The autophagy receptor p62/SQST-1 promotes proteostasis and longevity in C. elegans by inducing autophagy

Autophagy can degrade cargos with the help of selective autophagy receptors such as p62/SQSTM1, which facilitates the degradation of ubiquitinated cargo. While the process of autophagy has been linked to aging, the impact of selective autophagy in lifespan regulation remains unclear. We have recently shown in Caenorhabditis elegans that transcript levels of sqst-1/p62 increase upon a hormetic heat shock, suggesting a role of SQST-1/p62 in stress response and aging. Here, we find that sqst-1/p62 is required for hormetic benefits of heat shock, including longevity, improved neuronal proteostasis, and autophagy induction. Furthermore, overexpression of SQST-1/p62 is sufficient to induce autophagy in distinct tissues, extend lifespan, and improve the fitness of mutants with defects in proteostasis in an autophagy-dependent manner. Collectively, these findings illustrate that increased expression of a selective autophagy receptor is sufficient to induce autophagy, enhance proteostasis and extend longevity, and demonstrate an important role for sqst-1/p62 in proteotoxic stress responses.

GFP-tagged LGG-1(G116A) mutant protein is defective in lipidation and therefore autophagosome targeting 1 , rendering it diffuse in the cytosol, and is used to verify that GFP:: LGG-1/Atg8-positive punctae are representing autophagic vesicles. Unexpectedly, nerve-ring neurons expressing GFP-tagged LGG-1(G116A) showed a propensity to form some LGG-1/Atg8-lipidation independent structures. This may be due to the increased expression of LGG-1/Atg8 under the control of the pan-neuronal rgef-1 promoter, similar to the increased formation of LC3(G120A) punctae observed in cells transfected to overexpressing LC3 2 . The nature of these lipidation-independent GFP:: LGG-1/Atg8-positive structures is as of yet unknown and requires further investigation. We note, however, that expression of GFP:: LGG-1 or GFP:: LGG-1(G116A) in C. elegans under control of the endogenous lgg-1 promoter does not appear to cause any punctae formation in nerve-ring neurons under basal conditions 3 , suggesting that LGG-1(G116A) punctae in nerve-ring neurons may represent artifacts of the overexpression conditions, rather than physiologically relevant structures.
As previously reported, we found that hypodermal seam cells displays a high propensity to form lipidationindependent structures 3 , and that BafA injections did not increase the number of GFP:: LGG-1/Atg8-positive punctae observed in sqst-1(ok2892) mutants. Collectively, these results are consistent with autophagy being blocked in hypodermal seam cells of sqst-1(ok2892) mutants, and we did not further pursue this cell type for autophagy phenotypes in this study.
Source data are provided in the Source Data file.

Supplementary Figure 3: Healthspan analyses of sqst-1 mutants.
(a-b) Wild-type (WT, N2), sqst-1(ok2892) (a), and sqst-1(syb764) (b) animals were assayed from day 1 through day 10 of adulthood for thrashing ability in liquid. In such swimming assays, the number of body bends was counted for 20 s after animals were placed into a drop of M9 media. (a) Data are the mean ± 95% CI of N=15 animals; P(m)=0.8; P(y-intercept)<0.001 for sqst-1(ok2892) by linear regression comparison. The experiment was performed at least three times with similar results. (b) Data are the mean ± 95% CI of N=14 animals; P(m)=0.9; P(y-intercept)<0.02 for sqst-1(syb764) by linear regression comparison. The experiment was performed twice with similar results. While we currently have no explanation for this phenotype, we could speculate that the reduced number of GFP:: LGG-1/Atg8-positive punctae in neurons of sqst-1 mutants, implying reduced levels of sqst-1dependent autophagy, could change the innervation or neurotransmitter release required for locomotion. More experiments are needed to address this point.
(c-d) Wild-type (WT, N2), sqst-1(ok2892) (c), and sqst-1(syb764) (d) animals were assayed from day 1 through day 10 of adulthood for pharyngeal pumping. For pharyngeal pumping assays, the number of contractions in the terminal pharyngeal bulb was counted for 20 s. (c) Data are the mean ± 95% CI of N=14 animals; P(m)=0.8; P(yintercept)=0.7 for sqst-1(ok2892) by linear regression comparison. This experiment was performed three times with similar results. (d) Data are the mean ± 95% CI of N=14 animals; P(m)=0.9; P(y-intercept)=0.09 for sqst-1(syb764) by linear regression comparison. This experiment was performed twice with similar results. If no error bar is clearly visible, it was too small to be displayed.
(e-f) Wild-type (WT, N2) sqst-1(ok2892) (e), and sqst-1(syb764) (f) were assayed for progeny production. For reproductive span, the number of eggs and larvae produced per day per animal were counted. (e) Data are the mean ± 95% CI of N=10 animals. P>0.05 for all time points by two-way ANOVA with Tukey's multiple comparisons. This experiment was repeated twice with similar results. Our results are in contrast to a previous report listing a reduced brood size in sqst-1(ok2892) mutants 4 . This discrepancy could be either due to different laboratory conditions or the elimination of this phenotype after multiple backcrossing to wild-type animals. (c) Transcript levels of sqst-1 in wild-type (N2) animals maintained under control conditions (CTRL) or subjected to heat shock (HS) for 1 h at 36 °C on day 1 of adulthood (HS on day 1) on day 1, 3, 5 and 7 of adulthood. Data are the mean ± SEM of three biological replicates, each with three technical replicates, and are normalized to the mean expression levels of three housekeeping genes and relative to samples of CTRL animals on day 1 of adulthood. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 by one-way ANOVA with Dunnett's multiple comparisons test for comparison of age-related increase of sqst-1 mRNA levels (black asterisks), and by twoway ANOVA with Tukey's multiple comparisons test for comparison of heat shock-dependent increase of sqst-1 mRNA levels (grey asterisks). Source data are provided in the Source Data file. (a) Schematic of differential detergent extraction experiment. 1-day old WT and sqst-1(ok2892) animals expressing neuronal Q40 proteins (rgef-1p::Q40::yfp) were either subjected to a 1 h heat shock at 36°C, or kept at control conditions. After the heat shock, animals were placed on media plates containing FuDR to prevent progeny production. Animals were aged until day 7 of adulthood and effects of heat shock on aggregate formation was confirmed microscopically. 100 µl of worm pellet (~30,000 animals) were lysed by homogenization using a Fastprep homogenizer (3 x 6000 rpm, 45 s, with 30 s pause). 200 µg of protein lysate was subjected to ultracentrifugation (100,000 g, 20 min, 4°C). The supernatant was transferred to a fresh tube and kept for Western blot analysis. The pellet was resuspended in 100 µl lysis buffer with increasing SDS concentration (0.25%, 0.5%, 0.75%, 1.0%, or 2% SDS), followed by another centrifugation step (100,000 g, 20 min, 4°C). After the last centrifugation step, the pellet was dissolved in 100 µl urea buffer. The samples were loaded in corresponding order on an SDS-gel (Lane 1-7, as indicated in the figure). Error bars indicate 95% CI. ns: P>0.05, *P<0.05, ****P<0.0001, by two-way ANOVA with Tukey's multiple comparisons test. (b) Representative Western blot of SQST-1-overexpressing animals (sqst-1p::sqst-1::gfp, SQST-1 OE (1) and SQST-1 OE (2)), and neuronal SQST-1 overexpressing animals (rgef-1p::sqst-1::gfp, Neuronal SQST-1 OE (1)) fed bacteria expressing either empty vector (CTRL) or dsRNA targeting autophagy-related gene lgg-1/ATG8 (lgg-1 RNAi) for four generations to block autophagy. 50 animals were handpicked for each strain. Antibodies against GFP (a-GFP) and actin (a-Actin) were used. Quantification of the abundance of SQST-1::GFP upon reduction of lgg-1/ATG8 was normalized to actin and is depicted as relative to control conditions from three independent experiments. Error bars indicate SD. We note that SQST-1 abundance significantly increased upon reducing the levels of lgg-1/ATG8 using RNAi in animals overexpressing GFP-tagged::SQST-1 from integrated loci. This indicates that the tagged SQST-1 protein is functional in that it can serve as a substrate for autophagy 3 .
In contrast, there was no increase in the abundance of SQST-1 after subjecting animals neuronally overexpressing GFP-tagged::SQST-1 to lgg-1/ATG8 using RNAi, presumably because neurons are largely refractory to RNAi 5 .
Source data for b-d are provided in the Source Data file.

Supplementary Figure 9: Healthspan analyses of C. elegans strains overexpressing SQST-1.
Wild-type animals (WT), transgenic animals stably overexpressing either sqst-1p::sqst-1::gfp (SQST-1 OE (1) and (2)) or rgef-1p::sqst-1::gfp (Neuronal SQST-1 OE (1) and (2)) were assayed from day 1 through day 9 of adulthood for (a-b) thrashing ability in liquid, (c-d) pharyngeal pumping, (e-f) reproductive span, and (g) body length. (e-f) For reproductive span, the number of eggs and larvae produced per day per animal were counted. Data are the mean ± SEM of N=10 animals per strain. ****P<0.0001 and P>0.05 for all other time points by two-way ANOVA with Tukey's multiple comparisons. We note that SQST-1 overexpression under the control of the panneuronal promoter, but not the endogenous promoter, lead to a decrease in the number of progeny production, indicating a possible trade-off between longevity and progeny production in animals overexpressing SQST-1 in neurons. This experiment was repeated twice with similar results. If no error bar is visible, it was too small to be displayed. Error bars indicate SD. Where error bars are missing, 0% or 100% paralysis was observed on all plates. ns P>0.05, ****P<0.0001 by two-way ANOVA with Tukey's multiple comparisons. This experiment was performed at least four times with similar results. Source data for a-c are provided in the Source Data file.
Source data are provided in the Source Data file.

Quantitative reverse transcriptase PCR
Quantitative reverse transcriptase (RT) PCR was performed as previously described 8 . Briefly, total RNA was isolated from a synchronized population of ~2,000 one-day-old nematodes raised on OP50 bacteria and maintained under control conditions or subjected to heat shock for 1 h at 36°C.
For quantitative PCR analyses of older animals, the synchronized animals were washed off daily with M9 medium, adult animals were sedimented by gravity and the floating larvae were aspirated. This washing step was repeated until no more floating larvae were detected. The adult animals were re-seeded onto 10 cm NGM plates with OP50 bacteria or harvested on the desired day of adulthood. After harvesting, the animals were flash frozen in liquid nitrogen. RNA was extracted with TRIzol (Life Technologies), purified using a Qiagen RNeasy kit, and subjected to an additional DNA digestion step (Qiagen DNase I kit). Reverse transcription (1 µg RNA per sample) was performed using M-MuLV reverse transcriptase and random 9-mer primers (New England Biolabs) 9 .
Quantitative RT-PCR was performed using SYBR Green Master Mix in an LC480 LightCycler (Roche).
A standard curve was obtained for each primer set by serially diluting a mixture of different complementary DNAs and the standard curves were used to convert the observed CT values to relative values. Three biological samples were analyzed, each with three technical replicates. mRNA levels of target genes were normalized to the mean of the following housekeeping genes: ama-1 (large subunit of RNA polymerase II), nhr-23 (nuclear hormone receptor), cdc-42 (Rho-GTPase) and pmp-3 (putative ABC transporter) 9,10 . Primer sequences are listed in Supplementary Table 6. Data are displayed as relative values compared with controls. Data were analyzed using multiple t-test or one-way ANOVA (GraphPad Prism).

Healthspan parameters
Thrashing ability (i.e., swimming), pharyngeal pumping, progeny production, and body length were assayed as measures of healthspan. For swimming assays, animals on the indicated days of adulthood were transferred onto an NGM media plate containing a drop of M9 medium, and body bends of 14-20 animals were counted for 20 s on a Leica stereoscope. Pharyngeal pumping was measured on the indicated days of adulthood by counting the grinder movements in the terminal pharyngeal bulb of 14-20 animals for 15 s on a Leica stereoscope. For the assessment of a progeny production profile, 10 animals were singled on 6 cm NGM plates at the L4 larval stage and transferred daily onto fresh plates during the self-fertile reproductive span. The number of eggs/larvae produced by each animal per day was counted. For measurements of body length, 10-12 animals on day 1 of adulthood were anaesthetized with M9 medium containing 0.1% NaN3 and aligned for imaging with a Leica DFC310 FX camera. Image analysis was performed with ImageJ software (National Institutes of Health) by tracing the length of the individual animals. Data were analyzed by student's t-test or one-way ANOVA as applicable (GraphPad Prism). using Beckman rotor TLA100. The supernatant was transferred to a fresh tube and kept for Western Blot analysis. The pellet was resuspended in 100 µl lysis buffer with increasing SDS concentration (0.25%, 0.5%, 0.75%, 1.0%, 2% SDS), subjected to heating and cooling (15 min 95°C, 15 min 25°), and spun down again using the TL-10 ultracentrifuge. After the last centrifugation step, the pellet was dissolved in 100 µl urea buffer (30 mM Tris-HCl, pH 8.0, 150 mM NaCl, 2% SDS, 1% Triton X-100, 7 M Urea, 2 M Thiourea, 0.5 mM PMSF, Complete
Western Blots were quantified using Image Lab, version 6 by Bio-Rad Laboratories, Inc. Briefly, PolyQ protein intensity from all samples was set to 100% and the fraction of soluble and insoluble PolyQ was calculated. Data were analyzed by two-way ANOVA (GraphPad Prism).

Proteasome activity
Animals were bleached and left overnight in M9 buffer at 20°C for hatching. The next day, L1 larvae were seeded on 10 cm NGM plates coated with OP50 bacteria and kept at 20°C until day 1 of adulthood. glp-1(e2144) animals were kept at 25°C for development. On day 1 of adulthood animals were washed off and reseeded on plates containing 100 µg/ml 5-Fluoro-2′-deoxyuridine (FuDR) (Acros Organics) to prevent progeny development. On day 3 of adulthood animals were collected with M9 buffer, and the worm pellet was resuspended in proteasome buffer (50 mM Tris-HCl, pH 7.5, 10% glycerol, 5 mM MgCl2, 0.5 mM EDTA, 2 mM ATP, and 1 mM dithiothreitol).
Lysates were generated by glass bead disruption on ice using a Precellys 24 homogenizer (Bertin Technologies).
Debris were removed by two centrifugation steps, the first at 6000g for 5 min, the second at 13000g spin for 15 min, both at 4°C. Protein concentration was measured and a total of 20 μg of total protein lysate was transferred to a 96-well microtiter plate (BD Falcon) and incubated with fluorogenic Suc-LLVY-AMC substrate (Enzo, BML-P802) for the chymotrypsin activity of the 20S proteasome. Fluorescence (370 nm excitation, 445 nm emission) was monitored on a microplate fluorometer (Perkin Elmer EnSpire) every 5 min for 1 h at 25°C. The slope for the fluorescence increase is calculated by linear regression using Microsoft Excel.
At this time, 10-30 animals of each strain were placed on three-four 6 cm NGM plates and were incubated at 25°C until day 3 of adulthood. Paralysis was scored when animals did not move voluntarily.