Aneuploidy causes proteotoxic stress in
Saccharomyces cerevisiae.
ARtCHNEU
By
MASSACHUSCE
Ana Belen Oromendia
B.S. Biochemistry
University of Minnesota- Twin Cities
ETTYftg
TJUN 3 0 2014
LIBRA RIES
SUBMITTED TO THE DEPARTMENT OF BIOLOGY IN PARTIAL
FULLFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY IN BIOLOGY
AT THE
MASSACHUSSETTS INSTITUTE OF TECHNOLOGY
JUNE 2014
( Ana B. Oromendia. All rights reserved.
The author hereby grants to MIT permission to reproduce
and to distribute publically paper and electronic
copies of this thesis document in whole or in part in
any medium now know or hereafter created
Signature of author:
Certified by:
Signature redacted
I
Signature redacted
"'
Accepted by:
Department of Biology
June, 2014
I
1
A
Angelika Amon
Professor of Biology
Thesis Supervisor
Signature redact d
Michael Laub
Professor of Biology
Chair, Committee for Graduate Students, Microbiology Graduate Program
1
Aneuploidy causes proteotoxic stress in
Saccharomyces cerevisiae.
By
Ana Belen Oromendia
Submitted to the Department of Biology
on May 1", 2014 in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy in Biology
ABSTRACT
Gains or losses of entire chromosomes lead to aneuploidy, a condition tolerated
poorly in all eukaryotes analyzed to date. How aneuploidy affects organismal and cellular
physiology is only beginning to be understood. Aneuploidy also has a profound impact on
human health; it is the leading cause of mental retardation and spontaneous abortions and a
key characteristic of cancer, as more than 90% of all solid human tumors have aneuploid
genomes. Systematic analyses of aneuploid yeast and mouse cells suggested that aneuploidy
causes chromosome-specific effects elicited by the amplification of specific genes and
general aneuploidy-associated phenotypes Here I describe a phenotype that is shared by
most if not all aneuploid yeast cells- I find that aneuploid budding yeast cells are under
proteotoxic stress. I show that aneuploid strains are prone to aggregation of endogenous
proteins as well as of ectopically expressed hard to fold proteins such as polyQ stretchcontaining proteins. Prion conversion rates are also increased in most aneuploid yeast strains.
Protein aggregate formation in aneuploid yeast strains is likely due to limiting protein quality
control systems, since I present data showing that at least one chaperone family, Hsp90, is
compromised in many aneuploid strains. The link between aneuploidy and the formation
and persistence of protein aggregates has important implications for diseases such as cancer
and neurodegeneration.
Thesis Supervisor: Angelika Amon
Title: Professor of Biology
2
This thesis is dedicated with much love and admiration to Ana Maria Vigliocco.
"Ifyou are not lost you are at a place that someone else has aheady found..."
Junot Diaz
3
Acknowledgements
During the course of this thesis, many people have been invaluable in their support,
advice and encouragement. First and foremost, I would like to thank my advisor Angelika
Amon. Angelika: it has been a privilege to learn how to think about science from you- I
couldn't have asked for a better scientific role model. The Amon Lab was an amazing place
to learn how to be a scientist and I will take away with me many memories forged in the old
CCR and the KI building. Luke, Elcin, Matt, Leon, Folkert and Stefano were an invaluable
source of knowledge and technical expertise and always willing to discuss data when I
needed a sounding board. I would especially like to thank Jeremy and Michelle for teaching
me, and Stacie, Megan and Juliann for teaching me how to teach. Sarah was ever so patient
in helping me learn how to work with mammalian cells and scientific discussions with her
are some of my greatest memories. I'm thankful for the many fun times and wonderful
friendships I forged with Michelle, Sarah, Kristin, Stacie, Megan, Elcin, Luke Matt and
Jeremy.
A huge thank you goes to my committee members: Frank Solomon and Susan
Lindquist. Your input and support was greatly appreciated. I would also like to thank Randy
King for participating in my defense.
I am incredibly grateful to David Schauer and Alan Grossman for starting the
Interdepartmental Microbiology Graduate Program @ MIT and for including me in the
founding class. Their dedication to the program, and to my success while at MIT had no
bounds. I would especially like to thank Frank Solomon and Alan Grossman for their
continuous encouragement- thank you for being straightforward and honest and for
believing in me every step of the way; I cannot explain how much it has meant to me.
A very special thank you goes to the Massachusetts General Hospital and the
wonderful doctors and nurses there, in particular Dr. Christopher Oglivy and Dr. Patricia
Musolino. I truly could not have done this without you.
The support I have received from many friends during the last years both in the
form of lengthy conversations and late night drinks has been invaluable. I am better for
having had you in my life. Heather, Marina, Caro, Meche, Jordan and Cristian- thank you for
being my family in Boston. I would especially like to thank the MIT Micro dudes:Ben and
Tyler, who have been here with me from the very beginning. Saydi, there are no words for
how much your love and support has buoyed me through, thank you for always being my #1
cheerleader!
Finally, I would like to thank my family. You, and your unrelenting support and love
that has no bounds means more than I can ever explain. I thank you for encouraging the
curiosity and creativity that has led me to pursue science. To my siblings Mercedes, Clara,
Milagros and Manuel- you are my best friends, and your unwavering encouragement has
kept me going all these years. To my mom and dad: I am so incredibly grateful for all the
sacrifices you have made to give us choices; thank you for always being on my side and
encouraging my dreams. To Ana V: you inspire me as a scientist and as a person- I hope that
when I grow up I can be half the person you are.
4
Table of contents
ABSTRACT
2
ACKNOWLEDGEMENTS
4
TABLE OF CONTENTS
5
CHAPTER 1:
7
ANEUPLOIDY DISRUPTS CELLULAR BALANCE
8
Genome maintenance
Comparison between aneuploidy and polyploidy
Origins of whole-chromosome aneuploidy
Saccharomyces cerevisiaemodels of aneuploidy
Cellular consequences of aneuploidy
Aneuploidy results in reduced proliferation
Transcriptional response to aneuploidy
Aneuploidy results in proteome alterations
PROTEIN QUALITY CONTROL MAINTAINS THE PROTEOME
Protein Folding
Controlling Protein Aggregation
Protein Degradation
Cellular responses to acute proteotoxic stressors
ANEUPLOIDY, PROTEIN QUALITY CONTROL AND DISEASE
8
9
9
14
19
21
22
23
29
29
35
36
37
38
Aneuploidy in Cancer
Whole-organism aneuploidy
Aneuploidy and Neurodegeneration
39
40
43
Aneuploidy and aging
43
Concluding Remarks
44
References
46
CHAPTER 2:
50
Introduction
51
Results
53
Disomic yeast strains harbor a higher load of endogenous protein aggregates.
53
Adaptation to proteotoxic stress is delayed in disomic yeast strains.
57
Meiotic and mitotic chromosome mis-segregation leads to protein aggregate formation.
68
Aneuploid strains fail to efficiently fold the protein quality control sensor VHL.
Loss of UBP6 reduces aggregate burden in disomic yeast strains.
71
75
5
76
Hsp90 folding capacity is reduced in many disomic yeast strains.
Aneuploid strains are more susceptible to protein aggregates associated with human
disease.
80
Discussion
Why are aneuploid cells aggregate-prone?
Aneuploidy in cancer and neurodegenerative diseases.
87
88
90
Materials and Methods
91
Strains used in this study. All straisn are of the W303 background
97
References
107
CHAPTER 3:
111
Summary of key conclusions
112
Aneuploidy exhausts the cell's protein quality control capacity
Why are aneuploid cells aggregate-prone?
The folding capacity of chaperones is altered by genomic imbalances
Aneuploidy is a chronic stress, distinct form environmental proteotoxic stressors
114
114
118
119
The composition of protein aggregates in aneuploid yeast
122
Aneuploidy in mammalian cells alters protein quality control
127
Interface between aneuploidy, aging and neurodegeneration
129
References
132
6
Chapter 1:
Introduction
Sections of this introduction have been reproduced with permission from DMM
Oromendia, A and Amon, A 'A neuploidy: implicationsfor protein homeostasis and disease' DMM, in
press 2013
7
Homeostasis is at the crux of biology. Cells must maintain their karyotipic integrity
and, at the same time, ensure the maintenance of their proteome even when faced by
stressful growth conditions. I have found that the disruption of a balanced karyotype, i.e.
aneuploidy results in a disruption in protein homeostasis. This Introduction will expand first
on the consequences of aneuploidy, then on the cellular mechanisms that maintain protein
homeostasis and finally explore the interactions they share in the context of human disease.
ANEUPLOIDY DISRUPTS CELLULAR BALANCE
Genome maintenance
The maintenance of stable karyotype, i.e. number and identity of chromosomes, is
essential to the success of all species. Species exist with varying chromosomal copies, from
haploid (1 copy of each chromosome) to the most common diploid (2 copies) but some
plant species can have up to 12 copies of each chromosome. Regardless of ploidy, all
organisms carry an equal number of each chromosome ensuring a balanced genome in
which genes encoded on different chromosomes are present in the same number of copies.
It is this balance that gets disrupted in aneuploid cells.
Aneuploidy, defined as a karyotype that is not a whole multiple of the genomic
complement results in an 'unbalanced' genome in which chromosomes(s), or pieces of
chromosome(s)
are missing or supernumerary
and thus genes present on different
chromosomes are present in varying copy numbers. Several studies have now shown that
gene copy number is well correlated with gene expression and, for the most part, well
correlated with protein abundance- an imbalance in copy number results in an imbalance of
gene products that aneuploid cells are burdened with. Aneuploidy is generally not well
8
tolerated in nature, giving rise to developmental abnormalities of aneuploid organisms and
the impaired fitness of aneuploid cells in all species studied to date (reviewed in (Williams
and Amon 2009, Torres, 2008).
Comparison between aneuploidy and polyploidy
Whereas aneuploidy results in an unbalanced, abnormal number of chromosomes
and is poorly tolerated in nature, polyploidy does not. Polyploidy is a condition in which
cells contain a non-cognate, but balanced number of chromosomes- i.e. cells that of a
species that normally maintains a 2n karyotype being tetraploid (4n). Since the relative ratio
between gene products is maintained, there is no imbalance for the cell to contend with.
Polyploidy, to a degree, is well-tolerated and there are many well documented cases of cells
intentionally becoming polyploid to perform their function, such as human megakaryocytes
and Drosophila melanogaster salivary gland cells (Lacroix and Maddox 2012). It is clear that
while there is an optimal karyotype that each species has evolved to have, modifications that
alter chromosome number but maintain genomic balance are far less detrimental than those
that generate genomic imbalanceby altering the copy number of only a subset of
chromosomes.
Origins of whole-chromosome aneuploidy
During the course of cell division cells must replicate their DNA and then segregate
it equally so that each daughter cell maintains the same chromosomal content as the mother
cell. The cell employs a number of mechanisms to ensure that chromosome segregation has
occurred before cell division concludes. The process of chromosome segregation begins
when the replicated sister chromatids are linked via cohesin molecules. During prophase,
9
each pair of sister chromatids forms attachments to the mitotic spindle so that each
chromatid's kinetochore is attached to opposing spindle poles via microtubules.
In metaphase, sister chromatids are attached to opposing spindle poles and under
tension from pulling forces of rnicrotubules and cohesin molecules holding them together;
they are said to be bi-oriented. For accurate chromosome segregation, it is essential to
prevent cell cycle progression until all sister chromatid pairs are bi-oriented. The Spindle
Assembly Checkpoint (SAC) monitors chromatid attachment and tension and halts the cell
cycle until all sister chromatids are properly attached to the mitotic spindle. Once all of the
chromatids are appropriately attached, Separase cleaves the cohesin molecules and allows the
pulling microtubules to segregate individual chromatids to opposing poles (Figure 1).
Figure 1: The Spindle Assembly Checkpoint (SAC) ensures accurate chromosome
segregation
(a) Cohesion between sister chromatids is retained through metaphase until all attachments
to the spindle have been properly made. At the metaphase to anaphase transition, APCCDC20 stimulates the degradation of the inhibitory protein Securin, the degradation of
Securin frees Separase to cleave Cohesin. As the chromatids are attached to opposite spindle
poles and under tension, they move away from the metaphase plate as the spindle elongates.
(b) When chromosomes are not attached, or improperly attached to the spindle, there is a
lack of tension. It is this lack of tension, detected, in part, by the kinase Aurora B that
activates the Spindle Assembly Checkpoint. MAD2, along with other players, prevents the
ubiquitination of Securin by the APC-CDC20. Securin maintains Separase inactive, pausing
cell cycle progression. Bypass of the SAC can lead to progression through the cell cycle with
improper chromosome attachments resulting in aneuploidy.
10
Figure 1
A
Correct attachments
B
Tension
SAC OFF
Securi
Separase
incorrect attachments
No tension -
SAC ON
-'
AuroraB
MAD2
CDXSecurin
_.............................
......
r
Separase
Compromised SAC function or mis-regulated Separase activity invariably leads to
whole-chromosome aneuploidy because the cell cycle is not arrested in cells with unattached
or mis-attached chromosomes (Figure 2a). Defects in chromatid cohesion also result in
aneuploidy- each chromatid can segregate as it attaches to a microtubule, resulting in almost
random chromosome segregation (Figure 2b).
Chromatids can also form aberrant
kinetochore attachments that are difficult for the SAC to detect. Merotely, when a single
sister chromatid kinetochore is attached to microtubules from both spindle poles, is
especially difficult to detect as there is still ongoing tension. Often, these resolve by anaphase
and do not result in aneuploidy (Thompson and Compton 2008) (Thompson and Compton
2011) but when unequal merotelic attachments occur (kinetochore attached to more
microtubules emanating from one pole than from the other), aneuploidy is thought to ensue
(Figure 2c)
11
Errors in chromosome segregation in meiosis result in the creation of aneuploid
gametes, which can then lead to whole-organism aneuploidy. In mejosis, DNA replication is
followed by two rounds of chromosome segregation: first, in Meiosis I homologous
chromosomes segregate away from each other, and in Meiosis II sister chromatids segregate.
In order to accomplish these orchestrated segregation events, cells have altered the canonical,
mitotic
chromosome
segregation
program.
To
properly
segregate
homologues,
chromosomes undergo crossover events that physically link homologous chromosomes and
allow them to align at the Meiosis I metaphase plate, both sister kinetochores must also
coorient and attach to the same pole. Additionally, cohesion is lost in a stepwise manner,
with arm cohesion being lost first, to allow for homologue segregation in anaphase I and
centromere cohesion lost at a later stage to allow for sister chromatid segregation at
anaphase II. In metaphase II, sister kinetochores must bi-orient and attach to opposing poles
for sister chromatids to segregate to either pole (reviewed in (Miller et al. 2013)). Failure in
any of several meiotic chromosome segregation events can lead to mis-segregation, including
premature sister chromatid separation, failure to establish crossovers between homologous
chromosomes in Meiosis I and various chromosome attachment defects in either Meiosis I
or Meiosis II (Figure 2d).
Errors in chromosome
segregation can arise via many different means, and
understanding the consequences of these events on cellular physiology is of critical
importance. Aneuploidy has been shown to have severe consequences and to be detrimental
in most cases studied to date.
12
FIGURE 2: Whole chromosome aneuploidy arises through errors in mitosis or
meiosis (adapted from J. Siegel and Amon 2011)
Cells missegregate chromosomes in mitosis by: (a) mutations in the Spindle Assembly
Checkpoint (SAC) in which mis-attached kinetochores do not trigger a cell-cycle arrest, (b)
premature loss of sister chromatid cohesion where sister chromatids attach to spindle poles
and segregate randomly, and (c) merotelic attachments in which a single kinetochore attaches
to microtubules emanating from both poles. (e) Aneuploidy can also arise from errors in
chromosome segregation in either Meiosis I or Meiosis II.
Figure 2
A
Spindle Assembly
Checkpoint Mutations
B
Pre-mature Loss of
Chromatid Cohesion
C
Aberrant Kinetochore
Attachments
[ED
E I IC -c
*1
D
.
Meiotic Segregation
Errors
d
mis-segregation
durin meiosisi
Lurin
0
'
eiosis~
EN
[Nx
13
Saccharomyces cerevisiae models of aneuploidy
In this thesis I have used aneuploid Saccharomyces cerevisiae strains of various
karyotypes generated via three different methods (Figure 3). I generated highly aneuploid,
highly genomically unstable strains via triploid meiosis and using mutants that readily missegregate chromosomes during mitosis. Additionally, I used a set of stably aneuploid strains
that carry one extra chromosome that were generated by direct chromosome transfer. Using
this wide panel of aneuploid strains, I was able to ensure that the phenotypes observed are
not due to any particular karyotype nor to the method via which they were constructed; I am
confident that the phenotypes observed in the majority of the strains are consequences of
being aneuploid.
Saccharomyces cerevisiae strains that carry large, random, genomic imbalances were
created by inducing missegregation of chromosomes either in meiosis or in mitosis. I created
a triploid strain (3n, genotype a/a/a), induced it to undergo meiosis via starvation and
recovered the meiotic products (Figure 3a). Triploid cells induced to undergo meiosis
produce highly aneuploid progeny, with karyotypes ranging from diploid to highly aneuploid
(St Charles et al. 2010). The majority of the aneuploid progeny is inviable (Parry and Cox
1970), but some genetically unstable aneuploid strains can be obtained (Pavelka et al. 2010b)
(Sheltzer et al. 2011) (Zhu et al. 2012). As colony formation is a prerequisite for the recovery
of these strains, the aneuploidies that cause severe growth defects and do not form colonies
will not be analyzed. This approach to generating aneuploid strains is beneficial in that it
rapidly allows one to generate a pool of strains with high karyotype variability, as these cells
are highly unstable, one is limited to colony formation or single cell assays and must take
into account that the analysis will be biased towards 'healthier' aneuploidies that do not
impinge greatly on colony formation or growth.
14
Figure 3: Generating aneuploid Saccharomyces cerevisiae strains (Adapted from
(Siegel and Amon 2012))
Triploid strains induced to undergo meiosis produce highly aneuploid progeny (a). Using the
abortive matings of the karyogamy defective karl,15 strain, aneuploid strains can be
generated by single chromosome transfer and selection using markers placed at the same
locus on both chromosomes (b). Mitotic chromosome mis-seggregation can be induced by
shifting strains carrying temperature sensitive alleles of Iptl or Ndc1O (c)
Figure 3
B
k;
C 1 CN0i
15
N
[A
25 "C
selection~
N
{
marker 1
selecion
marker 2
_
Seieotionr 1 & 2
15
One can also generate random aneuploidies by inducing chromosome missegregation
during mitosis (Figure 3c). Strains harboring temperature-sensitive alleles of genes encoding
the kinetochore component Ndc10 or the SAC component Aurora B kinase, Ipli can be
arrested in G1 under permissive growth conditions, and induced to mis-segregate
chromosomes by shifting them to semi-permissive growth conditions. This treatment results
in dramatic chromosome mis-segregation, with 29-35% of cells being unable to correctly
segregate a chromosome that is marked by integrating a tandem array of tetO sequences. As
these strains also carry a TetR-green fluorescent protein (GFP) fusion, one can visualize the
tetR arrays and by extension, track chromosome segregation (GFP-dots) (Oromendia et al.
2012). As with aneuploid strains generated by meiotic chromosome mis-segregation, these
strains are highly unstable and are best employed for single cell assays or genetic synthetic
interaction analysis with other mutant strains.
In order to more carefully characterize aneuploidy and perform population based
assays, our lab developed a set of haploid yeast that carry an extra copy of one additional
chromosome ((Torres et al. 2007), Figure 3b); these strains have an n+1 karyotype and will
be referred to as disomes in this thesis. These disomic yeast strains with defined karyotypes
were generated via chromosome transfer from a donor cell to a recipient cell (Figure 4).
Disomic
strains are low-complexity
aneuploidies
(only carrying one supernumerary
chromosome) but, by adding selectable markers at the same locus in both copies of the
disomic chromosome, one can use double selection methods to ensure a stably propagating,
pure population of an aneuploid strain with a defined karyotype. These strains have proven
to be invaluable in understanding the effects of aneuploidy on cellular physiology, but due to
the method in which they are generated one can only create low-complexity (one or two
extra chromosomes) aneuploidies.
16
To comprehensively study the effects of aneuploidy on cellular physiology, I have
generated aneuploid strains in various different manners. I used strains that carry stable, lowcomplexity aneuploidies and unstable high-complexity aneuploid strains, strains resulting
from mitotic or mitotic chromosome mis-segregation and strains that can be maintained as
aneuploid via selection. Using this wide panel of aneuploidies I hope to elucidate the general
consequences that aneuploidy has on a cell.
Figure 4: Generating aneuploid strains via failed karyogamy matings (Adapted from
Torres, et al 2007)
Strains carrying extra chromosome were generated by a chromosome transfer strategy
described by Hugerat et al. (Hugerat and Simchen 1993) A HIS3 cassette is integrated at a
particular location on each chromosome using the PCR-based method described by
Longetine et al. (Longtine et al. 1998) The strain is then mated to a strain carrying the
karlA15 allele, which renders the strain defective in karyogamy (STEP 1).b In addition the
strain carries the cyb2-Q37E allele, which confers resistance to cycloheximide in a recessive
manner. The mating mixture was then plated on medium lacking histidine and containing
3pg/ml cycloheximide to select for the marked chromosome and to select against diploids
and heterokaryons. karz1l5 cells carrying the HIS3 marked chromosome were then mated
to cells that carried the kanMX6 cassette at the same genomic locus where the HIS3 was
integrated (STEP 2). This strain also carries the cani-100 allele, which confers resistance to
canavinine in a recessive manner. Matings were performed and the mating mixture was
plated on medium containing G418 and lacking histidine to select for the presence of the
disome. To select against mating events the medium also contained canavanine.
17
Figure 4
Step I
Mata, xxx::HIS3, LYS2, CYCH2, can 1-100
Mato, karlA15, lys2-801, cyh2-Q37E
xxx-,HIS3
Select for: CycR and -His
xxx:HIS3
Step 2
Mata, xxx::HIS3, LYS2, CYCH2, can 1-100
Mata, kar1A15, lys2-801, cyh2-Q37E
xxx::.kanMX6
xxx..HIS3
xxx HIS3
xxx HIS3
xxx katpMX6
Select for: CanR, -His and KanR
xxx kanMX6
18
Cellular consequences of aneuploidy
Systematic analyses of aneuploid yeast, mouse and human cells and studies on cancer
cell lines suggest that aneuploidy causes chromosome-specific effects that are elicited by the
increased (or decreased) number of copies of individual genes and/or combinations of a
small number of genes present on the aneuploid chromosome (Tang and Amon 2013).
Changes in the gene copy number of regulators of gene expression lead to further disruption
of cellular function. Surprisingly, recent studies have shown that aneuploidy also causes
chromosome-independent effects, which are a not a consequence of any specific gene
imbalance
but general consequences of harboring an unbalanced karyotype. These
phenotypes include a cell cycle delay in G1 (Torres et al. 2007; Stingele et al. 2012b;
Thorburn et al. 2013), metabolic alterations (Williams et al. 2008; Pavelka et al. 2010a),
genomic instability (Sheltzer et al. 2011; Zhu et al. 2012) and proteotoxicity (Torres et al.
2007; Tang et al. 2011; Oromendia et al. 2012; Stingele et al. 2012b) (Figure 5).
Understanding the origins of these phenotypes is important as this could provide insights
into how chromosome mis-segregation and the resulting imbalanced karyotype impacts
normal cell physiology and disease states. I describe a subset of these phenotypes in more
detail below.
19
Figure 5: Observed characteristics of aneuploid cells in yeast (a) and mammalian
cells (b). Adapted from (Siegel and Amon 2012). Blue boxes show observed physiological
stresses and pink boxes show conditional changes resulting from aneuploidy.
Figure 5
A
Yeast
Insabilenomic
ncreased Protein
Synthesis
Aneuploidy
(ProteinMetabolic
CProtein
Imbalance
r
Alterations
GSow
Energy Stress
roteotoxic
C Stress
nvironmental Stres
Response
R
Mammalian CE
An )uploidy
Metabolic
Alterations
increased
Hsp72 j
,Reactive
Oxygin
se dAutophag
----
E nergy Stres
S pecies
ATM Acvatio
Lp38 Activatio
~iva~on
+
20
A neuploidy results in reducedprolferation
Among the most prevalent and key phenotypes of aneuploid cells is their slower
proliferation relative to that of euploid cells. First described in fibroblasts derived from
individuals with Down's Syndrome (Segal and McCoy 1974), we now know that this
phenotype is a general consequence of aneuploidy. Thorough studies in aneuploid
Saccharomyces cerevisiae harboring an extra copy of one or two chromosomes (Torres et al.
2007) or derived from triploid meiosis (Pavelka et al. 2010b) and in SchiZosaccharomycespombe
aneuploid cells derived from triploid meiosis (Niwa et al. 2006) showed that, irrespective of
which chromosomes are present in excess, aneuploidy results in impaired proliferation.
Similarly, aneuploid MEFs containing an extra copy of chromosomes 1, 13, 16 or 19
(Williams et al. 2008) exhibit proliferation defects, and MEFs derived from mice carrying a
hypomorphic allele of the SAC component BUBR1 show slower proliferation than euploid
MEFs at later passages when aneuploidies are allowed to accumulate (Baker et al. 2004).
Aneuploid cells obtained by inducing meiotic non-disjunction, MEFs harboring mutations in
SAC component Bubi, or mutations that render the checkpoint component Cdc20 non
functional also exhibit proliferation defects and are outcompeted by euploid cells in growth
assays (Thompson and Compton 2008; Li et al. 2009).
The slow growth phenotype of aneuploid cells has been most extensively studied in
S. cerevisiae, in which a recent study has found that aneuploidy results in an extended G1
phase and a delay into entry of the cell cycle that correlates well with the size of the
supernumerary chromosome (Thorburn et al. 2013). This phenotype is dependent on the
proteornic consequences of aneuploidy, as strains carrying chromosome sized human DNA
fragments that can be replicated but do not produce any protein do not display a G1 delay.
Both cell growth (cell volume accumulation) and entry into the cell cycle appear to be
21
affected. Although most disomic yeast strains show a cell growth defect, there appear to be
no gross defects in global protein synthesis as measured by polysome profiling or
['S]methionine incorporation(Thorburn et al. 2013),although the effects of aneuploidy on
these processes may be too subtle to detect by these methods. The growth defect does not
appear to be due to diminished amino acid pools or reduced translational efficiency. 10 out
of 14 disomic strains analyzed showed a delay in cell cycle entry observed as an increase in
critical size (the size at which 50% of cells in a population have budded). All of the strains
analyzed show delayed accumulation of the G1 cyclin CLN2 mRNA, and it was shown that
high levels of CLN2 suppress the increase in critical size. Accumulation of Cln3, another G1
cyclin was also delayed in all disomes analyzed (Thorburn et al. 2013). It is yet unclear how
aneuploidy interferes with the accumulation of Cln3 and whether this is a gene specific effect
or a general response to aneuploidy. As it has been observed in almost all aneuploid strains, I
favor the idea that the G1 delay is a general consequence of aneuploidy. Interestingly, many
environmental stresses (including heat stress) have been shown to cause a transient G1
delay- it is possible that proteotoxic stress in aneuploid yeast is contributing to the G1 delay
observed.
Transcriptionalresponse to aneuploidy
Several lines of evidence suggest that cells respond to the aneuploid state. Most
aneuploid cells studied to date exhibit a transcriptional signature associated with slow growth
and stress (Torres et al. 2007; Sheltzer et al. 2012; Stingele et al. 2012b; Foijer et al. 2013).
Recent studies have shown that aneuploidy elicits a transcriptional response reminiscent of
the environmental stress response (ESR) in species as divergent as budding and fission yeast,
Arabidopsis thaliana, and human and mouse cell lines. The ESR consists of -300 genes that
22
are upregulated and ~600 genes that are downregulated by various exogenous stresses,
including heat shock or oxidative stress (Gasch et al. 2000). Most of these genes also vary in
expression in response to growth rate; inducing slow proliferation by nutrient limitation
mimics the ESR (Regenberg et al. 2006; Brauer et al. 2008). The high correlation between the
ESR-like response seen in aneuploid cells and the transcriptional response observed in slowgrowing S cerevisiae strains suggests that the transcriptional response observed in aneuploid
cells is, for the most part, due to the slow proliferation observed in aneuploidy (Sheltzer et al.
2012).
A neuploidy results in proteome alterations
The unbalanced genome caused by aneuploidy has been shown to translate into an
unbalanced proteome - that is to say that the changes in gene dosage for the most part result
in equivalent changes in protein levels (twice as much DNA results in twice as much protein,
Figure 5). Studies of Saccharomyces cerevisiae aneuploid strains show that the abundance of
approximately 80 percent of proteins changes in proportion to gene copy number (Pavelka
et al. 2010b; Torres et al. 2010). Interestingly, many of the proteins for which this is not true
are subunits of multimeric complexes (Torres et al. 2007). Indeed, often times, subunits that
are endogenously expressed in excess because of aneuploidy retain stoichiometric numbers
within multimeric complexes (Torres et al. 2007). Stingele and colleagues showed that this is
also true in human aneuploid cells {Stingele, 2012 #1121; Torres et al. 2010). Analysis of the
transcriptome and proteome of aneuploid human cells generated by chromosome transfer
showed that most genes are expressed according to their copy number, and proteins are
translated in strong correlation with the abundance of mRNA, resulting in a dramatic change
in cellular protein composition (Stingele, 2012 a).
23
Figure 6: DNA, mRNA and protein levels in yeast disomic for chromosome V (Data
from Torres et al 2007).
Disomic S. cerevisiae strains carry an active, replicating chromosome in one additional copy as
evidenced by comparative genome hybridization (CGH, top panel). The extra chromosome
is transcribed, as seen by the two-fold increase in mRNA present form that chromosome
(microarray, middle panel). The majority of the proteins encoded by the chromosome are
also expressed and can be found at close to 2 fold higher levels than those encoded by other
chromosomes (SILAC, bottom panel)
Figure 6
DISOME V
43-
DNA
o
.
2
(CGH)
-2 !hr i Chr V
-3
(Aray)
4-
-1
3
(RNA
-2
shr I Chr V
4.
PROTEIN
20
(SILAC)
1t
-2
br l ChrV
SChromosome Position
24
However, as in aneuploid yeast, human aneuploid cells were also found to maintain a
subset of proteins (enriched for complex subunits) at stoichiometric levels even if gene copy
number was altered.
The regulatory mechanisms responsible for this correcting process
have not been elucidated. Overall, these data suggest that, although some proteins are
maintained at stoichiometric levels, there is no general whole-chromosome 'gene dosage
compensation' mechanism for autosomes in yeast and mammals, as has been observed for
sex chromosomes. This might not be the case in all organisms, however. Aneuploid
Drosophila S2 cells have been reported to experience
dosage compensation
at the
transcriptional level by means of the male-specific lethal (MSL) complex and general
compensation mechanisms that compensate for differences in non-autosomal chromosome
copy number (Zhang et al. 2010). Further studies in Drosophilaaneuploid cells are needed to
determine the status of their proteome.
A key question resulting from the profound effects of aneuploidy on cellular protein
composition is whether the simultaneous changes in the relative ratios of many proteins
impacts upon the protein quality-control pathways of the cell. Chaperones and the
degradation machinery, the 26S proteasome, proteases and autophagy, ensure that all
proteins acquire their native conformation and prevent cellular toxicity by reducing the
number of aberrant interactions between proteins. In aneuploid cells, these protein qualitycontrol systems must not only attend to the excess proteins produced from additional
chromosomes, they must also support all excess subunits of complexes that are not in
stoichiometric ratios with their binding partners (Figure 7).
25
Figure 7. Aneuploidy causes proteotoxic stress. (a) Cells use protein quality-control and
feedback mechanisms to maintain subunit stoichiometries of complexes whose subunits are
encoded by different chromosomes. The protein quality-control (QC) machinery ensures
accurate folding and maintains complex subunits that lack a binding partner in a soluble state.
Eventually, excess and misfolded subunits must be degraded, as illustrated here by the yellow
subunit that has been produced in relative excess. (b) Changes in chromosome number in
aneuploid cells (shown here as disomy of the green chromosome) lead to a genomic
imbalance that results in stoichiometric protein imbalances. Every subunit encoded by an
unbalanced chromosome that functions in a protein complex lacks its binding partner(s) and
must rely on cellular chaperones to maintain solubility and, if no binding partner is found, on
the cellular proteases for its eventual degradation. This can lead to an increased burden on
the protein quality-control systems and the exhaustion of the cellular protein quality-control
machinery.
26
Figure 7
A EUPLOID CELLS
Chromosomes
lA
I
Complex Subunits
B
C
Protein Complex
ABC
A
X
W
Chaperone
QC
B ANEUPLOID CELLS
Chromosomes
1
11
111
Complex Subunits
A
B
C
Protein Complex
ABC
V7~
Many protein complex subunits are unstable unless bound to their partners, and will
often bind to cellular chaperones to remain soluble until they have formed the complex
(Boulon et al. 2010). Several previous studies have indeed hinted to the fact that aneuploidy
impacts protein quality-control systems. Budding yeast, mouse and human aneuploid cells
exhibit a transcriptional signature that is reminiscent of a stress response and slow growth
(Torres et al. 2007; Sheltzer et al. 2012; Stingele et al. 2012a). This transcriptional signature
includes upregulation of protein chaperones (Sheltzer et al. 2012). Human aneuploidies
generated by chromosome transfer were found to have a transcriptional stress signature
that
shows up-regulation of lysosome-mediated degradation and p62-dependent autophagy
27
(Stingele et al. 2012a; Stingele et al. 2013).
Furthermore, many haploid S. cerevisiae strains
harboring an additional chromosome (disomic yeast strains) were found to be sensitive to
chemical compounds that impair protein quality control; many disomic yeast strains are
sensitive to the proteasome inhibitor MG132, the ribosome poison cycloheximide and the
Hsp90 inhibitors radicicol and geldanamycin
(Torres et al. 2007). Mouse embryonic
fibroblasts (MEFs) trisomic for any of chromosomes 1, 13, 16 or 19 are more sensitive to
the Hsp90 inhibitor 17-AAG than are wild-type MEFs (Tang et al. 2011). These results can
be interpreted in that that the aneuploid state causes proteotoxic stress leading aneuploid
cells to rely more heavily on their protein quality control machinery. Thus, impairing
chaperone function via use of chemical chaperone inhibitors is more detrimental to cells that
are aneuploid than to cells that carry the appropriate number of chromosomes. This thesis
directly tests this possibility.
28
PROTEIN QUALITY CONTROL MAINTAINS THE PROTEOME
At the core of cellular biology is the process of converting genetic information into
proteins that both carry out the genetic program and provide structural integrity to the cell.
The central dogma of molecular biology describes the lifecycle of each individual protein
subunit. Protein coding genes are perpetuated in the genome as DNA. When necessary, the
DNA is transcribed into mRNA molecules, which are then translated by the ribosome into
polypeptides. In order to be functional the majority of polypeptides must acquire a welldefined three-dimensional structure (native structure) and, in many cases, bind to other
protein subunits to form a functional protein complex. Once the protein is no longer
necessary, it is degraded into individual amino acids that can then be recycled and used in the
fabrication of new polypeptides. This process is highly dynamic and energetically costly and
at the same time, is affected by almost all external cellular stressors; thus the process of
maintaining protein homeostasis (or proteostasis) is one of extreme balance and precision.
Protein synthesis is tightly controlled in cells, but in addition, protein folding and protein
degradation play an important role in maintaining proteostasis.
Protein Folding
The information necessary to acquire the native structure is encoded in the primary
amino acid sequence and thus, many proteins can fold unassisted in dilute solutions in vitro.
In the cellular mileu where the total protein concentration can be as high as 300 mg per ml,
acquiring native structure is much more challenging. Inter molecular interations are strongly
favored in vivo and since folding intermediates often expose hydrophobic patches, the
crowded cellular environment endangers newly synthesized proteins and unstable proteins
with high propensity to misfold. Exposed hydrophobic regions constantly pose a threat and
29
non-productive interactions that can result in misfolding and/or aggregation compete with
the formation of the native structure.
Both protein misfolding and aggregation are
detrimental and pose a significant burden to the cell and defects in these processes can result
in human disease (Reviewed in (Young et al. 2004; Taipale et al. 2010; Tyedmers et al. 2010).
In order to maintain proteostasis and mitigate the effects of heat and other stresses
on the proteome, cells have evolved a sophisticated network of protein chaperones. Protein
chaperones are intricately involved in the folding and maturation of a protein - from a
polypeptide exiting the ribosome acquiring the appropriate three-dimensional structure, to
assembly into the appropriate complexes. Molecular chaperone proteins bind to folding
intermediates, reducing the conformational space that can be explored and often times
preventing aberrant interactions by sequestering hydrophobic patches. Chaperones exist in
several structurally unrelated classes and have been classified into families according to their
type of enzymatic activity, the co-chaperones they require and the clients that they aid in
folding (Hard et al. 2011) and they are named according to their molecular size. Often times,
a single polypeptide will interact with different chaperones sequentially, each aiding in a
specific aspect of protein folding or complex assembly. It is important to bear in mind that
each chaperone family usually has multiple distinct members in each cellular compartment
serving to both increase chaperone diversity and ensure redundancy. I will briefly discuss the
specifics of the HSP90, HSP70, HSP60 (chaperonins) and small heat shock protein (sHSPs)
families here (Figure 8).
30
Figure 8: Molecular Chaperone Mechanisms (adapted from (Richter et al. 2010))
Chaperone model: In general, proteins fold via increasingly structured intermediates (L, L)
from the unfolded state (U) to the folded state (N). Protein chaperones bind proteins in
nonnative conformations. The shift from the high-affinity binding state to the low-affinity
release state is often triggered by ATP binding and hydrolysis. Hsp60/GroE: The GroE
machinery consists of two identical rings that enclose a central cavity each. Nonnative
protein is bound by the apical domains of the rings, and upon binding of ATP and the
cochaperone GroES (caps), the protein is encapsulated and released into the cavity. ATP
hydrolysis in one ring results in the release of GroES and substrate protein from the
opposite ring. During encapsulation the protein may fold partially or completely. Hsp70:
The Hsp70 system comprises two cochaperones, an activating protein (Hsp40/J-protein)
and a nucleotide exchange factor (NEF). The activating protein can bind the nonnative
protein and deliver it to Hsp70 forming a complex and stimulating its ATPase. The NEF
will induce the exchange of nucleotide accelerating the ATPase cycle and the client protein is
released Hsp90: In this chaperone system a large number of proteins work together. Often,
Hsp70 delivers the substrates to Hsp90. Cochaperones (shown here in purple and yellow)
modulate the system (shown here in purple and yellow). ClpB/Hsp104: This chaperone is
able to dissolve aggregates by actively pulling proteins through a central channel of the
hexameric structure. Refolding occurs upon release, and, to some extent, it can also occur in
cooperation with other chaperones. sHsps: sHps are oligomeric complexes that are often
activated, by heat or modifications. Many are believed to dissociate into smaller oligomers to
become active. sHsps can bind many nonnative proteins per complex. Release requires
cooperation with other ATP-dependent chaperones such as Hsp70.
31
Figure 8
chaperone model
GroES/HspGQ
Hsp7O
ClpBIHsplO4
sHsp
*rr
lowE
"Wegh%
affnitaLnt
AW~PI
Hsp9O
Hsp9O is at the center of maintaining proteostasis, forming a hub that controls many
important signaling pathways (Figure 8, reviewed in (Tfaipale et al. 2010)). Hsp9O functions
downstream of Hsp7O binding partially folded polypeptides and cooperating with many cochaperones and regulatory subunits to ensure structural maturation. The activity of Hsp9O is
ATP-dependent and closely coordinated with environmental perturbations. Hsp9O is poised
to be a buffering force in protein quality control- under normal growth conditions, as it is
present in vast excess and can be reduced to 10% of its natural abundance without
detrimental consequences to the cell (McClellan et al. 2007; Franzosa et al. 2011). Hsp9O
client lists have been notoriously hard to define, perhaps because Hsp9O's essential folding
roles seem to be in folding proteins that are central hubs of cellular processes such as
32
regulatory subunits of signal transduction cascades or kinases. Recent genome wide studies
have implicated Hsp90 in almost every cellular process from protein trafficking, secretion,
RNA processing, signal transduction to telomere maintenance and immunity.
The constitutive and inducible forms of Hsp70 are core players in protein quality
control (Figure 8, reviewed in (Richter et al. 2010)). Hsp70 (DnaK in E. coli) functions in
concert with Hsp40 (DnaJ in E. coli) and nucleotide exchange factors to, in an ATPdependent manner, aid in folding of nascent polypeptides and bind and release partially
folded substrates. Hsp70 binds to substrates via small stretches of hydrophobic amino acids,
exchanging rapidly in an ATP bound state and binding stably to substrates and Hsp40 after
ATP hydrolysis. Rapid cycles of binding and release restrict the conformational folding space
that the polypeptide is able to explore and allow for the rapid burial of hydrophobic patches
that can partake in aberrant interactions and form aggregates. If the protein is not folded
after interaction with the Hsp70/Hsp4O system it may be transferred to the specialized
compartment of chaperonins to continue its folding trajectory.
Chaperonins are large cage-like ring complexes that function by enclosing the folding
polypeptide (up to 60 kDa in size) and isolating it from all other proteins in the cell
(reviewed in Hard and Hayer-Hartl 2011). Group I chaperonins, GroEL in bacteria, Hsp60
in eukaryotes are two component systems, with the barrel of the cage formed by the
chaperonin and the lid being formed by GroES, in bacteria, or Hsp1O in the case of
eukaryotes
TRiC/CCT
(Figure 8, reviewed in (Dunn et al. 2001)). Group II chaperonins, the
system in eukaryotes
function under the same premise, but instead of
cooperating with another subunit to form a closed cage, they undergo conformational
changes to enclose the structure. Chaperonins function in an ATP dependent manner,
coordinating the encapsulation of the substrate with hydrolysis of the ATP molecule. The
33
encapsulated protein is free to fold in the chaperonin enclosure until it is released (10s in the
GroEL/ES system, longer in the TrIC/CCT complex). Still unfolded substrates can re-bind
and the process can be repeated until the protein has acquired its native fold or it is
transferred to a different chaperone. Although chaperonins do not actively assist in folding,
they have been shown to dramatically accelerate the speed of folding, probably by spatial
confinement and the prevention of aberrant interactions and aggregation with other proteins.
In S. cerevisiae the TrIC/CCT complex is essential for the folding of a small subset of
proteins, but within these are proteins of high abundance and extreme importance in
structural integrity of the cell such as actin and tubulin.
Small heat shock proteins (Figure 8, reviewed in (Richter et al. 2010)) are not as
cohesive of a protein family as HSP90 or HSP70 are. sHSPs are usually monomeric proteins
that bind to hydrophobic patches of amino acids. For the most part, their clients have not
been well defined, but they are thought to be unstable folding intermediates and that the
binding of sHSPs prevents aberrant interactions. sHSPs are thought to play a role in protein
complex formation, binding to one protein subunit and occluding the binding interface
(usually highly hydrophobic) until the binding partner is found and the complex is formed.
There is a vast network of proteins whose function is to ensure protein folding
within the cell. Protein chaperones are both diverse and specialized, and while some assist in
general folding of proteins, many have a defined subset of protein clients whose folding they
aid. As protein homeostasis is a process of utmost importance, protein chaperones also
maintain a large amount of redundancy, with many having obligate clients but being able to
assist in folding of others if necessary. To cope with severe folding stress- many chaperones
have two variants, one that is constitutively expressed at low levels and another whose
expression is induced by proteotoxicity. In summary, the cell has developed a robust system
34
of protein folding factors to minimize aberrant interactions between proteins and ensure
peptides acquire the appropriate 3-dimensional structure.
Controlling Protein Aggregation
When polypeptides cannot fold into their native structure and remain misfolded, if
they partially unfold after being properly folded or if they are terminally damaged by
oxidation or carbonylation, they become aggregate-prone. Assembly defects in protein
complexes, as would happen when a required subunit is not expressed, can also lead to
aggregation of the existing subunits as hydrophobic patches that would be buried within the
complex remain exposed and form aberrant interactions. In addition to folding assistance
provided by chaperones, the cell utilizes chaperones to solubilize protein aggregates and
utilizes diverse mechanisms to prevent toxicity from aggregated proteins.
The Hsp10O chaperone family is comprised by members of the AAA ATPases, most
notably ClpB in bacteria and Hsp104 in yeast. Both ClpB and Hsp104 have disaggregating
capabilities, using ATP hydrolysis to break apart protein aggregates (Figure 8). The
mechanism by which they do this is unclear, but they are thought to thread the misfolded
proteins through a central pore of their hexameric ring, leaving the client protein in an
unfolded state so that it can refold either on its own or assisted by the Hsp70/Hsp4O
machinery (Richter et al. 2010, Mogk, 2004, Tyedmers, 2010). Although no homologues of
Hsp104 have been found in higher eukaryotes, disaggregation activity has been attributed to
the mammalian chaperone system comprised of Hsp110 and Hsp70/Hsp40 (Shorter 2011).
Yeast cells also sequester certain types of protein aggregates, usually those that cannot be
refolded, in special compartments The JUNQ (juxtanuclear quality control compartment)
transiently accumulates aggregated proteins that are ubiquitinated and destined for
35
degradation whereas the IPOD (Insoluble protein deposit) houses insoluble terminatally
aggregated proteins such as polyQ or carbonylated proteins(Kaganovich et al. 2008). When a
cell is unable to disaggregate and refold aggregated proteins, degradation of the aggregated
proteins is a viable alternative to alleviate toxicity.
Protein Degradation
In order to cope with alterations in protein homeostasis, cells degrade excess,
misfolded and aggregated protein subunits and aberrant peptides by means of the Ubiquitin
Proteasome System (UPS) or via autophagy.
The 26S proteasome is the central macromolecular machine responsible for the
degradation of proteins and protein aggregates. Its functions are so essential to the cell that
partially inhibiting its function can lead to neurodegeneration and complete inhibition is
lethal (Bedford et al. 2008). The 26S proteasome is comprised of a core, barrel-like particle
(20S subunit) and two regulatory complexes (19S) that function as lids. Proteins are
recognized and targeted for degradation by E3 ubiquitin ligases that attach ubiquitin moieties.
Specialized proteins that contain UBL (ubiquitin like) and UBA (ubiquitin associated)
domains act as adaptors between the target protein (the ubiquitin moieties are bound by the
UBA domain) and the 19S cap (binds the UBL domains). The proteins targeted for
degradation are deubiquitinated, unfolded and threaded through the core particle. The
recognition and binding of a substrate to the 19S cap is an ATP dependent process, and the
ATP molecule is required for unfolding, but not translocation into the pore. Proteolysis
occurs in the core particle through a threonine-dependent nucleophilic attack and results in
short stretches of amino acids that can then be further processed by cytosolic proteases and
recycled into new polypeptides.
36
Protein degradation is mediated not only by the proteasome but cells can additionally
deploy autophagy as a means of protein quality control (Kubota 2009). Misfolded proteins
are sequestered into aggregates and, in a p62-dependent manner, are targeted for autophagy.
Autophagy utilizes double-membraned structures that engulph the cytosolic target proteins
forming an autophagosome which then fuses with the lysosome for degradation of their
content (Bukau et al 2010). A key player in autophagosome formation is the membrane
protein
LC3/Atg8;
upon
autophagy
induction,
LC3
is
conjugated
to
phosphatidylethanolamine and recruited to the membranes of the nascent autophagosome.
One of the many ways one can monitor autophagy is by assessing the number of LC3 foci,
or by assaying the abundance of LC3-II, the autophagosome-specific, lipidated form of LC3.
Cellular responses to acute proteotoxic stressors
In order to maintain protein homeostasis under acute insults to proteostasis, there
are transcriptional programs that cells implement when faced with abnormal quantities of
misfolded proteins. These transcriptional programs are distinct according to which cellular
compartment is being assaulted by protein misfolding but they are all transient, tailored to
temporary stressors. The main goal of these programs is to reduce the folding burden (by
reducing the number of polypeptides being produced) and to enhance the cell's folding
capacity (by increasing the number of protein chaperones). The best studied is the program
elicited by the general misfolding of cytosolic proteins elicited by exposure to high
temperature and thus named the 'heat hock response' (HSR). High temperatures result in
general protein misfolding which leads to the activation of the transcription factor HSF1
(heat shock factor 1) that then results in the up-regulation of a subset of genes enriched for
protein chaperones and the down-regulation of genes involved in protein synthesis
37
(reviewed in (Richter et al. 2010)). Hsfl is kept in an inactive complex together with
components of the Hsp90 chaperone system. In a state of heat shock, the high abundance of
misfolded proteins is thought to titrate away the chaperones bound to Hsfl. In complex
with chaperone, Hsf1 is found as a monomer but its release leads to homotrimerization and
transport into the nucleus. There, Hsf1 is hyperphosphorylated by several kinases
(Holmberg et al. 2001). Further modification events, like sumoylation, regulate the activity of
the final transcription factor complex (Hietakangas et al. 2003). Complex regulatory
feedback ensures that the response is transient so as to return to normal levels of protein
production and chaperone abundance once the proteotoxic stress has been relieved.
Misfolded proteins in the endoplasmic reticulum (ER) result in a similar, but distinct
response termed the Unfolded Protein Response (UPR). The UPR is also transient, and
results in the up-regulation of ER specific chaperones and a general, temporary, reduction in
protein synthesis (Walter and Ron 2011). Studies in mammalian cells have also recently
described the mitoUPR (Mitochondria Unfolded Protein Response). Details are far less clear,
but the essence of the response is the same: misfolded proteins in the mitochondria result in
a signal that translates to a temporary decrease in protein production and an increase in
protein quality control capacity (Haynes and Ron 2010).
In summary, there are well-
understood transcriptional programs that aid in coping with abrupt changes in misfolded
proteins caused by disruptions of protein homeostasis.
ANEUPLOIDY, PROTEIN QUALITY CONTROL AND DISEASE
The connection between aneuploidy and disease has been at the forefront of the
study of aneuploidy. David van Hansemann first described unbalanced mitoses in 1890.
Theodor Boveri (1912) expanded upon his early description of aneuploid sea urchin
38
embryos to postulate that aneuploid cells could result in tumor formation. Aneuploidy of
chromosome 21 was described as the cause of Down's syndrome by Lejeune in 1959
(Lejeune et al. 1959). Recent studies have described associations between the aneuploid state
and neurodegenerative diseases and aging. Here I expand upon the most common
conditions associated with aneuploidy.
Aneuploidy in Cancer
Aneuploidy is extremely prevalent in solid tumors, with 7 0- 9 0% estimated to have
an unbalanced karyotype (Weaver and Cleveland 2006; Duijf and Benezra 2013). Cancer cells
have also long been considered 'chaperone addicted' (Neckers 2002) and Hsp90 inhibitors
are currently being developed as chemotherapeutics (Wagner et al. 2013). The dependency of
tumors on chaperones has been attributed to the need to efficiently fold oncogene products,
which are often kinases and thus Hsp90 clients. However, the high levels of aneuploidy in
cancer cells, and the proteotoxic stress that stems from such aneuploidy, could provide an
additional explanation for their chaperone addiction. Further investigation of compounds
that increase chaperone burden or that inhibit the function of chaperones might lead to the
discovery of new cancer therapeutics with efficacy in a broad spectrum of human tumors.
The high degree of aneuploidy observed in cancers also begs the question of whether
cancer cells have evolved mechanisms that allow them to tolerate high levels of karyotypic
imbalances. One aneuploidy-tolerating mutation appears to be loss of p53 function. In
normal cells, chromosome mis-segregation leads to activation of the tumor suppressor p5 3 ;
the mechanisms whereby this occurs are still being elucidated and might be caused by
multiple aspects of chromosome mis-segregation (Pavelka et al. 2010a; Thompson and
Compton 2010; Janssen et al. 2011). Generating a comprehensive list of genetic alterations
39
that ameliorate the effects of aneuploidy and their characterization will shed light on tumor
evolution. It will allow us to address important questions such as when such mutations arise
with respect to aneuploidy and whether and how they contribute to tumorigenesis.
Compounds that neutralize aneuploidy-tolerating mutations could also provide new avenues
of cancer treatment.
Whole-organism aneuploidy
In addition to cancer, autosomal aneuploidy has been associated with numerous
human conditions that result in impaired development. In humans, three viable trisomies
have been described. An additional copy of chromosome 21 leads to Down syndrome,
chromosome 18 to Edward's syndrome and a trisomy of chromosome 13 to Patau syndrome.
Of these, only Down Syndrome individuals survive past childhood. It will be interesting to
determine whether protein quality-control systems are affected in individuals with these
constitutional aneuploidies. Chromosome 21 harbors the fewest genes of all human
chromosomes and might thus not cause a significant burden on the cellular protein qualitycontrol pathways. Determining the contribution of impaired protein homeostasis to the
pleiotropic phenotypes of this syndrome could nevertheless be warranted because Down
syndrome is strongly associated with a protein-folding disease. Individuals with Down
syndrome are predisposed to early-onset Alzheimer's Disease (AD). Although the main
cause of AD in Down syndrome individuals is likely to be the additional copy of the APP
gene encoded by chromosome 21 (reviewed in (Kingsbury et al. 2006), mice overexpressing
APP (which encodes amyloid beta A4 protein) do not fully recapitulate all the Alzheimer'slike phenotypes seen in Down syndrome mouse models (Cataldo et al. 2003). Conversely,
mouse models of Down syndrome that lack the APP gene still exhibit some of the
40
Alzheimer's-like pathologies (Table 1), suggesting that duplication of the A PP gene may not
be the only cause of early-onset Alzheimer's disease in Down syndrome individuals. Thus,
perhaps a reduced ability to maintain protein homeostasis contributes to the Alzheimer's
disease pathology in individuals with Down syndrome.
Table 1. Comparison of the phenotypes associated with transgenic mouse models of
Down's syndrome or Alzheimer's disease
The two mouse models of Down syndrome are Ts65Dn and TslCje. Ts65Dn mice are
trisomic for the distal region of chromosome 16 (92 genes homologous to human
chromosome 21 from APP to MXJ); this segment contains nearly two-thirds of the human
chromosome 21 homologous genes, including the Down syndrome critical region (DSCR)
and the APP gene. Ts65Dn mice are also trisomic for a segment of mouse chromosome 17
(60 genes) that is non-homologous to genes on human chromosome 21. TslCje mice are
trisomic for a smaller region of chromosome 16 that includes the DSCR but not APP (67
genes homologous to chromosome 21, from SODI to MX1, approximately two-thirds of the
trisomic region of Ts65Dn mice), and they are monosomic for the telomeric region of
mouse Chr 12 (seven genes) (Cataldo et al. 2003). In addition to these two mouse models of
Down syndrome, transgenic mice have been generated that harbor an additional copy of a
mutant form of APP (K670M/N671L) that has been identified in a Swedish family with
early-onset AD ('APP overexpression', Table 1).
Although many of the phenotypes are
shared between the mice, an increased copy of APP is not sufficient to recapitulate all of the
Alzheimer-related phenotypes of Down's syndrome mouse models.
TABLE 1
41
Phenotype
APP status
Down Syndrome
Alzheimer's disease (APP
Ts65Dn
TslCjc
overexpression)
3 genomic
2 genomic copies
High levels of mutant APP
copies
Cognitive abnormalities
YES
YES
YES
Age-related atrophy and
YES
NO
YES
YES
NO
NO
NO
NO
YES
degeneration of cholinergic
neurons
Age-related endosomal
pathologies
Extracellular $-amyloid
aggregates
In addition to the constitutive aneuploidies of chromosomes 13, 18 and 21,
mutations in genes encoding the spindle assembly checkpoint component BUBR1 or
centrosome components have been shown to lead to mosaic variegated aneuploidy (MVA), a
disease characterized by aneuploidies showing a random widespread distribution in the body
(Hanks et al. 2004; Snape et al. 2011). There are no published evaluations of proteotoxicity
in MVA cell lines, but, given that protein quality-control systems have also been shown to be
impaired in complexly aneuploid yeast strains (haploid strains that are aneuploid for more
than one chromosome) (Oromendia et al. 2012), it would be of interest to investigate
whether the same is true in the case of MVA patients and to determine how this contributes
to the disease phenotype.
42
Aneuploidy and Neurodegeneration
Finally, neurodegenerative diseases are protein-folding diseases. Alzheimer's disease,
Parkinson's
disease,
amyotrophic lateral sclerosis
(ALS),
spinocerebellar
ataxia and
Huntington's disease are all characterized by the misfolding and aggregation of specific
proteins. Intriguingly, aneuploid yeast strains were found to be more prone than wild type
strains to form aggregates of a hard-to-fold protein containing a polyQ stretch, which is also
considered a model for Huntington's disease. Expressing this polyQ protein also impairs
proliferation of aneuploid yeast strains more than that of euploid controls, indicating that
expression of a hard-to-fold protein affects the fitness of aneuploid cells (Oromendia et al.
2012). Could aneuploidy be a contributor to neurodegenerative protein-folding diseases?
Several studies have suggested that as many as 30% of embryonic neurons and 15-20% of
adult neurons harbor aneuploidies (Rehen et al. 2001; Rehen et al. 2005; Yurov et al. 2005;
Yurov et al. 2007). On the other hand, in a recent study that performed single-cell whole
genome sequencing of neurons, high levels of copy number variation (CNVs) but no
increased aneuploidy was described (McConnell et al. 2013). Why aneuploidy would be more
prevalent in neurons compared with cells of other tissues is unclear, but it would provide an
intriguing explanation for the prevalence of protein-folding diseases in this cell type. Future
studies and additional methods to assess aneuploidy in tissues will be necessary to assess the
degree and types of aneuploidy comprehensively in the brain and to determine the effects, if
any, of aneuploidy on neurodegenerative diseases.
Aneuploidy and aging
All organisms age, and this process is characterized by, among other phenotypes, the
following:
genomic
instability,
epigenetic
alterations,
deregulated
nutrient
sensing,
43
mitochondrial dysfunction and loss of proteostasis (reviewed in (Lopez-Otin et al. 2013).
Furthermore, aging is the primary risk factor for major human diseases, including cancer,
diabetes, cardiovascular disorders and neurodegenerative pathologies. Interestingly, recent
studies by van Deursen and coworkers have provided intriguing links between aneuploidy
and the aging process. They found that mice carrying hypomorphic alleles in the spindle
assembly checkpoint gene BUBRI, which also serves as a mouse model for MVA, harbor
high levels of aneuploidy (Baker et al. 2004). Remarkably, these animals age prematurely.
Mice carrying hypomorphic alleles of BUBRI prematurely develop phenotypes characteristic
of old age such as cataracts, sarcopenia, growth retardation, muscle wasting, fat loss and
cardiac arrhythmias (Baker et al. 2004; Wijshake et al. 2012; Baker et al. 2013a; Baker et al.
2013b). Intriguingly, overexpression of BUBRI has the opposite effects - it leads to a
reduction in chromosome mis-segregation and hence aneuploidy (Baker et al. 2013a), and
the animals live longer and have a longer life without ailments (health-span) Furthermore,
cardiac function is increased, and muscle and renal atrophy and glomerulosclerosis are
reduced (Baker et al. 2013). Exactly how aneuploidy might result in aging remains to be
determined, but I propose that the systemic impacts of aneuploidy on cell physiology, such
as proteotoxicity, as discussed here, together with metabolic changes and genomic instability,
are the source of aneuploidy-induced aging. It will be very interesting to determine whether
mutations that suppress the adverse effects of aneuploidy also delay aging and extend life
and health span.
Concluding Remarks
Aneuploidy has a profound impact on most, if not all, cellular functions. This thesis
is centered on the consequences of aneuploidy on the protein quality control mechanisms of
44
the cell and the implications this could have on our understanding of human diseases and
aging. Aneuploidy has been shown to cause proteotoxic stress in yeast and mammalian cells.
Proteotoxicity is a consequence of aneuploidy irrespective of the identity of the
supernumerary chromosomes, and thus it is a phenotype inherent to the aneuploid state
itself. In this thesis, I will describe the consequences that aneuploidy has on protein
homeostasis that lead to an increased prevalence of protein aggregates when compared to
euploid cells. I have found that, not only does aneuploidy lead to increased endogenous
protein aggregates but it precludes the folding of known protein substrates, and it directly
affects the folding capacity of at least one chaperone: Hsp90. In impinging protein quality
control, an unbalanced karyotype also sensitizes strains to hard to fold disease proteins and
renders them more susceptible to prion conversion.
The impact aneuploidy has on essential cellular processes such as protein quality
control could be exploited as, a new direction for treatment of the many ailments that are
connected to aneuploidy, either in cause or in consequence. The sensitivity of aneuploid cells
to disruptions in protein quality control could be used to develop therapeutic and treatment
protocols that selectively impale cells that have not maintained the original karyotype.
Additionally, it is interesting to ponder the thought of enhancing protein quality control
abilities of the cell (either via a chemical enhancer or gene therapy) to counter-act the
detrimental effects of an unbalanced karyotype in cases of whole-organism aneuploidy such
as Downs Syndrome or as a way to ameliorate the steady decline of aging cells.
Understanding the full impact of this condition on cells and organisms will not only
deepen our knowledge of the consequences of an imbalanced karyotype but will provide
fundamental insights into developmental disabilities such as Down syndrome and diseases
such as cancer. Exciting too is the possibility that it might also unveil the mysteries of aging.
45
References
Baker DJ, Dawlaty MM, Wijshake T, Jeganathan KB, Malureanu L, van Ree JH, CrespoDiaz R, Reyes S, Seaburg L, Shapiro V et al. 2013a. Increased expression of BubR1
protects against aneuploidy and cancer and extends healthy lifespan. Nature cell biology
15: 96-102.
Baker DJ, Jeganathan KB, Cameron JD, Thompson M, Juneja S, Kopecka A, Kumar R,
Jenkins RB, de Groen PC, Roche P et al. 2004. BubR1 insufficiency causes early
onset of aging-associated phenotypes and infertility in mice. Nature genetics 36: 744-
749.
Baker DJ, Weaver RL, van Deursen JM. 2013b. p21 both attenuates and drives senescence
and aging in BubR1 progeroid mice. Cell reports 3: 1164-1174.
Bedford L, Hay D, Devoy A, Paine S, Powe DG, Seth R, Gray T, Topham I, Fone K,
Rezvani N et al. 2008. Depletion of 26S proteasomes in mouse brain neurons causes
neurodegeneration and Lewy-like inclusions resembling human pale bodies. The
Journalof neuroscience : the oficialjournalofthe Sociey for Neurosience 28: 8189-8198.
Boulon S, Pradet-Balade B, Verheggen C, Molle D, Boireau S, Georgieva M, Azzag K,
Robert MC, Ahmad Y, Neel H et al. 2010. HSP90 and its R2TP/Prefoldin-like
cochaperone are involved in the cytoplasmic assembly of RNA polymerase II.
Molecular cell 39: 912-924.
Brauer MJ, Huttenhower C, Airoldi EM, Rosenstein R, Matese JC, Gresham D, Boer VM,
Troyanskaya OG, Botstein D. 2008. Coordination of growth rate, cell cycle, stress
response, and metabolic activity in yeast. Molecular biology of the cell 19: 352-367.
Cataldo AM, Petanceska S, Peterhoff CM, Terio NB, Epstein CJ, Villar A, Carlson EJ,
Staufenbiel M, Nixon RA. 2003. App gene dosage modulates endosomal
abnormalities of Alzheimer's disease in a segmental trisomy 16 mouse model of
down syndrome. The Journalof neuroscience : the officialjournalof the Sociey for Neurosczence
23: 6788-6792.
Duijf PH, Benezra R. 2013. The cancer biology of whole-chromosome instability. Oncogene
32: 4727-4736.
Dunn AY, Melville MW, Frydman
J.
2001. Review: cellular substrates of the eukaryotic
chaperonin TRiC/CCT. J Struct Biol 135: 176-184.
Foijer F, DiTommaso T, Donati G, Hautaviita K, Xie SZ, Heath E, Smyth I, Watt FM,
Sorger PK, Bradley A. 2013. Spindle checkpoint deficiency is tolerated by murine
epidermal cells but not hair follicle stem cells. Proceedings of the National Academy of
Sciences of the United States ofA merica 110: 2928-2933.
Franzosa EA, Albanese V, Frydman J, Xia Y, McClellan AJ. 2011. Heterozygous yeast
deletion collection screens reveal essential targets of Hsp90. PloS one 6: e2821 1.
Gasch AP, Spellman PT, Kao CM, Carmel-Harel 0, Eisen MB, Storz G, Botstein D, Brown
PO. 2000. Genomic expression programs in the response of yeast cells to
environmental changes. Molecular biology of the cell 11: 4241-4257.
Hanks S, Coleman K, Reid S, Plaja A, Firth H, Fitzpatrick D, Kidd A, Mehes K, Nash R,
Robin N et al. 2004. Constitutional aneuploidy and cancer predisposition caused by
biallelic mutations in BUB1B. Nature genetics 36: 1159-1161.
Hartl FU, Bracher A, Hayer-Hartl M. 2011. Molecular chaperones in protein folding and
proteostasis. Nature 475: 324-332.
Haynes CM, Ron D. 2010. The mitochondrial UPR - protecting organelle protein
homeostasis. Journalof cell science 123: 3849-3855.
46
Hietakangas V, Ahlskog JK, Jakobsson AM, Hellesuo M, Sahlberg NM, Holmberg CI,
Mikhailov A, Palvimo JJ, Pirkkala L, Sistonen L. 2003. Phosphorylation of serine 303
is a prerequisite for the stress-inducible SUMO modification of heat shock factor 1.
Molecular and cellular biology 23: 2953-2968.
Holmberg CI, Hietakangas V, Mikhailov A, Rantanen JO, Kallio M, Meinander A, Hellman J,
Morrice N, MacKintosh C, Morimoto RI et al. 2001. Phosphorylation of serine 230
promotes inducible transcriptional activity of heat shock factor 1. The EMBOjournal
20: 3800-3810.
Hugerat Y, Simchen G. 1993. Mixed segregation and recombination of chromosomes and
YACs during single-division meiosis in spo13 strains of Saccharomyces cerevisiae.
Genetics 135: 297-308.
Janssen
A, van der Burg M, Szuhai K, Kops GJ, Medema RH. 2011. Chromosome
segregation errors as a cause of DNA damage and structural chromosome
aberrations. Science 333: 1895-1898.
Kaganovich D, Kopito R, Frydman J. 2008. Misfolded proteins partition between two
distinct quality control compartments. Nature 454: 1088-1095.
Kingsbury MA, Yung YC, Peterson SE, Westra JW, Chun J. 2006. Aneuploidy in the normal
and diseased brain. Cellular and molecularlife sciences : CMLS 63: 2626-2641.
Kubota H. 2009. Quality control against misfolded proteins in the cytosol: a network for cell
survival. Journalof biochemistry 146: 609-616.
Lacroix B, Maddox AS. 2012. Cytokinesis, ploidy and aneuploidy. J Pathol226: 338-351.
Lejeune
J,
Turpin R, Gautier M. 1959. [Chromosomic diagnosis of mongolism]. Archives
francaisesdepediatrie16: 962-963.
Li M, Fang X, Wei Z, York JP, Zhang P. 2009. Loss of spindle assembly checkpointmediated inhibition of Cdc20 promotes tumorigenesis in mice. The Journal of cell
biology 185: 983-994.
Longtine MS, McKenzie A, 3rd, Demarini DJ, Shah NG, Wach A, Brachat A, Philippsen P,
Pringle JR. 1998. Additional modules for versatile and economical PCR-based gene
deletion and modification in Saccharomyces cerevisiae. Yeast 14: 953-961.
Lopez-Otin C, Blasco MA, Partridge L, Serrano M, Kroemer G. 2013. The hallmarks of
aging. Cell153: 1194-1217.
McClellan AJ, Xia Y, Deutschbauer AM, Davis RW, Gerstein M, Frydman J. 2007. Diverse
cellular functions of the Hsp90 molecular chaperone uncovered using systems
approaches. Cell131: 121-135.
McConnell MJ, Lindberg MR, Brennand KJ, Piper JC, Voet T, Cowing-Zitron C, Shumilina
S, Lasken RS, Vermeesch JR, Hall IM et al. 2013. Mosaic copy number variation in
human neurons. Science 342: 632-637.
Miller MP, Amon A, Unal E. 2013. Meiosis I: when chromosomes undergo extreme
makeover. CurrOpin Cell Biol25: 687-696.
Neckers L. 2002. Heat shock protein 90 is a rational molecular target in breast cancer. Breast
disease 15: 53-60.
Niwa 0, Tange Y, Kurabayashi A. 2006. Growth arrest and chromosome instability in
aneuploid yeast. Yeast 23: 937-950.
Oromendia AB, Dodgson SE, Amon A. 2012. Aneuploidy causes proteotoxic stress in yeast.
Genes Dev 26: 2696-2708.
Parry EM, Cox BS. 1970. The tolerance of aneuploidy in yeast. Genet Res 16: 333-340.
Pavelka N, Rancati G, Li R. 2010a. Dr Jekyll and Mr Hyde: role of aneuploidy in cellular
adaptation and cancer. Curr Opin Cell Biol22: 809-815.
47
Pavelka N, Rancati G, Zhu J, Bradford WD, Saraf A, Florens L, Sanderson BW, Hattem GL,
Li R. 2010b. Aneuploidy confers quantitative proteome changes and phenotypic
variation in budding yeast. Nature 468: 321-325.
Regenberg B, Grotkjaer T, Winther 0, Fausboll A, Akesson M, Bro C, Hansen LK, Brunak
S, Nielsen J. 2006. Growth-rate regulated genes have profound impact on
interpretation of transcriptome profiling in Saccharomyces cerevisiae. Cenome biology
7: R107.
Rehen SK, McConnell MJ, Kaushal D, Kingsbury MA, Yang AH, Chun
J.
2001.
Chromosomal variation in neurons of the developing and adult mammalian nervous
system. Proceedings of the NationalAcademy of Sciences of the United States of America 98:
13361-13366.
Rehen SK, Yung YC, McCreight MP, Kaushal D, Yang AH, Almeida BS, Kingsbury MA,
Cabral KM, McConnell MJ, Anliker B et al. 2005. Constitutional aneuploidy in the
normal human brain. The Journal of neuroscience : the oficial journal of the Societ for
Neuroscience 25: 2176-2180.
Richter K, Haslbeck M, Buchner J. 2010. The heat shock response: life on the verge of death.
Molecular cell 40: 253-266.
Segal DJ, McCoy EE. 1974. Studies on Down's syndrome in tissue culture. I. Growth rates
and protein contents of fibroblast cultures. J Cell Physiol 83: 85-90.
Sheltzer JM, Blank HM, Pfau SJ, Tange Y, George BM, Humpton TJ, Brito IL, Hiraoka Y,
Niwa 0, Amon A. 2011. Aneuploidy drives genomic instability in yeast. Science 333:
1026-1030.
Sheltzer JM, Torres EM, Dunham MJ, Amon A. 2012. Transcriptional consequences of
aneuploidy. Proceedings of the National Academy of Sciences of the United States of America
109: 12644-12649.
Shorter J. 2011. The mammalian disaggregase machinery: Hspl 10 synergizes with Hsp70 and
Hsp40 to catalyze protein disaggregation and reactivation in a cell-free system. PloS
one 6: e26319.
Siegel
JJ,
Amon A. 2012. New insights into the troubles of aneuploidy. Annual review of cell
and developmental biology 28: 189-214.
Snape K, Hanks S, Ruark E, Barros-Nunez P, Elliott A, Murray A, Lane AH, Shannon N,
Callier P, Chitayat D et al. 2011. Mutations in CEP57 cause mosaic variegated
aneuploidy syndrome. Naturegenetics 43: 527-529.
St Charles J, Hamilton ML, Petes TD. 2010. Meiotic chromosome segregation in triploid
strains of Saccharomyces cerevisiae. Genetics 186: 537-550.
Stingele J, Roder UW, Raasi S. 2012a. Surface plasmon resonance to measure interactions of
UbFs with their binding partners. Methods in molecularbiology 832: 263-277.
Stingele S, Stoehr G, Peplowska K, Cox J, Mann M, Storchova Z. 2012b. Global analysis of
genome, transcriptome and proteome reveals the response to aneuploidy in human
cells. Molecular systems biolog 8: 608.
Stingele S, Stoehr G, Storchova Z. 2013. Activation of autophagy in cells with abnormal
karyotype. A utophag 9: 246-248.
Taipale M, Jarosz DF, Lindquist S. 2010. HSP90 at the hub of protein homeostasis:
emerging mechanistic insights. Nature reviews Molecularcell biology 11: 515-528.
Tang YC, Amon A. 2013. Gene copy-number alterations: a cost-benefit analysis. Cell 152:
394-405.
Tang YC, Williams BR, Siegel
JJ,
Amon A. 2011. Identification of aneuploidy-selective
antiproliferation compounds. Cell 144: 499-512.
48
Thompson SL, Compton DA. 2008. Examining the link between chromosomal instability
and aneuploidy in human cells. The Journalof cell biology 180: 665-672.
-. 2010. Proliferation of aneuploid human cells is limited by a p53-dependent mechanism.
The Journalof cell biology 188: 369-381.
-.
2011.
Chromosome missegregation in human cells arises through specific types of
kinetochore-microtubule attachment errors. Proceedings of the National A cademy of
Sciences of the United States ofAmerica 108: 17974-17978.
Thorburn RR, Gonzalez C, Brar GA, Christen S, Carlile TM, Ingolia NT, Sauer U,
Weissman JS, Amon A. 2013. Aneuploid yeast strains exhibit defects in cell growth
and passage through START. Molecular biology ofthe cell24: 1274-1289.
Torres EM, Dephoure N, Panneerselvam A, Tucker CM, Whittaker CA, Gygi SP, Dunham
MJ, Amon A. 2010. Identification of aneuploidy-tolerating mutations. Cell143: 71-83.
Torres EM, Sokolsky T, Tucker CM, Chan LY, Boselli M, Dunham MJ, Amon A. 2007.
Effects of aneuploidy on cellular physiology and cell division in haploid yeast. Science
317: 916-924.
Tyedmers J, Mogk A, Bukau B. 2010. Cellular strategies for controlling protein aggregation.
Nature reviews Molecularcell biology 11: 777-788.
Wagner AJ, Chugh R, Rosen LS, Morgan JA, George S, Gordon M, Dunbar J, Normant E,
Grayzel D, Demetri GD. 2013. A Phase I Study of the HSP90 Inhibitor
Retaspimycin Hydrochloride ([PI-504) in Patients with Gastrointestinal Stromal
Tumors or Soft-Tissue Sarcomas. Clinical cancer research : an official journal of the
American A ssociationforCancer Research 19: 6020-6029.
Walter P, Ron D. 2011. The unfolded protein response: from stress pathway to homeostatic
regulation. Science 334: 1081-1086.
Weaver BA, Cleveland DW. 2006. Does aneuploidy cause cancer? Curr Opin Cell Biol18: 658-
667.
Wijshake T, Malureanu LA, Baker DJ, Jeganathan KB, van de Sluis B, van Deursen JM. 2012.
Reduced life- and healthspan in mice carrying a mono-allelic BubR1 MVA mutation.
PLoSgenetics 8: e1003138.
Williams BR, Amon A. 2009. Aneuploidy: cancer's fatal flaw? Cancerresearch 69: 5289-5291.
Williams BR, Prabhu VR, Hunter KE, Glazier CM, Whittaker CA, Housman DE, Amon A.
2008. Aneuploidy affects proliferation and spontaneous immortalization in
mammalian cells. Science 322: 703-709.
Young JC, Agashe VR, Siegers K, Hart FU. 2004. Pathways of chaperone-mediated protein
folding in the cytosol. Nature reviews Molecularcell biolog 5: 781-791.
Yurov YB, Iourov IY, Monakhov VV, Soloviev IV, Vostrikov VM, Vorsanova SG. 2005.
The variation of aneuploidy frequency in the developing and adult human brain
revealed by an interphase FISH study. The journal of histochemisty and gtochemisty
officialjournalof the Histochemisty Society 53: 385-390.
Yurov YB, Iourov IY, Vorsanova SG, Liehr T, Kolotii AD, Kutsev SI, Pellestor F,
Beresheva AK, Demidova IA, Kravets VS et al. 2007. Aneuploidy and confined
chromosomal mosaicism in the developing human brain. PloS one 2: e558.
Zhang J, Baker ML, Schroder GF, Douglas NR, Reissmann S, Jakana J, Dougherty M, Fu CJ,
Zhu
J,
Levitt M, Ludtke SJ et al. 2010. Mechanism of folding chamber closure in a group II
chaperonin. Nature 463: 379-383.
Pavelka N, Bradford WD, Rancati G, Li R. 2012. Karyotypic determinants of
chromosome instability in aneuploid budding yeast. PLoSgenetics 8: e1002719.
49
Chapter 2:
Aneuploidy causes proteotoxic stress in yeast.
Reproduced with permission from Genes & Development:
Oromendia, Ana B., Stacie E. Dodgson, and Angelika Amon. "Aneuploidy causes
proteotoxic stress in yeast." Cenes & development 26.24 (2012): 2696-2708
50
Introduction
Aneuploidy, defined as a karyotype that is not a multiple of the haploid complement,
results in an unbalanced genome. This imbalance is generally not well tolerated in nature, as
evidenced by the impaired fitness of aneuploid cells and organisms (reviewed in (Torres et al.
2008; Williams and Amon 2009)). The condition also has a profound impact on human
health. Aneuploidy is the leading cause of mental retardation and spontaneous abortions and
a key characteristic of cancer, as more than 90% of all solid human tumors harbor aneuploid
genomes (Weaver and Cleveland 2006).
Systematic analyses of aneuploid yeast and mouse cells suggest that aneuploidy
causes chromosome-specific effects that are elicited by the duplication/deletion of individual
genes or combinations of a small number of genes present on the aneuploid chromosome
(Torres et al. 2007; Pavelka et al. 2010; Torres et al. 2010; Tang et al. 2011). Aneuploid yeast
and mammalian cells also share a set of phenotypes, collectively called the aneuploidyassociated stresses (Torres et al. 2007; Williams et al. 2008; Tang et al. 2011), indicating that
the aneuploid state per se impacts cell physiology. Aneuploidy impairs proliferation of
budding and fission yeast cells as well as of mammalian cells under standard growth
conditions (Baker et al. 2004; Niwa et al. 2006; Torres et al. 2007; Thompson and Compton
2008; Williams et al. 2008; Li et al. 2009; Pavelka et al. 2010), with a delay at the G1 - S
phase transition being especially prominent (Niwa et al. 2006; Torres et al. 2007; (Stingele et
al. 2012). Whole chromosomal aneuploidies also lead to a transcriptional response. A gene
expression signature similar to the environmental stress response (ESR; (Gasch et al. 2000a))
in budding yeast has been observed in aneuploid budding and fission yeast strains,
A rabidopsis, mouse and human cells (Sheltzer et al. 2012). Lastly, aneuploid cells exhibit
phenotypes characteristic of disruption of protein homeoastasis. Aneuploid budding yeast
51
strains and trisomic mouse embryonic fibroblasts show increased sensitivity to compounds
that interfere with protein folding and turnover (Torres et al. 2007; Pavelka et al. 2010;
Torres et al. 2010; Tang et al. 2011). Understanding the phenotypes shared by many different
types of aneuploidies is of particular importance, as this could provide insights into how an
unbalanced karyotype impacts normal cellular physiology and disease states such as cancer.
Here we investigate the consequences of one of the general effects of aneuploidy disruption of cellular protein homeostasis.
Maintaining the proteome is essential for cell survival. Nascent peptides must be
folded, proteins that are unfolded have to be refolded, and terminally damaged proteins
must be degraded. Additionally, multi-protein complexes must be properly assembled. The
cell relies on molecular chaperones to aid in the folding and refolding of proteins and the
assembly of multi-protein complexes, as well as on the 26S proteasome and vacuolar
proteases to degrade proteins that are terminally misfolded (Tyedmers et al. 2010; Houck et
al. 2012).
The chaperone and ubiquitin-proteasome systems function in concert to ensure
protein homeostasis. When cells experience proteotoxic stress, that is, when protein qualitycontrol pathways such as the chaperone systems and the proteasomal degradation machinery
are compromised or overwhelmed, misfolded proteins are not eliminated and aggregates
form (Houck et al. 2012). Misfolded proteins not only inflict a fitness cost (Geiler-Samerotte
et al. 2011), they are also associated with human disease. Highly-structured aggregates have
been linked to neurodegenerative pathologies, including Huntington's, Alzheimer's and
Parkinson's diseases, as well as prion diseases such as Kuru and Creutzfeld-Jacob Syndrome
(reviewed in (Goedert et al. 2010)).
We previously generated 13 budding yeast strains harboring an additional copy of a
single yeast chromosome, called disomes. These strains exhibit, among other deleterious
52
phenotypes, increased sensitivity to high temperature, to inhibitors of protein synthesis and
folding, as well as chemical and genetic perturbation of proteasomal degradation (Torres et
al. 2007; Torres et al. 2010). Furthermore, we found that increasing proteoasomal
degradation by deleting the gene encoding the deubiquitinating enzyme Ubp6 improves the
proliferative abilities of a subset of disomic yeast strains. Together with the observation that
the additional chromosomes are actively transcribed and translated (Torres et al. 2007;
Pavelka et al. 2010; Torres et al. 2010), these studies suggest that aneuploidy alters the cell's
proteome resulting in proteotoxic stress and implicate the ubiquitin-proteasome pathway in
the survival of aneuploid cells. However, direct evidence of proteotoxicity in aneuploid cells,
and the role of chaperones in the generation of proteotoxic stress in aneuploid cells has thus
far been lacking. Here we show that aneuploid yeast cells are prone to protein aggregate
formation. Aneuploid yeast strains generated by a variety of different methods have defects
in aggregate clearance and exhibit increased sensitivity to aggregate-prone proteins. The
association between aneuploidy and protein aggregation uncovered in this study could have
important implications for the pathology and treatment of diseases such as cancer and
neurodegeneration, which have both been associated with aneuploidy.
Results
Disomicyeaststrains harbora higher load of endogenousprotein aggregates.
Introduction of whole chromosomes substantially alters the cell's proteome because
most genes present on the additional chromosome are expressed according to gene copy
number (Torres et al. 2007; Torres et al. 2010). This may impact protein homeostasis
mechanisms. To test this possibility we analyzed the subcellular localization of the
disaggregase Hsp104. Under standard growth conditions, the Hsp104 chaperone fused to
53
eGFP is diffusely localized throughout the cell, but the protein also co-localizes with protein
aggregates, manifesting as Hspl04-eGFP foci (Liu et al. 2010) (Figure 1A, B). All 13 disomic
strains analyzed showed a significant increase in the percentage of cells harboring Hsp104eGFP foci compared to the euploid control (Figure
1A, B). Increased aggregate formation
was not due to slowed proliferation caused by aneuploidy (Torres et al. 2007) because
temperature-sensitive cdc23-1 and cdc28-4 strains grow slowly at the permissive temperature
but do not harbor additional aggregates (Figure 1C). Our data further suggest that it is the
increased protein load generated from the additional chromosome that leads to increased
protein aggregation. We did not observe an increase in the percentage of cells with Hsp104eGFP foci in strains that contain yeast artificial chromosomes (YACs; Figure 1D) that carry
human DNA but generate no yeast proteins and very few if any other peptides and proteins
(Foote et al. 1989; Torres et al. 2007). As protein aggregates are the consequence of
misfolded proteins, our data suggest that aneuploid cells are challenged to fold proteins
efficiently and/or to process protein aggregates appropriately.
54
Figure 1: Disonic yeast strains harbor an increased protein aggregate load.
(A) Wild-type and disomic yeast strains containing an HSP04-eGFPfusion were grown to
exponential phase in YEPD, and the percentage of cells harboring Hspl04-eGFP foci was
determined (n=3; SEM, n=100 cells/time point; **P<0.005; ***P<0.0005; Student's t test).
Strain order: A31392, A31393, A31394, A31395, A31396, A31397, A31398, A31399,
A31400, A31401, A31402, A31403, A31404, A31405.
(B) Images of Hspl04-eGFP aggregates. Aggregates are in green, DNA in blue.
(C, D) WT (A25654), cdc23-1 (A29766) and cdc284 (A29765) strains (C) and strains
harboring YACs containing 580kb (A28922) or 670kb (A28925) of human DNA (D) were
grown as in (A) to determine the percentage of cells with Hspl04-eGFP aggregates.
(E) Quantification of Hspl04-eGFP foci in trisomic yeast strains grown as in (A). Strains in
order: A31406, A31407, A31408, A31409, A31410. Note that the number of cells with
Hspl04-eGFP aggregates in diploid cultures is lower than in haploid cultures. The basis for
this is at present unclear.
(F) Strains grown at 25*C were shifted to 37*C. The percentage of cells with Hspl04-eGFP
aggregates was determined at the indicated times after temperature shift (n=100 cells/time
point). Two replicas of this experiment are shown in Figure S1. Strains are the same as in (A).
55
Figure 1
B
A60-
25'C
x~
a.50-
T
U-
I
4030-
***
I
C)
2010.
I-
LI
z
04
C
-$\0
"
D
F
20-
L20-
U
C.R
10-
0.
10.
J-h-
0
0 0
-kV
E
L
U)
C6iC6,
10080U60
(9 40
C)
20J
0v
0
10080
60
40
20
0(
0
U
,C150
0
U-
iT
0.
U)
C.)
U,
Q0
0
<tp
\0
00
4$
~.
0
.00
+
U)
~_wr
Dis I
Dis VIII
-N+- Dis IX
-Ii- Dis X
-S- Dis XI
-a-
-0-
1
2
3
4
1
-+- Dis 11
-0-
-0- Dis XV
1
2
3
4
--
WT
Dis IV
-0- Dis V
--- Dis XI
--
u- 6040
20
00
Dis XIII
-- Dis XVI
100
-80
['1
-C
- -V
37'C
Eu
Eu
I
['1
25'C
-0- Dis XIV
2
3
4
Time (h)
56
We also found that, as with all other aneuploidy-associated phenotypes (Torres et al.,
2007), increasing ploidy suppressed aggregate formation. The percentage of cells harboring
Hspl04-eGFP foci in diploid strains carrying an additional chromosome (trisomic strains) is
significantly lower than that of haploid strains with an extra chromosome (compare Figure
1A and Figure 1E). Many subunits of protein complexes require the assistance of protein
chaperones to fold. These proteins then acquire a stable conformation by binding to the
complex's other subunits. If one of the components is present in excess and cannot exist
stably as an uncomplexed subunit, it requires the continuous assistance of chaperones to
prevent aggregation (Tyedmers et al. 2010). As a result, chaperones cannot assist other
folding reactions and the general folding capacity of the cell is reduced. The observation that
increasing ploidy reduces aggregate formation suggests that the proteotoxic stress in
aneuploid cells could, in part, be a result of stoichiometric imbalances caused by the proteins
encoded on the unbalanced chromosomes. Decreasing the ratio of uncomplexed proteins to
complexed proteins reduces the protein aggregate load of aneuploid yeast. The observation
that aggregate formation in many trisomic strains is not elevated compared to diploid
controls further suggests that cells either have the ability to compensate for some genomic
imbalances and/or that diploids have a higher folding capacity than haploids.
A daptation to proteotoxic stress is delayed in disomicyeaststrains.
If disomic yeast strains experience increased proteotoxic stress, they may be delayed
in responding or adapting to conditions that induce proteotoxicity. To test whether disomic
yeast strains are delayed in adapting to proteotoxic stress-inducing growth conditions, we
monitored Hspl04-eGFP foci after shift to high temperature (37*C). Virtually all wild-type
and disomic cells contained Hspl04-eGFP foci within an hour of temperature shift (Figure
57
1F). However, whereas wild-type cells cleared the aggregates by 4 hours, all disomes except
for disomes IV and XIV adapted to heat stress with slower kinetics (Figure 1F, 2). This
delayed adaptation to high temperature was not due to an inability to mount a heat-shock
response, as judged by microarray analysis of aneuploid cells adapting to thermal stress
(Figure 3). Activation of the unfolded protein response (UPR) in the endoplasmic reticulum
was also unaffected in aneuploid strains; splicing of the UPR gene HA C1 in the disomes was
similar to that in wild-type cells, both under normal conditions and under conditions when
the UPR is induced (Figure 4).
Figure 2: Behavior of Hspl04-eGFP aggregates in response to heat stress.
Two independent repeats of the experiment shown in Figure 1F are shown in (A) and (B).
Strains grown at 25*C in YEPD were collected on a filter and shifted to YEPD pre-warmed
to 37*C. The percentage of cells with Hspl04-eGFP aggregates was determined at the
indicated times after temperature shift (n=100/time point). Strains: A31392, A31393,
A31394, A31395, A31396, A31397, A31398, A31399, A31400, A31401, A31402, A31403,
A31404, A31405.
58
Figure 2
B
A
.5
0.
a 100-
+
-o-w-o-
4
80
601
1m2
0
3
Dish
Dis XIII
Dis XVI
Dis XV
--
0-*+
40
20
U,
_ ,__
o
1
0
,___
,___
3
2
Time (h)
Time (h)
3
4
3
0wr
W7
80 Dis I
-o- Dis VIII
-*- Dis IX
-a- Dis X
-.Dis XI
40
0o
Dis IV
-o- Dis V
-N- Dis XII
-o- Dis XIV
2
U100-
1
*1
404
20
1
2
Time (h)
,_
60.
o0
20
0
4
-100-+WT
D-0
0
T
Dis I
-o- Dis Vill
Dis IX
-c Dis X
Dis XI
6-N-
M
6-o-
Ile
Time (h)
-100
A
o
4
+ W
+ Dis|1
Dis XIII
-*Dis XVI
-o- Dis XV
10080
+WT
2
3
4
Time (h)
U100
P 80
60
WT
+ Dis IV
-o- Dis V
- Dis XII
-o- Dis XIV
+
20
o
1
2
3
4
Time (h)
59
Figure 3: The heat shock response is intact in disomic yeast strains.
WT and disomic yeast strains were grown at 25*C and shifted to 37*C. RNA samples were
taken 0, 5, 15 and 30 minutes after shift. RNA extracted from WT cells grown continuously
at 25*C was used as reference for all samples. Data were mined for genes involved in the
heat-shock response (Gasch et al. 2000a)(Gasch et al. 2000b), and those present in the extra
chromosome were removed from the analysis (grey boxes). Data set was split into those
genes that are upregulated and those that are down regulated in the heat-shock response. (A)
The expression of genes involved in the heat-shock response in disomic yeast after shifting
to 37*C. Yellow shows upregulated genes and blue shows downregulated genes. Shown are
the unclustered data after being zero-transformed. (B) The average expression changes of
up- (above x-axis) and down- (below x-axis) regulated genes displayed in (A) are shown.
60
Figure 3
A
WT
o
minutes at 37C
DisV
DisX
Dis XIV Dis
S 15 30 o s
; is 30 0 s 15 n 0 S 15 n 0
minutes at 37C
3,00
1.00
0.00
-1.00
1-.00
a
4)
a
a
-C
0
'M
S!
a
M3
a5
CO
0.
B
2-
I
1p
10
-,
0i>
-1
20
minutes at 370C
30
-.-
-
Dis. V
Dis X
Dis XIV
Dis XVI
-3-
61
Figure 4: The Unfolded Protein Response (UPR) in disomic yeast strains.
Wild type and disomic strains were grown at 25*C. The culture was split and one half was
treated with 2pg/ml of Tunicamycin (TM), a compound known to induce the UPR. Samples
were collected 2 hours later and the amount of unspliced (upper band) and spliced (lower
band) HA C1 was determined. The red bar indicates the location of the probe. Strains in
order are: WT (A22361), Dis I (A6863), Dis II (A6865), Dis VIII (A27036), Dis X (A21986),
Dis XI (A28266) Dis XIII (A21987) and Dis XIV (A28344).
Figure 4
WT
2 pg/pL Tm (2 hr)
I
Disi
-+1-+
WT7 Tis
Disi IDis
+
+
IVI DisVlDis VIII
+
-+
IXi D isX bisXI\iDisXViDisXVI
2 pg/pL Tm (2 hr)
62
Although aneuploid yeast strains can mount a heat-shock response, the proteotoxic
stress that we observe in disomic cells under standard growth conditions (25*C, YEPD) is
not sufficient to induce a canonical heat-shock response in most disomes (Figure 5) (Torres
et al. 2007). This lack of a heat shock response is expected, as this response is tailored to an
acute stressor and is transient in nature (Gasch et al. 2000b). In contrast, aneuploidy is a
chronic stress and adaptation to the aneuploid state may have taken place. It is however
noteworthy that disomes IV and XIV, which adapt to heat-shock with the same kinetics as
wild-type cells (Figure 1F, 2) upregulate genes involved in the heat-shock response even
under normal growth conditions (Figure 5). Analysis of the abundance of the chaperones
Hsp104, Ssal, Sse2 and Hsp42 further confirmed the absence of a canonical heat-shock
response (Figure 6). Although aneuploidy is not sufficient to induce the canonical heat-shock
response, most disomic yeast strains show a transcriptional response reminiscent of the
environmental stress response (Torres et al. 2007), which encompasses a subset of the heatshock response Gasch et al. 2000a). HSP1O4 RNA levels, for example, increase with degree
of aneuploidy (Sheltzer et al. 2012). Taken together, out results indicate that disomic yeast
strains experience proteotoxic stress that is evidenced by increased aggregate burden, both
under normal growth conditions and under conditions that induce proteotoxicity.
63
Figure 5: Heat shock signature in aneuploid strains grown in normal, unstressed
conditions.
Microarray data from aneuploid yeast strains grown in -His G418 medium at 25*C from
Torres et.al. (Torres et al. 2007) were mined for genes shown to be upregulated (A) and
downregulated (B) in the heat-shock response (Gasch et al. 2000b). A Heat Shock Score
was calculated by averaging the log2 disome/wt ratio of all heat shock induced or heat shock
down-regulated genes. Shown are the averages of all replicates available and the SEM. Note,
at present it is unclear why disomes XIII and XIV exhibit increased expression of genes upregulated in response to heat-shock but not of genes down-regulated upon heat-shock.
64
Figure 5
A
1.0-
K
0.80.6-
Ci)
0.40 c)
g) -
CL0
0.20.0'
FIW.I nn
-,-uzL 1-I
H
rL rL
1,
+
co
"C
B
0.20.0CD-0.2-
iuu
Yy
-40.4-
ri
rT-,
Z0, (2-0.6< 0'-0.81 .0
IT
\A
40,\-x
65
Figure 6: The abundance of several chaperones is unaltered in many disomic yeast
strains.
Hsp104, Ssal, Sse2 and Hsp42 protein levels were examined in wild-type cells and disomic
yeast grown at 25*C in YEPD. Pgkl was used as a loading control.
(a-g) Hspl04-eGFP levels were determined in WT (A31392), Dis I (A31393), Dis II
(A31394), Dis IV (A31395), Dis V (A31396), Dis VIII (A31397), Dis IX (A31398), Dis X
(A31399), Dis XI (A31400), Dis XII (A31401), Dis XIII (A31402), Dis XIV (A31403), Dis
XV (A31404), and Dis XVI (A31405) in 2-fold dilutions.
(h) Ssal-3HA levels were determined in WT (A32407), Dis II (A32408), Dis V (A32409),
Dis VIII (A32410), Dis X (A32411), Dis XI (A32412) Dis XIII (A32413), Dis XV (A32414)
and Dis XVI (A32415).
(i) Sse2-3HA levels were determined in WT (A32416), Dis V (A32417), Dis VIII (A32418),
Dis X (A32419), Dis XIII (A32420), Dis XV (A32421) and Dis XVI (A32422).
(j) Hsp42-3HA levels were determined in WT (A32423), Dis II (A32424), Dis VIII (A32425),
Dis XIII (A32426), Dis XIV (A32427) and Dis XVI (A32428).
66
Figure 6
A
WT
x
Dis I
Dis 11
x x xxx
x xx x
H
- > 5;
OL5
Hspl04-eGFP
-- .
-
Pgkl
B
WT
X)(
Dis IV
x
aoqrNj-
x
x
ooqwNj-
x
I,-
Hspl04-eGFP
Pgk1
C
Pgk1
I
Dis V
XX
xxxx
00oV(N-
..
Ssa1-3HA
0
-- 7 -- I Sse2-3HA
--- -- --- 1 Pgkl
-- -- *mom
1
Disv III
WT
Hspl04-eGFP
Pgk1
WI
D ~xx
xx
cqwCN
Dis X
F-
;A
-
I
0
V
Hsp42-3HA
Pgkl
Dis XI
x x xx x 4x x
0
r- - 00 rq --
-
144000
I HsplO4-eGFP
II Pgkl
E
WT
x
Dis XII
x
x
x
x
Dis XIII
xx
x
x
xx
IIHsp104-eGFP
"
"IWi S
F
WT
00
x 4x
CN
x
'-
iss
Dis XIV
Dis XV
x x xx
00 V
N
Pgkl
x xx x
O
-
V~ CN
1 Hsp104-eGFP
i 40wm--- - -*wow*
% Pgk1
MO""
G
~x
WT
Dis XVI
x4x
xx
mi
x
EIb
qmpow*wow
I
I Hsp104-eGFP
I
Pgkl
67
Meiotic and mitotic chromosome mis-segregationleads to protein aggregateformation.
Increased protein aggregation was not only observed in strains harboring single
chromosomal aneuploidies, but also in aneuploid cells that arose from meiotic and mitotic
non-disjunction. Triploid cells induced to undergo meiosis produce highly aneuploid
progeny with karyotypes ranging from diploid to highly aneuploid (St Charles et al. 2010).
The majority of the aneuploid progeny is inviable (Parry and Cox 1970), but some genetically
unstable aneuploid strains can be obtained (Sheltzer et al. 2011; Zhu et al. 2012). As colony
formation is a prerequisite for the analysis, we were only able to analyze those aneuploids
that were healthy enough to form colonies. Nevertheless, analysis of 19 products of triploid
meioses showed that the percentage of cells harboring Hspl04-eGFP foci was increased in
most strains (Figure 7A).
Chromosome mis-segregation during mitosis also resulted in aggregate formation.
Strains
harboring
temperature-sensitive
alleles
of genes
encoding the
kinetochore
component Ndc1O or the Aurora B kinase Ipli were arrested in G1 and released to progress
through the cell cycle at the semi-permissive temperature of 30'C. Under these conditions,
35% of ndclO-1 and 29% of ip/1-321 cells were unable to segregate a GFP-marked
chromosome IV (Figure 7B), indicating that dramatic chromosome mis-segregation occurs
under these growth conditions. Aggregate formation was increased as early as 3 hours after
release from the pheromone-induced G1 arrest (Figure 7C). Importantly, this increase in
Hspl04-eGFP foci depended on cell division. When ndclO-1 or ipll-321 cells were induced
to undergo a synchronous cell cycle at 30'C but chromosome segregation was prevented by
treating cells with the microtubule-depolymerizing drug nocodazole, Hspl04-eGFP focus
number did not increase (Figure 3C). We conclude that the percentage of cells harboring
68
Hspl04-eGFP-decorated protein aggregates is increased in most, if not all, aneuploid strains
and that aggregates form soon after chromosome non-disjunction.
Figure 7: Meiotic and mitotic chromosome
non-disjunction causes increased
Hspl04-eGFP focus formation.
(A) The percentage of cells with Hspl04-eGFP aggregates was analyzed in progeny of
diploid (A28220) or triploid (A28219) strains 3 days after germination (n=100 cells/strain).
Note that although many of the progeny from the triploid meioses will harbor multiple
aneuploidies, some will also be euploid or will have become euploid as they proliferate.
(B, C) Wild-type (A5244), ndclO-1 (A28204) and pll-321 (A16154) mutants harboring a
GFP-marked chromosome IV, and wild-type (A25654), ndclO-1 (A27681) and ipll-321
(A27682) mutants harboring Hspl04-eGFP, were arrested in GI with pheromone at 25*C
followed by release at 30*C either in the presence (10pg/ml; Arrested) or absence (Dividing)
of nocodazole. Samples were taken after 3 hours to determine the percentage of cells that
correctly segregated chromosome IV (B) and that harbored Hspl04-eGFP foci (C). We note
that the temperature shift during this experiment may inflate the percentage of cells
harboring aggregates in all strains, as they will be adapting to the temperature shift.
69
Figure 7
A
.5
-c
4.J
40
4-
908070-
0
60-
C-) C
5040-
C',
0
1-
C-
B
A
3020100
AA
A JA AA
A
U
.mE
U
*.
'A
AAA
A
A
6"..
A
I
Diploid F1
Triploid F1
100
L)
75.
cu
50.
0
0
CD>
25.
<n
0
WT
C
60-
.U
4-
-c 0L
U-
ndc1O-1
ipi 1-321
Dividi ng
Arres ted
**
40-
C',
C-) 0.
0
C)
20-
0-
WT
ndc1O-1
ipl1-321
70
A neuploid strainsfail to efficienty fold the protein quality controlsensor VHL
Protein aggregates in disomic yeast strains could be the result of proteins generated
from the additional chromosomes overwhelming and/or impairing chaperones. To test this
idea we challenged the cell's protein quality-control pathways using the well-studied substrate,
the human von Hippel-Lindau protein (VHL). VHL is unable to fold without its binding
partners ElonginB and ElonginC. When human VHL is expressed in yeast in the absence of
Elongin B and C, the protein is quickly ubiquitinated and degraded (McClellan et al. 2005;
Kaganovich et al. 2008) (Figure 8).
The quality-control pathways involved in the elimination of misfolded VHL are
known: folding-defective VHL is shuttled from Hsp70 to an Hsp90 complex that enables
degradation by the ubiquitin-proteasome system (McClellan et al. 2005). When any of these
pathways are defective, misfolded VHL forms aggregates that are seen as foci in cells
expressing VHL as a GFP fusion (Kaganovich et al. 2008) (Figure 9F.) Disomic yeast strains
grown under non-stress conditions (YEPRG, 25*C; Figure 9A) and under conditions of heat
stress (2 hours at 37'C; Figure 9B) showed increased VHL focus formation; slow-growing
mutants or strains carrying human DNA did not (Figure 9C, D). The failure to process
misfolded VHL-GFP was not specific to the disomic strains, but was also observed in
progeny of triploid meioses (Figure 9E). We note that haploid strains obtained from diploid
meioses harbored a higher percentage of cells with VHL-GFP foci than the haploid control
strain analyzed in the experiment shown in Figure 9A. We suspect that germination and
colony growth on selective medium places an increased burden on the cell's protein qualitycontrol systems compared to growth in liquid rich medium (YEPRG) at 25*C. We conclude
that targeting of VHL-GFP for proteasomal degradation is compromised in aneuploid cells,
either because the protein quality-control pathways (chaperones and/or the proteasome) of
71
the cells are defective or they are functional but overwhelmed by changes in the cell's
proteome caused by the aneuploid state.
Figure 8: Inhibiting the proteasome causes VHL-GFP aggregate accumulation in
aneuploid and euploid strains.
Wild-type and disomic yeast strains deleted for the multidrug transporter PDR5 and
expressing a VHL-GFP fusion were analyzed after a transient (2hr) exposure to the
proteasome inhibitor MG132 (80M) and the percentage of cells with VHL-GFP foci was
determined (n=3; SEM, n=100/time point).
(a) Strain order: A32076, A32078, A32080, A32081, A32082, A32083, A32084, A32086,
A32088, A32089
(b) Wild-type (A32076), cdc23-1 (A30461), cdc28-4 (A30462) and wild-type strains harboring
YACs carrying human DNA (A29969 and 29971).
Figure 8
A
nnr
1000
>;
LZ 75-
501 0 25-
0
B
10075
-C%
50
C
CO
- - - , , , , , , , , EM RM E
I
I
25
0
.
72
Figure 9: Aneuploid yeast display hallmarks of impaired protein quality control.
(A) Wild-type and disomic yeast strains deleted for the multidrug transporter PDR5 and
expressing a VHL-GFP fusion were grown in YEP 2%Raf 2%Gal at 25*C and the
percentage of cells with VHL-GFP foci was determined (n=3; SEM, n=300 cells/time point;
*P<0.05, **P<0.005; ***P<0.0005; Student's
t
test). Strain order: A32076, A32077, A32078,
A32079, A32080, A32081, A32082, A32083, A32084, A32085, A32086, A32087, A32088,
A32089
(B) Strains described in (A) were analyzed after a 2hr incubation at 37*C (n=100).
(C, D) cdc23-1 (A30461), cdc284 (A30462) and wild-type strains harboring YACs carrying
human DNA (A29969 and A29971) were grown at 25'C (C) or shifted for 2 hours to 37*C
(D) to analyze VHL-GFP focus formation.
(E) The percentage of cells with VHL-GFP foci was determined in progeny of diploid
(A28388) or triploid (A28389) strains 4 days after germination (n=100/colony).
(F) Images of VHL-GFP aggregates. Aggregates are in green, DNA in blue.
73
Figure 9
A
C
17.515.0-
aL'
) 10.0C-4
15.0
,,**
,***
I
12.5-
I
17.5
.g
5
*
7.5-
5.0o 2.5-
LL
9
12.5-
1
10.0
6
7.5
._'
5.0
S 2.5--
0
0
D
t21
C
0.0'
B
a
807060-
>0
I ,
5040-
404-
C')
E
9
*
.
0
*
>U)
3020100
0
-
F
9080-
A
A
70
A
60
U
GFP
GFP & DAPI
a-
AA
AA
AA
*
-J
3020100
0
70
60
50
40
30 -~
20
10
0 Q-ii
* *
A
0UDiploid F1
Triploid F1
74
Loss of UBP6 reduces aggregate burden in disomicyeast strains.
The high incidence of both endogenous and VHL protein aggregates in aneuploid
strains suggests that aneuploidy negatively impacts protein folding and/or degradation of
misfolded proteins. Previous studies showed that ubiquitin-proteasomal degradation is
important for the survival of aneuploid yeast strains (Torres et al. 2007; Torres et al. 2010).
Insufficient proteasome activity could also be responsible for increased aggregate formation
in aneuploid cells. A prediction of this hypothesis is that increasing proteasome function
decreases aggregate burden in aneuploid cells. To test this possibility we examined the
consequences of deleting UBP6 on Hspl04-eGFP focus formation in disomic yeast strains.
Ubp6 associates with the proteasome and removes ubiquitin chains from substrates. This
not only allows for the recycling of ubiquitin but also causes proteasome substrates to escape
degradation. This is evident from the analysis of cells lacking UBP6. Degradation of all
proteasome substrates analyzed to date is accelerated in such cells (Hanna et al. 2006; Peth et
al. 2009). We deleted UBP6 in disome V and disome XI cells, whose proliferation improves
when UBP6 is deleted, and in disome II cells, in which deleting UBP6 leads to decreased
proliferation (Torres et al. 2010). Deletion of UBP6 reduced aggregate burden in all three
disomic strains (Figure 10). This finding is consistent with our previous observation that
deletion of UBP6 causes attenuation of levels of proteins with high relative expression in all
disomic strains analyzed, irrespective of whether deletion of UBP6 improves proliferation
(Torres et al. 2010).
Interestingly, in disome V strains, aggregate burden was reduced to
almost wild-type levels when UBP6 was deleted (Figure 10). This finding raises the
interesting possibility that the increased proliferative abilities of disome V ubp6D cells are
due to a reduction in protein aggregates. We conclude that enhanced proteasomal
degradation reduces the aggregate burden in all disomic strains analyzed.
75
Figure 10: Increased proteasome activity decreases aggregate burden in disomic
strains.
Wild-type(A3369), disome II (A33370), disome V (A33371) and disome IX (A33372) cells
harboring a deletion of UBP6
and the HSP104-eGFP fusion were grown to exponential
phase in YEPD, and the percentage of cells harboring Hspl04-eGFP foci was determined
n=3; SEM, n=100 cells/time point.
Figure 10
.5
402 6
D
UBP6
0 I]ubp6
Cj)
0
WT
Dis 11
Dis V
Dis XI
Hsp90 folding capacity is reduced in many disomicyeast strains.
Are other protein quality-control systems also affected in aneuploid cells? Because
VHL is an Hsp90 client, we explored the in vivo folding activity of Hsp90 and found it to be
reduced in many disomic strains. Hsp90 is a highly abundant chaperone that, in concert with
co-chaperones, folds cytosolic proteins (McClellan et al. 2007; Franzosa et al. 2011).
Consistent with previous results using the Hsp90 inhibitor geldanamycin (Torres et al. 2007),
we found that several disomic yeast strains are more sensitive to the Hsp90 inhibitor
radicicol than the euploid control strain (Figure 11A).
76
To examine Hsp90 activity, we analyzed the in vivo folding activity of the well-studied
Hsp90 model substrate, the tyrosine kinase Src. Both c-src and the oncogenic form, v-src,
depend on Hsp90 for folding (Kimura et al. 1995; Nathan et al. 1997). For unknown reasons,
overexpression of v-src, but not c-src, is lethal in budding yeast (Xu and Lindquist 1993).
Compromising Hsp90 activity suppresses v-src folding and activity and, consequently, this
lethality (Nathan and Lindquist 1995; Nathan et al. 1997) (Figure 11B). We found that the
toxicity of v-src was diminished in many aneuploid strains (Figure 11C), suggesting a
reduction in Hsp90 activity. To further explore the activity of v-src, we took advantage of the
low levels of endogenous tyrosine phosphorylation in yeast that are dramatically increased
when v-src is expressed from the galactose-inducible GALI -10 promoter (Figure 11D). Total
tyrosine phosphorylation was reduced in disomes II, V, VIII and XII (Figure 11D),
correlating well with these cells' ability to form colonies under v-src-inducing conditions
(Figure 11C). The inability to generate active v-src was not due to decreased mRNA
expression, as v-src RNA levels were as high or higher in the disomes than in wild-type cells
(Figure 11 E). Reduced v-src activity was also observed in disomes I, XIII, XV and XVI as
judged by reduced tyrosine phosphorylation levels, but this decreased activity was not
sufficient to allow growth on v-src-inducing medium (Figure 11 C, D). Our results show that 8
out of 11 disomes exhibit reduced Hsp90 activity. Hsp90 may be overloaded by substrates
that rely on this chaperone to be folded. It is also possible that the activity of the Hsp90
folding machinery is reduced. Given that many different disomic strains exhibit decreased
Hsp90 activity, we favor the idea that the Hsp90 folding reservoir is depleted, rather than
inactive, in aneuploid strains. Hsp90 is thought to serve a limited number of clients under
normal growth conditions and to be present in excess (Borkovich et al. 1989; Neckers 2007).
It is therefore surprising that Hsp90 activity appears limiting in many disomic strains.
77
Perhaps under conditions of proteotoxic stress, Hsp90's folding repertoire is expanded. We
conclude that many aneuploid cells experience saturation of the Hsp90 system.
Figure 11: Hsp90 folding capacity is limiting in many disomic strains.
(A) Disomic yeast harboring a deletion of the multidrug transporter PDR5 were grown in
YEPD or YEPD containing 70 ptM radicicol at 30'C to determine their doubling time. Mean
and SEM of 3 replicates is shown. Strains in order are: A15549, A15551, A15553, A15555,
A15557, A15559, A15561, A15563, A15566, A15567, A15569, A15571, A15573.
(B) Schematic of v-src/c-src Hsp90 assay. Hsp90 is required to fold c-src and toxic v-sr. A
reduction in Hsp90 activity results in misfolded v-src and restores cell viability.
(C) Wild-type and disomic yeast strains carrying c-src or v-src under the galactose-inducible
CAL1-10 promoter were grown under conditions where expression is repressed (-URA 2%
Glu.) or induced (-URA 2% Raf. Gal.). 10-fold dilutions were plated. C-src strains in order
are: A32090, A32091, A32092, A32093, A32094, A32095, A32096, A32097, A32098,
A32099, A32100, A32101. V-src strains in order are: A32102, A32103, A32104, A32105,
A32106, A32107, A32108, A32109, A32110, A32111, A32112, A32113.
(D, E) Wild-type and disomic yeast strains harboring the GAL-v-src fusion, were grown in
YEP+ 2% raffinose. Galactose was added and the relative amount of v-src RNA (E) and total
tyrosine phosphorylation (D) was determined before and after 2 hours of v-src induction. Vsrc strains in the same order as in (C).
78
Figure 11
A
B
25-
.C20-
EJ YPD
csrc
0 70 pM Radicicol
csrc +,0
i!
E 15
10-
-ut-
-. -I.
W4
D
WT
Dis Dis Dis Dis Dis
2
2 0 2
II
0
V
2
0
2
VIII IX
0 2 0 2
WT Dis Dis Dis Dis Dis Dis
XIV XV VI
XI XII
0 210 210 210 2
0
202
0
2
m.
*.1wM
Z
.
<4-
2.5
0 hr induction
2 hr induclion
1.5-
S2-
1.0.
0.5
0
41
0 hr induction
2 hr induction
2.0
E 3-
\+4
-+-
Misfold csrc Viable
Viable
vsrc + Hsp9O -+
Fold vsrc
vsrc +060
Misfold vsrc Viable
-
II III
0.11 .
1
.9. Ell
lz qAC.,
4z : i
WT csrc
WT vsrc
Dis I csrc
Dis I vsrc
Dis 11 csrc
Dis 11 vsrc
Dis V csrc
Dis V vsr
WT csrc
WT vsrc
Dis X11 csrc
Dis XII vsrc
Dis XIII csrc
Dis XIII vsrc
Dis XIV csrc
Dis XIV vsrc
E
5-E
Fold csrc
WT csrc
WT7 vsrc
Dis Vill csrc
Dis Vill vsrc
Dis IX csrc
Dis IX vsrc
Dis X csrc
Dis XI vsrc
PpY
Kar2 11
-+
C
0
0
hr post
induction 0
+ Hsp9O
P,
0 0
WT csrc
WT vsrc
Dis XV csrc
Dis XV vsrc
Dis XVI csrc
Dis XVI vsrc
79
A neuploid strainsare more susceptible to protein aggregatesassociatedwith human disease.
Does aneuploidy also cause cells to be more susceptible to protein folding defects
associated with human disease? To address this question, we employed an assay that
measures the activity of the prion protein Sup35 and assessed toxicity associated with the
glutamine-rich protein Httl in disomic yeast strains.
The prion [PSI+] is formed by Sup35, a subunit of the translation terminator
complex. When Sup35 switches to the aggregated amyloid conformation [PSI+], much of
the protein becomes unavailable to terminate translation, causing read-through of stop
codons (Liebman and Sherman 1979). Because the basal conversion frequency of Sup35 to
its
prion form is low (10-1-107), we used strains carrying a variant of the SUP35 gene
(SUP35-R2E2) that increases the conversion frequency (Liu and Lindquist 1999; Cox et al.
2003) to study the effects of aneuploidy on SUP35 activity. We then utilized an assay where
conversion from [psi-] to [PSI+] results in read-through of three stop codons upstream of
GFP, allowing expression of the fluorescent protein (Tyedmers et al. 2008). Single colonies
obtained from a frozen stock were inoculated into rich medium, and the percentage of
fluorescent cells was determined immediately after inoculation and after 8 and 24 hours. All
disomic strains tested showed an increase in the fraction of cells expressing GFP (Figure
12A, B, Figure 13). Attempts to visualize the Sup35 aggregates by SDD-AGE (SemiDenaturing Detergent-Agarose Gel Electrophoresis) were unsuccessful. We therefore cannot
exclude the possibility that mechanisms other than prion conversion lead to the observed
increase in the percentage of GFP-positive cells in the disomic strains. We, however, favor
the interpretation that Sup35 aggregates are in the detergent-soluble, small oligomer stage
that precedes the large amyloid aggregates detectable by SDD-AGE (Halfmann et al. 2010)
80
and/or that aggregates comprise a small fraction of total Sup35 protein and are thus
undetectable by this technique.
Figure 12: Disomic yeast strains shown increased expression of the Sup35 prion
reporter.
Single colonies of wild-type and disomic strains carrying the SUP35-R2E2 allele and a GFP
construct preceded by 3 stop codons were inoculated into SC medium and the percentage of
fluorescent cells was determined after 0, 8 and 24 hours by flow cytometry (A). Shown is the
ratio of GFP+ cells after 8 and 24 hours of growth to GFP+ cells immediately after
inoculation (0 hr). The mean and SEM of at least 12 single colonies are depicted. Strains in
order are: A31114, A29843, A29845, A29846, A29847, A29848, A31110, A31111, A31112,
A29450, A31113.
(B) Strains in (A) and A22361 (no GFP control) were grown for 24 hours and total protein
was extracted. GFP protein levels were analyzed by Western blot analysis. Pgkl was used as
a loading control.
81
Figure 12
A
12U)
C/)
0~
4.
0
E8hr/Ohr
M24hr/Ohr 11
I
Isin
4 It
liftni
ill
B
GFP
I
Pgkl
82
Figure 13: Sup35 activity in disomic yeast strains.
Single colonies of wild-type and disomic strains carrying the SUP35-R2E2 allele and a GFP
construct harboring 3 stop codons were inoculated into YEPD medium and the percentage
of fluorescent cells was determined immediately after inoculation and after 8 and 24 hours of
growth by FACS. The average of at least 12 single colonies is shown and data are plotted as
a percentage of GFP positive cells of a total of 30,000 cells/sample. Strains in order are:
A31114, A29843, A29845, A29846, A29847, A29848, A31110, A31111, A31112, A29450,
A31113.
Figure 13
100-
80-
U)
04
L1
Time =0
Time = 8
Time = 24
60-
4
20-
0goEl
83
Huntington's disease is a neurodegenerative
disease
associated with
P-sheet
aggregates comprised mainly of the Huntingtin (Httl) protein (Goedert et al. 2010). Toxicity
and disease phenotypes require that the poly-glutamine (polyQ) stretch in its N terminus
expand beyond 38 repeats (Duyao et al. 1993; 1993). We used yeast strains expressing a 17
amino acid fragment of Httl exon 1 with polyQ tracts of varying length to determine the
susceptibility of aneuploid yeast to expanded polyQ stretches (Duennwald et al. 2006). In
euploid cells, Httl harboring 25 glutamine residues (25Q) is not toxic when expressed from
the galactose-inducible CALi-10 promoter, but Httl harboring 46 or 72 Qs causes toxicity
(Duennwald et al. 2006). All disomes tested, except for disome VIII, exhibited increased
sensitivity to Htt1-polyQ expression compared to the euploid control (Figure 14A). Httl46Q-CFP aggregates are also accumulated more readily in many of the disomic strains
analyzed (Figure 14B). The lack of sensitivity of disome VIII is most likely due to reduced
expression of the construct (Figure 14C). We conclude that the proteotoxicity that afflicts
aneuploid yeast cells can predispose them to the accumulation of protein aggregates
associated with human diseases.
84
Figure 14: Disomic yeast strains exhibit increased sensitivity to Huntingtin polyQaggregates.
(A) Wild-type and disomic strains harboring a CAL-FLAG-HTT1(17AA)25QApro-CFP,
GAL-FLA C-HTF
T(17AA)46QApro-CFP
or
GAL-FLA G-HTT(17AA)72QApro-CFP
construct were grown under conditions where expression is repressed (YEPD) or induced
(YPRG). 10-fold dilutions were plated. 25Q strains in order are: A32114, A32115, A32116,
A32117, A32118, A32119, A32120. 46Q strains in order are: A32121, A32122, A32123,
A32124, A32125, A32126, A32127, A32128. 72Q strains in order are: A32129, A32130,
A32131, A32132, A32133, A32134, A32135.
(B) Wild-type and disomic strains harboring the GAL-FLA G-HTT(17AA)46QApro-CFP
construct were grown for 8 hours in the presence of galactose to determine the percentage
of cells with Httl-46Q-CFP foci (n=100). Shown are mean and SEM of 3 independent
experiments.
(C) Expression of the CAL-HTIT1(17aa)-FLAG- n9-CFPconstructs. Strains were grown in
YEP 2% raffinose to OD 600= 0.2 when
2%
galactose was added. RNA was extracted from
samples taken after two hours and the amount of H7T1-nQ-CFP RNA was determined via
Northern blot analysis.
85
Figure 14
A
OFF
ON
OFF
ON
OFF
ON
WT
Dis 11
Dis V
Dis Vill
Dis XI
Dis XIll
Dis XVI
25Q
B
72Q
30-
U0
0
46Q
I
20-
6
~10-
0~
0
~%>~'
C
'
A I1
VV 1
Dis Dis Dis Dis Dis Dis
11 1 V IilI Al IAIi JAVI hours
0 21 0210 210 210 210 202
-N
Vqs
post nd.
25Q
rRNA
1 0
*0
46Q
NS
rRNA
4
I
0*
9
p
72Q
cam en~ amerRNA
86
Discussion
Our studies of aneuploid yeast have revealed the dramatic effect of an unbalanced
karyotype on cellular protein homeostasis. All aneuploid strains, irrespective of how they
were generated or their karyotype, showed an increased protein aggregate burden. Aneuploid
strains are prone to aggregation of endogenous proteins as well as of ectopically expressed
hard-to-fold proteins such as polyQ stretch-containing proteins. We do not know which
proteins comprise the aggregates observed in aneuploid cells. Obligate chaperone clients
present in excess due to aneuploidy could accumulate and then either form aggregates
themselves or interfere with the folding of other chaperone clients. Identifying aggregate
constituents will distinguish between these two non-mutually exclusive possibilities.
Two protein quality-control systems, the proteasome and the Hsp90 chaperone,
appear to be limiting in many aneuploid yeast strains. Increasing proteasome function by
deleting UBP6 led to a decrease in aggregate burden in all aneuploid yeast strains analyzed.
We also found that the Hsp90 substrate v-src was less active in many disomic strains
indicating that Hsp90 activity is limiting, which could contribute to aggregate formation in
these strains. This latter result is surprising as Hsp90 is highly abundant and its activity
thought to be in excess in cells (Borkovich et al. 1989; Neckers 2007). Perhaps this is not the
case, especially under conditions of proteotoxic stress. Hsp90 may not be the only
chaperone system limiting in aneuploid cells. We speculate that other protein folding
pathways are also saturated with different aneuploidies impacting different chaperone
families to varying degrees depending on the identity of the proteins encoded on the extra
chromosomes. Examining the activity of the different folding pathways in different disomic
yeast strains will test this idea.
87
Why are aneuploid cells aggregate-prone?
Aggregates could be a result of overwhelmed folding pathways, or they could stem
from reduced chaperone activity. Both alternatives are possible, but given that the aneuploid
chromosomes are actively expressed, it is likely that excess proteins produced from the
aneuploid chromosomes occupy chaperones and thereby reduce their availability to assist in
the folding of their other clients. What determines the extent of aggregate formation is not
yet known. It does not appear to correlate with either the degree of aneuploidy (by DNA
content), total protein in excess, delay in G1 or doubling time. However, it is important to
bear in mind that our aggregate analyses do not measure absolute amounts of aggregated
proteins in cells nor are they able to distinguish toxic from non-toxic aggregates. The
environmental stress response, which encompasses part of the heat-shock response,
correlates with degree of aneuploidy (Torres et al. 2007) (Sheltzer et al. 2012). We propose
that protein aggregate burden correlates with the number of obligate chaperone clients, and
hence with the distribution of their encoding genes in the genome.
It may seem surprising that the cell's protein quality-control pathways cannot
compensate for the presence of a single additional chromosome, which depending on
chromosome size, results in 2 - 12 percent of the genome being imbalanced. Previous studies
showed that even small amounts of misfolded proteins place a burden on the cell's protein
quality-control systems and hence adversely affect cellular fitness. Expression of a single
misfolded cytosolic protein at less than 0.1% of total protein leads to a significant decrease
in proliferative abilities and the induction of a cytoplasmic unfolded protein response
(Geiler-Samerotte et al. 2011).
Importantly, the generation of misfolded proteins requiring the assistance of the
cell's protein quality-control pathways is a common occurrence in aneuploid cells. It is well
88
established that many subunits of protein complexes only acquire a stable conformation by
binding to other subunits of the complex (Imai et al. 2003; Boulon et al. 2010). Thus, every
single polypeptide produced by genes located on aneuploid chromosomes that normally has
a binding partner is - in the disomes - in excess. For example, if in euploid cells 1 percent of
a subunit of a heterodimeric protein complex is present in excess due to variability in subunit
expression and must be eliminated, the number of proteins that needs to be eliminated rises
to 102% in cells that carry an additional copy of the gene encoding one of the two subunits.
This scenario applies to all proteins encoded on the extra chromosome that require a binding
partner to acquire a stable conformation. This dramatic change in protein stoichiometries,
we propose, leads to an increased burden on the protein quality-control pathways of the cell.
Individual subunits present in excess require the continuous assistance of chaperones,
preventing chaperones from assisting other folding reactions and reducing the general
folding capacity of the cell and thus interfere with their essential function of mediating
folding of essential proteins (Hard et al. 2011). The fact that the aggregate phenotype was
ameliorated when the ratio of uncomplexed proteins to properly complexed proteins was
decreased by increasing base ploidy (as in trisomic strains) suggests that the proteotoxicity
observed in aneuploids is indeed in part the result of the protein stoichiometry imbalances
caused by aneuploidy, although it is also possible that diploid cells are more efficient at
clearing aggregates. We furthermore propose that an additional burden on the protein
quality-control machinery is generated by the overproduction of proteins encoded on the
extra chromosomes that require chaperones for their function, such as protein kinases and
WD40 repeat proteins. In summary, aneuploidy impacts protein homeostasis in multiple
ways, so that even small unbalanced chromosomes have a significant impact on the cell's
protein quality control systems.
89
A neuploidy in cancerand neurodegenerative diseases.
Our results have important implications for how we think about the impact of
aneuploidy on human disease. Solid tumors, which are highly aneuploid, have long been
deemed chaperone-addicted (Neckers 2007; Workman et al. 2007; Powers et al. 2008).
Eliminating HSF1, the master regulator of the heat-shock response, results in a lower
incidence of tumors in mice (Dai et al. 2007). This dependence on chaperones has been
attributed to the importance of chaperones for the folding of oncogenes. Our studies
suggest that the aneuploid nature of tumors contributes to their dependence on chaperones
such as Hsp90 for survival. We further suggest that aneuploidy could contribute to
neurodegenerative diseases such as Huntington's or Alzheimer's Diseases. The human brain
is a naturally aneuploid organ, with one third of fetal neurons and 10% of adult neurons
being aneuploid (Rehen et al. 2001; Rehen et al. 2005; Yurov et al. 2007a; Yurov et al. 2007b).
Our finding that aneuploidy causes proteotoxic stress including polyQ aggregate formation,
raises the interesting possibility that aneuploidy reduces the cell's capacity to eliminate
protein aggregates and/or increases the propensity for aggregate formation. Thus, their
aneuploid nature may predispose neurons to protein aggregation diseases. Interestingly, mice
chimeric for trisomy 16 (one of the mouse models of Down's syndrome) have been
previously associated with increased susceptibility and poor prognosis when injected with the
Scrapie prion protein (Epstein et al. 1991). Further studies of the proteotoxicity associated
with aneuploidy could therefore provide important insights into tumorigenesis and
neurodegenerative diseases and may even pave the way for the development of novel
treatments for these diseases.
90
Materials and Methods
Strains andplasmids:
Strains used in this study are described in Table S1 and are derivatives of W303. Strains were
constructed using PCR-based methods described by Longtine et al. (Longtine et al. 1998).
The generation of disomic strains has been described previously (Torres et al. 2007).
Karyotypes of all disomic and trisomic strains were confirmed by comparative genome
hybridization (Torres et al. 2007). YACs used in this study have been previously described
(Foote et al. 1989). The pGAL-VHL-GFP
fusion is described in Kaganovich et. a.
(Kaganovich et al. 2008). CA L-HYT1(I7aa)-FLAG-25Q-CFP and CA L-HT1(I7aa)-FLA G46Q-CFP and GAL-HT1(17aa)-FLAG-72,Q-CFP are described in Duennwald et aL
(Duennwald et al. 2006).
A naysis of endogenousprotein aggregatesin disomicyeaststrains.
For analysis of endogenous aggregates, strains carrying an Hspl04-eGFP fusion were grown
in YEPD medium. Exponentially-growing cells were fixed in 3.7% formaldehyde by adding
0.1 ml 37% formaldehyde to 1 mL of cells. Cells were then permeabilized in 1%
Triton/Potassium Phosphate, washed and resuspended in KPi/Sorbitol. The percentage of
cells harboring Hspl04-eGFP foci was determined in at least 100 cells per sample. Foci were
defined as GFP dots that were visible without the aid of a camera.
A na/ysis of endogenous aggregatesin progenj of diploid and trpiloidmeioses.
Diploid and triploid cells were sporulated. Tetrads were dissected on YEPD plates. Colonies
that grew up were diluted in water and Hspl04-eGFP foci were analyzed as described above
91
in at least 100 cells/colony. We note that since this analysis relies on colony growth, the
most severe aneuploids that cannot form colonies cannot be analyzed.
A na/ysis of endogenous aggregates upon chromosome mis-segregation.
Haploid strains harboring the temperature-sensitive ndclO-1 or
alleles were arrested
/pl1-321
in YEPD + lOg/ml a-factor at room temperature. After 90 minutes, 5pg/ml a-factor was
added. 180 minutes after the initial a-factor addition, cells were washed and released into pre
warmed (30*C) YEPD. To half of the culture 15 pig/ml nocodazole was added. Hspl04eGFP foci were analyzed 3 hours after release from the G1 arrest as described above.
High temperature adaptationtime courses.
Cells were grown to exponential phase in YEPD at 25*C. Cells were collected by filtration
and resuspended in pre-warmed (37*C) YEPD. Samples were taken at the indicated times
(Oh is immediately before shifting temperature). Hspl04-eGFP foci were counted in at least
100 cells/time point as described above.
A naysis of cellularproteinqualiy control using the VHL-GFP reporter.
Strains deleted for the multidrug transporter PDR5 harboring the GZAL-VHL-GFP fusion
and split in two. Half
were grown at 25*C in YEP 2% Raffinose 2% Galactose to OD e0=0.2
6
the culture was maintained at 25*C and the other half was shifted to 37'C. Samples were
taken 2 hours later, fixed with 3.7% formaldehyde and the percentage of cells harboring
VHL-GFP foci was determined. GFP foci were counted without the aid of a camera, and
any cell with a visible focus was counted as a cell harboring a focus. At least 300 cells were
counted for cultures grown at 25*C and at least 100 for cultures grown at 37*C.
92
A naysis of VHL aggregates in progeny of diploid and triploidmeioses.
Diploid and triploid cells were sporulated and tetrads were dissected on plates lacking
leucine containing 2% raffinose and 2% galactose. Colonies were resuspended in water and
the percentage of cells with VHL-GFP foci was determined as described above. As this
analysis relies on colony growth, the most severe aneuploids that cannot form colonies
cannot be analyzed.
Effects of radicicolon the growth rate ofdisomicyeast strains.
Disomes deleted for the multidrug transporter PDRS were inoculated at OD60 0 =0.1 in
YEPD in 96-well plates either lacking or containing 70ptM radicicol in freshly made medium.
OD 60 was measured every 15 minutes on a plate reader (Synergy2, Biotek) for 24 hours.
Doubling times were calculated using the exponential growth phase of each culture.
Assessing v-src and Httl-poyQ glutamine toxicity:
Tenfold serial dilutions were prepared and spotted onto the appropriate medium: medium
lacking uracil and containing either 2% glucose or 2% raffinose, 2% galactose for GAL--src,
GAL-c-src containing strains and YEPD or YEP 2% Raf 2% Gal for GAL-HTT1(17aa)FLA C-nQ-CFPharboring strains. Plates were imaged after 3 days of growth at 25 0 C.
Western blot analyses.
Cells were harvested by adding an equal volume of 10% trichloroacetic acid to the cell
culture and incubated on ice for at least 20 minutes. Cells were then washed with 1.5mL of
acetone. The dried pellet was resuspended in 100 pLL of protein breakage buffer (50mM Tris,
93
pH 7.5, 1mM EDTA, 2.75 mM DTT, and Roche Complete protease inhibitor, used per the
manufacturer's instructions). 100 ptL of glass beads were added and the cells broken by
beating for 2.5 minutes on a Biospec mini-bead beater. 50 piL of 3X SDS sample buffer were
added, the samples boiled for 5 minutes, then centrifuged for 5 minutes. An equal volume of
lysate was loaded onto 10% SDS polyacrylamide gel, electrophoresed, and transferred to a
nitrocellulose membrane. Hspl04-eGFP was detected using a mouse anti-GFP antibody (L8, Clontech) at a 1:1,000 dilution. Pgk1 was detected using a mouse anti-Pgkl antibody (A6457, Molecular Probes) at a 1:5,000 dilution. Ssal-3HA, Sse2-3HA and Hsp42-3HA were
detected using a mouse anti-HA antibody (HA.11, Covance) at a 1:1,000 dilution. Total
phosphotyrosine levels were detected using a mouse anti phosphotyrosine antibody (4G10,
Millipore) at a 1:1,000 dilution. The secondary antibody was a sheep anti-mouse antibody
coupled to horseradish peroxidase (NA931, GE Healthcare) and used at a 1:2,000 dilution.
Kar2 was detected using a rabbit anti-Kar2 antibody at a 1:200,000 dilution and followed by
donkey anti-rabbit antibody coupled to horseradish peroxidase (NA9340, GE Healthcare)
used at a 1:2,000 dilution. Bands were detected using Amersham ECL Plus Substrate
according to the manufacturer's instructions.
Determination of Sup35function.
The GFP read-through assay to assess Sup35 function was performed essentially as
described previously (Tyedmers et al. 2008). Briefly, strains were streaked from frozen stocks
on selective medium (-His, G418) and allowed to grow to single colonies for 3 days. Single
colonies were then resuspended in 200 p1 SC, and 100 p was analyzed immediately by flow
cytometry to determine %GFP-positive cells (n>10,000). The remaining 100p were used to
inoculate 3-ml SC cultures, which were grown at room temperature and maintained in
94
exponential phase. Samples were taken for flow cytometry analysis after 8 and 24 hours of
growth and the percentage of GFP-positive cells was determined.
A nafysis of lyQ aggregates.
Strains harboring the GAL-HTT1(17aa)-FLAG46Q-CFPconstruct were grown at 25'C in
YEP 2% Raffinose 2% Galactose to OD60 () =0.4. Samples were taken, fixed with 3.7%
formaldehyde and the percentage of cells harboring polyQ-CFP foci was determined. CFP
foci were counted without the aid of a camera, and any cell with a visible focus was counted
as a cell harboring a focus. At least 100 were counted for each replicate.
Cell Imaging.
For the analysis of Hsp104-eGFP and VHL-GFP foci, cells were fixed in 3.7%
formaldehyde by adding 0.1 ml of 37% formaldehyde to 1 mL cells. Cells were then
permeabilized in 1%
Triton/Potassium phosphate washed and re-suspended in DAPI
/KPi/Sorbitol Microscopy was performed using a Zeiss Axioplan 2 microscope with a
Hamamatsu OCRA-ER digital camera. Image analysis was performed with Openlab 4.0.2
software.
Northern blot analysis.
Total RNA was purified by phenol extraction and isopropanol precipitation as described in
(Hochwagen et al. 2005). 10pg of total RNA were separated on a 1.1% agarose gel
containing 6% formaldehyde and 40 mM MOPS (pH 7.0). Gels were blotted in 1OX SSC
(1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate [pH = 7.0]) onto Hybond-XL
95
membranes (Amersham Biosciences). Blots were probed overnight with radioactively labeled
HA C1 or GAL-HTT1(17aa)-FLA C-nQ-CFPspecific probes.
Heat Shock Microarrys.
Strains were grown in YEPD at 25 0 C to OD 60 =0.2, collected by filtration and shifted to
pre-warmed (37*C) YEPD. Samples were collected 0, 5, 15, and 30 min after temperature
shift. WT grown at 25 0 C was used as a reference for all samples. RNA preparation and
microarrays were performed as described previously (Torres et al. 2007). Briefly, total RNA
was ethanol precipitated and further purified over RNeasy columns (Qiagen). 325ng of RNA
were labeled using the Agilent Low RNA Input Fluorescent Linear Amplification Kit.
Reactions were performed as directed except half the recommended reaction volume and
one quarter the recommended Cy-CTP amount was used. Dye incorporation and yield were
measured with a Nanodrop spectrophotometer. Equal amounts of differentially-labeled
control and sample cDNA were combined such that each sample contained at least 2.5pmol
dye. Samples were fragmented, combined with hybridization buffer, and boiled for 5 minutes,
and applied to a microarray consisting of 60mer probes for each yeast open reading frame
(Agilent). Microarrays were rotated at 60*C for 17 hours in a hybridization oven (Agilent).
Arrays were then washed according to the Agilent SSPE wash protocol, and scanned on an
Agilent scanner. The image was processed using the default settings with Agilent Feature
Extraction software. All data analysis was performed using the resulting log 2 ratio data, and
filtered for spots called as significantly over background in at least one channel. Data were
normalized to account for the extra chromosomes as previously described (Torres et al.
2007). Data were mined for genes that comprise the heat-shock response Gasch et al. 2000a).
96
The full dataset has been deposited in the Gene Expression Omnibus under the accession:
GSE40073.
Acknowledgements
We thank B. Vincent, D. Jarosz and M. Duennwald for reagents;
J.
Boulin for technical
assistance; and S. Lindquist, F. Solomon, and members of the Amon Lab for suggestions
and critical reading of this manuscript. This work was supported by the National Institute of
Health (GM056800 to A.A) and a Ludwig Fund Graduate Fellowship (to A.O.). A.A is an
investigator of the Howard Hughes Medical Institute.
Strains used in this study. All straisn are of the W303 background
Strain Number
A22361
A6863
A6865
A24367
Disome
A28265
V
A27036
VIII
A13975
IX
A21986
X
A28266
XI
A12694
XII
A21987
XIII
-
I
II
IV
Relevant Genotype
MATa, adel::HIS3, lys2::KanMX6
MATa, adel::HIS3, ade1::KanMX6
MATa, lys2::HIS3, lys2::KanMIX6
MATa, trpl::HIS3, trpl::KanMX6
MATa, canl::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMX6
MATa, intergenic region (119778-119573) between
YHRO06W and YHRO07C::HIS3, intergenic region (119778119573) between YHRO06W and YHR007C::KanMX6
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6
MATa, intergenic region (322250-322350) between
YJL061W and YJL060W::HIS3, intergenic region (322250322350) between YJL061W and YJL060W::KanMX6
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKL006W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6
MATa, adel6::HIS3, ade16::KanMX6
MATa, intergenic region (309200-309300) between
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
97
YMR018W::KanMX6
A28344
XIV
A27930
A27096
A31392
A31393
A31394
A31395
XV
XVI
A31396
V
A31397
VIII
A31398
IX
A31399
X
-
I
II
IV
MATa, intergenic region (622880-622980) between
YNLO05C and YNLO04W::HIS3, intergenic region (622880622980) between YNLO05C and YNL0O4W::KanMX6
MATa, leu9::HIS3, leu9::KanMX6
MATa, met12::HIS3, metl2::KanMX6
MATa, adel::HIS3, lys2::KanMX6, HSP1 04-eGFP:KanMX6
MATa, adel::HIS3, adel::KanMX6, HSP104-eGFP:KanMX6
MATa, lys2::HIS3, lys2::KanMX6, HSP104-eGFP:KanMX6
MATa, trp1::HIS3, trpl::KanMX6, HSP104-eGFP:KanMX6
MATa, cani::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMIX6, HSP104eGFP:KanMX6
MATa, intergenic region (119778-119573) between
YHR0O6W and YHR0O7C::HIS3, intergenic region (119778119573) between YHRO06W and YHRO07C::KanMX6
MATa, intergenic region (430900-431000) between
YKL006C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKL0O6W::KanMX6, HSP104-eGFP:KanMX6
MATa, intergenic region (322250-322350) between
YJL061W and YJL060W::HIS3, intergenic region (322250322350) between YJL061W and YJL060W::KanMX6,
HSP104-eGFP:KanMX6
A31400
XI
A31401
XII
A31402
XIII
A31403
XIV
A31404
XV
A31405
XVI
A31406
-
A31407
2n +11
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6, HSP104-eGFP:KanMX6
MATa, adel6::HIS3, ade16::KanMX6, HSP104eGFP:KanMX6
MATa, intergenic region (309200-309300) between
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6, HSP104-eGFP:KanMX6
MATa, intergenic region (622880-622980) between
YNLO05C and YNLO04W::HIS3, intergenic region (622880622980) between YNLO05C and YNLO04W::KanMX6,
HSP104-eGFP:KanMX6
MATa, leu9::HIS3, leu9::KanMX6, HSP104-eGFP:KanMX6
MATa, metl2::HIS3, met12::KanMX6, HSP104eGFP:KanMX6
MATa/a,MATa, adel::HIS3, lys2::KanMX6, HSP104eGFP:KanMX6
MATa/a, lys2::HIS3, lys2::KanMX6,lys2::LEU2, HSP104eGFP:KanMX6
98
MATa/cc, intergenic region (322250-322350) between
YJL061W and YJL060W::HIS3, intergenic region (322250322350) between YJL061W and YJL060W::KanMX6,
intergenic region (322250-322350) between YJL061W and
YJL060W::LEU2, HSP104-eGFP:KanMX6
MATa/a, intergenic region (430900-431000) between
YKL006C-A and YKL0O6W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6, ntergenic region (430900-431000)
between YKLO06C-A and YKLO06W::LEU2, HSP104eGFP:KanMX6
MATa/a, leu9::HIS3, leu9::KanMX6, leu9::LEU2,HSP104eGFP:KanMX6
MATa/a, HSP104-eGFP:KanMX6
MATa/a/a, HSP104-eGFP:KanMX6
MATa, hspl04::hspl04-eGFP-KAN adel::HIS3, lys2::KAN
/YAC-6
MATa, hspl04::hsplO4-eGFP-KANadel::HIS3, lys2::KAN
/YAC-3
MATa,, cdc28-4
A31408
2n+X
A31409
2n+XI
A31410
2n+XV
A28220
A28219
-
A28922
YAC-6
A28925
YAC-3
A29765
-
A29766
-
A25654
-
MATa, Hspl04-eGFP::KanMX
A32076
-
MATa, adel::HIS3, lys2::KanMX6, pdr5::TRP1, YCP:
A32077
I
MATa, adel::HIS3, adel::KanMX6 pdr5::TRP1, YCP:
pGAL-VHL-GFP:LEU2
A32078
II
A32079
IV
A32080
-
Hspl04-eGFP::KanMX
V
MATa,, cdc23-1, Hspl04-eGFP::KanMX
pGAL-VHL-GFP:LEU2
MATa, lys2::HIS3, lys2::KanMX6 pdr5::TRP1, YCP: pGAL-
VHL-GFP:LEU2
MATa, trpl::HIS3, trpl::KanMX6 pdr5::TRP1, YCP: pGAL-
VHL-GFP:LEU2
MATa, canl::HIS3, intergenic region (187520-187620)
between YERO15W and YER016W::KanMX6 pdr5::TRP1,
YCP: pGAL-VHL-GFP:LEU2
A32081
ViII
MATa, intergenic region (119778-119573) between
YHRO06W and YHRO07C::HIS3, intergenic region (119778119573) between YHRO06W and YHRO07C::KanMX6
pdr5::TRP1, YCP: pGAL-VHL-GFP:LEU2
MATa, intergenic region (430900-431000) between
A32082
IX
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6, pdr5::TRP1. YCP: pGAL-VHLGFP:LEU2
MATa, intergenic region (322250-322350) between
A32083
X
1_
YJL061W and YJL060W::HIS3, intergenic region (3222501 322350) between YJL061W and YJL060W::KanMX6
99
pdr5::TRP1, YCP: pGAL-VHL-GFP:LEU2
A32084
XI
MATa, intergenic region (430900-431000) between
YKL006C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKL006C-A and
YKL0O6W::KanMX6, pdr5::TRP1, YCP: pGAL-VHLGFP:LEU2
MATa, adel6::HIS3, adel6::KanMX6 pdr5::TRP1, YCP:
A32085
XII
A32086
XIII
A32087
XIV
A32088
XV
A32089
XVI
A30465
YAC-5
Mata, YCP: pGAL-VHL-GFP:LEU2, pdr5::TRP1,
adel::HIS3, lys2::KAN/YAC3
A30467
YAC-6
Mata, YCP: pGAL-VHL-GFP:LEU2, pdr5::TRP1,
pGAL-VHL-GFP:LEU2
MATa, intergenic region (309200-309300) between
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6 pdr5::TRP1, YCP: pGAL-VHLGFP:LEU2
MATa, intergenic region (622880-622980) between
YNLO05C and YNLO04W::HIS3, intergenic region (622880622980) between YNLO05C and YNLO04W::KanMX6,
pdr5::TRP1, YCP: pGAL-VHL-GFP:LEU2
MATa, leu9::HIS3, leu9::KanMX6 pdr5::TRP1, YCP: pGALVHL-GFP:LEU2
MATa, metl2::HIS3, metl2::KanMX6 pdr5::TRP1, YCP:
pGAL-VHL-GFP:LEU2
adel::HIS3, lys2::KAN/YAC6
A30461
Mata, YCP: pGAL-VHL-GFP:LEU2, pdr5::TRP1,
A30462
Mata, YCP: pGAL-VHL-GFP:LEU2, pdr5::TRP1,
adel::HIS3, lys2::KAN, cdc23-1
MATa/a,YCP: pGAL-VHL-GFP:LEU2, pdr5::TRP1
MATa/a/a,YCP: pGAL-VHL-GFP:LEU2, pdr5::TRP1
MATa, adel::HIS3, lys2::KanMX6, pdr5::TRP1
MATa, adel::HIS3, adel::KanMX6, pdr5::TRP1
MATa, lys2::HIS3, lys2::KanMX6, pdr5::TRP1
MATa, trpl::HIS3, trpl::KanMX6, pdr5::TRP1
adel::HIS3, lys2::KAN, cdc28-4
A28388
A28389
A15549
A15551
Al 5553
Al 5555
I
II
IV
Al 5557
vMATa,
A15559
ViII
canl::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMX6, pdr5::TRP1
MATa, intergenic region (119778-119573) between
YHRO06W and YHRO07C::HIS3, intergenic region (119778119573) between YHRO06W and YHRO07C::KanMX6,
pdr5::TRP1
A15561
IX
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6, pdr5::TRP1
100
A15563
X
A15566
XII
A15567
xIII
MATa, intergenic region (322250-322350) between
YJL061W and YJL060W::HIS3, intergenic region (322250322350) between YJL061W and YJL060W::KanMX6,
pdr5::TRP1
MATa, adel6::HIS3, adel6::KanMX6, pdr5::TRP1
MATa, intergenic region (309200-309300) between
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6, pdr5::TRP1
Al 5569
XIV
A15571
A15573
A32090
A32091
A32092
XV
XVI
A32093
V
-
I
II
MATa, intergenic region (622880-622980) between
YNLO05C and YNLO04W::HIS3, intergenic region (622880622980) between YNLO05C and YNL0O4W::KanMX6,
pdr5::TRP1
MATa, leu9::HIS3, leu9::KanMX6, pdr5::TRP1
MATa, met12::HIS3, met12::KanMX6, pdr5::TRP1
MATa, adel::HIS3, lys2::KanMX6, YCP: pGAL-csrc:URA3
MATa, adel::HIS3, adel::KanMX6, YCP: pGAL-csrc:URA3
MATa, lys2::HIS3, lys2::KanMX6, YCP: pGAL-csrc:URA3
MATa, canl::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMX6, YCP: pGALcsrc:URA3
A32094
ViII
MATa, intergenic region (119778-119573) between
YHRO06W and YHRO07C::HIS3, mtergenic region (119778119573) between YHRO06W and YHRO07C::KanMX6,
YCP: pGAL-csrc:URA3
MATa, intergenic region (430900-431000) between
A32095
IX
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX, YCP: pGAL-csrc:URA36
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6, YCP: pGAL-csrc:URA3
MATa, adel6::HIS3, adel6::KanMX6, YCP: pGAL-
A32096
XI
A32097
XII
A32098
XIII
A32099
XIV
A32100
XV
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6, YCP: pGAL-csrc:URA3
MATa, intergenic region (622880-622980) between
YNLO05C and YNLO04W::HIS3, intergenic region (622880622980) between YNLO05C and YNLO04W::KanMX6,
YCP: pGAL-csrc:URA3
MATa, leu9::HIS3, leu9::KanMX6, YCP: pGAL-csrc:URA3
A32101
XVI
MATa, metl2::HIS3, metl2::KanMX6, YCP: pGAL-
A32102
-
MATa, adel::HIS3, lys2::KanMIX6, YCP: pGAL-vsrc:URA3
csrc:URA3
MATa, intergenic region (309200-309300) between
csrc:URA3
101
A32103
A32104
I
II
A32105
V
A32106
ViII
A32107
IX
A32108
XI
A32109
XI
A321 10
XIII
A321 11
XIV
A32112
XV
A32113
XVI
A31114
-
A29843
I
A29844
II
A29845
V
A29846
ViII
A29847
IX
MATa, adel::HIS3, adel::KanMX6, YCP: pGAL-vsrc:URA3
MATa, lys2::HIS3, lys2::KanM.X6, YCP: pGAL-vsrc:URA3
MATa, cani::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMX6, YCP: pGALvsrc:URA3
MATa, intergenic region (119778-119573) between
YHR0O6W and YHR007C::HIS3, intergenic region (119778119573) between YHRO06W and YHRO07C::KanMX6,
YCP: pGAL-vsrc:URA3
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX, YCP: pGAL-vsrc:URA36
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6, YCP: pGAL-vsrc:URA3
MATa, adel6::HIS3, adel6::KanMX6, YCP: pGALvsrc:URA3
MATa, intergenic region (309200-309300) between
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6, YCP: pGAL-vsrc:URA3
MATa, intergenic region (622880-622980) between
YNLO05C and YNLO04W::HIS3, intergenic region (622880622980) between YNLO05C and YNLO04W::KanMX6,
YCP: pGAL-vsrc:URA3
MATa, leu9::HIS3, leu9::KanMX6, YCP: pGAL-vsrc:URA3
MATa, netl2::HIS3, metl2::KanMX6, YCP: pGALvsrc:URA3
MATa, adel::HIS3, lys2::KanMX6, ura3::
stop2xEGFP::URA, sup35::sup35-R2E2
MATa, adel::HIS3, adel::KanMX6, ura3::
stop2xEGFP::URA, sup35::sup35-R2E2
MATa, lys2::HIS3, lys2::KanMX6, ura3:: stop2xEGFP::URA,
sup35::sup35-R2E2
MATa, canl::HIS3, intergenic region (187520-187620)
between YERO15W and YERO16W::KanMX6, ura3::
stop2xEGFP::URA, sup35::sup35-R2E2
MATa, intergenic region (119778-119573) between
YHRO06W and YHRO07C::HIS3, intergenic region (119778119573) between YHRO06W and YHR07C::KanMX6, ,
ura3:: stop2xEGFP::URA, sup35::sup35-R2E2
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKL06W::KanMX6, , ura3:: stop2xEGFP::URA,
102
sup35::sup35-R2E2
X
A29848
MIATa, intergenic region (322250-322350) between
YJL061W and YJL060W::HIS3, intergenic region (322250322350) between YJL061W and YJL060W::KanMX6, , ura3::
stop2xEGFP::URA, sup35::sup35-R2E2
XI
A31110
MATa, intergenic region (430900-431000) between
YKL0O6C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6, , ura3:: stop2xEGFP::URA,
sup35::sup35-R2E2
A31 111
XII
A29849
XIII
MATa, adel6::HIS3, adel6::KanMX6, , ura3::
stop2xEGFP::URA, sup35::sup35-R2E2
MATa, intergenic region (309200-309300) between
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6, ura3:: stop2xEGFP::URA,
sup35::sup35-R2E2
XCIV
A31112
MATa, intergenic region (622880-622980) between
YNLO05C and YNLO04W::HIS3, intergenic region (622880622980) between YNLO05C and YNLO04W::KanMX6,,
ura3:: stop2xEGFP::URA, sup35::sup35-R2E2
-
kMATa, leu9::HIS3, leu9::KanMX6, ura3::
stop2xEGFP::URA, sup35::sup35-R2E2
MATa, metl2::HIS3, metl2::KanMX6, ura3::
stop2xEGFP::URA, sup35::sup35-R2E2
MATa, adel::HIS3, lys2::KanMX6, GAL-FLAG-
A321 15
II
MATa, lys2::HIS3, lys2::KanMX6, GAL-FLAGHttl(17AA)25QApro-CFP:URA3
A32116
V
MATa, canl::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMX6, GAL-
A29450
XV
A31113
XVI
A32114
A32114
_
-
Httl(17AA)25QApro-CFP:URA3
FLAG-Htt1(17AA)25QApro-CFP:URA3
A32117
VIII
MATa, intergenic region (119778-119573) between
YHRO06W and YHRO07C::HIS3, intergenic region (119778119573) between YHRO06W and YHRO07C::KanMX6
A32118
XI
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLOO6W::KanMX6, GAL-FLAG-Httl(17AA)25QAproCFP:URA3
A32119
XIII
MATa, intergenic region (622880-622980) between
YNLO05C and YNLO04W::HIS3, intergenic region (622880622980) between YNLO05C and YNLO04W::KanMX6,
GAL-FLAG-Httl(17AA)25QApro-CFP:URA3
103
A32120
XVI
A32121
~
A32122
1MATa,
A32123
V
A32124
VIII
A32125
XI
A32126
XIII
A32127
XIIV
A32128
XVI
A32129
-
A32130
II
A32131
V
A32132
ViII
A32133
XI
MATa, met12::HIS3, met12::KanMX6, GAL-FLAGHttl(17AA)25QApro-CFP:URA3
MATa, adel::HIS3, lys2::KanMX6, GAL-FLAGHttl(17AA)46QApro-CFP:URA3
lys2::HIS3, lys2::KanMX6, GAL-FLAGHttl(17AA)46QApro-CFP:URA3
MATa, canl::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMX6, GALFLAG-Httl(17AA)46QApro-CFP:URA3
MATa, intergenic region (119778-119573) between
YHRO06W and YHRO07C::HIS3, intergenic region (119778119573) between YHRO06W and YHRO07C::KanMX6
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6, GAL-FLAG-Httl(17AA)46QAproCFP:URA3
MATa, intergenic region (309200-309300) between
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6, GAL-FLAG-Httl(17AA)46QAproCFP:URA3
MATa, intergenic region (622880-622980) between
YNLO05C and YNLO04W::HIS3, intergenic region (622880622980) between YNLO05C and YNLO04W::KanMX6,
GAL-FLAG-Httl (1 7AA)46QApro-CFP:URA3
MATa, metl2::HIS3, met12::KanMX6, GAL-FLAGHttl(17AA)46QApro-CFP:URA3
MATa, adel::HIS3, lys2::KanMX6, GAL-FLAGHttl(17AA)72QApro-CFP:URA3
MIATa, lys2::HIS3, lys2::KanMIX6, GAL-FLAGHttl (17AA)72QApro-CFP:URA3
MIATa, canl::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMX6, GALFLAG-Httl(17AA)72QApro-CFP:URA3
MATa, intergenic region (119778-119573) between
YHRO06W and YHRO07C::HIS3, intergenic region (119778119573) between YHRO06W and YHRO07C::KanMX6
GAL-FLAG-Htt (17AA)72QApro-CFP:URA3
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6, GAL-FLAG-Httl(17AA)72QAproCFP:URA3
104
A32134
XIII
A32135
XVI
A32407
A32408
II
A32409
V
A32410
ViII
A32411
X
A32412
XI
A32413
XIII
A32414
A32415
A32416
XV
XVI
A32417
V
A32418
ViII
A32419
X
A32420
XIII
A32421
XV
-
MATa, intergenic region (309200-309300) between
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6, GAL-FLAG-Httl(17AA)72QAproCFP:URA3
MATa, metl2::HIS3, met12::KanMX6, GAL-FLAGHttl(17AA)72QApro-CFP:URA3
MATa, adel::HIS3, lys2::KanMX6, SSA1-3HA:TRP1
MATa, lys2::HIS3, lys2::KanMX6 SSA1-3HA:TRP1
MATa, canl::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMX6, SSA13HA:TRP1
MATa, intergenic region (119778-119573) between
YHRO06W and YHRO07C::HIS3, intergenic region (119778119573) between YHRO06W and YHRO07C::KanMX6,
SSA1-3HA:TRP1
MATa, intergenic region (322250-322350) between
YJL061W and YJL060W::HIS3, intergenic region (322250322350) between YJL061W and YJL060W::KanMX6, SSA13HA:TRP1
MATa, intergenic region (430900-431000) between
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKLO06W::KanMX6, SSA1-3HA:TRP1
MATa, intergenic region (309200-309300) between
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6, SSA1-3HA:TRP1
MATa, leu9::HIS3, leu9::KanMX6, SSA1-3HA:TRP1
MATa, metl2::HIS3, metl2::KanMX6, SSA1-3HA:TRP1
MATa, adel::HIS3, lys2::KanMX6, SSE2-3HA:TRP1
MATa, canl::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMX6,SSE23HA:TRP1
MATa, intergenic region (119778-119573) between
YHRO06W and YHRO07C::HIS3, intergenic region (119778119573) between YHRO06W and
YHR07C::KanMX6,SSE2-3HA:TRP1
MATa, intergenic region (322250-322350) between
YJL061W and YJL060W::HIS3, intergenic region (322250322350) between YJL061W and YJL060W::KanMX6, SSE23HA:TRP1
MATa, intergenic region (309200-309300) between
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6,
MATa, leu9::HIS3, leu9::KanMX6, SSE2-3HA:TRP1
105
A32422
A32423
A32424
XVI
II
A32425
ViII
MATa, metl2::HIS3, met12::KanMX6, SSE2-3HA:TRP1
MATa, adel::HIS3, lys2::KanMX6, HSP42-3HA:TRP1
MATa, lys2::HIS3, lys2::KanMX6, HSP42-3HA:TRP1
MATa, intergenic region (119778-119573) between
YHRO06W and YHR007C::HIS3, intergenic region (119778119573) between YHR0O6W and
YHR0O7C::KanMX6,HSP42-3HA:TRP1
MATa, intergenic region (309200-309300) between
A32426
XIII
A32427
XIV
A32428
XVI
A33369
-
YMR017W and YMR018W::HIS3, intergenic region
(309200-309300) between YMR017W and
YMR018W::KanMX6, HSP42-3HA:TRP1
MATa, intergenic region (622880-622980) between
YNL0O5C and YNLO04W::HIS3, intergenic region (622880622980) between YNLO05C and YNL0O4W::KanMX6,
HSP42-3H-A:TRP1
MATa, metl2::HIS3, met12::KanMX6, HSP42-3HA:TRP1
MATa, adel::HIS3, lys2::KanMX6, ubp6::TRP1, HSP104-
A33370
II
MATa, lys2::HIS3, lys2::KanMIX6, ubp6::TRP1, HSP104-
V
MATa, canl::HIS3, intergenic region (187520-187620)
between YER015W and YER016W::KanMX6, ubp6::TRP1,
A33371
eGFP:KanMX6
eGFP:KanMX6
HSP104-eGFP:KanMX6
MATa, intergenic region (430900-431000) between
A33372
IX
YKLO06C-A and YKLO06W::HIS3, intergenic region
(430900-431000) between YKLO06C-A and
YKL06W::KanMX6,ubp6::TRP1, HSP104-eGFP:KanMX6
106
References
Baker DJ, Jeganathan KB, Cameron JD, Thompson M, Juneja S, Kopecka A, Kumar R,
Jenkins RB, de Groen PC, Roche P et al. 2004. BubR1 insufficiency causes early
onset of aging-associated phenotypes and infertility in mice. Nature genetics 36: 744-
749.
Borkovich KA, Farrelly FW, Finkelstein DB, Taulien
J,
Lindquist S. 1989. hsp82 is an
essential protein that is required in higher concentrations for growth of cells at
higher temperatures. Molecularand cellular biology 9: 3919-3930.
Boulon S, Pradet-Balade B, Verheggen C, Molle D, Boireau S, Georgieva M, Azzag K,
Robert MC, Ahmad Y, Neel H et al. 2010. HSP90 and its R2TP/Prefoldin-like
cochaperone are involved in the cytoplasmic assembly of RNA polymerase II.
Molecular cell 39: 912-924.
Cox B, Ness F, Tuite M. 2003. Analysis of the generation and segregation of propagons:
entities that propagate the [PSI+] prion in yeast. Genetics 165: 23-33.
Dai C, Whitesell L, Rogers AB, Lindquist S. 2007. Heat shock factor 1 is a powerful
multifaceted modifier of carcinogenesis. Cell 130: 1005-1018.
Duennwald ML, Jagadish S, Muchowski PJ, Lindquist S. 2006. Flanking sequences
profoundly alter polyglutamine toxicity in yeast. Proceedings of the NationalAcademy of
Sciences of the United States ofAmerica 103: 11045-11050.
Duyao M, Ambrose C, Myers R, Novelletto A, Persichetti F, Frontali M, Folstein S, Ross C,
Franz M, Abbott M et al. 1993. Trinucleotide repeat length instability and age of
onset in Huntington's disease. Naturegenetics 4: 387-392.
Epstein CJ, Foster DB, DeArmond SJ, Prusiner SB. 1991. Acceleration of scrapie in trisomy
16----diploid aggregation chimeras. Ann Neurol 29: 95-97.
Foote SJ, Thompson JK, Cowman AF, Kemp DJ. 1989. Amplification of the multidrug
resistance gene in some chloroquine-resistant isolates of P. falciparum. Cell 57: 921-
930.
Franzosa EA, Albanese V, Frydman J, Xia Y, McClellan AJ. 2011. Heterozygous yeast
deletion collection screens reveal essential targets of Hsp90. PloS one 6: e2821 1.
Gasch AP, Spellman PT, Kao CM, Carmel-Harel 0, Eisen MB, Storz G, Botstein D, Brown
PO. 2000a. Genomic expression programs in the response of yeast cells to
environmental changes. Molecular biology of the cell 11: 4241-4257.
-. 2000b. Genomic expression programs in the response of yeast cells to environmental
changes. Molecularbiology of the cell 11: 4241-4257.
Geiler-Samerotte KA, Dion MF, Budnik BA, Wang SM, Hard DL, Drummond DA. 2011.
Misfolded proteins impose a dosage-dependent fitness cost and trigger a cytosolic
unfolded protein response in yeast. Proceedings of the NationalAcademy of Sciences of the
United States ofA merica 108: 680-685.
Goedert M, Clavaguera F, Tolnay M. 2010. The propagation of prion-like protein inclusions
in neurodegenerative diseases. Trends Neurosci 33: 317-325.
Group HsDCR. 1993. A novel gene containing a trinucleotide repeat that is expanded and
unstable on Huntington's disease chromosomes. The Huntington's Disease
Collaborative Research Group. Cell 72: 971-983.
Halfmann R, Alberti S, Lindquist S. 2010. Prions, protein homeostasis, and phenotypic
diversity. Trends in cell biology 20: 125-133.
107
Hanna J, Hathaway NA, Tone Y, Crosas B, Elsasser S, Kirkpatrick DS, Leggett DS, Gygi SP,
King RW, Finley D. 2006. Deubiquitinating enzyme Ubp6 functions noncatalytically
to delay proteasomal degradation. Cell 127: 99-111.
Hard FU, Bracher A, Hayer-Hartl M. 2011. Molecular chaperones in protein folding and
proteostasis. Nature 475: 324-332.
Hochwagen A, Wrobel G, Cartron M, Demougin P, Niederhauser-Wiederkehr C, Boselli
MG, Primig M, Amon A. 2005. Novel response to microtubule perturbation in
meiosis. Mol Cel/Biol25: 4767-4781.
Houck SA, Singh S, Cyr DM. 2012. Cellular responses to misfolded proteins and protein
aggregates. Methods in molecularbiology 832: 455-461.
Imai J, Maruya M, Yashiroda H, Yahara I, Tanaka K. 2003. The molecular chaperone Hsp90
plays a role in the assembly and maintenance of the 26S proteasome. The EMBO
journal22: 3557-3567.
Kaganovich D, Kopito R, Frydman J. 2008. Misfolded proteins partition between two
distinct quality control compartments. Nature 454: 1088-1095.
Kimura Y, Yahara I, Lindquist S. 1995. Role of the protein chaperone YDJ1 in establishing
Hsp90-mediated signal transduction pathways. Science 268: 1362-1365.
Li M, Fang X, Wei Z, York JP, Zhang P. 2009. Loss of spindle assembly checkpointmediated inhibition of Cdc20 promotes tumorigenesis in mice. J Cell Biol 185: 983-
994.
Liebman SW, Sherman F. 1979. Extrachromosomal psi+ determinant suppresses nonsense
mutations in yeast. J Bacteriol139: 1068-1071.
Liu B, Larsson L, Caballero A, Hao X, Oling D, Grantham J, Nystrom T. 2010. The
polarisome is required for segregation and retrograde transport of protein aggregates.
Cell 140: 257-267.
Liu JJ, Lindquist S. 1999. Oligopeptide-repeat expansions modulate 'protein-only' inheritance
in yeast. Nature 400: 573-576.
Longtine MS, Fares H, Pringle JR. 1998. Role of the yeast Gin4p protein kinase in septin
assembly and the relationship between septin assembly and septin function. The
Journalof cell biology 143: 719-736.
McClellan AJ, Scott MD, Frydman J. 2005. Folding and quality control of the VHL tumor
suppressor proceed through distinct chaperone pathways. Cell 121: 739-748.
McClellan AJ, Xia Y, Deutschbauer AM, Davis RW, Gerstein M, Frydman J. 2007. Diverse
cellular functions of the Hsp90 molecular chaperone uncovered using systems
approaches. Cell131: 121-135.
Nathan DF, Lindquist S. 1995. Mutational analysis of Hsp90 function: interactions with a
steroid receptor and a protein kinase. Mol Cell Biol15: 3917-3925.
Nathan DF, Vos MH, Lindquist S. 1997. In vivo functions of the Saccharomyces cerevisiae
Hsp90 chaperone. ProcNat/A cad Sci USA 94: 12949-12956.
Neckers L. 2007. Heat shock protein 90: the cancer chaperone. J Biosci 32: 517-530.
Niwa 0, Tange Y, Kurabayashi A. 2006. Growth arrest and chromosome instability in
aneuploid yeast. Yeast 23: 937-950.
Parry EM, Cox BS. 1970. The tolerance of aneuploidy in yeast. Genet Res 16: 333-340.
Pavelka N, Rancati G, Zhu J, Bradford WD, Saraf A, Florens L, Sanderson BW, Hattem GL,
Li R. 2010. Aneuploidy confers quantitative proteome changes and phenotypic
variation in budding yeast. Nature 468: 321-325.
Peth A, Besche HC, Goldberg AL. 2009. Ubiquitinated proteins activate the proteasome by
binding to Uspl4/Ubp6, which causes 20S gate opening. Mo Cell 36: 794-804.
108
Powers MV, Clarke PA, Workman P. 2008. Dual targeting of HSC70 and HSP72 inhibits
HSP90 function and induces tumor-specific apoptosis. Cancer Cell 14: 250-262.
Rehen SK, McConnell MJ, Kaushal D, Kingsbury MA, Yang AH, Chun
J.
2001.
Chromosomal variation in neurons of the developing and adult mammalian nervous
system. Proceedings of the NationalAcademy of Sciences of the United States of America 98:
13361-13366.
Rehen SK, Yung YC, McCreight MP, Kaushal D, Yang AH, Almeida BS, Kingsbury MA,
Cabral KM, McConnell MJ, Anliker B et al. 2005. Constitutional aneuploidy in the
normal human brain. The Journal of neuroscience : the official journal of the Society for
Neuroscience 25: 2176-2180.
Sheltzer JM, Blank HM, Pfau SJ, Tange Y, George BM, Humpton TJ, Brito IL, Hiraoka Y,
Niwa 0, Amon A. 2011. Aneuploidy drives genomic instability in yeast. Science 333:
1026-1030.
Sheltzer JM, Torres EM, Dunham MJ, Amon A. 2012. Transcriptional consequences of
aneuploidy. Proceedings of the NationalAcademy of Sciences of the United States of America
109: 12644-12649.
St Charles J, Hamilton ML, Petes TD. 2010. Meiotic chromosome segregation in triploid
strains of Saccharomyces cerevisiae. Genetics 186: 537-550.
Stingele S, Stoehr G, Peplowska K, Cox J, Mann M, Storchova Z. 2012. Global analysis of
genome, transcriptome and proteome reveals the response to aneuploidy in human
cells. Mol Syst Biol 8: 608.
Tang YC, Williams BR, Siegel
JJ,
Amon A. 2011. Identification of aneuploidy-selective
antiproliferation compounds. Cell 144: 499-512.
Thompson SL, Compton DA. 2008. Examining the link between chromosomal instability
and aneuploidy in human cells. The Journalof cell biology 180: 665-672.
Torres EM, Dephoure N, Panneerselvam A, Tucker CM, Whittaker CA, Gygi SP, Dunham
MJ, Amon A. 2010. Identification of aneuploidy-tolerating mutations. Cell143: 71-83.
Torres EM, Sokolsky T, Tucker CM, Chan LY, Boselli M, Dunham MJ, Amon A. 2007.
Effects of aneuploidy on cellular physiology and cell division in haploid yeast. Science
317: 916-924.
Torres EM, Williams BR, Amon A. 2008. Aneuploidy: cells losing their balance. Genetics 179:
737-746.
Tyedmers J, Madariaga ML, Lindquist S. 2008. Prion switching in response to environmental
stress. PLoS Biol6: e294.
Tyedmers J, Mogk A, Bukau B. 2010. Cellular strategies for controlling protein aggregation.
Nat Rev Mol Cell Biol 11: 777-788.
Weaver BA, Cleveland DW. 2006. Does aneuploidy cause cancer? Curr Opin Cell Biol 18: 658-
667.
Williams BR, Amon A. 2009. Aneuploidy: cancer's fatal flaw? CancerRes 69: 5289-5291.
Williams BR, Prabhu VR, Hunter KE, Glazier CM, Whittaker CA, Housman DE, Amon A.
2008. Aneuploidy affects proliferation and spontaneous immortalization in
mammalian cells. Science 322: 703-709.
Workman P, Burrows F, Neckers L, Rosen N. 2007. Drugging the cancer chaperone HSP90:
combinatorial therapeutic exploitation of oncogene addiction and tumor stress. Ann
NYA cad Sci 1113: 202-216.
Xu Y, Lindquist S. 1993. Heat-shock protein hsp90 governs the activity of pp60v-src kinase.
Proceedingsof the NationalA cademy of Sciences of the United States ofA merica 90: 7074-707 8.
109
Yurov YB, Iourov IY, Vorsanova SG, Liehr T, Kolotii AD, Kutsev SI, Pellestor F,
Beresheva AK, Demidova IA, Kravets VS et al. 2007a. Aneuploidy and confined
chromosomal mosaicism in the developing human brain. PloS one 2: e558.
Yurov YB, Vorsanova SG, Iourov IY, Demidova IA, Beresheva AK, Kravetz VS,
Monakhov VV, Kolotii AD, Voinova-Ulas VY, Gorbachevskaya NL. 2007b.
Unexplained autism is frequently associated with low-level mosaic aneuploidy. J Med
Genet 44: 521-525.
Zhu
J,
Pavelka N, Bradford WD, Rancati G, Li R. 2012. Karyotypic determinants of
chromosome instability in aneuploid budding yeast. PLoSgenetics 8: e1002719.
110
Chapter 3:
Conclusions and Future Directions
111
Summary of key conclusions
The study of aneuploidy has advanced tremendously in the last few years, in large
part due to systematic and rigorous analyses of aneuploid cells with diverse karyotypes. The
ability to generate a diverse panel of aneuploid Saccharoyces cerevisiae strains (both stable
aneuploid strains of a defined karyotype and highly aneuploid albeit genomically unstable
strains) and the many tools that have been developed to study cellular biology in this model
organism have, in no doubt, enabled the field to pursue questions that had previously never
been addressed. Using this approach, we have been able to separate the phenotypes that are
consequences of specific karyotypic imbalances, i.e. due to the karyotype itself and those that
are general consequences of harboring an imbalanced genome, irrespective of which
chromosomes are in excess. The research in this thesis expands upon our knowledge of how
aneuploidy interacts with, and the consequences it has on the basic cellular process of
protein homeostasis. Understanding the relationship between the maintenance of genomic
integrity and proteostasis has provided us with great insight that may, in part, explain why
aneuploidy has such striking effects on cellular fitness. Since protein homeostasis is a
dilemma that is encountered by all cells and the regulatory programs that govern it are well
conserved, the research discussed in this thesis may shed light on the effects of aneuploidy in
higher eukaryotes including humans. Furthermore, understanding the consequences of
aneuploidy on cellular physiology, and in particular on proteostasis could have a profound
effect on our view and treatment of human pathologies that harbor unbalanced karyotypes
such as Down's Syndrome and most solid tumors.
Protein production is carefully controlled in all cells. As this process requires an
extraordinary amount of energy and resources for both translating the peptide and folding it,
the cell rarely produces superfluous protein molecules. The protein quality control
112
machinery (chaperones and proteasome) are well calibrated to ensure protein homeostasis in
normal growth conditions and there are transcriptional programs that upregulate the cell's
quality control capacity that are initiated in conditions of acute proteotoxic stress. The
amount of protein molecules produced is regulated both at the transcription and translation
levels, and thus protein subunits that coalesce into protein complexes are generated at
stoichiometric levels. That is to say that if protein A binds forms a heterotrimeric complex
with two molecules of protein B, there will be approximately twice as many molecules of
protein B than of protein A in the cell. As I showed in Chapter 2, aneuploidy poses a severe
problem for protein homeostasis. The chromosomes in excess are, for the most part,
translated into protein that must then be folded and degraded by the quality control
machinery. Not only does the cell have to cope with excess protein molecules, it must tackle
the stoichiometric imbalances that are generated when complex subunits are expressed on
chromosomes that are present in different copy numbers. Aneuploidy results in the
formation of protein aggregates and we speculate that this is due to the increased folding
burden on the quality control machinery. I showed this effect specifically on the ubiquitous
chaperone Hsp90 and we determine that aneuploidy appears to exhaust Hsp90's folding
capacity. Furthermore, aneuploidy sensitizes cells to the detrimental effects of toxic hard-tofold proteins associated with human neurodegenerative disease suggesting that this
phenomenon may have implications beyond the study of cancer and developmental disease.
In this section, I will discuss several issues that arise from this research, centering the
discussion on 4 central questions: (1) How does aneuploidy exhaust the cell's folding
capacity? (2) What proteins comprise the protein aggregates found in aneuploid strains? (3)
How is the mammalian proteome affected by aneuploidy? (4) Is there a relationship between
aneuploidy, aging and neurodegenerative disease?
113
Aneuploidy exhausts the cell's protein quality control capacity
Why are aneuploid cells aggregate-prone?
In Chapter 2, I described how the alterations in the proteome brought along by
aneuploidy result in proteotoxic stress and lead to an accumulation of protein aggregates.
Aggregates could be a result of overwhelmed folding pathways or could stem from reduced
chaperone activity. . In vitro activity studies are needed to determine if the folding activity of
chaperones purified from aneuploid strains is significantly different than that of euploid
yeast strains. Both alternatives are possible, but given that chaperone activity is crucial to the
viability of the cell and that the aneuploid chromosomes are actively expressed, it is likely
that excess proteins produced from the aneuploid chromosomes occupy chaperones and
thereby reduce their availability to assist in the folding of their other clients What deternines
the extent of aggregate formation in aneuploid cells is not yet known. It does not appear to
correlate with either the degree of aneuploidy (by DNA content), total protein in excess,
delay in G1, or proliferation rate. However, it is important to bear in mind that our aggregate
analysis by HSP104-eGFPfoci does not measure absolute amounts of aggregated proteins in
cells and, more importantly, is not able to distinguish toxic from nontoxic aggregates. It is
interesting to note that the Environmental Stress Response (ESR) (Gasch et al. 2000), which
encompasses part of the heat- shock response, correlates with degree of aneuploidy (Torres
et al. 2007, Sheltzer, 2012 #4286). I propose that protein aggregate burden correlates with
the number of obligate chaperone clients and hence with the distribution of their encoding
genes in the genome. This hypothesis has proven difficult to test as our understanding of
client specificity for many chaperones is poor and we do not yet have a comprehensive list
of obligate chaperone clients. Additionally, chaperones have redundant functions and
114
although a client may generally have a defined folding pathway, it will get folded by
alternative means if those chaperones are non-functional or overwhelmed (Hard et al. 2011).
It may seem surprising that the cells cannot maintain proteostasis under the presence
of a single additional chromosome, which, depending on chromosome size, results in
20
/-
12% of the genome being imbalanced. Previous studies showed that even small amounts of
non-functional, seemingly non-toxic, misfolded proteins (Ura3, YFP) place a burden on the
cell's protein quality-control systems and hence adversely affect cellular fitness. Expression
of a single misfolded cytosolic protein at <0.1% of total protein leads to a significant
decrease in proliferative abilities and the induction of a cytoplasmic unfolded protein
response (Geiler-Samerotte et al. 2011). As aneuploid cells express full chromosomes that
are comprised of proteins that are functional, it is not surprising that even a 2% increase in
protein production placed on the protein quality control machinery is great could be
detrimental especially when consider the overproduction of proteins encoded on the extra
chromosomes that require chaperones for their function, such as protein kinases and WD40
repeat proteins.
It is important to keep in mind that excess protein molecules is not the only
proteomic alteration that aneuploid cells must tolerate. It is well established that many
subunits of protein complexes only acquire a stable conformation by binding to other
subunits of the complex
(Imai et al. 2003; Boulon et al. 2008). Thus, every single polypeptide
produced by genes located on aneuploid chromosomes that normally has a binding partner
is-in the disomes-in excess. Every single subunit that requires a chaperone to maintain
solubility until it can bind to its binding partner (that, in case of aneuploid cells, does not
exist) will occupy the chaperone indefinitely until it is degraded. I pose forth a model in
which it is this dramatic change in protein stoichiometries that leads to an increased burden
115
on the protein quality-control pathways of the cell (Figure 1). Individual subunits present in
excess require the continuous assistance of chaperones, titrating chaperones away from
assisting other folding reactions and reducing the general folding capacity of the cell, and
thus interfering with their essential function of mediating folding of essential proteins (Hard
et al. 2011). This model is supported by my finding that the aggregate phenotype was
ameliorated when the ratio of uncomplexed proteins to properly complexed proteins was
decreased by increasing base ploidy (as in trisomic 2n+1 strains, see Figure 1). This data
strongly suggests that the proteotoxicity observed in aneuploids is indeed in part the result of
the protein stoichiometry imbalances caused by aneuploidy, although it is also possible that
diploid cells are more efficient at clearing aggregates. In summary, aneuploidy impacts
protein homeostasis in multiple ways so that even small un-balanced chromosomes have a
significant impact on the cell's protein quality-control systems.
116
FIGURE 1 Stoichiometric imbalance increases the burden on quality control
pathways.
Aneuploidy causes stoichiometric imbalances, resulting in protein complex subunits that do
not have binding partners available. Partially folded subunits (represented in blue here) often
bind to chaperones to maintain stability while they await a binding partner, but if their
partner is non-existent, they will be bound to chaperones until they are degraded. When the
basal ploidy is increased (from n to 2n in this example) but the number of excess
chromosomes remains the same, the relative imbalance is reduced and the burden on the
cell's quality control machinery is also reduced.
Figure 1
Haploid
(n) cell
Clients
Protein complex
DNA
A
Aneuploid
(n+1) cell
OW
qW
VW 1W vw
W
4M Ift
1
1"W IM
Diploid
(2n) cell
Aneuploid
(2n+1) cell
1 /
5/
Q.CtrI.
capacity
2/
I.
I
f
4f
6/2
4
JO
-U
117
The folding capacity of chaperonesis alteredby genomic imbalances
In this thesis, I have shown that aneuploidy results in dramatic consequences for the
quality control machinery of the cell. Using assays that monitor Hsp90 activity in vivo I
showed that 8 out of 11 disomes (haploid +1 chromosome) tested showed reduced Hsp90
folding activity (Oromendia et al. 2012). This was surprising given that Hsp90 has been
shown to be highly abundant in yeast, representing 1-2% of total protein (Borkovich et al.
1989; Neckers 2007). It is not yet known why Hsp90 activity is limiting in so many different
disomic yeast strains. I can envision two possible scenarios: either aneuploidy affects the
chaperone's intrinsic catalytic activity and causes a reduction in folding capacity or the
presence of excess client proteins brought upon by aneuploidy depletes the Hsp90 reservoir.
In vitro folding assays are needed to exclude the possibility of a reduction in the folding
activity of the Hsp90 molecules but it would be surprising if 8 different aneuploidies caused
an intrinsic defect in a chaperone as essential as Hsp90.
Hsp90 is unlikely to the only chaperone system limiting in aneuploid yeast strains. I
hypothesize that, when analyzed, most chaperone families will be limiting in at least some
aneuploid strains. As with Hsp90, it would be of special interest to perform in vivo or in
whole extract chaperone assays to be able to assess whether the folding capacity (as opposed
to inherent biochemical activity) is impaired. In vivo/in extract protein folding assays have
been developed for numerous chaperone families: to assay TRiC/CCT activity in vivo one
can measure the relative ratio of native and misfolded actin (native actin binds with high
affinity to DNAseI coated beads). De novo folding of luciferase, and thus its activity,
requires Hsp70; by measuring luminescence after expression of luxAB, one can determine
the in vivo folding capacity of Hsp70. To measure the disaggregation capacity of Hsp104,
one can express luciferase and heat shock cells. The heat shock will cause aggregation of
118
luciferase and loss of luminescence, the recovery of luciferase activity (and luminescence) is
directly dependent on Hsp104. Performing these assays on a panel of aneuploid strains will
enable us to generate a data set that analyzes the activities of various chaperones when
different proteins are present in excess. Different aneuploidies will, of course, affect the
various chaperone families differently depending on the proportion of clients for a particular
chaperone affected by a given aneuploidy.
A neuploidy is a chronic stress, distinctform environmentalproteotoxicstressors
The proteotoxic stress that aneuploidy brings upon cells is distinct
from
environmental proteotoxic stressors. An environmental stressor such as high heat, or sharp
transitions into osmotic stress usually brings upon a sharp transition between a state of
proteostasis and one of severely altered and mis-folded proteins. For example, a 30 minute
exposure to 42C results in misfolding of over 50% of the yeast proteome (Richter et al.
2010). This abrupt change results in an immediate need for increased quality control capacity
relieved by the transcriptional up-regulation of chaperones and down-regulation of protein
synthesis that comprise the heat-shock response and similar responses in other organelles
(UPR, mitoUPR). These responses are costly to the cell, not only in chaperone production
and function but also in the lack of biomass produced while translation is dampened. As
environmental stressors are most often severe but transient, the transcriptional responses are
tailored to resolve proteotoxic stress and rapidly return to basal level so the cell can continue
with its normal function. As such, the heat shock response and other similar responses are
transient transcriptional programs tailored for resolving acute proteotoxic stress.
In Chapter 2, we show that although, when prompted, disomic yeast are capable of
mounting both the heat shock and unfolded protein response with similar kinetics as wild119
type cells they appear to be unable to maintain protein homeostasis under normal growth
conditions. This is not surprising as aneuploidy results in a mild, but persistent proteotoxic
stress. Saccharoyces cerevisiae cells that are exposed to a severe heat shock (42*C for 15 min)
accumulate large amounts of protein aggregates which can be quantified both by the
percentage of cells that harbor HSP104-eGFP aggregates (100% of cells contain many large
eGFP foci) or via a simple fractionation and visualization of total protein in the pellet
fraction on an SDS-PAGE gel stained for total protein with Coomassie stain (Figure 2). In
contrast, both the percentage of cells that contain HSP104-eCFP aggregates (Oromendia,
2012) and the accumulation of proteins in the pellet fraction are far less severe in aneuploid
cells. Whereas
environmental proteotoxic stressors result in transient but dramatic
transformations of the proteome, aneuploidy leads to persistent but relatively mild
alterations in protein homeostasis. We believe that it is this difference that prevents the
canonical transcriptional responses to protein misfolding (HSR, UPR) from being evoked.
The contrast between the proteotoxic stress caused by environmental insults and that
caused by aneuploidy prompts us to question whether general upregulation of protein quality
control machinery would be sufficient to ameliorate proteotoxicity in aneuploid yeast and,
perhaps, result in an improvement in proliferation rates. Torres et al. (Torres et al. 2007)have
previously shown that enhancing the degradation of highly abundant proteins by the
proteasome by deleting the ubiquitin ligase UBP6 is sufficient to improve growth rates in a
subset of disomes. We later showed that this deletion could ameliorate, but not fully
suppress the accumulation of protein aggregates in disomic yeast (Oromendia, 2012). In
future work, it will be of interest to assess the effects of enhancing the quality control
capacity of the cell by increasing the abundance or activity of protein chaperones on
aneuploid cells. This could be achieved by simply increasing the abundance of one, or a
120
subset of protein chaperones by increasing the copy number or placing them under the
control of a strong, constitutive promoter but as many chaperones function in large
complexes with many co-chaperones and accessory proteins an up-regulation of a subset of
chaperones may not be sufficient to improve protein quality control. An alternative
approach is to artificially induce a heat shock response in aneuploid cells; this would result of
up-regulation of a suite of chaperones and cofactors that are already poised to rescue cells
from proteotoxic distress. The master regulator of the heat shock response in S. cerevisiae is
the transcription factor Hsfl. Hyperphosphorylation and the resulting activation of Hsf1
results in the transcription of a large suite of genes involved in protein folding, carbohydrate
metabolism, and energy generation (reviewed in (Richter et al. 2010). Both the hyperactive
allele of HSFJ, hsf1147-833 (Sorger 1990) and the overexpression of an upstream regulator
GIP2 (Yeger-Lotem et al. 2009) have been shown to cause overexpression of HSF1 targets.
Generating aneuploid strains that contain these alleles and assessing whether an artificially
induced heat shock response reduces the percentage of cells harboring protein aggregates or
improves proliferative capacity would enable us to determine whether increasing protein
quality control capacity is sufficient to counteract the detrimental effects that aneuploidy has
on the proteome. Additionally, this would suggest that aneuploidy is simply over-burdening
the quality control machinery and titrating away chaperones from proteins that require them.
As mentioned before, the canonical responses to protein misfolding are energetically costly
and thus one major caveat of this reasoning is that, if increasing quality control capacity in
the cell by over-expressing chaperones or artificially inducing a heat shock response is too
taxing and requires too much of the cells' energy stores we may not be able to observe an
improvement in proliferation nor a reduction in protein aggregate accumulation.
121
The composition of protein aggregates in aneuploid yeast
Aneuploidy results in the accumulation of protein aggregates under conditions of
normal, non-stress growth conditions (Chapter 2) and this phenotype is exacerbated under
conditions of mild proteotoxic stress. This finding brings about an important issue that is yet
unresolved: which proteins comprise the protein aggregates and is there a reason why they
end their life in protein aggregates?
Two non-exclusive models could explain the identity of the proteins that result in the
protein aggregates in aneuploid cells: the aggregates in each aneuploid strain could be
comprised mainly of proteins that are encoded by the chromosome in excess or, the
aggregates found in all strains could have similar composition irrespective of the karyotype
of the cell. In the latter case, the aggregates could be formed mainly by proteins that are
especially difficult to fold and that are obligate clients of chaperones and when these are
over-burdened they do not manage to acquire native structure and terminate as aggregated
folding intermediates.
I have now managed to develop an aggregate purification protocol that is both
reproducible and comprehensive and that shows a clear differential between wild type and
disomic strains (Figure 2). Using this purification method, identifying the proteins that make
up the protein aggregates observed in aneuploid cells is now achievable via quantitative mass
spectrometry. Using the Stable Isotope Labeling by Amino acids in Cell Culture (SILAC)
technique one can purify aggregates from differentially labeled wild type and aneuploid
cultures and after mass spectrometry determine quantitatively which proteins comprise the
aggregate fraction of the cell. Once one has done this with a panel of different aneuploidies
that each have a different chromosome in excess, one can use bioinformatics to analyze the
datasets and determine whether they are more similar to each other (the identity of the
122
proteins that aggregate is irrespective of the karyotype) or they are more similar to the
proteins encoded by the particular chromosome present in additional copies (suggesting that
most proteins that terminate in aggregates were proteins that were in excess).
Elucidating the identity of the proteins that conform the protein aggregates in
aneuploid yeast may glean insight on how the cell determines which proteins will be
aggregated. If the protein composition of aggregates is different according to karyotype and
the proteins that aggregate are those that are present in excess because of the presence of an
additional copy of the chromosome that encodes for them, one can imagine two nonexclusive mechanisms via which they could terminate as aggregated forms. Given that they
are overabundant, proteins present in excess might be, by sheer stochasticity, more likely to
end up in protein aggregates, especially if they are highly abundant proteins in their natural
copy number state.
123
FIGURE 2- Purification of protein aggregates from Aneuploid yeast.
Protein aggregates were purified from a haploid yeast strain (WT) and strains disomic for
chromosome II and chromosome VIII carrying an Hspl04-GFP fusion protein. As a
positive control we used wild type cells grown at 25C and heat-shocked for 30 minutes at
42'C (A). Cells were grown in rich medium (YPD) at 25 0 C (B), 30 0 C (C) and 34 0 C (D) to
OD6
0
1 and 25 mL were collected for fractionation. Cells were incubated for 20 min at 25*C
in Lysis buffer (20mM NaPI pH6.8, 10mM DTT, 1mM EDTA, 0.1% Tween, 1x Roche
Protease Inhibitor, 1mM PMSF and 2.5mg/ml 20T Zymolyase) and then lysed by sonication
(two rounds of 8x, level 4, 50% on a Branson Sonifier). The lysate was spun for 20 min at
1600rpm and the protein concentration was equalized to 3mg/ml (Total Protein Sample, T).
The aggregates were then fractionated by spinning for 20 min 16000xg (Supernatant Sample,
S), washing twice with 2% NP40 in 20 mM NaPI, 1x protease inhibitor, 1mM PMSF,
sonicating 6x, level 4 50% duty cycle in between washes. . The pellet fraction was spun again
and washed with 20 mM NaPi, 1x protease inhibitors, 1 mM PMSF and sonicated 4x, level 2,
65% duty cycle and spun a final time at 16 0 0 0 xg for 20 min. The pellet fraction (P) was
resuspended in 8M urea. Total, Supernatant and Pellet fractions were run on an SDS-PAGE
gel and stained with Coomassie. We also performed an anti-GFP western blot to detect
Hspl04-GFP
124
Figure 2
A
0
YPD 25 C
30min 420C
B
Coomassie a-GFP
T S P T S P
YPD 250C
YPD 250C
Pellet
Sup.
Total
WT 11 VIll W1 11Vill WT 11 Vill
Total
Sup.
Pellet
WT 11 Vill WT 11 Vill WT II Vill
Iw
YPD 300C
YPD 300C
Pellet
Sup.
Total
WT 1I Vill WT 11 Vill WT 11 Vill
Sup.
Pellet
Total
WT iI Vill WT iI Vill WT iI Vill
C
LL.
Asia
D
YPD 34*C
YPD 340C
Pellet
Sup.
Total
WT 11 Vill WT 11 Vill WT 11 Vill
Pellet
Sup.
Total
WT 11 VIII WT it Vill WT II Vill
Apof
a.
LL
Irv Wop
40
125
Alternatively, proteins that are overabundant may be actively partitioned into
aggregates as means to reduce aberrant interactions that may cause toxicity and to decrease
burden on the quality control pathways of the cell. Although protein aggregates have long
been thought of as toxic species that hamper cellular function, reports in the literature
suggests that they may, at least in cases of protein folding diseases, be cytoprotectant.
Perhaps aneuploid cells recognize that there is an excess of a subset of proteins (because
their coding sequence is present in excess) and there is an active process that attempts to
sequester these protein subunits to limit the damage they can cause. This would require an
active sorting process where the cell would be able to recognize that there are protein units
that are not being utilized and then selectively target those for aggregation and presumably
later degradation. While there have been no post-translational modifications described that
specifically target proteins for aggregation, the addition of ubiquitin chains to proteins has
been shown to mark them for protein degradation by the UPS system. One can take
advantage of the di-Gly remnant after trypsinization of the isopeptide bond formed between
the Lysine on the target protein and the C-terminal Glycine of ubiquitin to immunopurify
and identify via mass spectrometry those proteins that were ubiquitinated when the sample
was collected (Kim et al. 2011). Using di-Gly mass spectrometry one could determine if
there is an overrepresentation of proteins encoded by the extra chromosome in the subset of
ubiquitinated proteins, which could suggest pre-emptive targeting for degradation previous
to protein misfolding.
In an alternative model where the composition of protein aggregates is irrespective
of the genomic imbalance present, the determinant for which proteins constitute the protein
aggregates found in aneuploid strains could be the difficulty that they have in acquiring a
stable, soluble native structure. Limited protein folding capacity would result in a higher
126
proportion of those more 'demanding' proteins to remain unfolded/misfolded and a higher
proportion of the more easily folded ones to remain soluble. As this scenario would not
require the cell to sort its proteins into those that are present in excess and those that are not,
I believe it to be more likely. Purification and identification of the protein aggregate
constituents will inform us whether either one, or both of these models is accurate.
Aneuploidy in mammalian cells alters protein quality control
As I have shown in Chapeter 2, aneuploidy dramatically alters protein quality control
in Saccharomyces cerevisiae- there is a pressing need in the field to determine if the same is true
in aneuploid mammalian cells. Similarly to aneuploid S. cerevisiae (Torres et al. 2010),
aneuploid human cell lines created by chromosome transfer also appear to fully transcribe
the tetrasomic chromosomes (average mRNA aneuploid to diploid log2 ratio of the extra
chromosome is 1.09) and to, at least partially translate it (average protein aneuploid to
diploid ratio of proteins encoded by the tetrasomic chromosome is 0.69) (Stingele et al.
2012). Interestingly, as in aneuploid yeast cells, there appears to be a subset of proteins
(between 20 and 25% in both human and yeast), that are retained at disomic levels; as their
mRNA is present at copy number levels, it appears that they're expression is controlled at
the protein level (Torres et al. 2010) (Stingele et al. 2012). The vast majority of proteins that
are not present in levels reflecting the increased copy number are members of protein
complexes. The mechanism via which this translational control occurs has not yet been
determined, but it is an active area of research in the field.
Although there have not been any reports of protein aggregation in mammalian
aneuploid cells, recent work from our lab and others has suggested that aneuploidy could
also be disturbing protein homeostasis in mammalian cells both in similar and different ways
127
than it is altered in yeast. Aneuploidy in all species analyzed shows a common transcriptional
response reminiscent of the ESR (Sheltzer et al. 2012), this response includes upregulation of
a subset of protein chaperones. In aneuploid yeast, this transcriptional up regulation of
chaperones does not result in a higher protein abundance of any of the chaperones tested as
a general response to aneuploidy (Chapter 2). Although this is also true for most chaperones
tested in trisomic mouse embryonic fibroblasts (MEFs) trisomic for chromosomes 13, 16 or
19, all trisomic MEFs harbor consistently higher abundance of the inducible isoform of
Hsp70: Hsp72 (Tang et al. 2011). We have also determined that trisomic MEFs appear to
show an altered heat shock response, with many genes upregulated to a higher degree than
wild type (Y.C Tang and S. Pfau, unpublished results). It remains to be seen if this is due to
differential kinetics of the heat shock response or if the response maintains its kinetics but
its intensity increased to aid in clearance excess misfolded proteins.
I, and others have shown that aneuploidy sensitizes yeast cells to proteasome
inhibitors and that increasing proteasme activity partially ameliorates the protein aggregation
phenotypes (Chapter 2, (Torres et al. 2007)). In contrast, trisomic MEFs have not been
found to be sensitive to the proteasome inhibitor Bortezamid, suggesting that mammalian
aneuploidy does not add stress to the proteasome system (YC Tang, unpublished
observations). This is unsurprising as rather than relying on the proteasome for protein
degradation, mammalian cells deploy autophagy as a major means to deal with aggregated
proteins (Tyedmers et al. 2010). In fact, mouse and human aneuploid cells were shown to
have increased levels of the autophagosome marker LC3-II (Stingele et al. 2013). Similar to
yeast, aneuploid cells are unable to maintain quality control- in mammalian cells this is
observed as an accumulation of autophagosomes
in the lysosome. (S. Santaguida,
unpublished results).
128
It is clear that both mammalian and yeast aneuploid cells are under proteotoxic stress,
but the pathways that are affected appear to differ. Further studies characterizing the role of
protein chaperones in mammalian aneuploidy would provide a more comprehensive picture
of the general effects of aneuploidy on eukaryotic cells. It would also be exciting to
hyperactivate the heatshock response in either aneuploid MEFs or human lines and assess
the effects of increasing quality control capacity on the proliferation of these cells.
Interface between aneuploidy, aging and neurodegeneration
Recent work has described a relationship between karyotypic imbalances and aging.
Mice that carry a hypomorphic allele of the spindle assembly checkpoint protein BUBRI and
thus missegregate chromosomes readily show signs of progeria (premature aging) (Baker et
al. 2004; Wijshake et al. 2012; Baker et al. 2013b). Conversely, mice that overexpress BUBR1
show reduced chromosome mis-segregation rates and extended lifespans (Baker et al. 2013a).
These data strongly suggest a relationship between the process of aging and of aneuploidy;
the nature of this relationship is still poorly understood. It is interesting to note that one of
the defining characteristics of an aging cell is the breakdown of protein homeostasis and the
accumulation of protein aggregates (Lopez-Otin et al. 2013), as this is the same phenotype
we observe in aneuploid cells it is tempting to posit whether the breakdown in protein
quality control is causal to aging. If so, premature aging seen in the hypomorphic BUBR1
mice could be the result of the proteotoxic stress and resulting breakdown in protein quality
control caused by aneuploidy. In depth studies of the quality control capacity of cells
carrying a hypomorphic allele of BUBR1 are needed to start assessing the validity of this
model. As hypomorphic BUBR1 mice are also the mouse model for Mosaic Aneuploid
Variegated (MVA) Syndrome, it would be of interest to assess the protein quality control
129
capacity in cultured cells derived from MVA patients. MVA is a pediatric syndrome
characterized by the early onset of tumors, but it can also be described as a progeria
syndrome as patients show symptoms such as growth retardation and cataracts that are
reminiscent of those seen in premature aging disorders. Although reduced lifespan has been
shown for mice carry hypomorphic
BUBR1 alleles, a causal relationship between
chromosome mis-segregation or aneuploidy and aging of cells has yet to be described. There
are no other reports of mouse models of chromosome mis-segregation leading to changes in
lifespan but most studies did not have as a goal to determine aneuploidy's effect on lifespan
and mice were sacrificed in their youth. More detailed studies with the goal of unveiling any
possible relationship between chromosome mis-segregation and aging need to be done to
elucidate the potential relationship between them.
Age has been shown to be the most predictive and predisposing factor for common
neurodegenerative ailments such as Alzheimer's, Parkinson's and Lou Gherig's disease.
These diseases all share similar etiologies: a misfolded protein that wreaks havoc on cellular
homeostasis causing neuronal death.
It is unclear why age is such a preponderant
predisposing factor, but it is believed that the breakdown in protein quality control and
accumulation of protein aggregates in aged cells may lead to a reduction in quality control
capacity and to the misfolding and formation of toxic disease proteins.
Concluding Remarks
Aneuploidy severely impacts cellular physiology, affecting almost every cellular
process. In this thesis I have shown that proteomic alterations and imbalances caused by
aneuploidy negatively impact the cell's ability to maintain protein homeostasis. I have
elucidated the impact of aneuploidy of the ubiquitous chaperone Hsp90 and determined that
130
aneuploidy sensitizes cells to situations that demand high folding capacity. Together, the data
in this thesis demonstrate that, in addition to gene specific effects, a state of chromosomal
imbalance has dramatic consequences for the cellular protein quality control pathways. Much
remains to be learned, especially in understanding the mechanisims by which aneuploidy
disturbs protein homeostasis and the cellular attempts to ameliorate this disturbance. It will
be important to glean molecular insight and to understand which quality control systems, if
any, are more severely affected and determine if there are ways to ameliorate these defects. It
is my hope that with further mechanistic understanding of this process, ongoing research
will contribute to our understanding, and potentially to the treatment of disease conditions
for which aneuploidy is a central part of their etiology, be they cancer or developmental
syndromes.
131
References
Baker DJ, Dawlaty MM, Wijshake T, Jeganathan KB, Malureanu L, van Ree JH, CrespoDiaz R, Reyes S, Seaburg L, Shapiro V et al. 2013a. Increased expression of BubR1
protects against aneuploidy and cancer and extends healthy lifespan. Nature cell biology
15: 96-102.
-.
2013b. Increased expression of BubR1 protects against aneuploidy and cancer and extends
healthy lifespan. Nat Cell Biol15: 96-102.
Baker DJ, Jeganathan KB, Cameron JD, Thompson M, Juneja S, Kopecka A, Kumar R,
Jenkins RB, de Groen PC, Roche P et al. 2004. BubR1 insufficiency causes early
onset of aging-associated phenotypes and infertility in mice. Nature genetics 36: 744-
749.
Borkovich KA, Farrelly FW, Finkelstein DB, Taulien
J,
Lindquist S. 1989. hsp82 is an
essential protein that is required in higher concentrations for growth of cells at
higher temperatures. Molecularand cellular biology 9: 3919-3930.
Boulon S, Marmier-Gourrier N, Pradet-Balade B, Wurth L, Verheggen C, Jady BE, Rothe B,
Pescia C, Robert MC, Kiss T et al. 2008. The Hsp90 chaperone controls the
biogenesis of L7Ae RNPs through conserved machinery. The Journalof cell biology 180:
579-595.
Gasch AP, Spellman PT, Kao CM, Carmel-Harel 0, Eisen MB, Storz G, Botstein D, Brown
PO. 2000. Genomic expression programs in the response of yeast cells to
environmental changes. Molecular biology of the cell 11: 4241-4257.
Geiler-Samerotte KA, Dion MF, Budnik BA, Wang SM, Hard DL, Drummond DA. 2011.
Misfolded proteins impose a dosage-dependent fitness cost and trigger a cytosolic
unfolded protein response in yeast. Proceedings of the NationalAcademy of Sciences of the
United States of-A merica 108: 680-685.
Hard FU, Bracher A, Hayer-Hartl M. 2011. Molecular chaperones in protein folding and
proteostasis. Nature 475: 324-332.
Imai J, Maruya M, Yashiroda H, Yahara I, Tanaka K. 2003. The molecular chaperone Hsp90
plays a role in the assembly and maintenance of the 26S proteasome. The EMBO
journal22: 3557-3567.
Kim W, Bennett EJ, Huttlin EL, Guo A, Li J, Possemato A, Sowa ME, Rad R, Rush J,
Comb MJ et al. 2011. Systematic and quantitative assessment of the ubiquitinmodified proteome. Molecular cell 44: 325-340.
Lopez-Otin C, Blasco MA, Partridge L, Serrano M, Kroemer G. 2013. The hallmarks of
aging. Cell 153: 1194-1217.
Neckers L. 2007. Heat shock protein 90: the cancer chaperone. Journalof biosciences 32: 517-
530.
Oromendia AB, Dodgson SE, Amon A. 2012. Aneuploidy causes proteotoxic stress in yeast.
Genes Dev 26: 2696-2708.
Richter K, Haslbeck M, Buchner J. 2010. The heat shock response: life on the verge of death.
Molecular cell 40: 253-266.
Sheltzer JM, Torres EM, Dunham MJ, Amon A. 2012. Transcriptional consequences of
aneuploidy. Proceedings of the NationalAcademy of Sciences of the United States of A merica
109: 12644-12649.
Sorger PK. 1990. Yeast heat shock factor contains separable transient and sustained
response transcriptional activators. Cell 62: 793-805.
132
Stingele S, Stoehr G, Peplowska K, Cox J, Mann M, Storchova Z. 2012. Global analysis of
genome, transcriptome and proteome reveals the response to aneuploidy in human
cells. Molecular systems biolog 8: 608.
Stingele S, Stoehr G, Storchova Z. 2013. Activation of autophagy in cells with abnormal
karyotype. A utophag 9: 246-248.
Tang YC, Williams BR, Siegel JJ, Amon A. 2011. Identification of aneuploidy-selective
antiproliferation compounds. Cell144: 499-512.
Torres EM, Dephoure N, Panneerselvam A, Tucker CM, Whittaker CA, Gygi SP, Dunham
MJ, Amon A. 2010. Identification of aneuploidy-tolerating mutations. Cell143: 71-83.
Torres EM, Sokolsky T, Tucker CM, Chan LY, Boselli M, Dunham MJ, Amon A. 2007.
Effects of aneuploidy on cellular physiology and cell division in haploid yeast. Science
317: 916-924.
Tyedmers J, Mogk A, Bukau B. 2010. Cellular strategies for controlling protein aggregation.
Nature reviews Molecularcell biology 11: 777-788.
Wijshake T, Malureanu LA, Baker DJ, Jeganathan KB, van de Sluis B, van Deursen JM. 2012.
Reduced life- and healthspan in mice carrying a mono-allelic BubR1 MVA mutation.
PLoS genetics 8: e1003138.
Yeger-Lotem E, Riva L, Su LJ, Gitler AD, Cashikar AG, King OD, Auluck PK, Geddie ML,
Valastyan JS, Karger DR et al. 2009. Bridging high-throughput genetic and
transcriptional data reveals cellular responses to alpha-synuclein toxicity. Nature
genetics 41: 316-323.
133