermediates of the kynurenine pathway can act as
agonists or antagonists of certain glutamate receptors (Moroni 1999).
Imbalances of the kynurenine pathway metabolism are implicated in
numerous neurodegeneration diseases, such as Huntington's disease,
Alzheimer's disease, ALS, and AIDS-dementia (Stone et al. 2001; Stone et
al. 2003).
We isolated mutants in a kynurenine monooxygenase geneflu-1 in a
screen for C. elegans mutants with elevated intestinal autofluorescence. We
obtained at least three alleles of theflu-1 gene, all of which have mutations
in the coding region offlu-1 and likely result in loss-of-function of this
gene. flu-1 was first characterized by Babu (1974), who isolated aflu-1
mutant in a similar screen. We have now determined the molecular lesion in
the original mutant strain and shown thatflu-1 encodes a conserved
metabolic enzyme kynurenine monooxygenase. Kynurenine
monooxygenase catalyzes the conversion of kynurenine to 3hydroxykynurenine. Accordingly, biochemical studies by Babu and
colleagues showed thatflu-1 mutants have decreased levels of 3175
hydroxykynurenine and increased levels of anthranilic acid, another product
of kynurenine.
The idea behind our screen was to identify mutants that show
increased accumulation of aging pigment lipofuscin and are aging
prematurely. However, several observations suggest thatflu-1 is not
involved in the control of aging rate in C. elegans. First, the lifespan of a
flu-1 (n4433) mutant, which has an early stop mutation in the flu-1 gene, is
not shorter than that of wild type. Theflu-i (el 002) mutant also has wildtype lifespan. Two otherflu-1 alleles, n4427 and n4432 confer shorter
lifespans than the wild type; however, both mutants have missense
mutations influ-1. The amino acid mutated in n4427 is identical to the
amino acid mutated in e1002, even though the resulting substitutions are
different. These data suggest that short lifespan observed in n4427 and
n4432 may not be specific to theflu-i loss-of-function. However, it is still
possible that mutations in n4432 and n4427 result in a dominant-negative
effect and cause stronger lifespan phenotype than n4433 loss-of-function.
Rescue experiments reintroducing wild-typeflu-1 gene into n4432 and
n4427 strains should clarify this issue.
Animals mutated influ-1 show no gross abnormalities apart from
high fluorescence accumulation. None of the flu-1 mutants showed
behavioral abnormalities typical of premature aging mutants, such as
progressive decline in locomotion and pharyngeal pumping with age (see
Chapter 3 Figure 6), suggesting thatflu-1 does not generally control the rate
of aging in C. elegans, and at most is involved in regulation of some aspects
of the nematode aging.
176
CONCLUSIONS AND FUTURE DIRECTIONS
To conclude, flu-] mutants may not be involved in aging regulation,
but they may be useful in the studies of the kynurenine pathway and
tryptophan biology. In particular, additional genetic screens for animals
with abnormal fluorescence accumulation may result in identification of
mutants in other steps of these pathways. Screens for suppressors of high
fluorescence accumulation may identify upstream and downstream
regulators of the pathway and shed light on the structural aspects of the
FLU-1 kynurenine monooxygenase. It would be interesting to determine
whetherflu-1 mutants have any defects in signal transduction in the nervous
system, as young adults or as aging adults, that would parallel the
abnormalities caused by deregulation of tryptophan metabolism in
mammals. Finally, efforts are underway to develop therapeutic agents that
target the kynurenine pathway (Schwarcz 2004). The profound increase in
autofluorescence in worm kynurenine pathway mutants may make C.
elegans nematodes useful for screening of potential drug candidates that
target enzymes of the kynurenine pathway.
177
References
Babu, P. 1974. Biochemical genetics of Caenorhabditiselegans. Molec Gen
Genet 135: 39-44.
Moroni, F. 1999. Tryptophan metabolism and brain function: focus on
kynurenine and other indole metabolites. Eur JPharmacol375(1-3):
87-100.
Schwarcz, R. 2004. The kynurenine pathway of tryptophan degradation as a
drug target. Curr Opin Pharmacol4(1): 12-17.
Siddiqui, S.S. and Babu, P. 1980. Kynurenine hydroxylase mutants of the
nematode Caenorhabditiselegans. Mol Gen Genet 179(1): 21-24.
Stone, T.W., Behan, W.M., Jones, P.A., Darlington, L.G., and Smith, R.A.
2001. The role of kynurenines in the production of neuronal death,
and the neuroprotective effect of purines. JAlzheimers Dis 3(4): 355366.
Stone, T.W., Mackay, G.M., Forrest, C.M., Clark, C.J., and Darlington,
L.G. 2003. Tryptophan metabolites and brain disorders. Clin Chem
Lab Med 41(7): 852-859.
Wicks, S.R., Yeh, R.T., Gish, W.R., Waterston, R.H., and Plasterk, R.H.
2001. Rapid gene mapping in Caenorhabditiselegans using a high
density polymorphism map. Nat Genet 28(2): 160-164.
178
TABLES
Table 1. Three of the seven screen isolates with mutations that map to
chromosome V have coding mutations in the gene encoding FLU-1 R07B7.5
kinurenine 3-monooxygenase.
Allele
Mapped to
R07B7.5
Mean lifespan,
Decline in
Dominance chromosome
genotype
days 200 C / 250 C locomotion, pumping
wild type
na
na
WT
21/12
normal
e1002
strong SD
V
A137T
19/11, **
n4427
strong SD
V
A137V
16/11, p<0.05
normal
n4428
strong SD
V
WT*
19/11, **
normal
n4429
SD
V
WT*
19/11, **
normal
n4431
SD
V
WT*
19/11, **
normal
n4432
strong SD
V
P407L
12/10, p<0.05
normal
n4433
strong SD
V
Q 115ochre
19/11, **
normal
n4436
strong SD
V
WT*
16/11, p<0.05
normal
nd
SD, semidominant; na, not applicable; WT, wild type; *, only sequences of
coding regions were determined; ** , not significant (p>0.05).
179
FIGURES
Figure 1. Chromosome V mutants accumulate high amounts of
autofluorescence. Autofluorescence was quantitated in extracts of 3-dayold and 7-day-old animals of above genotypes. For each genotype/age, 100
or more animals were used to make the extract, and the amounts were
normalized by protein content. Fluorescence was measured using
spectrofluorimeter, with excitation of 350nm. The values in the table
represent the emission peak at 420 nm.
180
Figure 1.
12000 '
10000
r
8000
I
6000
C
'a
4M0o
] day 3
l day 7
C
)00
2]
0
I N2
N2
I
d•-la•
daf-16lf)
I
n4427 V
181
n4432 V
Figure 2. Mapping and cloning offlu-1. A. 3-point positional mapping
was used to mapflu-1 to a region between lon-3 and rol-4. B. SNP mapping
allowed to narrow down the region to the one shown, covered by 11
cosmids. Injections of 8 of these cosmids (shown in black) did not result in
rescue of the fluorescence accumulation phenotype. C. RNAi was used to
knock down genes present in the area defined by SNP mapping. RNAi of
R07B7.5 (red) resulted in strong accumulation of autofluorescence. D.
Structure of R07B7.5 flu-] as predicted in Wormbase with sites of
mutations and amino acid changes in screen isolates.
182
Figure 2.
tlr46
-. 6
w23
dPYI
tLc.6
4
,v1i
0-2
I Or
I)
--2
-2.36
K12G1 1
7.29
3.74 4.27
13.18
F58H1
C13C4
D2023
R31
R07B7
C52E4
C08F12
F10C2
C 02C8
3 recombinants
5 recombinants
RNAi:
-R311
R313
- --.3 .4 .5
-
.2
A
A
r
R07B7.5
I
0.1 kb
R07B7
F1OC2
R31.2
'
A
.14-4 - .4 .5
.6
-- *
4-.3
.7
A
I
n4433, Q15ochre n4427, A237V
e1002, A237T
183
A
I
A
In4432, P407L
Figure 3.flu-1 encodes a 3-kynurenine mooxygenase, a conserved
metabolic enzyme. Alignment of the C. elegans FLU-i R07B7.5 protein
with its closest C. briggsae homolog, and with the human 3-kinurenine
monooxygenase 1 Kmo-1. C. elegans FLU-1 is 48% identical and 69%
similar to Kmo- 1.
184
Fig•ure 3.
oeFLU-lp
obFLU-1p
human kmo-1
MDSSVIQ
ceFLU-1lp
cbFLU-1p
human kio-1
--n4433, Q>ochre
obFLU-lp
E
to---UIIII
human kmo-1
Illllllli~Plrm*rP41s~
oeFLU-lp
obFLU.lp
human kmo-1
S-"f 4 4 27,A>V; e1002, A>T
oeFLU-lp
obFLU-lp
human kmo-1
B~lldCrlClr
ceFLU-1p
obFLU-lp
humnm krno-1
(S H C
oeFLU-lp
obFLU-p
human kmo-1
s
77-
L
~rm
-
FA
-
D
nUDL
Q
LCLP
U
~rp( ~I
PslPrrrr~ IllrsrrrslPs~rssrro~
f-- n4432, P>L
ceFLU-lp
obFLU-lp
human kinmo-i
- IM 1
0
FF
I:
MS
IceFLU-1lp
obFLU-lp
human kmo-1
-UI
-P F
P-
RPWMiWIRHFRNTTCFPRKRVDSLEQISLISRLLSHSEILOT
Figure 4. Kynurenine metabolic pathway in tryptophan metabolism.
A. Enzymatic reaction catalyzed by FLU-1. Kinurenine 3-monooxygenase
catalyzes the convertion of L-kinurenine into 3-hydroxykinurenine. B.
Kinurenine 3-monooxygenase in Tryptophan metabolism. In addition to
serving as a building block for proteins, tryptophan can be metabolized by a
variety of pathways leading to generation of biologically active molecules
such as hormones, neurotransmitters and enzymatic cofactors. Kinurenine
3-monooxygenase catalyzes one of the reactions of the Kinurenine pathway
that leads to the production of Quinolinic acid, a precursor of NAD.
186
Figure 4.
A.
kynurenine 3monooxygenase
L-kyurankm
"M"ydroxykynmrenkin
B.
PROTEINS
A
Tryptamine
5-Hydroxytryptopha
kynureine
3-monooxygenase
Kynarenine
3-HydroxyI
Acetyl-c oA1
Anthranili acij[dE
Quinolinic Acid
NAD + NADP
Nicatinamide
ADPR
187
Melatonin
188
Appendix 1.
A Stress Response Pathway involving Sirtuins, Forkheads,
and 14-3-3 proteins.
Ala Berdichevsky and Leonard Guarente
This chapter was published:
Berdichevsky, A. and Guarente, L. 2006.
Cell Cycle 5(22): 1165-1177. Review.
Acknowledgments: We thank Nicholas Bishop and Robert H. Horvitz for
critical comments on this perspective.
189
Introduction. Sir2 and mechanisms of life span determination
The conserved protein deacetylase Sir2 can affect life span of yeast,
flies, and worms: introducing extra copies of the sir2 gene results in life
span extension, while deletion of sir2 shortens lifespan (Blander and
Guarente 2004). It is likely that revealing how sir2 affects longevity in
simple organisms can improve our understanding of mechanisms of human
aging.
What do we know about mechanisms of life span extension? From studies
of worms many mitochondrial genes have emerged as important for life
span determination: worms with loss-of-function mutations in genes
encoding components of the mitochondrial electron transport chain have
much longer life spans. However, the development and behavior of these
mutants are slowed down, resulting in a "slow-living" phenotype (Lee et al.
2003; Aguilaniu et al. 2005). Another conserved mechanism of life span
extension involves the insulin-like pathway, attenuation of which can
extend life span of worms, flies, and even mice (Kenyon 2005). In this
pathway, a signaling cascade originates at the insulin-like receptor and
eventually results in phosphorylation and exclusion from the nucleus of
transcription factor DAF-16, the activation of which is required for the very
long life span and marked stress-resistance of insulin-like signaling C.
elegans mutants (Lin et al. 1997). The best-known means of extending life
span, which works in every species tested, including primates, is calorie
restriction (CR). Reducing calorie intake has been also shown to alter
metabolism and delay aging-related disease (Koubova and Guarente 2003).
Which, if any, of the above mechanisms might sir2 affect, and how is
it acting? In S. cerevisiae,sir2 acts to prevent recombination at the rDNA
loci, reducing the number of extrachromosomal DNA circles that cause
190
aging in yeast (Guarente 2000). This mechanism, however, is specific to
yeast cells, implying that in higher eukaryotes sir2 functions in a different
way to affect life span. In flies, sir2 may act as part of a calorie restriction
longevity pathway and respond to the histone deacetylase protein encoded
by Rpd3 (Rogina and Helfand 2004).
There is some controversy about whether C. elegans sir-2.1 (the
closest homologue of yeast sir2) is required for longevity induced by calorie
restriction in worms. A recent report (Wang and Tissenbaum 2006) showed
genetic analysis with a sir-2.1 mutant that suggested that sir-2.1 may be
essential for CR-induced life span extension, while others did not see the
same effect in similar and different experimental settings (N. Bishop, and
M. Viswanathan, unpublished observations). It has been proposed in the
past that sir-2.1 acts through the insulin-like pathway (Tissenbaum and
Guarente 2001; Tissenbaum and Guarente 2002). This hypothesis stemmed
from the fact that the forkhead transcription factor DAF-16, encoded by a
downstream gene in the insulin-like pathway that is required for life span
extension by low insulin-like signaling, is also required for life span
extension by sir-2.1 overexpression in worms (Tissenbaum and Guarente
2001). However, the mechanism of how sir-2.1 extends worm life span was
not known until a short time ago.
14-3-3 proteins: new players in life span determination
Recently, we reported that C. elegans SIR-2.1 forms a complex with
two proteins of a conserved 14-3-3 family, PAR-5 and FTT-2
(Berdichevsky et al. 2006). Studies of mammalian cells suggested that 143-3 proteins, which frequently serve as adaptors modulating activity and/or
localization of their binding partners (Tzivion et al. 2001), interact with the
191
mammalian DAF-16 homologue FOXO, and that this interaction retains
FOXO in the cytoplasm, inactive (Brunet et al. 1999). Is this also true for
worms? Indeed, when we down-regulate 14-3-3 proteins in C. elegans,
DAF-16 becomes constitutively nuclear, implying that worm 14-3-3
proteins can act similarly to mammalian 14-3-3 to sequester DAF-16 in the
cytoplasm (Berdichevsky et al. 2006). One model that could explain the
requirement for 14-3-3 proteins for the long life span granted by SIR-2.1
overexpression is that SIR-2.1 acts by preventing 14-3-3 from sequestering
DAF-16. This appears not to be the case, however, because overexpression
of SIR-2. 1 does not result in relocalization of DAF- 16 into the nucleus.
Also, despite the evident role of 14-3-3 in DAF-16 subcellular localization,
low levels of 14-3-3 proteins are not beneficial for longevity. Instead, low
14-3-3 levels cause shortening of life span in wild-type worms, and actually
eliminate the extension of life span by sir-2.1 overexpression, indicating
that 14-3-3 genes act in the same life span determination pathway as sir-2.1,
downstream of sir-2.1 (Berdichevsky et al. 2006). Therefore, like SIR-2.1
and DAF-16, 14-3-3 proteins act positively on lifespan. More recently,
Wang et al (2006) also reported SIR-2.1/14-3-3 physical interaction, as well
as requirement for 14-3-3 proteins for the life span extension mediated by
SIR-2.1 overexpression. Notably, they demonstrate that overexpression of
14-3-3 proteins causes extension of life span in a daf-16-dependent manner,
much like SIR-2.1 overexpression (Wang et al. 2006). This result further
supports the overall positive role of 14-3-3 proteins in C. elegans longevity.
192
Mechanism of SIR-2.1 effect on lifespan
We found that C. elegans SIR-2.1 protein can interact with DAF- 16.
This interaction is readily detectable when the worms are stressed by heat
shock, but in the absence of heat shock (under normal conditions) a SIR2. 1/DAF- 16 complex was not detected. 14-3-3 proteins, which interact with
both SIR-2.1 and DAF-16 independent of heat shock, play a vital role in the
formation of the SIR-2.1/DAF-16 complex, as reducing 14-3-3 levels
greatly decreases the stress-induced interaction between DAF-16 and SIR2.1. It seems likely, therefore, that SIR-2.1 overexpression results in
increased levels of complex formation between SIR-2.1 and DAF-16,
leading to DAF- 16 activation (Figure 1). This hypothesis is supported by
the fact that expression of a canonical DAF- 16 target gene, sod-3, as well as
other DAF-16 targets, is elevated in worms overexpressing sir-2.1, and 143-3 proteins are again required for this effect (Berdichevsky et al. 2006).
Therefore, in C. elegans, SIR-2.1 and 14-3-3 can act as DAF-16 coactivators, binding to DAF- 16 and triggering transcription of its targets
upon stress. 14-3-3 proteins either physically bridge SIR-2.1 and DAF-16,
or promote their interaction by modifying one or both proteins in response
to stress.
In mammalian cells, a similar complex between SirTI (a homologue
of SIR-2. 1) and FOXO (a homologue of DAF-16) has been reported, and
stress is also important for this interaction (Brunet et al. 2004; Motta et al.
2004). However, it is unknown whether mammalian 14-3-3 is required for
complex formation between SirTi and FOXO. Also, in mammals SirTi can
directly deacetylate FOXO, and this deacetylation can modulate FOXO
activity by repressing some of FOXO's target genes while activating others
(Brunet et al. 2004). So far, there is no evidence for SIR-2.1-mediated
193
repression of DAF- 16 targets in worms, but the physical interaction
between the two proteins is conserved from worms to mammals. It is
tempting to speculate that human SirT1 may also modulate response to
stress and organismal aging through interaction with DAF-16/FOXO.
SIR-2.1 pathway for life span extension.
Thus, sir-2.1, 14-3-3 genes and daf-16 act together to determine
longevity, but do they act within the insulin-like pathway? If that were the
case, one would expect 14-3-3 and sir-2.1 to be required for the life span
extension by reduction in insulin-like signaling. However, the long life span
of mutants defective in the insulin-like receptor DAF-2 does not depend on
sir-2.1 or 14-3-3 (Berdichevsky et al. 2006; Wang et al. 2006; Wang and
Tissenbaum 2006). Thus sir-2.1 and 14-3-3 do not act downstream of the
insulin-like receptor to extend life span. Like stress, low insulin-like
signaling results in accumulation of DAF-16 in the nucleus. However,
unlike stress, low insulin-like pathway activity does not promote the
formation of SIR-2. I1/DAF- 16 complex (Berdichevsky et al. 2006). Taken
together, these results indicate that sir-2.1 and 14-3-3 likely act to promote
longevity in parallel to the insulin-like pathway, in a separate stressdependent pathway that also converges on daf-16. Is stress then good for
the worm? It may be: a mild heat shock early in life can produce life span
extension of up to 30% (Apfeld et al. 2004). It will be interesting to
determine whether this effect requires 14-3-3 and/or SIR-2.1.
Emerging group of DAF-16 co-factors
Several studies suggest that nuclear localization of DAF- 16 is not
sufficient for life span extension and stress resistance: as noted above,
194
worms with low 14-3-3 levels have constitutively nuclear DAF-16 yet they
are short-lived and hypersensitive to stress. This implies that either
phosphorylated DAF-16 that accumulates in the nucleus upon reduction in
14-3-3 levels is inactive, or that another cofactor is required for DAF- 16
activity. Overexpression of a constitutively nuclear form of DAF- 16 lacking
all the Akt phosphorylation sites also does not lead to extended longevity
and stress resistance (Lin et al. 2001), supporting the notion that other
cofactors play a vital role in DAF-16 activation. Binding of such cofactors
to DAF-16 may depend on DAF-16 localization and possibly
posttranscriptional modifications of DAF-16. For 14-3-3/DAF-16
interaction, phosphorylation of DAF-16 at the Akt sites, which requires
activity of the insulin-like signaling cascade, is essential (Figure 1) (Cahill
et al. 2001). Stress may lead to additional modifications of DAF-16 and/or
SIR-2.1, promoting their interaction. Several recent reports bring to light
new factors in the stress-induced relocalization of DAF- 16 into the nucleus.
Jun N-terminal kinase (JNK) can directly phosphorylate DAF-16 and
facilitate its nuclear localization following heat shock (Oh et al. 2005). It
would be interesting to see whether phosphorylation by JNK is required for
SIR-2. 1/DAF-16 interaction and life span extension by SIR-2.1 or 14-3-3
overexpression.
Upon oxidative stress, MST1 kinase (CST-1 in C. elegans)
phosphorylates DAF-16/FOXO and promotes its nuclear translocation
(Lehtinen et al. 2006). MST1 phosphorylation disrupts FOXO's interaction
with 14-3-3 proteins, suggesting that 14-3-3 proteins may not act as DAF16 co-activators following oxidative stress. It is plausible that different
stressors will induce distinct modifications that determine the specificity of
the cofactor (Figure 2). A recent report suggests that bar-], a worm
195
homologue of beta-catenin, may act as a DAF-16 co-activator upon
oxidative stress: like SIR-2.1 and 14-3-3 proteins during heat stress, betacatenin/BAR-1 can bind FOXO/DAF-16 and activate transcription of its
targets when worms are subjected to oxidative stress (Essers et al. 2005).
Under low insulin-like signaling, an unphosphorylated DAF-16
accumulates in the nucleus. In this case, 14-3-3 and SIR-2.1 do not bind
DAF-16, but another cofactor, SMK-1 (a homologue of mammalian
Smekl), does. This nuclear protein is required for life span extension by
low insulin-like signaling (Wolff et al. 2006).
Another pathway for life span determination that converges on DAF16 involves signaling through germline cells and the somatic gonad: worms
lacking the germline live 40% longer than their fertile counterparts, and this
effect depends on DAF- 16 as well as another nuclear co-factor, KRI- 1
(Hsin and Kenyon 1999; Berman and Kenyon 2006).
So it seems that DAF-16 does not act alone, but rather selects its
allies depending on the situation, an interesting fact considering the number
of longevity pathways that converge on DAF-16. So far, the search for
DAF-16 transcriptional targets has been limited to genes with expression
that is altered when DAF-16 is activated by reducing insulin-like signaling.
It is possible that under different conditions, specific cofactor binding
determines which of the variety of DAF- 16 targets will be induced. Further
experiments should clarify this issue.
196
References
Aguilaniu, H., Durieux, J., and Dillin, A. 2005. Metabolism, ubiquinone
synthesis, and longevity. Genes Dev 19(20): 2399-2406.
Apfeld, J., O'Connor, G., McDonagh, T., Distefano, P.S., and Curtis, R.
2004. The AMP-activated protein kinase AAK-2 links energy levels
and insulin-like signals to lifespan in C. elegans. Genes Dev.
Berdichevsky, A., Viswanathan, M., Horvitz, H.R., and Guarente, L. 2006.
C. elegans SIR-2.1 interacts with 14-3-3 proteins to activate DAF-16
and extend life span. Cell 125(6): 1165-1177.
Berman, J.R. and Kenyon, C. 2006. Germ-cell loss extends C. elegans life
span through regulation of DAF-16 by kri-1 and lipophilic-hormone
signaling. Cell 124(5): 1055-1068.
Blander, G. and Guarente, L. 2004. The Sir2 family of protein deacetylases.
Annu Rev Biochem 73: 417-435.
Brunet, A., Bonni, A., Zigmond, M.J., Lin, M.Z., Juo, P., Hu, L.S.,
Anderson, M.J., Arden, K.C., Blenis, J., and Greenberg, M.E. 1999.
Akt promotes cell survival by phosphorylating and inhibiting a
Forkhead transcription factor. Cell 96(6): 857-868.
Brunet, A., Sweeney, L.B., Sturgill, J.F., Chua, K.F., Greer, P.L., Lin, Y.,
Tran, H., Ross, S.E., Mostoslavsky, R., Cohen, H.Y., Hu, L.S.,
Cheng, H.L., Jedrychowski, M.P., Gygi, S.P., Sinclair, D.A., Alt,
F.W., and Greenberg, M.E. 2004. Stress-dependent regulation of
FOXO transcription factors by the SIRT1 deacetylase. Science
303(5666): 2011-2015.
Cahill, C.M., Tzivion, G., Nasrin, N., Ogg, S., Dore, J., Ruvkun, G., and
Alexander-Bridges, M. 2001. Phosphatidylinositol 3-kinase signaling
197
inhibits DAF-16 DNA binding and function via 14-3-3-dependent
and 14-3-3-independent pathways. J Biol Chem 276(16): 1340213410.
Essers, M.A., de Vries-Smits, L.M., Barker, N., Polderman, P.E.,
Burgering, B.M., and Korswagen, H.C. 2005. Functional interaction
between beta-catenin and FOXO in oxidative stress signaling.
Science 308(5725): 1181-1184.
Guarente, L. 2000. Sir2 links chromatin silencing, metabolism, and aging.
Genes Dev 14(9): 1021-1026.
Hsin, H. and Kenyon, C. 1999. Signals from the reproductive system
regulate the lifespan of C. elegans. Nature 399(6734): 362-366.
Kenyon, C. 2005. The plasticity of aging: insights from long-lived mutants.
Cell 120(4): 449-460.
Koubova, J. and Guarente, L. 2003. How does calorie restriction work?
Genes Dev 17(3): 313-321.
Lee, S.S., Lee, R.Y., Fraser, A.G., Kamath, R.S., Ahringer, J., and Ruvkun,
G. 2003. A systematic RNAi screen identifies a critical role for
mitochondria in C. elegans longevity. Nat Genet 33(1): 40-48.
Lehtinen, M.K., Yuan, Z., Boag, P.R., Yang, Y., Villen, J., Becker, E.B.,
DiBacco, S., de la Iglesia, N., Gygi, S., Blackwell, T.K., and Bonni,
A. 2006. A conserved MST-FOXO signaling pathway mediates
oxidative-stress responses and extends life span. Cell 125(5): 9871001.
Lin, K., Dorman, J.B., Rodan, A., and Kenyon, C. 1997. daf-16: An HNF3/forkhead family member that can function to double the life-span
of Caenorhabditis elegans. Science 278(5341): 1319-1322.
198
Lin, K., Hsin, H., Libina, N., and Kenyon, C. 2001. Regulation of the
Caenorhabditis elegans longevity protein DAF-16 by insulin/IGF-1
and germline signaling. Nat Genet 28(2): 139-145.
Motta, M.C., Divecha, N., Lemieux, M., Kamel, C., Chen, D., Gu, W.,
Bultsma, Y., McBurney, M., and Guarente, L. 2004. Mammalian
SIRT1 represses forkhead transcription factors. Cell 116(4): 551-563.
Oh, S.W., Mukhopadhyay, A., Svrzikapa, N., Jiang, F., Davis, R.J., and
Tissenbaum, H.A. 2005. JNK regulates lifespan in Caenorhabditis
elegans by modulating nuclear translocation of forkhead transcription
factor/DAF-16. Proc Natl Acad Sci U S A 102(12): 4494-4499.
Rogina, B. and Helfand, S.L. 2004. Sir2 mediates longevity in the fly
through a pathway related to calorie restriction. Proc Natl Acad Sci U
S A 101(45): 15998-16003.
Tissenbaum, H.A. and Guarente, L. 2001. Increased dosage of a sir-2 gene
extends lifespan in Caenorhabditis elegans. Nature 410(6825): 227230.
-. 2002. Model organisms as a guide to mammalian aging. Dev Cell 2(1): 9-
19.
Tzivion, G., Shen, Y.H., and Zhu, J. 2001. 14-3-3 proteins; bringing new
definitions to scaffolding. Oncogene 20(44): 6331-6338.
Wang, Y., Oh, S.W., Deplancke, B., Luo, J., Walhout, A.J., and
Tissenbaum, H.A. 2006. C. elegans 14-3-3 proteins regulate life span
and interact with SIR-2.1 and DAF- 16/FOXO. Mech Ageing Dev.
Wang, Y. and Tissenbaum, H.A. 2006. Overlapping and distinct functions
for a Caenorhabditis elegans SIR2 and DAF-16/FOXO. Mech
Ageing Dev 127(1): 48-56.
199
Wolff, S., Ma, H., Burch, D., Maciel, G.A., Hunter, T., and Dillin, A. 2006.
SMK- 1, an essential regulator of DAF- 16-mediated longevity. Cell
124(5): 1039-1053.
200
Figures:
Figure 1: SIR-2.1 mechanism of life span extension.
(a) SIR-2.1 binds DAF-16 in the 14-3-3-dependent manner. We propose
that SIR-2.1 can bind and activate DAF-16 in the nucleus. 14-3-3 likely
bridges between SIR-2.1 and DAF- 16, or modifies one or both of them in
response to stress to promote their interaction. (b) In the absence of 14-3-3
proteins, SIR-2.1 cannot bind DAF-16, and DAF-16 remains inactive
despite nuclear localization, resulting in shorter life span and sensitivity to
stress. (c) When SIR-2.1 is overexpressed, more DAF-16/SIR-2.1
complexes are formed, leading to DAF-16 activation and life span
extension.
201
Figure 1.
a. Wild type + stress
SIR-2.
14-3-3
...
DAF-16
active
r
b. 14-3-3A
~IP..9 I
Life span
S•tressRI
c. SIR-2.1 ole
SIR-2.1
Life spar
Stress R
202
4"14-3-3
DAF-16
active
· ·---
--·- ---
Figure 2: DAF-16 can be activated by different cofactors.
Under normal conditions, the majority of DAF-16 molecules are present in
the cytoplasm, inactive, bound and sequestered by 14-3-3 proteins. When
insulin-like signaling is low, unphosphorylated DAF-16 is unable to bind
14-3-3 and accumulates in the nucleus, where it can interact with, and get
activated by SMK-1. In worms that lack a germline, DAF-16 also
accumulates in the nucleus, and is activated by KRI-1 protein. It is
unknown whether such nuclear DAF-16 is still phosphorylated at the Akt
sites and bound by 14-3-3. It is also not known what other modifications
may be involved in DAF-16 activation and/or nuclear localization in
germlineless worms. Stress also results in accumulation of DAF-16 in the
nucleus. Heat stress-induced relocalization of DAF-16 likely involves
phosphorylation by JNK. Since the insulin-like pathway is still active in this
case, DAF-16 is phosphorylated at the Akt sites and can interact with 14-33 and SIR-2. 1, which activate DAF-16 in this case. Finally, activation of
DAF-16 by oxidative stress may involve phosphorylation by MSTl/CST-1
kinase and interaction with BAR- 1. In every case, lack of a cofactor
abolishes DAF-16-dependent life span extension and resistance to stress.
203
~-C4DAF-I6'j
SITR-2.·'3?.,·
DAF-16
IBAR-1t
DA~LF-16e
active
r..
Stress
I
Not
DAF-16
active
SMK-I
1443-3
DAF-16
active
nucleus
cytoplasm
Figure 2
204