Effects of HSP70 chaperones Ssa1 and Ssa2 on Ste5 scaffold and the mating mitogen-activated protein kinase (MAPK) Pathway in Saccharomyces cerevisiae

Ste5 is a prototype of scaffold proteins that regulate activation of mitogen-activated protein kinase (MAPK) cascades in all eukaryotes. Ste5 associates with many proteins including Gβγ (Ste4), Ste11 MAPKKK, Ste7 MAPKK, Fus3 and Kss1 MAPKs, Bem1, Cdc24. Here we show that Ste5 also associates with heat shock protein 70 chaperone (Hsp70) Ssa1 and that Ssa1 and its ortholog Ssa2 are together important for Ste5 function and efficient mating responses. The majority of purified overexpressed Ste5 associates with Ssa1. Loss of Ssa1 and Ssa2 has deleterious effects on Ste5 abundance, integrity, and localization particularly when Ste5 is expressed at native levels. The status of Ssa1 and Ssa2 influences Ste5 electrophoresis mobility and formation of high molecular weight species thought to be phosphorylated, ubiquitinylated and aggregated and lower molecular weight fragments. A Ste5 VWA domain mutant with greater propensity to form punctate foci has reduced predicted propensity to bind Ssa1 near the mutation sites and forms more punctate foci when Ssa1 Is overexpressed, supporting a dynamic protein quality control relationship between Ste5 and Ssa1. Loss of Ssa1 and Ssa2 reduces activation of Fus3 and Kss1 MAPKs and FUS1 gene expression and impairs mating shmoo morphogenesis. Surprisingly, ssa1, ssa2, ssa3 and ssa4 single, double and triple mutants can still mate, suggesting compensatory mechanisms exist for folding. Additional analysis suggests Ssa1 is the major Hsp70 chaperone for the mating and invasive growth pathways and reveals several chaperone-network proteins required for mating morphogenesis.

the swinging bucket Sorvall RT 6000B centrifuge at 5,000 rpm (3,000 x g) for 10 min at 4 o C. 296 Supernatants were transferred equally to two chilled eppendorf tubes, saving one as a control for 297 total input. The other tube of sample was centrifuged in a refrigerated Eppendorf 5415C centrifuge 298 at 14,000 rpm (16,000 x g) for 15 minutes and then the supernatant was transferred to a new 299 tube. An equal volume of lysis buffer was added to each pellet. Protein concentration was 300 determined and 5x loading buffer added to samples for running on an SDS polyacrylamide gel.  ascospores were recovered and switched to MATa with a pGAL-HO plasmid, then ura3-clones 317 that had lost the plasmid were used for analysis (EYL1798). Strains were transformed with 318 pYEE121 (FUS3-HA5-CEN-URA3) and pYEE128.30-1 (fus3K42R-HA5-CEN-URA3) for kinase 319 assays. Approximately 150 mls of cells were grown to A600 of 0.7-0.8 and processed exactly as 320 described using an extraction buffer that contained 250 mM NaCl. Transformations and growth 321 were done at room temperature.

322
Immunofluorescence microscopy 323 Indirect immunofluorescence and live cell imaging were performed exactly as described

339
Ste5 co-purifies with Ssa1 340 We purified proteins that co-immunoprecipitate (co-IP) with over-expressed Ste5-Myc9 341 from logarithmically dividing cells grown at 30 o C before and after a factor treatment. We used low 342 NaCl extraction conditions and the 9E10 monoclonal antibody to the Myc epitope. A number of 343 proteins co-IP'd specifically with Ste5-Myc9, but not with untagged Ste5. Figure 1B  indicates six bands that are most prominent: Ste5-Myc9 (~115 kDa), band "a" also labeled as 345 ~70 kDa, bands "b", "c" (which are ~68-58 kDa) and IgG2a (~50 kDa )(Fig1B). The signal 346 transduction partners of Ste5 that are closest in size to bands "a", "b" and "c" are Ste7 (57,723.7 347 Da) and Ste11 (80,718.8 Da) whereas Fus3 (40,770.5 Da) and Kss1 (42,680.6 Da) may be 348 obscured by the wide IgG2a band. Ste5-Myc9 is largely full-length and the +a factor sample 349 migrates more slowly (Fig 1B, lane 3), presumably from hyper-phosphorylation (38). Thus, the 350 preparation of Ste5-Myc9 was intact and the a factor treatment induced a robust pheromone 351 response. Band "a" is present in an amount similar to that of Ste5-Myc9, both in absence and

358
The two other bands, "b," "c," are present in sub-stoichiometric amounts compared to  Focus was placed on band "a" due to its high abundance. A large-scale IP of Ste5-Myc9 was 360 performed, the ~70 kDa band "a" was excised from the ge, proteolyzed with trypsin, and 361 microcapillary HPLC-ion trap tandem mass spectrometry was done on two trypsin-generated 362 peptides. The tandem mass spectra of two major HPLC peaks identified residues 170-185 363 (IINEPTAAAIAYGLDK) of Hsp70 isoforms Ssa1, Ssa2, Ssa3 and Ssa4 (Fig 1C in the co-IP when more Ste5-Myc9 is in the whole cell lysate (Fig 2A,B compare lanes 2,3), and 375 the correct migration position of Ssa1-GFP for its predicted size which is distinct from GFP-

376
Cdc24, a positive control (Fig 2B, lane 4). HA3-Ste5 also co-immunoprecipitated with Ssa1-GFP Ile-Glu(But)-Ala-Leu-H (aldehyde)) during whole cell lysate preparation. In a native gel, a broad 386 diffuse banding pattern is detected for full-length Ste5-Myc9 that shifts to higher apparent 387 molecular weight with a factor treatment in a manner that requires signaling through the mating 388 MAPK cascade is blocked by a ste11D mutation (Fig 2F, 38). Ste5-Myc9 also accumulates high 389 molecular weight species that are likely ubiquitinylated, because they enrich in presence of 390 proteasome inhibitor ( Fig 2F) and disappear when Cdc28 CDK is mutated (Fig 2G).

401
Overexpression of Ste5-Myc9 also leads to accumulation of proteolyzed fragments whose 402 abundance declines in a cdc28-4 ts mutant, presumably from loss of the ubiquitinylation-mediated 403 degradation ( Fig 2G, lanes 6-9). 404 We determined the effect of a ssa1D ssa2D double mutation on the abundance and gel   (Table S1). Ste5-Myc9 was expressed from a CEN plasmid in selective 433 medium containing 2% dextrose in one of the MATa ssa1D ssa2D ascospore strains (the whole 434 cell lysates were prepared with Z-Ile-Glu(But)-Ala-Leu-H (aldehyde) as in Fig 2F). In the wild type  Remarkably, when expressed at native levels, Ste5-Myc9 was nearly undetectable at 30 o C, 37 o C, 438 and 42 o C in ssa1D ssa2D strains compared to in wild type ( Fig. 3D Loss of Ssa1 and Ssa2 alters the integrity of Ste5 455 We have previously shown the Ste5 exists in a high molecular weight complex that 456 sediments with kinases in a glycerol gradient and is likely also associated with cytoskeletal 457 proteins. We examined the distribution of Ste5 in supernatant and pellet fractions of our whole 458 cell lysate preparations from WT SSA1 SSA2 (EY3407) and ssa1D ssa2D (EY3409) strains, using 459 an initial brief 3,000 x g centrifugation followed by a 16,000 x g centrifugation that is known to   (Table S3). Thus, Ssa1 and Ssa2 are required for 546 efficient recruitment of Ste5 to the cell periphery even when the level of Ste5 is increased or when 547 the Ste5-Myc9 derivative has enhanced ability to be recruited. Collectively, these findings suggest 548 the integrity of Ste5 is not optimal for being recruited to the plasma membrane.

550
Ssa2-GFP is thought to localize in the cytoplasm and the nucleus whereas Ssa1-GFP is 551 thought to localize only in the cytoplasm (94). We found Ssa1-GFP to localize throughout the 552 cytoplasm and nucleus as shown by absence of nuclear exclusion ( Fig 6B). Moreover, Ssa1-GFP 553 was clearly enriched in the nucleus in addition to being in the cytoplasm in the ssa1D ssa2D strain 554 that lacks endogenous Ssa1 and Ssa2 ( Fig 6B). Therefore, Ssa1 is likely to have substrates in 555 the cytoplasm and the nucleus. compared to wild-type before and during a factor treatment at room temperature (Table 1).

562
Ste5(1-242)-GFP2 which has the NLS and RING-H2 also accumulated in fewer nuclei in ssa1D 563 ssa2D cells (Table 1) whereas TAgNLS-GFP2 and TAgNLS-NES-GFP2 which have an additional 564 NLS from SV40 TAg efficiently accumulated in nuclei of ssa1D ssa2D cells (Table S3). These 565 results imply that Ssa1 and Ssa2 may stimulate nuclear accumulation or nuclear retention of Ste5 566 and may do so at least in part through the first 242 residues of Ste5.  (Table 2), suggesting either that nuclear import 572 was blocked or the nuclear pool was degraded or exported (Table 2).

573
In some S288c and W303a wild type strain backgrounds, elevated temperatures did    The ssa1D ssa2D ssa4 triple and ssa1D ssa2D ssa3 ssa4 quadruple mutants were inviable and 653 died as unbudded haplospores as previously found. The ssa1 and ssa2 single mutants gave rise 654 to colonies that were 1.96-fold and 2.75-fold larger, respectively, than the ssa1D ssa2D double Stronger mating pathway activation is needed for shmoo formation than for G1 arrest (84).

662
The morphologies of ssa1D ssa2D cells in vegetative growth were similar to wild-type but the  Table S5). The ssa1D ssa2D cells had a weaker morphological response to a factor than 666 wild type SSA1D SSA2D cells. Although the ssa1D ssa2D double mutant cells could arrest in G1 667 23 phase, fewer cells formed shmoos compared to wild type (Fig 7B-G; FigS7). Most noteworthy was 668 that many more ssa1D ssa2D cells were round enlarged rather than the classic pear shape with 669 tapered projection. The cells that had some shmoo morphology had less emerged shorter 670 projections that were broader (e.g. Fig 7C, Table S9, 29.2% round unbudded ssa1D ssa2D cells 671 versus 6.8% SSA1D SSA2D cells. In Fig 7B,  directions rather than towards a single polarization site. 676 We also looked at cells solely dependent on Ssa1 for survival for their ability to arrest in 677 G1 phase and form shmoos. Ssa1 was overexpressed in a ssa1D ssa2D ssa4 strain with a 678 GAL1prom-SSA1-CEN (EYL342) by growth in medium containing 2% galactose, then Ssa1 was 679 depleted through glucose repression and then cells were exposed to a factor or mock buffer and mutant to ~24% of wild type at all concentrations of a factor tested (Fig 9A-B). An ssa1 ssa3 ssa4 696 triple mutant solely dependent on Ssa2 had 3.4% of wild type basal FUS1::ubiYlacZ activity and, 697 after a factor stimulation, had ~45-49% wild type FUS1::ubiYlacZ activity (Fig 9B; lack of viability 24 prevented analysis of ssa2 ssa3 ssa4 triple and a ssa1D ssa2D ssa3 ssa4 quadruple mutants).

699
Therefore, Ssa1 and Ssa2 are equivalently required for full expression of FUS1:UbiYlacZ, with 700 little contribution from Ssa3 and Ssa4.

701
To assess ability to cell cycle arrest in presence of a factor, we performed halo assays.

702
The ssa1D ssa2D strain was also less efficient at being inhibited by a factor in halo assays and 703 formed smaller more turbid halos of growth inhibition compared to wild type (Fig 9C-D).

704
Collectively, the reduced shmoo response, FUS1::ubiYlacZ activity and G1 arrest reveal that 705 mating pathway activation is impaired in the ssa1D ssa2D double mutant. 706 We assessed the ability of a factor activated Fus3 to phosphorylate associated substrates vegetative growth and at all time points after addition of 50nM a factor (Fig 10A, Table S2). The 722 decrease in the band signal was not from less protein loaded based on ribosomal protein Tcm1 723 and Ponceau S staining (data not shown). Analysis of more ssa1D ssa2D cell extract did not 724 reveal basal Fus3-P, whereas basal Kss1-P was detected in both wild type and ssa1D ssa2D 725 whole cell extracts (Fig 10C). Densitometry of these representative immunoblots confirmed that 726 the ssa1D ssa2D strain had undetectable basal levels of active Fus3 and less active Fus3 during 727 a factor stimulation compared to wild type (Table S2). Kss1 activation during a factor stimulation 728 was reduced, although less impaired than that of Fus3 in the ssa1D ssa2D strain (Fig10B-C), but 729 the abundance of Kss1 was not reduced (Fig 10E, Table S2). obviously reduced in the ssa1D ssa2D double mutant (Fig 4C, Fig 10 B,D, TableS2) even when 746 Ste5 was overexpressed (Table S2). Fus3 abundance appeared to be approximately the same  Table S2). By contrast, Ste7-Myc abundance was not obviously altered in the ssa1D ssa2D 26 mutant, (Fig 10F, Table S2). The relative abundance of phosphorylated and unphosphorylated 763 species of Ste7-Myc was the same in wild type and ssa1D ssa2D strains (Fig 10E,  analysis.

770
The ssa1D ssa2D double mutant can mate 771 We compared the mating of wild-type, ssa1, ssa2, ssa3, ssa4 single, double and triple 772 mutants in a standard plate assay in which patches of cells are mated to a lawn of wild-type cells.
773 Surprisingly, there was no obvious decrease in mating ability for any of the mutants compared to 774 wild-type at room temperature or 30 o C in BAR1 or bar1 backgrounds (Fig 8D, Table S4). Given  components in the mating and invasive growth pathways (50%, 57% respectively) ( Table S6).

806
None of the other regulators we examined are predicted to interact with a significant number of 807 proteins in the mating and invasive growth pathways (Table S6).  Table   814 S6,S7, Fig S6). Mutant ydj1 cells (101) did not efficiently arrest in G1 phase and were often 815 severely misshapen (both unbudded and budded cells) and formed fewer shmoos that were often 816 enlarged and aberrant, with projections that were longer, wider, curved or bent (Fig 7I, FigS6E-F, 817 Table S6, Tables S6, S8), adding to known defects in Axl1 biogenesis and a factor processing 818 (102). The Sse1 Hsp110 protein (40, 103) is predicted to interact with many mating and invasive 819 growth proteins, whereas its paralog Sse2 is predicted to interact with few (Tables S6). 820 Surprisingly, the sse1 mutant had a nearly normal pheromone response for G1 arrest and shmoo 821 morphogenesis, whereas the sse2 mutant failed to form shmoos although it could undergo G1 822 arrest and cell enlargement ( Fig S7, Table S8). Perhaps its paralog Sse2 compensates in the 823 sse1 mutant but not vice versa. Mutation of Ssz1 Hsp70 that modulates Zuo1, a ribosome-824 associated J protein, caused defects in both G1 arrest and shmoo formation ( Fig S7, Table S8).
Therefore, Ssz1 may provide unique functions for mating morphogenesis although it is not 826 predicted to interact with many mating pathway proteins (Table S6). Surprisingly, the ssb1 mutant 827 underwent nearly normal G1 arrest and shmoo formation with some shmoos longer than wild-828 type ( Fig 7K; Table S8), contrasting its predicted interactions with many mating pathway proteins 829 (Table S6). Perhaps its paralog Ssb2 compensates for loss of Ssb1. Interestingly, the sti1 830 mutation in the Hsp90 cochaperone (104) increased both the number of shmoos and the length 831 of shmoo projections compared to wild type (Fig 7G, Fig S7, Table S8), indicating either enhanced 832 shmoo formation or reduced downregulation of shmoo formation. Currently, Sti1 has no known 833 interactions with mating proteins (Table S6). In summary, Fes1, Ydj1, Sse2, and Ssz1 are needed derivatives accumulated at the cell cortex (Fig 6, TableS3), suggesting the capacity of dimers of 857 Ste5 to bind to anchors at the plasma membrane is impaired. Less Ste5 accumulated in nuclei 858 including Ste5-Myc9, Ste5(1-242)-GFP2, which has the bipartite NLS (Table 1), GFP-Ste5 and 859 TAgNLSK128T-Ste5-Myc9 (Table S3). We infer from these findings that Ssa1 and Ssa2 repeatedly found that Ssa1 and Ssa2 promoted overexpressed Ste5 to accumulate as high 884 molecular weight species that are known to be ubiquitinylated ( Fig. 2; 37). Thus, we do not rule appear to be in the cytoplasm and near the nucleus (Fig 6, FigS2,   , ssa1D ssa2D, ssa3 ssa4, ssa1 ssa3 ssa4, ssa2 ssa3 ssa4, ssa1D 1116 ssa2D ssa3. G RT, 6 hour mating for WT, ssa2 ssa2.   Table 1. Localization of Ste5 to cell periphery and nucleus in wild type and ssa1D ssa2D 1142 cells.      Table   1662 S5, Table S6 Table S7 and Table S8 for tallies of morphology.  Table S8      F G A region of homology for active ERK antibodies (anti-phospho-p42 p44 Ab) 23 +/-7.0 __________________________________________________________________________ 1 Strains: SSA1 SSA2 is EY3376 is EY3136 ura3-+ (pSKM19 STE5-MYC9 URA3 2µ), ssa1 ssa2 is EY3379 is EY3141 ura3-ssa1::HIS3 ssa2::LEU2 + (pSKM19 STE5-MYC9 URA3 2µ). SSA1 SSA2 FY23 + (pSKM19 STE5-MYC9 URA3 2µ), EYL2712 YMY467 (1,2,3) is EYL457 = YMY467 = [RAY914 MATa cdc24-1::LoxPHIS5SpLoxP his3-200 leu3-3, 112 ura3-52 trp1-901 lys2-801 gal2 suc2-9 (pRS414 TRP1-CEN-CDC24)] + (pSKM10 STE5-MYC9 URA3 2µ. Cells were grown in SC-uracil with 2% dextrose to mid logarithmic phase (A600 0.4-0.6) then equal A600 units of cells were pelleted and resuspended in pre-warmed YEPD medium and incubated with shaking at the indicated temperature. Cells were collected and prepared for indirect immunofluorescence using 9E10 monoclonal antibody and DAPI. 2 N>C means the 9E10 signal in the nucleus is stronger than in the cytoplasm. ~133-400 cells were counted for data points. The Ste5-Myc9 signal was weaker in the ssa1 ssa2 strains than in the wild type strains. 3 At 55 o C (for 1-3 hours) cells of this strain background appear enlarged with a large vacuole devoid of Ste5-Myc9 signal and some Ste5-Myc9 signal beneath cell periphery. The cells appeared to be dying. 4 A comparison of FUS1prom-lacZ b galactosidase levels in S288c (FY23) grown at room temperature and 37 o C was also done. Cells grown logarithmically at room temperature then exposed to + aF for 2 hours at room temperature had 135.6 +/-1.0 S.E. FUS1-lacZ units whereas the same strain grown at 37 o C for 3 hours then exposure to aF for 2 hours at 37 o C had 122.5+/-9.0 S.E. FUS1-lacZ units.  ________________ _____________________________________________________________ 1 Cells were grown in SC selective medium containing 2% dextrose, either lacking uracil, tryptophan and /or leucine at 30 o C or at the indicated temperature. Cells were grown to logarithmic phase at an A 600 of 0.2-0.5 adjusted for equal cell density, ~A 600 0.5 then treated with a factor. Where indicated cells were transferred to pre-warmed medium at 37 o C and incubated for 3 hours. For the GAL1 promoter, cells were pre-grown in pre-warmed medium containing 2% raffinose prior to being transferred to pre-warmed medium containing 2% galactose and then grown for 5 hours. 2 Cells were fixed with 10% formaldehyde and processed for indirect immunofluorescence to detect Myc tagged proteins and DAPI stained DNA or visualized live for GFP tagged proteins as described (S3). Most analysis used captured images from published and unpublished work. 3 Nuclear accumulation means that the 9E10 signal in the nucleus is equal to or greater than the 9E10 signal in the cytoplasm. Nuclear exclusion means that the 9E10 signal is less than the 9E10 signal in the cytoplasm leaving a visible "hole" of no or little signal. 4 The cortical accumulation in all cases was asymmetric and if a shmoo was present, it was located at the shmoo tip. 5 Ste5(1-242)-GFP2 was visualized live in cells grown at RT, 37 o C and by indirect immunefluorescence on fixed cells that were grown at 30 o C. The GFP + cytoplasmic pool is more prominent than the nuclear pool in fixed cells than in live cells. 6 nsp1ts strains were grown in medium containing 2% raffinose then shifted to medium containing 2% galactose for 3 hours at room temperature, then shifted to medium containing 2% dextrose for 1 hour then shifted to 37 o C for 3 hours or kept at room temperature.