Neuronal CBP-1 is Required for Enhanced Body Muscle Proteostasis in Response to Reduced Translation Downstream of mTOR

2.2. RNAi Experiments

RNAi knockdown treatments were performed as described in [40]. RNAi bacteria strains included empty vector L4440 (Addgene, Cambridge, MA, USA), ifg-1 (M110.4), cbp-1 (R10E11.1), daf-15 (C10C5.6, gifted from Han Lab), and rict-1 (F29C12.3) (Ahringer library, Source BioScience, Nottingham, UK). RNAi empty vector L4440 is referred to in text as control RNAi. RNAi was carried out from day 0 of adulthood, defined by the period immediately following L4 stage, but before animals have developed visible germline. RNAi empty vector L4440 is referred to in text (Figs. 1,​,22,​,33,​,4,4, Supplementary Figs. 1–4) as control RNAi. In Fig. 4 and RNAi screens shown in Supplementary Fig. 4, a dual RNAi strategy was used in which day 0 adults were placed on RNAi for ifg-1 or control for the first two days before being transferred to plates containing RNAi for cbp-1 or other screen target genes for 5 days prior to analysis. For a full list of RNAis used in screens shown in Supplementary Fig. 4, refer to Supplementary Table 1. For all assays except RNAi screen, animals were syncronized via 2–5 hours timed egg lays, duirng which gravid adults were allowed to lay eggs, before all animals were removed from plate, leaving behind only eggs. For the RNAi screen, animals were synchronized via bleaching gravid animals to obtain similarly staged embryos.

Open in a separate window

Fig. 1.

Reduced translation in germline or neurons primes the heat shock response (HSR) and increases thermotolerance.

(A) Time course of heat shock gene expression in N2 animals under ifg-1 RNA interference (RNAi) normalized to control RNAi at each time point shown, beginning at the onset of adulthood. Unpaired t-tests using Welch’s correction were conducted on ΔCts for each gene at each time point (Supplementary Table 3). (B) Heat shock gene expression of tissue-specific strains on ifg-1 RNAi normalized to control RNAi at day 7 of adulthood. From left to right, panels show N2 wild type animals, MAH23 (germline-specific RNAi strain), TU3335 (neuron-specific RNAi strain), WM118 (body muscle-specific RNAi strain), VP303 (intestine-specific RNAi strain), and NR222 (hypodermis-specific RNAi strain). Two-way ANOVAs for each strain were used comparing ΔCts of all genes (Supplementary Table 4). (C) Thermotolerance of tissue-specific RNAi strains after 1 week on RNAi. Unpaired t-tests using Welch’s correction were run at each time point for each strain (Supplementary Table 5). (D) Similar to (B) except that heat shock gene expression was measured 1 hour after exposure to 35 °C for 4 hours. Two-way ANOVAs were conducted for each strain comparing heat shock gene expression (Supplementary Table 4). Error bars represent means ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Open in a separate window

Fig. 2.

Reducing translation in the germline or neurons improves motility and restores youthful transcription of muscle maintenance genes.

(A) Paralysis assay of proteinopathy model expressing alpha-synuclein in body muscle (strain NL5901) on control or ifg-1 RNAi from the onset of adulthood. Kaplan-Meier survival curves were plotted for paralysis assays and compared using the Mantel-Cox log rank test, **** p < 0.0001. Additional replicates show similar results (Supplementary Table 6). (B) Representative microscope images of YFP-tagged alpha-synuclein in NL5901 on control and ifg-1 RNAi for 7 days. Scale bar = 200 μM. Images were collected from at least 9 animals per replicate across 4 replicates. (C) Quantification and comparison of fluorescence in NL5901 animals from (B) using an unpaired t-test with Welch’s Correction (Supplementary Table 7). (D) Western blot probing for alpha-synuclein and beta-tubulin proteins in NL5901 animals comparing conditions from (B) using an unpaired t-test with Welch’s Correction (Supplementary Table 8). (E) Expression of muscle structural and regulatory genes in NL5901 animals on control vs. ifg-1 RNAi at day 7 of adulthood compared using two-way ANOVA (Supplementary Table 9). (F) Aging-related changes in muscle-related gene expression of N2 animals on control RNAi using two-way ANOVA, **** p < 0.0001. (G) Expression of body muscle genes in N2 animals on control or ifg-1 RNAi from day 1 to day 7 of adulthood compared using two-way ANOVA, p < 0.0001. (H) From left to right, panels show body muscle gene expression on control or ifg-1 RNAi for the first 7 days of adulthood in MAH23 (germline-specific RNAi strain), TU3335 (neuron-specific RNAi strain), WM118 (body muscle-specific RNAi strain), VP303 (intestine-specific RNAi strain), and NR222 (hypodermis-specific RNAi strain). Two-way ANOVA was used (Supplementary Table 10). Error bars represent means ± SEM. * p < 0.05, ** p < 0.01, **** p < 0.0001.

Open in a separate window

Fig. 3.

Lowering translation in the germline or neurons upregulates muscle-related gene expression and improves motility in a model of body muscle proteotoxicity.

(A) Expression of muscle-related genes after 7 days of ifg-1 RNAi in the proteinopathy model strain expressing alpha-synuclein in body muscle crossed with tissue-specific RNAi strains (strains ANR149, ANR168, ANR153). Comparisons between animals treated with control or ifg-1 RNAi were carried out using two-way ANOVAs. (B) Paralysis assays carried out for strains and conditions in (A). Kaplan-Meier survival curves were plotted for paralysis assays and compared using the Mantel-Cox log rank test. Replicates for each strain were conducted with similar results (see Supplementary Table 6). Error bars represent means ± SEM. ** p < 0.01. **** p < 0.0001.

Open in a separate window

Fig. 4.

cbp-1 RNAi in neurons both dampens muscle-related gene expression changes as well as enhanced survival to proteotoxicity under low translation conditions.

(A) Expression of genes enriched in muscle in the neuron-specific RNAi strain TU3335 crossed with NL5901 (ANR168). Day 1 adults were placed on control or ifg-1 RNAi for 2 days before transfer to control or cbp-1 RNAi for 5 days. Two-way ANOVAs were conducted comparing ΔCts of the genes shown (Supplementary Table 11). (B) Western blot of alpha-synuclein (top) and total protein measured from Ultraviolet (UV) shadowing (bottom) for conditions in (A). Quantification and comparison of alpha-synuclein was run using unpaired t-tests with Welch’s Correction (right) (Supplementary Table 12). (C) Paralysis assay of the strain and conditions from (A). Kaplan-Meier survival curves were plotted for the paralysis assay and compared using the Mantel-Cox log rank test. p < 0.0001 (blue asterisk, ctl vs. ifg1:ctl) and p < 0.0001 (yellow asterisk, ctl:cbp-1 vs. ifg-1:cbp-1) (Supplementary Table 13). (D) Thermotolerance assays (5 hours at 37 °C) using RNAi conditions from (A) in the neuron-specific RNAi strain TU3335. Comparisons at different time points were carried out with unpaired t-tests using Welch’s correction compared to control RNAi (Supplementary Tables 14,15). Error bars represent means ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

2.3. Thermotolerance

Approximately 120 synchronized animals were maintained for each condition at 20 °C until exposure to heat stress (35 °C for 4 hours or 37 °C for 5 hours). Animals were allowed to recover for 1 hour before being scored for survival. Thereafter, survival was scored at daily intervals.

2.4. RNA Processing and Qualitative Polymerase Chain Reaction (qPCR)

RNA was isolated using Trizol Reagent (Fisher Scientific, Waltham, MA, USA) following the manufacturer’s instructions using 1–200 worms per sample. RNA samples were then processed with either Sureprep RNA Cleanup and Concentration Kit (Fisher BioReagents, Fair Lawn, NJ, USA) or RNA Clean and Concentrator (Zymo Research, Irvine, CA, USA). 200 ng RNA was reverse transcribed using QuantiTect Reverse Transcription Kit (Qiagen, Valencia, CA, USA). qPCR was performed in technical duplicate or triplicate using SYBR FAST qPCR Master Mix (Kapa Biosystems, Cape Town, South Africa) on a LightCycler 480 (Roche Applied Science, Indianapolis, IN, USA). Target gene mRNA expression was normalized to the housekeeping gene cdc-42 expression. Relative expression was determined by normalizing to control samples. Primer sequences are provided in Supplementary Table 2.

2.5. Paralysis Analysis

Synchronized NL5901, ANR149, ANR153, and ANR168 strains were maintained on control or ifg-1 RNAi plates starting at adulthood and transferred fresh RNAi plates daily. On days of paralysis measurement, 100–150 adults were transferred to a new RNAi plate with their bodies aligned in a straight line. After 10 minutes, worms that had not moved from their original location were gently tapped with a sterile platinum wire 2–3 times on the head. Worms able to move only their head (from the pharynx bulb to the tip of the head) were scored as paralyzed. Worms able to move between the pharynx and tail were considered not paralyzed. Worms unable to respond to touch were scored as dead. Paralyzed or dead worms were removed from the plate on the day of paralysis measurement.

2.6. Western Blotting

Synchronized NL5901 and ANR168 strains were maintained as they were in paralysis assays until day 7 when total protein extraction occurred. To determine the levels of alpha-synuclein proteins, western blotting was performed in triplicate. Total protein extraction and preparation was performed as previously described in [ 11]. Alpha-synuclein was detected using an anti-alpha-synuclein mouse monoclonal antibody (1:500 dilution) (Santa Cruz biotechnology, Santa Cruz, CA, USA). Beta-tubulin was detected using an anti-beta-tubulin mouse monoclonal antibody (E7, 1:500 dilution) (DSHB, Iowa City, IA, USA). Peroxidase-conjugated goat anti-mouse IgG secondary antibody (1:5000 dilution) was from Pierce (Rockford, IL, USA). The density of the bands was determined using ImageJ software (1.54i, NIH LOCI, University of Wisconsin (Madison, WI, USA)) and normalized according to beta-tubulin or total protein.

2.7. Imaging

Synchronized NL5901 animals were maintained were maintained on control or ifg-1 RNAi plates starting at adulthood and transferred fresh RNAi plates daily. Visualization and quantification of alpha-synuclein:: YFP expression was conducted after 7 days of control or ifg-1 RNAi exposure. Individual worms were mounted on a 2% agarose pad and immobilized in a drop of 25 mM levamisol. Worms were imaged on a Leica M 165 FC microscope with the YFP filter (excitation 510/20 nm, emission 560/40 nm). The intensity of YFP fluorescence was measured in ImageJ by closely tracing around individual worms. In total, approximately 30–40 worms were used per condition.

3.2. Reducing Translation in Neurons or the Germline Improves Motility and Restores Youthful Transcription of Muscle Maintenance Genes

The fact that selectively lowering translation in neurons or the germline improve survival from challenge with heat indicates that they are likely to improve maintenance of protein folding in other tissues. Testing this possibility requires a tissue-specific model of protein unfolding stress. For this, we employed motility assays using strain NL5901, which expresses alpha-synuclein fused with YFP in body muscle. This tissue-specific expression is due to the fact that the transgene is controlled by the major heavy myosin chain unc-54 gene promoter. Aging animals lose the ability to maintain proteostasis, which is highly exacerbated in body muscle in this strain due to expression of alpha-synuclein, resulting in paralysis in worm middle-age [47]. Here, we used paralysis to monitor the onset of proteostatic collapse in body muscle.

Prior to looking at tissue-specific effects, we tested the effect of whole body attenuation of ifg-1 during adulthood, which increased the motile period in NL5901 by an average of 44% (Fig. 2A). Unexpectedly, we observed that low translation led to increased alpha-synuclein::YFP fluorescence intensity by the end of the first week of adulthood (Fig. 2B,​,C).C). Increased protein expression was observed also in a Western blot probed with a monoclonal antibody specific for alpha-synuclein (Fig. 2D). Thus, despite increased motility under this condition, lowering systemic translation resulted in increased total protein expression for alpha-synuclein::YFP.

We wondered whether this result could be explained by increased transcriptional activity of the major heavy myosin chain gene (unc-54) promoter used to drive body muscle expression of alpha-synuclein. After seven days of ifg-1 RNAi started at the onset of adulthood, the transcript level of endogenous unc-54 was increased by more than 10-fold (Fig. 2E). In addition to unc-54, we tested expression of several other muscle structural and regulatory genes including muscle actin (act-4), minor myosin heavy chain (myo-3), troponin C (pat-10), tropomyosin (lev-11), myogenic transcription factors (hlh-1, unc-120), and the unc-23 gene encoding a negative regulator of proteosomal degradation. Expression of these genes increased in NL5901 subjected to ifg-1 RNAi. To rule out the possibility of this phenomenon being specific to this strain, we tested wild-type N2 animals with ifg-1 RNAi. A previous study showed that unc-54 decreased in the first week of adulthood in C. elegans [48]. We also observed decreased expression at the end of the first week of adulthood in wild-type animals for unc-54 and several other muscle structural and regulatory genes (Fig. 2F). Transcript expression of these muscle specific genes showed that lowering translation reversed the age-related loss of muscle specific gene expression in wild-type animals (Fig. 2G). Tissue-specific RNAi strains demonstrated that increased muscle gene expression resulted from lowering translation in the germline or neurons (Fig. 2H). Interestingly, lowering translation selectively in body muscle tissue did not induce body muscle expression significantly (Fig. 2H).

Because selectively lowering translation in neurons and germline resulted in HSR priming (Fig. 1) and an increase in muscle specific transcripts (Fig. 2), we hypothesized that selectively lowering translation in neurons or germline in the alpha-synuclein proteotoxicity model could extend their motile period and delay disease onset. We also wondered what effect lowering translation selectively in body muscle, where proteotoxicity occurs, could improve conditions in this model. Thus, we crossed the corresponding tissue-specific RNAi strains with NL5901. We confirmed that muscle structural and regulatory genes increased when translation was reduced selectively in the germline or neurons (Fig. 3A). Interestingly, slightly increased expression of muscle structural and regulatory genes was also observed when translation was selectively lowered in muscle (Fig. 3A). This result differs slightly from the results with wild-type N2 animals (Fig. 2H), which may be due to the constitutively perturbed muscle proteostasis resulting from expression of alpha-synuclein. The paralysis assay demonstrated that lowering translation in the germline or neuronal tissue resulted in the greatest increase in motility with age, whereas lowering translation in muscle resulted in a small protective effect (Fig. 3B). In summary, low translation in the neurons or germline improved body muscle proteostasis.

4.2. The Connection between Translational Regulation and Proteostasis

At the heart of improved muscle function and enhanced resistance to unfolded cellular protein resulting from low translation is reversal of age-related diminution of proteostatic maintenance. Many factors are involved in maintaining proteostasis, which involves synthesis, folding, and turnover of cellular protein. Nascent peptide chains emerging from the ribosome are managed by folding chaperones and are subject to ubiquitination, making translation a hub for all these processes [54]. The fact that everything from protein synthesis to degradation is regulated at the same location makes translation a critical process for maintaining this balance. We show that limiting protein synthesis controlled by an essential nutrient-responsive translation factor, eIF4G/IFG-1, acts as a lever to restore robustness of stress responsiveness through the HSR. Previously, we showed that at least one other proteostasis mechanism, the ER unfolded protein response, is enhanced when translation is reduced in a manner that depends on the gene encoding the HSR transcription factor, hsf-1 [11]. Experiments with single-celled organisms and mammalian tissue culture showed that intracellular stress signaling pathways can cross-activate one another to maintain cellular proteostasis [30,55–62]. Thus, HSR priming may improve function of other proteostasis mechanisms, a possibility that future experiments will need to address.

This work was supported by grants from the National Institutes of Health (R01AG062575) and by the Morris Scientific Discovery Fund. Research reported in this publication was also supported by an Institutional Development Award (IDeA) from the National Institute of General Medical Sciences of the National Institutes of Health under grant numbers P20GM0103423 and P20GM104318. Some strains were provided by the CGC, which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440).