ABSTRACT
We developed a competition assay to compare, in a quantitative manner, the
ability of human nucleotide excision repair (NER) to recognise structurally different forms of DNA damage. This assay uses a NER substrate consisting of M13 double-stranded DNA with a single and uniquely located acetylaminofluorene (AAF) adduct, and measures the efficiency by which multiply damaged plasmid DNA
competes for excision repair of the site-directed modification. To validate this assay, we tested competitor DNA containing defined numbers of either AAF adducts or UV radiation products. In both cases, repair of
the site-directed NER substrate was inhibited in a damage-specific and dose-dependent manner. We then exploited this competition assay to
determine the susceptibility of bulky adozelesin-DNA adducts to human NER.
Nucleotide excision repair (NER) is a multistep process that catalyses incision
of damaged DNA strands on either side of the lesions and removes DNA damage as
a component of oligonucleotide segments (
1
-
4
). The correct sequence is then reestablished by DNA repair synthesis, followed
by ligation of the repair patches to the preexisting strands. In eukaryotes,
NER incises DNA at the 3rd to 5th phosphodiester bond 3' to the lesion, and at the 21st to 25th phosphodiester bond on the 5' side, yielding damage-containing oligomers of 27-29 residues in length (
2
,
5
).
Most eukaryotic cells depend on NER to remove the principal forms of DNA damage induced by ultraviolet (UV) radiation (
2
,
3
,
6
). However, NER-defective yeast or human (xeroderma pigmentosum) cells display
hypersensitivity to killing by a very broad spectrum of genotoxic agents (
7
,
8
), indicating that this system is involved in the excision of virtually any kind
of bulky DNA modification.
In vitro
studies with human cell extracts confirmed that NER is a versatile mechanism
that processes a wide range of base adducts, such as, for example, those
generated by the carcinogen
N
-acetoxy-2-acetylaminofluorene or the chemotherapeutic agents
cis
-diamminedichlorplatinum and psoralen (
9
-
12
). On the other hand, it was found that these adducts are recognised and removed
by human NER with markedly different efficiencies (
13
). A heterogeneous NER response is also indicated by the kinetics of
photoproduct removal in UV-irradiated human cells. In fact, pyrimidine(6-4)pyrimidone photoproducts are excised with a significantly shorter half-life than cyclobutane pyrimidine dimers (
14
,
15
). In the present study, we have set up a novel assay to analyse the substrate
preferences of human NER in a quantitative manner
.
This assay is based on the ability of damaged plasmids to compete with a site-specific acetylaminofluorene (AAF) adduct. Using this competition assay, we observed that UV radiation products are
recognised 30 times less effectively than the AAF modification, while bulky
adducts produced by adozelesin, an antitumor antibiotic of the
cyclopropylpyrroloindole type, are largely refractory to recognition by human
NER
in vitro
.
T4 polynucleotide kinase and T4 DNA ligase were purchased from Gibco-BRL. Restriction enzymes were purchased from New England Biolabs. Creatine
phosphokinase and ribonuclease A were from Boehringer Mannheim.
The 19 nucleotide (nt) oligomer 5'-ACCACCCTTCGAACCACAC-3' was phosphorylated by incubation with ATP and T4
polynucleotide kinase, and reacted with
N
-acetoxy-2-acetylaminofluorene (NCI Chemical Carcinogen Reference
Standard Repository) to form an AAF adduct at the single guanine residue (
10
).
The method used to produce site-specifically modified M13 DNA was adapted from Comess
et al.
(
16
), and is based on the construction of a gapped intermediate consisting of a circular (+)strand and a
linear (-)strand (
17
). This gapped intermediate contains a single-stranded region of 19 nt that is complementary to the 19mer
oligonucleotide described above. Briefly, the M13 derivative M13mp19G was
obtained by ligating the synthetic duplex:
5'-GGTGTGGTTCGAAGGGTGGT-3'
3'-ACGTCCACACCAAGCTTCCCACCAGATC-5'
into the
Pst
I-
Xba
I site of M13mp19 double-stranded DNA, followed by transfection of the ligation product into
Escherichia coli
strain DH5[alpha]F'. The resulting single-stranded DNA provides the (+)strand of the gapped
intermediate. The complementary (-)strand was obtained by ligation of the duplex:
5'-GACGTCGATATCGTGCA-3'
3'-ACGTCTGCAGCTATAGCACGTGATC-5'
into the
Pst
I-
Xba
I site of M13mp19 DNA, generating a second M13 derivative designated M13mp19Hb5. The insert in this DNA molecule contains
the restriction sites for
Aat
II (5'-GACGT'C) and
Apa
LI (5'-G'TGCAC). After amplification in strain DH5[alpha]F', M13mp19Hb5 DNA was linearised by digestion with
Aat
II and
Apa
LI, and the obtained large fragment was separated by gel filtration chromatography. M13mp19G single-stranded DNA (typically 1.2 mg) and the large fragment of M13mp19Hb5 (200 [mu]g) were coincubated for 3 min at 95oC, followed by 15 min at 65oC and 2 h at room temperature in a volume of 0.6 ml containing
60 mM Tris-HCl, pH 8.0, 100 mM NaCl and 10 mM EDTA. The resulting gapped circular
DNA was purified by benzoylated naphthoylated DEAE cellulose (Sigma) and
ethanol precipitation as indicated (
18
). Modified 19mer oligonucleotides were then ligated into this intermediate in
0.9 ml reactions containing 30 [mu]g M13 gapped DNA, 0.5 [mu]g phosphorylated oligonucleotides and 50 U T4 DNA ligase in 50 mM Tris-HCl, pH 7.6, 10 mM MgCl
2
, 1 mM ATP and 5% (w/v) polyethylene glycol 8'000. After incubations of 4 h at 16oC, covalently closed duplex DNA was purified by CsCl gradient
centrifugation. Control M13 substrates were constructed by ligating unmodified
19mers into the gapped DNA.
Plasmid pUC19 was prepared by alkaline lysis from
Escherichia coli
strain DH5[alpha] grown without chloramphenicol amplification, and purified by CsCl
followed by 5-20% sucrose gradient centrifugation (
19
). To obtain AAF-DNA adducts, pUC19 (50 [mu]g/ml) was reacted with 0.1 mM
N
-acetoxy-2-acetylaminofluorene in 2 mM sodium citrate, pH 7.0, at 25oC for 3 h. The unreacted carcinogen was then extracted
five times with ether as described (
20
), and DNA was precipitated with ethanol and repurified through a 5-20% sucrose gradient. Using radioactively labelled [
3
H]
N
-acetoxy-2-acetylaminofluorene (NCI Chemical Carcinogen Reference
Standard Repository), we determined that this modification protocol produced
10.2 +- 0.9 AAF adducts/pUC19 molecule. For UV irradiation, aliquots of 20 [mu]l containing pUC19 plasmids (100 [mu]g/ml) in 10 mM Tris-HCl, pH 8.0, 1 mM EDTA were placed on ice in an open
Petri dish and irradiated at a dose rate of 2 J
.
m
-2.
s
-1
using a germicidal lamp with peak output at 254 nm. The UV fluence was
monitored with a Steritest dosimeter. Adozelesin-DNA adducts were prepared by incubating pUC19 (333 [mu]g/ml) with 3.3 [mu]M adozelesin in 1.5 mM sodium citrate, pH 7.0, and 15 mM NaCl for
120 min at 25oC, followed by ethanol precipitation and 5-20% sucrose gradient centrifugation. In aqueous solutions, the UV
absorption spectrum of adozelesin is characterised by two peaks at 315 and 370
nm, with a molar extinction coefficient of 21 000 at 370 nm (data not shown). This latter value was used to estimate the frequency of adozelesin
modification after UV spectrophotometry of the purified drug-plasmid DNA complexes (13.1 DNA adducts/pUC19).
HeLa cell extracts were prepared by the method of Manley
et al.
(
21
). Reactions (50 [mu]l) were slightly modified from Hansson
et al.
(
10
) and contained HeLa cell extract (80 [mu]g of proteins), 50 or 100 ng M13 DNA, various amounts of competitor pUC19 DNA, 45 mM HEPES, pH 7.8, 70 mM KCl, 7.4 mM MgCl
2
, 0.9 mM dithiothreitol, 0.4 mM EDTA, 3.4% glycerol, 2 mM ATP, 20 [mu]M each of dATP, dGTP and dTTP, 8 [mu]M dCTP, 2.0 [mu]Ci [[alpha]-
32
P]dCTP (3000 Ci/mmol), 40 mM phosphocreatine, 2.5 [mu]g creatine phosphokinase and 18 [mu]g bovine serum albumin. After 3 h at 30oC, reactions were stopped by the addition of EDTA to 20 mM. The samples were incubated at 37oC with ribonuclease A (80 [mu]g/ml) for 10 min, SDS to 0.5% and proteinase K to 190 [mu]g/ml were then added, and the mixtures incubated for a
further 45 min at 37oC. DNA was extracted, digested with
Ava
II,
Sma
I and
Pst
I, and analysed by 20% polyacrylamide gel electrophoresis and autoradiography.
This enzymatic restriction produces
Sma
I-
Pst
I fragments of 37 base pairs (bp) containing the NER patches. Nucleotide incorporation within this 37mer fragment was quantified by scanning densitometry
of the X-ray films on a Molecular Dynamics Computing Densitometer using ImageQuant
software.
As illustrated in Figure
1
, the repair competition assay is performed by coincubating in the same reaction
a site-directed NER substrate and competitor DNA (plasmid pUC19) damaged with the
lesion of interest. The assay measures the efficiency by which NER of the
substrate is inhibited by the presence of DNA damage on the competitor. NER is the sole excision repair mechanism for bulky base adducts
such as the AAF modification (
2
,
9
,
11
,
22
). Thus, competition for repair of a site-specific AAF adduct is expected if pUC19 contains lesions that are also
recognised by the NER system.
To validate the competition assay, we first tested the situation in which both
substrate and competitor DNA contain the same lesion, i.e.
AAF-DNA adducts. For that purpose we mixed singly modified M13 DNA substrate
with pUC19 containing 10.2 +- 0.9 AAF adducts/molecule (see Materials and methods for adduct quantification). This mixture of substrate (100 ng)
and competitor DNA (50 ng) was incubated in HeLa cell extract in the presence
of
32
P-labelled deoxynucleotides. Figure
2
A shows reactions containing site-specifically modified M13 DNA (lanes 3 and 4, in duplicate), and control reactions containing unmodified substrate (lanes 1 and 2). Among the three fragments obtained by restriction
digestion, only the 37mer
Sma
I-
Pst
I fragment displayed ~30-fold increased nucleotide incorporation in response to the site-directed AAF adduct, indicating NER of the bulky modification.
This NER response was absent in reactions performed with extracts from XPA or XPC cells (data not shown). Nucleotide
incorporation observed in the two long fragments of 330 and 6920 bp represents background DNA synthesis unrelated to NER, presumably initiated at
nicks generated by unspecific nucleases (
10
,
22
,
23
). Lanes 5 and 6 demonstrate that the specific NER response within the 37mer
fragment was completely suppressed by the addition of 50 ng pUC19 competitor
DNA containing an average of 10.2 AAF modifications. In contrast, the
unspecific background synthesis into the long fragments of 330 and 6920 bp was
not reduced by the presence of this competitor (lanes 5 and 6).
UV radiation products are the most widely tested form of DNA damage. We therefore calibrated our system with competitor DNA exposed to
different doses of UV radiation at 254 nm wavelength. M13 DNA substrate containing the site-directed AAF adduct (50 ng) was mixed with an identical amount of pUC19 competitor DNA that
was either untreated or UV-irradiated. No inhibition of nucleotide incorporation into the 37mer
fragment of M13 DNA was obtained in the presence of unmodified pUC19 (see Fig.
4
), demonstrating that NER is not suppressed by undamaged competitor. However, nucleotide incorporation into the 37mer fragment was progressively
reduced with increasing exposure of competitor DNA to UV light (Fig.
3
). Thus, UV radiation products effectively compete with the AAF substrate for
NER. Half-maximal inhibition was found at doses ~450 J
.
m
-2
, while ~20% of control activity was detected at 900 J
.
m
-2
. This dose dependence yielded a nearly linear relationship between 200 and 1200
J
.
m
-2
when the data was plotted on a logarithmic scale (Fig.
3
). In a previous report, UV treatment at 450 J
.
m
-2
has been shown to induce a total of 15-16 UV radiation products on a plasmid of 3658 bp (
24
). Irradiation of the smaller plasmid pUC19 with an identical dose is expected
to induce 11-12 UV photoproducts/pUC19. Taken together, UV-irradiated competitor DNA containing 11-12 lesions/plasmid inhibited repair of the AAF substrate to ~50% at a M13 substrate to pUC19 competitor mass ratio
of 1:1, corresponding to a molar ratio between the two DNA molecules of 1:2.7.
Adozelesin is the lead compound for a series of synthetic DNA-reactive agents that have entered clinical trials for potential use as
antitumor drugs (
25
,
26
). These agents alkylate DNA at N3 of adenine with high selectivity and
considerable sequence specificity, and generate bulky adducts that are mainly
confined within the minor groove (
27
,
28
).
Plasmid pUC19 was reacted with adozelesin to obtain an average of 13.1
adducts/plasmid molecule, as estimated from the spectrophotometric analysis of purified drug-DNA complexes (see Materials and methods). When tested in the competition assay, adozelesin-modified DNA was unable to compete with the repair of AAF adducts
on the substrate, even when added in large excess (Fig.
4
). Specifically, adozelesin-modified pUC19 was tested at substrate to competitor mass ratios of 1:1
and 1:3, corresponding to molar ratios of 1:2.7 and 1:8.1, but no competitive
inhibition was observed. The small inhibition found at the 1:8.1 molar ratio
was not significantly different from that observed with identical amounts of
undamaged pUC19. In parallel control reactions, AAF- or UV-damaged (500 J
.
m
-2
) pUC19 were effective competitors (Fig.
4
).
The repair competition assay requires a site-directed NER substrate (M13 double-stranded DNA with a single AAF adduct in a unique sequence),
competitor DNA (multiply damaged plasmid pUC19), and a NER-proficient extract prepared from HeLa cells. Although the assay was
developed with human cell extracts, it should be applicable to rodent,
Xenopus
egg, or yeast extracts. The assay determines the capacity of a particular DNA
lesion located on plasmid pUC19 to inhibit repair of the site-specific AAF adduct by competing for repair factors. As its main
advantage, this novel method provides a biochemical assay for quantitative
comparisons between structurally distinct DNA lesions.
The authors thank Barbara Zweifel for outstanding technical assistance and Dr J. P. McGovren (The Upjohn Company, Kalamazoo, Michigan) for
the generous gift of adozelesin. This work was supported by grant 31-40307.94 from the Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung and by the Wolfermann-Naegeli-Stiftung, Zürich.
REFERENCES
Return