Mutations in Nonessential eIF3k and eIF3l Genes Confer Lifespan Extension and Enhanced Resistance to ER Stress in Caenorhabditis elegans

Abstract

The translation initiation factor eIF3 is a multi-subunit protein complex that coordinates the assembly of the 43S pre-initiation complex in eukaryotes. Prior studies have demonstrated that not all subunits of eIF3 are essential for the initiation of translation, suggesting that some subunits may serve regulatory roles. Here, we show that loss-of-function mutations in the genes encoding the conserved eIF3k and eIF3l subunits of the translation initiation complex eIF3 result in a 40% extension in lifespan and enhanced resistance to endoplasmic reticulum (ER) stress in Caenorhabditis elegans. In contrast to previously described mutations in genes encoding translation initiation components that confer lifespan extension in C. elegans, loss-of-function mutations in eif-3.K or eif-3.L are viable, and mutants show normal rates of growth and development, and have wild-type levels of bulk protein synthesis. Lifespan extension resulting from EIF-3.K or EIF-3.L deficiency is suppressed by a mutation in the Forkhead family transcription factor DAF-16. Mutations in eif-3.K or eif-3.L also confer enhanced resistance to ER stress, independent of IRE-1-XBP-1, ATF-6, and PEK-1, and independent of DAF-16. Our data suggest a pivotal functional role for conserved eIF3k and eIF3l accessory subunits of eIF3 in the regulation of cellular and organismal responses to ER stress and aging.

Fig 1. Loss of eif-3.K or eif-3.L suppresses larval lethality of xbp-1 mutants on P. aeruginosa.

(A) Development assay monitoring the growth and viability of the indicated genotypes on E. coli or P. aeruginosa at 25°C. 50–100 eggs were laid on each plate and following 72h the fraction reaching the L4 larval stage or older were counted. Error bars reflect the S.D. of 3 plates. A Student’s t-test was used to assess significance: **P<0.01, ***P < 0.001. (B) Transmission electron microscopy of L3 larvae cultivated on E. coli or P. aeruginosa at 60,000x magnification visualizing intestinal or hypodermal ER morphology. Scale bar, 500nm. (C) Survival curves of L4 larvae of the indicated genotypes at 25°C following transfer to plates containing P. aeruginosa. Two biological replicates were performed with similar results.

Fig 2. eif-3.K and eif-3.L are the only nonessential subunits in C. elegans, and mutants lacking these subunits do not have phenotypes associated with attenuated protein translation.

(A) Individual eIF3 subunits were knocked down for one generation by RNAi feeding at 20°C. Eggs were then laid on the normal laboratory food source E. coli OP50 and evaluated for their ability to reach the larval stage L4 or older after 72h at 20°C. Inset: Interaction diagram of eIF3 subunits based on mass spectrometry of Zhou et al. [22]. Gray subunits represent the functional core of the mammalian complex sufficient to allow for initiation in vitro by Masutani et al. [26]. (B) Progeny of eif-3.J heterozygotes were assayed for sterility and then genotyped. Number of worms scored for each genotype is indicated in parentheses. (C) Developmental time course of the indicated genotypes. Populations were synchronized by egg lay and monitored over time for their development to the L4 larval stage or older. (D) Brood size of the indicated genotypes at 20°C. Error bars represent the S.D. of 10 animals. (E) Egg-laying rate was determined by transferring worms to new E. coli plates in 12h periods, and progeny were counted following 24h incubation at 20°C. Error bars represent the S.D. of 10 animals.

Fig 3. eif-3.K mutants do not have attenuated bulk translation.

(A) Polysome profiles of the indicated genotypes, harvested at the L4 stage. Identical quantities of total RNA were loaded onto sucrose gradients, and absorbance was measured at 254 nm. (B) Relative translation assayed by ribosome profiling, using whole yeast lysates as an internal standard. Whole lysates from S. cerevisiae were added to whole C. elegans lysates at a ratio of 1:20 based on total RNA concentration. Following ribosome profiling, the relative abundance of ribosome protected fragments (RPF) was determined by dividing the total number of reads that map unambiguously to the C. elegans genome by those that map unambiguously to the S. cerevisiae genome. Raw counts are presented in S1 Table. Statistical significance was assessed by the Student’s t-test.

Fig 4. Loss of eif-3.K or eif-3.L confers lifespan extension in C. elegans in a manner distinct from depletion of essential eIF3 subunits.

(A-C) Survival curves of the indicated genotypes at 25°C. Times indicated are days post-L4 stage. Lifespan statistics and replicate data is presented in S2 Table. For RNAi lifespan experiments, survival curves of the indicated genotypes transferred to control (gfp) or eif-3.A RNAi bacteria as L4 larvae. Worms were grown at 25°C for 3 days and shifted to 20°C for the remainder of the experiment.

Fig 5. Loss of eif-3.K or eif-3.L hyper-activates the DAF-16 transcription factor, which is required for lifespan extension.

(A) Survival curves of the indicated genotypes at 25°C. Times indicated are days post-L4 stage. Lifespan statistics and replicate data is presented in S2 Table. (B) Relative expression of sod-3, mtl-1, and dod-3 in the indicated genetic backgrounds as assayed by qRT-PCR of synchronous L4 larvae. Error bars represent the S.D. of three replicates. (C) Total fluorescence of the sod-3p::GFP transgene in worms of the indicated genotypes was quantitated at several adult timepoints. Error bars represent the S.D. of 10 worms. (D) Composite image depicting representative images of Day 3 adult worms of each indicated genotype expressing the transgene sod-3p::GFP.

Fig 6. Mutants lacking eif-3.K or eif-3.L are resistant to tunicamycin, independent of daf-16 and regulators of the Unfolded Protein Response.

(A) Animals were scored for their ability to reach the L4 stage or older 72h after 50–100 eggs were laid on plates containing 0, 2, or 5 μg/mL tunicamycin. Error bars reflect the S.D. of three plates. Two biological replicates were performed with similar results. (B) hsp-4 induction was measured by qRT-PCR after L4 larvae were transferred to plates containing 10 μg/mL tunicamycin for 4h. Error bars reflect the SEM of three replicates. A Student’s t-test was used to assess significance: *P < 0.05, **P < 0.01.

Fig 7. EIF-3.K::GFP is ubiquitously expressed in C. elegans and is localized to the cytosol.

(A) Lifespan curves of wild-type, the eif-3.K mutant, and a strain containing a fluorescently-tagged allele of eif-3.K. (B) Developmental assay monitoring the growth and viability of the indicated genotypes on P. aeruginosa at 25°C. Error bars reflect the S.D. of 3 plates. (C) Fluorescent micrograph of EIF-3.K::GFP at 10x magnification. (D) Fluorescent confocal image depicting intestinal nuclei of L4 larvae of the indicated genotypes at 20x magnification. Arrows indicate intestinal nuclei.