Polyglutamine Toxicity Is Controlled by Prion
Composition and Gene Dosage in Yeast
He Gong1¤a, Nina V. Romanova1¤b, Kim D. Allen1¤c, Pavithra Chandramowlishwaran1, Kavita Gokhale1,
Gary P. Newnam1, Piotr Mieczkowski2, Michael Y. Sherman3, Yury O. Chernoff1*
1School of Biology, Georgia Institute of Technology, Atlanta, Georgia, United States of America, 2School of Medicine, The University of North Carolina at Chapel Hill,
Chapel Hill, North Carolina, United States of America, 3Department of Biochemistry, Boston University School of Medicine, Boston, Massachusetts, United States of
Polyglutamine expansion causes diseases in humans and other mammals. One example is Huntington’s disease. Fragments
of human huntingtin protein having an expanded polyglutamine stretch form aggregates and cause cytotoxicity in yeast
cells bearing endogenous QN-rich proteins in the aggregated (prion) form. Attachment of the proline(P)-rich region targets
polyglutamines to the large perinuclear deposit (aggresome). Aggresome formation ameliorates polyglutamine cytotoxicity
in cells containing only the prion form of Rnq1 protein. Here we show that expanded polyglutamines both with (poly-QP) or
without (poly-Q) a P-rich stretch remain toxic in the presence of the prion form of translation termination (release) factor
Sup35 (eRF3). A Sup35 derivative that lacks the QN-rich domain and is unable to be incorporated into aggregates
counteracts cytotoxicity, suggesting that toxicity is due to Sup35 sequestration. Increase in the levels of another release
factor, Sup45 (eRF1), due to either disomy by chromosome II containing the SUP45 gene or to introduction of the SUP45-
bearing plasmid counteracts poly-Q or poly-QP toxicity in the presence of the Sup35 prion. Protein analysis confirms that
polyglutamines alter aggregation patterns of Sup35 and promote aggregation of Sup45, while excess Sup45 counteracts
these effects. Our data show that one and the same mode of polyglutamine aggregation could be cytoprotective or
cytotoxic, depending on the composition of other aggregates in a eukaryotic cell, and demonstrate that other aggregates
expand the range of proteins that are susceptible to sequestration by polyglutamines.
Citation: Gong H, Romanova NV, Allen KD, Chandramowlishwaran P, Gokhale K, et al. (2012) Polyglutamine Toxicity Is Controlled by Prion Composition and Gene
Dosage in Yeast. PLoS Genet 8(4): e1002634. doi:10.1371/journal.pgen.1002634
Editor: Tricia R. Serio, Brown University, United States of America
Received September 9, 2011; Accepted February 21, 2012; Published April 19, 2012
Copyright: ? 2012 Gong et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the NIH (grant R01GM058763 to YOC, subaward to YOC on grant R01GM093294, and grant R01GM086890 to MYS) and by
the Hereditary Disease Foundation. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: email@example.com
¤a Current address: Department of Human Genetics, Emory University School of Medicine, Atlanta, Georgia, United States of America
¤b Current address: Department of Biology, Emory University, Atlanta, Georgia, United States of America
¤c Current address: School of Science, Health, and Technology, Medgar Evers College – CUNY, Brooklyn, New York, United States of America
A variety of human neurodegenerative disorders are associated
with expansions of polyglutamine (poly-Q) repeats in certain
proteins [1,2]. One well known example is Huntington’s disease
(HD), which is caused by an expansion of the poly-Q stretch,
located within the N-terminal stretch of the essential protein called
huntingtin (Htt) . Poly-Q expansion promotes formation of
aggregates by the proteolytic Htt fragments containing an
expanded poly-Q stretch [4,5]. As the poly-Q-expanded N-
terminal region of Htt is shown to aggregate and produce HD-like
neurodegeneration in the mouse model, it is clear that this region
is sufficient for reproducing the characteristic features of poly-Q
aggregation and toxicity [6,7]. Poly-Q associated pathologies can
not be explained solely by the loss of the cellular function of a
respective protein, e. g. Htt (for review, see ). Sequestration of
other essential proteins by poly-Q aggregates was proposed to be a
possible mechanism for toxicity [1,8]. However, different exper-
imental models suggested different candidates for sequestration
[9–12], which decreased enthusiasm for the sequestration model.
To complicate matters further, expanded poly-Q proteins form
various types of aggregates in mammalian cells [13,14]. In the case
of Htt, both nuclear and cytoplasmic aggregates were found
[4,15,16]. Their contributions to poly-Q pathogenicity remain a
topic of intense discussion [17,18]. At least, most researchers agree
that one type of cytoplasmic aggregated structure, so-called
‘‘aggresome’’, plays a cytoprotective role via assembling poly-Q
expanded Htt at one site and possibly promoting its autophagy-
dependent clearance [19–22]. The aggresome is located perinu-
clearly, associated with the centrosome, and assembled with
participation of the microtubular cytoskeleton. Other misfolded
proteins can also be sequestered into an aggresome, indicating that
this structure serves as a universal quality control depot for
aggregating proteins [19,23–27].
Experimental assays for studying the molecular mechanism of
poly-Q aggregation and toxicity have been developed in the yeast
Saccharomyces cerevisiae [28–34]. It has been shown that cytoplasmic
aggregation and toxicity of the chimeric protein, generated by a
fusion of the expanded poly-Q stretch of Htt to the green
fluorescent protein (GFP), is facilitated by the presence of the
PLoS Genetics | www.plosgenetics.org1April 2012 | Volume 8 | Issue 4 | e1002634
endogenous yeast prions, [PIN+] and/or [PSI+] [30,35]. In the
absence of a prion, aggregates of this construct were rare, and no
significant cytotoxicity was detected. However, in the presence of a
prion, multiple peripherally located aggregates were formed, and
cytotoxicity was observed . The prions [PIN+] and [PSI+] are
self-perpetuating aggregates of the endogenous yeast proteins
Rnq1 (unknown function) and Sup35 (translation termination, or
release factor, also called eRF1), respectively (for review, see ).
Both of these proteins contain QN-rich prion domains (PrDs) that
are responsible for aggregation properties (for review, see [37,38]).
It is likely that pre-existing prion aggregates nucleate aggregation
of poly-Q expanded huntingtin. In the case of the Rnq1 prion, it
was shown that poly-Q aggregates sequester some cytoskeletal
components and inhibit endocytosis, which apparently contributes
to cytotoxicity . Inhibition of endocytosis was also detected in
mammalian cells expressing poly-Q . As mammalian Htt has
been proposed to play a role in vesicle trafficking , these results
are likely relevant to human HD.
Flanking sequences modulate poly-Q toxicity [32,42]. In yeast
strains containing the Rnq1 prion, cytotoxicity was eliminated by
using a longer Htt fragment, which includes a proline (P)-rich
stretch in addition to poly-Q. This P-rich stretch was shown to
target aggregated poly-Q protein into a single perinuclear
microtubule-dependent deposit, co-localized with the spindle body
(yeast counterpart of a centrosome) and therefore resembling a
mammalian aggresome . The cytoprotective role of the
aggresome, as opposed to cytotoxicity of some other types of
aggregates, recapitulates the situation previously observed in
mammalian cells [20,21,23].
While the prion form of Sup35 protein ([PSI+]) also promotes
poly-Q toxicity in the yeast assay , the mechanism for this
toxicity has not been studied in detail previously. In our current
work, we demonstrate that [PSI+]-dependent poly-Q toxicity is not
counteracted by aggresome formation, but is ameliorated by an
increased dosage of some components of the translational
termination machinery. These data show that targets of poly-Q
toxicity and the cytoprotective potential of the aggresome depend
on the composition of endogenous aggregated proteins in a
Aggresome formation does not ameliorate
polyglutamine toxicity in the presence of Sup35 prion
To distinguish between the different patterns of poly-Q
aggregation in yeast, we have employed the previously described
constructs (Figure 1A) that produce the N-proximal region of Htt,
fused to the FLAG epitope at the N-terminus and the green
fluorescent protein (GFP) at the C-terminus. The N-terminal Htt
region included the poly-Q stretch, which is either followed (poly-
QP) or not followed (poly-Q) by the P-rich region. The poly-Q
expanded versions (103Q and 103QP) contained a stretch of 103
glutamine residues, which corresponds to a severe form of
Huntington’s disease, while control non-aggregating versions
(25Q and 25QP) contained 25 glutamine residues. As there was
no difference in the effects of 25Q and 25QP, some of the figures
show only the 25Q control. As described previously , the
103Q construct was toxic to yeast strains containing either [PIN+]
(Rnq1 protein in a prion form) or [PSI+] (Sup35 protein in a prion
form), with a combination of both prions showing an additive
effect (Figure 1B). Also in agreement with previous observations
, the 103QP construct was not toxic to the strains containing
only Rnq1 prion. Surprisingly, the 103QP construct was toxic to
the strains containing the Sup35 prion, independently in the
presence or absence of Rnq1 prion (Figure 1B). Fluorescence
microscopy confirmed that 103QP preferentially formed a single
perinuclear aggregate deposit (aggresome) in the cells containing
Rnq1 and/or Sup35 prions, while 103Q produced multiple
peripheral aggregates (Figure 1C). Therefore, the ability of poly-
QP to form an aggresome was not affected by the Sup35 prion,
however, amelioration of toxicity by the aggresome was impaired.
These data show that the mechanism of polyglutamine toxicity,
promoted by the Sup35 prion, is different from the mechanism of
polyglutamine toxicity promoted by the Rnq1 prion. Unless stated
otherwise, all further experiments were performed in the strains
containing the [PIN+] prion, thereby comparing the [PSI+] and
[psi2] derivatives so that we could distinguish the effects attributed
specifically to the [PSI+] prion.
Polyglutamine accumulation leads to Sup35
Notably, when Sup35NM, tagged with DsRed, and 103QP-
GFP are co-overproduced in the [PSI+] strain, most of the
Sup35NM-DsRed is eventually assembled into one large deposit,
that is either partially or completely overlapping with the 103QP
aggresome (Figure 1D). This indicates a possibility of sequestration
of Sup35 by the aggresome, and agrees with the previous
observation , confirmed by us (Figure 1E) that polyglutamines
promote aggregation of a fraction of Sup35, even in a [psi2] strain.
Notably, 103QP toxicity in the [PSI+] cells was ameliorated by
introducing the Sup35 derivative (designated Sup35C) that lacks
the N-terminal (prion) and middle domains and therefore, is
functional but unable to be incorporated into the prion aggregates
(Figure 1F). Expression of Sup35C also decreased toxicity of 103Q
in the [PIN+PSI+] strain down to the levels observed in the [PIN+
psi2] strain. However, Sup35C did not affect toxicity of 103Q in
[PIN+psi2]. These data confirm that in the [PSI+] strain (but not in
the [PIN+psi2] strain), sequestration of Sup35 contributes to
Polyglutamine diseases, including Huntington disease, are
associated with expansions of polyglutamine tracts,
resulting in aggregation of respective proteins. The
severity of Huntington disease is controlled by both DNA
and non–DNA factors. Mechanisms of such a control are
poorly understood. Polyglutamine may sequester other
cellular proteins; however, different experimental models
have pointed to different sequestered proteins. By using a
yeast model, we demonstrate that the mechanism of
polyglutamine toxicity is driven by the composition of
other (endogenous) aggregates (for example, yeast prions)
present in a eukaryotic cell. Although these aggregates do
not necessarily cause significant toxicity on their own, they
serve as mediators in protein sequestration and therefore
determine which specific proteins are to be sequestered
by polyglutamines. We also show that polyglutamine
deposition into an aggresome, a perinuclear compartment
thought to be cytoprotective, fails to ameliorate cytotox-
icity in cells with certain compositions of pre-existing
aggregates. Finally, we demonstrate that an increase in the
dosage of a sequestered protein due to aneuploidy by a
chromosome carrying a respective gene may rescue
cytotoxicity. Our data shed light on genetic and epigenetic
mechanisms modulating polyglutamine cytotoxicity and
establish a new approach for identifying potential
therapeutic targets through characterization of the en-
dogenous aggregated proteins.
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org2April 2012 | Volume 8 | Issue 4 | e1002634
Figure 1. Polyglutamine toxicity and aggregation in the yeast strains with various prion compositions. A – Polyglutamine constructs
used in this work. All constructs were under the control of the galactose-inducible promoter (PGAL), and contained the FLAG epitope, N-terminal 17
amino acid residues and poly-Q stretch of human Htt, and were fused to the gene coding for green fluorescent protein (GFP) at C-terminus. Numbers
indicate length of poly-Q stretch. Poly-QP constructs also contained the proline-rich region of Htt (designated as P), immediately following the poly-Q
stretch. B – Expanded poly-Q without a P-rich region (103Q), expressed under the PGALpromoter on -Ura/Gal medium, is toxic in the presence of
either [PIN+] or [PSI+] (or both), with two prions showing an additive effect. In contrast, expanded poly-Q with a P-rich region (103QP) is toxic only in
the presence of [PSI+]. The 25Q construct, not exhibiting toxicity under these conditions, is shown as a control. The 25QP construct (not shown)
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org3April 2012 | Volume 8 | Issue 4 | e1002634
Isolation of the anti-polyQ-toxic (AQT) mutants
Next, we looked for other genetic factors influencing the [PSI+]-
dependent polyglutamine toxicity. As the ubiquitin-proteasome
system (UPS) is known to influence poly-Q effects in mammalian
cells, we have studied poly-Q toxicity in the yeast strain with a
deletion of the UBC4 gene, coding for one of the major yeast
ubiquitin-conjugating enzymes . Ubc4D did not improve
growth in the presence of polyglutamines (Figure 2A) however
the ubc4D 103Q strain produced spontaneously arising fast-
growing papillae (Figure 2B). Three independent papillae were
analyzed further, and each was confirmed to stably reproduce the
anti-toxic phenotype (Figure 2B and 2C), and also to ameliorate
toxicity of 103QP (Figure 2D). These derivatives were named AQT
for Anti-polyQ Toxicity, with respective phenotype designated as
Aqt+. Specific effect of AQT on toxicity caused by expanded
polyglutamines was especially pronounced after longer periods of
incubation (Figure 2C). All AQT derivatives retained the [PIN+]
and [PSI+] prions (data not shown). In each derivative, the Aqt+
phenotype was dominant (see Figure 2E as an example) and
segregated in a Mendelian fashion in meiosis (Table S1). All
pairwise genetic crosses between three independent AQT deriva-
tives produced 4 AQT: 0 wild-type pattern of segregation in the
vast majority of tetrads (Table S2), indicating that all AQT
derivatives are formally confined to a single genetic locus.
Reintroduction of the wild-type UBC4 gene into the AQT strain
decreased but did not completely eliminate amelioration of
toxicity, indicating that ubc4D strengthens the Aqt+phenotype
but is not required for its manifestation (Figure 2F).
Despite their anti-toxic effect, AQT derivatives retained the
typical mode of cytologically detectable aggregation for both 103Q
(multiple peripheral aggregates, Figure 3A) and 103QP (single
perinuclear aggregate, Figure 3B), indicating that amelioration of
toxicity is not due to a lack of aggregation. Overproduction of
Sup35 protein or its prion domain, Sup35N, is known to inhibit
growth of the [PSI+] strains [37,38]. This effect was ameliorated in
AQT derivatives (Figure 3C). The AQT strains also exhibited
additional phenotypes that were not directly related to ameliora-
tion of toxicity, such as compensation of the ubc4D-mediated
temperature sensitivity and loss of the invasive growth capability
(Figure S1). AQT also slightly increased the growth of the [PIN+
psi2] ubc4D strain expressing 103Q, especially after shorter
incubation periods (Figure S1D), possibly due to increased
robustness of the AQT ubc4D strain in the stressful conditions.
AQT derivatives are disomics of chromosome II
Genetic crosses aimed at characterizing the inheritance of AQT
revealed that AQT is centromere-linked (Figure S2A). Moreover,
when the original AQT derivative, obtained in the strain bearing
the ubc4D::HIS3 disruption was mated to the wild-type strain
bearing the ubc4D::KanMX disruption (causing resistance to the
antibiotic G418), both AQT and KanMX markers segregated 2:2 in
meiosis as expected, while majority of tetrads did show 3:1 or 4:0
segregation for the His+phenotype, indicative of the presence of
two copies of the HIS3 allele in the cross (Figure 4A). Notably, all
AQT spores (76 total) obtained from tetrads with a 2:2 ratio for
KanMX were His+. The simplest scenario compatible with these
ratios is that the AQT derivative is a disomic of chromosome II,
where the ubc4D:: HIS3 allele is located. As chromosome
segregation is controlled by the centromere, this would also
explain the centromere linkage of AQT. Indeed, fractionation of
the yeast chromosomes via CHEF (Contour-clamped Homoge-
neous Electric Field) gel electrophoresis confirmed that each of the
three independent AQT derivatives contains an additional copy of
the chromosome II band (Figure S2B). Extra-copy of chromosome
II also co-segregated with AQT in tetrad analysis (data not shown).
Microarray-based analysis of genomic DNA of the three original
AQT derivatives and two AQT meiotic segregants confirmed that
each of these strains contains an extra copy of every piece of the
coding material in chromosome II (Figure 4B). Overall, our data
demonstrate that AQT is associated with an extra copy of
Extra copy of the SUP45 gene is responsible for the
amelioration of [PSI+]-dependent polyglutamine toxicity
in the AQT derivatives
Sequential deletion analysis of the extra copy of chromosome II
in the AQT strain confined the gene responsible for amelioration of
toxicity to the region of the right arm, located between positions
528161 and 537490 (Figure 4C). This region contains 5 ORFs,
including the essential gene SUP45, that codes for a translation
termination factor Sup45, or eRF1, working together with Sup35
(eRF3) . We have disrupted the copy of the SUP45 gene,
located on the duplicated chromosome II in the AQT strain, and
have shown that this disruption eliminates the anti-toxicity effect
on both 103Q and 103QP (Figure 4D). Notably, other
phenotypes, associated with chromosome II disomy but not
related to amelioration of polyglutamine toxicity in the [PSI+]
strain, including slightly increased growth of the [PIN+psi2] strain
in the presence of 103Q, were not affected by sup45D (Figure S1B,
S1C and S1E).
Western blot analysis confirmed that the AQT derivative
contains more Sup45 protein, compared to the isogenic wild type
strain (Figure 4E). This increase was more profound in the ubc4D
than in the UBC4+background. This explains why the AQT effect
was better seen in ubc4D. Thus, an increase in Sup45 levels due to
the presence of an extra-copy of the SUP45 gene is responsible for
the anti-toxic (Aqt+) phenotype in the [PSI+] background.
Plasmid-mediated overproduction of a release factor also
ameliorates polyglutamine toxicity
Next, we checked if an increase in the Sup45 levels, produced
by means other than duplication of chromosome II, would also
ameliorate the [PSI+]-dependent polyglutamine toxicity. Indeed,
introduction of the centromeric plasmid bearing the SUP45 gene
under its own (Figure 5A) or galactose-inducible (Figure 5B)
promoter (in the latter case, under inducing conditions) amelio-
rated toxicity of both 103Q and 103QP. Anti-toxic effect of
plasmid-borne SUP45 was clearly detected in both ubc4D and
behaved in the same way as 25Q. C – 103Q and 103QP form multiple peripheral aggregates and single aggregate (aggresome), respectively, in cells
containing either or both prions ([PIN+] and/or [PSI+]), as visualized by fluorescence microscopy. Perinuclear location of aggresome (not shown) was
confirmed by DAPI staining as described previously . D – Overexpressed Sup35NM-DsRed (red) forms large clumps in the [PSI+] cells, that overlap
with the 103QP-GFP aggresome (green), as pointed by arrows. E - Expression of 103Q or 103QP promotes aggregation of Sup35 in the [psi2] strain as
seen by an increase of pellet (P) versus supernatant (S) fraction, in comparison to the respective strain expressing 25Q. Centrifugation analysis was
followed by Western blotting and immunostaining with the Sup35 antibody. F - Expression of the Sup35 derivative, lacking the prion and middle
domains (Sup35C), decreases 103Q and 103QP toxicity in the [PIN+PSI+] strain but does not influence 103Q toxicity in the [PIN+psi2] strain. SUP35C
gene was under control of the endogenous SUP35 promoter. Serial decimal dilutions were spotted onto -Ura/Gal medium.
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org4April 2012 | Volume 8 | Issue 4 | e1002634
Figure 2. Isolation and characterization of anti-polyQ toxicity (AQT) derivatives. A – Ubc4D has no significant effect on toxicity of 103Q or
103QP in the [PIN+PSI+] background. Serial decimal dilutions were spotted onto -Ura/Gal medium. B – Papillae arise spontaneously in the ubc4D [PIN+
PSI+] strain expressing 103Q, and are able to stably maintain the anti-polyQ-toxic phenotype after colony purification. These papillae were designated
as AQT. C – Comparison of the growth curves of [PIN+PSI+] ubc4D strains that differ by polyglutamine constructs and by the presence or absence of
AQT. Growth was measured by optical density at 600 nm in the liquid –Ura medium with galactose and raffinose instead of glucose. At least 3
independent cultures were characterized per each combination. Error bars represent standard deviations. D – AQT ameliorates 103QP toxicity. E –
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org5 April 2012 | Volume 8 | Issue 4 | e1002634
UBC4+backgrounds, indicating that it is less sensitive to the
presence of Ubc4 protein, compared to the chromosomal extra
copy. For both plasmids, Sup45 overproduction was confirmed by
protein analysis (Figure 5C and 5D). Ability of the extra-copy of
SUP45 to ameliorate polyglutamine toxicity was abolished by a
deletion of 19 C-terminal amino acids (Figure 5E), that impairs
Sup45 function in translation and interaction with Sup35 , or
by the missense mutation sup45-103, T62C (Figure 5F) that also
impairs Sup45 function in translation termination . Thus,
Sup45 ability to ameliorate toxicity depends on the same sequence
elements that control its function in translational machinery.
Aggregation patterns of polyQ and release factors in the
wild-type and AQT strains
As both polyglutamines and prion form of Sup35 form SDS-
resistant polymers in the yeast cells, we have checked if patterns of
their aggregation are influenced by the presence of an extra copy
of SUP45. Both 103Q and 103QP proteins exhibit a broad range
of distribution of the SDS-resistant polymers by size, as
demonstrated by semi-denaturing agarose gel electrophoresis
(SDD-AGE), with 103QP containing more protein in the higher
molecular weight (MW) fraction (Figure 6A). This result confirms
that the aggresome, formed by 103QP, contains insoluble protein
aggregates, in contrast to the juxtanuclear quality control
compartment (JUNQ) observed in the yeast cells with a defect of
the ubiquitin-proteasome system . Neither 103Q nor 103QP
polymer distribution was significantly affected by AQT (Figure 6A).
In the wild type [PSI+] cells containing non-expanded polygluta-
mines (25Q), Sup35 prion polymers were distributed within a
relatively narrow range of sizes (Figure 6B). However, in the
presence of either 103Q or 103QP, size range of the Sup35
polymers was increased and higher molecular weight (MW)
polymers were accumulated, suggesting that some Sup35 could be
associated with 103Q (or 103QP) polymers, therefore partly
following their distribution. Notably, the Sup35 polymer size range
became narrower in the presence of AQT, and the high MW
fraction, which depends on 103Q/QP, disappeared (Figure 6B).
This suggests that the extra dosage of Sup45 somewhat
counteracts the increase in size of the Sup35 polymers and
possibly, their interaction with polyglutamines.
We have also checked effects of polyglutamines and gene dosage
on patterns of Sup45 aggregation. Sequestration of Sup45 by the
Sup35 prion aggregates is known to contribute to cytotoxicity of
overproduced Sup35 in [PSI+] strains . We could not detect
aggregate-associated Sup45 by SDD-AGE (data not shown),
apparently because it is not converted into an amyloid form and
is therefore released after SDS treatment. However, centrifugation
analysis demonstrated that presence of either [PSI+] prion or 103Q
protein resulted in the shift of a fraction of Sup45 protein to the
pelletable (aggregate-associated) form, with both prion and 103Q
together having an additive effect (Figure 6C). 103QP did not
exhibit any observable effect on Sup45 aggregation in the [psi2]
strain, however it further increased Sup45 aggregation in the
presence of [PSI+]. Remarkably, proportion of the pelletable
versus soluble Sup45 was decreased in the AQT (disomic) [PSI+]
strain expressing 103Q or 103QP, compared to the identical strain
not possessing disomy (Figure 6D). This showed that an increase in
Sup45 levels counteracted its sequestration by aggregates.
Overall, our data indicate that both release factors, Sup35 and
Sup45, are sequestered by polyglutamine aggregates in the [PSI+]
cells, and that excess Sup45 not only restores supply of functional
Sup45 but also changes the mode of Sup35 aggregation.
Polyglutamines and translational readthrough
As our data point to sequestration of release factors as a
mechanism of polyglutamine toxicity, we have checked if
polyglutamines increase translational readthrough of stop codons.
For this purpose, the chimeric constructs bearing a stop codon
between the PGK1 and lacZ ORFs have been employed.
Surprisingly, no increase in translational readthrough (measured
by b-galactosidase activity) has been detected in the presence of
103Q (Table S3). One possible explanation of these data is that
damage to translational machinery, caused by the aggregation and
sequestration of release factors in the presence of polyglutamines,
is so severe that translation is arrested and not proceeding beyond
the stop codon. Another (but not mutually exclusive) possibility is
that cytotoxicity is related to non-translational functions of Sup35/
45. Indeed, it has been reported that the immediate consequence
of the severe shortage of a release factor in yeast is not translational
defect per se, but cytoskeleton damage leading to cell death .
Prion role in polyglutamine toxicity
Our data demonstrate that mechanism of polyglutamine
toxicity depends on the prion composition of the cell. In fact, it
appears that polyglutamine protein is not the toxicity agent itself,
but rather amplifies the effects of the endogenous prion aggregates
by sequestering them and making them more rigid. In case of
Rnq1 prion, sequestration of Rnq1 by polyglutamines also leads to
sequestration of the cytoskeletal proteins, interacting with Rnq1,
and subsequent impairment of endocytosis [39,40]. In this case,
toxicity is relieved by re-localization of polyglutamines to an
aggresome that removes polyQ (and possibly Rnq1) from the
endocytic sites . However, in case of Sup35 prion, relocaliza-
tion is not sufficient for amelioration of toxicity. This could be due
to the fact that Sup35 itself is an essential protein, and/or due to its
normal distribution all over the cytoplasm, making it impossible to
define specific toxicity sites. Additive action of Rnq1 and Sup35
prions on 103Q toxicity in the absence of the P-rich region also
confirms that their cytotoxic effects are at least partly independent
of each other.
Remarkably, our data show for the first time that the aggresome
is not always cytoprotective. Moreover, it is possible that formation
of the aggresome in the cell containing an essential protein in the
form of self-perpetuating amyloid (prion) is itself cytotoxic due to
sequestration of this essential protein. However, it remains
uncertain whether toxicity is primarily driven by sequestration of
Sup35 into an aggresome, or by its sequestration into the smaller
polyglutamine aggregates remaining in the cytoplasm. While some
Sup35 is definitely detected in the aggresome (Figure 1D), we don’t
know if the functional fraction of Sup35 is sequestered there.
Moreover, while it is obvious that some fraction of Sup35 should
retain function in the [PSI+] strain, as elimination of Sup35 is
lethal , it remains unknown whether this functional compo-
nent of Sup35 is represented by residual non-aggregated Sup35,
smaller oligomers, or both. However, it is more likely that a
fraction of oligomeric Sup35 remains functional, as amount of
monomeric Sup35 retained by the strong [PSI+] strains, that is
essentially at the limit of detection, seems too low for maintaining
AQT is dominant (all strains are [PIN+PSI+] and ubc4D homozygotes). F – Reintroduction of the UBC4 gene under galactose-inducible promoter on a
multicopy plasmid partly suppresses but does not completely eliminate anti-toxic effect of AQT. -Ura/Gal plates are scored on panels D, E and F.
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org6 April 2012 | Volume 8 | Issue 4 | e1002634
viability. Indeed, Sup35C regions are not included in the amyloid
core and may stay enzymatically active, so that only size,
composition and/or location of the aggregate would modulate
its functionality in the cellular context. Changes in the distribution
of Sup35 polymers by size in the presence of polyglutamines
(Figure 6B), clearly show that certain alterations of Sup35
aggregation patterns, making them more similar to poly-Q
aggregation patterns, coincide with toxicity.
Modulation of polyglutamine toxicity by Sup45 dosage
Amelioration of [PSI+]-dependent cytotoxicity by extra-dosage
of the Sup35 functional partner, Sup45, confirms that toxicity
results from sequestration of release factor(s) by polyglutamine
aggregates. Possibly Sup35, containing a QN-rich domain, is
sequestered directly, while Sup45 is sequestered via its interaction
with Sup35. Indeed, ability of polyglutamines to facilitate
aggregation of endogenous QN-rich proteins even in a non-prion
strain has been reported previously [43,52] and confirmed by us
(Figure 1E and Figure 6D), and it was shown that Sup35 prion
aggregates produced at high levels cause toxicity via sequestering
Sup45 . In agreement with these data, AQT (i. e., extra copy of
SUP45) ameliorates both polyglutamine toxicity (Figure 2) and
toxicity of excess Sup35 (Figure 3C) in the [PSI+] cells. There is
probably a competition for the Sup35/Sup45 complex between
polyglutamine aggregates and functional sites (ribosome etc.) at
which the Sup35/Sup45 complex is supposed to act. Therefore, an
increased abundance of Sup45 not only increases a proportion of
non-sequestered Sup45 but also partly counteracts sequestration of
Sup35, that can be seen as a change in size distribution of the
Sup35 polymers (Figure 6B). Hence, the antitoxic effect of excess
Role of ubc4 deletion in AQT detection
The AQT derivatives were originally detected in the strain,
lacking the major ubiquitin-conjugating enzyme Ubc4. One
obvious reason for this is that amelioration of toxicity by AQT is
more pronounced in the absence of Ubc4 (Figure 2F), making
detection of the anti-toxic papillae easier. It is possible that Ubc4
promotes ubiquitination and subsequent degradation of a fraction
of excess Sup45, that agrees with a more profound increase in
Sup45 protein levels, which was detected in the ubc4D strain
bearing an extra-copy of SUP45, in comparison to the isogenic
UBC4+strain (Figure 4E). In addition, ubc4D may influence
patterns of polyglutamine aggregation and/or ability of poly-
glutamines to sequester other proteins. Indeed, defects of the
ubiquitin system are known to promote aggresome formation in
mammalian cells , and ubc4D influences formation and
aggregation of the [PSI+] prion yeast, as well as levels of some
Hsps and patterns of their interactions with prion aggregates .
Another, although not necessarily exclusive explanation of
increased AQT appearance in ubc4D cells is that a lack of Ubc4
may affect chromosome segregation and/or recombination,
therefore increasing the frequency of chromosome non-disjunc-
tion. Ubiquitination and ubiquitin-dependent protein degradation
are involved in regulation of DNA repair and chromosome
segregation [54,55]. Ubc4 is implicated in ubiquitination of
histones , and ubc4D is shown to affect proper segregation of
some yeast plasmids . Persisting variations of the chromosome
II size in AQT strains (for example, see Figure S2B) and occasional
appearance of weak additional bands on the CHEF gels of the
ubc4D strains (data not shown) speak in favor of a detrimental effect
of ubc4D on chromosome stability. It is possible that the presence
of the foreign DNA (KanMX insertion) on chromosome II of the
ubc4D strain aids in destabilization of this specific chromosome.
Figure 3. Effects of AQT on polyglutamine aggregation and
Sup35 toxicity. A and B – Typical aggregation patterns of 103Q
(multiple dots, A) and 103QP (single clump, B) are not affected by AQT,
as confirmed by fluorescence microscopy. C – AQT mutant ameliorates
toxicity of excess Sup35 or Sup35N in the [PSI+] strain. Sup35 and
Sup35N proteins were expressed from centromeric plasmid under
control of the galactose-inducible promoter. Cells were grown on the -
Ura/glucose medium selective for the plasmid for 1 day. Serial decimal
dilutions were plated onto -Ura/Gal medium.
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org7 April 2012 | Volume 8 | Issue 4 | e1002634
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org8April 2012 | Volume 8 | Issue 4 | e1002634
Irrespective of the mechanism of the ubc4D effect, this deletion is
required for neither maintenance of the chromosome II extra copy
nor toxicity amelioration. The UBC4+strains with an extra copy of
chromosome II were obtained by genetic cross and dissection and
continued to maintain a disomy (data not shown), and ameliora-
tion of toxicity by excess Sup45 was still detected in the UBC4+
strain, at a lower level in case of chromosome II extra copy
(Figure 2F) and at more profound level for the plasmid-mediated
excess Sup45 that is less sensitive to the presence of Ubc4
Relevance of yeast data to human polyglutamine
Involvement of translational machinery in HD has been
suspected from some results in mammalian systems . It
remains unknown if polyglutamines can sequester the human
homologs of Sup35 and Sup45 (respectively, eRF3 and eRF1), as
mammalian ortholog of Sup35 does not have a QN-rich domain.
However, our results could be relevant to mammalian systems in a
more general way. About 40% of the variation in the age of HD
onset, in cases where the polyglutamine repeat is of the same
length, is due to DNA variation . Our work provides a
potential explanation for such a variation by demonstrating that
changes in the abundance of the sequestered protein(s), occurring
via alteration of either gene dosage or gene expression, can
modulate polyglutamine toxicity. A non-DNA component of
variations in polyglutamine toxicity can be explained by
differences in the composition of other aggregated proteins (e. g.
endogenous self-perpetuating aggregates or prions) present in the
cell. Our results show that prion composition of the cell not only
drives polyglutamine toxicity but also determines a pathway via
which polyglutamines influence cell physiology, as proteins already
associated with the other aggregates are more likely to be
sequestered by polyglutamines. Mammalian cells contain a variety
of proteins with the prion-like QN-rich domains, and machinery
for propagation of the QN-rich protein aggregates exists in
mammals . Protein aggregation can also be induced by
oxidative damage and other stresses. It was reported that
artificially generated b-rich aggregates may sequester other
proteins . It is therefore entirely possible that organisms or
tissues (or both) differ by the aggregate composition, and this in
turn influences their susceptibility to polyglutamine disorders.
Composition of endogenous aggregates may also modulate which
proteins are sequestered by polyglutamines, as proteins associated
with other aggregates interacting with polyglutamines are more
likely to be sequestered, like Sup45 in the cells containing the
Sup35 prion. This could explain why different groups are coming
out with different conclusions in regard to both mechanisms of
polyglutamine toxicity and contributions of different types of
Materials and Methods
Yeast strains and growth conditions
The Saccharomyces cerevisiae strains, used in this study and listed in
Table S4, are derivatives of GT81 series  of the prototype
haploid genotype ade1 his3 leu2 lys2 trp1 ura3, with different mating
types and various prion compositions. The individual gene
deletions were made by using PCR-mediated transplacement with
the cassette bearing either Schizosaccharomyces pombe HIS5 gene, an
ortholog of S. cerevisiae HIS3 gene (thus designated in this paper as
HIS3), or bacterial kanrgene, which causes resistance to G418 in
yeast . Spontaneous AQT mutants were initially obtained in
the strain GT349 (MATa ubc4D::HIS3 [PIN+PSI+]), as described in
Standard yeast media, procedures (including transformation,
phenotype scoring, velveteen replica plating, mating and sporu-
lation), and growth conditions were used . Yeast cultures were
grown at 30uC except for the temperature-sensitivity assays
(employing 39uC). Tetrad dissection was performed by using the
MSM System 300 micromanipulator from Singer Instrument Co.
Ltd. Analysis of yeast chromosomes by CHEF (Contour-clamped
Homogeneous Electric Field) is described in Text S1.
Polyglutamine toxicity was detected as growth inhibition on the
synthetic dropout medium with galactose instead of glucose where
polyglutamine constructs were selectively maintained and induced.
As most of our polyglutamine constructs were expressed from
plasmids bearing the URA3 marker, the plasmid-selective galactose
medium (-Ura/Gal) was used, unless stated otherwise. Polygluta-
mine toxicity becomes more evident after longer incubation
periods, as also confirmed by growth curves (see Figure 2C).
Typically, velveteened plates were scanned following 5–10 days of
incubation after a second passage on galactose medium, while
spotted from solution (without dilutions) plates were scanned after
3–5 days, and serial decimal dilutions spotted onto galactose
medium were scanned after 2–3 weeks.
Major plasmids used in this study are described in Text S1. A
list of plasmids is available in Table S5.
Construction of chromosomal deletions
Strategy of making chromosomal deletions is described in Text
Figure 4. AQT derivatives are disomic for chromosome II, and extra-copy of SUP45 is responsible for antitoxicity. A – Tetrad analysis of
a diploid obtained from mating of the AQT strain bearing the ubc4D::HIS3 transplacement, to the strain bearing the ubc4D::KanMX transplacement,
demonstrates presence of at least 2 copies of the HIS3 gene versus one copy of the KanMX gene. This can be concluded from the fact that majority of
tetrads produce more than 2 His+spores, in contrast to the typically 2:2 segregation by G418 resistance caused by KanMX. All AQT spores in this cross
were His+(not shown). B – Hybridization of total DNA to a complete DNA microarray of the S. cerevisiae genome confirms that all the coding material
of chromosome II is duplicated in the AQT strain. Comparison is performed according to CLAC (CLuster Along Chromosome) consensus plot. For
procedure, see Text S1. C – Sequential deletion mapping of the chromosome II extra copy in the AQT strain. The AQT#7 derivative (see Figure S2B)
was used in these experiments. Each numbered region corresponds to a respective deletion. Deletions eliminating the antitoxicity phenotype in the
[PSI+] background are shown as boxes filled in black. All deletions were verified by PCR. Five ORFs located within region 2.1a were each deleted
individually; among those deletions, only deletion of SUP45 eliminated AQT as shown on panels B and C. D – Elimination of the antitoxic effect on
103Q and 103QP by the sup45 deletion in AQT strain. Serial decimal dilutions were spotted onto -Ura/Gal medium. E – Sup45 protein levels are
elevated in the AQT strain, more profoundly in ubc4D background than in the presence of wild type UBC4 gene (UBC4+). Sup45p level is shown
relative to the isogenic monosomic (non-AQT) control in each case. Ade2 protein was used as the loading control. At least 3 measurements with
independent cultures were performed in each case. Error bars correspond to standard deviations. In each case the difference in Sup45 levels between
the AQT and non-AQT strain is statistically significant as confidence limits do not overlap, and differences between the UBC4+and ubc4D strains are
statistically significant according to t-criterium (PHo,0.01).
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org9April 2012 | Volume 8 | Issue 4 | e1002634
Fluorescence microscopy was performed according to standard
techniques, as described in Text S1.
Protein isolation and electrophoresis are described in Text S1.
Semi-Denaturing Detergent-Agarose Gel Electrophoresis (SDD-
AGE), used to fractionate the SDS-resistant protein polymers
according to their sizes, was performed according to the standard
protocol  with slight modifications. Proteins were diluted in
2% SDS, incubated for 5 min at room temperature before
loading, run in the 1.5% agarose gel with 0.1% SDS in 1X TAE
buffer containing 0.1% SDS, transferred to nitrocellulose
membrane (Whatman) by capillary blotting, and reacted to
appropriate antibody. Assay for b-galactosidase activity was
performed according to the standard protocol , except that
cell debris was removed by centrifugation to avoid light scattering
before the OD reading at 420 nm was taken.
Description of antibodies used in this study can be found in Text
DNA microarray analysis
Gene copy number was determined by hybridization to the
complete DNA microarray of the Saccharomyces cerevisiae genome, as
described previously . Detailed information can be found in
the Text S1.
Figure 5. Modulation of polyglutamine toxicity by the plasmid-borne release factor genes. A - An extra copy of SUP45 gene, located on
the centromeric plasmid under endogenous promoter, ameliorates toxicity of 103Q in the ubc4D [PSI+] strain, as seen from serial decimal dilutions
plated onto the galactose medium selective for both poly-Q and SUP45 (or control) plasmids. B – Amelioration of [PSI+]-dependent polyglutamine
toxicity by a plasmid-borne extra copy of SUP45 gene is detected for both endogenous (CEN-SUP45) galactose-inducible (GAL-SUP45) promoters, for
both 103Q and 103QP constructs, and in both ubc4D and UBC4+strains. Antitoxic effect of the plasmid-borne SUP45 gene in the ubc4D strain is
comparable to antitoxic effect of AQT. Toxicity was scored on the galactose medium selective for both poly-Q and SUP45 (or control) plasmids. C and
D – Centromeric plasmids with SUP45 gene under endogenous (C) or galactose-inducible PGAL(D) promoters increase levels of Sup45 protein
(Sup45p) both [UBC4+] and ubc4D strains. Cultures were grown in liquid -Ura -Leu glucose (C) or -Ura -Leu galactose/raffinose (D) medium. Ade2
(Ade2p) protein is shown as a loading control. E and F – Plasmids, expressing the SUP45 alleles with either C-terminal deletion, SUP45DC19 (that
abolishes Sup45 function and interaction with Sup35) (E) or missense mutation sup45-103, T62C (that impairs Sup45 function) (F) from the
endogenous SUP45 promoter, do not ameliorate 103Q and 103QP toxicity, as scored on the galactose medium selective for both plasmids.
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org10 April 2012 | Volume 8 | Issue 4 | e1002634
A – ubc4D causes complete inhibitionof growth at 39uC. AQT partly
compensates for this defect of growth. Cultures were incubated in
SUP45-independent phenotypes associated with AQT.
the liquid YPD medium for 1 day and serial decimal dilutions were
spotted onto a YPD plate. B – Deletion of SUP45 does not affect
compensation of temperature resistance by AQT. C - The invasive
growth phenotype is eliminated by AQT in a SUP45-independent
manner. Cells were patched on a YPD plate and grown for 2 days.
The plate was scanned before and after gentle wash under running
water for 3 min. Similar effect of AQT was observed in the UBC4+
strain (not shown). D – AQT slightly increases growth of the [PIN+
psi2] ubc4D strain in –Ura/galactose+raffinose medium in the
presence of 103Q, as seen after relatively short periods of
incubation. It is not known if this effect is specific to 103Q or is a
consequence of the general increase in robustness of the AQT strain
in these conditions. Cultures were grown in liquid –Ura/glucose
medium for 1 day, and washed 3 times prior to the induction of
103Q in –Ura/galactose+raffinose medium, starting with the
inocula of the same concentration. Serial decimal dilution were
spotted onto –Ura/glucose medium after 24 hrs of growth. E –
Deletion of the extra copy of SUP45 gene does not eliminate the
AQT effect on growth in the presence of 103Q in the [PIN+psi2]
strain, confirming that the molecular basis of this phenotype is
different from the antitoxicity detected in the [PSI+] background.
the extra-copy of chromosome II. A – Tetrad analysis of the
diploid heterozygous by both AQT and met3D (a centromere-linked
marker on chromosome X) demonstrates that AQT is centromere-
linked, as seen from low proportion of tetratypes (T) in comparison
to parental (PD) and non-parental (NPD) ditypes (P,0.001). AQT
is scored by growth on –Ura/Gal medium in the presence of 103Q
plasmid, and met3D is scored by lack of growth in the absence of
methionine (-Met). Similar results (not shown) were obtained after
sporulating and dissecting diploids generated by mating the AQT
strain to the isogenic strains of the opposite mating type,
containing disruptions of the centromere-linked genes met28
(chromosome IX) or met14 (chromosome XI). For an explanation
of tetrad types, see ref. . B – Chromosome fractionation by
CHEF (left), followed by Southern blotting (right) demonstrates the
presence of the extra copy of chromosome II in all AQT
derivatives. Chromosome II bands are indicated by arrows on
the CHEF gel, and visualized by hybridization to the labeled
fragment of SSA3 gene (located on chromosome II) on Southern
blot. Per each independent AQT derivative (designated as AQT
#2, AQT #7 and AQT #9) and wild-type control, two isolates are
tested. An extra-band chromosome II was also co-inherited with
AQT in meiosis (not shown). Notably, electrophoretic mobilities of
duplicated chromosomes varied among AQT derivatives, and in
one AQT derivative (#2) the difference was detected between two
isolates. Variations in electrophoretic mobilities of chromosome II
copies were also detected after meiosis of the AQT-containing
diploids (data not shown). As all isolates contain a duplication of
the whole coding material of chromosome II (see Figure 4B),
variations in electrophoretic mobility are apparently due to
repetitive non-coding elements or may reflect exchanges of
material between non-homologous chromosomes.
Additional evidence for the association of AQT with
mated to the isogenic wild type (WT) ubc4D strain of the opposite
Mendelian inheritance of AQT. Each AQT strain was
* In parentheses are numbers of tetrads showing the respective
ratio. ** One exceptional tetrad with 3:1 ratio was recovered.
Recombination test for allelism of AQT derivatives.
Figure 6. Effects of AQT on aggregation of release factors. A and
B – Fractionation of the polyQ/QP-GFP (A) and Sup35 (B) polymers by
sizes in the ubc4D [PSI+] strains either with (AQT) or without (WT) AQT.
Polymers were separated by semi-denaturing agarose gel electropho-
resis (SDD-AGE, see Text S1). Filter obtained from one and the same gel
was reacted to either GFP (A) or Sup35C (B) antibodies. Polyglutamines
alter distribution of Sup35 polymers, and this effect is counteracted by
AQT. Experiment has been repeated with 3 independent cultures per
each combination, each time with the same result. C – Expression of
103Q promotes aggregation of Sup45 protein (Sup45p) in the ubc4D
[PIN+psi2] strain, and expression of either 103Q or 103QP increases
aggregate-associated fraction of Sup45 in the ubc4D [PSI+] strain, as
detected by an increase in the pellet (P) versus supernatant (S) fraction
in comparison to the respective strain expressing 25Q. Centrifugation
was followed by Western blotting and reaction to the Sup45 antibody.
D – Proportion of soluble (supernatant, S) versus aggregate-associated
(pellet, P) Sup45 protein is significantly increased in AQT ubc4D [PSI+]
strain, compared to the identical non-AQT (WT) strain, as determined by
centrifugation analysis, followed by Western blotting and reaction to
the Sup45 antibody.
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org11April 2012 | Volume 8 | Issue 4 | e1002634
103Q. Cultures were grown in -Ura -Trp glucose medium to early
stationary phase. Cells were washed 3 times before being
transferred to -Ura-Trp/galactose+raffinose medium for 24-hr
induction. Three independent cultures were tested. Differences are
not statistically significant (PHo.0.05).
UGA readthrough in the absence and presence of
Yeast strains. * Chromosome II disomics.
List of plasmids.
tions; construction of chromosomal deletions; fluorescence mi-
croscopy; protein analysis; antibodies; electrophoretic separation
Supporting Materials and Methods. Plasmid construc-
of yeast chromosomes; DNA microarray analysis. References ,
, , , and [67–80] are quoted.
We thank K. Lobachev for guidance in CHEF gel electrophoresis; A.
Romanyuk for guidance in SDD-AGE experiments; F. Storici for guidance
in deletion analysis; D. Kiktev for plasmid construction; C. Kubicek for
technical help; D. Bedwell, M. Tuite, and G. Zhouravleva for plasmids and
Conceived and designed the experiments: YOC NVR PM. Performed the
experiments: HG NVR KDA PC KG GPN PM. Analyzed the data: HG
NVR KDA PM MYS YOC. Contributed reagents/materials/analysis
tools: PM MYS. Wrote the paper: YOC HG PM. Helped to edit and
format the manuscript: GPN MYS.
1. Shao J, Diamond MI (2007) Polyglutamine diseases: emerging concepts in
pathogenesis and therapy. Hum Mol Genet 16: R115–123.
La Spada AR, Taylor JP (2010) Repeat expansion disease: progress and puzzles
in disease pathogenesis. Nat Rev Genet 11: 247–258.
The Huntington’s Disease Collaborative Research Group (1993) A novel gene
containing a trinucleotide repeat that is expanded and unstable on Huntington’s
disease chromosomes. Cell 72: 971–983.
DiFiglia M, Sapp E, Chase KO, Davies SW, Bates GP, et al. (1997) Aggregation
of huntingtin in neuronal intranuclear inclusions and dystrophic neurites in
brain. Science 277: 1990–1993.
Lunkes A, Lindenberg KS, Ben-Haiem L, Weber C, Devys D, et al. (2002)
Proteases acting on mutant huntingtin generate cleaved products that
differentially build up cytoplasmic and nuclear inclusions. Mol Cell 10: 259–269.
Mangiarini L, Sathasivam K, Seller M, Cozens B, Harper A, et al. (1996) Exon 1
of the HD gene with an expanded CAG repeat is sufficient to cause a progressive
neurological phenotype in transgenic mice. Cell 87: 493–506.
Stack EC, Kubilus JK, Smith K, Cormier K, Del Signore SJ, et al. (2005)
Chronology of behavioral symptoms and neuropathological sequela in R6/2
Huntington’s disease transgenic mice. J Comp Neurol 490: 354–370.
Zoghbi HY, Orr HT (1999) Polyglutamine diseases: protein cleavage and
aggregation. Curr Opin Neurobiol 9: 566–570.
Steffan JS, Kazantsev A, Spasic-Boskovic O, Greenwald M, Zhu YZ, et al.
(2000) The Huntington’s disease protein interacts with p53 and CREB-binding
protein and represses transcription. Proc Natl Acad Sci USA 97: 6763–6768.
10. Hay DG, Sathasivam K, Tobaben S, Stahl B, Marber M, et al. (2004)
Progressive decrease in chaperone protein levels in a mouse model of
Huntington’s disease and induction of stress proteins as a therapeutic approach.
Hum Mol Genet 13: 1389–1405.
11. Ravikumar B, Vacher C, Berger Z, Davies JE, Luo S, et al. (2004) Inhibition of
mTOR induces autophagy and reduces toxicity of polyglutamine expansions in
fly and mouse models of Huntington disease. Nat Genet 36: 585–595.
12. Yamanaka T, Tosaki A, Miyazaki H, Kurosawa M, Furukawa Y, et al. (2010)
Mutant huntingtin fragment selectively suppresses Brn-2 POU domain
transcription factor to mediate hypothalamic cell dysfunction. Hum Mol Genet
13. Ross CA, Poirier MA (2004) Protein aggregation and neurodegenerative disease.
Nat Med 10 Suppl: S10–17.
14. Hands SL, Wyttenbach A (2010) Neurotoxic protein oligomerisation associated
with polyglutamine diseases. Acta Neuropathol 120: 419–437.
15. Davies SW, Turmaine M, Cozens BA, DiFiglia M, Sharp AH, et al. (1997)
Formation of neuronal intranuclear inclusions underlies the neurological
dysfunction in mice transgenic for the HD mutation. Cell 90: 537–548.
16. Scherzinger E, Lurz R, Turmaine M, Mangiarini L, Hollenbach B, et al. (1997)
Huntingtin-encoded polyglutamine expansions form amyloid-like protein
aggregates in vitro and in vivo. Cell 90: 549–558.
17. Ross CA, Wood JD, Schilling G, Peters MF, Nucifora FC, Jr., et al. (1999)
Polyglutamine pathogenesis. Philos Trans R Soc Lond B Biol Sci 354:
18. Ross CA, Tabrizi SJ (2011) Huntington’s disease: from molecular pathogenesis
to clinical treatment. Lancet Neurol 10: 83–98.
19. Johnston JA, Ward CL, Kopito RR (1998) Aggresomes: a cellular response to
misfolded proteins. J Cell Biol 143: 1883–1898.
20. Waelter S, Boeddrich A, Lurz R, Scherzinger E, Lueder G, et al. (2001)
Accumulation of mutant huntingtin fragments in aggresome-like inclusion
bodies as a result of insufficient protein degradation. Mol Biol Cell 12:
21. Taylor JP, Tanaka F, Robitschek J, Sandoval CM, Taye A, et al. (2003)
Aggresomes protect cells by enhancing the degradation of toxic polyglutamine-
containing protein. Hum Mol Genet 12: 749–757.
22. Olzmann JA, Li L, Chin LS (2008) Aggresome formation and neurodegener-
ative diseases: therapeutic implications. Curr Med Chem 15: 47–60.
23. Kopito RR (2000) Aggresomes, inclusion bodies and protein aggregation.
Trends Cell Biol 10: 524–530.
24. Garcia-Mata R, Gao YS, Sztul E (2002) Hassles with taking out the garbage:
aggravating aggresomes. Traffic 3: 388–396.
25. Olzmann JA, Li L, Chudaev MV, Chen J, Perez FA, et al. (2007) Parkin-
mediated K63-linked polyubiquitination targets misfolded DJ-1 to aggresomes
via binding to HDAC6. J Cell Biol 178: 1025–1038.
26. Chin LS, Olzmann JA, Li L (2010) Parkin-mediated ubiquitin signalling in
aggresome formation and autophagy. Biochem Soc Trans 38: 144–149.
27. Bondzi C, Brunner AM, Munyikwa MR, Connor CD, Simmons AN, et al.
(2011) Recruitment of the oncoprotein v-ErbA to aggresomes. Mol Cell
Endocrinol 332: 196–212.
28. Krobitsch S, Lindquist S (2000) Aggregation of huntingtin in yeast varies with
the length of the polyglutamine expansion and the expression of chaperone
proteins. Proc Natl Acad Sci USA 97: 1589–1594.
29. Muchowski PJ, Schaffar G, Sittler A, Wanker EE, Hayer-Hartl MK, et al. (2000)
Hsp70 and Hsp40 chaperones can inhibit self-assembly of polyglutamine
proteins into amyloid-like fibrils. Proc Natl Acad Sci USA 97: 7841–7846.
30. Meriin AB, Zhang X, He X, Newnam GP, Chernoff YO, et al. (2002)
Huntington toxicity in yeast model depends on polyglutamine aggregation
mediated by a prion-like protein Rnq1. J Cell Biol 157: 997–1004.
31. Duennwald ML, Jagadish S, Giorgini F, Muchowski PJ, Lindquist S (2006) A
network of protein interactions determines polyglutamine toxicity. Proc Natl
Acad Sci USA 103: 11051–11056.
32. Duennwald ML, Jagadish S, Muchowski PJ, Lindquist S (2006) Flanking
sequences profoundly alter polyglutamine toxicity in yeast. Proc Natl Acad Sci
USA 103: 11045–11050.
33. Duennwald ML (2011) Polyglutamine misfolding in yeast: toxic and protective
aggregation. Prion 5: 285–290.
34. Mason RP, Giorgini F (2011) Modeling Huntington disease in yeast:
perspectives and future directions. Prion 5: 269–276.
35. Gokhale KC, Newnam GP, Sherman MY, Chernoff YO (2005) Modulation of
prion-dependent polyglutamine aggregation and toxicity by chaperone proteins
in the yeast model. J Biol Chem 280: 22809–22818.
36. Wickner RB, Edskes HK, Shewmaker F, Nakayashiki T, Engel A, et al. (2007)
Yeast prions: evolution of the prion concept. Prion 1: 94–100.
37. Tuite MF, Cox BS (2003) Propagation of yeast prions. Nat Rev Mol Cell Biol 4:
38. Inge-Vechtomov SG, Zhouravleva GA, Chernoff YO (2007) Biological roles of
prion domains. Prion 1: 228–235.
39. Meriin AB, Zhang X, Miliaras NB, Kazantsev A, Chernoff YO, et al. (2003)
Aggregation of expanded polyglutamine domain in yeast leads to defects in
endocytosis. Mol Cell Biol 23: 7554–7565.
40. Meriin AB, Zhang X, Alexandrov IM, Salnikova AB, Ter-Avanesian MD, et al.
(2007) Endocytosis machinery is involved in aggregation of proteins with
expanded polyglutamine domains. FASEB J 21: 1915–1925.
41. Caviston JP, Holzbaur EL (2009) Huntingtin as an essential integrator of
intracellular vesicular trafficking. Trends Cell Biol 19: 147–155.
42. Wang Y, Meriin AB, Zaarur N, Romanova NV, Chernoff YO, et al. (2009)
Abnormal proteins can form aggresome in yeast: aggresome-targeting signals
and components of the machinery. FASEB J 23: 451–463.
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org12April 2012 | Volume 8 | Issue 4 | e1002634
43. Urakov VN, Vishnevskaya AB, Alexandrov IM, Kushnirov VV, Smirnov VN,
et al. (2010) Interdependence of amyloid formation in yeast: implications for
polyglutamine disorders and biological functions. Prion 4: 45–52.
44. Seufert W, Jentsch S (1990) Ubiquitin-conjugating enzymes UBC4 and UBC5
mediate selective degradation of short-lived and abnormal proteins. EMBO J 9:
45. Stansfield I, Jones KM, Kushnirov VV, Dagkesamanskaya AR, Poznyakovski AI,
et al. (1995) The products of the SUP45 (eRF1) and SUP35 genes interact to
mediate translation termination in Saccharomyces cerevisiae. EMBO J 14:
46. Kallmeyer AK, Keeling KM, Bedwell DM (2006) Eukaryotic release factor 1
phosphorylation by CK2 protein kinase is dynamic but has little effect on the
efficiency of translation termination in Saccharomyces cerevisiae. Eukaryot Cell 5:
47. Moskalenko SE, Zhuravleva GA, Soom M, Shabel’skaia SV, Volkov KV, et al.
(2004) Characterization of missense mutations in the SUP45 gene of Saccharomyces
cerevisiae encoding translation termination factor eRF1. Genetika 40: 599–606.
48. Kaganovich D, Kopito R, Frydman J (2008) Misfolded proteins partition
between two distinct quality control compartments. Nature 454: 1088–1095.
49. Vishveshwara N, Bradley ME, Liebman SW (2009) Sequestration of essential
proteins causes prion associated toxicity in yeast. Mol Microbiol 73: 1101–1114.
50. Valouev IA, Kushnirov VV, Ter-Avanesyan MD (2002) Yeast polypeptide chain
release factors eRF1 and eRF3 are involved in cytoskeleton organization and cell
cycle regulation. Cell Motil Cytoskeleton 52: 161–173.
51. Dagkesamanskaya AR, Ter-Avanesyan MD (1991) Interaction of the yeast
omnipotent suppressors SUP1(SUP45) and SUP2(SUP35) with non-mendelian
factors. Genetics 128: 513–520.
52. Derkatch IL, Uptain SM, Outeiro TF, Krishnan R, Lindquist SL, et al. (2004)
Effects of Q/N-rich, polyQ, and non-polyQ amyloids on the de novo formation of
the [PSI+] prion in yeast and aggregation of Sup35 in vitro. Proc Natl Acad
Sci U S A 101: 12934–12939.
53. Allen KD, Chernova TA, Tennant EP, Wilkinson KD, Chernoff YO (2007)
Effects of ubiquitin system alterations on the formation and loss of a yeast prion.
J Biol Chem 282: 3004–3013.
54. Pagano M (1997) Cell cycle regulation by the ubiquitin pathway. FASEB J 11:
55. Al-Hakim A, Escribano-Diaz C, Landry MC, O’Donnell L, Panier S, et al.
(2010) The ubiquitous role of ubiquitin in the DNA damage response. DNA
Repair 9: 1229–1240.
56. Singh RK, Kabbaj MH, Paik J, Gunjan A (2009) Histone levels are regulated by
phosphorylation and ubiquitylation-dependent proteolysis. Nat Cell Biol 11:
57. Sleep D, Finnis C, Turner A, Evans L (2001) Yeast 2 microm plasmid copy
number is elevated by a mutation in the nuclear gene UBC4. Yeast 18: 403–421.
58. King MA, Hands S, Hafiz F, Mizushima N, Tolkovsky AM, et al. (2008)
Rapamycin inhibits polyglutamine aggregation independently of autophagy by
reducing protein synthesis. Mol Pharmacol 73: 1052–1063.
59. Wexler NS, Lorimer J, Porter J, Gomez F, Moskowitz C, et al. (2004)
Venezuelan kindreds reveal that genetic and environmental factors modulate
Huntington’s disease age of onset. Proc Natl Acad Sci USA 101: 3498–3503.
60. Krammer C, Kryndushkin D, Suhre MH, Kremmer E, Hofmann A, et al.
(2009) The yeast Sup35NM domain propagates as a prion in mammalian cells.
Proc Natl Acad Sci USA 106: 462–467.
61. Olzscha H, Schermann SM, Woerner AC, Pinkert S, Hecht MH, et al. (2011)
Amyloid-like aggregates sequester numerous metastable proteins with essential
cellular functions. Cell 144: 67–78.
62. Chernoff YO, Galkin AP, Lewitin E, Chernova TA, Newnam GP, et al. (2000)
Evolutionary conservation of prion-forming abilities of the yeast Sup35 protein.
Mol Microbiol 35: 865–876.
63. Longtine MS, McKenzie A, 3rd, Demarini DJ, Shah NG, Wach A, et al. (1998)
Additional modules for versatile and economical PCR-based gene deletion and
modification in Saccharomyces cerevisiae. Yeast 14: 953–961.
64. Sherman F (2002) Getting started with yeast. Methods Enzymol 350: 3–41.
65. Halfmann R, Lindquist S (2008) Screening for amyloid aggregation by Semi-
Denaturing Detergent-Agarose Gel Electrophoresis. J Vis Expdoi 10.3791/838.
66. Miller JH (1972) Experiments in Molecular Genetics. Cold Spring Harbor, NY:
Cold Spring Harbor Laboratory.
67. Lemoine FJ, Degtyareva NP, Lobachev K, Petes TD (2005) Chromosomal
translocations in yeast induced by low levels of DNA polymerase a model for
chromosome fragile sites. Cell 120: 587–598.
68. Le Goff C, Zemlyanko O, Moskalenko S, Berkova N, Inge-Vechtomov S, et al.
(2002) Mouse GSPT2, but not GSPT1, can substitute for yeast eRF3 in vivo.
Genes Cells 7: 1043–1057.
69. Ter-Avanesyan MD, Kushnirov VV, Dagkesamanskaya AR, Didichenko SA,
Chernoff YO, et al. (1993) Deletion analysis of the SUP35 gene of the yeast
Saccharomyces cerevisiae reveals two non-overlapping functional regions in the
encoded protein. Mol Microbiol 7: 683–692.
70. Serio TR, Cashikar AG, Moslehi JJ, Kowal AS, Lindquist SL (1999) Yeast prion
[psi+] and its determinant, Sup35p. Methods Enzymol 309: 649–673.
71. Laney JD, Hochstrasser M (2003) Ubiquitin-dependent degradation of the yeast
Mata2 repressor enables a switch in developmental state. Genes Dev 17:
72. Firoozan M, Grant CM, Duarte JA, Tuite MF (1991) Quantitation of
readthrough of termination codons in yeast using a novel gene fusion assay.
Yeast 7: 173–183.
73. Storici F, Resnick MA (2006) The delitto perfetto approach to in vivo site-directed
mutagenesis and chromosome rearrangements with synthetic oligonucleotides in
yeast. Methods Enzymol 409: 329–345.
74. Sikorski RS, Hieter P (1989) A system of shuttle vectors and yeast host strains
designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics
75. Newnam GP, Birchmore JL, Chernoff YO (2011) Destabilization and recovery
of a yeast prion after mild heat shock. J Mol Biol 408: 432–448.
76. Allen KD, Wegrzyn RD, Chernova TA, Me `uller S, Newnam GP, et al. (2005)
Hsp70 chaperones as modulators of prion life cycle: novel effects of Ssa and Ssb
on the Saccharomyces cerevisiae prion [PSI+]. Genetics 169: 1227–1242.
77. Narayanan V, Mieczkowski PA, Kim HM, Petes TD, Lobachev KS (2006) The
pattern of gene amplification is determined by the chromosomal location of
hairpin-capped breaks. Cell 125: 1283–1296.
78. Argueso JL, Westmoreland J, Mieczkowski PA, Gawel M, Petes TD, et al. (2008)
Double-strand breaks associated with repetitive DNA can reshape the genome.
Proc Natl Acad Sci USA 105: 11845–11850.
79. DeRisi JL, Iyer VR, Brown PO (1997) Exploring the metabolic and genetic
control of gene expression on a genomic scale. Science 278: 680–686.
80. Wang P, Kim Y, Pollack J, Narasimhan B, Tibshirani R (2005) A method for
calling gains and losses in array CGH data. Biostatistics 6: 45–58.
Poly-Q Toxicity and Release Factors
PLoS Genetics | www.plosgenetics.org 13April 2012 | Volume 8 | Issue 4 | e1002634