| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
REVIEW ARTICLE |
1 Department of Biology, Ghent University, Belgium.
2 Institute for Behavioral Genetics, Department of Integrative Physiology, University of Colorado at Boulder.
Address correspondence to Jacques R. Vanfleteren, Department of Biology, Ghent University, K.L. Ledeganckstraat 35, B-9000 Ghent, Belgium. E-mail: Jacques.Vanfleteren{at}UGent.be
|
A |
|---|
|
|
|---|
Despite a long period of research on CR, much remains unknown about the mechanisms underlying its life extension. The major theories that were proposed to explain the healthy effects of CR include (a) theories that predict reduced ROS (reactive oxygen species) production, (b) theories that predict increased ROS defense, and (c) theories predicting altered signaling. These theories are not mutually exclusive. The disposable soma theory explains the life extension of calorically restricted animals from an evolutionary point of view: It states that restricted animals shift energy use from reproductive effort to somatic maintenance, resulting in slower aging and allowing them to reproduce when conditions improve, even at ages at which they would normally have been postreproductive (1012).
The nematode Caenorhabditis elegans has been widely used for studying development, molecular biology, genetics, and aging. However, surprisingly few experiments on CR have been carried out in C. elegans, perhaps due to the absence of a definitive method for subjecting C. elegans to CR. Several approaches have, however, been taken in an attempt to study the physiology, metabolism, and molecular biology of restricted worms. Because most treatments aimed at reducing calories change nutrient composition as well, the term dietary restriction (DR) instead of CR is used throughout the text when referring to C. elegans.
| C. ELEGANS NUTRIENT UPTAKE |
|---|
|
|
|---|
|
In the laboratory, C. elegans is generally cultured on Escherichia coli bacteria that are seeded on agar plates. Generally, the slow growing OP50 strain, which is auxotrophic for uracil, is used because the thin bacterial lawn makes microscopic study of C. elegans easier. A thin live E. coli lawn is thus considered as the normal, nonrestricted diet of the worm. C. elegans can also be grown in liquid suspension cultures with E. coli as the food source, but vigorous shaking is needed if the depth of the medium exceeds a few millimeters, to prevent hypoxic stress to the worms (15).
| SUBJECTING C. ELEGANS TO DR |
|---|
|
|
|---|
Klass (20) decreased the bacterial concentration in suspension culture to impose DR to C. elegans. He found a mean life-span extension of 60% when the bacterial density was decreased from 109 to 108 bacterial cells per milliliter (higher concentrations led to decreased life span and decreased reproductive capacity, likely caused by hypoxic stress). Under these conditions, progeny production was decreased more than fourfold. Hosono and colleagues (21) reduced bacterial concentrations on agar plates by decreasing the amounts of bactopeptone. They also observed life extension but no reduction of reproductive capacity or body volume. The advantage of restricting worms by reducing bacterial concentration is that this treatment can be applied in a quantitative manner. Reduced bacterial uptake can also be obtained genetically by using mutants with a reduced pumping rate. Such Eat mutants have a starved appearance, and they were used by Lakowski and Hekimi (22) to study the genetics of DR. [Note that in C. elegans nomenclature, Eat refers to the phenotype, eat-1(ad427) to the gene (allele) and EAT-1 to the protein encoded by eat-1.] These authors found that most Eat mutants were indeed long-lived, with a maximal life extension of about 50%. However, smaller or no effects on life span were found in some other laboratories (M. Keaney and D. Gems, personal communication, 2002). A likely explanation is that these mutants experience DR depending on the environmental conditions, such as the thickness of the bacterial lawn. For example, we found life extension when the Eat mutants were grown in liquid culture, but not on plates (K. Houthoofd, unpublished data, 2002). Perhaps reduced bacterial intake of Eat mutants is not limiting when they are grown on plates with plenty of E. coli.
C. elegans can also be grown in axenic media (axenic: grown in the absence of any other contaminating species). One example of such a medium is CbMM (Caenorhabditis briggsae maintenance medium), a defined medium containing 54 compounds (23). A more frequently used, semidefined, axenic medium is composed of yeast extract and soy peptone (24,25). A sterol and heme source must be added to axenic media because C. elegans is not able to synthesize the porphyrin skeleton of heme. The heme requirement was originally met by adding tissue extracts (e.g., liver extract or chicken embryo extract). Later it was found that pure hemoglobin is a suitable supplement (26). Sterols are supplied sufficiently as impurities in yeast extract, soy peptone, and the heme source. When grown in axenic media, the life span of worms is about twice as long relative to populations maintained with E. coli (27). Axenically cultured worms have a much slower development and a severely affected brood size compared to those grown in monoxenic culture conditions, consistent with nutrient deprivation seen in other DR regimens. A DR effect of axenic culture is also suggested by the observation that worms grown in axenic medium show metabolic alterations similar to those seen in worms that are restricted by eat-mutation or by lowering bacterial food supply (25,28). Because axenic media are sterile, life extension is partially caused by the absence of pathogenic bacteria. Axenic medium is a rich medium, and it therefore seems contradictory that worms that are grown axenically are subject to DR. Possible explanations are that axenic medium cannot be taken up by the worms efficiently, either because worms are filter feeders, spitting out most of the liquid medium, that compounds are not taken up efficiently by the intestine, or that one or more critical compounds are underrepresented or missing from axenic medium. Note, however, that there is most probably no malnutrition, because many essential growth compounds are supplied in large quantities. Unfortunately, axenic medium cannot be used if one wants to partially reduce energy intake.
Finally, one can also knock down the expression of transporters that are required for nutrient uptake. For example, RNAi (RNA-mediated interference) against nac-2 (transporter of di- and tricarboxylates) or nac-3 (transporter of dicarboxylates) leads to a life extension of 19% and 15%, respectively (29,30). PEP-2 is a proton-dependent carrier responsible for the uptake of di- and tripeptides. Mutation in pep-2 leads to a smaller body size and reduced developmental rate and fertility but, surprisingly, does not increase life span (31). The NHX-2 Na+/H+ exchanger is needed to prevent acidification of the cytoplasm (32). RNAi against nhx-2 leads to a number of phenotypes, including a 40% increased life span, probably because uptake of di- and tripeptides is inhibited when the cytoplasmic pH drops.
| POSSIBLE MECHANISMS OF DR-MEDIATED LIFE EXTENSION |
|---|
|
|
|---|
Reduced Metabolic Rate?
In an attempt to study the interaction between DR and long-lived mutants, Lakowski and Hekimi (22) constructed double mutants of eat-2 and clk-1, a mutant with a slowed behavior and long life span (39,40). They found that mutation in the clk-1 gene could not further extend the life span of eat-2 and they concluded that both mutations lengthen the C. elegans life span via the same mechanism. Because it was believed that clk-mutants were long-lived due to a reduction of metabolic rate, it was concluded that DR postpones aging by lowering metabolic rate. However, later studies showed that clk-mutants had no reduced respiration or heat production rate (41,42), undermining the proposed hypothesis.
Houthoofd and colleagues (25,28) measured respiration and heat-production rate directly in worms that were restricted by E. coli dilution, eat-mutation, or growth in axenic medium. Surprisingly, they found that DR leads to an increase in metabolic rate, at least when expressed per unit of body mass. They next determined the ATP (adenosine triphosphate) content and the reductive capacity in these worms, and found that DR causes lower ATP concentrations and higher reductive capacity. These results can be explained assuming that the high metabolic rate is needed for the synthesis of molecules that are freely available in the diet of ad libitum-fed worms, but absent or at a lower concentration in restricted worms. These anabolic reactions would require ATP and reductive reactions. Another possible energy-demanding process is increased protein turnover, as observed in mammals and yeast subjected to DR (43,44). Increased respiration rate as a response to DR is in accordance with experiments in yeast. Lin and colleagues (45) have shown that DR causes a shift from fermentative towards respirative metabolism and that life extension caused by DR depends on activity of the tricarboxylic acid cycle.
It should be stressed that a higher aerobic metabolism is not necessarily linked to increased free radical generation, as ROS production is dependent on the inner mitochondrial membrane potential. On the contrary, membrane potential and ROS production are inversely related with respiratory activityhigh in resting mitochondria and low in actively respiring mitochondria (4648). Uncoupling proteins can also lower the membrane potential, again leading to a lowering of free radical generation. For instance, Speakman and colleagues (49) found that individuals with the highest metabolism, in a mouse population, had the highest mitochondrial uncoupling rate and the longest life span.
The ROS production rate in response to the nutritional regime has not been tested yet in C. elegans. Mitochondria from CR rodents had a lower membrane potential and produced less ROS (reviewed in 35,50).
Increased Stress Resistance?
Houthoofd and colleagues (25,27) also determined the resistance to oxidative and heat stress in worms that were grown under axenic conditions. They found that restricted worms had a superior resistance to both stresses. Moreover, this resistance was accompanied by higher activities of superoxide dismutase (SOD) and catalase, two enzymes that are involved in the breakdown of ROS. pep-2 mutants have a higher heat tolerance and an increased resistance to oxidative stress (31), but are not long lived, as mentioned previously. An upregulation of stress resistance in response to DR could be the consequence of a hormetic response (51,52). In this view, DR is seen as a low intensity stressor, and animals subjected to DR react to this stressor by upregulating the stress defense system which also protects them against aging.
Reduced Ins/IGF-1 Signaling?
If increased stress defense is necessary for DR-induced life extension, the DR response is likely mediated by a signaling pathway that regulates the expression of a life-extending program in response to the nutritional status of the organism. Good candidates are the JNK (c-Jun amino-terminal kinase) (53) and the Ins/IGF-1-like signaling pathways. Both converge on the transcription factor DAF-16. The Ins/IGF-1-like signaling pathway is an evolutionary conserved pathway that shows homology with the insulin (Ins) and insulin-like-growth factor-1 (IGF-1) mammalian pathways, and which regulates the aging rate in C. elegans, Drosophila, and mice (54). Inactivation of this pathway by mutation in the Ins/IGF-1 receptor daf-2 or in one of the downstream genes (e.g., age-1) in the worm results in a substantial life-span extension that is dependent on the transcription factor DAF-16. DAF-16 accumulates in the nucleus of worms with reduced signaling activity, resulting in the increased expression of many genes that confer resistance to stress and enhance mean and maximum life span (5558). Several observations suggested a role for the Ins/IGF-1 pathway in life-span extension caused by DR: (a) The Ins/IGF-1 pathway is also important for the formation of dauers (a long-lived and stress-resistant larval stage) in C. elegans, and food availability is one of the regulating factors for dauer formation. (b) Mutants with defective olfactory perception are long lived, and this longevity phenotype is dependent on the DAF-16 transcription factor, suggesting the involvement of the Ins/IGF-1 pathway in this process (59). Moreover, insulin-like peptides are mainly expressed in neuronal cells in the head region of the worm (60). (c) Ins/IGF-1 activity in intestinal tissue, which is responsible for the uptake of nutrients and for the production of vitellogenins, is necessary and sufficient to regulate the worm's life span (61). (d) The germline also regulates Ins/IGF-1 signaling: Worms lacking germline proliferation are long-lived, and this phenotype is also dependent on DAF-16 (62). Because DR causes lower brood size, it is likely that germline proliferation is reduced. (e) In vertebrates, insulin production is dependent on the glucose concentration in the blood stream. (f) Mutants with reduced Ins/IGF-1 signaling activity share several phenotypes, such as smaller body size and increased stress resistance, with restricted individuals. Thus, it seems plausible that food regulates the expression of certain insulin-like peptides that downregulate the activity of the Ins/IGF-1 pathway to shift energy usage from reproduction to somatic maintenance, extending the life span.
C. elegans is a useful model organism with which to test the role of Ins/IGF-1 signaling in the DR response. Many longevity mutants with impaired activity of the Ins/IGF-1 pathway are available and can be used for testing if Ins/IGF-1 signaling and DR act via common or different mechanisms. Several authors have used this method to investigate the role of Ins/IGF-1 signaling in DR responses. Johnson and colleagues (63), for example, cultured the longevity mutant age-1 in liquid culture with different E. coli concentrations and found that the life span of age-1 could also be extended by DR, indicating that the Ins/IGF-1 pathway is not needed for mediating DR life span effects. Lakowski and Hekimi (22) made double mutants of eat-2 and daf-2, and found additive life-extending effects, again an indication of different mechanisms of life-span extension. Consistent with these results, Houthoofd and colleagues (27,64,65) cultured daf-2 mutants in axenic medium and found that daf-2 mutants live substantially longer in axenic medium. Axenic culture substantially enhanced the upregulation of the stress defense program and altered metabolism of worms in which Ins/IGF-1 signaling was reduced (27,66). Meissner and colleagues (31) found that the life span of daf-2 mutants can be further extended by an additive mutation in pep-2, and that additive effects were also seen on heat tolerance of the double mutant. One difficulty with this approach, however, is that the long-lived Ins/IGF-1 mutants are "reduction of function" mutants and thus still have residual activity. One could therefore argue that a further life extension of Ins/IGF-1 mutants is the result of a further reduction of Ins/IGF-1 activity (67).
More definitive conclusions have been obtained by studying daf-16 mutants. A variety of mutant alleles are available for this gene, including null and almost null mutants, and these mutations completely suppress the long life span of the long-lived Ins/IGF-1 mutants. If the life extension caused by DR is mediated by Ins/IGF-1 signaling, then mutation in daf-16 should suppress the long life span of DR worms. DR-mediated life-span extension was not suppressed by daf-16 mutations when DR was imposed by using eat-2 mutants (22). The elevated stress defense of pep-2 mutants was also unaffected by mutation in the daf-16 gene (31). Finally, daf-16 failed to suppress the life extension, metabolism, and stress resistance of worms that were cultured in axenic medium (27,66). These results are consistent with the cytosolic localization of DAF-16 in eat-2 mutants and in wild-type worms grown in axenic medium (27,56). The predicted role of Ins/IGF-1 signaling was therefore contradicted by experimental testing (Figure 1). The dependence of JNK signaling on DAF-16 (53) similarly argues against its potential role in mediating the effect of DR on life span.
These findings are at odds with the observation that flies carrying a mutation in the IRS (insulin receptor substrate) gene chico respond to DR less efficiently than wild-type flies. Thus, in flies DR may act by downregulating Ins/IGF-1 activity (68). In rodents, CR leads to lower insulin, IGF-1, and growth hormone concentrations in the blood stream (6). CR feeding also results in a decreased proton motive force and ROS production, and these effects are reversed by subjecting CR animals to a short period of insulin treatment (50). However, Bartke and colleagues (69) reported that CR further extends the life span of Ames dwarf mice, which produce less growth hormone and IGF-1 (among other things). These authors therefore concluded that mutants with reduced activity of the growth hormoneIGF-1 axis slow down the aging process by a mechanism that is different from CR. However, this conclusion was criticized by Clancy and colleagues (68), who argue that Ames dwarf mice have residual IGF-1 activity that could be further decreased by DR, resulting in a further life extension.
Altered DNA Silencing?
The life span of C. elegans can also be extended by extra copies of sir-2.1 (70) or by resveratrol, a stimulator of SIR proteins (71). sir-2.1 is a homologue of the yeast SIR2 gene that encodes a histone deacetylase, which increases the life span of yeast mother cells (72,73). It is believed that CR increases that life span of yeast mother cells by regulating Sir2 activity, because the life span of Sir2 mutants does not respond to variations in caloric uptake (74). In humans, sirt1 is also dependent on Foxo3a, a daf-16 homologue (75). However, longevity induced by increased SIR-2.1 activity in C. elegans is dependent on DAF-16 (70), and because DR promotes longevity independently from DAF-16 in the worm, SIR activity seems unlikely to mediate the DR response in C. elegans.
Reduced TOR Signaling?
Another likely candidate for mediating a DR response is the TOR (target of rapamycin) pathway. In mammals and fruit flies, TOR senses the cellular amino acid pool and regulates cell growth by controlling transcription, translation, and protein degradation through the activation of S6K and eIF-4E (76). A mutation in let-363, the C. elegans TOR homologue, or in daf-15, the C. elegans homologue of RAPTOR (regulatory associated protein of TOR) causes arrest and death as dauer-like larvae (77). Heterozygous daf-15 mutants are long-lived (77). Knocking down let-363 by RNAi from the first day of adulthood also leads to increased life span, and this effect does not require DAF-16 (78). Evidence for the role of TOR signaling in the DR response comes from Meissner and colleagues (31). These authors found that RNAi against let-363 did not extend the life span of pep-2 mutants. This is consistent with pep-2 acting upstream of TOR via PI3K (phosphoinositide-3-OH kinase) and AKT signaling.
In mammals, TOR activity is stimulated by insulin (and other growth factors) via PI3K and Akt (retroviral oncogen v-akt related protein; also called PKB or RAC) (79), whereas in C. elegans, DAF-16 controls the expression of daf-15 (77) and pep-2 (31,58). Thus, Ins/IGF-1 and TOR signaling might cooperate in a complex metabolic control circuit that optimizes metabolism and life span as a function of nutrient availability.
Interestingly, AMPK (AMP-activated protein kinase) activation leads to a decrease in mammalian TOR activity as measured by S6K phosphorylation (80). AMPK proteins are potential candidates for the regulation of life span under DR conditions. The C. elegans genome contains two aak-genes. Overexpression of aak-2 extends life span after being activated by a high AMP/ATP ratio. Because reduced caloric intake is likely to increase this ratio, it seems plausible that aak-2 mediates DR-induced life extension (81). This hypothesis has not been tested directly, however. AAK-2 functions independently from DAF-16 to regulate life span. However, aak-2 acts downstream of daf-2. Thus, it seems highly plausible that TOR regulates the aging rate via stimulation by nutrients and by sensing the AMP/ATP ratio in the cell (Figure 1).
Conclusions
Several methods have been used to study the effects of DR on the life span, stress resistance, metabolism, and activity of signaling pathways in C. elegans. The increased life span that is seen in restricted worms is accompanied by increased resistance to environmental stressors and elevated activity of stress-defense enzymes. Several signaling pathways have been proposed to mediate this kind of retardation of the aging process. One has mostly focused on Ins/IGF-1 signaling as a potential regulator of DR-induced life extension, but experimental verification has refuted this hypothesis. TOR signaling, however, is a more likely candidate, but this research area has not been explored completely yet. Because TOR signaling is predicted to affect various metabolic processes, it would be interesting to test life span, metabolism, protein turnover, and stress defense of worms with reduced TOR activity that are grown under replete and DR conditions.
|
F |
|---|
|
|
|---|
Received March 18, 2005
Accepted April 18, 2005
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
W.-L. Yen and D. J. Klionsky How to Live Long and Prosper: Autophagy, Mitochondria, and Aging Physiology, October 1, 2008; 23(5): 248 - 262. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. J. Szewczyk, I. A. Udranszky, E. Kozak, J. Sunga, S. K. Kim, L. A. Jacobson, and C. A. Conley Delayed development and lifespan extension as features of metabolic lifestyle alteration in C. elegans under dietary restriction J. Exp. Biol., October 15, 2006; 209(20): 4129 - 4139. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|