The most well-validated antiaging interventions involve two major nutrient-sensing pathways: IIS and TOR. Although strong alterations of either of these signaling pathways can cause adverse effects, including embryonic lethality, cancer, and diabetes, milder downregulation can be beneficial for health and longevity in all organisms so far tested (Piper et al., 2008
). An important consequence of using IIS or TOR manipulations as antiaging interventions is that they can also delay the progression of numerous diseases associated with old age, for example, cancer, neurodegeneration, cardiovascular disease, and diabetes (Bishop and Guarente, 2007; Piper et al., 2008
). Hence, clarifying the role of the TOR pathway in aging, as well as its relationship with IIS, could provide valuable insights for developing treatments for age-related diseases.
The aim of this study was to determine if a pharmacological intervention to reduce TOR signaling had antiaging effects in Drosophila
and to identify the underlying mechanisms. Rapamycin was chosen because it is the most specific and best-studied TOR kinase inhibitor available (Hartford and Ratain, 2007; Wullschleger et al., 2006
). Importantly, rapamycin analogs have already been approved for human use as immunosuppressant drugs and are currently under clinical trials for use as anticancer agents (Guertin and Sabatini, 2009
We administered rapamycin to adult Drosophila
and demonstrated a dose-dependent downregulation of TOR activity after just 1 day of treatment by measuring the phosphorylation status of the TORC1 target, S6K. TOR inhibition by rapamycin was ubiquitous, since the levels of phosphorylated S6K were similarly downregulated in all main segments of the fly body. Notably, we found that continuous rapamycin treatment from early adulthood resulted in a robust and reproducible extension of life span, independent of the genetic and cytoplasmic background and sex of the flies. Rapamycin treatment in yeast has been previously shown to extend replicative (Medvedik et al., 2007
) and chronological (Powers et al., 2006
) life span, and a recent study has shown that rapamycin also extends life span in mice when administered late in life (Harrison et al., 2009
). These data, together with our observation that rapamycin treatment also extends life span in Drosophila
, suggest that the antiaging effects of rapamycin are evolutionarily conserved.
Rapamycin treatment that increased life span also reduced female fecundity. However, rapamycin also extended life span of sterile ovoD
females, strongly suggesting that reduced fertility alone is not sufficient for rapamycin-mediated life span extension. In C. elegans
also, downregulation of TOR, ribosomal proteins, S6K, or elF genes decreases fecundity, and reduced TOR signaling can extend life span in sterile mutants (Hansen et al., 2007
). In addition, a growing body of evidence suggests that reduced fecundity and longevity can be uncoupled (Grandison et al., 2009; Partridge et al., 2005
TOR kinase is a central component of two protein complexes: TORC1 and TORC2. Rapamycin is generally considered to be a specific inhibitor of TORC1, although in some cell lines prolonged rapamycin treatment can also inhibit TORC2 activity (Sarbassov et al., 2006
). Rapamycin treatment significantly reduced S6K phosphorylation, indicative of reduced TORC1 activity. However, we saw no change in phosphorylation of Akt at a TORC2-specific phosphorylation site, suggesting that TORC2 activity remained unchanged. Both TORC1 and TORC2 interact with components of the IIS pathway: the TORC2 complex phosphorylates and activates Akt kinase; conversely, Akt phosphorylates and functionally inactivates the TOR pathway suppressor protein TSC2. In addition, S6K inhibits IRS at the level of translation, transcription, and phosphorylation, thereby exerting a negative feedback loop on IIS (Um et al., 2006; Wullschleger et al., 2006
). Reduced IIS could therefore contribute to extension of life span by rapamycin. However, we did not detect any measurable change in GSK3α/β phosphorylation, an important downstream target of the IIS pathway, suggesting that rapamycin treatment did not alter IIS output.
We performed epistasis experiments to explore the functional interactions between TOR and IIS in determination of life span. When life span extension was maximized by rapamycin treatment, the effects of IIS manipulation were dependent upon the degree of IIS downregulation. Thus, while rapamycin treatment further extended the life span of flies heterozygous for chico1
, a null mutation in the gene encoding the single Drosophila
IRS homolog, it shortened the life span of long-lived chico1
homozygous flies. Rapamycin treatment was hence beneficial under conditions of weaker IIS downregulation, possibly indicating deleterious combined effect with stronger IIS downregulation. Alternatively, extension of life span by rapamycin may require chico1
. Rapamycin did not alter the life span of flies in which the insulin-producing mNSCs were partially ablated, possibly because they are intermediate in degree of downregulation of IIS. Since life span under rapamycin treatment was maximized, the data from chico1
heterozygotes suggest that IIS and TOR regulate life span, at least in part, by nonoverlapping mechanisms. In C. elegans,
life span extension of daf-2
(the worm insulin receptor) mutants was not modified by RNAi against TOR, suggesting that these two pathways may have overlapping downstream targets (Hansen et al., 2007; Vellai et al., 2003
). The IIS and TOR pathways may converge on a common downstream target, with severe inhibition of the two pathways causing detrimental effects, for example, through the induction of apoptosis (Talapatra and Thompson, 2001
). An important downstream effector of IIS-mediated life span extension is the Forkhead transcription factor Foxo/DAF-16. However, in C. elegans
, the life span of daf-16
null mutants can be extended by reductions in TOR signaling (Hansen et al., 2007
), and, similarly, rapamycin treatment can extend the life span of dFOXO null flies (C.S. and L.P., unpublished data), suggesting that Foxo is not required for rapamycin to extend life span.
Elevated levels of autophagy are generally considered to be beneficial for the prevention of aging, due to increased rates of removal of damaged molecules and organelles (Klionsky, 2007
). For example, upregulation of Atg8
in fly neurons extends life span and is associated with decreased amounts of insoluble ubiquinated and oxidatively damaged proteins (Simonsen et al., 2008
). Furthermore, downregulation of autophagy in C. elegans
shortens the life span of long-lived daf-2
mutant worms (Hansen et al., 2008; Melendez et al., 2003
), suggesting that autophagy is required for the longevity effects of IIS mutants. However, higher levels of autophagy alone do not appear to be sufficient for increased life span, because daf-16
mutation blocks daf-2
longevity but does not reduce autophagy levels (Hansen et al., 2008
). Rapamycin induces autophagy, possibly by altering the interaction of TORC1 with the autophagy proteins ATG13 and ATG1 (Chang and Neufeld, 2009; Ravikumar et al., 2004
). We have shown that downregulation of autophagy blocks rapamycin-mediated life span extension. Similar observations have also been made for chronological life span in yeast (Alvers et al., 2009
). It is interesting to note that, although autophagy seems to be an important contributor to longevity (Toth et al., 2008
), we did not observe shortening of life span upon reduced expression of Atg5
, which is in agreement with the previously published observations (Ren et al., 2009
Reduced protein translation can also have an antiaging effect. Mutations in genes encoding ribosomal proteins, S6K, or translation initiation factors, which are involved in translation, can all extend life span, in yeast, worms, and flies (Kaeberlein and Kennedy, 2008
). However, the precise mechanisms underlying the antiaging effects of reduced translation remain elusive, although several possible mechanisms have been proposed. For example, energy saved by lowering translation may be reinvested in longevity-promoting processes; stresses imposed by mutations in the translation machinery may result in cap-independent translation and enrichment of different sets of proteins; and improved protein homeostasis and better removal of damage may allow for increased life span when translation is altered (Kaeberlein and Kennedy, 2008
). Rapamycin has been shown to be a potent repressor of translation; both microarray and proteomic analyses have demonstrated that rapamycin significantly decreases translation of mRNA-encoding initiation factors and ribosomal proteins (Grolleau et al., 2002; Guertin et al., 2006
). This effect on translation is mediated by the downstream targets of TORC1: S6K and 4E-BP (Fingar and Blenis, 2004; Wullschleger et al., 2006
). We have shown that both a constitutive upregulation of S6K activity and the absence of 4E-BP block rapamycin-mediated life span extension, suggesting that downregulation of protein translation is an important mediator of the effects of rapamycin on life span.
Thus, we have demonstrated that the positive in vivo effects of rapamycin on life span are mediated by the TORC1-dependent downstream processes, autophagy and protein translation. However, it is not yet clear if the effects of autophagy and protein translation are additive or if their combined effects are required to extend life span. Downregulation of S6K alone is sufficient to extend life span in flies (Kapahi et al., 2004
). However, S6K activity modulates the activity of ATG1 kinase, an important regulator of autophagy, which may thus be affected in addition to translation (Lee et al., 2007
). Thus, these processes may act in concert to extend life span: autophagy may reduce cellular damage and thereby provide cells with ATP and amino acids, which can subsequently be used for cap-independent translation, stimulating the synthesis of proteins that are important for stress resistance and, perhaps, longevity.
Interestingly, rapamycin treatment increased median and maximum life span under DR. Life span of rapamycin-treated flies was increased at all food concentrations, suggesting that the effect of rapamycin on life span is at least partially independent of the effects of DR. Therefore, rapamycin treatment appears to capture all of the advantages of DR plus additional benefits. Moreover, although rapamycin-treated flies still responded to DR, the DR response was less pronounced than in controls, and, in particular, the rapamycin-treated flies were more resistant to the effects of full feeding on mortality. The mechanisms underlying DR and its interactions with the TOR pathway are complex. For example, DR in C. elegans
cannot extend the life span of long-lived TOR RNAi worms, but it does nevertheless extend the life span of mutants that have reduced levels of S6K and ribosomal proteins (Bishop and Guarente, 2007; Hansen et al., 2007; Mair and Dillin, 2008
). In Drosophila
, life span extension by DR cannot be increased further by ubiquitous overexpression of the TOR suppressor, TSC2 (Kapahi et al., 2004
). However, our data demonstrate that rapamycin treatment may involve additional longevity assurance pathways, because it can increase life span beyond the maximum achieved by DR. Interestingly, it was recently shown that 4E-BP extends life span under DR by enhancing mitochondrial activity (Zid et al., 2009
), which may underlie part of the beneficial effects of rapamycin.
In conclusion, we have shown that rapamycin treatment delays aging, an effect more pronounced than DR and that could further increase life span in combination with mild downregulation of IIS, firmly demonstrating that rapamycin could be used to study aging in flies. Future studies to investigate the effects of rapamycin in disease model systems may reveal potential common therapies for a wide range of age-related conditions.