journal orf I~nderr.3d.3ua.e R-ese-arch
,,Olumie :*, isSue . - l.a3rch I.'ULs
An Episomal Model for Aging in Saccharomyces cerevisae
Saccharomyces cerevisiae, commonly known as a baker's yeast, has become a key model organism for studying
the molecular basis of cellular aging. Aging mechanisms in yeast have been studied for years, and one
hypothesis suggests that extrachromosomal rDNA circles (ERCs) accumulate in the mother cell ultimately causing
its senescence. To test this hypothesis, autonomously replicating (ARS) plasmids that "mimic" ERCs
were constructed. Plasmid transmittance was monitored by measuring mitotic stability. Our results show that
ARS-plasmids and ERCs "compete" for retention in the mother cell. Furthermore, we also showed that ARS-
plasmids compete with each other when in the presence of one another. Since ERCs retained in the mother cell
are rarely passed on to daughter cells due to an asymmetric inheritance process known as mother inheritance
bias, we contend that ERCs and ARS-plasmids accumulate due to this mother cell bias inheritance mechanism
during the aging process in yeast.
In 1996, Saccharomyces cerevisiae, commonly known as baker's yeast, became the first eukaryote to have
its genome fully sequenced. Like yeast, humans are also eukaryotes, and thus many similarities exist in
their genomes. It is estimated that one third of the yeast genes are related to human genes. 1
All eukaryotes exhibit similarities in their cellular anatomy, and even similarities in cell function. Yeast DNA
contains stretches of subunits called bases which encode for proteins that have also been found in human DNA.
These protein sequences must play an important role in cell function because they have been conserved by
evolution for the billion years that separates humans from yeast. Importantly, certain yeast and human genes
seem to be similar in aging functions. Human diseases, such as the Werner Syndrome, which is caused by a
mutation to the human WRN protein, have yeast analogs. The Werner Syndrome is a genetic disease that leads
to premature aging. Yeast cells with mutations in the SGS1 protein show accelerated aging like those with the
Werner Syndrome. Studying the yeast genome and its aging mechanisms will give insight to similar
aging mechanisms in humans.2
S. cerevisiae is an excellent model with which to study aging because it is a budding yeast as shown in Fig.
1. Budding is the process by which the mother cell divides and gives rise to a bud which becomes the daughter
cell. This division is an asymmetric process which means that the mother cell and daughter cell are not identical
and can be tracked individually. The age of the mother cell is determined by the number of daughter cells
it produces. A mother cell has a limited capacity to produce daughter cells, and the decline in this capacity with
each generation is known as replicative aging.3
Figure 1. Budding Yeast Cell Division. Mother cell(M) gives rise to daughter (D) cells in an
Aging mechanisms in yeast have been studied for years, and one hypothesis suggests that extrachromosomal
rDNA circles (ERCs) accumulate in the mother cell causing its senescence[mssl] . ERCs are the product
of homologous recombination at the ribosomal DNA (rDNA) locus, which is transcribed to form ribosomal RNA
(rRNA). Since ERCs contain origins of replication, each ERC has the ability to replicate each cell cycle. ERCs
are retained by mother cells and rarely passed on to the daughter cells due to an asymmetric inheritance
process known as mother cell bias. As the ERCs continue to accumulate, it has been hypothesized that cell
functions are hindered, including replication mechanisms, eventually leading to cell senescence.45
\--- s0; D-- * (
young cell exclusion/ rep
of ERC as)
lcatlon old ce
Figure 2. ERC Model for Regulating Yeast Life Span. ERCs are the product of homologous recombination
at the ribosomal DNA (rDNA) locus, which is transcribed to form ribosomal RNA (rRNA). They
are retained in the mother cell during mitosis and their accumulation is linked to replicative
aging. Obtained from Ref. 4.
However, not all hypotheses agree that the accumulation of ERCs is the cause of aging. Others believe that
ERC accumulation is more of an effect of aging[mssl] . Shortened lifespan in yeast may also be due to
mutations impairing DNA replication, recombination, and repair. By this view, ERC accumulation has little to do
with cell senescence. Previous work by Falcon and Aris tested the ERC hypothesis using a "plasmid-based
model." Their work demonstrated that plasmids accumulate in mother cells and reduce lifespan[mss2]. They
showed that plasmid copy number increases with mother cell age and that inheritance of plasmids is affected by
ERC levels in the cell. Furthermore, their work also suggests that ERCs are not merely an effect of aging. To
continue probing this controversy, we have designed an experiment similar to that of Falcon and Aris with
different plasmids. The two types of plasmids used are ARS plasmids and ARS/CEN plasmids, which are shown
Figure 3. ARS1 and ARS1/CEN4 Plasmids.
Autonomously replicating sequence (ARS) plasmids, which act most like ERCs, are circular DNA molecules
with replication origins. In previous experiments they have exhibited a strong mother cell bias, making them an
ideal plasmid to mimic ERCs. It is predicted that the ARS plasmids will accumulate in the mother cell and reduce
Autonomously replicating/centromeric sequence (ARS/CEN) plasmids have a centromeric DNA region that
attaches the plasmid to the mitotic spindle during replication. This mechanism ensures that the plasmid is
effectively passed on to the daughter cell. Like the ARS plasmids, ARS/CEN plasmids also contain a replication
origin. However, since ARS/CEN plasmids do not exhibit mother cell bias, they should be distributed throughout
the population and not accumulate in any given cell. Thus, it is predicted that ARS/CEN plasmids will not
accumulate or reduce lifespan in the mother cell.
Each plasmid will also contain a nutritional marker unique to its type so that cells can be grown under the
desired selection. TRPI, URA3, and LEU2 will be used as the markers. If cells are grown in the absence of
tryptophan, only the cells that inherited a plasmid with the TRPI marker will grow. The same holds true for uracil
and the URA3 marker, and leucien and the LEU2 marker.
The plasmids were transformed into strains of varying levels of ERCs to monitor the interaction between plasmids
and ERCs. To quantify this, the inheritance of the plasmids has to be measured by determining the mitotic
stability. Our results show that ARS plasmids are not inherited well by the daughter cells and thus accumulate in
the mother cell. In contrast, the ARS/CEN plasmids are inherited by the daughter cells and little accumulation in
the mother cell occurs.
These results[mssl] support the findings of Falcon and Aris which state that ARS plasmids and ERCs have
similar inheritance mechanisms. Further research shows that when two ARS plasmids are introduced to the
same strain, the mitotic stability of each plasmid is affected. This suggests that ARS plasmids compete
among themselves much like ERCs compete with the ARS plasmids. It also supports the idea that mother cells have
a limited capacity for episomes and thus once that critical threshold is reached, the mother cell begins to
pass plasmids and ERCs onto the daughter cells[mss2][mss3].
Yeast Strains and Plasmids
Plasmids were transformed into three different yeast strains. The wildtype strain was used to monitor ERC
and plasmid interactions in the typical yeast strain. Two mutant strains, SIR2A and foblA, were used to
determine the effect different levels of ERCs would have on the plasmids. The SIR2A strain has a deleted SIR2
gene. Expression of the Sir (silent information regulator) protein complex has been correlated with yeast life
span. Yeast strains with a deleted SIR2 show a reduction in life span, while overexpression of SIR2 shows an
increase in life span. We chose to work with SIR2A because the deletion of the SIR2 gene leads to an increase in
ERC levels and a reduction in life span as compared to the wildtype. The foblA, on the other hand, is known to
have a reduced level of ERCs and therefore an increased life span as compared to the wildtype.
The wildtype strain, W303AR5 (MAT_ leu2-3,112 his3-11,15 ura3-1 ade2-1 trpl-1 canl-100 RAD5 ADE2::rDNA,
[cir+]), was obtained from D.A. Sinclair. The SIR2A strain used has a genotype of RMY206-5B SIR2 MAT
SIR2::HIS3 hmr::TRPlleu2-3, 112 his3-11, 15 ade2-1 ura3-1 trpl-1 canl-100 [RAD5+]. The foblA strain's
genotype is MAT_ fobl::URA3 leu2-3, 112 his3-11,15 ade2-1 ura3-1 trpl-1 canl-100 [RAD5+]. Both strains
have the same parent strain HKY580-10D.
Six plasmids were used throughout these experiments. Descriptions of these plasmids can be found in Table
1. Plasmids pJPA113, pJPA133, and pAF31 are ARS plasmids. The ARS/CEN plasmids were pJPA136, pJPA116,
and pAF32. The plasmids were transformed in the yeast strains using a standard lithium acetate method
Table 1. Descriptions of Plasmids Used in Experiments
Plasmid Origin, Insert Marker Backbone Size (kb)
pJPA113 ARS1 URA3 pRS306 4,575
pJPA116 ARS1, CEN4 URA3 pRS306 5,316
pJPA133 ARS1 LUE2 pRS306 5,698
pJPA136 ARS1, CEN4 LEU2 pRS306 6,468
pAF31 ARS1 URA3, ADE2 pJPA113 6,828
pAF32 ARS1, CEN4 URA3, ADE2 pJPA116 7,568
In the lithium acetate method for transforming yeast, the yeast are first grown in YPD medium to an optical
density (OD) between 0.2 and 0.4 as measured by a UV/Vis spectrometer at 600 nm. (The YPD medium consists
of Bacto Yeast Extract, Bacto Peptone, and Dextrose.) About 2 OD units are needed per plasmid
transformation where an OD600 of 0.4 in 25 mLs is equivalent to about 10 OD units. Once enough cells have
been gathered, they are washed with deionized water and kept on ice. The pellets are then washed with 1 mL of
TE/LiOAc. For 2 mL of TE/LiOAc, 200 pL 10X TE, 200 pL 1M LiOAc, and 1.6 sterile ddH20 were used where 10X TE
is made of 100 mM Tris-HCI and 10 mM EDTA at pH of 7.5.
After washing the pellet with TE/LiOAc, the cells are resuspended in 100 pL of TE/LiOAc per 1-2 OD units. In
a microfuge tube, 5 pL of 10 mg/mL carrier DNA (HMW, ssDNA, sonicated), 0.5-5 pL transforming DNA, and 100
pL of yeast cells are mixed for each plasmid transformation. A fresh solution of PEG/LiOAc is prepared using 200
pL 10X TE, 200 pL 1M LiOAc, and 1.6 mL of 50% PEG 3400 for 2 mL of solution. To each microfuge tube, 0.6 mL
of the PEG/LiOAc solution is added and the tubes are rotated end-over-end for 30 minutes. Afterwards, the tubes
are placed in a water bath at 426C for 15 minutes. Finally, the cells are resuspended in 200 _L of 1X TE of which
100 _L is plated on the appropriate medium.
The mitotic stability of each plasmid was determined in order to know about its inheritance. The mitotic stability
is defined as the percentage of colony forming units that contain the plasmid in a given population. Each plasmid
was transformed into W303AR5, SIR2A, and foblA. Five transformants were then grown for each plasmid in
each strain for 2 days at 300C in selective synthetic dextrose (SD) liquid medium. The SD medium is made
of Dextrose, Yeast Nitrogen Base, and Ammonium Sulfate. The OD of each cell culture was then measured
to determine the volume of cells needed to plate approximately 200-250 colony forming units on nonselective
SD medium. This was accomplished using the fact that an OD between 1.1 and 1.5 corresponds to approximately
0.5 - 1 X 107 colony forming units. After 200-250 colony forming units were plated, they were incubated for 2-
3 days. Once the cells had grown to a countable size (1-2 mm diameter), they were replica plated onto selective
and nonselective plates. Replica plating is a procedure by which cells from colonies on the original ("master")
plate are transferred to new plates. This is done by placing felt sheets on a circular metal block with a slightly
smaller diameter than that of the plates. The plates are then placed on the felt so that each colony forming
unit transfers cells to the felt. The new plates are then placed on the felt so that cells will be transferred to them as
a replica of the original plate. These plates are incubated at 300C as well and left to grow for 3-4 days. For
mitotic stability counts, the plates are replica plated onto selective and nonselective plates. The selective
plates contain all the amino acids necessary for the cells to grow except for the amino acid that the plasmid
contains. For instance, pJPA133 has a LEU2 selective marker. Therefore, to grow it in selection means to grow it
on medium without LEU2 so that only the cells that contain the plasmid can survive.
After the cells are at a countable size, the numbers of cells on the selective and nonselective plates are counted.
The ratio of cells on selective to nonselective plates is the mitotic stability. This ratio serves as a measure of how
well a plasmid is inherited by a population. A high mitotic stability indicates that the mother cell effectively passes
on the plasmid to the daughter cells. A low mitotic stability suggests that the plasmid has asymmetric inheritance
and that the mother cell is prone to retaining the plasmid.
RESULTS AND DISCUSSION
The results obtained support the results obtained previously by Falcon and Aris [mssl]. New conclusions as to
how plasmids interact with one and other are also presented in these results. They show that plasmids affect
each other's inheritance and suggest that their inheritance mechanisms are similar to ERCs.
80 -- - -
40- - - -
30 - -
20 - - -
10 - -
pJPAI33 pJPA133 pJPAI33 pJPA136 pJPA136 pJPA136
Figure 4. Mitotic Stability Results for pJPA133 and pJPA136.
Table 2. Mitotic Stability Results for pJPA133 and
Plasmid and Strain Mitotic Stability
W303 pJPA133 15.55
foblA pJPA133 20.98
sir2A pJPA133 24.50
The first plasmids tested were pJPA133 and pJPA113. They were transformed with the wildtype strain, W303, and
the mutant strains, fobl_ and sir2_. The results obtained from the mitotic stability experiment are shown below
in Table 2 and in Figure 4. The ARS plasmid pJPA133 acted as predicted in all three strains. The mitotic stability
of the plasmid was highest in the sir2_ mutant. This is expected because the sir2_ strain has a high level of
ERCs. ARS plasmids are known to act like ERCs and thus the rapid accumulation of plasmids and ERCs leads
to impairment of the mother cell bias. This result suggests that the mother cell has a limited capacity for
episomes. As such, it is reasonable to conclude that the competition of ARS plasmids and ERCs leads to
the transmission of the episomes to the daughter cells. Conversely, fobl_ has a reduced level of ERCs as
compared to sir2_, and thus the plasmid does not 'compete' with ERCs to the same extent as in sir2_. This
difference in "level of competition" is reflected in the results since the mitotic stability of pJPA133 in fobl_ is
lower than in sir2_. The mother cell has a higher capacity to retain the plasmid in the fobl_ strain since there
are less ERCs for the plasmid to compete with. The ARS/CEN plasmid acted as predicted and did not accumulate
in any of the strains.
pAF31 pAF31 pLAF32 pAP~2 pAPn
Figure 5. Mitotic Stability Results for pAF31 and pAF32.
Table 3. Mitotic Stability Results for pAF31 and
Strain and Plasmid Mitotic Stability
W303 pAF31 15.66
fobl pAF31 16.2
sir21 pAF31 29.15
W303 pAF32 93.91
fob11 pAF32 95.17
sir2l pAF32 96.5
The mitotic stability results for pAF31 and pAF32 showed similar trends to those described above. Again, the
ARS plasmid had the highest mitotic stability in the sir2_ strain which is known to have increased levels of ERCs.
The mitotic stabilities of pAF31 in the wildtype and fobl_ strains were significantly lower than that in the sir2_.
This supports the conclusions made earlier with pJPA133. The mitotic stability of pAF32 was in the 90th percentile
for all three strains as predicted. CEN plasmids are known not to accumulate in any given cell and thus are
inherited efficiently throughout the population.
113 133 113+133*113+133* 136+116* 136 116
133 113 136 116 113 133
Figure 6. Mitotic Stability Results for Double Transformants.
Table 4. Mitotic Stability Results for Single and
Double Transformants with W303
(MS of plasmid in bold) Mitotic Stability (MS)
113 + 133
133 + 113
113 + 136
133 + 116
136 + 113
Finally, we tested the mitotic stability of plasmids in a double transformant environment. Plasmids pJPA113
and pJPA133 are both ARS plasmids which are predicted to compete with ERCs for retention in the mother
cell. Plasmids pJPA116 and pJPA136 are CEN plasmids which do not accumulate in the mother cell. We
transformed W303 with all four plasmids alone, and then double transformed W303 with combinations of
the plasmids. Our results showed that ARS plasmids in the presence of another ARS plasmid have an
increased mitotic stability. The mitotic stability of pJPA113 increased from 17% when alone to 25% when in
the presence of another ARS plasmid, pJPA133. Similar results were obtained for pJPA133 in the presence
of pJPA113. Normally, the mitotic stability of pJPA133 is 14%, but when transformed with pJPA113, its
mitotic stability increased to 21%. The mitotic stability of the ARS plasmids was unaffected when combined with
CEN plasmids [mssl].
The results show that there is a clear relationship between the accumulation of episomes and the mitotic stability
of ARS plasmids. The mitotic stability results of the single transformants with pJPA133 and pAF31 both show that
in strains with high ERC levels, the plasmid has a greater chance of getting passed on to the daughter cell. This
gives us insight to the aging mechanisms relevant to episomes. It suggests that mother cells have a limited
capacity for episomes, and that as this limit is reached, the mother passes on episomes to the daughter
cells. Competition among ARS plasmids and ERCs might also be an explanation for the increased mitotic stability
of ARS plasmids in high ERC level environments. If the mother cell does have a limited capacity for episomes,
then both plasmids and ERCs cannot be retained by the mother cell as it ages. This leads to a 'competition'
among the episomes to stay in the mother cell.
Even more interesting is the effect that two ARS plasmids have on the mitotic stability of the individual
plasmids. Both plasmids compete for retention in the mother cell but neither one seems to be dominant. The
mitotic stability of both plasmids increases by similar amounts when in the presence of each other. When the
ARS plasmids are in the presence of a CEN plasmid, neither mitotic stability is affected. This suggests that
the competition among the ARS plasmids is similar to that among ERCs and ARS plasmids. Thus, it is reasonable
to conclude that the inheritance mechanisms of ERCs are the same as those of ARS plasmids.
These conclusions are important to the aging field because they give insight to mechanisms involved with
cellular senescence. Understanding these mechanisms in yeast will make studies in higher organisms possible
since many analogous genes have been found among eukaryotes. Although ERCs have not been characterized
in humans, it is believed that all DNA has the ability to form ERCs. Thus, understanding the role that ERCs play
in aging may lead to understanding how DNA is affected by aging and how this leads to senescence in yeasts
and higher organisms.
I would like to thank the following people for their invaluable help and support: Dr. John Aris, Alaric Falcon, Lauren
N. Elliot, Gloria Rios, and Joshua Herr.
1. "International Team Completes DNA Sequence of Yeast". National Human Genome Research Institute Press
Releases. April 24, 1996. Back
2. Sinclair, David A., Mills K, and Guarente Leonard. "Accelerated aging and nucleolar fragmentation in yeast
sgsl mutants". Science, vol. 277 (1997): 1259-60. Back
3. Sherman, Fred. "Getting Started with Yeast". Methods of Enzymolgy. ol. 350. New York: Academic Press, 2002. Back
4. Falcon, Alaric A. and Aris, John P. "Plasmid Accumulation Reduces Life Span in Saccharomyces cerevisiae".
The Journal of Biological Chemistry, vol. 278 (2003): 41607-41617. Back
5. Sinclair, David A. and Guarente, Leonard. "Extrachromosomal rDNA Circles -A Cause of Aging in Yeast". Cell, vol.
91 (1997): 1033-1042. Back
Back to the Journal of Undergraduate Research
College of Liberal Arts and Sciences I University Scholars Program I University of Florida I
The Foisrrdanom for F iir Gwor N G.I
ï¿½ University of Florida, Gainesville, FL 32611; (352) 846-2032.