4970–4976 Nucleic Acids Research, 1997, Vol. 25, No. 24
 1997 Oxford University Press
A structure-specific endonuclease from cauliflower
(Brassica oleracea var. botrytis) inflorescence
Seisuke Kimura, Mihoko Kai, Hiroyuki Kobayashi1, Atsushi Suzuki1, Hiroshi Morioka1,
Eiko Otsuka1 and Kengo Sakaguchi*
Department of Applied Biological Science, Faculty of Science and Technology, Science University of Tokyo,
2641 Yamazaki, Noda-shi, Chiba-ken 278, Japan and 1Faculty of Pharmaceutical Science, Hokkaido University,
Sapporo, Japan
Received August 26, 1997; Revised and Accepted October 20, 1997
ABSTRACT
A protein with structure-specific endonuclease activity
has been purified to near homogeneity from cauliflower
(Brassica oleracea var. botrytis) inflorescence through
five successive column chromatographies. The protein
is a single polypeptide with a molecular mass of 40 kDa.
Using three different branched DNA structures (flap,
pseudo-Y and stem–loop) we found that the enzyme, a
cauliflower structure-specific endonuclease, cleaved
the single-stranded tail in the 5′-flap and 5′-pseudo-Y
structures, whereas it could not incise the 3′-flap and
3′-pseudo-Y structures. The incision points occur
around the single strand–duplex junction in these DNA
substrates and the enzyme leaves 5′-PO4 and 3′-OH
termini on DNA. The protein also endonucleolytically
cleaves on the 3′-side of the single-stranded region at the
junction of unpaired and duplex DNA in the stem–loop
structure. The structure-specific endonuclease activity is
stimulated by Mg2+ and by Mn2+, but not by Ca2+. Like
mammalian FEN-1, the protein has weak 5′→3′ doublestranded DNA-specific exonuclease activity. These
results indicate that the cauliflower protein is a plant
structure-specific endonuclease like mammalian FEN-1
or may be the plant alternative.
INTRODUCTION
DNA replication, recombination and repair are key processes to
maintain integrity of the genome. We are interested in proteins
that are associated with these processes in higher plants,
basidiomycetes and Drosophila (1–10). Repair of some types of
DNA double-strand breaks is thought to proceed through flap
DNA structure intermediates and, consequently, repair requires a
DNA structure-specific endonuclease to identify and cleave the
flap structure (11). Structure-specific endonucleases perform
structure-specific cleavage at the junction between the 5′
unannealed tail and that of a primer annealed to a template and
play a important role in these processes (11,12). Several different
branched DNA structures, such as the flap structure, have been
prepared and structure-specific nucleases such as mammalian flap
endonuclease-1 (FEN-1) have been found (11). FEN-1 is essential
for maturation of Okazaki fragments in DNA replication (13,14)
and recognizes a specific 5′-flap structure. Homologs of FEN-1 have
been identified in budding (15,16) and fission yeast (17). The yeast
genetic studies provide evidence that the FEN-1 homologs are
involved in DNA replication, repair and, possibly, recombination.
FEN-1 is a homolog of a Saccharomyces cerevisiae gene, RAD2.
XPG–RAD2 are essential for the incision steps of nucleotide
excision repair (NER) of UV-damaged DNA (12,18–20). We
recently found a structure-specific endonuclease that recognizes a
UV-induced (6–4) photoproduct lesion in DNA molecules in
Drosophila (7, Kai et al., in preparation). It was also thought to have
an important role in the (6–4) photoproduct repair process.
Deoxyinosine 3′-endonuclease, an Escherichia coli repair enzyme
that recognizes and cleaves DNA containing deoxyinosine and
base mismatches, could also cleave the 5′-single-stranded tails of
flap and pseudo-Y structures (21). These results indicate a variety
of structure-specific endonucleases, each of which recognizes one
or more damaged specific DNA structures. Each of the newly
found structure-specific endonucleases could, therefore, be a key
to study unknown repair reactions in which damaged DNA
structures are different.
We have extensively screened the structure-specific endonucleases in the branches of the evolutionary tree. Our primary
concern in this study was to identify a structure-specific endonuclease in higher plants. Although plants are constantly bombarded
by UV light (22), little is known about plant DNA structure-specific
endonucleases involved in the UV lesion repair process. We chose
cauliflower inflorescence as a study material because it is a
promising biochemical material in higher plants. Cauliflower
inflorescence can be easily collected year round and exhibits
hypertrophic differentiation with continuous growth by mitotic
division (23). Secondly, the apical portion of the inflorescence
consists of small cells and has a much higher DNA content than the
underlying axial tissues (24,25). Finally, the cauliflower has a close
relationship to Arabidopsis thaliana, which has become a model for
plant molecular genetics (26).
In the present study we found a structure-specific endonuclease
in cauliflower inflorescence. The cauliflower structure-specific
endonuclease, like FEN-1 and XPG–RAD2, cleaves the singlestranded tail in 5′-flap structures. Saccharomyces cerevisiae
RAD1–RAD10, which are respectively homologs of XPF–ERCC1,
are also thought to be a structure-specific endonuclease.
RAD1–RAD10 or XPF–ERCC1 form a heterodimeric complex
*To whom correspondence should be addressed. Tel: +81 471 24 1501; Fax: +81 471 23 9767; Email [email protected]
4971
Nucleic Acids
Acids Research,
Research,1994,
1997,Vol.
Vol.22,
25,No.
No.124
Nucleic
having structure-specific endonuclease activity with a polarity
opposite to XPG–RAD2 (27). The polarity of this cauliflower
structure-specific endonuclease suggests that it is in the FEN-1
family. We describe herein our findings and the first extensive
purification of a plant structure-specific endonuclease.
MATERIALS AND METHODS
Materials
The Bradford reagents and molecular weight markers for
SDS–PAGE were purchased from BioRad. AF-Blue Toyopearl
650ML was purchased from Tosoh. Sephadex G-25, Mono Q HR
5/5, Mono S HR 5/5, HiTrap desalting column and Superose 6
were from Pharmacia. Double-stranded DNA–Sepharose was
made in our laboratory according to the instruction manual for
HiTrap affinity columns (Pharmacia). DE52 was from Whatman
Biosystem Ltd. Centricon-10 was purchased from Amicon.
[γ-32P]ATP and [α-32P]ddATP (3000 Ci/mmol) were purchased
from Amersham. T4 polynucleotide kinase (PNK), calf thymus
terminal deoxynucleotidyl transferase (TdT) and bacterial alkaline
phosphatase (BAP) were obtained from TaKaRa. All other
reagents were from Sigma or Wako.
Oligonucleotides
The following oligonucleotides were used to construct various
DNA substrates.
Oligonucleotide (oligo) A, 5′-CAGCAACGCAAGCTTG-3′;
oligo B, 5′-GTCGACCTGCAGCCCAAGCTTGCGTTGCTG-3′;
oligo C, 5′-ATGTGGAAAATCTCTAGCAGGCTGCAGGTCGAC-3′;
oligo D, 5′-ACCGTCTTGAGGCAGAGT-3′;
oligo E, 5′-CACGTTGACTGAATC-3′;
oligo F, 5′-GGACTCTGCCTCAAGACGGTAGTCAACGTG-3′;
oligo G, 5′-CACGTTGACTACCGTCTTGAGGCAGAGTCC-3′;
oligo H, 5′-GCCAGCGCTCGGTTTTTTTTTTTTTTTTTTTTTTCCGAGCGCTGGC-3′;
oligo I, 5′-TAGCAGGCTGCAGGTCGAC-3′;
oligo J, 5′-AGGCTGCAGGTCGAG-3′.
Preparation of flap substrates and derivatives
The substrates used in this study were prepared according to the
methods described by Harrington et al. (28) with minor changes.
Radiolabeled oligonucleotide C was annealed to oligonucleotides A
and B to generate the 5′-flap structure. Radiolabeled oligonucleotide
C was annealed to oligonucleotide B to generate the 5′-pseudo-Y
structure. Radiolabeled oligonucleotide E was annealed to oligonucleotides D and F to generate the 3′-flap structure. Radiolabeled
oligonucleotide E was annealed to oligonucleotide F to generate the
3′-pseudo-Y structure. To make the stem–loop structure oligonucleotide H was labeled and annealed. Radiolabeled oligonucleotide F was used as single-stranded DNA. Radiolabeled
oligonucleotide F was annealed to oligonucleotide G to produce
double-stranded DNA. Radiolabeled oligonucleotide I or J was
annealed to oligonucleotides A and B to generate the 5′-flap
structure with a 5 or 1 nt strand. The annealing reactions were carried
out by mixing the oligonucleotides in an annealing buffer (20 mM
Tris–HCl, pH 7.4, 150 mM NaCl), followed by incubation at 95C
for 2.5 min, then slow cooling to room temperature.
4971
Structure-specific endonuclease assay
The structure-specific endonuclease activity was measured in a
20 µl reaction containing 50 mM Tris–HCl, pH 7.5, 1 mM MgCl2,
1 mM DTT and 0.5 pmol 32P-labeled DNA substrate
(10 000–50 000 c.p.m.). After incubation at 37C for 60 min the
reaction was stopped by adding 10 µl gel loading buffer
(90% formamide, 5 mM EDTA, 0.1% bromophenol blue,
0.1% xylene cyanol) and heated at 95C for 5 min. Then the
reaction was loaded onto a 20% polyacrylamide gel containing 7 M
urea in TBE buffer and electrophoresis was performed. The reaction
products were visualized by autoradiography or quantified using a
BAS2000 Bio imaging analyzer (Fuji). One unit of activity was
defined as the amount of enzyme required to cleave 1 pmol 5′-flap
DNA substrate in 30 min under standard structure-specific
endonuclease assay conditions.
Buffers
Buffer A contained 50 mM Tris–HCl, pH 7.5, 1 mM EDTA,
5 mM 2-mercaptoethanol and 15% glycerol. Buffer B was
identical to buffer A except that 0.01% NP-40 was added. Buffer
C was identical to buffer B except that the glycerol concentration
was decreased to 10%.
Enzyme purification
Purification of the cauliflower structure-specific endonuclease
was performed at 4C by the following operations. Unless stated
otherwise all buffers contained three different types of protease
inhibitor: 1 mM PMSF and 1 mg/ml each leupeptin and pepstatin.
Approximately 30 g fresh cauliflower inflorescence tissues were
suspended in 20 ml buffer A containing 0.4 M KCl after being
ground with a mortar and pestle. The suspension was fully disrupted
with a French press and sonication (20 kHz, 3 × 20 s), then
centrifuged at 30 000 g for 15 min. The supernatant was desalted
through Sephadex G-25 (Fraction I). Fraction I was loaded onto a
DEAE–cellulose column equilibrated with buffer A. The column
was washed with a sufficient amount of buffer A and fractions were
collected with a 200 ml linear gradient of 0–0.5 M KCl in the same
buffer. The structure-specific endonuclease activity was eluted as a
broad peak centered at 0.2 M KCl. The active fraction was pooled
and dialyzed against buffer B (Fraction II). Fraction II was loaded
onto an AF-Blue Toyopearl column equilibrated with buffer B. The
column was washed with 100 ml buffer B, then eluted using 130 ml
of a 0–0.5 M KCl gradient in buffer B. The structure-specific
endonuclease activity was eluted at 0.2 M KCl. The active fraction
was pooled and dialyzed against buffer C (Fraction III). Fraction III
was loaded on a Mono Q HR 5/5 column equilibrated with buffer
C and then the column was washed with 10 ml buffer C. Fractions
were collected with 30 ml of a 0–0.4 M NaCl gradient in buffer C.
The active fraction eluted at 0.1 M NaCl was collected and desalted
on a HiTrap desalting column equilibrated with buffer C (Fraction
IV). Fraction IV was loaded on a Mono S HR 5/5 column
equilibrated with buffer C. The column was washed with 10 ml
buffer C, then eluted with a 20 ml linear gradient of 0–0.5 M NaCl
in the same buffer. The structure-specific endonuclease activity was
eluted at 0.25 M NaCl as a tight peak. The active fraction was
collected and desalted on a HiTrap desalting column equilibrated
with buffer C (Fraction V). A final purification step was achieved
using dsDNA–Sepharose column chromatography and the
structure-specific endonuclease activity was eluted at 0.2 M NaCl
4972 Nucleic Acids Research, 1997, Vol. 25, No. 24
Figure 1. DNA substrates used. Oligonucleotides that carry the labels are
underlined. The asterisk indicates the position of the 32P label.
in 10 ml of a 0–0.5 M NaCl gradient in buffer C. The combined
active fraction was dialyzed against buffer C (Fraction IV) and
used in the subsequent experiments.
Figure 2. Molecular weight determination of the enzyme. (A) SDS–PAGE
analysis of purified cauliflower structure-specific endonuclease. The concentration
of the SDS–polyacrylamide gel was 10% and the gel was stained with silver nitrate.
Standard marker proteins are indicated by arrows to the left of the panel (in kDa).
(B) Gel filtration through a Superose 6 column by FPLC. Fraction VI was
concentrated using Centricon-10, then loaded onto the column. The arrow
indicates the position at which the structure-specific endonuclease activity was
found. The molecular weight makers were β-amylase (200 kDa), alcohol
dehydrogenase (150 kDa), BSA (67 kDa), carbonic anhydrase (29 kDa) and
cytochrome c (12.4 kDa).
Other methods
SDS–PAGE was performed with a 10% gel using the standard
Laemmli methods (29). After electrophoresis the gel was silver
stained as described by Wray et al. (30). The protein concentration
was determined using the BioRad protein assay with γ-globulin as
the standard.
RESULTS AND DISCUSSION
Purification of cauliflower structure-specific endonuclease
The structure-specific endonuclease was purified from cauliflower
inflorescence as described in Materials and Methods. FEN-1 was
found using the 5′-flap structure as substrate (11). We also used it
in the first place to assay the column fractions. Purification was by
column chromatography using DEAE–cellulose, AF-Blue Toyopearl, Mono Q HR 5/5, Mono S HR 5/5 and finally a dsDNA–
Sepharose (HiTrap affinity column) column. The chromatographic
behavior of the cauliflower structure-specific endonuclease was
monitored using the 5′-flap structure (Fig. 1) as substrate under
standard assay conditions. The results of purification are
summarized in Table 1.The structure-specific endonuclease activity
before Mono Q column chromatography could not be measured
precisely.
The active fractions from the dsDNA–Sepharose column were
combined, temporarily designated Fraction VI and then analyzed by
SDS–PAGE. As shown in Figure 2A, Fraction VI was found to be
composed of a single polypeptide with a molecular mass of 40 kDa
and the enzyme was sufficiently pure as judged by the results of
SDS–PAGE analysis. To determine the native molecular mass of the
structure-specific endonuclease Fraction VI was concentrated using
Centricon-10 and then loaded onto a Superose 6 gel filtration
column for FPLC. The cauliflower structure-specific endonuclease
was eluted at a position representing a calculated molecular mass of
40 kDa (Fig. 2B), indicating that the enzyme is a 40 kDa
monopolypeptide.
Fraction VI, as the purified cauliflower structure-specific
endonuclease, was used in subsequent experiments. The enzyme
in Fraction VI has been kept for over a month at –80C without
apparent loss of activity.
Table 1. Purification of cauliflower structure-specific endonuclease
Fraction
Total
volume
(ml)
Total
protein
(mg)
Total
activity
(U)
Specific
activity
(U/mg)
Yield
(%)
I. Crude extract
II. DEAE–cellulose
III. AF-Blue
Toyopearl
50
35
20
600
120
14
NDa
ND
ND
ND
ND
ND
–
–
–
IV. Mono Q
V. Mono S
VI. ds
DNA–Sepharose
3.0
2.0
1.0
0.2
0.064
0.019
378
155
104
1890
2422
5477
100
41
27.5
aND,
not determined.
4973
Nucleic Acids
Acids Research,
Research,1994,
1997,Vol.
Vol.22,
25,No.
No.124
Nucleic
Cauliflower structure-specific endonuclease cleaves 5′-flap
DNA structure
In accordance with the methods described by Harrington et al. (11),
we hybridized 5′-32P-labeled oligonucleotide C or 3′-32P-labeled
oligonucleotide C to the non-radiolabeled oligonucleotides A and B
(Figs 1 and 3A). This structure, called the 5′-flap structure, contains
19 nt of the 5′ overhanging single-stranded sequence (Fig. 1). We
similarly hybridized 5′- or 3′-radiolabeled oligonucleotide E to
non-radiolabeled oligonucleotides D and F to form the 3′-flap
structure, which contains a 3′ overhanging single-stranded sequence
(Fig. 1). The 5′-flap structure terminates with a 5′ single-stranded
end and, conversely, the 3′-flap structure has a flap strand that
terminates with a 3′ single-stranded end.
To confirm whether the cauliflower structure-specific endonuclease mediates cleavage of the flap structures mixtures of the
cauliflower structure-specific endonuclease (0.5 U) and 5′- or 3′-flap
structure (0.5 pmol) were incubated at 37C and then heated for 3
min at 95C in the presence of formamide to inactivate the enzyme
and to separate the hybridized DNA strands. The mixture was run,
along with radiolabeled DNA size markers, on a 20% polyacrylamide gel containing 7 M urea to prevent rehybridization of the
DNA strands during electrophoresis. The gel was subjected to
autoradiography to reveal the radiolabeled DNA species. The
presence of DNA size markers in the polyacrylamide gel enabled us
to determine the size of the cleavage products precisely. As shown
in Figure 3, treatment of the 5′-flap structure of the 5′-32P-labeled
oligonucleotide or the 3′-32P-labeled oligonucleotide with the
cauliflower structure-specific endonuclease resulted in formation of
4973
one major and two minor products (Fig. 3A). As the major product
of the 5′-flap structure with the 5′-32P-labeled oligonucleotide the
19 nt single-stranded product was the result of a cut at the junction
of the duplex DNA and the single-stranded DNA of the 5′-flap
structure, and the two minor products were 18 and 20 nt
single-stranded oligonucleotides which were cut on either side of the
19 nt single-stranded product (lanes 1–6 in Fig. 3A). When the
5′-flap structure with the 3′-32P-labeled oligonucleotide was used
the major and minor products also appeared as one and two bands
respectively (lanes 7–12 in Fig. 3A). The major product was 15 nt
from the 3′-end of oligonucleotide C, also the result of a cut at the
junction in the flap structure, and the minor products were 1 nt each
side of the 15 nt product from the 3′-end. One 32P-labeled
nucleiotide was added to the 3′-end using TdT and [α-32P]ddATP.
From the sizes of the cleavage products it could be concluded that
the major site of cleavage is located at the single strand–duplex
junction and that the minor sites are 1 base inside the duplex region
and 1 base into the single-stranded region. To determine whether
flap strand length affects the structure-specific endonuclease activity
or not we varied the length of the flap strand of the 5′-flap structure
(Fig. 3B). We found that the cauliflower structure-specific
endonuclease was capable of cleaving both 1 and 5 nt flaps
efficiently (Fig. 3B, lanes 1–6). This result demonstrates that
cleavage is independent of flap strand length. We were unable to
detect any endonucleolytic cleavage of the 3′-flap structure (lanes
1–3 in Fig. 5) and, as described later in this report, no single-stranded
oligonucleotide-dependent cleavage by the cauliflower structurespecific endonuclease was shown (Fig. 6).
Figure 3. Cauliflower structure-specific endonuclease cleaves 5′-flap structures. (A) Cleavage of 5′-flap structure by cauliflower structure-specific endonuclease.
5′-Flap structures with a 5′-32P-labeled oligonucleotide (lanes 1–6) and 5′-flap structures with a 3′-32P-labeled oligonucleotide (lanes 7–12) were incubated with 0.5 U
cauliflower structure-specific endonuclease for 0 (lanes 1 and 7), 5 (lanes 2 and 8), 10 (lanes 3 and 9), 30 (lanes 4 and 10) or 60 min (lanes 5 and 11) and 60 min in
the presence of 1 mM ATP (lanes 6 and 12). Asterisks indicate the position of the 32P label. The radiolabeled cleavage products were revealed by autoradiography
after electrophoresis of the reaction mixtures. The numbers indicate the position and size of the oligonucleotide standards. (B) Effect of flap strand length on cleavage.
5′-Flap structures with a 19 (lanes 1 and 2), 5 (lanes 3 and 4) or 1 nt strand (lanes 5 and 6) were incubated with 0.5 U cauliflower structure-specific endonuclease for
0 (lanes 1, 3 and 5) or 10 min (lanes 2, 4 and 6).
4974 Nucleic Acids Research, 1997, Vol. 25, No. 24
Table 2. Biochemical properties of cauliflower structure-specific endonuclease
Assay conditions
Complete reaction
No protein control
+ 0.1% SDS
DTT
Mg2+
Mg2+, + 1 mM Mn2+
Mg2+, + 1 mM Ca2+
+ 1 mM ATP
Activity
1.0
<0.01
<0.01
0.5
0.3
0.7
0.36
0.85
5′-Flap DNA (0.5 pmol) was incubated for 10 min at 37C with 0.5 U cauliflower
structure-specific endonuclease and 1 mM MgCl2 in the complete reaction.
Conditions deviating from the complete reaction are indicated in the left hand
column. The right hand column displays the relative reaction rate in comparison with
the complete reaction
Replication-related FEN-1 cleavage and cutting by the NERrelated endonucleases, such as RAD2–XPG and RAD1–RAD10/
XPF–ERCC1, are flap strand-specific and independent of flap
strand length (11,12), suggesting that the tertiary structure of
DNA is universally important for site recognition and flap strand
length may be irrelevant to function. The cauliflower structurespecific endonuclease appears to be a plant alternative in this
category of proteins.
Biochemical properties of the enzyme
The 5′-flap structure-dependent biochemical properties of the
cauliflower structure-specific endonuclease are shown in Table 2.
Mg2+ ions were quite important for the reaction, but not essential
(Table 2). Compared with the complete reaction, activity without
Mg2+ ions was reduced to 30%. Ca2+ ions were unable to support
activity, but Mn2+ ions could support moderate activity. The
cleavage pattern in the presence of Mn2+ ions was the same as that
in the presence of Mg2+ ions (data not shown). In the presence of
1 mM ATP (lanes 6 and 12 in Fig. 3) the structure-specific
endonuclease activity of the enzyme was reduced to 85%
compared with the complete reaction (Table 2). Since ATP can
chelate divalent metal ions, activity might be reduced. The
enzyme is thus ATP independent.
Cauliflower structure-specific endonuclease has 5′→3′
exonuclease activity on double-stranded DNA
As shown in Figure 4A, the cauliflower structure-specific endonuclease could not cleave single-stranded DNA, indicating that it has
no detectable single-stranded DNA-dependent endonuclease or
exonuclease activity. However, the cauliflower structure-specific
endonuclease when incubated with double-stranded DNA produced
a monomer band of the 32P-labeled 5′-end nucleotide (Fig. 4B).
This indicates that the cauliflower structure-specific endonuclease has 5′→3′ double-stranded DNA-specific exonuclease
activity and this function is also known in FEN-1 (11).
To confirm the specificity of the enzyme for the flap DNA
structure a competitor analysis was performed. The structurespecific endonuclease activity was measured in a 20 µl reaction with
0.5 U cauliflower structure-specific endonuclease, 0.5 pmol
32P-labeled 5′-flap DNA as the substrate and 0.5 or 1.5 pmol
competitor. As shown in Figure 4C, single-stranded DNA and
double-stranded DNA did not compete with the activity at all, but
Figure 4. Specificity of cauliflower structure-specific endonuclease for flap
structures. (A) Single-stranded DNA was incubated with 0.5 U enzyme for 0,
5, 10, 30 or 60 min (lanes 1–5 respectively). (B) Double-stranded DNA was
incubated with 0.5 U enzyme for 0, 5, 10, 30 or 60 min (lanes 1–5 respectively).
The radiolabeled cleavage products were revealed by autoradiography after
electrophoresis of the reaction mixtures. (C) Competitor analysis. The
structure-specific endonuclease activity of the enzyme was measured with
0.5 pmol radiolabeled 5′-flap structure. Single-stranded DNA, double-stranded
DNA and 5′-flap structure served as competitors at concentrations of 0.5 (left
columns) and 1.5 pmol (right columns) respectively.
the non-labeled 5′-flap DNA competed efficiently. This result
indicates that the cauliflower structure-specific endonuclease shows
a preference for 5′-flap DNA much more than double-stranded
DNA as the substrate, although it is a 5′→3′ double-stranded
DNA-specific exonuclease.
Cauliflower structure-specific endonuclease cleaves
5′-pseudo-Y and stem–loop structures
XPG–RAD2 and FEN-1 efficiently cleave the 5′ overhanging
single-strand tail in the flap structures at and around the single
strand–duplex junction, but FEN-1 cannot cleave 5′- or 3′-pseudo-Y
structures (see Fig. 1). XPG–RAD2 has been shown to recognize
and cleave such a pseudo-Y structure. To test this we also made
pseudo-Y DNA structures. When the 5′-32P-radiolabeled
oligonucleotide C was hybridized to the partially homologous
non-radiolabeled oligonucleotide B it gave the 5′-pseudo-Y
structure (Fig. 1). The 5′-pseudo-Y structure has a 5′ overhanging
single-stranded tail in its structure. In addition to the 5′-pseudo-Y
structure we hybridized the 5′-radiolabeled oligonucleotide E to
4975
Nucleic Acids
Acids Research,
Research,1994,
1997,Vol.
Vol.22,
25,No.
No.124
Nucleic
4975
Figure 5. Ability of cauliflower structure-specific endonuclease to cleave
various substrates. The 5′-pseudo-Y (lanes 1–3), 3′-flap (lanes 4–6),
3′-pseudo-Y (lanes 7–9) and stem–loop (lanes 10 and 11) structures were
incubated with 0.5 U enzyme for 0 (lanes 1, 4, 7 and 10), 10 (lanes 2, 5, 8 and
11) or 30 min (lanes 3, 6 and 9). The radiolabeled cleavage products were
revealed by autoradiography after electrophoresis of the reaction mixtures.
the non-labeled oligonucleotide F to form the 3′-pseudo-Y
structure, which contains a 3′ overhanging single-stranded tail
(Fig. 1). The two pseudo-Y structures were incubated with the
cauliflower structure-specific endonuclease under the reaction
conditions employed for cleavage of the flap structures. As shown
in Figure 5, the cauliflower structure-specific endonuclease was
able to cleave the 5′-pseudo-Y structure (lanes 1–3 in Fig. 5),
although not efficiently. The enzyme could not cleave the 3′-flap or
3′-pseudo-Y structures (lanes 4–9 in Fig. 5). Using the 3′-pseudo-Y
structure no detectable cleavage of the 3′ overhanging single-strand
was observed, even in the presence of excess cauliflower structurespecific endonuclease (data not shown). Thus the inability of the
cauliflower structure-specific endonuclease to cleave the 3′-flap and
3′-pseudo-Y structures further confirms its specificity for the 5′-flap
structure.
XPF–ERCC1 and RAD1–RAD10 complexes are known to be
structure-specific endonucleases responsible for 5′ incision
during repair and to cut a stem–loop DNA structure in vitro
consisting of a 22 nt single-stranded loop and a duplex stem of
12 bp, shown in Figure 1 (27). The XPF–ERCC1 and
RAD1–RAD10 complexes cleave 2, 3 and 4 phosphodiester
bonds away from the 5′-side of the duplex DNA in the stem–loop
structure (27). Conversely, XPG cleaves the structure on the
3′-side, at the phosphodiester bond on the stem–loop border and
one bond into the stem (27). To determine whether the cauliflower
structure-specific endonuclease can cleave the stem–loop structure
we also used a partially self-complementary oligonucleotide to
form the stem–loop structure shown in Figure 1. The substrate was
end-labeled at the 5′-terminus and the reaction products analyzed by
comparison with DNA size markers to map the exact sites of
cleavage. The cauliflower structure-specific endonuclease
Figure 6. Analysis of DNA ends generated by cauliflower structure-specific
endonuclease. (A) Experimental scheme. Non-labeled 5′-flap structure was
incubated with the enzyme under standard conditions. The mixture of
oligonucleotides after cleavage was incubated with or without BAP. The BAP
was removed by phenol extraction. The 3′-ends of the oligonucleotides were
labeled with [α-32P]ddATP and calf thymus terminal deoxytransferase or the
5′-ends of the oligonucleotides were labeled with [γ-32P]ATP and T4
polynucleotide kinase. The products were analyzed by 20% PAGE and
autoradiography. (B) The 3′-end-labeled 19mer products and the 5′-end-labeled
14mer product were visualized.
specifically cleaved this substrate on the 3′-side of the singlestranded region only at the phosphodiester bond on the stem–loop
border, indicating that the cauliflower structure-specific endonuclease makes the 3′ nick without further processing (lanes 10 and
11 in Fig. 5). The ability of the enzyme to cleave the stem–loop
structure at this special point indicates its similarity to XPG, although
the cutting site of the cauliflower structure-specific endonuclease
was only a single site (lanes 10 and 11 in Fig. 5).
Analysis of DNA ends generated by the enzyme
To determine the cut ends of the oligonucleotide generated by the
cauliflower structure-specific endonuclease we analyzed it by the
following methods (Fig. 6A). The non-labeled 5′-flap structure,
4976 Nucleic Acids Research, 1997, Vol. 25, No. 24
vegetable used here are bombarded by severe sunlight containing
UV for much longer periods than are animals or yeast. They cannot
avoid this bombardment or escape from it. This plant enzyme may
therefore help to elucidate the biochemistry of the DNA repair
system for UV-damaged DNA. Using the purified cauliflower
structure-specific endonuclease we have tried to clone the gene and
expect to do so soon, along with monoclonal antibodies. In this
report we describe the purification and the structure-specific
properties of this new plant structure-specific endonuclease as a
first step in a series of plant UV lesion repair studies.
ACKNOWLEDGEMENTS
We thank Mr Hirokazu Seto of our Department for his technical
advice concerning cauliflower cultivation. This work was
supported in part by a Grant-in-Aid (no. 362-0163-08277223)
from the Ministry of Education, Science and Culture (Japan).
Figure 7. Summary of cleavage sites. The arrows indicate the sites of cleavage.
which is composed of oligonucleotides A–C, was incubated with
the enzyme under standard conditions. Oligonucleotide C (33 nt)
could be separated into two fragments, 19mer of the 5′-end side
and 14mer of the 3′-end side, because the cleavage site is at the
junction of the single-stranded and duplex DNA. Next, the
mixture of oligonucleotides A–C after cleavage was incubated
with or without bacterial alkaline phosphatase (BAP), which can
remove the phosphate from both the 5′- and 3′-end of the the DNA
strand. Then the BAP was removed by phenol extraction. The
3′-ends of the oligonucleotides were then labeled with
[α-32P]ddATP and calf thymus terminal deoxytransferase or the
5′-ends of the oligonucleotides were labeled with [γ-32P]ATP and
T4 polynucleotide kinase. The products were analyzed by
20% PAGE and autoradiography. The 3′-end-labeled 19mer and
the 5′-end-labeled 14mer could be visualized (Fig. 6B). If the
phosphate is not present at the termini of the oligonucleotides
only the termini will be labeled by 32P. As shown in Figure 6B,
the 19mer was labeled by 32P with or without BAP treatment,
suggesting that the 3′-end generated by the enzyme has no
phosphate. In contrast, the 14mer was labeled only when the BAP
treatment was performed, suggesting that the 5′-end generated by
the enzyme has a phosphate. These results indicate that the
cauliflower structure-specific endonuclease leaves 5′-PO4 and
3′-OH on the 5′-flap DNA.
Based on the data of this study, the cleavage sites on various
substrates of the cauliflower structure-specific endonuclease are
summarized in Figure 7. The cauliflower structure-specific
endonuclease could not cleave either the 3′-flap or 3′-pseudo-Y
structures but could cleave the 5′-flap and 5′-pseudo-Y structures,
indicating that the cauliflower structure-specific endonuclease
acts on the 5′ overhanging single-stranded tail in both the 5′-flap
and 5′-pseudo-Y structures but does not act on the 3′ overhanging
region in the 3′-flap or 3′-pseudo-Y structures. The cauliflower
structure-specific endonuclease can also cleave the stem–loop
structure on the 3′-side of the junction of unpaired and duplex
DNA. The cut ends of the oligonucleotides generated by the
cauliflower structure-specific endonuclease are the 5′-PO4 and
3′-OH termini.
This study represents the first report and the first extensive
purification of a structure-specific endonuclease in higher plants.
Although it has not often been pointed out, higher plants like the
REFERENCES
1 Sakaguchi,K. and Boyd,J.B. (1985) J. Biol. Chem., 260, 10406–10411.
2 Lu,B.C. and Sakaguchi,K. (1991) J. Biol. Chem., 266, 21060–21066.
3 Matsuda,S., Takami,K., Sono,A. and Sakaguchi,K. (1993) Chromosoma,
102, 631–636.
4 Takami,K., Matsuda,S., Sono,A. and Sakaguchi,K. (1994) Biochem. J., 299,
335–340.
5 Gomi,K. and Sakaguchi,K. (1994) Biochem. Biophys. Res. Commun., 198,
1232–1239.
6 Aoyagi,N., Matsuoka,S., Furunobu,A., Matsukage,A. and Sakaguchi,K.
(1994) J. Biol. Chem., 169, 6045–6050.
7 Kai,M., Takahashi,T., Todo,T. and Sakaguchi,K. (1995) Nucleic Acids Res.,
23, 2600–2607.
8 Matsuda,S., Sakaguchi,K., Tsukada,K. and Teraoka,H. (1996) Eur.
J. Biochem., 237, 691–697.
9 Aoyagi,N., Oshige.M., Hirose,F., Kuroda,K., Matsukage,A. and
Sakaguchi,K. (1997) Biochem. Biophys. Res. Commun., 230, 297–301.
10 Sawado,T. and Sakaguchi,K. (1997) Biochem. Biophys. Res. Commun., 232,
454–460.
11 Harrington,J.J. and Lieber,M.R. (1994) EMBO J., 13, 1235–1246.
12 Habraken,Y., Sung,P., Parakash,L. and Parakash,S. (1995) J. Biol. Chem., 270,
30194–30198.
13 Ishimi,Y., Claude,A., Bullock,P. and Hurwitz,J. (1988) J. Biol. Chem., 263,
19723–19733.
14 Turchi,J.J., Huang,L., Murante,R.S., Kim,Y. and Bambara,R.A. (1994)
Proc. Natl. Acad. Sci. USA, 91, 9803–9807.
15 Jacquier,A., Legrain,P. and Dujon,B. (1992) Yeast, 8, 121–32.
16 Sommers,C.H., Miller,E.J., Dujon,B., Prakash,S. and Prakash,L. (1995)
J. Biol. Chem., 370, 4193–4196.
17 Murray,J.M., Tavassoli,M., Al-harithy,R., Sheldrick,K.S., Lehmann,A.R.,
Carr,A.M. and Watts,F.Z. (1994) Mol. Cell. Biol., 14, 4878–4888.
18 MacInnes,M.A., Dickson,J.A., Hernendez,R.R., Learmonth,D., Lin,G.Y.,
Mudgett,J.S., Park,M.S., Shauer,S., Reynolds,R.J., Strniste,G.F. and Yu,J.Y.
(1993) Mol. Cell. Biol., 13, 6393–6402.
19 Scherly,D., Nouspikel,T., Corlet,J., Ucla,C., Bairoch,A. and Clarkson,S.G.
(1993) Nature, 363, 182–185.
20 Matsunaga,T., Mu,D., Park,C.-H., Reardon,J.T. and Sancar,A. (1995)
J. Biol. Chem., 270, 20862–20869.
21 Yao,M. and Kow,Y.W. (1996) J. Biol. Chem., 271, 30672–30676.
22 Staplenton,A.E., Thornber,C.S. and Walbot,V. (1997) Plant Cell Environ.,
20, 279–290.
23 Medford,J.I., Elmer,J.S. and Klee,H.J. (1991) Plant Cell, 3, 359–370.
24 Fukata,H. and Fukasawa,H. (1982) J. Biochem., 91, 1337–1342.
25 David,R.S. (1995) Curr. Biol., 5, 361–363.
26 Meyerowiz,E.M. (1987) Annu. Rev. Genet., 21, 93–111.
27 Sijber,M.S., de Laat,W.L., Ariza.,R.A., Biggerstaff,M. Wei,Y.-F., Moggs,J.G.,
Carter,K.C., Shell,B.K., Evans,E., de Jong,M.C., Rademakers,S., de Rooij,J.,
Jaspers,N.G.J., Hoeijmakers,J.H.J. and Wood,R.D. (1996) Cell, 86, 811–822.
28 Harrington,J.J. and Lieber,M.R. (1994) Genes Dev., 8, 1344–1355.
29 Laemmli,U.K. (1970) Nature, 227, 680–685.
30 Wray,W., Boulikas,T., Wray,V.P. and Hancock,R. (1981) Anal. Biochem.,
118, 197–203.