IMPACT is a GCN2 inhibitor that limits lifespan in Caenorhabditis elegans
- Rafael C. Ferraz†1,
- Henrique Camara†1, 2,
- Evandro A. De-Souza1, 2,
- Silas Pinto1, 2,
- Ana Paula F. Pinca1, 2,
- Richard C. Silva3,
- Vitor N. Sato1,
- Beatriz A. Castilho3 and
- Marcelo A. Mori1, 2Email author
© Ferraz et al. 2016
Received: 13 June 2016
Accepted: 18 August 2016
Published: 7 October 2016
The General Control Nonderepressible 2 (GCN2) kinase is a conserved member of the integrated stress response (ISR) pathway that represses protein translation and helps cells to adapt to conditions of nutrient shortage. As such, GCN2 is required for longevity and stress resistance induced by dietary restriction (DR). IMPACT is an ancient protein that inhibits GCN2.
Here, we tested whether IMPACT down-regulation mimics the effects of DR in C. elegans. Knockdown of the C. elegans IMPACT homolog impt-1 activated the ISR pathway and increased lifespan and stress resistance of worms in a gcn-2-dependent manner. Impt-1 knockdown exacerbated DR-induced longevity and required several DR-activated transcription factors to extend lifespan, among them SKN-1 and DAF-16, which were induced during larval development and adulthood, respectively, in response to impt-1 RNAi.
IMPACT inhibits the ISR pathway, thus limiting the activation of stress response factors that are beneficial during aging and required under DR.
Overall decreases in fertility and mortality rates have aged the world’s population . Generalized and progressive tissue deterioration with aging is associated with impaired stress responses and results in increased risk of chronic diseases . The increasing number of elderly people has therefore raised public health concerns and highlighted the necessity of interventions to attenuate age-related dysfunctions.
Dietary restriction (DR) is one of the best characterized strategies that promote healthy aging. DR is defined as a reduction in food intake without malnutrition. This nutritional intervention prolongs lifespan in a variety of species , while also reducing fertility [4, 5]. DR acts by promoting stress response pathways and preventing age-related functional decline, delaying the appearance of cardiovascular diseases, type 2 diabetes, and neurodegeneration in mammals, including non-human primates . In C. elegans, different protocols of DR elicit different degrees of lifespan extension and stress resistance by often distinct and complementary mechanisms . The genetic model of DR – the eat-2 mutant – has a decreased pumping rate, thus ingesting less bacteria, and requires the FoxA transcription factor PHA-4  and the dimethoxy ubiquinone hydroxylase CLK-1  to extend its lifespan. Distinct and overlapping factors are involved in lifespan extension by bacterial DR (bDR) and liquid DR (lDR) – two protocols of bacterial dilution in liquid cultures. bDR also depends on PHA-4  and is partially dependent on the energy sensing kinase AMPK subunit AAK-2  and on the FoxO transcription factor DAF-16 [8, 10], while lDR depends on the NRF2 transcription factor homolog SKN-1 [11, 12]. The complete absence of bacteria requires the heat shock transcription factor HSF-1 to promote longevity , while dilution of medium peptone increases lifespan through AAK-2 and DAF-16 , and serial dilution of bacteria on semi-solid medium depends on the same factors in addition to CLK-1 [7, 14]. Stress resistance induced by amino acid restriction in mice [15, 16] and lifespan extension induced by DR or by inhibition of the worm homolog of the nutrient sensing kinase mTOR (LET-363) in C. elegans  also require the General Control Nonderepressible 2 kinase (GCN2 in mammals or GCN-2 in C. elegans).
GCN2 is an evolutionarily conserved serine/threonine kinase that senses amino acid restriction through binding to uncharged tRNAs . In most metazoans, GCN2 is one of the four known kinases that phosphorylate eukaryotic initiation factor 2, α subunit (eIF2α) to inhibit translation initiation . This leads to translation of an alternative and functional open reading frame of the ATF4 gene (atf-5 in C. elegans) [20, 21]. ATF4 is a transcription factor that targets stress response genes to confer an adaptive cellular response . This pathway is often referred to as the Integrated Stress Response (ISR) pathway, given that it integrates cellular response to multiple stress signals such as nutrient restriction (via GCN2), endoplasmic reticulum stress (via PERK), viral infection (via PKR) and metal deprivation (via HRI) [19, 22, 23]. Thus, GCN2-induced ISR activation in response to amino acid starvation serves as an important interface between nutritional cues and cellular resilience .
The mechanism of activation of GCN2 has been well characterized in S. cerevisiae, where Gcn2 is the only kinase that phosphorylates eIF2α. For Gcn2 to be activated in vivo it must interact with Gcn1 through its RWD domain, when both proteins are tethered to the ribosome . It is thought that Gcn1 functions to transfer to Gcn2 uncharged tRNAs that enter the A site of the ribosomes, thus activating it . Genetic studies in yeast revealed Yih1 (Yeast IMPACT Homolog 1) as a potent suppressor of Gcn2 function . Yih1 and its mammalian homolog IMPACT harbor a RWD domain that competes with Gcn2/GCN2 for the binding to Gcn1/GCN1 [27, 28]. Yih1 or IMPACT overexpression in yeast or mammalian cells, respectively, abrogates amino acid restriction-induced eIF2α phosphorylation [26, 28]. Like the other members of the GCN2 pathway, Yih1/IMPACT is found in virtually all eukaryotes and in all cells examined [29–31]. In mice, IMPACT is especially abundant in neurons [28, 32]. In these cells, down-regulation of IMPACT inhibits in vitro neuritogenesis by enhancing the basal levels of GCN2 activation, suggesting an important role for IMPACT in neuronal development . It was less clear, however, if IMPACT plays any role in organismal biology. Our hypothesis in the present study was that IMPACT acts to limit stress response at the whole organism level, hence counteracting the beneficial effects of DR and favoring age-related dysfunction. We used C. elegans to test this hypothesis and found that the worm homolog of IMPACT (henceforth named IMPT-1) is an inhibitor of GCN-2 that suppresses eIF2α phosphorylation even during fed states. impt-1 knockdown led to activation of the ISR, improved stress resistance, reduced fertility, decreased food intake, and extended longevity in the worms. These phenotypes resembled DR in all aspects, but DR was additive to impt-1 RNAi to extend lifespan. Consistent with a DR mimetic, impt-1 down-regulation activated key downstream players in the DR pathway. Our results highlight IMPT-1 as an important negative regulator of longevity in C. elegans, setting up the stage for a better characterization of its functions in mammalian aging and providing a new target for lifespan regulation.
Identification and functional characterization of the C. elegans IMPACT homolog
To identify the worm homolog of IMPACT, we browsed the C. elegans proteome for hits with close similarity to Mus musculus IMPACT protein. This analysis led us to C. elegans Y52B11A.2, which shared 33 % amino acid identity and 51 % amino acid similarity (E value = 7e–55) with mouse IMPACT (Additional file 1: Figure S1a) and displayed the UPF0029 and RWD domains (Additional file 1: Figure S1b), which are conserved among IMPACT proteins across the evolutionary spectrum. We henceforth named this protein IMPT-1 and its gene impt-1. A loss-of-function mutation in the impt-1 gene  rendered dead larvae when in homozygosis (data from the Caenorhabditis Genetics Center and our own observations). We therefore studied the heterozygous worms – a balanced strain called VC2511 or impt-1 +/– – and wildtype worms treated with impt-1 RNAi. Impt-1 mRNA levels were reduced by 77 % in impt-1 +/– mutants and by 84 % upon impt-1 RNAi in comparison to their respective controls (Additional file 2: Figure S2).
Impt-1 inhibition extends lifespan and confers stress resistance
To assess the responsiveness of stress pathways in impt-1 knockdown models, we measured thermotolerance and oxidative stress resistance. Upon chronic, mild temperature stress (28 °C), impt-1 +/– mutants lived 22 % longer than wildtype worms, while impt-1 RNAi-treated worms lived 8 % longer than their controls (Additional file 4: Figure S4a, b). Furthermore, 8-hour exposure to the pro-oxidant agent sodium arsenite at 5 mM or 7.5 mM revealed that impt-1 +/– mutants were more resistant to oxidative damage than N2 wildtype and gcn-2 mutant worms (Additional file 4: Figure S4c). This phenotype was also observed in N2 worms treated with impt-1 RNAi, and it was completely abrogated when impt-1 was silenced in gcn-2 mutants (Additional file 4: Figure S4d). Interestingly, gcn-2 mutants were more resistant to 7.5 mM arsenite than N2 worms (Additional file 4: Figure S4c), demonstrating a baseline stress resistance phenotype as observed in GCN2 knockout mice . Together, these data show that impt-1 knockdown promotes longevity and stress resistance in a gcn-2 dependent manner.
Impt-1 limits lifespan extension conferred by DR
Components of the DR pathway are activated by impt-1 knockdown in an age-dependent manner
Given that most of these mutations (including in the impt-1 gene) affect development, this could be the stage in which impt-1 and these DR genes interact to influence lifespan. Furthermore, developmentally impaired mutants might indirectly mitigate the ability of impt-1 knockdown to prolong lifespan if this phenotype is simply a manifestation of developmental traits. We therefore tested if silencing these genes and other DR-modulated genes only during adulthood could block longevity induced by impt-1 knockdown. When DR genes were silenced during the adult life of impt-1 +/– mutants, we found that aak-2 RNAi blunted impt-1 +/– extended longevity, while let-363 RNAi did not further increase it (Additional file 7: Figure S7c,d). Furthermore, impt-1 +/– mutants treated with daf-16 RNAi during adulthood not only failed to exhibit increased lifespan, but lived 7 % less than wildtype worms exposed to the same RNAi (Fig. 5b). However, loss of other DR genes (e.g., skn-1, hsf-1, and atf-5) after development did not affect impt-1 +/– longevity (Fig. 5c and Additional file 7: Figure S7e, f). The fact that all of the DR-related gene mutations studied here blocked the effects of impt-1 RNAi treatment starting at L1, but only some DR-related genes affected impt-1 +/– longevity when silenced during adulthood, suggested to us that impt-1 might control lifespan by interacting with some genes during adult life (e.g., daf-16, aak-2, and let-363) and others during the larval stages (e.g., skn-1, hsf-1, and atf-5). Consistent with this notion, when N2 and impt-1 +/– worms were maintained in control or skn-1 RNAi from L1 to L4 larval stages, and then transferred to plates with control RNAi for the rest of their adult life, larva-restricted skn-1 RNAi not only blocked lifespan extension of impt-1 +/– mutants but also reduced it by 12 % in comparison to N2 worms on the same regimen (Fig. 5d).
Next, we tested if impt-1 knockdown affected DAF-16 activity by measuring the expression of Pdaf-16::daf-16::GFP and the DAF-16 target construct Psod-3::GFP in response to impt-1 RNAi. The expression of Psod-3::GFP during the first days of adulthood was higher in worms treated with impt-1 RNAi when compared to control RNAi, with the biggest difference being at day 3 (Fig. 6b, c). The expression of Psod-3::GFP in larvae subjected to the same protocol was not different until the final larval stage (L4), demonstrating that DAF-16 activation by impt-1 knockdown starts at late larval/early adult stages (Additional file 8: Figure S8c). We also observed a trend towards nuclear localization of DAF-16 in day 3 worms under impt-1 RNAi (Fig. 6d). Taken together, these data demonstrate that several mediators of DR and the activation of stress response transcription factors are necessary to explain longevity induced by impt-1 knockdown, and this occurs in an age-dependent manner.
Finally, to assess if impt-1 expression was controlled by these DR-related transcription factors, we measured impt-1 mRNA in daf-16, skn-1, and hsf-1 loss-of-function mutants (Additional file 9: Figure S9a) and in a condition where they were activated, i.e., during DR (Additional file 9: Figure S9b). There were no significant differences whatsoever, indicating that impt-1 is constitutively expressed under these conditions and is not transcriptionally regulated by these stress-associated transcription factors. Nonetheless, impt-1 expression decreased with aging regardless of the diet (Additional file 9: Figure S9b) – a feature that could help activation of the ISR in aged individuals.
A similar phenomenon is observed when worms are offered excess NAD+ starting at late embryonic development [40, 41]. These protocols prolong lifespan by a mechanism that requires activation of the mitochondrial unfolded protein response (UPRmt) first observed in day 1 adults and that consequentially leads to DAF-16 translocation and sod-3 transcription in day 3 adults. Lifespan extension induced by NAD+ donors also requires the sirtuin SIR-2.1 and SKN-1 [41, 42], similarly to impt-1 knockdown-induced longevity (Fig. 5 and data not shown). Interestingly, mitochondrial stress induced by inhibition of certain components of the electron transport chain (i.e., isp-1 and clk-1) promotes eIF2α phosphorylation during development, activates UPRmt and prolongs lifespan in C. elegans, and this requires GCN-2 . Moreover, mild mitochondrial stress  or stimulation of UPRmt  during the larval stages are sufficient to render worms long lived. We thus suggest that multiple metabolic pathways can converge into GCN-2 and the ISR pathway during development to regulate lifespan and IMPT-1, as an inhibitor of the GCN-2 pathway, limits their effects.
How exactly the activation of GCN-2 and the ISR pathway initiates downstream events during development that culminate in lifespan extension is not absolutely clear. Since expressions of the ISR transcription factor ATF-5 and the SKN-1 target gene gst-4 are more prematurely up-regulated by impt-1 RNAi and both ATF-5 and SKN-1 are required during larval stages for the longevity phenotype of impt-1 knockdown models, we propose that these two transcription factors are acting in concert to trigger a longevity pathway. Indeed, in human cells, ATF4 (the mammalian ATF-5 homolog) has been shown to dimerize with NRF2 (the mammalian SKN-1 homolog) to activate the enhancer of the heme oxygenase-1 gene . In C. elegans, SKN-1 and ATF-5 are mutually regulated in response to endoplasmic reticulum (ER) or oxidative stress . Importantly, atf-5 is a SKN-1 target gene and atf-5 ablation prevents ER stress from inducing transcription of skn-1 and its target genes . Thus, these studies and our observations provide evidence to suggest that SKN-1 function is controlled by the ISR. However, NRF2 is also a direct substrate of the ER stress-regulated eIF2α kinase PERK, and it has been proposed to mediate PERK-dependent cell survival independently of eIF2α phosphorylation in mouse cells . Whether SKN-1 and ATF-5 act together or in parallel to control lifespan of C. elegans is a matter for future studies.
The importance of IMPT-1 and GCN-2 during larval stages is further supported by delayed development in the impt-1(ok3233) heterozygous mutants and mid-larval arrest in impt-1(ok3233) homozygous mutants and in N2 worms treated with krs-1 RNAi from eggs ( and confirmed by our own data). Given that both krs-1 and impt-1 ablations activate ISR when initiated in early development (Fig. 1 and ), these interventions could represent a stress signal that inhibits progression through the larval stages [48–50]. However, GCN-2 depletion by RNAi does not reverse the larval arrest phenotype of impt-1(ok3233) homozygous mutants (data not shown), indicating independence of GCN-2. Similarly, the knockdown of IMPACT in mouse neuronal cell lines partially affects neuritogenesis in a GCN2-independent manner . In addition, it has been shown that Yih1 controls yeast cell cycle independently of Gcn2 or Gcn1 . These phenotypes are in sharp contrast with the effects of impt-1 knockdown on longevity and stress resistance, which require both gcn-2 and gcn-1, evidencing the broad, yet poorly understood roles of IMPACT.
Impt-1 knockdown by RNAi or by its heterozygous null mutation render worms with very similar phenotypes but also with some fundamentally different ones. Both interventions decrease impt-1, reduce pharyngeal pumping, increase lifespan, and promote stress resistance at similar levels. On the other hand, while impt-1 +/– reduces brood size, delays development and decreases triglyceride content, impt-1 RNAi does not affect these parameters, or in the case of developmental timing even accelerates it. Impt-1 RNAi also does not promote eIF2α phosphorylation as impt-1 +/– does, at least not in the time point that we analyzed, but it increases ATF-5 expression, which is directly downstream of eIF2α. These distinctions may be explained by different hypotheses, including maternally inherited traits, RNAi delivery constraints or timing. More importantly, they highlight the fact that these phenotypes are not completely associated, and that impt-1 knockdown can increase lifespan without continuous hyperphosphorylation of eIF2α or affecting fertility, development, and fat accumulation.
In a broad spectrum, the phenotypes of impt-1 knockdown models can resemble the characteristics of dietary restricted animals, i.e., increased lifespan, improved stress response, delayed development, reduced fertility, decreased fat accumulation, and diminished food intake. Additionally, the levels of IMPT-1 and uncharged tRNAs seem to limit each other’s effects. For example, when impt-1 is decreased (e.g., in impt-1 +/– mutants or impt-1 RNAi) but uncharged tRNAs levels are low (e.g., during the fed state), lifespan extension is limited, as observed when uncharged tRNA levels increase (e.g., during DR) in the presence of high levels of IMPT-1 (e.g., in N2 worms or control RNAi). However, when DR and impt-1 knockdown are combined, lifespan is extended to its maximum under these circumstances. This could be interpreted through the perspective that DR and impt-1 knockdown act in parallel pathways. However, the genetic interaction of impt-1 with essentially all downstream components of the DR pathway suggests otherwise, that DR and impt-1 indeed share the same route to control lifespan. One alternative explanation is that impt-1 knockdown releases more GCN-1 to interact with GCN-2 and respond more efficiently to uncharged tRNAs that are more abundant during DR. Therefore, loss of impt-1 would have an effect under the fed state, and this effect would be exacerbated under DR. Considering that impt-1 knockdown (Fig. 2), eat-2  and krs-1 RNAi (Additional file 6: Figure S6) all increase lifespan in GCN-2 or GCN-1 dependent manners, we suggest that these interventions converge into activating GCN-2 to exert their effects.
However, how can impt-1 knockdown increase GCN-2 activation under the fed state? A similar effect was observed in mouse neuronal cells, which express high levels of IMPACT. The knockdown of IMPACT in these cells resulted in increased levels of phosphorylated GCN2, the active form of the kinase, and of phosphorylated eIF2α . It is possible that the resulting increase in GCN-1 available for interacting with GCN-2 will render GCN-2 more sensitive to the basal levels of uncharged tRNAs that occur naturally in the cells, as a result of each cycle of translation elongation as they exit the ribosomes.
And why would DR and krs-1 RNAi increase lifespan when applied to adult worms while impt-1 knockdown requires activation of SKN-1 during development? One possible explanation is a model where stoichiometric balance determines the effects. In such a model, uncharged tRNA levels are sufficiently high during development (perhaps due to increased protein synthesis) and IMPT-1 becomes rate limiting, exerting an important inhibitory pressure on GCN-2. On the other hand, uncharged tRNAs are less abundant in adults and therefore IMPT-1 is dispensable for GCN-2 regulation in the absence of a proper stimulus. This model is supported by the important functions of IMPT-1 during development and by the fact that krs-1 RNAi during adulthood adds to impt-1 +/– to prolong lifespan. Curiously, while krs-1 RNAi applied to adult N2 worms increases lifespan in our hands, another study reported decreased lifespan of N2 worms under the same RNAi clone . These discrepancies might be related to slightly different protocols that could sensitize or desensitize the GCN-2 pathway. One such difference in our study is the use of FUdR, which inhibits progeny.
Reduced food intake is usually the primary cause of DR. The fact that impt-1 knockdown causes the pharyngeal pumping rate to be decreased indicates that this phenotype may also explain or at least partially contribute to longevity. Interestingly, brain-specific activation of GCN2 and phosphorylation of eIF2α leads to an aversive behavior against amino acid deficient diets [52–54]. Hyperactivation of the GCN2 pathway may therefore represent an evolutionarily conserved hub to control food intake and metabolic adaptation in response to changes in nutrient availability.
In summary, reduction of IMPT-1 in worms is sufficient to trigger the ISR pathway and confer phenotypic characteristics that resemble DR, including extended longevity. Further studies are required to confirm the importance of IMPT-1 in vertebrates, but its highly conserved function and the broad evolutionary spectrum where the IMPACT/GCN2 pathway actuates gives us hints that our findings might be applied to more complex organisms. If so, IMPACT may serve as an important drug target to mimic DR in humans, in a way that could dissociate its beneficial effects from its heavy demands.
Identification of IMPACT in C. elegans
The C. elegans homolog of mouse IMPACT was identified using Blastp from NCBI. We browsed proteins in the C. elegans genome (Taxid: 6239) sharing sequence-similarity with Mus musculus IMPACT (GenBank: BAA35139.1) and analyzed predicted domains of Y52B11A.2 gene product of C. elegans by DELTA-BLAST from NCBI.
Construction of the impt-1 RNAi vector
Construction of impt-1 RNAi plasmid was similar to that reported elsewhere [55, 56]. A region corresponding to an exonic part of the impt-1 gene of N2 worms was amplified by PCR using the forward primer 5′-AATCTAGACTGCCGGGCAACACATAAT-3′ and the reverse primer 5′-AACTCGAGTATCCGCTTTCATTCGATCC-3′. The resulting PCR product was cloned into the L4440 vector using the XbaI and XhoI sites (regions in bold) and the Quick Ligation kit (New England Biolabs). Plasmids were transformed into E. coli TOP10 (One Shot® iTOP10 Chemically Competent E. coli). Recombinant clones were selected using ampicillin. Plasmids were extracted using QIAGEN Plasmid Mini Kit and sequenced to confirm the presence of the insert. Plasmids were used to transform E. coli HT115(DE3), which was used to feed the nematodes.
Strains and maintenance of C. elegans
C. elegans strains were cultured at 20 °C in streptomycin-supplemented (100 μg/mL) Nematode Growth Medium (NGM) with a lawn of E. coli OP50-1 as the food source, unless stated otherwise . The following strains were obtained from the Caenorhabditis Genetics Center (CGC): wild-type N2 Bristol, DA1116 eat-2(ad1116), RB976 gcn-2(ok871), ST60 gcn-1(nc40), VC2511 impt-1(ok3233) I/hT2 [bli-4(e937) let-?(q782) qIs48], MR507 aak-2(rr48), DR26 daf-16(m26), EU31 skn-1(zu135), PS3551 hsf-1(sy441), VP303 rde-1(ne219);kbIs7[nhx-2p::rde-1 + rol-6(su1006)], NR350 rde-1(ne219);kzIs20[(hlh-1p::rde-1) + pTG95(sur-5p::NLS::GFP)], TU3401 sid-1(pk3321);uIs69[pCFJ90(myo-2p::mCherry) + unc-119p::sid-1], CF1553 muIs84[(pAD76) sod-3::GFP + rol-6], TJ356 zIs356 [daf-16p::daf-16a/b::GFP + rol-6], and CL2166 dvIs19[(pAF15)gst-4p::GFP::NLS]. The strains atf-5(tm4397), its respective N2 control, and the Patf-5::GFP transgenic line were kindly provided by Dr. Keith Blackwell and Dr. David Ron.
For RNAi experiments, NGM was supplemented with ampicillin (100 μg/mL), tetracycline (12.5 μg/mL), and isopropyl β-D-1-thiogalactopyranoside (IPTG; 1 mM) and seeded with HT115(DE3) E. coli (OD 1.5 concentrated 10×). Expression of dsRNA was induced by addition of 1 mM IPTG to concentrated bacteria right before seeding the plates .
C. elegans protein lysates were obtained using approximately 700 synchronized day 0 adults. For positive control, drops of 5 mM DTT or vehicle were applied onto the worms and incubated at 20 °C for 3 hours. Worms were washed with sterile Milli-Q water and after decantation, the supernatant containing water and bacteria was removed. RIPA buffer (50 mM Tris pH:8.0, 150 mM NaCl, 1 % NP-40, 0.5 % sodium deoxycholate, 0.1 % SDS) was added and worms were shattered using metallic beads (Bullet Blender, Next Advance). Beads were removed and samples centrifuged at 17950 × g for 10 min at 4 °C. Supernatant was removed to obtain protein-containing lysates. Protein quantitation was performed using the BCA kit (Thermo Fisher). Immunodetection of phosphorylated eIF2α and total eIF2α was performed as described previously . The antibodies anti-Ser(P)52-eIF2α (Invitrogen, 44728G) and anti-Sui2 (eIF2α)  were used to detect phosphorylated and total eIF2α, respectively.
Approximately 150 synchronized young-adult worms were transferred to lifespan plates (NGM supplemented with 50 μM FUdR) containing OP50-1 or HT115(DE3) bacteria. For RNAi, worms were kept in control RNAi (L4440) or impt-1 RNAi starting from L1, and then transferred as young adults to lifespan plates containing the same bacteria as the initial food source. For RNAi only in adults, worms were kept in OP50-1 and then transferred as young adults to lifespan plates with HT115(DE3) expressing control vector or impt-1 RNAi. For RNAi in larvae, worms were kept in control RNAi or skn-1 RNAi and then transferred as young adults to lifespan plates with HT115 expressing control RNAi. Plates were maintained at 20 °C and monitored daily . For heat stress experiments, adult worms were kept at 28 °C. Dead worms were scored until the death of the last worm and the ones that crawled out of the agar, or showed bag-of-worms or exploded vulva were censored.
Brood size analysis
Brood size experiments were similar to previous reports . Synchronized hermaphrodites were maintained in OP50-1 or HT115(DE) expressing control vector or impt-1 RNAi and then transferred individually at the L4 stage to a well of a 24-well plate containing NGM media and the same bacteria used during development. Worms were transferred daily to a new well and scored until the end of the reproductive phase. After 24 h, for each day, larvae and eventually dead eggs that were laid by each worm were counted. Eight worms were used in each replicate.
Plates containing hundreds of synchronized day 1 adult worms were washed with 2 mL of sterile milli-Q water and transferred to Eppendorf conical tubes. After decantation, the supernatant containing water and bacteria was removed and 250 μL of a 5 % NP-40 solution was added. The worm suspension was heated at 75 °C for 10 minutes in a thermoshaker at 800 rpm. Samples were then vortexed and cooled at room temperature for 10 minutes. To eliminate the insoluble particles, tubes were centrifuged at 12000 × g for 5 minutes at 4 °C. The supernatant containing the lipids was then transferred to a new tube for triglyceride quantification. Triglyceride quantitation was performed using the LabTest kit according to the manufacturer’s instructions. Lipid extract (2 μL) or standard control (200 mg/dL) were added to 200 μL of detection buffer and the solution was heated at 37 °C for 10 minutes . The product was quantified at 490 nm using a plate reader (EL808 Ultra Microplate Reader). Concentration of triglycerides was estimated according to the standard control and then normalized by protein levels, as measured by the BCA kit using the same extract.
Developmental timing assay
Time of development was performed as previously described , with modifications. Briefly, several adult hermaphrodites were transferred to NGM plates containing OP50-1 or HT115 expressing impt-1 RNAi or control vector and let to lay eggs for 1–12 hours, depending on the experiment. Adults were then killed and, in some cases, when there were too many eggs on the plate, the eggs were transferred to new plates. Plates were monitored every 6 hours until the appearance of the first adult of the progeny and then every hour. Adults were counted and removed until all the worms reached adulthood.
Synchronized worms were maintained in NGM containing OP50-1 or HT115(DE3) for RNAi experiments. Pharyngeal pumping was scored at different developmental stages, as indicated. Pharynx movements were observed using a light stereoscope, and counted during a period of 10 seconds. Values were multiplied by 6 to the obtain pumping min–1 rate. At least two replicates of each developmental stage were performed and 10 worms were scored per day .
GFP reporter analysis
Synchronized L1 reporter strains (Patf-5::GFP, Psod-3::GFP, Pgst-4::GFP or Pdaf-16::daf-16::GFP) were maintained in control or impt-1 RNAi from L1 stage [40, 65]. Worms were transferred to a well of a 96-well plate containing 50 μL of M9 media (22 mM Na2HPO4, 22 mM KH2PO4, 85 mM NaCl, 1 mM MgSO4) at different time points, and immobilized with 0.1 % sodium azide right before images were acquired using the InCell Analyzer 2200 microscope (GE HealthCare). For atf-5 expression, drops of 5 mM DTT or vehicle were applied onto young adult worms for 3 hours prior to imaging. Fifteen worms of each group were transferred to new plates and photographed using a SteREO DiscoveryV8 modular stereoscope (Zeiss). Images were analyzed using ImageJ and mean intensity or integrated density of each worm was calculated. For the Pdaf-16::daf-16::GFP construct, the frequency of worms with high; medium or low levels of nucleated DAF-16::GFP were assessed qualitatively by a blind investigator. The worms were classified based on the intensity of nuclear DAF-16::GFP and the number of cells with nucleated versus cytoplasmic DAF-16::GFP.
Approximately 150 synchronized day 0 worms were collected and the total RNA was extracted using the TRIzol Reagent (Life Technologies), following the manufacturer’s instructions. RT-qPCR was performed as previously described . Briefly, the synthesis of cDNA was conducted using the High-Capacity cDNA Reverse Transcription kit (Life Technologies) using 1 μg of total RNA. After cDNA synthesis, gene expression was quantified by real-time PCR in an ABI 7000 detection system (Applied Biosystems) using the Maxima SYBR-Green Master Mix (Fermentas). The following primers were used: impt-1 forward 5′-CGAGGTTCGGAGTAAGGAAG-3′, impt-1 reverse 5′-CCCCATCATCCACACAATC-3′, his-10 foward 5′-GCAATTCGTCGTCTCGC-3′ and his-10 reverse 5′-GACTCCACGGGTTTCCT-3′. Impt-1 expression was normalized to his-10 expression. Specificity of amplification was tested by melting analysis.
For lifespan and heat stress analyses, survival comparisons between groups were made using the log-rank test. For comparisons between two means, a two-tailed Student t test was performed. Variance was tested and a non-parametric test was used when variance was different between groups. For more than two means, one-way analysis of variance (ANOVA) was conducted. For experiments with two independent variables, means were compared by two-way ANOVA. When no statistic interaction was observed, an overall independent variable effect was reported. In cases where statistic interaction between variables was detected, multiple comparison post-tests were performed and main effects reported. Some means were composed by pooled data from all replicates and are indicated as so. Comparisons were made using Graph Pad Prism 6 and null hypothesis was rejected when P < 0.05.
We thank Dr. T. Keith Blackwell and Dr. David Ron for providing the Patf-5::GFP reporters, the atf-5(tm4397) mutants and their control strains. The other strains were provided by the Caenorhabditis Genetics Center (CGC), which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440), or the international C. elegans Gene Knockout Consortium. We also thank Dr. Popi Syntichaki for the krs-1 RNAi clone. This research was financially supported by CNPq (444424/2014-8, 474397/2011-4) and FAPESP (2010/52557-0, 2015/01316-7, 2015/04264-8, 2012/24490-4, 2009/52047-5, 2014/17145-4, 2012/04064-0, 2014/25068-0, 2014/25270-3, 2014/10814-8).
MAM coordinated the project, planned the experiments and drafted the manuscript. RCF and HC performed the study and drafted the manuscript. APFP performed tissue-specific impt-1 RNAi lifespan assays. SP and VNS trained RCF and HC in C. elegans handling and helped in multiple experiments. SP performed oxidative stress assays. EAS performed GFP reporter assays. RCS performed the immunoblotting assays of phosphorylated eIF2α. BAC provided essential reagents and expertise to study IMPACT function. All authors contributed to the final manuscript.
The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Christensen K, Doblhammer G, Rau R, Vaupel JW. Ageing populations: the challenges ahead. Lancet. 2009;374(9696):1196–208.View ArticlePubMedPubMed CentralGoogle Scholar
- Kennedy BK, Berger SL, Brunet A, Campisi J, Cuervo AM, Epel ES, Franceschi C, Lithgow GJ, Morimoto RI, Pessin JE, et al. Geroscience: linking aging to chronic disease. Cell. 2014;159(4):709–13.View ArticlePubMedPubMed CentralGoogle Scholar
- Roth GS, Ingram DK, Lane MA. Caloric restriction in primates and relevance to humans. Ann N Y Acad Sci. 2001;928:305–15.View ArticlePubMedGoogle Scholar
- Crawford D, Libina N, Kenyon C. Caenorhabditis elegans integrates food and reproductive signals in lifespan determination. Aging Cell. 2007;6(5):715–21.View ArticlePubMedGoogle Scholar
- Dirks AJ, Leeuwenburgh C. Caloric restriction in humans: potential pitfalls and health concerns. Mech Ageing Dev. 2006;127(1):1–7.View ArticlePubMedGoogle Scholar
- Fontana L, Partridge L, Longo VD. Extending healthy life span--from yeast to humans. Science. 2010;328(5976):321–6.View ArticlePubMedPubMed CentralGoogle Scholar
- Greer EL, Brunet A. Different dietary restriction regimens extend lifespan by both independent and overlapping genetic pathways in C. elegans. Aging Cell. 2009;8(2):113–27.View ArticlePubMedPubMed CentralGoogle Scholar
- Panowski SH, Wolff S, Aguilaniu H, Durieux J, Dillin A. PHA-4/Foxa mediates diet-restriction-induced longevity of C. elegans. Nature. 2007;447(7144):550–5.View ArticlePubMedGoogle Scholar
- Lakowski B, Hekimi S. The genetics of caloric restriction in Caenorhabditis elegans. Proc Natl Acad Sci U S A. 1998;95(22):13091–6.View ArticlePubMedPubMed CentralGoogle Scholar
- Houthoofd K, Braeckman BP, Johnson TE, Vanfleteren JR. Life extension via dietary restriction is independent of the Ins/IGF-1 signalling pathway in Caenorhabditis elegans. Exp Gerontol. 2003;38(9):947–54.View ArticlePubMedGoogle Scholar
- Bishop NA, Guarente L. Two neurons mediate diet-restriction-induced longevity in C. elegans. Nature. 2007;447(7144):545–9.View ArticlePubMedGoogle Scholar
- Moroz N, Carmona JJ, Anderson E, Hart AC, Sinclair DA, Blackwell TK. Dietary restriction involves NAD(+)-dependent mechanisms and a shift toward oxidative metabolism. Aging Cell. 2014;13(6):1075–85.View ArticlePubMedPubMed CentralGoogle Scholar
- Steinkraus KA, Smith ED, Davis C, Carr D, Pendergrass WR, Sutphin GL, Kennedy BK, Kaeberlein M. Dietary restriction suppresses proteotoxicity and enhances longevity by an hsf-1-dependent mechanism in Caenorhabditis elegans. Aging Cell. 2008;7(3):394–404.View ArticlePubMedPubMed CentralGoogle Scholar
- Greer EL, Dowlatshahi D, Banko MR, Villen J, Hoang K, Blanchard D, Gygi SP, Brunet A. An AMPK-FOXO pathway mediates longevity induced by a novel method of dietary restriction in C. elegans. Curr Biol. 2007;17(19):1646–56.View ArticlePubMedPubMed CentralGoogle Scholar
- Ravindran R, Loebbermann J, Nakaya HI, Khan N, Ma H, Gama L, Machiah DK, Lawson B, Hakimpour P, Wang YC, et al. The amino acid sensor GCN2 controls gut inflammation by inhibiting inflammasome activation. Nature. 2016;531(7595):523–7.View ArticlePubMedPubMed CentralGoogle Scholar
- Peng W, Robertson L, Gallinetti J, Mejia P, Vose S, Charlip A, Chu T, Mitchell JR. Surgical stress resistance induced by single amino acid deprivation requires Gcn2 in mice. Sci Transl Med. 2012;4(118):118ra111.View ArticleGoogle Scholar
- Rousakis A, Vlassis A, Vlanti A, Patera S, Thireos G, Syntichaki P. The general control nonderepressible-2 kinase mediates stress response and longevity induced by target of rapamycin inactivation in Caenorhabditis elegans. Aging Cell. 2013;12(5):742–51.View ArticlePubMedPubMed CentralGoogle Scholar
- Sonenberg N, Hinnebusch AG. Regulation of translation initiation in eukaryotes: mechanisms and biological targets. Cell. 2009;136(4):731–45.View ArticlePubMedPubMed CentralGoogle Scholar
- Donnelly N, Gorman AM, Gupta S, Samali A. The eIF2alpha kinases: their structures and functions. Cell Mol Life Sci. 2013;70(19):3493–511.View ArticlePubMedGoogle Scholar
- Vattem KM, Wek RC. Reinitiation involving upstream ORFs regulates ATF4 mRNA translation in mammalian cells. Proc Natl Acad Sci U S A. 2004;101(31):11269–74.View ArticlePubMedPubMed CentralGoogle Scholar
- Lu PD, Harding HP, Ron D. Translation reinitiation at alternative open reading frames regulates gene expression in an integrated stress response. J Cell Biol. 2004;167(1):27–33.View ArticlePubMedPubMed CentralGoogle Scholar
- Harding HP, Zhang Y, Zeng H, Novoa I, Lu PD, Calfon M, Sadri N, Yun C, Popko B, Paules R, et al. An integrated stress response regulates amino acid metabolism and resistance to oxidative stress. Mol Cell. 2003;11(3):619–33.View ArticlePubMedGoogle Scholar
- Wek RC, Jiang HY, Anthony TG. Coping with stress: eIF2 kinases and translational control. Biochem Soc Trans. 2006;34(Pt 1):7–11.View ArticlePubMedGoogle Scholar
- Guo F, Cavener DR. The GCN2 eIF2alpha kinase regulates fatty-acid homeostasis in the liver during deprivation of an essential amino acid. Cell Metab. 2007;5(2):103–14.View ArticlePubMedGoogle Scholar
- Sattlegger E, Hinnebusch AG. Separate domains in GCN1 for binding protein kinase GCN2 and ribosomes are required for GCN2 activation in amino acid-starved cells. EMBO J. 2000;19(23):6622–33.View ArticlePubMedPubMed CentralGoogle Scholar
- Sattlegger E, Swanson MJ, Ashcraft EA, Jennings JL, Fekete RA, Link AJ, Hinnebusch AG. YIH1 is an actin-binding protein that inhibits protein kinase GCN2 and impairs general amino acid control when overexpressed. J Biol Chem. 2004;279(29):29952–62.View ArticlePubMedGoogle Scholar
- Kubota H, Sakaki Y, Ito T. GI domain-mediated association of the eukaryotic initiation factor 2alpha kinase GCN2 with its activator GCN1 is required for general amino acid control in budding yeast. J Biol Chem. 2000;275(27):20243–6.View ArticlePubMedGoogle Scholar
- Pereira CM, Sattlegger E, Jiang HY, Longo BM, Jaqueta CB, Hinnebusch AG, Wek RC, Mello LE, Castilho BA. IMPACT, a protein preferentially expressed in the mouse brain, binds GCN1 and inhibits GCN2 activation. J Biol Chem. 2005;280(31):28316–23.View ArticlePubMedGoogle Scholar
- Bittencourt S, Pereira CM, Avedissian M, Delamano A, Mello LE, Castilho BA. Distribution of the protein IMPACT, an inhibitor of GCN2, in the mouse, rat, and marmoset brain. J Comp Neurol. 2008;507(5):1811–30.View ArticlePubMedGoogle Scholar
- Cambiaghi TD, Pereira CM, Shanmugam R, Bolech M, Wek RC, Sattlegger E, Castilho BA. Evolutionarily conserved IMPACT impairs various stress responses that require GCN1 for activating the eIF2 kinase GCN2. Biochem Biophys Res Commun. 2014;443(2):592–7.View ArticlePubMedGoogle Scholar
- Hirose T, Horvitz HR. The translational regulators GCN-1 and ABCF-3 act together to promote apoptosis in C. elegans. PLoS Genet. 2014;10(8):e1004512.View ArticlePubMedPubMed CentralGoogle Scholar
- Roffe M, Hajj GN, Azevedo HF, Alves VS, Castilho BA. IMPACT is a developmentally regulated protein in neurons that opposes the eukaryotic initiation factor 2alpha kinase GCN2 in the modulation of neurite outgrowth. J Biol Chem. 2013;288(15):10860–9.View ArticlePubMedPubMed CentralGoogle Scholar
- C. elegans Deletion Mutant Consortium. Large-scale screening for targeted knockouts in the Caenorhabditis elegans genome. G3. 2012;2(11):1415–25.View ArticleGoogle Scholar
- Schindler AJ, Baugh LR, Sherwood DR. Identification of late larval stage developmental checkpoints in Caenorhabditis elegans regulated by insulin/IGF and steroid hormone signaling pathways. PLoS Genet. 2014;10(6):e1004426.View ArticlePubMedPubMed CentralGoogle Scholar
- Szewczyk NJ, Udranszky IA, Kozak E, Sunga J, Kim SK, Jacobson LA, Conley CA. Delayed development and lifespan extension as features of metabolic lifestyle alteration in C. elegans under dietary restriction. J Exp Biol. 2006;209(Pt 20):4129–39.View ArticlePubMedGoogle Scholar
- Nehrke K. A reduction in intestinal cell pHi due to loss of the Caenorhabditis elegans Na+/H+ exchanger NHX-2 increases life span. J Biol Chem. 2003;278(45):44657–66.View ArticlePubMedGoogle Scholar
- Hansen M, Flatt T, Aguilaniu H. Reproduction, fat metabolism, and life span: what is the connection? Cell Metab. 2013;17(1):10–9.View ArticlePubMedPubMed CentralGoogle Scholar
- Link CD, Johnson CJ. Reporter transgenes for study of oxidant stress in Caenorhabditis elegans. Methods Enzymol. 2002;353:497–505.View ArticlePubMedGoogle Scholar
- Park SK, Tedesco PM, Johnson TE. Oxidative stress and longevity in Caenorhabditis elegans as mediated by SKN-1. Aging Cell. 2009;8(3):258–69.View ArticlePubMedPubMed CentralGoogle Scholar
- Mouchiroud L, Houtkooper RH, Moullan N, Katsyuba E, Ryu D, Canto C, Mottis A, Jo YS, Viswanathan M, Schoonjans K, et al. The NAD(+)/sirtuin pathway modulates longevity through activation of mitochondrial UPR and FOXO signaling. Cell. 2013;154(2):430–41.View ArticlePubMedPubMed CentralGoogle Scholar
- Schmeisser K, Mansfeld J, Kuhlow D, Weimer S, Priebe S, Heiland I, Birringer M, Groth M, Segref A, Kanfi Y, et al. Role of sirtuins in lifespan regulation is linked to methylation of nicotinamide. Nat Chem Biol. 2013;9(11):693–700.View ArticlePubMedPubMed CentralGoogle Scholar
- Houtkooper RH, Mouchiroud L, Ryu D, Moullan N, Katsyuba E, Knott G, Williams RW, Auwerx J. Mitonuclear protein imbalance as a conserved longevity mechanism. Nature. 2013;497(7450):451–7.View ArticlePubMedPubMed CentralGoogle Scholar
- Baker BM, Nargund AM, Sun T, Haynes CM. Protective coupling of mitochondrial function and protein synthesis via the eIF2alpha kinase GCN-2. PLoS Genet. 2012;8(6):e1002760.View ArticlePubMedPubMed CentralGoogle Scholar
- Dillin A, Hsu AL, Arantes-Oliveira N, Lehrer-Graiwer J, Hsin H, Fraser AG, Kamath RS, Ahringer J, Kenyon C. Rates of behavior and aging specified by mitochondrial function during development. Science. 2002;298(5602):2398–401.View ArticlePubMedGoogle Scholar
- He CH, Gong P, Hu B, Stewart D, Choi ME, Choi AM, Alam J. Identification of activating transcription factor 4 (ATF4) as an Nrf2-interacting protein. Implication for heme oxygenase-1 gene regulation. J Biol Chem. 2001;276(24):20858–65.View ArticlePubMedGoogle Scholar
- Glover-Cutter KM, Lin S, Blackwell TK. Integration of the unfolded protein and oxidative stress responses through SKN-1/Nrf. PLoS Genet. 2013;9(9):e1003701.View ArticlePubMedPubMed CentralGoogle Scholar
- Cullinan SB, Zhang D, Hannink M, Arvisais E, Kaufman RJ, Diehl JA. Nrf2 is a direct PERK substrate and effector of PERK-dependent cell survival. Mol Cell Biol. 2003;23(20):7198–209.View ArticlePubMedPubMed CentralGoogle Scholar
- Baugh LR. To grow or not to grow: nutritional control of development during Caenorhabditis elegans L1 arrest. Genetics. 2013;194(3):539–55.View ArticlePubMedPubMed CentralGoogle Scholar
- Johnson TE, Mitchell DH, Kline S, Kemal R, Foy J. Arresting development arrests aging in the nematode Caenorhabditis elegans. Mech Ageing Dev. 1984;28(1):23–40.View ArticlePubMedGoogle Scholar
- Hu PJ. Dauer. WormBook. http://www.wormbook.org/chapters/www_dauer/dauer.pdf. Accessed 29 Aug 2016.
- Silva RC, Dautel M, Di Genova BM, Amberg DC, Castilho BA, Sattlegger E. The Gcn2 regulator Yih1 interacts with the cyclin dependent kinase Cdc28 and promotes cell cycle progression through G2/M in budding yeast. PLoS One. 2015;10(7):e0131070.View ArticlePubMedPubMed CentralGoogle Scholar
- Hao S, Sharp JW, Ross-Inta CM, McDaniel BJ, Anthony TG, Wek RC, Cavener DR, McGrath BC, Rudell JB, Koehnle TJ, et al. Uncharged tRNA and sensing of amino acid deficiency in mammalian piriform cortex. Science. 2005;307(5716):1776–8.View ArticlePubMedGoogle Scholar
- Maurin AC, Benani A, Lorsignol A, Brenachot X, Parry L, Carraro V, Guissard C, Averous J, Jousse C, Bruhat A, et al. Hypothalamic eIF2alpha signaling regulates food intake. Cell Rep. 2014;6(3):438–44.View ArticlePubMedPubMed CentralGoogle Scholar
- Maurin AC, Jousse C, Averous J, Parry L, Bruhat A, Cherasse Y, Zeng H, Zhang Y, Harding HP, Ron D, et al. The GCN2 kinase biases feeding behavior to maintain amino acid homeostasis in omnivores. Cell Metab. 2005;1(4):273–7.View ArticlePubMedGoogle Scholar
- Tamas I, Hodges E, Dessi P, Johnsen R, Vaz GA. A combined approach exploring gene function based on worm-human orthology. BMC Genomics. 2005;6:65.View ArticlePubMedPubMed CentralGoogle Scholar
- Timmons L, Fire A. Specific interference by ingested dsRNA. Nature. 1998;395(6705):854.View ArticlePubMedGoogle Scholar
- Stiernagle T. Maintenance of C. elegans. WormBook. http://www.wormbook.org/chapters/www_strainmaintain/strainmaintain.html. Accessed 29 Aug 2016.
- Ahringer J. Reverse genetics. WormBook. http://www.wormbook.org/chapters/www_introreversegenetics/introreversegenetics.pdf. Accessed 29 Aug 2016.
- Hashimoto NN, Carnevalli LS, Castilho BA. Translation initiation at non-AUG codons mediated by weakened association of eukaryotic initiation factor (eIF) 2 subunits. Biochem J. 2002;367(Pt 2):359–68.View ArticlePubMedPubMed CentralGoogle Scholar
- Sutphin GL, Kaeberlein M. Measuring Caenorhabditis elegans life span on solid media. J Vis Exp. 2009;(27). doi:10.3791/1152.
- Nehammer C, Podolska A, Mackowiak SD, Kagias K, Pocock R. Specific microRNAs regulate heat stress responses in Caenorhabditis elegans. Sci Rep. 2015;5:8866.View ArticlePubMedPubMed CentralGoogle Scholar
- Valente V, Maia RM, Vianna MC, Paco-Larson ML. Drosophila melanogaster lipins are tissue-regulated and developmentally regulated and present specific subcellular distributions. FEBS J. 2010;277(22):4775–88.View ArticlePubMedGoogle Scholar
- Govindan JA, Jayamani E, Zhang X, Mylonakis E, Ruvkun G. Dialogue between E. coli free radical pathways and the mitochondria of C. elegans. Proc Natl Acad Sci U S A. 2015;112(40):12456–61.View ArticlePubMedPubMed CentralGoogle Scholar
- Raizen D, Song BM, Trojanowski N, You YJ. Methods for measuring pharyngeal behaviors. WormBook. http://www.wormbook.org/chapters/www_measurepharyngeal/measurepharyngeal.html. Accessed 29 Aug 2016.
- Crook-McMahon HM, Olahova M, Button EL, Winter JJ, Veal EA. Genome-wide screening identifies new genes required for stress-induced phase 2 detoxification gene expression in animals. BMC Biol. 2014;12:64.View ArticlePubMedPubMed CentralGoogle Scholar
- Mori MA, Raghavan P, Thomou T, Boucher J, Robida-Stubbs S, Macotela Y, Russell SJ, Kirkland JL, Blackwell TK, Kahn CR. Role of microRNA processing in adipose tissue in stress defense and longevity. Cell Metab. 2012;16(3):336–47.View ArticlePubMedPubMed CentralGoogle Scholar
- DuRose JB, Tam AB, Niwa M. Intrinsic capacities of molecular sensors of the unfolded protein response to sense alternate forms of endoplasmic reticulum stress. Mol Biol Cell. 2006;17(7):3095–107.View ArticlePubMedPubMed CentralGoogle Scholar