Hyperactivation of the proteasome core particle in Caenorhabditis elegans protects against proteotoxic stress and extends lifespan

Many age-related diseases (ARDs) including virtually all neurodegenerative diseases (NDs) are characterized by the accumulation of proteins that are thought to significantly contribute to disease pathogenesis. One of the cell’s primary systems for the degradation of misfolded/damaged proteins is the Ubiquitin Proteasome System (UPS), and its impairment is implicated in essentially all NDs. Thus, upregulating this system to combat NDs has garnered a great deal of interest in recent years. Various animal models have focused on increasing the total proteasome levels, but thus far, none have focused on intrinsic activation of the proteasome itself. With this in mind, we constructed a, first to our knowledge, animal model that endogenously expresses a hyperactive open-gate proteasome in Caenorhabditis elegans ( C. elegans ). The gate-destabilizing mutation introduced into the nematode germline created a viable nematode population with substantially enhanced proteasomal peptidase and unstructured protein degradation activity. These CRISPR edited nematodes showed a significantly increased lifespan and substantial resistance to oxidative/proteotoxic stress with surprisingly mild consequential phenotypes. These results show that introducing a constitutively active proteasome into a multicellular organism is feasible and suggests targeting the proteasome gating mechanism as a valid approach for future ARD research efforts in mammals.


INTRODUCTION
Ageing, as defined by Gompertz law, is accompanied by the exponential increase in death rate with age 1 . This increase in death rate may be caused by many coinciding organismal, cellular, and molecular factors, but one specific inevitability that afflicts humans as we age is the disruption of cellular proteostasis leading to the buildup of damaged and aggregate prone proteins [2][3][4] . The buildup of such non-native proteins is thought to lead to ageing phenotypes as well as age related diseases (ARDs) 2,4-6 . Two systems exist in the cell that are responsible for degrading misfolded and damaged proteins: the ubiquitin proteasome system (UPS) and the lysosome. The UPS is highly regulated and is responsible for degrading individual misfolded, damaged, or unneeded proteins, while the lysosome is responsible for degrading larger cargo including organelles and large protein aggregates. Many ARDs, including virtually all neurodegenerative diseases are characterized by protein misfolding and accumulation, which is thought to be a result of decreased UPS function [7][8][9][10][11][12][13] . The definitive cause of decreased proteasome function has yet to be determined. One explanation is that proteasome activity decreases with age, which could lead to protein accumulation, especially later in life when neurodegenerative diseases primarily occur; a wide range of literature supports this hypothesis 7,[9][10][11][13][14][15][16][17][18][19][20] . In fact, synthetic proteasome impairment in mice and rats alone can cause pathologies and symptoms associated with neurodegenerative diseases [21][22][23][24][25] . To better understand the potential mechanisms of proteasome impairment in disease understanding the structure and architecture of this molecular machine is needed.
The 20S proteasome, the proteolytic component of the UPS, is a barrel like structure that has four stacked heptameric rings (⍺7, β7, β7, ⍺7) 26 . The ⍺-rings' N-termini form "gates" that deter non-specific degradation by interacting with one another to form a folded structure over the central pore blocking nonspecific substrate entry 27 . The β-rings house the proteolytic sites that cleave proteins. Several different regulatory caps that bind to the 20S proteasome exist in the cell to aid in regulating protein degradation by the proteasome, and many of these caps have been extensively characterized [28][29][30][31][32] . The 19S regulatory particle (RP) is one of the primary regulatory caps in the cytosol. It binds to one or both ends of the 20S proteasome creating the 26S proteasome. The 19S RP consists of a base and lid region. The base consists of a hexameric ring of AAA-ATPases for protein unfolding, and the lid region contains ubiquitin receptors and deubiquitinases for targeting substrates for degradation [33][34][35][36] . The C-termini of the ATPase subunits contain a HbYX (hydrophobic, tyrosine, any amino acid) motif, which docks into intersubunit pockets in the 20S complex 37 . Binding of the HbYX motif can cause conformational changes in the 20S α-subunits that trigger gate-opening, which is required for substrate entry into the 20S for protein degradation.
Many previous studies have reported that one cause of proteasomal inhibition in aging and disease states is the presence of oligomeric proteins and other types of protein aggregates 8,15,[38][39][40][41] . Our lab has recently shown that conformationally specific oligomeric forms of misfolded neurodegenerative disease associated proteins (i.e. Amyloid Beta, alpha Synuclein, and Huntingtin) can bind to and inhibit 20S proteasome activity by stabilizing a closed gate conformation even in the presence of the gate-opening HbYX motif 42 . However, this inhibition can be reversed with saturating levels of peptides baring the HbYX motif, highlighting the potential therapeutic opportunity in targeting the proteasomal gating mechanism 42 .
Recent studies have targeted proteasome activation or upregulation as a type of therapy to combat neurodegenerative disease or increase resistance to cellular stress 43 . More specifically, studies have shown that overexpression of the 20S ß5 subunit increases total proteasome levels resulting in an increase in lifespan and resistance to cellular and organismal stressors [44][45][46] . While these demonstrate the protective effects of increasing proteasome expression, this study sought to stimulate the intrinsic activity of endogenous proteasomes. A previous study in yeast has shown that an eleven-residue truncation of the ⍺3's N-terminus creates a 20S with a constitutively open proteasomal gate, which leads to dramatically increased 20S proteasome activity 27 . More recently, it was shown that the expression of this proteasome construct in the mammalian HEK293 cell-line leads to resistance to proteotoxic stress induced by tau overexpression 47 . It is important to note that this construct was endogenously introduced in Saccharomyces cerevisiae, whereas in HEK293 cells, the truncated subunit was exogenously overexpressed on a wild type α3 subunit background 27,47 . To our knowledge no animal model has ever been made with a similar gate-opening mutation. Creating this mutant in a multicellular organism poses many potential issues since the proteasome and its regulation is imperative for almost every cellular process including immune response, signal transduction, development, metabolism, and progression through the cell cycle [48][49][50]  showed that expression of this open-gate proteasome in yeast caused a hindered ability to exit stationary phase resulting in reduced population growth following nutrient deprivation. However, under normal conditions, population growth and cell division are normal 54 . In the present study, we successfully generate the first ever multicellular organism, Caenorhabditis elegans, that expresses an endogenous constitutively active proteasome through direct genome editing. We examine how this open-gate proteasome affects C. elegans biology and impacts its lifespan and resistance to oxidative and proteotoxic stresses.

Mutation Design
The 7 α-subunits of the eukaryotic proteasome contain N-terminal regions that fold over the central pore to block and allow for regulation of substrate entry. The N-termini of each alpha subunit, while highly conserved, differ slightly in sequence, length, and structure, and therefore, play unique roles in regulating gate closure. One commonality between all the eukaryotic alpha subunit N-termini is the presence of a YDR motif (Tyr-Asp-Arg[Ser]), which is essential for creating the hydrogen bonds responsible for the stability of a closed gate conformation 26,27,52,55,56 . The N-terminus of α3 is uniquely important for gating in that it extends across the length of the entry pore acting as an anchor by providing hydrogen bonding between the other alpha subunit N-termini resulting in a stable closed gate conformation 26 (Fig. 1A).
Previous studies in yeast and mammalian cells show that deletion of α3's N-terminal region including the YDR motif destabilizes the closed gate conformation resulting in a constitutively open gate and increased proteolytic activity 47,54 . Given the high sequence conservation of the α3 N-terminus among eukaryotes, we hypothesized that making this mutation in C. elegans would yield comparable results (Fig. 1B). Using Co-CRISPR Cas9 technology with the assistance of InVivo Biosystems (Oregon, USA), we were able to generate the first viable multicellular organism with this proteasomal gate-opening mutation.
Two identical mutant clones were generated separately to control for off-target editing resulting in the strains: COP1857 pas-3(knu746 [NTD del]) & COP1858 pas-3(knu747 [NTD del]), which we will refer to as α3ΔN throughout this study. In these mutants, 36 base pairs were deleted and replaced with 12 base pairs of new coding resulting in an 8 residue N-terminal deletion of the α3 subunit including the YDR motif (Supplemental Fig. 1A). The insertion of this new coding region allows for specific mutant identification using PCR. (Supplemental Fig. 1B).
Following the CRISPR, we performed whole genome sequencing on both strains to confirm successful α3 N-terminal truncation, and a bioinformatic analysis confirmed no editing in offtarget regions for both strains (Supplemental Fig. 1D). We next sought to determine if proteasome expression levels differed between α3ΔN and WT. As determined by western blot, the mutant proteasome showed no difference in expression compared to WT ( Fig. 1C; Supplemental Fig. 2). This is important to note as any differences in proteasome activity detected is not due to changes in expression but to the activity of the proteasome itself.

Proteasome Activity
After confirming the successful α3 N-terminal truncation and unchanged expression levels, we sought to determine whether this mutation did, in fact, cause increased proteasome activity in α3ΔN compared to WT. 20S activity can be measured in-vitro using small fluorogenic substrates as well as fluorescently labeled intrinsically disordered proteins (IDPs). Using the highly proteasome specific substrate, suc-LLVY-AMC, we observed a 13-fold increase in α3ΔN 20S activity in lysates compared to WT ( Fig. 2A). This data shows clear proteasome activation, but further analyses are required to determine the mechanism responsible. Proteasome activity can be stimulated through either gate opening or allosteric regulation of the internal active sites.
To confirm total proteasome activation has occurred via gate opening, it is important to confirm A. Top view of S. cerevisiae 20S (1YRP) α-ring with α3 subunit in blue and its N-terminal gating residues illustrated in red. B. α3 N-terminal sequence alignment of C. elegans, H. Sapiens, and S. cerevisiae showing homology, the YDR region demarcated between the red lines, and specific residues deleted to form a "gateless" 20S. C. SDS-PAGE followed by western blot probing for 20S α-subunits (Enzo, MCP231) using γ-tubulin as a loading control (N=3, P>0.05).
that activity is increased for all proteolytic active sites inside the 20S and not just one. Within the degradation chamber exists three distinct active sites that cleave at specific areas of proteins based on chemical identity: hydrophobic (chymotrypsin-like), Basic (trypsin-like), and Acidic (caspase-like). Incubating lysates with substrates targeted to each of these active sites in vitro, suc-LLVY-AMC (chymotrypsin-like), Z-LLE-AMC (caspase-like), and Boc-LRR-AMC (trypsinlike), we found WT activity was as much as 98% less active compared to α3ΔN (Fig. 2A&B).
Stimulation of all three of these active sites suggests activation via gate-opening rather than allosteric regulation of the active sites.
To further corroborate the gate opening mechanism and that the activity detected was indeed proteasomal, we ran lysates on a native gel and detected activity by incubating the gel in buffer containing suc-LLVY-AMC (in-gel activity). We then used SDS, which is a detergent that has been shown to induce gate opening at low concentrations (0.02%), to stimulate the WT to a more open gate state. The in-gel activity assay showed drastically higher 20S signal in both α3ΔN clones compared to WT, to the point where the WT signal was hardly detectable (Fig.   2C). The 26S could also been seen on this gel and its activity was also higher in the α3ΔN To determine the extent to which α3ΔN in C. elegans opens the 20S gate, we looked at activation capacity of the 26S proteasome. Typically, the 26S proteasome can be activated by 4 to 5-fold in the presence of ATP 51 , which induces 20S gate opening. When we added ATP to WT or α3ΔN lysates, we found that it stimulated suc-LLVY-AMC peptidase activity by 5.5-fold for WT and 2.3-fold for α3ΔN lysates ( Fig. 2D; Supplemental Fig. 4). This indicates that the mutation may not cause completely stable gate opening in these proteasomes since the 26S can become more active when saturated with ATP (Fig. 2E). The fact that the α3ΔN 26S proteasome can be further stimulated ATP indicates that these proteasomes may still offer a level of negative regulation that may contribute to the viability of these mutant populations. It is Figure 2. In-vitro proteasome activity assays shows strong activation of nematode α3ΔN 20S. All activity assays were performed using nematode lysates from synchronized young-adult populations.
possible that a completely open gate would be lethal in a multicellular organism given the crucial role of proteasomal regulation during development (see below for more). Previously, a different open-gate mutant had been generated in yeast with a truncation in both the α3 and α7 N-termini (α3α7ΔN) resulting in more extensive activation than α3ΔN alone 54 . We also attempted this double N-termini mutation in C. elegans and were only able to produce heterozygous populations as homozygosity appeared to be embryonic lethal. This observation is consistent with the hypothesis, that our α3ΔN 20S mutant is not fully active, which may contribute to the successful generation of this organism.
Thus far, we have measured proteolysis in-vitro using only small fluorogenic peptides, which is useful because it provides information about degradative capacity and the degree to which α3ΔN elicits gate opening. However, it does not necessarily provide direct evidence α3ΔN 20S's ability to degrade longer, more physiologically relevant substrates such as IDPs. To test α3ΔN C20S's ability to degrade IDPs, we used changes in fluorescence polarization (ΔmP) of the unstructured fluorescein isothiocyanate labeled casein (FITC-casein). When we incubated α3ΔN lysates with FITC-Casein, we observed a 4-fold higher ΔmP than that of WT ( Fig. 2E&F) demonstrating the increased capacity of α3ΔN to degrade unfolded proteins or IDPs. Taken together, the data shown thus far show an overall increase in proteasomal degradative capacity in α3ΔN. This characteristic may be helpful in protecting against organismal stressors known to illicit protein destabilization, unfolding and aggregation.

Lifespan and Other notable phenotypes
After confirming the viability and proteasome activation in α3ΔN, we sought to characterize their phenotypic differences relative to WT that might arise from a hyperactive 20S proteasome. A wide range of literature reports that proteasome activity declines with age in many model systems 7,[9][10][11][13][14][15][16][17][18][19][20] , and its inhibition leads to a dramatically decreased lifespan in C.
elegans 57 although, one study did show contrary results 58 . In addition, many long-lived nematode mutants are characterized by increased proteasome activity 59 , thus, we asked if hyperactivation of the proteasome core particle could impact lifespan. Visualized using a Kaplan-Meier curve, we found that open-gate mutants have a median lifespan of 20 days (about 3 weeks) compared to 17 days (about 2 and a half weeks) for WT nematodes which corresponds to a 20% increase in lifespan for α3ΔN (Fig 3A, P<0.0001). This is consistent with previous studies where proteasome upregulation increases cellular viability in mammalian systems 60 and extends the lifespans of c elegans and drosophila 44,46 . However, the mechanisms of activation studied in the context of lifespan extension have relied on β5 subunit overexpression to increase total proteasome levels, not activation of the proteasome itself. The data shown here confirms that intrinsic proteasome activation via gate opening leads to lifespan extension as well.
While open-gate proteasome expression resulted in no obvious structural abnormalities and an increased lifespan, we did notice a slower population growth in α3ΔN. To investigate this, we counted the number of viable offspring and found a >90% decrease in fecundity for α3ΔN compared to WT (Fig. 3b). This contrasts with a previous study that increased proteasome levels in C. elegans through active site overexpression and found that increasing proteasome levels elicited a 12% increase in fecundity 45 . This dichotomy demonstrates that upregulation of proteasome amounts with fully intact gating residues is not physiologically analogous to proteasome activation via gate-opening. Clearly, the loss of the ability to close the proteasome gate has a negative impact on reproductive and/or developmental systems, however, the disruptions are limited as some embryos do survive to adulthood.
Larval development timing is important to consider when using C. elegans because they may respond entirely differently to perturbations from one larval stage to another. Slower development is also associated with increased lifespan in many strains 61 so we sought to determine if α3ΔN's development timeline was delayed. C. elegans pass through four larval stages (L1-L4) before reaching adulthood, each of which consisting of a "lethargus" period, where feeding and locomotion are transiently arrested during the molting process 62 . These changes in activity can be measured and used to detect developmental milestones and calculate a timeline to adulthood. Using specially designed nematode WMicrotracker TM (InVivo Biosystems) to detect nematode activity levels in liquid culture, we found no significant timeline differences between α3ΔN and WT with both strains going through their developmental "lethargus" periods synchronously and both reaching adulthood within 50 hours (Fig. 3C). In addition to identical developmental progression, we also found that the two strains remained the same length to each other and grew at the same rate through each developmental stage measured, which further supports the consistent development timeline. However, after reaching adulthood and throughout the gravid period, we found that α3ΔN remained consistently shorter in length than WT (Fig. 3D). The precise reason for this in unclear but could be attributed to fewer eggs present in α3ΔN as described above or a decreased cell size due to differences in overall food consumption. We calculated E. coli (OP50) consumption rates by measuring daily changes in OD600 and found that α3ΔN consumed significantly less than WT (Fig. 3E). Because α3ΔN consumed less, we then measured pharyngeal pumping (pumps/min) to determine if this correlated with α3ΔN's decrease in overall food consumption and found that α3ΔN had consistently lower pumping frequency compared to WT over the first five days of adulthood (Fig.   3F). It is tempting to draw connections between reduced caloric intake (potentially due to increased satiation), and increased lifespan, as this is a constant theme in many other studies in other model systems 63 , including humans, but these connections are only hypothetical here and will require further investigation to be conclusive about direct links. Nonetheless, the data presented here demonstrates that expression of a hyperactive open gate 20S in C. elegans is viable, increases lifespan, and the only "major" physiological defect found is decreased fecundity, as the other phenotypic differences shown here are relatively small, even though statistically significant.

Resistance to oxidative stress
Given that the α3ΔN proteasomes in these nematodes show an increased ability to degrade unfolded proteins and that α3ΔN expression extends lifespan, we sought to determine whether this strain displayed resistance to oxidative stress, which induces general protein damage and unfolding. Paraquat is a potent herbicide, which produces reactive oxygen species in eukaryotes through disruption of the mitochondrial electron transport chain 64 . This compound can be toxic to many organisms, including humans. Recently, one study has found a positive correlation between paraquat exposure risk of Parkinson's disease 65,66 . For this reason, this compound has been used in studies with eukaryotes including c elegans as a toxin induced Parkinson's disease model 65,67,68 . In the context of proteostasis, oxidative stress has been observed to cause the dissociation of 19S from the 20S 69 and that the 20S is primarily responsible for the degradation of oxidatively damaged proteins [69][70][71][72][73] . With this in mind, we sought to determine whether our opengate strain displayed resistance to paraquat induced oxidative stress. We

Resistance to Heat Shock
Most eukaryotic proteins have evolved to function optimally at specific and restrictive temperature ranges. Elevated temperatures destabilize protein tertiary structure causing unfolding and subsequent protein aggregation. Under these conditions, cells undergo a heat shock response (HSR), where highly conserved heat shock proteins (HSPs) are upregulated and act as chaperones to attempt to quickly refold proteins. Unlike its function during oxidative stress, the role of the 20S proteasome in the HSR is poorly understood. There is evidence that proteasome activity increases during mild heat stress in human skin fibroblasts due to enhanced association with an 11S proteasome activator, which induces gate opening 74 . With this in mind and given that α3ΔN 20S expression during oxidative stress is protective, we sought to determine whether its expression is also protective under heat induced stress. The HS was performed by subjecting synchronized young adults to a 37˚C heat shock for 2 or 2.5 hours with an 18-hr recovery period followed by survival and paralysis scoring. After the 2-hour heat shock, ~30% of the WT population was paralyzed while the mutant strain remained largely unaffected ( Fig 5A) (P=0.0002). When combined with survival data (Fig. 5B), 36% of the WT population was adversely affected by a 2-hour HS compared to only 2% for α3ΔN (Fig. 5C). Since 2-hours at 37˚C was not sufficient to cause significant death in either strain (Fig. 5B, p=0.2395), we increased the HS by 30 minutes and repeated the analysis. After 2.5 hours at 37˚C, α3ΔN paralysis increased slightly to ~5%, and paralysis of the WT population remained similar to the 2-hour exposure (Fig. 5D) (P=0.0017). The mortality rate, however, was more significantly impacted; 50% WT population died compared to only 20% of the α3ΔN population (Fig. 5E) (P=<0.0001). Taken together, the 2.5-hour HS adversely impacted 75% of the WT population but only 25% of the mutant population (Fig. 5F). After determining that α3ΔN expression is protective during heat stress, we wanted to determine the role of protein clearance in this protection.
Western blots probing for ubiquitin 1 hour after the 2.5-hour HS showed that WT had a 26-fold increase in poly-ubiquitin accumulation and α3ΔN only had a 10-fold increase (Fig 5G;   Supplemental Fig. 5). Therefore, the accumulation of poly-ubiquitinated proteins in α3ΔN was less than half that of WT in both HS conditions. These results demonstrate that opening the 20S proteasome gate can stimulate the degradation of proteins that are unfolded due to thermal stress. These results are also consistent with our hypothesis that open-gate proteasome expression provides protection under heat induced stress by preventing protein accumulation.

Mass Spectroscopy
Global proteome changes caused by the hyperactive proteasome expression was characterized by quantitative proteomics based on tandem mass tag-mass spectroscopy (TMT-MS). Protein lysates were extracted from three biological replicates of 8-day old adult populations from WT and α3ΔN. We chose 8-day old adults because both strains are post reproductive, thus preventing the confounding variable of differing amounts of eggs/oocytes in WT vs α3ΔN during the reproductive period since α3ΔN appear to produce fewer eggs. At this stage, they would also be nearing "old-age" where ageing phenotypes begin to appear and ageing related proteomic changes can be detected 58,76 . The six samples were trypsin digested and labeled with TMT6plex Isobaric Label Reagent (Thermo), fractionated using HPLC, subjected to MS, and analyzed against C. elegans protein database using Maxquant resulting in the identification of 2843 proteins that were present in both strains (Fig 6A-C). Proteins of relative quantitation (α3ΔN/WT), with P<0.05, were divided into 1.5-fold upregulated or downregulated categories based on fold change (FC). Within these parameters, 18 proteins were upregulated, and 32 proteins were downregulated (Fig 6C).
We first sought out to determine if there were any changes detected in the abundance of proteasome subunits and other proteins related to the UPS due to open-gate expression.
Interestingly, the data indicated that 7 of the 14 19S subunits and just 1 of the 14 20S subunits were slightly upregulated in α3ΔN with a p-value<0.05. However, the maximum upregulation was only 18%, which is quite small and does not reach the usual minimum threshold of a 1.5-fold change. Among additional proteasomal regulators, PA200/blm10 also showed a statistically significant upregulation, but only by 18% (P=0.04). 1 of the 3 C. elegans P97/VCP homologs, cdc-48.2, appeared to be significantly upregulated by 1.29-fold (P=0.03), which is interesting since cdc-48 plays an enigmatic role in some types of ubiquitin-dependent protein degradation.
Taken together, the gate-opening mutations cause very minimal impact (<1.5-FC) on the levels of components involved with the proteasome machinery, including the proteasome itself.
Due to α3ΔN's enhanced ability to degrade unstructured proteins using lysates in-vitro, we also examined differences in IDP levels. Using the online software, Regression-based Accurate Prediction of protein Intrinsic Disorder content (RAPID) 77 , we determined the predicted level of intrinsic disorder for all proteins detected regardless of P-value and correlated them to their MS-based Log FC (Supplemental Fig. 6A). We did not detect any notable differences IDP levels for either up or downregulated protein groups. We also did not detect any notable differences when we plotted only statistically significant protein level differences ( Supplemental   Fig. 6B). These somewhat surprising results are likely due to compensatory mechanisms (e.g. increased IDP synthesis), but future more specific and targeted analysis will be required to make this conclusion definitively.
Using DAVID bioinformatics database, we performed a gene set enrichment analysis (GSEA) on genes encoding proteins in each group of significant differentially expressed proteins (≥1.5-fold, p<0.05) to identify enrichment of specific gene ontology terms (GO-terms). In the downregulated group, we identified "response to heat" as significantly enriched GO-term (FDR=0.01) (Supplemental Table 1 (Table 1). All these proteins are involved in the unfolded protein response (UPR), which is upregulated under conditions of stress due to heat, oxidative damage, and ageing. More specifically, proteins in the HSP-16 family act as passive ligands to prevent aggregation and are known to be upregulated under conditions of oxidative stress 78 as well as during treatment with protein damaging compounds 79 . HSP-17 has been shown to prevent aggregation of specific proteins, although its mechanism of action remains unclear 80 . Previous studies have shown that sHSPs are upregulated in many long-lived mutants including daf-2 81,82 , which is consistent with the observation that increased proteostasis protects from aging. Therefore, it is tempting to hypothesize that opening the proteasome gate in our mutants is directly enhances proteostasis, which causes a decreased need for proteostasis by other mechanisms (e.g., HSPs). The decrease in HSPs seen here suggests that it is unlikely that our mutant increases lifespan in a components. To the contrary, we did not detect significant changes in any daf-16 downstream factors or other proteostasis systems (chaperones, etc.). Therefore, while at first, it appears counterintuitive that sHSPs would be downregulated in our longer living mutant, this result is expected if opening the proteasome gate by itself increases the capacity of cellular proteostasis by accelerating the degradation of oxidatively damaged and unfolded proteins, thus reducing the need for downstream HSPs. This is seen clearly in our HS data showing decreased accumulation of polyubiquitinated proteins in α3ΔN after HS. In addition, the analysis also uncovered a downregulation of two proteins shown to be involved in the oxidative stress response, LEC-10 (↓1.6-FC, P=0.003) and MSRA-1 (↓1.6-FC, P=0.02) ( Table 1), which is consistent with α3ΔN's resistance to oxidative stress discussed above. Interestingly, we also identified a family of proteins downregulated in α3ΔN that are specifically named for being proteins upregulated in the long-lived daf-2 mutant: PUD-1, PUD-2, PUD-3, and PUD-4 (Table 1). In fact, of all the proteins differentially regulated in daf-2, PUD-1 and PUD-2 are the most upregulated of them all with a 4-fold increase in abundance 83 .

Function Gene Ratio α3ΔN/WT (Fold Change) p-value
However, the PUD proteins do not appear to be directly involved in the longevity or stress resistance seen in daf-2 and are independent of the daf-16 pathway 84 . Additionally, PUD-1 and PUD-2 appear to be regulated similarly to HSPs as they are upregulated after a temperature increase from 20˚C to 27˚C. However, they do not show any chaperone activity in vitro, so their role within the proteome remains unclear 84 . Nonetheless, the differences in PUD expression between α3ΔN and daf-2 may shed light on the cellular processes affected in these mutants.
Given that the PUD family and HSPs are induced in a similar way, and both HSPs and PUDs appear to be downregulated in α3ΔN, this further suggests that the daf-2 pathway is unaffected in our open-channel mutants.
To examine GSEs in α3ΔN's upregulated protein population, we followed the same procedure as above, but did not detect any significantly enriched categories (Supplemental Table 1). Therefore, we cannot make any definitive associations with potential GO-terms. There are several proteins, however, that are slightly above 1.2-fold upregulated (P<0.05) that are associated with reproduction and germ cell development. We hesitate to make any strong connections, but these proteins could be associated with α3ΔN's decreased fecundity discussed above. We should also note that the data identified one uncharacterized protein encoded by Y49G5A.1 with a notably higher upregulation (3.4-fold, P<0.0001) compared to the rest, which are mostly increased by less than 2-fold. According to WormBase (WS280), the protein contains a serine-type endopeptidase inhibitor domain, but it is not known whether this protein inhibits proteasome activity or if its upregulation is a compensatory mechanism attempting to counteract the increased proteasome activity. Generally, open-gate proteasome expression did not appear to cause many drastic changes in the global proteome, but the changes we do see align with our phenotypic observations and alludes to an overall healthier proteome.

CONCLUSION
In this study, we successfully generated a C. elegans animal model endogenously Many proteins downregulated in α3ΔN are proteins that are usually upregulated during stress (e.g., heat shock & oxidative stress) indicating a relative increase in proteostasis. Together, our data have shown here that expression of a hyperactive open-gate proteasome in a simple multicellular organism is not only feasible, but also increases lifespan and resistance to proteotoxic stress. These findings therefore support the hypothesis that activating proteasome function via inducing gate opening could be a viable and useful approach to increase proteostasis, and to potentially treat neurodegenerative diseases whereby proteostasis and protein degradation are perturbed.

Strains & Maintenance
Standard methods of culturing C. elegans were followed (Stiernagle, 2006 the solution to neutralize the reaction and the tubes were centrifuged at 3,000xg for 1 minute to pellet the eggs. Eggs were washed with 10mL sterile M9, centrifuged again for at 3,000xg for 1 minute, and M9 was discarded. The pelleted eggs were added to fresh 15mL conical tubes containing 7.5mL sterile M9 and incubated overnight with light agitation to allow for hatching. The resulting L1 arrested nematodes were harvested within 24hrs, plated on E. coli (OP50) seeded plates, and incubated in a 20˚C incubator until desired stage.

CRISPR-Cas9 approach
The CRISPR/Cas9 system was used with homology-directed repair (HDR) as described previously 86 to precisely delete amino acids 2-9 of pas-7. The strains produced are COP1857

In-gel Proteasome activity assay and western blot.
Native-PAGE in-gel 20S/26S proteasome activity assay was performed as described previously 42 . Protein Lysates (20µg) were mixed with Novex TM Tris-glycine sample buffer (Invitrogen) and separated on NuPAGE TM 3-8% Tris Acetate gels (Invitrogen) using Novex TM TRIS-glycine native running buffer (Invitrogen) containing 1mM DTT, 2mM ATP, and 5mM MgCl2 at 4˚C and 50V overnight. The gels were then incubated with slight agitation in proteasome activity buffer [50mM Tris-HCL pH 7.4, 2mM DTT, 5% glycerol, 2mM ATP, 5mM MgCl2, and 100µm suc-LLVY-AMC) for 30min at 20˚C and visualized using ethidium bromide imaging protocol on G:BOX XX9 (Syngene). Gate-opening was induced by incubating in activity buffer plus 0.02% SDS for 30min at 20˚C and imaged as described above. the WMT software and plotted using GraphPad (Prism). The data shown is the 2 nd order smoothing of the raw data.

Food Consumption
Food consumption was measured as described previously 90

Pharyngeal pumping
Pharyngeal pumping was manually counted using Stemi 508 (Zeiss) at 50X magnification in 10s intervals to determine pumps per minute. Statistical significance for each day was calculated using an unpaired Student's t-test in GraphPad (Prism) and a value of p<0.05 was considered statistically significant (N=15).

Paraquat treatment
Paraquat was used to measure oxidative stress sensitivity as described previously 91 .
Synchronized L4 or Day 1 adult nematodes were placed on 5 NGM plates containing 100mM paraquat and seeded with OP50. Each individual nematode was counted every hour for 25hrs and scored as dead or censored. Each Plate was counted as a replicate within each strain and statistical differences at each timepoint was calculated using an unpaired t-test (Prism 9). A Value of p<0.05 was considered statistically significant. Experiments were performed twice under blinding conditions.

Heat-shock treatment
Heat-Shock experiments were designed based on the methodological considerations discussed in Zevian & Yanowitz, 2014 92 . For each strain, synchronized, 120 young-adult worms were divided between 6 OP50 seeded NGM plates (20 worms/plate) at room temperature. Plates were placed in a 37°C incubator for either 2-hrs or 2.5-hrs then returned to 20°C for a 17-hr recovery period. After recovery, the individual nematodes were counted and scored as dead, paralyzed, or censored. Death was characterized by absolutely no movement following mechanical stimuli, and paralysis was characterized by the inability to move more than half its body length. Each plate counted as a replicate for each condition and strain. Unpaired t-tests were used to identify significant differences between strains. A value of p<0.05 was considered statistically significant. The 2-hour and 2.5-hour conditions were performed with separate populations and experiments were performed under blinding conditions.
Membranes were blocked for 1hr at room temperature in TBST+3% BSA, briefly rinsed with TBST, incubated with primary antibodies overnight at 4˚C, washed with TBST (3 x 5min), Incubated with secondary antibodies for 1hr at room temp, and washed with TBST (3 x 5min).

Mass Spectroscopy
Global protein levels were measured on 8-day old adult populations using tandem mass tagmass spectroscopy (TMT-MS). Three independent populations from each strain were collected on day 8 of adulthood, washed with M9 several times to remove bacteria, and the pellets were sent on dry ice to Creative Proteomics (NY, USA) for MS analysis. Detailed MS methods can be found in supplemental files. Gene set enrichment analysis on proteins differentially express by at least 1.5-fold was performed using DAVID bioinformatics database. We used our list of all proteins detected as the reference background. A false discovery rate (FDR) of less than 0.05 was considered statistically significant.

Statistical analysis
The data were analyzed using unpaired Student's t-tests and log-rank tests (Prism). For all statistical analyses, a value of p < 0.05 was considered significant.