Sci. Aging Knowl. Environ., 31 August 2005
Vol. 2005, Issue 35, p. pe26
[DOI: 10.1126/sageke.2005.35.pe26]


Interfering with Longevity

Siu Sylvia Lee

The author is in the Department of Molecular Biology and Genetics at Cornell University, Ithaca, NY 14850, USA. E-mail:SSL29{at}

Key Words: RNA interference • RNAi screen • C. elegansdaf-2daf-16 • mitochondria


The ability to rapidly reduce gene activity using RNA-mediated interference (RNAi) has substantially facilitated the dissection of gene function, often in ways that were not possible or feasible previously. In Caenorhabditis elegans, RNAi can be implemented via microinjection (1), soaking (2), or feeding (3). In the latter method (termed "feeding RNAi"), worms are fed Escherichia coli that express a specific double-stranded RNA (dsRNA). By a mechanism that is not well understood, worms are able to take up the dsRNA, which triggers RNAi and leads to down-regulation of the corresponding gene. Because of its ease of administration and compatibility with routine life-span analysis, this method in particular has been widely exploited by the C. elegans research community studying aging, leading to an explosion of new findings on C. elegans longevity.

RNAi-Based Longevity Screens

Several recent studies reported using genomic-scale feeding RNAi to identify a large number of new candidate longevity genes (4-6) (see Melov Perspective) (Fig. 1). Hamilton et al. (5) and Hansen et al. (4) independently screened the genome-wide RNAi library generated in the Ahringer lab (7, 8). This library harbors RNAi clones corresponding to over 80% of all the predicted open reading frames in the C. elegans genome. Both groups focused on RNAi clones that induce life-span extension. The specific screening procedures employed by the two groups were similar, but slightly different: Whereas Hamilton et al. (5) performed the screen with a wild-type strain and exposed worms to RNAi starting at the L1 larval stage, Hansen et al. (4) used a sterile strain [fer-15(b26); fem-1(hc17)] and exposed worms to RNAi starting at the embryonic stage. For the ease of high-throughput screening, both groups monitored the maximal life span exhibited by each RNAi-treated worm population. Upon multiple rounds of retesting, Hamilton et al. identified 91 RNAi clones that induced significant life-span extension and Hansen et al. identified 29 RNAi clones that induced significant life-span extension.

View larger version (34K):
[in this window]
[in a new window]
Fig. 1. Large-scale feeding RNAi has been used to systematically identify genes that, when down-regulated, extend C. elegans life span.

In addition to identifying some of the previously known longevity genes (such as those that encode components of the daf-2/insulin-like signaling pathway), both screens revealed a long list of new candidate longevity genes. These candidate genes encode proteins involved in diverse cellular processes, ranging from metabolism to stress responses, signal transduction, and the regulation of gene expression (4, 5). Interestingly, although a large number of new genes were identified, genetic interaction studies indicated that the majority of these new genes were associated with genetic pathways previously known to affect C. elegans longevity, in particular the daf-16-dependent pathway, the sir-2.1-dependent pathway, or the caloric restriction-related pathway (9). This finding raises the interesting question of whether the major pathways that determine C. elegans life span have already been identified. Even though manipulations of many genes can influence life span, they may do so by modulating a limited number of key pathways.

It is interesting to note that the two RNAi screens yielded largely distinct sets of candidate longevity genes (4, 5). Some of the differences likely arose because of the slightly different screening conditions, including strain background, the time at which animals were first exposed to RNAi, and the statistical cutoffs used to determine which clones significantly enhanced life span. However, the majority of the difference is probably a result of the intrinsic limitations associated with the feeding RNAi technique. Feeding RNAi is prone to producing false negatives (that is, a particular clone fails to cause an effect even though down-regulation of the corresponding gene will produce a specific phenotype) and can lead to quite variable results; variations between experiments, as well as within a single experiment, have been commonly observed (10, 11). This finding in fact suggests that many more longevity genes are likely to be discovered when the genome-wide RNAi library is screened to saturation.

Although we do not yet know whether in-depth analysis of the new candidate longevity genes will provide important insights into the regulation of C. elegans life span, these systematic longevity gene hunts represent important advances in C. elegans aging-related research. With the exception of age-1 (12) (see Johnson Subfield History), most of the C. elegans longevity genes we knew of prior to these studies were first identified and cloned based on various developmental phenotypes caused by mutations in these genes, and they were subsequently discovered, often serendipitously, to modulate life span as well. Because life span is a phenotype assayed at the population level, positional cloning of a longevity locus requires assaying the life span of hundreds to thousands of worm populations, an endeavor that demands exceptional labor, patience, and time. The implementation of the feeding RNAi technique makes large-scale surveys of longevity genes much more feasible. Additional RNAi screens using different RNAi libraries (7, 8, 11), different genetic backgrounds, and different target phenotypes such as delayed or premature aging, will most definitely be carried out in the near future. Together, these screens will provide a more comprehensive view of the genes that can influence C. elegans life span and will have important implications for the general understanding of longevity control.

Functional Validation by RNAi

RNAi also greatly assists the validation of findings obtained from other genomic-wide studies. Multiple microarray and informatic experiments have been carried out to identify genes regulated by the key longevity determinant DAF-16/FOXO (13-16), a transcription factor that acts downstream of the insulin-like receptor DAF-2 (see Antebi Perspective). RNAi adds to the power of these gene expression analyses by providing a means to quickly assay for the functional relevance of the candidate DAF-16 downstream genes. For example, Murphy et al. (14) performed microarray-based studies to identify DAF-16-regulated genes. The authors subsequently used the feeding RNAi technique to knock down expression from each of the DAF-16-regulated genes and analyzed whether any RNAi treatment might induce life-span extension in normal worms or suppress the prolonged life span displayed by daf-2 mutant worms. They reported the identification of more than 30 DAF-16-induced genes that normally promote longevity, including stress response genes and antimicrobial genes, and about 20 DAF-16-repressed genes that normally limit life span, including an insulin peptide. McElwee et al. (13) combined microarray studies, informatics, and RNAi to identify a protease that might act downstream of DAF-16 to promote life span (see Larsen Perspective and Kaeberlein Perspective). Lee et al. (15) and Ookuma et al. (16) both combined informatics and RNAi to identify putative DAF-16 direct target genes that are functionally important for life-span regulation (see "Extra CRISPy, Please"). The combination of gene expression studies with functional analysis greatly increases the possibility of recovering biologically important players. This enhancement is particularly important for multifunctional pathways such as the daf-2/insulin-like pathway, which regulates diverse physiological outputs in C. elegans. RNAi allows for the rapid functional identification of the specific downstream effectors that are most relevant to longevity.

RNAi Reveals the Roles of Essential Genes in Longevity

An additional bonus associated with RNAi is that it usually produces a hypomorphic phenotype. This feature is particularly useful in assessing the roles of essential genes in longevity. Strong loss-of-function mutations in this class of genes will result in developmental defects that preclude an investigation of their roles in life span. The ability of RNAi-based methods to identify such genes is evident in three studies showing that RNAi knockdown of many genes that play a role in mitochondrial function, including those that encode certain subunits of the electron transport chain (ETC) and a frataxin (6, 17, 18), leads to prolonged C. elegans adult life span (see "Long Life Starts Early"). This is likely a result of partial loss of gene function induced by RNAi, as null mutation of an ATP synthase subunit or a frataxin causes larval arrest (17, 19). Although previous forward genetic studies (that is, studies that begin with a random mutagenesis screen and end with the identification of the gene mutation that causes a phenotype of interest) have revealed that mutations in clk-1, which encodes an enzyme involved in ubiquinone synthesis, and isp-1, which encodes a complex III iron-sulfur protein, cause life-span extension (20, 21), the findings from these RNAi studies suggest that perhaps a general impairment of mitochondrial function, and not necessarily the specific gene mutation, is sufficient to extend C. elegans life span. Consistent with this idea, Dillin et al. showed that treating worms with the pharmacological agent antimycin A, which inhibits the activity of complex III in the ETC, similarly prolongs C. elegans life span (18). The investigation of the molecular mechanisms of how mitochondrial dysfunction extends life span will likely provide novel insights into the complex relation between metabolism, oxidative stress, and longevity. It is possible that some of the new candidate longevity genes identified in the recent genomic RNAi screens (4, 5) are also essential genes whose roles in longevity were missed previously because of developmental defects caused by mutations in these genes. Of course, the strength of the RNAi technique is also a weakness. A major limitation of RNAi is in fact that it usually only induces partial loss of gene function, which simply might not be sufficient to elicit a detectable phenotype for many genes. This shortcoming likely contributes to the high rate of false negatives associated with RNAi screens.

Mapping Temporal Regulation of Longevity by RNAi

RNAi is also versatile in that the timing of administration can be varied according to experimental needs. In a series of elegant experiments, Dillin et al. used the feeding RNAi technique to inactivate components of the insulin-like pathway (22), as well as the subunits of the ETC (18), to determine whether there is a specific time during life that down-regulation of these pathways will induce life-span extension (Fig. 2). These investigations revealed an interesting distinction: Insulin-like signaling must be down-regulated specifically during the first 4 days of adulthood to prolong life span, whereas the ETC must be down-regulated early in development to extend adult life span. Such specific temporal requirements suggest that, whereas the insulin-like receptor daf-2 signals through adulthood to determine life span, alterations in mitochondrial function might induce developmental changes early in life that persist to influence life span in adults. A fascinating possibility is that some aspects of longevity might be determined very early during development, whereas other aspects of longevity might remain plastic and could be further influenced throughout adulthood.

View larger version (6K):
[in this window]
[in a new window]
Fig. 2. A gene function can be inactivated at different times using RNAi, allowing the mapping of the temporal requirement of a gene in longevity control. Furthermore, genes essential for early development can be inactivated later in life to determine their specific roles in adult life span.

Temperature sensitive (ts) mutants have long been very useful tools that allow for the investigation of the activity of essential genes, as well as for the analysis of the temporal requirements of gene activities. However, the isolation of ts mutations can be labor intensive, and RNAi has again substantially facilitated these lines of investigation. RNAi, however, requires varying amounts of time to take effect, a limitation that undoubtedly obscures the temporal resolution of the experiments. Furthermore, the effect of RNAi is long-lasting. Unlike a ts mutation, the effect of which is reversed fairly quickly upon switching to the permissive temperature, the effect of RNAi knockdown persists, sometimes through generations. There is currently no simple way to terminate the effect of RNAi once dsRNA has been introduced into cells or animals. Dillin et al. (18, 22) attempted to terminate RNAi by inactivating dicer, a critical component of the RNAi machinery. The potential caveat is that dicer has multiple functions, including involvement in the processing of numerous micro RNAs (short RNAs generated from longer precursors, most of which are of unknown function), and its inactivation is likely to have consequence far beyond simply terminating the effect of the administered dsRNA. Therefore, whereas RNAi can be very useful in mapping the time that gene inactivation has to be initiated, it is less useful in defining the time that gene inactivation can be terminated.

Another major problem with RNAi is "cross-interference," in which a particular dsRNA molecule influences the activity of more than one gene. The specificity of RNAi has been a confusing issue, and the risk of off-target knockdowns might in fact be gene/sequence-specific (23, 24). It might be possible to circumvent "cross-interference" by confirming an observed phenotype with multiple independent RNAi constructs that are composed of distinct sequences. Different dsRNA sequences would likely induce distinct off-target interference that would result in variable secondary phenotypes.


In just a few years, RNAi has become an irreplaceable tool for investigating gene function. Although RNAi has its limitations, the good news is that many of these limitations stem from our ignorance of this fascinating biological process. With the ongoing intense research into the molecular mechanisms of RNAi, a more thorough understanding of this mysterious process will be a reality in the near future, undoubtedly also bringing about new improvements to this powerful technology. RNAi will continue to take center stage in aging-related research in a variety of model organisms, and an abundance of new insights into the molecular control of longevity should be forthcoming.

August 31, 2005
  1. A. Fire, S. Xu, M. K. Montgomery, S. A. Kostas, S. E. Driver, C. C. Mello, Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391, 806-811 (1998).[CrossRef][Medline]
  2. H. Tabara, A. Grishok, C. C. Mello, RNAi in C. elegans: Soaking in the genome sequence. Science 282, 430-431 (1998).[Free Full Text]
  3. L. Timmons, A. Fire, Specific interference by ingested dsRNA. Nature 395, 854 (1998).[CrossRef][Medline]
  4. M. Hansen, A. L. Hsu, A. Dillin, C. Kenyon, New genes tied to endocrine, metabolic, and dietary regulation of lifespan from a Caenorhabditis elegans genomic RNAi screen. PLoS Genet. 1, e17 (2005).[CrossRef]
  5. B. Hamilton, Y. Dong, M. Shindo, W. Liu, I. Odell, G. Ruvkun, S. S. Lee, A systematic RNAi screen for longevity genes in C. elegans. Genes Dev. 19, 1544-1555 (2005).[Abstract/Free Full Text]
  6. S. S. Lee, R. Y. N. Lee, A. G. Fraser, R. S. Kamath, J. Ahringer, Gary Ruvkun, A systematic RNAi screen identifies a critical role for mitochondria in C. elegans longevity. Nat. Genet. 33, 40-48 (2003).[CrossRef][Medline]
  7. R. S. Kamath, A. G. Fraser, Y. Dong, G. Poulin, R. Durbin, M. Gotta, A. Kanapin, N. Le Bot, S. Moreno, M. Sohrmann et al., Systematic functional analysis of the Caenorhabditis elegans genome using RNAi. Nature 421, 231-237 (2003).[CrossRef][Medline]
  8. A. G. Fraser, R. S. Kamath, P. Zipperlen, M. Martinez-Campos, M. Sohrmann, J. Ahringer, Functional genomic analysis of C. elegans chromosome I by systematic RNA interference. Nature 408, 325-330 (2000).[CrossRef][Medline]
  9. C. Kenyon, The plasticity of aging: Insights from long-lived mutants. Cell 120, 449-460 (2005).[CrossRef][Medline]
  10. F. Simmer, C. Moorman, A. M. van der Linden, E. Kuijk, P. V. van den Berghe, R. S. Kamath, A. G. Fraser, J. Ahringer, R. H. Plasterk, Genome-wide RNAi of C. elegans using the hypersensitive rrf-3 strain reveals novel gene functions. PLoS Biol. 1, E12 (2003).[Medline]
  11. J. F. Rual, J. Ceron, J. Koreth, T. Hao, A. S. Nicot, T. Hirozane-Kishikawa, J. Vandenhaute, S. H. Orkin, D. E. Hill, S. van den Heuvel et al., Toward improving Caenorhabditis elegans phenome mapping with an ORFeome-based RNAi library. Genome Res. 14, 2162-2168 (2004).[Abstract/Free Full Text]
  12. D. B. Friedman, T. E. Johnson, A mutation in the age-1 gene in Caenorhabditis elegans lengthens life and reduces hermaphrodite fertility. Genetics 118, 75-86 (1988).[Abstract/Free Full Text]
  13. J. McElwee, K. Bubb, J. H. Thomas, Transcriptional outputs of the Caenorhabditis elegans forkhead protein DAF-16. Aging Cell 2, 111-121 (2003).[CrossRef][Medline]
  14. C. T. Murphy, S. A. McCarroll, C. I. Bargmann, A. Fraser, R. S. Kamath, J. Ahringer, H. Li, C. Kenyon, Genes that act downstream of DAF-16 to influence the lifespan of Caenorhabditis elegans. Nature 424, 277-283 (2003).[CrossRef][Medline]
  15. S. S. Lee, S. Kennedy, A. C. Tolonen, G. Ruvkun, DAF-16 target genes that control C. elegans life-span and metabolism. Science 300, 644-647 (2003).[Abstract/Free Full Text]
  16. S. Ookuma, M. Fukuda, E. Nishida, Identification of a DAF-16 transcriptional target gene, scl-1, that regulates longevity and stress resistance in Caenorhabditis elegans. Curr. Biol. 13, 427-431 (2003).[CrossRef][Medline]
  17. N. Ventura, S. Rea, S. T. Henderson, I. Condo, T. E. Johnson, R. Testi, Reduced expression of frataxin extends the lifespan of Caenorhabditis elegans. Aging Cell 4, 109-112 (2005).[CrossRef][Medline]
  18. A. Dillin, A.-L. Hsu, N. Arantes-Oliveira, J. Lehrer-Graiwer, H. Hsin, A. Fraser, R. Kamath, J. Ahringer, C. Kenyon, Rates of behavior and aging specified by mitochondrial function during development. Science 298, 2398-2401 (2002).[Abstract/Free Full Text]
  19. W. Y. Tsang, L. C. Sayles, L. I. Grad, D. B. Pilgrim, B. D. Lemire, Mitochondrial respiratory chain deficiency in Caenorhabditis elegans results in developmental arrest and increased life span. J. Biol. Chem. 276, 32240-32246 (2001).[Abstract/Free Full Text]
  20. J. Feng, F. Bussiere, S. Hekimi, Mitochondrial electron transport is a key determinant of life span in Caenorhabditis elegans. Dev. Cell 1, 633-644 (2001).[CrossRef][Medline]
  21. B. Lakowski, S. Hekimi, Determination of life-span in Caenorhabditis elegans by four clock genes. Science 272, 1010-1013 (1996).[Abstract]
  22. A. Dillin, D. K. Crawford, C. Kenyon, Timing requirements for insulin/IGF-1 signaling in C. elegans. Science 298, 830-834 (2002).[Abstract/Free Full Text]
  23. S. P. Persengiev, X. Zhu, M. R. Green, Nonspecific, concentration-dependent stimulation and repression of mammalian gene expression by small interfering RNAs (siRNAs). RNA 10, 12-18 (2004).[Abstract/Free Full Text]
  24. P. C. Scacheri, O. Rozenblatt-Rosen, N. J. Caplen, T. G. Wolfsberg, L. Umayam, J. C. Lee, C. M. Hughes, K. S. Shanmugam, A. Bhattacharjee, M. Meyerson et al., Short interfering RNAs can induce unexpected and divergent changes in the levels of untargeted proteins in mammalian cells. Proc. Natl. Acad. Sci. U.S.A. 101, 1892-1897 (2004).[Abstract/Free Full Text]
Citation: S. S. Lee, Interfering with Longevity. Sci. Aging Knowl. Environ. 2005 (35), pe26 (2005).

Science of Aging Knowledge Environment. ISSN 1539-6150