Office
of Biological and Environmental Research
DOE
Lowdose Radiation Program Workshop V
Abstract
Title: Mechanisms of Low-Dose Inducible DNA Repair and the
Adaptive Response
Authors: Helen Budworth, Janice Pluth, Hengameh
Zahed Kargaran, Sophia Chernikova, Eric Campeau, Isi Tolliver,
Altaf Sarker, and Priscilla K. Cooper
Institutions: Life
Sciences Division, Lawrence Berkeley National Laboratory,
Berkeley, CA 94720
Neither the underlying mechanism nor the
generality and extent of a protective effect to subsequent
radiation-induced or endogenous damage (the radio-adaptive
response) that can be induced by exposure to low level ionizing
radiation (LLIR) is well understood at present. Although
DNA damage signaling and DNA repair processes appear likely
to be involved, the detailed nature and regulation of repair
processes responsive to LLIR are still largely unknown, and
important interconnections between the relevant repair pathways
largely remain to be elucidated. We have proposed a requirement
for transcription-coupled repair (TCR) in the adaptive response
through an LLIR-inducible base excision repair (BER) process.
Our working hypothesis is that the salient damage induced
both by LLIR and endogenous sources is oxidative base damage
and strand breaks that, when encountered by a replication
fork, result indirectly in double-strand breaks (DSBs) that
require recombinational repair coordinated with recruitment
of the TC-BER machinery for removal of the lesion and restoration
of fork progression. We further hypothesize that the efficiency
of this coupled process is improved by induction of the relevant
repair processes, in particular BER. Specifically, we are
(1) investigating the relationship between transcription-coupled
repair (TCR), inducible base excision repair (BER), and the
adaptive response, and (2) investigating the possibility
that the TCR machinery is required for repair of replication-blocking
oxidative lesions induced either by endogenous sources or
by LLIR and that it is recruited to stalled replication forks
as well as to blocked transcription complexes.
Our studies
of TCR in the context of the adaptive response are focused
on two key proteins, XPG and CSB, each of which has been
shown by our colleague M. Weinfeld to be required for LLIR-inducible
BER. XPG is a multifunctional DNA repair protein that is
a central player in both of the global excision repair processes
that repair lesions throughout the genome – nucleotide
excision repair (NER) and BER – as well as in the preferentially
rapid repair of lesions in transcribed strands of active
genes by TCR. Its functions in TCR and/or BER are essential
for normal postnatal development and viability. CSB is the
protein most commonly defective in the fatal TCR disorder
Cockayne syndrome (CS), which can also arise from mutations
affecting the CSA protein, XPG, or the XPB and XPD helicase
components of the basal transcription factor TFIIH. CSB is
a DNA-dependent ATPase with chromatin remodeling activity
and is recruited to RNA Polymerase II (RNAP II) stalled during
elongation by encountering DNA damage. Our recent studies
have shown that XPG also recognizes stalled RNAP II and that
XPG and CSB, which interact directly with each other, cooperatively
form supracomplexes with the stalled polymerase. We are approaching
the test of our hypothesis relating the TCR machinery to
the adaptive response by use of multiple approaches, as outlined
below.
Consistent with a role for XPG in LLIR-inducible
repair events and perhaps in the adaptive response, we have
preliminary evidence that XPG protein levels are increased
following exposure to an inducing dose of 5 cGy X-rays in
a lymphoblast cell line that had previously been shown to
undergo adaptation, and also in the same line after exposure
to low concentrations of H2O2. The generality of this finding
is currently being tested using fibroblasts. Strikingly,
a recent transcription profiling study by M. Coleman and
A. Wyrobek (pers. comm.) identified expression of only two
DNA repair proteins as correlating with the adaptive response – XPG
and ATM. This observation takes on additional significance
in view of our finding that XPG is phosphorylated constitutively
and hyperphosphorylated after IR, with both DNA-PKcs and
ATM being implicated as responsible kinases by metabolic
labeling studies using mutant cell lines. In order to study
the effect of adaptive regimens and damage signaling on repair
of oxidative DNA damage by XPG and CSB, we have developed
a rapid host cell reactivation assay that quantitates luciferase
reporter gene expression from an oxidatively damaged plasmid
transfected into human cells. In agreement with M. Weinfeld’s
finding using a capillary electrophoresis assay that efficient
removal of thymine glycol depends on XPG and CSB, we have
shown that luciferase expression from a plasmid treated with
OsO4 to induce thymine glycols is significantly lower in
cells from CS patients that are defective in XPG or CSB than
in cells from normal individuals. This approach is now being
used to determine whether exposure of the cells to an adaptive
LLIR regimen prior to transfection increases gene expression
from the oxidatively damaged plasmid and to test the dependence
of an induced increase on XPG and/or CSB. We will use the
same system to examine dependence on the signaling protein
ATM as well as on DNA-PKcs, since we have found that XPG
exists in a complex with DNA-PKcs in the cell. As an alternative
approach to studying the dependence of adaptation on TCR
proteins and ATM, we are developing conditions for application
of our ultrasensitive FACS-based method for quantitating γH2AX
levels to detect the occurrence of adaptation. In order to
eliminate the possibility of confounding uncontrolled factors
arising from differences in genetic background in either
assay, we have developed a lentivirus system that allows
very efficient introduction of inducible expression vectors
for shRNA in order to deplete XPG (or other target proteins)
by RNA interference. We will thus be able to directly test
a requirement for XPG in adaptation by comparing the same
cell line with or without normal levels of XPG.
Using the
lentiviral system to express shRNA against XPG in U2OS
human cells, we have achieved a very substantial reduction
in XPG protein levels. Strikingly, the cells in which XPG
has been knocked down have a greatly reduced growth rate,
suggesting the possibility that XPG is required for normal
traversal through S phase. The growth defect is currently
being further characterized, but this observation is consistent
with the hypothesis that XPG is involved in resolution of
replication forks stalled by encounters with endogenous oxidative
lesions in DNA. To test this idea more directly, we have
biochemically fractionated cell lysates by successive extraction
with increasing salt and detergent concentrations to separate
the cellular proteins into soluble, chromatin-bound, and
nuclear matrix-associated fractions. We find that in asynchronous
cultures XPG translocates from the soluble fraction to both
the high-salt chromatin fractions and the nuclear matrix
fraction in response to DNA damage, and we postulate that
the chromatin-bound XPG is engaged in global repair whereas
the nuclear matrix-associated XPG is associated with either
stalled transcription or stalled replication forks. In agreement
with this interpretation and the idea that replication forks
encounter significant numbers of endogenous lesions during
every S-phase, we have found that in S-phase cells a significant
portion of XPG is in the nuclear matrix fraction without
DNA damaging treatments. The biochemical fractionation procedure
will be employed to investigate effects of an adaptive LLIR
regiment on the kinetics and extent of XPG translocation
in response to challenge doses as well as the possible dependence
of the translocation on damage signaling proteins including
ATM. We anticipate that the combination of approaches described
here will provide significant new information on the relationship
between inducible BER and the radio-adaptive response and
their importance in resolution of stalled replication forks.
