Calcium-responsive transactivator (CREST) toxicity is rescued by loss of PBP1/ATXN2 function in a novel yeast proteinopathy model and in transgenic flies

Proteins associated with familial neurodegenerative disease often aggregate in patients’ neurons. Several such proteins, e.g. TDP-43, aggregate and are toxic when expressed in yeast. Deletion of the ATXN2 ortholog, PBP1, reduces yeast TDP-43 toxicity, which led to identification of ATXN2 as an amyotrophic lateral sclerosis (ALS) risk factor and therapeutic target. Likewise, new yeast neurodegenerative disease models could facilitate identification of other risk factors and targets. Mutations in SS18L1, encoding the calcium-responsive transactivator (CREST) chromatin-remodeling protein, are associated with ALS. We show that CREST is toxic in yeast and forms nuclear and occasionally cytoplasmic foci that stain with Thioflavin-T, a dye indicative of amyloid-like protein. Like the yeast chromatin-remodeling factor SWI1, CREST inhibits silencing of FLO genes. Toxicity of CREST is enhanced by the [PIN+] prion and reduced by deletion of the HSP104 chaperone required for the propagation of many yeast prions. Likewise, deletion of PBP1 reduced CREST toxicity and aggregation. In accord with the yeast data, we show that the Drosophila ortholog of human ATXN2, dAtx2, is a potent enhancer of CREST toxicity. Downregulation of dAtx2 in flies overexpressing CREST in retinal ganglion cells was sufficient to largely rescue the severe degenerative phenotype induced by human CREST. Overexpression caused considerable co-localization of CREST and PBP1/ATXN2 in cytoplasmic foci in both yeast and mammalian cells. Thus, co-aggregation of CREST and PBP1/ATXN2 may serve as one of the mechanisms of PBP1/ATXN2-mediated toxicity. These results extend the spectrum of ALS associated proteins whose toxicity is regulated by PBP1/ATXN2, suggesting that therapies targeting ATXN2 may be effective for a wide range of neurodegenerative diseases.

in cultured neurons and causes retina degeneration in transgenic Drosophila [48,52]. Also, CREST co-immunoprecipitates with FUS in neuron lysates [48,52]. In cultured cells, transformed CREST forms predominantly nuclear dots with some cytoplasmic foci. Also, like other ALS associated proteins, CREST is recruited to stress-induced ribonucleoprotein particle (RNP) granules [52,53]. While many proteins must reversibly aggregate in physiological granules as part of their normal cellular function, this ability to aggregate could lead to the formation of pathological aggregates. Indeed, TDP-43 and FUS have prion-like domains and have been found in cytoplasmic aggregates in the neurons of patients with ALS. While no such aggregates have yet been reported for CREST, it has an unstructured prion-like domain predicted to form amyloid-like aggregates [48]. CREST in cell culture was slightly resistant to Triton-X100 but was solubilized by SDS; in addition it formed nuclear dots in a concentrationdependent manner [52].
Here we establish a new ALS model in yeast. We show that this human chromatin remodeling complex component and transcriptional activator is toxic and forms largely nuclear aggregates that stain with Thioflavin T when expressed in yeast. We also show that CREST toxicity is increased by the endogenous yeast [PIN + ] prion and reduced by deletion of the HSP104 chaperone that is required for propagation of many yeast prions [4]. Deletion of the TDP-43 toxicity modifier, PBP1, likewise reduces CREST toxicity and this finding was also confirmed in a CREST transgenic fly model. Both hsp104Δ and pbp1Δ alter CREST aggregation. Finally, overexpression of PBP1 causes a large portion of CREST to leave the nucleus and co-localize with PBP1 containing cytoplasmic dots in yeast. Likewise, in mammalian cells, cytoplasmic aggregates of CREST sequester ATXN2. Overall, our results establish PBP1/ATXN2 as a potent modifier of CREST toxicity in vivo.

Expression of human CREST in yeast is toxic
Growth of yeast transformed with control GAL-GFP vectors was compared with yeast expressing fusions of human CREST cDNA and GFP from a GAL promoter on three types of vectors: a Yeast CEN plasmid (YCp) containing a centromere and expressed as one or a few copies in each cell ( Fig 1A); a Yeast Episomal plasmid (YEp) containing elements of the endogenous yeast 2μ episome and expressed in high copy in some cells but absent in other cells ( Fig 1B); or a Yeast Integrating plasmid (YIp) that was integrated into the genome (Fig 1C). Growth inhibition is evident but slight when CREST-GFP was expressed from a CEN vector ( Fig 1A) and more inhibition is seen when episomal ( Fig 1B) or integrating vectors ( Fig 1C) were used. Expression of the GAL controlled genes was turned on with a constitutively expressed fusion of the human estrogen receptor hormone-binding domain, the yeast GAL4 DNA-binding domain and the VP16 viral transcriptional activator, which was activated by the addition of βestradiol [54]. The high level of β-estradiol used here (2 μM) in glucose media results in more expression from the GAL promoter than that obtained by traditional induction on 2% galactose without β-estradiol [55]. Note the reduced induction of CREST caused by galactose relative to β-estradiol results in reduced toxicity (S1 Fig).

CREST forms largely nuclear and occasionally cytoplasmic amyloid-like foci in yeast
Similar to mammalian cells [52], CREST-GFP expressed in yeast largely appeared as dots in nuclei, but smaller cytoplasmic foci were also visible and were more frequent after longer induction times (Fig 2A). Nuclear localization was confirmed by using DAPI stain or a

Fig 1. Expression of human CREST in yeast is toxic. Shown is the growth of cells carrying vectors encoding
GAL-CREST-GFP and a fusion protein that promotes induction of the GAL promoter in the presence of estradiol (pβest) spotted on plasmid selective glucose plates without estradiol where CREST-GFP expression is Non-induced, or with 2 μM β-estradiol where CREST-GFP expression is β-Estradiol Induced. Duplicates are independent transformants or integrants. Cells grown on plasmid selective glucose plates were diluted to OD 600 = 2 and then serially diluted 10-fold and spotted on the plates shown. The pβ-est plasmid expresses the fusion activator that is responsive to βestradiol, and YCpGAL-ccdB (pGAL C ) was used as its control. (A) CREST-GFP expressed from a CEN plasmid inhibits growth. YCpGAL-CREST-GFP (pCREST C ) and control YCp-GAL-GFP (pGFP C ) vectors were used. (B) CREST-GFP expressed from a 2μ plasmid inhibits growth. YEpGAL-CREST-GFP (pCREST E ) and the control YEpGAL-GFP (pGFP E ) vectors were used. (C) Integrated CREST-GFP is toxic. Strain with integrated YIpGAL-CREST-GFP (pCREST I ) and parent strain lacking the integrated plasmid but containing control YEpGAL-GFP (LEU2) was used. Strain 74D-694 [PIN + ] was used for the CEN and integrated vector experiments (A and C) while strain BY4741 [PIN + ] was used for the 2μ vector experiments (B). constitutively expressed nuclear HTB1-mCh marker (Fig 2B and S2 Fig) which co-localized with CREST-GFP. CREST foci also stained with Thioflavin T that marks amyloid-like protein ( Fig 2C). For controls we show TDP-43 and FUS aggregates that, respectively do not and do, stain with Thioflavin T [56]. In addition, the bulk of the CREST protein is found in the pellet of lysates while the non-aggregated proteins, GFP and PGK, remain in the supernatant (Fig 2D).
When CREST was overexpressed in mammalian cells, it formed large dot-like nuclear structures and was also present in cytoplasmic aggregates [52]. When CREST is expressed in yeast for 5-6 h, most of the protein is found in nuclear dots with just occasional cytoplasmic foci (see Figs 2 and 3 left). However, when cells with nuclear CREST are stressed by incubation at high temperature there is a dramatic increase in the appearance of cytoplasmic foci (Fig 3  middle) that partially co-localize with Ded1-mCh stress granule and EDC3-mCh P-body markers. Following a return to 30˚C, cytoplasmic CREST aggregates disappear along with stress granules and P-bodies (Fig 3 right).

The presence of the [PIN + ] prion or HSP104 enhances toxicity of CREST
Since the presence of the [PIN + ] prion causes enhanced toxicity of the human genes huntingtin/polyQ, TDP-43 and FUS [28,36,37,57], we asked if [PIN + ] would likewise increase toxicity of CREST and found that it did ( Fig 4A). However, CREST and [PIN + ] aggregates did not colocalize ( Fig 4B). Also, CREST aggregated in both [PIN + ] and [pin -] cells (Figs 2A and 4D). Interestingly, CREST toxicity was reduced even further when the HSP104 chaperone required for the propagation of many yeast prions [4,58] was deleted in [pin -] cells (Fig 4C left). Deletion of HSP104 also caused the appearance of many nuclei with diffuse CREST-GFP with or without dots instead of nuclear CREST-GFP dots on a black background, and reduced the number of cells showing any cytoplasmic CREST aggregation (Fig 4D). CREST-GFP was found largely in the pellet in lysates of both hsp104Δ and HSP104 yeast (Fig 4E). The higher level of CREST-GFP in the hsp104Δ cultures (Fig 4C right) could reflect the reduced toxicity of CREST.

Expression of human CREST activates yeast FLO genes and causes flocculation
While working with GAL-CREST-GFP and GAL-GFP transformants, we noticed that expression of CREST-GFP caused cells in liquid culture to settle very rapidly compared to the GFP controls. In a measured experiment, we found that overexpression of CREST caused flocculation (cell clumping), while overexpression of TDP-43 or FUS did not ( Fig 5A).
Since CREST increased yeast flocculence just as the non-prion form of the SWI1 chromatin remodeling protein does [59], we hypothesized that CREST, like SWI1, remodels yeast chromatin structure causing a release of silencing of telomerically located and epigenetically controlled [60] FLO genes, resulting in flocculation. To test this we examined the effect of CREST on the telomerically located FLO1 and the epigenetically controlled FLO11 promoters, using ura3Δ strains with either the FLO1 or FLO11 open reading frames (ORFs) replaced with the URA3 ORF reporter [59]. In the absence of CREST these strains were auxotrophic for uracil indicating that both the FLO1 and FLO11 promoters were off. However, upon CREST expression on galactose medium, both strains gained the ability to grow in the absence of uracil, indicating that the FLO1 and FLO11 promoters were indeed both activated by CREST (Fig 5B).  [8,17,55,61]. Here we detect conversion into [PSI + ], which causes readthrough of nonsense codons, by the appearance of adenine prototrophy despite the presence of the nonsense mutation ade1-14. We found that unlike [PIN + ], or overexpression of a variety of QN-rich yeast proteins [8], overexpression of CREST did not cause the appearance of fluorescent SUP35NM-GFP rings or adenine prototrophy. This was true even when PBP1 was also overexpressed (S3 Fig), which, as described below in the last research section, drives CREST-DsRED aggregates into the cytoplasm.

PBP1/dAtx2 enhances toxicity of CREST in yeast and flies
Overexpression of yeast's PBP1, a homolog of human ATXN2, enhances, while PBP1 deletion reduces, the toxicity of TDP-43 expressed in yeast [42]. This led to the discovery that ATXN2 intermediate-length polyglutamine expansions, which are associated with increased ATXN2 protein levels, are a risk factor for ALS [42]. Work in a variety of model organisms including mice, now suggests that ATXN2 is a modifier of some neurodegenerative diseases [43].
We thus tested the effect of PBP1 level on the toxicity of CREST in yeast. Overexpression of PBP1 itself was toxic in our system (Fig 6A) so the effect on CREST toxicity could not be determined. However, we clearly show that deletion of PBP1 reduces toxicity in either a [PIN + ] or [pin -] background ( Fig 6B) and this is associated with an increase in CREST-GFP solubility ( Fig 2D). Indeed, lysates of pbp1Δ cells expressing CREST-GFP had more GFP reactive protein in the supernatant vs. pellet fractions while the reverse was true for PBP1 lysates. Furthermore the CREST-GFP in pbp1Δ supernatant showed a lot of degradation compared with CREST-GFP in the pellet or in either supernatant or pellet fractions of the isogenic PBP1 strain (Fig 2D). This is likely because soluble CREST-GFP is more susceptible to degradation than aggregated CREST-GFP found in the pellet. We also saw a significant reduction in the number of cells with cytoplasmic CREST aggregates in the presence of pbp1Δ (Fig 6C). Taken together the data imply that a smaller fraction of total CREST-GFP is aggregated in either nuclei or cytoplasm in the absence of PBP1. Below is shown the % of cells containing CREST-GFP that contained one or more cytoplasmic dots. All of these cells also had nuclear dots. (B) Co-localization of wild-type CREST-GFP aggregates and DAPI nuclear stain. Cells in (a) induced for 6 h were stained with DAPI. (C) CREST aggregates stained with the Thioflavin T (ThT). Top row: BY4741 transformed YEpGAL-CREST-YFP (CREST-YFP) and pβ-est was induced for 6 h with 2 μM β-estradiol. Next three rows strain L3491 transformed with YEpGAL-CREST-YFP (CREST-YFP), YCpGAL-TDP-43-YFP (TDP-43-YFP) as negative control, or YCpGAL-FUS-YFP (FUS-YFP) as a positive control, were grown in plasmid selective SRGAL medium overnight. ThT fluorescence was detected with a CFP filter. CFP was recolored green and YFP recolored red to facilitate visualization of merged pictures. Cell boundaries are drawn. (D) CREST-GFP is found mostly in pellets in WT but mostly in supernatant of pbp1Δ strains. Three independent transformants of (WT) and (pbp1Δ) with pβ-est and either YEpGAL-CREST-GFP (CREST) or control YEpGAL-GFP (GFP) were grown on plasmid selective glucose medium, induced with 2 μM β-estradiol for 6 h and harvested. Total lysates were separated into Supernatant and Pellet by centrifugation. Samples were run on SDS-PAGE. Three gels were immunoblotted with anti-GFP, stripped and reprobed with anti-PGK for loading control. All gels showed a ratio of more GFP reactive material in pellets of WT vs. supernatant and the reverse for pbp1Δ. Since ratios are used loading differences are not relevant. https://doi.org/10.1371/journal.pgen.1008308.g002 In order to validate the observed effect of the ATXN2 ortholog PBP1 in an independent in vivo model, we used transgenic Drosophila which possesses an ortholog of the human ATXN2 gene, dAtx2. Overexpression of untagged wild-type CREST using GAL4-UAS system in transgenic lines results in severe retinal degeneration [52]. For our studies we used dAtx2 deletion fly line (dAtx2 06490) [62]. Homozygous deletion of dAtx2 was reported to cause disorganized and decreased photoreceptor neurons, but there was no effect in the heterozygous state [62]. Heat stress causes relocalization of some CREST aggregates from the nucleus to the cytoplasm. CREST partially co-localizes with the DED1 stress granule and EDC3 P-body markers during heat-stress and returns to the nucleus following heat-stress. YCpDED1-mCh and YCpEDC3-mCh transformants of [PIN + ] 74D-694 with integrated GAL-CREST-EGFP were grown in plasmid selective SRaf (2%) to OD 600 = 0.3 when CREST-EGFP was induced by addition of 2% galactose for 5 h. Cells were then stressed for 20 min at 46˚C. CREST-EGFP, DED-mCh and EDC3-mCh were examined before (30˚C), during (46˚C) and 3 h following (30˚C) heat stress.
https://doi.org/10.1371/journal.pgen.1008308.g003   YEpGAL-CREST-GFP (pCREST E ) or empty GFP control YEpGAL-GFP (pGFP E ) and all carrying pβ-est expressing the fusion activator responsive to β-estradiol, were spotted and grown on plasmid selective media with (β-est Induced), and without (Non-induced), βestradiol. (C) Deletion of PBP1 alters CREST aggregation. Transformants were grown for 2 d on medium with β-estradiol and were then examined. Counts were performed on 3 independent transformants each to determine the fraction of cells with CREST-GFP that had nuclei with dots on a black background, nuclei with dots on a diffuse background and one or more cytoplasmic dots.
We crossed CREST overexpressing (transgenic, tg) flies with heterozygous dAtx2 flies (dAtx2 06490-/+ ) and analyzed the double tg progeny. Histological analysis confirmed normal retinal morphology in dAtx2 06490-/+ flies, whereas complete disruption of the regularly ordered arrays of photoreceptor neurons was observed in the retina in CRESTwt tg flies, consistent with the previous report [52]. Strikingly, this severe retinal degenerative phenotype in CRESTwt tg flies was largely rescued by downregulation of dAtx2 (CRESTwt tg; dAtx2 06490-/+ line) (Fig 7A and 7B). This rescue was not caused by reducing the expression of CREST ( Fig  7C). Thus, human CREST toxicity in flies is largely mediated by endogenous dAtx2.

PBP1/ATXN2 and CREST co-localize in cytoplasmic foci in yeast and human cells
Finally, we asked whether potentiated toxicity of CREST in the presence of PBP1 might be linked to the aggregation-prone nature of the two proteins. We found a dramatic increase in cytoplasmic CREST dots when PBP1 was overexpressed in yeast ( Fig 8A). This was not due to an increase in CREST protein level ( Fig 8B). We also found the CREST-DsRed dots co-localized with PBP1-GFP in cytoplasmic foci ( Fig 8C). However these foci do not appear to be RNP granules because they generally lack DED1 and EDC3 (S4 Fig).
We next examined the localization of human ATXN2 in CREST-overexpressing human cells. CREST-GFP forms nuclear aggregates in cells expressing high levels of the protein, and in the~10% of cells with the highest levels of CREST-GFP, cytoplasmic aggregates also develop [52]. We found that cytoplasmic CREST-GFP aggregates sequester endogenous ATXN2 in human neuroblastoma cells under these conditions (Fig 8D middle), whereas in non-transfected (NT) cells and in cells lacking these aggregates, ATXN2 remains diffusely distributed (Fig 8D, top and bottom). Thus, CREST and PBP1/ATXN2 co-localize in the cytoplasm, which may serve as one of the mechanisms of PBP1/ATXN2-mediated toxicity.

Discussion
Our data show that human CREST expressed in yeast shares properties with other ALS-associated proteins and yeast prions: CREST, FUS and yeast prions form amyloid-like foci; CREST, TDP-43 and FUS toxicity and yeast prion appearance are enhanced by the yeast prion [PIN + ]; TDP-43 and CREST toxicity is reduced by deletion of PBP1; CREST, TDP-43 and FUS are associated with cytoplasmic RNP granules. Like maintenance of yeast prions, but unlike effects on FUS or TDP-43 [31,32], CREST toxicity and aggregation is dependent on the presence of HSP104 chaperone.
In mammals, endogenous CREST is a member of a chromatin remodeling complex and thereby activates transcription. When expressed in yeast, we found that CREST releases telomeric silencing and causes flocculation. Thus we propose that even in yeast the heterologous CREST protein remodels chromatin causing transcriptional activation of telomerically located genes including those that cause flocculation (e.g. FLO1, FLO5, FLO9 and FLO10) as well as the epigenetically controlled gene FLO11. Likewise, a yeast chromatin remodeling complex protein, SWI1, is required for the expression of telomerically located and epigenetically controlled FLO genes [63]. Interestingly, SWI1 can form infectious prion aggregates [12] reminiscent of the amyloid CREST aggregates seen here.
We first described [PIN + ], the prion form of the RNQ1 protein, because its presence allowed the efficient conversion of the SUP35 protein to its prion form, [PSI + ] [8,17,64]. It was later shown that [PIN + ] is also required for the efficient aggregation and toxicity of polyQ in yeast [28]. The mechanisms causing these effects are still unknown. Some data support the hypothesis that the [PIN + ] aggregates of RNQ1 cross-seed aggregation of heterologous proteins such as SUP35 and polyQ. Likewise some, but not all, evidence suggests that the [PIN + ] aggregates bind proteins such as chaperones that would otherwise inhibit aggregation of the heterologous SUP35 or polyQ protein, thereby enhancing their aggregation [18][19][20][21][22][23][24][25]27] We recently showed that toxicity of TDP-43 and FUS is slightly enhanced by the presence of [PIN + ] [37,57]. Both of these proteins contain Q/N-rich regions and co-aggregate with polyQ disease protein alleles of huntingtin [65,66]. Surprisingly, we did not detect a clear effect of [PIN + ] on TDP-43 or FUS aggregation. Likewise, CREST is a Q-rich protein. We show here that the toxicity of CREST is also slightly enhanced by [PIN + ], again with no clear effect on aggregation. Apparently the toxicity of the aggregates can be altered without a noticeable change in their appearance. On the other hand, deletion of either HSP104 or PBP1, which both cause dramatic rescues of CREST toxicity, were each accompanied by reduced frequency of CREST cytoplasmic foci, more diffuse CREST in nuclei and, for pbp1Δ, reduced CREST sedimentation.
Surprisingly, despite the fact that we clearly detect more diffuse CREST in the presence of hsp104Δ, there was no increase in CREST degradation or in the level of CREST in supernatant fractions as was seen for pbp1Δ. Also, immunoblots of pbp1Δ lysates reproducibly show that CREST-GFP in the supernatant is degraded much more in the pbp1Δ vs. WT PBP1 strain. CREST-GFP in the pellet fractions was stable in either PBP1 or pbp1Δ strains. This suggests that pbp1Δ causes more CREST to remain unaggregated and subject to degradation. Indeed there was more CREST-GFP plus GFP breakdown product in the supernatant vs. pellet in the pbp1Δ while the reverse was true for the PBP1 strain. While pbp1Δ also rescues toxicity of TDP-43, effects on TDP-43 aggregation have not been reported [42].
Since the HSP104 chaperone is required to shear the [PSI + ] prion aggregates thereby creating ends where growth occurs, inhibition of HSP104 initially causes [PSI + ] prion polymers to enlarge and, for newly made SUP35-GFP, to not efficiently join the aggregates remaining diffuse instead [67]. Possibly something similar is happening to CREST-GFP in the absence of HSP104. It is also possible that HSP104 is directly involved in CREST aggregate formation.
A recent finding that deletion of PBP1 causes a reduction in the level of RNQ1 protein [68] could have explained the reduced CREST, TDP-43 and FUS toxicity seen in pbp1Δ cells if the lowered RNQ1 level reduced the potency of [PIN + ]. However, we show here that this is not the case because pbp1Δ reduces CREST toxicity even in the absence of [PIN + ] (Fig 6E). It remains possible that toxicity is reduced by pbp1Δ by reducing RNQ1 levels regardless of [PIN + ] status.
We have shown that the presence of certain QN-rich aggregates can substitute for [PIN + ] and allow the efficient induction of [PSI + ] by overexpression of SUP35 [8]. This fact, plus the observation that RNQ1, the [PIN + ] prion protein, although not a homolog, is the yeast protein most similar to CREST, prompted us to ask if overexpression of CREST could substitute for [PIN + ]. However, CREST did not provide any [PIN + ] activity even when PBP1 was overexpressed driving CREST into the cytoplasm.
Prion aggregates enhance de novo formation of heterologous prion, presumably by a crossseeding mechanism. Curiously, however, if heterologous prion is already present, prion aggregates instead promote loss of the heterologous prion [20]. One mechanism for this has been shown to be prion aggregate binding to, and therefore titrating, chaperones needed to propagate the heterologous prion [24]. Likewise, CREST has been shown to suppress polyQ-mediated huntingtin aggregation and toxicity [69]. Many of the ALS associated proteins, e.g. TDP-43, FUS and TAF15 are soluble nuclear RNA/DNA binding proteins that regulate mRNA splicing and/or stability and contain prionlike unstructured domains and nuclear/cytoplasmic trafficking signals. In neurons of patients with ALS/FTD these proteins have been found in cytoplasmic aggregates instead of their normal nuclear location. This suggests that impaired nucleocytoplasmic transport may be a general mechanism for neurotoxicity [3,[70][71][72][73][74][75]. Indeed, impairing nuclear import of TDP-43 enhances neurotoxicity [76].
While TDP-43 and FUS quickly form cytoplasmic aggregates in yeast with little or no nuclear localization [31,32,37,57,[77][78][79], we found here that CREST appeared in nuclear aggregates with only occasional cytoplasmic foci. While the ALS-associated proteins TDP-43, FUS and SOD1 form aggregates in patients, model systems and in vitro, these aggregates are sometimes but not always classical amyloids. In yeast, aggregates of FUS, but not TDP-43, stain with Thioflavin T that is indicative of amyloid-like protein [56]. Although both nuclear and cytoplasmic CREST foci appear to stain with Thioflavin T, it is nonetheless tempting to speculate, by analogy with FUS and TDP-43, that the CREST cytoplasmic foci are more toxic than nuclear CREST. Indeed reduction of CREST-GFP toxicity caused by deletion of either PBP1 or HSP104 also reduced the frequency of cells with cytoplasmic CREST-GFP aggregates.
We found that heat stress induced CREST to leave the nucleus and partially co-localize with RNP granules. This is consistent with previous findings in mammalian cells that CREST becomes mislocalized to the cytoplasm under stressful conditions (oxidative stress) and becomes recruited to cytoplasmic stress granules [52]. Furthermore, as seen for classic stress granule proteins, CREST returned to the nucleus in yeast following the heat stress. Many proteins associated with neurodegenerative disease contain intrinsically disordered regions that are required for their appearance in RNP granules. Evidence has been accumulating that RNP granules are incubators for the formation of pathogenic protein aggregates associated with neurodegenerative disease. It has been proposed that localization of proteins such as TDP-43, FUS and C9ORF72-encoded dipeptide repeat GR50 in high concentration in stress granules likely in a gel form, promotes their conversion into pathogenic amyloid-like aggregates [80][81][82][83][84].
In support of this model, RNP-granule formation has been shown to be strongly associated with neurodegeneration. This was accomplished with the aid of ATXN2, a nucleic-acid-binding protein that is required for RNP-granule formation. Three ALS associated proteins, TDP-43, FUS and C9ORF72-encoded GR dipeptide, have been shown to co-localize with ATXN2 in RNP granules [42,80] in yeast, human or Drosophila S2 cells. Also, either reducing the level of ATXN2 or deleting one of its intrinsically disordered regions both reduced RNP-granule formation and reduced neurodegeneration caused by TDP-43, C9ORF72 dipeptides or FUS [80,[85][86][87]. These results help explain why expansions of the polyQ regions in ATXN2 that increase the stability, and thus likely the level of ATXN2, are associated with neurodegenerative disease (susceptibility to ALS and type 2 spinocerebellar ataxia).
Here, we report similar effects of PBP1/ATXN2 on CREST: deletion of PBP1 reduces toxicity of CREST in yeast and fly models of CREST proteinopathy; ATXN2 is recruited into cytoplasmic CREST aggregates in mammalian cells and overexpression of PBP1 causes CREST to co-localize with PBP1 in cytoplasmic foci in yeast. Overexpression of PBP1 has previously been shown to induce the formation of cytoplasmic PBP1 foci [88] that recruit TORC1 thereby down regulating TORC1 function [89]. PBP1 foci may similarly attract CREST which may promote CREST's conversion to a more toxic form. In sum, we showed that CREST is yet another protein, in addition to TDP-43, that is highly sensitive to ATXN2 dosage indicating that ATXN2 likely has a more general role as a modifier of neurotoxicity.
Our results extend the spectrum of ALS-associated proteins affected by ATXN2 to include CREST. This supports the hypothesis that therapies that target ATXN2 may be effective for a wide range of neurodegenerative diseases.

Yeast strains and plasmids
Yeast strains and plasmids used are listed in Tables 1 and 2. Unless otherwise stated, yeast strain BY4741 (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0) was used. Yeast transformation was by the lithium acetate method [90]. L3491 bearing an integrated copy of ADH1-HTB1-mCh was used to check CREST nuclear localization [37]. FLO1 (L3580) and FLO11 (L3582) promoter reporter strains were respectively made from DY755 and DY758. DY755 and DY758 are Ura + because FLO8, which is disrupted in these strains, is required for regulation that inhibits the FLO1 and FLO11 promoters [59,91]. We obtained pop-outs of the HIS3 disruptions of FLO8 by selecting for Urarevertants on FOA plates [92] and checking these for those that had become His -. Several such pop-outs were obtained and tested and they all behaved identically. Unless otherwise stated, all overexpression plasmids were driven by the GAL1 promoter. Gateway technology [93] was used to make CREST entry clone pDONR-CREST with BP reactions between pDONR221 (Invitrogen, Cat# 12536017) and a CREST fragment amplified from

Western blot analysis
CREST was detected in pre-cleared lysates separated by SDS-PAGE and blotted as described previously [96]. For analysis of proteins in total lysate vs. supernatant and pellet, the pellet from 90 μg of total lysate centrifuged at 50K rpm for 6 min was resuspended in water to the initial volume of total lysate centrifuged. The same sample volumes of total lysate, supernatant and pellet were then separated by SDS-PAGE. Blots were developed with GFP mouse antibody from Roche Applied Science (Indianapolis, IN) and PGK antibody from life technologies (Frederick, MD).

Visualization of aggregates, nuclei and co-localization studies
Fluorescently labeled protein aggregates, were visualized in cells with a Nikon Eclipse E600 fluorescent microscope (100X oil immersion) equipped with FITC, YFP and mCh (respectively, chroma 49011, 49003 and 49008) filter cubes. To visualize nuclei, cells with the integrated HTB1-mCh nuclear marker were used [37] or nuclei were stained with 1 μg/ml 4 0 ,6 0diamidino-2-phenylindole (DAPI) in 1XPBS for 10 min following fixing and permeabilizing cells for 1 h in 60% ethanol in PBS.

Flocculation assay
Flocculation was assayed by resuspending a sample of each 2 d culture grown in plasmid selective SRGal in 50 mM EDTA in water. Since flocculation requires the presence of Ca 2+ ions [97], EDTA removed clumping allowing cells' OD 600 to be accurately measured. Then another sample of cells that was not treated with EDTA was appropriately normalized to OD 600 = 10. Cells were photographed after they were allowed to settle for 15 min. The level of precipitation of cells and the reduction in the OD of the supernatants (diluted 10X) is a measure of the flocculence.

Thioflavin T Staining
Yeast cells were stained with Thioflavin T according to a protocol adapted from [31] with

Generation and characterization of transgenic flies
Generation of the human CREST line used in this study has been described earlier [52]. gmr-GAL4, UAS-lacZ and dAtx2 06490 lines were obtained from the Bloomington Drosophila stock center. Fly stocks were raised on standard Drosophila medium at 20˚C. Crosses between the Drosophila strains were carried out at 25˚C using standard procedures. For histochemical analyses, heads of 5-day-old adult transgenic flies were dissected, collected, briefly washed in phosphate buffered saline (PBS), and fixed with 4% paraformaldehyde containing 0.1% Triton X-100 at room temperature for 2 h. After a brief wash in PBS, tissues were dehydrated by graded ethanol, cleared in butanol and embedded in paraffin. Four-micrometer thick coronal sections were stained with hematoxylin and eosin (H&E). Between 8 and 10 flies were analyzed per genotype, and 3 sections were analyzed for each fly. For Western blot analysis, heads of 5-day-old flies were lysed in Laemmli sample buffer containing 2% SDS. Anti-CREST (Proteintech, 12439-1-AP) and anti-α-tubulin (Sigma, DM1A) antibodies were used.