The PP2A/4/6 subfamily of phosphoprotein phosphatases regulates DAF-16 during aging and confers resistance to environmental stress in postreproductive adult C. elegans

Insulin and insulin-like growth factors are longevity determinants that negatively regulate Forkhead box class O (FoxO) transcription factors. In C. elegans mutations that constitutively activate DAF-16, the ortholog of mammalian FoxO3a, extend lifespan by two-fold. While environmental insults induce DAF-16 activity in younger animals, it also becomes activated in an age-dependent manner in the absence of stress, modulating gene expression well into late adulthood. The mechanism by which DAF-16 activity is regulated during aging has not been defined. Since phosphorylation of DAF-16 generally leads to its inhibition, we asked whether phosphatases might be necessary for its increased transcriptional activity in adult C. elegans. We focused on the PP2A/4/6 subfamily of phosphoprotein phosphatases, members of which had been implicated to regulate DAF-16 under low insulin signaling conditions but had not been investigated during aging in wildtype animals. Using reverse genetics, we functionally characterized all C. elegans orthologs of human catalytic, regulatory, and scaffolding subunits of PP2A/4/6 holoenzymes in postreproductive adults. We found that PP2A complex constituents PAA-1 and PPTR-1 regulate DAF-16 during aging and that they cooperate with the catalytic subunit LET-92 to protect adult animals from ultraviolet radiation. PP4 complex members PPH-4.1/4.2, SMK-1, and PPFR-2 also appear to regulate DAF-16 in an age-dependent manner, and they contribute to innate immunity. Interestingly, SUR-6 but no other subunit of the PP2A complex was necessary for the survival of pathogen-infected animals, suggesting that a heterotypic PP4 complex functions during aging. Finally, we found that PP6 complex constituents PPH-6 and SAPS-1 contribute to host defense during aging, apparently without affecting DAF-16 transcriptional activity. Our studies indicate that a set of PP2A/4/6 complexes protect adult C. elegans from environmental stress, thus preserving healthspan. Therefore, along with their functions in cell division and development, the PP2A/4/6 phosphatases also appear to play critical roles later in life.


68
Perhaps more than any other, the Insulin and IGF-signaling (IIS) pathway has 69 been unequivocally established as a primary modulator of lifespan across evolutionarily diverse species (1). Genetic and behavioral changes that modify IIS pathway activity reached L4 on NGM plates seeded with OP50. Worms subjected to heat stress or UV 232 irradiation remained on RNAi plates until they died. For bacterial infection assays, worms 233 were transferred from the RNAi plates at either the L4 stage or at Day 6 of adulthood to 234 plates containing Pseudomonas aeruginosa where they remained throughout the 235 duration of the assay. 236 Pseudomonas aeruginosa (PA14) infection assays 237 Infection assays were carried out as described in Tan   Thermotolerance assay 248 After being synchronized by sodium hypochlorite treatment as described above, 249 L1 larvae were dropped on to RNAi plates seeded with a clone expressing dsRNA to 250 target a given gene of interest. Worms were maintained on those plates until the L4 larval 251 stage when they were split into two groups. The first group of worms was subjected to 252 thermal stress as described previously with some modifications (43). Briefly, 253 approximately 100 L4 worms were distributed evenly between three pre-seeded 3.5 cm RNAi food plates containing FUdR and then incubated at 35ºC for the duration of the heat 255 treatment and scored every few hours for survival with animals being counted as alive if 256 they responded to gentle prodding with a wire pick. The second group of worms, 257 consisting of ~2000 animals, was divided among three 6 cm pre-seeded RNAi + FUdR 258 plates by chunking and were maintained at 20˚C. At the sixth day of adulthood, ~100 of 259 these animals were subjected to heat stress according to the same procedure as outlined 260 above for L4 thermal stress assays.

297
In silico analysis of the C. elegans PP2A/4/6 phosphoprotein phosphatases 298 In both yeast and mammals significant structural and functional similarities 299 between members of the PP2A/4/6 family of phosphoprotein phosphatases exist. In addition, proteomic and smaller scale biochemical studies have revealed non-canonical 301 associations between subunits of the holoenzyme complexes (32,51,52). To ask how 302 the PP2A/4/6 proteins in C. elegans compare to their human counterparts we first took 303 an in silico approach. Alignments of the sequences of the three human catalytic 304 subunits PP2Ac, PP4c and PP6c with their C. elegans orthologs revealed strong 305 homology between them (Fig. S1, Table S1). For example, LET-92 and PPH-4.1 are 306 both greater than 80% identical to PP2Ac and PP4c, respectively. Key structural motifs 307 and regulatory residues are especially well-conserved both within catalytic subunits of 308 the PP2A/4/6 family and across species. Based on these similarities, we anticipated that 309 our studies would demonstrate significant parallels between these phosphatases in 310 worms and humans. 311 We next asked whether C. elegans PP2A/4/6 proteins were expected to 312 associate with the same binding partners as their human orthologs in vivo. To answer 313 this question we performed database searches, using SMK-1 and PPH-4.1 as 314 representative subunits for our analyses. We first used Wormbase to compile a list of 315 PPH-4.1 interactors (Table S2). Drawing mainly from biochemically verified 316 associations, high throughput yeast two-hybrid data, and interaction networks generated 317 in silico, this list captures a broad spectrum of potential interactors. Included among 318 them are not only proteins that could be members of a trimeric complex along with PPH-319 4.1 but also regulators, functional analogs, and possible substrates that may be 320 dephosphorylated by PPH-4.1. In addition to SMK-1, two other worm homologs of PP4 321 regulatory subunits, PPFR-1 and PPFR-2, were included as potential PPH-4.1 322 interactors. The only other worm orthologue of a PP4 regulatory subunit, F46C5.6, was absent from the list of PPH-4.1 interactors found on Wormbase. Both orthologs of the 324 yeast Tip41p-Tap42p regulatory system that modulates all members of the PP2A/4/6 325 family were identified as PPH-4.1 interactors, namely PPFR-4 and ZK688.9, 326 orthologous to mammalian ⍺4 and TIPRL, respectively. One other notable potential 327 PPH-4.1 interactor is PPH-6, which encodes the C. elegans orthologue of PP6c, the 328 catalytic subunit of the related but distinct PP6 complex. When we compared the list of 329 potential PPH-4.1 interactors with the list of proteins predicted by Wormbase to interact 330 with SMK-1, we expected to find significant overlap, especially among orthologues of 331 the PP4 complex. Instead, although PPH-4.1 was listed as an interactor of SMK-1 the 332 only other potential PP4 complex member that was in common to the list of interactors 333 for both PPH-4.1 and SMK-1 was the regulatory subunit PPFR-1 (Table S3).

334
To gain further insight into the possible relationships between SMK-1, PPH-4.1, 335 and their potential binding partners, we performed interaction network analysis using 336 Genemania. All homologues of the PP2A/4/6 family earmarked as possible interactors 337 of PPH-4.1 or SMK-1 on Wormbase along with the modulator PPFR-4 were included.

338
This analysis yielded additional putative PPH-4.1 interactors ( Fig S2) n/a n/a PP6-ARS-A/PPP1R65 n/a n/a ANKRD44 PP6-ARS-B n/a n/a ANKRD52 PP6-ARS-C n/a n/a fluorescence microscopy (Fig. 1A). Compared to worms at the L4 larval stage, the 423 expression of GFP driven by the promoter of lys-7 in the intestines was substantially 424 higher in Day 6 adults, consistent with an age-dependent increase in the transcriptional activity of DAF-16. In examining a population of animals we were able to assign 426 individuals to one of three categories based on their GFP expression pattern (Fig. 1A).

427
In general, we observed robust and uniform GFP expression along the length of the 428 intestine among the majority of individuals in a synchronized population of Day 6 adults.

429
These animals were scored as "high" GFP expressers. However, in a proportion of   RNAi.
Continuing to examine the functional parallels between the function of SMK-1 471 under low IIS conditions and during normal aging, we next asked whether SMK-1 is 472 required to confer resistance to environmental stress in adult C. elegans. Consistent 473 with the Plys7::gfp results, knocking down either smk-1 or daf-16 had no effect on the 474 ability of L4 animals to resist bacterial infection, UV irradiation, or thermal stress ( Fig.   475 1C, E, G; Table 2). On the other hand, both smk-1 and daf-16 were necessary for Day 6 476 adults to withstand challenge with either P. aeruginosa infection (Fig. 1D) or with UV 477 irradiation ( Fig. 1F), as knockdown of either gene reduced the median survival of 478 animals exposed to these insults (   Members of the PP2A/4/6 subfamily contribute to innate immunity during aging 542 We found that smk-1 is necessary to confer resistance to bacterial infection in 543 Day 6 adults, as is DAF-16 ( Fig 1D). This result leads to the prediction that pph-4.1, 544 encoding the catalytic subunit of the PP4 complex, is also required for host defense 545 during aging. Further, based on the results of our in vivo reporter assay for DAF-16 546 transcriptional activity, we wondered whether PP2A subunits may also function in innate 547 immunity later in life. We addressed these possibilities as part of our characterization of 548 the PP2A/4/6 family in adults challenged with P. aeruginosa. In a phenotype reminiscent 549 of the smk-1 knockdown, RNAi targeting homologues of the PP4 catalytic subunit pph-550 4.1 or its paralog pph-4.2 reduced the ability of Day 6 adults to resist bacterial infection 551 but had no effect on the susceptibility of L4 larvae to pathogen (Fig. 3A   Along with pph-4.1/4.2 and smk-1, RNAi inhibition of another PP4 regulatory 581 subunit homologue, ppfr-2 (PP4R2 in humans), also reduced the median lifespan of 582 worms infected at Day 6 of adulthood but had no effect on the ability of L4 larvae to 583 resist infection (Fig. 3A, B, G, H). No other PP4 regulatory subunits were found to be 584 required for innate immunity in Day 6 adults ( Fig. 3H; Fig S3). These results imply that 585 PPFR-2 may act in conjunction with SMK-1 and PPH-4.1 to protect adult C. elegans 586 from bacterial pathogens.
We found genes encoding putative subunits of other protein phosphatase 588 complexes to also be important for host defense in adult C. elegans. In addition to the 589 effect of inhibiting the expression of PP4 subunits, one of the ten RNAi treatments 590 directed against C. elegans orthologs of PP2A subunits enhanced the susceptibility of 591 adult animals to P. aeruginosa infection. Knockdown of sur-6, the sole worm homolog of 592 human PP2A B/B55 regulatory subunits, significantly reduced the median lifespan of 593 infected Day 6 adults but had no effect on larvae (Fig. 3C, D, G, H; Table 2). We found 594 no evidence of a role for other orthologs of PP2A complex subunits in adult or larval 595 innate immunity ( Fig. 3H; Fig. S4). In an unexpected result, inhibiting the expression of 596 two orthologues of PP6 complex members pph-6, the catalytic subunit, and saps-1, a 597 regulatory subunit, accelerated the rate of death from infection when worms were 598 challenged with P. aeruginosa at Day 6 of adulthood but not at the L4 larval stage (Fig.   599 3E, F, G, H). This was surprising because none of the PP6 members appear to be 600 required for the increased expression of plys7::gfp in adults, a departure from smk-1-like 601 phenotypes (Fig. 2B). Taken together, our results indicate that the entire PP2A/4/6 602 family plays a role in innate immunity during aging and that the PP6 complex may do so 603 without influencing the transcriptional output of DAF-16. 604 SMK-1 and the PP2A complex confer UV resistance in postreproductive C. elegans 605 We found SMK-1 to be important not only for host defense in Day 6 animals, but 606 also for resistance to ultraviolet irradiation, consistent with its role in daf-2(e1370) 607 mutants ( Fig 1F; Table 2) (43,60). We therefore expected that other components of the 608 PP4 complex would be required for survival upon exposure to UV light during aging.

609
This was not the case. No canonical members of the PP4 holoenzyme, including the catalytic subunits, seem to contribute to UV tolerance in C. elegans as knocking them 611 down did not affect the survival of irradiated L4 larvae or Day 6 adults (Fig. 4A 613 the scaffolding protein paa-1 and the regulatory subunit pptr-1 caused Day 6 adult 614 animals exposed to UV light to die more rapidly, leading to significant reductions in their 615 median lifespans (Fig. 4 C, D, G, H; Table 2). Other orthologs of PP2A subunits do not 616 appear to play a role in conferring resistance to UV light (Fig. S6). Targeting PP6 617 complex subunits by RNAi was also inconsequential to the ability of adults to resist UV  The PP2A regulatory subunit SUR-6 promotes resistance to heat stress in adult worms 642 Despite its roles in conferring resistance to bacterial pathogens and ultraviolet 643 irradiation, SMK-1 does not contribute to thermotolerance in adult C. elegans even 644 though DAF-16 does (Fig. 1G, H). In like manner, we found no evidence to suggest that  (Fig. 5G, H; Fig. S8). On its own this result suggests that in contrast to  adult C. elegans maintained at elevated temperature (Fig 5E, F, G, H). Knockdown of orthologs of the catalytic, regulatory, or scaffolding subunits of PP2A did not affect the 680 survival of adults challenged by thermal stress (Fig. 5H; Fig. S9). Although inhibiting 681 pph-6 expression increased the resistance of L4 worms the thermal stress (Fig. 5C), 682 this effect did not persist into adulthood (Fig. 5D). Upon exposure to high temperature, 683 the survival of Day 6 animals treated with RNAi targeting pph-6 or saps-1 was 684 comparable to the control group, (Fig. 5H; Fig. S10). Our do not implicate a complete  Table S2). We therefore asked whether they might modulate the PP2A/4/6 family during aging in C. elegans. Since our 703 data indicate that the PP2A/4/6 family is important for resistance to environmental  Fig. S12). We found no evidence of 712 a role for Y71H2AM.20 in protecting animals from thermal stress ( Fig. 6G; Fig. S13).
713 Surprisingly, instead of making adult animals more resistant, when ppfr-4 was targeted 714 by RNAi, Day 6 C. elegans were more sensitive to UV irradiation, yet the effect on the 715 relative median survival of these animals fell just short of the significance threshold (p= 716 0.0548) (Fig. 6A, B, E; Table 2). The only resistance phenotype that we found to be 717 associated with ppfr-4 knockdown was when L4 animals were challenged with P. 718 aeruginosa but not when adults were infected (Fig. 6 C, D, F). Our observations suggest 719 that there may be some differences between the roles of this class of PP2A/4/6 720 regulators in C. elegans and mammals, but the lack of statistical significance in our data 721 prevents us from drawing definitive conclusions.  The FoxO transcription factor DAF-16 is a major longevity determinant in C. 744 elegans. Animals with constitutively active DAF-16 live longer and are more youthful 745 than their wildtype counterparts, suggesting that DAF-16 contributes not only to lifespan 746 but also to healthspan (62). Accordingly, one category of DAF-16 transcriptional targets 747 are genes that confer resistance to stress, thus shielding animals from long-lasting cellular damage (7). While exposure to environmental insults induces DAF-16 to 749 upregulate these genes, recent evidence suggests that DAF-16 becomes activated in 750 age-dependent manner in animals that have not been challenged with a stressor (8,9).

751
Here we investigated the mechanism by which DAF-16 is activated during aging in C.    members that are all part of a second PP complex. In particular, we found Day 6 adults 910 to be susceptible to ultraviolet irradiation upon treatment with RNAi targeting not only 911 the PP2A complex members let-92, pptr-1, and paa-1 but also the PP4 subunit smk-1. 912 Similarly, sur-6 was the only PP2A subunit whose knockdown resulted in enhanced 913 susceptibility to bacterial infection, as did RNAi targeting the PP4 complex members 914 pph- 4.1/4.2, smk-1, and ppfr-2. In each case a regulatory subunit seems to be operating 915 separately from its cognate catalytic subunit. These results suggest that there may be 916 alternative forms of the PP2A and PP4 complexes that include at least one regulatory 917 subunit of the family that is not typically associated with that complex's eponymous 918 catalytic subunit.

919
In some cases we found that a gene's requirement for the age-dependent 920 increase in plys-7::GFP expression was apparently uncoupled from any role that it may In the course of characterizing the entire PP2A/4/6 subfamily in our studies, we 954 uncovered a new role for the PP6 complex in innate immunity during aging in C. 955 elegans. Prior to our work on PPH-6, the C. elegans ortholog of PP6c, it had only been    highlighted by coloring the names of those tissues (red, green, and blue, respectively) in 1654 the "Anatomical sites of expression" column.