TOXICOLOGICAL SCIENCES 113(1), 77–84 (2010)
Advance Access publication October 5, 2009
Rapamycin Inhibits Yeast Nucleotide Excision Repair
Independently of Tor Kinases
Melvin V. Limson,* and Kevin S. Sweder†,1
*Education Office, American Physiological Society, Bethesda, Maryland 20814-3991; and †Research and Development,
Bristol-Myers Squibb Company, East Syracuse, New York 13057-5050
1To whom correspondence should be addressed at Research and Development, Bristol-Myers Squibb Company, 6000 Thompson Road, East Syracuse,
NY 13057-5050. Fax: (315) 432-2172. E-mail: firstname.lastname@example.org.
Received April 16, 2009; accepted September 2, 2009
The yeast target of rapamycin (Tor) kinases, Tor1 and Tor2,
belong to the phosphatidylinositol 3-kinase–related family of
proteins, which are involved in the cellular response to DNA
damage and changes in nutrient conditions. In contrast to yeast,
many eukaryotes possess a single Tor kinase. Regardless of the
number of Tor kinases in an organism, two distinct complexes
involving Tor proteins exist in eukaryotes, TORC1 and TORC2.
The yeast TORC1, containing Tor1 or Tor2, is sensitive to the
antibiotic rapamycin. The yeast TORC2 is insensitive to
rapamycin. We examined the influence of rapamycin treatment
upon yeast transcription–coupled nucleotide excision repair in
a gene transcribed by RNA polymerase II. We also examined tor
mutants for their ability to perform transcription-coupled repair
in the absence or presence of rapamycin. Ostensibly lacking
TORC1 and TORC2 function, a tor1tor2tsmutant grown at the
nonpermissive temperature exhibited similar rates of repair as the
wild-type strain. However, repair of both strands in genes
decreases in the wild-type strain and the tor1tor2tsmutant
exposed to rapamycin. Rapamycin may be inhibiting DNA repair
independently of the Tor kinases. In yeast, FPR1 encodes the
rapamycin-binding protein Fpr1 that inhibits the TORC1 kinase
in the presence of rapamycin. Fap1 competes with rapamycin for
Fpr1 binding. Deletion of the FPR1 or FAP1 gene abolishes the
inhibitory effect of rapamycin on repair. Thus, the decreased
repair observed following rapamycin treatment is independent of
TORC1/2 function and likely due to a function of Fap1. We
suggest that Fap1 and peptidyl-prolyl isomerases, particularly
Fpr1, function in the cellular response to genotoxic stress. Our
findings have clinical implications for genetic toxicities associated
with genotoxic agents when coadministered with rapamycin.
Key Words: Fap1; Fpr1; NER; Tor; transcription-coupled repair.
The conserved phosphatidylinositol 3-kinase (PI3K)–related
family of proteins is involved in cellular responses to
extracellular conditions and DNA damage. Members of the
family include ATM (a gene mutated in the disease ataxia
telangiectasia), DNA-PKcs (the catalytic subunit of the DNA-
dependent protein kinase), Mec1, Tel1, and Rad3 (Hoekstra,
1997). Another group of PI3K-related proteins, the mammalian
and yeast target of rapamycin (Tor) kinases, are involved in cell
growth control and nutrient sensing (Schmelzle and Hall,
2000). The Tor proteins, encoded by TOR1 and TOR2, were
first discovered in yeast as the targets of rapamycin, an immu-
nosuppressant agent that inhibits T-cell proliferation in mam-
malian cells via mechanisms elucidated in yeast (Heitman
et al., 1991). The Tor proteins exist in two distinct protein
complexes, TORC1 and TORC2 (De Virgilio and Loewith,
2006). TORC1 can contain Tor1 or Tor2 and is sensitive to
rapamycin. TORC2 contains only Tor2 and is insensitive to
rapamycin. When rapamycin is bound to its target in yeast, the
peptidyl-prolyl isomerase Fpr1, the Fpr1-rapamycin complex
inhibits the kinase activity of the TORC1 complex. Alteration
of the rapamycin-binding site of Fpr1, or the Fpr1-rapamycin–
binding site in Tor proteins, confers resistance to rapamycin
(Lorenz and Heitman, 1995; Zaragoza et al., 1998).
In a genetic screen for genes encoding dosage suppressors
of rapamycin toxicity, FAP1 was identified and characterized as
encoding an Fpr1-associated protein (Kunz et al., 2000). Fap1
is a homolog of the human transcriptional repressor NF-X1,
which is involved in the regulation of major histocompatibility
genes during the inflammatory response (Kunz et al., 2000; Song
et al., 1994). The interaction of Fap1 with Fpr1 is competitively
disrupted by the addition of rapamycin (Alarcon and Heitman,
1997; Dolinski and Heitman, 1999; Kunz et al., 2000). Thus, the
effects of rapamycin can also be attributed to the release of Fap1
or other factors from Fpr1 that subsequently affect transcription.
Tor1 and Tor2, as part of TORC1, function in the cellular
response to nutrient conditions or stress via translational and
transcriptional control and can be blocked by rapamycin
(Rohde et al., 2001; Schmelzle and Hall, 2000). The TORC1
pathway coordinates aspects of ribosome biogenesis, such as
ribosomal RNA (rRNA) synthesis by RNA polymerase (RNA
pol) I, processing of the 35S rRNA precursor, and regulation of
ribosomal protein gene expression by RNA pol II (Kuruvilla
et al., 2001; Powers and Walter, 1999; Raught et al., 2001).
The mammalian Tor (mTor) kinases have been implicated as
? The Author 2009. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved.
For permissions, please email: email@example.com
signal transduction components that respond to DNA damage to
induce cytokine gene expression in normal human epidermal
keratinocytes, and Tor kinases are regulated in a p53-dependent
manner (Budanov and Karin, 2008; Yarosh et al., 2000).
A highly conserved mechanism of DNA repair in eukaryotes
is nucleotide excision repair (NER), which removes bulky
DNA adducts such as cyclobutane pyrimidine dimers (CPDs)
induced by ultraviolet (UV) radiation (Friedberg et al., 1995).
In genes actively transcribed by RNA pol II, CPDs are
removed from the transcribed strand faster than the non-
transcribed strand in a process known as transcription-coupled
DNA repair (TCR), a subpathway of NER. The level of repair
in the nontranscribed strand is similar to repair of the entire
genome, a process known as global genomic repair. We
examined if rapamycin treatment impacts the Tor signaling
pathway, or other components associated with Tor (such as
Fpr1 or Fap1), to affect the cellular response to DNA damage
as shown for other members of the PI3K family.
MATERIALS AND METHODS
Strains, media, and growth. All strains were kindly provided by M. N.
Hall (University of Basel, Basel, Switzerland): JK9-3da (wild-type parental
strain: MATa leu2-3, 112 ura3-52 rme1 trp1 his4 GAL þ HMLa), MH349-3d
(same as parental, tor1::LEU2-4), SH121 (same as parental, tor2::ADE2-3/
YCplac111::tor2-21ts), SH221 (same as parental, tor1::HIS3-3 tor2::ADE2-3/
YCplac111::tor2-21ts), JHY1-2c (same as parental, fpr1::URA3-1), and
MH347-2D (same as parental, fap1::LEU2-2). Synthetic dextrose (SD) is
minimal media (0.17% yeast nitrogen base, 0.5% ammonium sulfate, and 2%
glucose). Growth of strains was on SD supplemented with the appropriate
amino acids. Rapamycin (Sigma, St. Louis, MO) was made 1 mg/ml in the
vehicle (90% ethanol/10% Tween-20) and added to cultures at a final
concentration of 0.2 lg/ml. Strains were grown overnight in minimal media
supplemented with appropriate amino acids. Each strain was diluted 10-fold in
fresh media and grown at 30?C, the permissive temperature for the conditional
mutants, for 1 h prior to taking optical density (OD600nm) readings. The cultures
either remained at 30?C or were shifted to 37?C, the nonpermissive temperature
for the conditional mutants. The OD600nmof each culture in different conditions
was taken by spectrophotometer analysis over a period of 12 h. For treatment
with 0.2 lg/ml of rapamycin, all strains were grown for 5 h to early log phase at
30?C prior to the addition of rapamycin or vehicle.
UV irradiation and strand-specific repair assay. The detailed protocol for
the strand-specific repair assay in yeast was previously described (Sweder and
Hanawalt, 1992). Mid-logarithmic cultures were grown in SD supplemented
with the appropriate amino acids and then collected by centrifugation. Cells
were resuspended (~1.0 3 107cells/ml) in PBS. Temperature-sensitive tor2ts
and tor1tor2tsmutants were grown overnight at the permissive temperature,
diluted 1:10 in fresh media, grown at the permissive temperature (30?C) for 1 h,
and shifted to the nonpermissive temperature (37?C) for 5 h before UV
irradiation. Additionally, wild-type, fpr1, fap1, and tor1tor2tsstrains were
treated with 0.2 lg/ml rapamycin 1.75 h prior to UV irradiation and returned to
their original medium throughout the remainder of the experiment.
Cells suspended in PBS were irradiated with predominantly 254 nm UV
light at 60 J/m2using a germicidal lamp (American Ultraviolet Co., Lebanon,
IN). To measure the initial incidence of DNA damage, an aliquot was taken
immediately after the irradiation (0 min of repair) and prepared for cell lysis,
while the remainder of the irradiated cells were collected by centrifugation and
resuspended in the original media at the appropriate temperature to allow time
for repair. Aliquots were taken at 15, 30, 60, and 90 min and prepared for cell
lysis. Spheroplasts were made by adding Zymolyase 100T (ICN, Costa Mesa,
CA) to digest the cell walls. The DNA was extracted, purified, and digested with
a restriction endonuclease (HincII) resulting in fragments of 5.0 kb for RPB2.
After restriction digestion, DNA from each time point was ethanol precip-
itated, resuspended, and divided in half. One half was treated with the CPD-
specific T4 endonuclease V (TEV), which makes a single-stranded break at the
site of each CPD. Such strand breaks will reduce the amount of the full-length
restriction fragments when electrophoresed under alkaline conditions. The other
half was mock treated and used to quantify the amount of DNA extracted from
each time point. The samples were denatured, electrophoresed through an alkaline
agarose gel, and transferred onto nylon membranes (Hybond Nþ, GE Healthcare,
Using RNA probes specific for either the transcribed or the nontranscribed
strand, it is possible to determine the rates and extent of repair for the individual
strands of the gene of interest. The membranes were sequentially hybridized
and exposed to x-ray film with one32P-radiolabeled RNA probe (as described
below) at a time. Assuming a Poisson distribution of initial DNA damage, the
number of CPDs per fragment can be calculated from the ratio of the hybrid-
ization signal intensities for the TEV-treated samples versus mock-treated con-
trols as quantified with the application National Institutes of Health Image 1.62.
Plasmid construction and probes. For a gene transcribed by RNA pol II,
an ~1.0-kb fragment of RPB2 was previously cloned into the Bluescript pKSþ
vector (Stratagene, La Jolla, CA), which has a T7 and T3 promoter flanking the
insert as well as an EcoRI and XhoI restriction site at the opposite ends of the
insert (Sweder and Hanawalt, 1992). Plasmid pKS212 was linearized by
cleaving with XhoI or EcoRI and incubated with (a-32P) cytidine triphosphate
(GE Healthcare, Piscataway, NJ), ribonucleotide triphosphates, and T7 RNA
pol or T3 RNA pol, respectively, under conditions recommended by the man-
ufacturer to generate strand-specific radioactive RNA probes (transcribed or
nontranscribed strand, respectively) for RPB2 (Sweder and Hanawalt, 1992).
Statistical analysis of repair. For each strain and experimental treatment
condition, the initial rate of repair was estimated using a linear function to describe
the relationship between %CPD values and time (0–30 min). A no-intercept model
was utilized since values at time 0 were used to calculate subsequent %CPD
values. The slope estimates were then compared to evaluate differences in
initial rate of repair between rapamycin treatment and no treatment or between
each mutant strain and wild type for each treatment condition. The treatment
conditions are identified as rapamycin-transcribed and -nontranscribed strands
and vehicle-transcribed and -nontranscribed strands in Figures 1B and 3 and
30? transcribed and 30? nontranscribed strands and 37? transcribed and 37?
nontranscribed strands in Figure 2. In addition, for each strain, the slope esti-
mates were compared to evaluate differences between treatment conditions:
difference between temperature for each strand, differences between rapamycin
treatment (þ RM vs. ?RM) for each strand, and differences between strands
(nontranscribed vs. transcribed) for each temperature or þ RM and ?RM
treatment. Comparisons were made at the 5% significance level. In addition,
comparisons of mean %CPD were made at each time point (15–90 min). In the
context of a one-way ANOVA model, separately by time and treatment con-
dition, comparisons between the mean %CPD for mutant strains versus wild-
type strain were made using t-tests. In a two-way ANOVA, separately by strain,
comparisons within treatment conditions were made (transcribed vs. non-
transcribed strands for each temperature [30?C and 37?C] or rapamycin
treatment [þ RM and ?RM] and temperature differences or rapamycin treat-
ment differences for both transcribed and nontranscribed strands) using t-tests.
Each t-test was evaluated at the 5% significance level.
Strand-Specific Repair of UV-Induced DNA Damage in the
Wild Type Treated with Rapamycin
The Tor kinases have been implicated in the cellular
response to DNA damage (Budanov and Karin, 2008; Yarosh
LIMSON AND SWEDER
et al., 2000). We measured repair in cells where Tor activity was
repressed biochemically with rapamycin, a well-characterized
inhibitor of the TORC1 kinase (Cardenas et al., 1999). Cultures
of the wild-type strain were grown in the absence or presence of
0.2 lg/ml rapamycin for 1.75 h prior to and after UV irradiation
as described in ‘‘Materials and Methods’’ section. Representa-
tive autoradiograms of strand-specific repair assays for the RPB2
gene in the wild type grown at 30?C in the absence or presence
of rapamycin are shown in Figure 1A. The ratio of the
hybridization intensities in the TEV-treated DNA versus the
mock-treated DNA samples increased faster for the transcribed
strand than for the nontranscribed strand of RPB2 in the wild-
type strain (Fig. 1A). The number of CPDs removed was cal-
culated (from the hybridization signal intensities from multiple
experiments), and the percent repair over time for the RPB2
gene in the wild-type strain in the absence or presence of
rapamycin are shown graphically in Figure 1B.
Linear slope estimates (0–30 min) were made based on a
no-intercept model and were then compared to evaluate
differences in initial rate of repair in the wild-type strain in
rapamycin (bottom panel). Exponentially growing wild-type cultures at 30?C were sham treated or treated with rapamycin, were UV irradiated at 60 J/m2, and then
incubated at the respective temperatures for the times indicated. DNA purified from the cells was digested with the restriction endonuclease HincII to generate
a 5.0-kb fragment containing RPB2. Restricted DNA was digested with TEV (þ) or mock treated (?) and then electrophoresed through 0.5% alkaline agarose.
DNA was transferred to Hybond Nþ membranes and sequentially hybridized with an RNA probe specific for either the transcribed or the nontranscribed strand of
the RPB2 gene. (B) Rapamycin decreases repair of RPB2 after UV damage in a wild-type strain. Strand-specific repair after UV irradiation in wild-type cells grown
at 30?C and treated with rapamycin (n) or vehicle (d) for 1.75 h. The average percent repair of the transcribed (–––) and nontranscribed (- - -) strands is calculated
from at least three repair experiments. Vertical bars at each point indicate SE of the mean.
(A) Representative autoradiograms from strand-specific repair assays in RPB2 in the wild-type strain in the absence (top panel) or presence of
DNA REPAIR INHIBITION BY RAPAMYCIN
the absence or presence of rapamycin. In the absence of
rapamycin, the rate of repair of the transcribed strand is ~3.5-
fold faster than that of the nontranscribed strand (Fig. 1B). In the
presence of rapamycin, the rate of repair of the transcribed
strand is ~2.1-fold faster than that of the nontranscribed strand.
By 90 min, repair reaches 80% with rapamycin versus 96%
without treatment. Thus, there was a significant reduction in the
rate of repair of the transcribed strand of RPB2 due to rapamycin
treatment (p < 0.01). The level of repair in the nontranscribed
strand in either gene is also reduced about 12% from 30 to 90
min following irradiation, although the rates of repair of the
nontranscribed strands are not statistically different (p > 0.05).
Thus, wild-type cells that were exposed to rapamycin showed
a decrease in the rate of repair in the RPB2 gene.
Strand-Specific Repair of UV-Induced DNA Damage in tor
Having demonstrated a decrease in repair following treatment
with the TORC1 inhibitor rapamycin, we next measured repair
in cells where Tor activity was shut off genetically using
deletion and/or conditional tor mutants. Cultures of the wild-
type strain and the tor mutant strains at the permissive (30?C)
and nonpermissive (37?C) temperatures were UV irradiated as
described in ‘‘Materials and Methods’’ section.
For the RPB2 gene, the wild type and each mutant at 30?C
and 37?C exhibit TCR where the transcribed strand is repaired
faster than the nontranscribed strand (Fig. 2). At 30?C, the
transcribed strand in the wild-type strain is repaired ~2.5-fold
faster than the nontranscribed strand (p < 0.01). Similarly, at
37?C, repair of the transcribed strand is about 2.5-fold greater
than that of the nontranscribed strand (p < 0.01). However, the
rate of repair of the transcribed and the nontranscribed strands
is significantly greater at 37?C than at 30?C with p < 0.01 and
p < 0.05, respectively. The tor1 mutant displays repair rates
similar to the wild-type strain with the exception that repair of
the nontranscribed strand was not statistically elevated at 37?C
compared to the repair observed at 30?C (Fig. 2).
(C) tor2ts, and (D) tor1tor2tsmutants grown at 30?C (d) and 37?C (:). The average percent repair of the transcribed (–––) and nontranscribed (- - -) strands is
calculated from at least three repair experiments. Vertical bars at each point indicate SE of the mean.
Mutations in TOR1 and TOR2 have no effect on TCR in a class II gene. Strand-specific repair of a class II gene, RPB2, in (A) the wild type, (B) tor1,
LIMSON AND SWEDER
Repair of RPB2 in the tor2tsand tor1tor2tsmutants,
containing a single plasmid-borne copy of the same conditional
allele of TOR2 (tor2-21), is similar to the repair observed in the
wild-type and tor1 strains with some subtle differences. Repair
of the transcribed strands is faster than that of the non-
transcribed strands at 30?C and 37?C (p < 0.01 and p < 0.01,
respectively). Interestingly, the initial repair rates of the
transcribed strand in the tor2tsand tor1tor2tsmutants at both
30?C and 37?C were intermediate between the repair observed
in the wild-type and tor1 strains at 30?C and 37?C.
Consequently, the significantly faster repair of the transcribed
strand at 37?C relative to repair at 30?C seen in wild-type and
tor1 (Fig. 2) strains is not apparent (i.e., significant) in the
tor2tsand tor1tor2ts(Fig. 2). As in the tor1 mutant, there is no
significant change in repair of the nontranscribed strand of
RPB2 at 37?C compared to 30?C in either the tor2tsor the
Strand-Specific Repair of UV-Induced DNA Damage in the
tor1tor2tsMutant at the Nonpermissive Temperature
Treated with Rapamycin
The minimal differences in repair between the wild-type and
tor mutant strains lead us to directly test whether or not
rapamycin may inhibit repair independently of Tor1 or Tor2.
When the tor1tor2tsmutant, growing at the nonpermissive
temperature for 3.25 h and lacking TORC1 and TORC2 activity,
was treated with 0.2 lg/ml rapamycin for an additional 1.75 h,
repair of both strands in the RPB2 gene surprisingly decreased
(Fig. 3). In the RPB2 gene, the rate of repair of the transcribed
strand is very rapid over the first 30-min postirradiation and
significantly faster than repair of the nontranscribed strand over
the same period (p < 0.05). Under the same conditions and in
the presence of rapamycin, the rate of repair of the transcribed
strand is reduced significantly (p < 0.05) relative to repair in the
absence of rapamycin. Strikingly, repair rates for the non-
transcribed strand in the presence of rapamycin were also greatly
diminished relative to repair rates in the absence of rapamycin
(p < 0.01). As described above in the experiments measuring
repair in the tor1tor2tsmutant at the nonpermissive and
permissive temperatures, repair of the nontranscribed strand of
RPB2 in tor1tor2tsis not affected by a temperature shift. Thus,
repair in both strands of RPB2 is significantly reduced in the
tor1tor2tsmutant treated with rapamycin and grown at the
nonpermissive temperature and lacking TORC1/2 activity.
Strand-Specific Repair of UV-Induced DNA Damage in an
curs, weexaminedrepairinmutantslacking a rapamycin-binding
a tor1tor2tsmutant but not in an fpr1 or fap1 mutant. Strand-specific repair after
UV irradiation in (A) fpr1 mutant grown at 30?C and treated with rapamycin
(n) or vehicle (h) for 1.75 h, (B) fap1 mutants grown at 30?C and treated with
rapamycin (¤) or vehicle (e) for 1.75 h, and (C) tor1tor2tsmutants grown at
37?C and treated with rapamycin (n) or vehicle (:) for 1.75 h. The average
percent repair of the transcribed (–––) and nontranscribed (- - -) strands is
Rapamycin decreases repair of RPB2 after UV damage in
calculated from at least three repair experiments. Vertical bars at each point
indicate SE of the mean. The dashed lines lacking symbols in (A and B)
represent repair of the transcribed strand in the wild-type strain treated with
rapamycin (see Fig. 1B).
DNA REPAIR INHIBITION BY RAPAMYCIN
protein, Fpr1. The inhibitory effect of rapamycin on the TORC1
kinase is dependent upon binding of the rapamycin-Fpr1
complex. Simultaneously, when rapamycin binds to Fpr1,
rapamycin competitively releases proteins that interact with
Fpr1. Two such Fpr1-binding factors are DNA-binding proteins:
the high-mobility group (HMG) protein Hmo1 and the putative
transcription factor Fap1.
We first determined if disruption of the FPR1 gene
influences the rates and extent of NER in UV-irradiated cells.
Cultures of the fpr1 mutant strain were mock treated or treated
with rapamycin prior to UV irradiation and during the time
allowed for repair. In the rapamycin-resistant fpr1 mutant,
rapamycin does not significantly affect the repair of either
strand in RPB2 (Fig. 3). In the presence or absence of
rapamycin, repair in the transcribed strand of RPB2 is not
significantly different than the wild type without rapamycin (p >
0.05) but is greater than the repair in the rapamycin-treated
wild-type strain (p < 0.05).
In the presence or absence of rapamycin, the rate of repair in
the nontranscribed strand of RPB2 in fpr1 is also not
significantly different than the wild type in the presence or
absence of rapamycin (Figs. 1B and 3). In the presence of
rapamycin, the level of repair in the nontranscribed strand in an
fpr1 mutant approximates that of the nontranscribed strand of
RPB2 in the wild type treated with rapamycin (p > 0.05).
Similarly, in the absence of rapamycin, repair in the non-
transcribed strand of RPB2 in fpr1 is not significantly different
than that of the wild type treated with rapamycin (p > 0.05).
Strand-Specific Repair of UV-Induced DNA Damage in
a fap1 Mutant
Following release from Fpr1, Fap1 may function as
a transcriptional repressor, like its mammalian homolog, to
inhibit transcription and thereby inhibit TCR. Such an indirect
effect on TCR can be measured in our assay by using a fap1
mutant. In a fap1 mutant, transcription of genes inhibited by
Fap1 would not be repressed. Thus, rapamycin treatment in
a fap1 mutant should not affect the inhibition of repair
normally caused by Fap1. However, rapamycin bound to Fpr1
in a fap1 mutant will still inhibit the TORC1 activity regulating
transcription (N.B. fap1 disruption mutants are as rapamycin
sensitive as the wild-type parental strain; Kunz et al., 2000).
Thus, the inhibition of transcription that is indirectly measured
by repair would be reflected in a similar rate of repair between
an fap1 mutant and an fpr1 mutant.
Cultures of the fap1 mutant strain were mock treated or
treated with rapamycin, irradiated with UV light, and allowed
time for repair. There is little or no difference in the level of
repair of either strand in RPB2 after UV damage when fap1 is
treated with rapamycin or mock treated (Fig. 3). Although the
initial rate of repair of the nontranscribed strand in the fap1
mutant without rapamycin is initially less than that when
treated with rapamycin (p < 0.05), the nontranscribed strands
from both conditions reach only about 45% repair by 90 min
(Fig. 3). Repair rates of the transcribed strand in the first 30 min
in the presence or absence of rapamycin are not significantly
different (p > 0.05) and are similar to the repair in the trans-
cribed strand in the wild type without rapamycin (p > 0.05). In
contrast, repair rates of the transcribed strand in the presence of
rapamycin are significantly different than that for the trans-
cribed strand in the wild-type strain treated with rapamycin
(p < 0.01). Additionally, at later time points, the level of repair
in the fap1 mutant is approximately equal to the level of repair
in the fpr1 mutant treated with rapamycin or not, reaching
about 80% by 90 min. Thus, rapamycin does not affect the rate
of repair for RPB2 in a fap1 mutant, which has an overall rate
of repair in RPB2 near that of the wild type and fpr1 mutant.
We examined the effect of rapamycin treatment upon strand-
specific repair of UV-induced damage in a gene transcribed by
RNA pol II in a repair-proficient Saccharomyces cerevisiae
strain. By treating yeast with rapamycin, we observed a decrease
in TCR. Like most laboratory strains, the parental strain JK9-
3da exhibits classical TCR in which the transcribed strand of
class II genes is repaired faster than the nontranscribed strand.
In the presence of rapamycin repair rates of the transcribed
strand of RPB2 were significantly reduced, and repair rates
of the nontranscribed strand were slightly reduced. Though
rapamycin inhibits ribosomal protein gene transcription by
RNA pol II, other genes transcribed by RNA pol II are not
repressed (Hardwick et al., 1999; Shamji et al., 2000). Thus, the
effect on TCR within the RPB2 gene is not a direct inhibition of
transcription. Additionally, we do not think that the effect of
rapamycin on TCR in wild-type cells simply reflects the
inhibitory effect of rapamycin on cell growth. We and others
have published data demonstrating that cultures that have been
arrested in the cell cycle due to cycloheximide treatment,
temperature-sensitive mutants, or hydroxyurea treatment have
no deficit in TCR related to untreated samples (Conconi et al.,
2002; Lommel et al., 2000; Sweder and Hanawalt, 1992).
Having shown that rapamycin inhibits the repair rate of
transcribed strands, we then examined whether or not this
inhibition was mediated through the Tor signaling pathway. We
inhibited Tor kinase activity genetically, by using yeast mutants
containing a disrupted TOR1 allele or a conditional temperature-
sensitive TOR2 allele or both. At the nonpermissive temperature
(37?C), tor1tor2tsmutants lack both TORC1 and TORC2 kinase
activities. We found increased repair of RPB2 in the wild-type
strain at 37?C relative to 30?C, consistent with increased rates
for cellular processes at higher temperatures. We found subtle
differences in the rates of repair among the tor1, tor2ts, and
tor1tor2tsmutants at the permissive and nonpermissive temper-
atures, but these differences were not statistically significant
relative to the wild-type strain. Additionally, we observed an
unambiguousinhibitionofrepair at thenonpermissive
LIMSON AND SWEDER
temperature when the tor1tor2tsstrain was treated with
rapamycin prior to and during the time of repair. Rapamycin-
resistant mutants TOR1-1 and TOR2-1 also showed marked
reduction of repair in the presence of rapamycin (data not
shown). We conclude that the rapamycin-induced inhibition of
repair is not a result of Tor kinase activity inhibition.
In addition to inhibiting Tor kinase activity, rapamycin
treatment causes Fpr1 to release factors that have the potential
to affect transcription and/or affect repair, such as Fap1.
Therefore, we examined repair of fpr1 and fap1 disruption
mutants in the absence or presence of rapamycin. Unlike the
wild-type strain, where rapamycin treatment reduced repair of
the transcribed strand significantly, the overall levels of repair
for both strands of the gene examined in the rapamycin-resistant
fpr1 mutant and the fap1 mutant, treated or not with rapamycin,
are similar to the levels of repair observed in the wild-type strain
in the absence of rapamycin. That Fpr1 associates with Fap1 in
the absence of rapamycin may provide an explanation for this
observation (Kunz et al., 2000). In the presence of rapamycin,
the cytoplasmic Fpr1 binds to rapamycin and is sequestered
from its functional interaction with Fap1. Fap1 then relocates to
the nucleus and interacts with DNA as a putative transcriptional
repressor for as yet unidentified genes (Kunz et al., 2000).
Simultaneously, the Fpr1-rapamycin complex inhibits TORC1
activity and affects transcriptional regulation of other genes.
In the rapamycin-resistant fpr1 mutant, rapamycin is neither
able to inhibit the TORC1 complex nor regulate factors
associated with Fpr1. In this case, Fap1 is not bound to Fpr1.
However, Fap1 has been observed to remain in the cytoplasm
in cells lacking Fpr1 (Kunz et al., 2000). In the fap1 mutant,
Fap1 is not available to repress transcription (or as yet
unknown functions) in the presence or absence of rapamycin,
despite being as rapamycin sensitive as the wild-type parental
strain (recall that the fap1 mutant still possesses functional Fpr1
protein that can bind rapamycin and inhibit Tor activity). Thus,
in either mutant, wild-type levels of repair were observed. The
decrease of repair in RPB2 observed after rapamycin treatment
in the wild-type strain led us to conclude that the control of the
NER response regulated by rapamycin is twofold. First,
repression of TCR is not due to the inhibition of Tor regulation.
Second, rapamycin competitively releases factors interacting
with Fpr1 that subsequently repress TCR.
Another likely candidate that interacts with Fpr1 and
competes with rapamycin is Hmo1, a HMG protein in yeast
(Dolinski and Heitman, 1999; Lu et al., 1996). As a chromatin-
associated HMG-like protein, Hmo1 binds DNA by structure
rather than sequence specificity. As Cardenas et al. (1999)
suggested, Fpr1 might regulate the assembly of Hmo1
complexes or Hmo1-DNA interactions. The same phenotype
may exist in the absence of Fpr1, as in the fpr1 mutant, where
Hmo1 can oligomerize and affect DNA repair mechanisms.
Perhaps, Fpr1 may have other unidentified associated factors
that may contribute to regulating TCR and interfere with the
cellular response to DNA damage.
In summary, we expected that conditional mutants lacking
Tor function would behave like rapamycin-treated cells with
respect to repair deficiencies. Instead, we observed repair rates
of a class II gene in tor1tor2ts(and in the other tor mutants)
similar to repair rates in the wild-type strain. In contrast,
rapamycin treatment of the wild-type strain and the tor1tor2ts
mutant reduced NER, indicating that rapamycin was affecting
a factor other than the TORC complexes. We suspected the
rapamycin-binding protein Fpr1, and/or factors associated with
Fpr1, may indirectly inhibit repair by affecting transcription
independently of the Tor pathway. Fpr1 may have a role in
regulating DNA-binding proteins, such as Fap1 and Hmo1,
which may contribute to regulating transcriptional activity or
DNA accessibility. In the presence of rapamycin, these factors
are competitively released from Fpr1 and affect transcription or
the DNA damage response pathway in a manner yet to be
determined and subsequently affect DNA repair.
The findings reported here have important implications for
clinical applications of inhibitors of mTor, like rapamycin.
Coadministration of mTor inhibitors with genotoxic agents, as
in an oncology indication, will likely result in greater amounts
of DNA damage persisting into S phase such that DNA
replication will fix the damage into mutations or cell death may
occur. While such outcomes may result in tumor regression,
enhanced mutagenesis in normal tissues will likely increase
carcinogenesis in the long term.
Public Health Service grant (R29 GM53717 to K.S.S.),
a Minority Predoctoral Fellowship (F31 CA833430 to M.V.L.),
and the Initiative for Minority Student Development (R25
GM55145) from the National Institutes of Health.
The authors thank Carol Gleason for statistical analyses of
the repair data and Lori Lommel, Sean Gregory, Kiran Madura,
Joseph Heitman, and Michael N. Hall for helpful discussions
and critical reading of the manuscript.
Alarcon, C. M., and Heitman, J. (1997). FKBP12 physically and functionally
interacts with aspartokinase in Saccharomyces cerevisiae. Mol. Cell. Biol. 17,
Budanov, A. V., and Karin, M. (2008). p53 target genes sestrin1 and sestrin2
connect genotoxic stress and mTOR signaling. Cell 134, 451–460.
Cardenas, M. E., Cruz, M. C., Del Poeta, M., Chung, N., Perfect, J. R., and
Heitman, J. (1999). Antifungal activities of antineoplastic agents: Saccha-
romyces cerevisiae as a model system to study drug action. Clin. Microbiol.
Rev. 12, 583–611.
DNA REPAIR INHIBITION BY RAPAMYCIN
Conconi, A., Bespalov, V. A., and Smerdon, M. J. (2002). Transcription- Download full-text
coupled repair in RNA polymerase I-transcribed genes of yeast. Proc. Natl.
Acad. Sci. U.S.A. 99, 649–654.
De Virgilio, C., and Loewith, R. (2006). Cell growth and control: little
eukaryotes make big contributions. Oncogene 25, 6392–6415.
Dolinski, K. J., and Heitman, J. (1999). Hmo1p, a high mobility group 1/2
homolog, genetically and physically interacts with the yeast FKBP12 prolyl
isomerase. Genetics 151, 935–944.
Friedberg, E., Walker, G. C., and Siede, W. (1995). In DNA Repair and
Mutagenesis. ASM Press, Washington, DC.
Hardwick, J. S., Kuruvilla, F. G., Tong, J. K., Shamji, A. F., and
Schreiber, S. L. (1999). Rapamycin-modulated transcription defines the
subset of nutrient- sensitive signaling pathways directly controlled by the Tor
proteins. Proc. Natl. Acad. Sci. U.S.A. 96, 14866–14870.
Heitman, J., Movva, N. R., and Hall, M. N. (1991). Targets for cell cycle arrest
by the immunosuppressant rapamycin in yeast. Science 253, 905–909.
Hoekstra, M. F. (1997). Responses to DNA damage and regulation of cell cycle
checkpoints by the ATM protein kinase family. Curr. Opin. Genet. Dev. 7,
Kunz, J., Loeschmann, A., Deuter-Reinhard, M., and Hall, M. N. (2000).
FAP1, a homologue of human transcription factor NF-X1, competes with
rapamycin for binding to FKBP12 in yeast. Mol. Microbiol. 37, 1480–1493.
Kuruvilla, F. G., Shamji, A. F., and Schreiber, S. L. (2001). Carbon- and
nitrogen-quality signaling to translation are mediated by distinct GATA-type
transcription factors. Proc. Natl. Acad. Sci. U.S.A. 98, 7283–7288.
Lommel, L., Gregory, S. M., Becker, K. I., and Sweder, K. S. (2000).
Transcription-coupled DNA repair in yeast transcription factor IIE (TFIIE)
mutants. Nucleic Acids Res. 28, 835–842.
Lorenz, M. C., and Heitman, J. (1995). TOR mutations confer rapamycin
resistance by preventing interaction with FKBP12-rapamycin. J. Biol. Chem.
Lu, J., Kobayashi, R., and Brill, S. J. (1996). Characterization of a high
mobility group 1/2 homolog in yeast. J. Biol. Chem. 271, 33678–33685.
Powers, T., and Walter, P. (1999). Regulation of ribosome biogenesis by the
rapamycin-sensitive TOR-signaling pathway in Saccharomyces cerevisiae.
Mol. Biol. Cell 10, 987–1000.
Raught, B., Gingras, A. C., and Sonenberg, N. (2001). The target of rapamycin
(TOR) proteins. Proc. Natl. Acad. Sci. U.S.A. 98, 7037–7044.
Rohde, J., Heitman, J., and Cardenas, M. E. (2001). The TOR kinases link
nutrient sensing to cell growth. J. Biol. Chem. 276, 9583–9586.
Schmelzle, T., and Hall, M. N. (2000). TOR, a central controller of cell growth.
Cell 103, 253–262.
Shamji, A. F., Kuruvilla, F. G., and Schreiber, S. L. (2000). Partitioning the
transcriptional program induced by rapamycin among the effectors of the Tor
proteins. Curr Biol. 10, 1574–1581.
Song, Z., Krishna, S., Thanos, D., Strominger, J. L., and Ono, S. J. (1994). A
novel cysteine-rich sequence-specific DNA-binding protein interacts with the
conserved X-box motif of the human major histocompatibility complex class
II genes via a repeated Cys-His domain and functions as a transcriptional
repressor. J. Exp. Med. 180, 1763–1774.
Sweder, K. S., and Hanawalt, P. C. (1992). Preferential repair of cyclobutane
pyrimidine dimers in the transcribed strand of a gene in yeast chromosomes
and plasmids is dependent on transcription. Proc. Natl. Acad. Sci. U.S.A. 89,
Yarosh, D. B., Cruz, P. D., Dougherty, I., Bizios, N., Kibitel, J., Goodtzova, K.,
Both, D., Goldfarb, S., Green, B., and Brown, D. (2000). FRAP DNA-
dependent protein kinase mediates a late signal transduced from ultraviolet-
induced DNA damage. J. Invest. Dermatol. 114, 1005–1010.
Zaragoza, D., Ghavidel, A., Heitman, J., and Schultz, M. C. (1998).
Rapamycin induces the G0 program of transcriptional repression in yeast
by interfering with the TOR signaling pathway. Mol. Cell. Biol. 18,
LIMSON AND SWEDER