An intervention resembling caloric restriction prolongs life span and retards aging in yeast.

Abstract

The yeast Saccharomyces cerevisiae has a finite life span that is measured by the number of daughter cells an individual produces. The 20 genes known to determine yeast life span appear to function in more than one pathway, implicating a variety of physiological processes in yeast longevity. Less attention has been focused on environmental effects on yeast aging. We have examined the role that nutritional status plays in determining yeast life span. Reduction of the glucose concentration in the medium led to an increase in life span and to a delay in appearance of an aging phenotype. The increase in life span was the more extensive the lower the glucose levels. Life extension was also elicited by decreasing the amino acids content of the medium. This suggests that it is the decline in calories and not a particular nutrient that is responsible, in striking similarity to the effect on aging of caloric restriction in mammals. The caloric restriction effect did not require the induction of the retrograde response pathway, which signals the functional status of the mitochondrion and determines longevity. Furthermore, deletion of RTG3, a downstream mediator in this pathway, and caloric restriction had an additive effect, resulting in the largest increase (123%) in longevity described thus far in yeast. Thus, retrograde response and caloric restriction operate along distinct pathways in determining yeast longevity. These pathways may be exclusive, at least in part. This provides evidence for multiple mechanisms of metabolic control in yeast aging. Inasmuch as caloric restriction lowers blood glucose levels, this study raises the possibility that reduced glucose alters aging at the cellular level in mammals.

Full text

Page 1

An intervention resembling caloric restriction
prolongs life span and retards aging in yeast1
JAMES C. JIANG, EWA JARUGA, MARINA V. REPNEVSKAYA, AND S. MICHAL JAZWINSKI2
Department of Biochemistry and Molecular Biology, Louisiana State University Health Sciences
Center, New Orleans, Louisiana 70112, USA
SPECIFIC AIMS
We addressed the hypothesis that caloric restriction
acts at the cellular level to extend longevity and
postpone senescence in eukaryotes, and provides
one of multiple mechanisms of metabolic control in
aging. The effect of progressive reduction in the
glucose or amino acids concentration of the growth
medium on the life span and aging of individual
Saccharomyces cerevisiae cells has been examined, and
the interaction of this caloric restriction effect with
the retrograde response pathway, which signals the
functional status of the mitochondrion and deter-
mines longevity, has been investigated.
PRINCIPAL FINDINGS
1. Lowering the glucose concentration increases
mean and maximum life span of yeast cells in both
broth and synthetic medium
Reduction of the glucose concentration in a modi-
fied broth routinely used for culturing yeast resulted
in an increase in the life span of individual cells,
measured by the number of daughters they pro-
duced, which was the more extensive the greater the
reduction in nutrient levels. Corresponding in-
creases in the mean and maximum life span were
observed—up to 75% in some experiments. Similar
results have been obtained with three different yeast
strains. The effect on life span of reducing glucose
levels was duplicated in a standard, chemically de-
fined growth medium. Extension of life span by as
much as 81% was found. However, the increased
longevity was abrogated when glucose was reduced
beyond a certain concentration, indicating that be-
low this point glucose becomes limiting and malnu-
trition sets in.
2. Reduction of glucose levels postpones the
development of a senescent phenotype during the
yeast life span
One of the manifestations of aging in yeast is an
increase in generation time, which is the interval
between consecutive buddings of individual cells.
The appearance of this age-related phenotype was
delayed when life span was extended by lowering
glucose levels in the medium, as measured by the
rate of bud production. After a brief lag, there
appeared to be a small increase in this rate at the
lower glucose concentrations.
3. Extension of life span by adjustment of glucose
levels does not depend on the retrograde response
Metabolic control plays a role in determining yeast life
span, as evidenced by the increased longevity afforded
by the induction of the retrograde response, a form of
interorganelle communication that signals the func-
tional status of the mitochondrion to the nucleus
resulting in changes in the expression of nuclear genes
that encode a variety of mitochondrial, cytoplasmic,
and peroxisomal proteins. To ascertain whether the
effect on life span of lowering glucose levels is medi-
ated by the retrograde response, the expression of the
CIT2 gene, a diagnostic of the retrograde response, was
assessed. Not only were CIT2 transcript levels not
increased under conditions that extend life span, they
were reduced by 49%.
The RTG genes are required for the retrograde
response. Rtg2p promotes the formation of an active
heterodimeric Rtg1p-Rtg3p transcription factor by
transducing mitochondrial signals that affect the
phosphorylation state and subcellular localization of
Rtg3p. Deletion of the RTG2 gene did not suppress
the life span extension caused by reduction of glu-
cose levels (Fig. 1A), suggesting the lack of a require-
ment for the retrograde response for life span exten-
sion. In fact, the small decrease in life span at high
glucose concentrations that has been often observed
1To read the full text of this article, go to http://www.
fasebj.org/cgi/doi/10.1096/fj.00–0242fje To cite this arti-
cle, use (September 8, 2000) An intervention resembling
caloric restriction prolongs life span and retards aging in
yeast. FASEB J. 10.1096/fj.00-0242fje
2Correspondence: Department of Biochemistry and Molec-
ular Biology, Louisiana State University Health Sciences Cen-
ter, 1901 Perdido St., Box P7–2, New Orleans, LA 70112, USA.
E-mail: sjazwi@lsuhsc.edu
2135 0892-6638/00/0014-2135/$02.25 © FASEB

Page 2

on various growth media on deletion of this gene was
eliminated at low glucose levels (Fig. 2A).
It is possible that the prolonged life span resulting
from reduced glucose levels by-passed the requirement
for Rtg2p but was still dependent on the Rtg1p-Rtg3p
transcription factor. To examine this possibility, we
deleted RTG3. This deletion did not prevent the exten-
sion of life span upon reduction of glucose concentra-
tion in the growth medium (Fig. 2B). There was a 55%
increase in mean life span of the rtg3? strain even
without a reduction in glucose levels. Thus, the expres-
sion of genes under the control of the Rtg1p-Rtg3p
transcription factor suppresses longevity to an extent
under these growth conditions, in contrast to the
requirement of the Rtg pathway for optimal longevity
when the retrograde response signal is present in yeast
whose mitochondria are not fully functional. The in-
crease in mean life span seen upon reduction of
glucose levels (80%) was additive (123%) to that ob-
tainedupondeletionofRTG3(Fig.2B),generatingthe
largest increase in longevity described thus far in yeast.
These results have been observed in two different yeast
strains, where the details of the induction of the
retrograde response differ but its effect on life span is
the same.
4. Decreasing the amino acids concentration of the
growth medium promotes an increase in the mean
and maximum life span of yeast
The extension of life span detected on reduction of
glucose concentration in broth was effective only
when the levels of other nutrients were decreased,
suggesting that manipulation of these nutrients, es-
pecially amino acids, might result in life span exten-
sion. Indeed, decreasing amino acids concentration
while maintaining glucose levels resulted in an in-
creased longevity (up to 95%) that was larger the
greater the reduction in the nonessential amino
acids. This life span extension does not operate
through the retrograde response pathway, because it
did not entail the induction of CIT2 and it was not
suppressed by deletion of RTG2. Thus, it is not the
decreased availability of a specific nutrient but rather
the restriction of the caloric content of the growth
medium that plays a role in life span extension.
CONCLUSIONS
This study demonstrates that yeast life span can be
modulated physiologically. The life span extension
observed upon manipulation of nutritional status
possesses many of the hallmarks of caloric restriction
in mammals. First, this effect is continuous, operat-
ing like a rheostat—the greater the reduction in
nutrient concentrations, the larger the life extension
(up to a point). This evokes a mechanism involving
changes in flux through a pathway rather than a
threshold effect that triggers an all-or-none re-
sponse. Second, the increased longevity is associated
with a postponement in the appearance of an aging
phenotype. Third, the effect is not due to limitation
of a specific nutrient. On the basis of these similar
features, we tentatively call this mechanism of life
span extension in yeast caloric restriction.
The increase in life span upon reduction of glu-
cose levels does not appear to be simply due to
release of the cells from glucose repression, because
the life span extension seen is continuous as glucose
concentration is lowered progressively. Further-
more, the glucose concentrations at which the max-
Figure 1. The retrograde response is not required for exten-
sion of life span by the reduction of glucose levels. A)
Decreased glucose concentration caused an increase in mean
and maximum life span (P? ?0.001) in chemically defined
medium. There was no significant effect on life span on 0.1%
glucose of deletion of RTG2 (P?0.74). Similarly, no effect on
maximum life span was observed (P?0.05). B) Deletion of
RTG3 increased mean life span 55% on 2% glucose in
chemically defined medium. The life extension was signifi-
cant (P?0.00001). As before, growth on 0.1% glucose in-
creased mean life span. The deletion of RTG3 resulted in an
additive increase in mean life span above the control. The
additional enhancement of
(P?0.01). The corresponding differences in maximum life
span were significant (P? ?0.001).
longevitywas significant
2136Vol. 14November 2000 JIANG ET AL.The FASEB Journal

Page 3

imal effect on longevity was observed were quite
different in broth and in chemically defined me-
dium. Finally, life span extension by limitation of
amino acids is clearly not the result of release from
glucose repression.
A reduction in blood glucose levels is an inevita-
ble, early response to caloric restriction in mammals.
The fact that lowering the concentration of glucose
available to individual yeast cells has such a marked
effect on their longevity and aging suggests that a key
component of the anti-aging effect of caloric restric-
tion in mammals may be changes elicited by glucose
at the cellular level. The reduction of glucose levels
delayed the decline in budding rate normally seen
during aging. The calorie-restricted yeast displayed a
higher level of metabolic activity than the controls,
all other things being equal, as indicated by an
increased budding rate. This supports the notion
that a change in the characteristics of fuel use
underlies the caloric restriction effect.
Caloric restriction did not induce the retrograde
response; in fact, a substantial reduction in the
expression of the diagnostic gene for this response,
CIT2, was found. This suggests that caloric restriction
operates by down-regulating at least part of the
retrograde response. Significantly, the extension of
life span by caloric restriction did not require RTG2.
Caloric restriction removed any need of this gene for
a normal life span, suggestive of a positive relation-
ship with certain features of the retrograde response.
Deletion of RTG3 also did not suppress the caloric
restriction effect; instead, it potentiated a maximal
increase in longevity that was additive to caloric
restriction. Thus, the retrograde response and ca-
loric restriction represent two distinct metabolic
mechanisms of life extension (Fig. 2). The results
suggest some separation of functions of the RTG2
and RTG3 mediators of the retrograde response.
They are consistent with the hypothesis that caloric
restriction diminishes or prevents certain aspects of
the retrograde response and that certain aspects of
the retrograde response have a negative effect on the
increased longevity provided by caloric restriction, in
effect preventing the full benefits of caloric restric-
tion (Fig. 2). This does not rule out some overlap
between the longevity effectors induced by the ret-
rograde response and by caloric restriction.
There are similarities between the metabolic ef-
fects of the retrograde response in yeast and of the
daf mutants of Caenorhabditis elegans that display
extended adult longevities. These metabolic effects
are reminiscent of some of the features of the
respective dispersal forms of yeast and worms, the
spore and the dauer larva, which represent evolu-
tionarily conserved responses to environmentally un-
favorable conditions. The retrograde response, how-
ever, appears to be a compensatory adaptation to the
deleterious effects of the internal stress of mitochon-
drial dysfunction that can accumulate with age (Fig.
2). In contrast, caloric restriction may constitute an
adjustment that prevents the deficits associated with
aging. Although clues of its operation exist, evidence
for the presence of the retrograde response in
organisms other than yeast is currently lacking. A
fundamental question concerning caloric restriction
is whether the reduction in calories triggers re-
sponses that enhance survival, or whether it is the
elimination of excess nutrients that have a negative
effect on life span that is responsible.
The present study makes apparent the utility of yeast
to elucidate the molecular mechanisms underlying
caloric restriction, and it may bridge the gap between
studiesofagingacrosswidephylogeneticdivisions.Our
results provide evidence for multiple mechanisms of
metabolic control in aging. Inasmuch as caloric restric-
tion lowers blood glucose levels, this study raises the
possibility that reduced glucose alters aging at the
cellular level in mammals.
Figure 2. Multiple metabolic mechanisms of aging. The
retrograde response and caloric restriction are two nonover-
lapping pathways, which constitute a compensatory adapta-
tion and a preventive adjustment to the dysfunction and
deficits of aging, respectively. These pathways interact at the
level of the longevity effectors. The interaction can be either
positive or negative (see text). The extent of the overlap
between the effectors of both pathways is not known at
present. Rtg2 and Rtg3 are mediators of the retrograde
response, and ‘Cal’ is a hypothetical mediator of the caloric
restriction response.
2137 CALORIC RESTRICTION AND YEAST AGING