Nucleic Acids Research, Vol. 19, No. 2 257
Sequence-specific cleavage of double-stranded DNA
caused by X-ray ionization of the platinum atom in the
Pt-Bis-netropsin—DNA complex
S.L.Grokhovsky* and V.E.Zubarev1
VAEngelhardt Institute of Molecular Biology, The USSR Academy of Sciences, Moscow 117984 and
1
M.V.Lomonossov Moscow State University, Chemical Department, Moscow 117234, USSR
Received November 9, 1990; Accepted December 11, 1990
ABSTRACT
An analog of the antibiotic netropsin containing two
netropsin-like fragments linked covalently via a
platinum atom has been synthesized. DNase I and
hydroxyl radical footprlnting studies have shown that
this compound binds at selective sites on a DNA
restriction fragment with a known nucleotide sequence.
After X-ray Irradiation of Pt-bis-netropsln—DNA
complexes a platinum-mediated cleavage of DNA is
observed at specific DNA sites. This enables one to
determine the location of the synthetic llgand on the
DNA with a precision of about one nucleotide. The
cleavage activity seems to be related to the emission
of Auger electrons from the platinum atom that cause
rupture of the deoxyribose residues on the two DNA
strands near the position of the platinum atom in the
complex.
INTRODUCTION
There is much interest in the design and synthesis of chemical
compounds capable of interacting specifically with defined
nucleotide sequences in DNA [1,2]. These compounds can
penetrate through cellular membranes and might affect
specifically the activity of certain genes in bacterial and eukaryotic
cells and serve as effective antiviral and antitumor agents. In this
regard the development of new methods for determining the
preferred binding sites of ligands on the surface of the DNA
double helix is of great importance. One example is the classical
footprinting method that involves cleavage of the ligand complex
with end-labeled DNA restriction fragments with a DNA cleaving
agent and subsequent analysis of the cleavage products by means
of high resolution gel electrophoresis [ 3 - 5 ] . In addition to
different variants of the footprinting technique, Dervan et al. [6]
has proposed an affinity cleaving method. They have introduced
a DNA cleaving group into a synthetic sequence-specific ligand,
thereby converting a sequence-specific DNA binding molecule
to a bifunctional molecule capable of binding and cleaving DNA
at the recognition site. Dervan et al. have synthesized analogs
of the AT-specific antibiotic netropsin [7] tethered to the iron
* To whom correspondence should be addressed
chelator, EDTA [6]. These analogs bind noncovalently to doublestranded DNA. Like the naturally occurring antibiotic bleomycin,
these compounds can cleave phosphodiester bonds at or near the
site of ligand binding.
Another approach to the design of synthetic DNA cleaving
molecules involves attachment of electrophiles or photosensitive
groups to a sequence-specific DNA binding molecule. However,
for estimation of the sequence specificity shown by a DNA
binding molecule these methods are not preferred over the
classical footprinting technique due to a strong selectivity of the
chemical alkylating reaction [7 — 10].
In order to determine the location of the preferred ligand
binding sites on DNA a more helpful approach seems to be a
technique which makes use of a physical process that may cause
rupture of any bond in the DNA. An obvious example is the
approach developed by Martin and Holmes [11] who found that
introduction of radioactive ]25l into the sequence-specific dye
Hoechst 33258 converts the dye molecule to a sequence-specific
DNA-cleaving agent. The cleavage activity arises from emission
of very low energy Auger electrons due to the decay of the
radioactive iodine. The decay was accompanied by multiple cuts
of phosphodiester bonds on the two DNA strands.
The idea of the present approach to synthesis of sequencespecific DNA-cleaving agents is based on the known capacity
of various atoms to adsorb X-rays with markedly different
efficiencies. We have synthesized a sequence-specific DNAbinding analog of the antibiotic netropsin in which two netropsinlike fragments are covalently linked via the cis-diammineplatinum(II) residue [12,13]. This synthetic ligand will be referred
to as Pt-bis-netropsin. After irradiation of the Pt-bis-netropsin—
DNA complex with X-rays selective cutting of DNA is observed
at the preferred ligand binding sites.
MATERIALS AND METHODS
End labelled DNA fragments were obtained by cleaving pUC-9
plasmid with Sau 96 I ('BRL') or Dde I ('Boehringer Mannheim')
with subsequent separation of DNA fragments in 5%
polyacrylamide gels. In order to introduce a radioactive label at
258 Nucleic Acids Research, Vol. 19, No. 2
one of the fragment's 3'-end, the appropriate [a-32P]NTP
('Isotop', Tashkent) and Klenow fragment of Escherichia coli
DNA polymerase ('Boehringer Mannheim') were used [14]. The
A and B fragments (166 and 616 base pairs respectively) contain
nucleotides: 1484—1650 (fragment A) and nucleotides
2053-2669 (fragment B). In the plasmid pBR-322 these
sequences exist between nucleotides 3159 — 3325 and
3728-4344, respectively.
Preparation of the Complex. A 5 fi\ portion of a solution of the
DNA labelled fragment (3,000-10,000 cpm) and unlabeled
DNA pUC-9 at a concentration of 4 10~5M (base pair) in
10 mM Tris-HCl (pH 7.5), 0.25 M NaCl, 10% glycerol directly
before irradiation was mixed with 5 /tl Pt-bis-netropsin [12]
solution at concentration 1) 0 ; 2) 1 • 10"5M; 3) 3.3- 10~6M and
4) l.l-10~ 6 M in the same buffer in 0.5 ml 'Eppendorf
polypropylene tubes.
X-Ray irradiation. For irradiation a 5BXV6 X-ray lamp (USSR,
tungsten anode , U = 35 KeV, 1=80 mA, dose capacity = 15
KRad/min) was used. The tungsten polychromatic X-ray
irradiation is represented by quanta with energies ranging from
a few KeV to 35 KeV with a maximum at 22.5 KeV. The
irradiation was carried out at 10°C for 2 min in polypropylene
tubes with approximately 1 mm thick walls. Under these condition
practically complete filtration of soft X-ray irradiation (with
energy less than 10 KeV) take place.
After irradiation 90 /d of a solution containing 50 mM TrisHCl buffer (pH 7.5), 0.15 M NaCl, 10 mM EDTA and 10 ^g/ml
tRNA were added. The DNA was purified by extraction with
phenol, precipitated with ethanol, washed with 70% ethanol,
dried, dissolved in 1.5 /tl of formamide-dye mixture, heated
1 min at 90°C and applied to an 8% denaturing polyacrylamide
gel 40 cm long with a gradient thickness of 0.15-0.45 mm [15].
Electrophoresis proceeded for 75 min at 45 WT (2000 V). Before
exposure the gel was fixed with 10% acetic acid and dried on
glass pretreated with 'Bind-silane' ('LKB'). Cleavage products
were identified by comparison with an A + G Maxam-Gilbert
sequencing ladder [4].
DNase I footprinting was carried out according to a standard
procedure [3] with the above-listed concentrations of the ligand
under conditions identical to those used for experiments with Xray irradiation.
Hydroxyl radical footprinting was carried out as reported in ref.
[16].
Autoradiographs were scanned with a LKB 2222-010 Ultroscan
XL laser densitometer using an aperture width of 0.05 mm with
collection of data points every 0.04 mm interval of lane.
RESULTS
Fig. 1 shows chemical formulae of netropsin and Pt-bis-netropsin
that contains the two netropsin-like fragments linked covalently
via the cis-diammineplatinum(II) residue. In the Pt-bis-netropsin
molecule the dimethylamino group was substituted for the
netropsin amidine group and the netropsin guanidylacetic acid
residue replaced by a glycine.
In Fig.2 the DNase I and hydroxyl radical cleavage protection
profiles [16] and cleavage profiles obtained after X-ray irradiation
with and without 1 M glycerol generated by Pt-bis-netropsin on
:OKH(CH2
CH,
-2HC1
CH,
(I )
NH2CH2COHH
H,N
Pt
2+
» 2H0l
jN< C H ,
Pr
HjN
) 3 N(CHj) 2 »HNO 3
Pr
Pr
(II )
Fig.l. Chemical formulae of netropsin (I) and Pt-bis-netropsin (II).
a DNA restriction fragment A (166 nucleotides) are presented.
In the left and right parts of the picture the cleavage protection
patterns generated by Pt-bis-netropsin on the two DNA chains
are shown. The concentrations of Pt-bis-netropsin were about
one molecule per 4, 12 and 36 base pairs, respectively, in the
presence of 0.25 M NaCl.
The densitometer tracing (Fig.3a and 3b) enables one to
measure the extent of protection from DNase I and hydroxyl
radical cleavage of DNA caused by Pt-bis-netropsin at the central
part of the DNA fragment.
Comparing cleavage protection profiles of Pt-bis-netropsin
complexes obtained with DNase I and hydroxyl radicals, one can
conclude that Pt-bis-netropsin protects a number of AT-rich sites
on the DNA. The protected regions are indicated in Fig.4. Upon
X-ray irradiation of the free DNA fragment in glycerol-free
solution the DNA is randomly cleaved in close similarity with
the cleavage pattern of the same fragment generated by hydroxyl
radicals in the EDTA-Fe2+/H2O2/ascorbate reagent system.
After irradiation of the Pt-bis-netropsin—DNA complexes the
following effects were observed: i) a decrease of the rate of nonspecific DNA cleavage associated with the strong absorption of
X-ray irradiation by platinum atoms, ii) a small protection from
cleavage of the sites on the DNA fragment to which Pt-bisnetropsin binds and iii) appearance of new cutting sites near the
center of the Pt-bis-netropsin binding sites (Fig 3c).
Upon X-ray irradiation of the free DNA fragment and its
complexes with Pt-bis-netropsin in the presence of 1 M glycerol,
a strong decrease in the non-specific cleavage is observed.
However, the cleavage rate at the center of the Pt-bis-netropsin
binding site is either not changed or even increased (Fig. 2, 3c
and 3d).
In the nucleotide sequence of the DNA fragment shown in
Fig.4 the cleavage sites observed at low bis-netropsin
concentrations (less than one molecule per 36 base pairs) are
indicated with long arrows. Additional cleavage sites seen at
higher ligand concentration are shown by short arrows. One can
see that each cutting site on the upper DNA strand is correlated
with the appropriate cutting site on the lower DNA strand with
Nucleic Acids Research, Vol. 19, No. 2 259
ONus I
A
1 + 2
4
C
3
5
EDTA/Fo2+
A
1 + 2
4
C
3
5
X-R»y»
X-Rays
+ glycerol
A
A
1 + 2
4
1 + 2
4
C 3
5
C 3
5
1590
B
/-Rays
1640
•
1630
•
1620
•
1610
•
1600
-
1590
-
1580
•
1560
-
1550
1600
1540
•
1530
•>
1610
1620
-
1520
1630
FIg.2. Gel radioautograms. The figure shows the action of different DNA-cleaving agents on Pt-bis-netropsin complexes with the fragment in which the top strand
(A) or the bottom strand (B) of the DNA was radiolabelled at the 3'-end. Lane (1): Untreated fragments, Lane 'A+G': A + G sequence reaction, Lane (2): the
treatment of free fragment, Lanes (3), (4), (5): treatment of Pt-bis-netropsin complexes with the fragment at molar ratio of Pt-bis-netropsin to DNA base pairs of
1/4, 1/12 and 1/36, respectively.
a shift by 3 nucleotides in the 3'-direction. The cutting sites occur
in the DNA regions which are protected by Pt-bis-netropsin
from DNase I and hydroxyl radicals in the EDTAFe 2+ /H 2 (V a sc°rbate system.
Fig.5 shows the cleavage pattern obtained after X-ray
irradiation of naked DNA fragment B (616 base pair) and of its
complexes with Pt-bis-netropsin in the presence of 1 M glycerol.
In Fig.6 the observed cleavage sites on this DNA fragment are
indicated.
DISCUSSION
Upon radiolysis of dilute aqueous solutions the major active
particles capable of initiating disruption of DNA chains appear
to be H • atoms, hydrated electrons and OH radicals which are
formed with yields 0.55, 2.7 and 2.7 unit/100 eV, respectively
[17]. As established at present [18], the disruption of DNA upon
radiolysis occurs via formation of intermediate radical centers
which are formed upon elimination of hydrogen from ribose
residues or as a result of attachment of radicals to the double
bonds of DNA bases. In a air-saturated aqueous solution the
concentration of oxygen is about 2.5-10~ 4 mole/1. Under these
conditions the H • atoms and hydrated electrons are completely
transferred to Oj —radicals. The disruption of the DNA chain
can be initiated only by OH and 62— radicals [18]. The
primary radical centers forming on the DNA are capable of
binding oxygen with formation of the macroradicals of superoxide
type. The monomolecular dissociation of superoxide
macroradicals leads to DNA cleavage. Upon X-ray irradiation
of a DNA fragment a strong non-specific cleavage of DNA is
observed in close similarity to the hydroxyl radical cleavage
pattern observed in the presence of EDTA-Fe2+/H202/ascorbate
[ 16]. In the presence of Pt-bis-netropsin one could observe a small
protection of DNA from cleavage at or near the sites of ligand
binding (Fig.3c).
On adding 1M glycerol to the reaction mixture OH radicals
are transformed into relatively unreactive radicals of RO2 — type
[17]. On irradiating the DNA solution in the presence of glycerol
only a few cuts are observed (see Fig.5). It is interesting that
the position of these cuts is sequence-dependent: cleavage is
observed before the cytosine residues in sequences 5'-PuCPu-3'
and 5'-PuCC-3' (Pu = purine).
Upon X-ray irradiation of the DNA complexed with Pt-bisnetropsin, new cutting sites are observed near the center of the
260
Nucleic Acids Research, Vol. 19, No. 2
•
:f.
V
U
5'-A I C C I I 1 T M A
* A
A.
T T M M K T C M t
l t ! I A « A I C M T C ! M
S'yi.T C C I I I t A A «TT T A A A A A T G A A G t T
C A A I C T A A
A
a,
M *\
3"-I A G C A A A A T Itltfl
• Mil; jH" V
5
•!,«. ; » , ) ,
i«rc'e«MlWrAVrr'» w «ri v i v iHi A c 11 c i M v t T T A G T ? A e A T T
T T T T T A C T I C A A A A T T T A G T T A G A T T
X-rmut VtU Clyccrol
X-ray* Uitkoat Clycsrol
V - \ 1 ' '\*
\ * rr-A ** A ** J! a .*• fS *f * "
' v vV v v' v v v ; XJ Jiw/ J y- •) j v y:
\
S'-A
^ A A/AAAAAAAAAAAA
^ VV AV/ A
V ^ VA VA \ /VA\ A
A AA AAAA'
3'-I
VAA/IAAAA
T C C T T T T A A A T T A A A A A T C A A C T T T T A A A T C A A T C I A A
WAWWVV
W^A
A C C A A A A T T T A A T T T I T A C T T C A A A A I I t A C T T A C A T T
S'-A
-t
TCCTTITAAAtTAA
ACCAAAATTTAATTTTTACTTCAAAATTTACITACATT
Fig.3. Denshometer scans corresponding to the central parts of the gel autoradiograms in Fig.2. The number of scan lines corresponds to the same number lanes
of gels in Fig.2 with index 'a' for top and 'b' for bottom strands. (A) DNase I footprinting. (B) Hydroxyl radical footpnnting. (C) Cleavage pattern produced by
X-ray irradiation of Pt-bis-netropsin—DNA complexes in the absence and (D) in the presence of 1M glycerol.
sites of the expected ligand location, their share diminishing with
the decrease in the amount of ligand up to their disappearance
when the ligand concentration at an ionic strength of 0.25 M NaCl
becomes less than one ligand molecule per 50 base pairs.
The experimental results obtained by us and the literature data
allow one to suggest that specific cutting of the DNA can be
caused by local emission of the energy of the irradiation at places
where the platinum atoms are located on the surface of DNA.
It is well-known that absorption of photons of comparatively
mild X-ray irradiation by atoms in the medium is accompanied
by the photoeffect in the internal electron shells of the atoms of
a solute substance [17]. The probability of photon capture is
determined to a large degree by the ratio of a absorption
coefficients for all atoms present and energies of photons. As
a result, depending on which atom in the given molecule absorbs
the photon, the fate of the molecule can be different. Absorption
of the X-ray photon whose energy is dozens of KeV by an atom
with a large cross section leads to the emission of the
photoelectron from one of the internal (K, L, M etc) shells.
Subsequent filling of the 'hole' in the electron shell is
accompanied either by emission of an X-ray photon, or by
emission of several Auger-electrons from external shells of the
atom with formation of a multi-charged positive ion.
The low-energy Auger-electrons are thermalized in the vicinity
of the positive ion, losing energy upon ionization and excitation
of the molecules of the surrounding medium with formation of
short-lived products of radiolysis. So, at this stage, the formation
of a radical center on the DNA double helix is possible. Besides,
as a result of the neutralization of the positive ion which occurs
in parallel with the 'Auger-radiolysis', a great amount of energy
(dozens of eV) is released, which is sufficient for disruption of
several chemical bonds (The energy of a chemical bond equals
3—4 eV on average). Finally these processes lead to the
disruption of DNA.
A possible contribution of the Auger-mechanism resulting from
selective absorption of the energy by the atoms making-up the
molecule was first noted by Platzman in 1952. However, this
idea was substantiated experimentally much later (see Review
[17]). In the last 15-20 years the Auger-mechanism of molecular
decay under X-ray action was well-substantiated. In one of the
works of Halpern [18] it was called 'micro surgery" on biomolecules. This mechanism of molecular decay, including
molecular biological structures, underlies numerous variants of
anticancer therapy with the aid of X-ray and synchrotron
irradiation [17] including a photon X-ray therapy [19].
Upon X-ray irradiation of a dilute aqueous solution of DNA
X-rays are absorbed mainly by water molecules. Due to the low
concentration of DNA in solution and a small difference in the
number of electrons between atoms C, N, O and P the absorbance
of X-rays by DNA molecules is negligible as compared to that
Nucleic Acids Research, Vol. 19, No. 2 261
5 • -AGGGATTTTGGTCATGAGATTATCAAJWAGGATCrrCACCTAGATCCrrTTAAATTAAAA3 • - ' » " • • AAAACCAGTACTCTAATAGTTTTTCCTAG AAGTGGATCTAOGAAAATTTAATTTT
1568
5 -ATaAAO'm^fflTOQt'cTAAAGTATATATGAGTAAACTTGGTCTGicAGTTACCAATG3 ' -TACTTCAAAATTTAGTTAGATTTCATATATACTCATTTGAACCAGACTGTCAATGGTTAC
1628
2500 -
2460 EDTA / Fo2+
2430 ->
5 ' -AGGGATTTTGGTCATGAGATTATCAAAAAGGATCTTCACCTAGATCCTfTfAAATTAAAA3- -»,»>./UAACCAGTACTCTMTAGTTTTTCCTAGAAaTGaATCTAGGAAAATTTAATTTT
•• • m
HBWff.3
1568
5-ATGAAGTTiTATCATCAATCTAAAaTATAfATGAGTAAACTTGGTCTGACAGTTACCAATaTTTOKTC
3 ' -TACTTCAAAATTTAGTTAGATTrCATATATACTCATTTGAACCAGACTGTCAATGGTTAC
1628
2400 -
2380 -
II
.. . .•
*
5 • -TAAGOGATTTTGaTCATGAaATTATCAAAAAOGATCTTCACCTAGATCCTTTTAAiTTAA3-.»,t»»»HAAACCAGTACTCTAATAGTTTTTCCTAaAAGTGaATCTAGOAAAATTTAATT
tt
••
••••
2360 -»
2460
2340 -
{...'....
i
*.
•..».
•
•.
•••••
t
•• t ••• •
t ••
2320 -»
Flg.4. A part of the plasmid pUC-9 sequence. The data are taken from lane 3
on Fig 2 which were scanned by a laser densitometer. Following normalization
on the total areas of all bands on the gel, the relative inhibition of cutting was
calculated and plotted in the form of a histogram. The density of each bar
approximates the extent of protection from cleavage. Sites which are protected
from DNase I and hydroxyl radicals cleavage are black. The arrows indicate the
major sites of DNA cleavage resulting from X-ray irradiation of Pt-bis-netropsin
complexes with the restriction fragment. Sequences to which Pt-bis-netropsin binds
preferentially are underlined.
2300 -
2280
of the surrounding medium. In the presence of platinum atoms
the picture is quite different. For a photon with energy ranging
from 15 to 35 KeV, the mass absorption coefficient of the Pt
atom is more than 20-200 times greater than that of the atoms
C, N, O and P (estimated according to [17]). So, upon
polychromatic X-ray irradiation with the maximum photon energy
in the region of 22.5 KeV the probability of seizure of the X-ray
photon by the Pt atom exceeds by nearly 200 times the probability
of seizure by any other atom in the Pt-bis-netropsin—DNA
complex. So, nearly all the amount of the absorbed energy is
emitted from the platinum atom.
Filling the 'hole' in the L-shell of the Pt atom occurs in
accordance with the selection rules as a result of the electron
transition from one of the above-lying levels, with either X-ray
quantum or Auger-electron being emitted. The emitted X-ray
quantum can be subsequently seized by the same atom with
Auger-electron emission. For the Pt atom the photoeffect from
the L-shell causes a cascade of Auger-electrons with the
probability 0.65 [11] within a time interval on the order of
2480
2
5 ' -AAATOAAGTTTTAAATCAATCTAAAaTATATATGAaTAAACTTGGTCTGACAGTTACCAA3 -TTTACTTCAAAATTTAGTTAGATTTCATATATACTCATTTGAACCAGACTOTCAATGKTT
2260
2500
1
i
2520
I
i
2540
1
I
«- 2560
r_ -- *• 2580
Fig.5. Gel radkautograms.(a) The figure shows the action of X-ray irradiation
on the Pt-bis-netropsin complexes with fragment B labelled at the 3'-end in the
presence of glycerol. (b) The same DNA fragment labelled at 3'-end of the opposite
polvnucleotide strand. Lane (1): Untreated fragments, Lane 'A+G': A + G
sequence reaction, Lane (2): irradiation of the free DNA fragment, Lanes (3),
(4), (5): irradiation of the Pt-bis-netropsin complexes with DNA at a molar ratio
of Pt-bis-netropsin to DNA base pair ratio of 1/4, 1/12 and 1/36, respectively
10" 15 sec. It may be accompanied by the emission of 11
external electrons with the formation of a positive ion Pt +13 .
The Auger-electrons have energies ranging from 0.1 KeV to
several KeV. As a result of energy losses on ionization and
262 Nucleic Acids Research, Vol. 19, No. 2
excitation of the medium (water molecules, DNA fragments)
Auger-electrons are shown to decrease to thermal energies when
they travel over distances of about 100 run in the medium. So,
it
in the neighborhood of the Pt atom local radiolysis might take
place. On the other hand, the appearance of a multi-charged
positive ion Pt n+ induces a series of neutralization reactions,
which are connected with the redistribution of the boundary
orbitals of the Pt atom in the Pt-bis-netropsin molecule. Finally
these processes lead to breakage of certain DNA chemical bonds.
Analysis of the Pt-bis-netropsin binding sites shows that in our
method cleavage of the polynucleotide chain occurs preferentially
only at one nucleotide, unlike the cleavage pattern obtained by
netropsin analogs tethered to an iron chelator.
From the X-ray data for DNA-netropsin complexes [20-22],
it follows that netropsin binds in the minor groove to a run of
four AT base pairs where the width of the minor groove is smaller
than in the mixed sequence DNA regions [21]. In the central
part of the Pt-bis-netropsin there are two positive charges on the
Pt atom. This seems to favor a closer approach between the Pt
atom and the deoxyribose residues in the opposite DNA strands.
The characteristic time for formation of the multicharged Pt ion
upon X-ray irradiation and its subsequent neutralization and for
cleavage of DNA, seems to be relatively small compared to the
dissociation time of the Pt-bis-netropsin—DNA complex [23].
Therefore, for each binding site one can observe preferentially
a single cutting site on each of the two DNA strands. Such an
accuracy of cleavage allows us to compare Pt-bis-netropsin
affinities for adjacent and even partially overlapped binding sites
on DNA. This seems to be an advantage of our approache over
various modifications of the footprinting and affinity cleavage
techniques The latter usually give a more integral picture
concerning the distribution pattern of ligand on DNA. For
example, as one can see from Fig.3d, fragment A contains the
two overlapped strong binding sites with sequences TTTTAAAT
and AATTAAAA. At low concentrations of bis-netropsin when
each sequence is complexed with a single Pt-bis-netropsin
molecule, the cut sites are observed at both sequences. At the
saturation levels of binding a shift of the cleavage site by one
nucleotide is observed at the second sequence. It seems likely
that occupation of the stronger affinity binding site located in
the first sequence disturbs the binding of a second molecule which
appears to be partially bound (only one half of the bis-netropsin
is attached to DNA).
I
2240
I
I
t
I
I
AAC911 C I ^ 9 < ^ ^ C 9 A A A A C T C T C A A 9 9 A T C T T A C C 9 < J 1911 9A9ATC5A(;
iAAteccc<icTTTTcAcA(^ccTA(iAAT<i(ic(iAcAACTCTA(i<h-cAA(icTAc^
1
1
1 1
t
it
I
236Q
1
CTTT
AACCCACTC9T9CACCCAACT9ATCTTCA9CAT
\
t
CTCdil I I IT(JTCCTTCCt!;i I 1 I Acfojcfll I I I I ICCCTTATTCCC6CT(iT(icCTTTACAA
t
t
l
i
l
t
II
248Q
»_;
U—
25IQ
>_
I
1
9AATACTCATACTCTTCCI 111 IC^TATTATT9AA9CATTTATCA999TTATTt7TCTcA
CTTATcA<^AT<iA6AA&AAAAA(iTTATAATAAcrrctTAAATAfiTcccAATAAcA(iA(iT
_
1—. I
U258Q
AAAAATAAACAAATA999rfrTC
ACTCCCCTATCTATAAACTTACATAAATCiii1 i A r n < i i i iATccccAA&icccfrrCTA
It is not easy to find the optimal sequence for binding of lowmolecular-weight molecules due to their comparatively low
binding specificity. At high levels of binding, Pt-bis-netropsin
binds practically to any sequence which contains more than three
successive AT base pairs. The possibility to accurately determine
Fig.6. A part of the sequence of DNA fragment B. The arrows indicate the cleavage
sites arising upon action of X-ray irradiation. Sequences to which Pt-bis-netropsin
binds preferentially are underlined.
Table 1. A) Nucleotide composition of 85 cleavage sites at a molar Pt-bis-netropsin/DNA base pair ratio of about 1/41.
Base
A
T
C
G
Consensus
1
23
21
24
17
N
8*
24
16
25
20
N
22
47
10
6
T
27
24
14
20
T/A
10
57
9
9
T
13
58
14
0
T
57
19
1
8
A
45
10
18
12
A
30
20
23
12
T/A
10
11
12
41
29
10
5
A
18
28
20
19
N
20
10
13
33
N
B) Nucleotide composition of 12 cleavage sites at a molar Pt-bis-netropsin/DNA base pair ratio of about 1/361.
Base
A
T
C
G
1
3
6
2
1
2
1
2
7
2
3
4
8
0
0
4
5
6
0
1
5
2
10
0
0
6
1
11
0
0
7
12
0
0
0
8*
8
0
3
1
9
4
6
1
1
Data are obtained in the presence of 0.25 M NaCl. The cleavage site is shown with the asterisk.
10
5
6
0
1
11
1
4
3
4
12
3
5
2
2
Nucleic Acids Research, Vol. 19, No. 2 263
the location of the preferred binding sites for Pt-bis-netropsin
using X-ray irradiation allows us to perform a rapid screening
of a large number of preferred binding sites on DNA at different
ligand concentrations in order to determine sequence preferences
of the binding reaction. We have analyzed the cleavage patterns
observed upon X-ray irradiation of Pt-bis-netropsin complexes
with a number of DNA restriction fragments containing a total
of over 3 • 103 base pairs. Table 1 contains the sequences near
the cutting sites in which the rate of cleavage markedly exceeds
the rate of nonspecific cleavage on the both strands of the DNA
fragments. The upper part of the Table lists the frequencies of
the occurrence of the bases in the vicinity of cutting site at high
Pt-bis-netropsin concentrations.
In positions 1,2, 11 and 12 the occurrence of each of the four
bases is very close to that expected for a random nucleotide
sequence. However, in the eight central positions adenine and
thymine bases occur predominantly. The presence of guanine and
cytosine bases in these positions can be explained if only one
netropsin-like fragment of the Pt-bis-netropsin is attached to
DNA. Our data are consistent with the inference that Pt-bisnetropsin binds preferentially to the pseudosymmetrical sequence
5'-TXTTAAYA-3' where X and Y are adenine or thymine bases,
although cytosine and guanine may also occur in these positions.
In the lower part of the Table the frequencies of the occurrence
of the various bases at the preferred binding sites observed at
low bis-netropsin concentrations are listed. In the sequence shown
in Fig.3 there are three strong binding sites between nucleotides
1556-1563, 1561-1568 and 1565-1572. It should be noted
that the practically complete absence of sequences with GC base
pairs at the central part of the recognition sites for Pt-bis-netropsin
can be explained by the fact that upon binding of bis-netropsin
to such sequences the Pt atom is located too far from the
deoxyribose residues, presumably, due to steric hindrance caused
by the guanine 2 amino group.
The presence of four thymine and four adenine residues in the
left- and right-hand parts of the recognition sequence proves that
each fragment of the bis-netropsin molecule binds in an
asymmetrical manner to DNA. Fig.7 illustrates the proposed
model for the strongest type of DNA complex with Pt-bisnetropsin. Like netropsin, Pt-bis-netropsin binds in the DNA
minor groove. It occupies 8 base pairs upon binding and forms
6 bifurcated hydrogen bonds connecting each amide group with
the two adjacent AT base pairs.
The 3' end-labeled DNA fragments obtained after X-ray
irradiation of DNA complexes with Pt-bis-netropsin exactly
comigrated with the 5'-phosphate-ended Maxam-Gilbert sequence
markers, suggesting that cutting sites induced by X-rays also
contain 5'-phosphates.
This demonstrates that the deoxyribose residues adjacent to the
Pt atom are attacked. The absence of new bands upon piperidine
treatment of the DNA fragments obtained after X-ray irradiation
of Pt-bis-netropsin complex with DNA (not shown) suggests that
the bases are not damaged [24]. This is also supported by the
fact that cleavage occurs nearly identical with nearly the same
efficiency at all sites with no dependence on the nature of the
base attached to the attacked sugar.
Introduction of a heavy atom in a DNA binding ligand and
irradiation of its complexes with X-rays allows one to determine
the sequence preferences for binding of the ligand to DNA.
It should be noted that our approach can be easily applied to
the determination of sequence preferences shown by other DNAbinding molecules. The observed cleavage mechanism can be
used to construct and synthesize other compounds that may cut
DNA at specific sites. These results are also of interest for
3- 6-
3' 3Fig.7. (A) Scheme illustrating the general plan of the structure of the Pt-bis-netropsin—DNA complex. The straight lines symbolize DNA base pairs. The circles
symbolize deoxyribose residues in each of the two DNA strands, whereas the bands represent netropsin-like fragments of Pt-bis-netropsin. The crosses indicate positions
of cleavage sites arising after irradiation with X-rays of the Pt-bis-netropsin complex. (B) Model demonstrating the bifurcated hydrogen bonds.
264 Nucleic Acids Research, Vol. 19, No. 2
radiobiology and X-ray therapy of tumors. Introduction into a
sequence-specific ligand of an inert group, capable of triggering
a mechanism of DNA modification only under X-ray irradiation
might be of interest for cancer therapy.
ACKNOWLEDGMENTS
The authors are grateful to Drs. G.Gursky, A.Zasedatelev,
A.Zhuze, B.Gottich, L.Bugaenko and R.Shafer for valuable
discussions and N.Sidorova for isolation of the plasmid.
REFERENCES
1. Gursky, G.V., Zasedatelev, A.S., Zhuze, A.L., Khorlin, A.A., Grokhovsky,
S.L., Streltsov, S.A., Surovaya, A.N., Nikitin, S.M., Krylov, A.S.,
Rctchinsky, V.O., Mikhailov, M.V., Beabealashvily, R.Sh. and Gottikh,
B.P. (1983), Cold Spring Harbor Symp. Quant. Biol., 47, 361-378.
2. Zimmer, C. and Wahnert, U. (1986), Prog. Biophys. Mo). Biol. 47, 31 -112.
3. Galas, D.J. and Schmitz, A. (1978), Nucleic Acids Res., 5, 3157-3170.
4. Maxam, A.M and Gilbert, W.A. (1980), Methods Enzymol. 65, 499.
5. Nielsen, P.E. (1990), J. Mol. Recognition 3, 1-25.
6. Dervan, P.B. (1986), Science, 232, 464-471.
7. Zasedatelev, A.S., Zhuze, A.L., Zimmer, Ch., Grokhovsky, S.L.,
Tumanyan, V.G., Gursky, G.V. and Gottikh, B.P. (1976), Dokl. Akad.
Nauk SSSR, 231, 1006-1009.
8. Baker, B.F. and Dervan, P.B. (1985), J. Am. Chem. See. 107, 8266-8268.
9. Rill, R.L., Marsch, G.A. and Graves, R. (1989), J. of Biomol. Structure
& Dynamics 7, 591-605.
10. Warpehoski, M.A. and Hurley, L.H. (1988), Chem. Rev. Toxicol. 1,
315-333.
11. Martin R.F. & Holmes N. (1983), Nature 302, 452-454.
12. Grokhovsky, S.L., Zhuze, A.L. and Gottikh, B.P. Bioorgan. Khem.(USSR)
in press.
13. Grokhovsky, S.L. and Zubarev, V.E. (1990), Dokl. Akad. Nauk SSSR, 313,
No.6.
14. Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982), Molecular Cloning:
A Laboratory Manual. Cold Spring Harbor University Press, Cold Spring
Harbor, N.Y.
15. Kraev, A.S. (1988), Molek. Biol. (USSR) V.22, 1164-1197.
16. Tullius, T.P. and Dombroski, B.A. (1986), PNAS 83, 5469- 5473.
17. Halpern A. In: Uses of Synchrotron Radiation in Biology. (Ed.
H.B.Stuhrmann), Academic Press, London (1982), 255-283.
18. Henglein A., Schnabel W., Wendenburg J. (1969), Einfuhrung in die
Strahlenchemie. Akademie Verlag, Berlin 195-211.
19. Shinohara, K., Ohara, H., Kobayashi, K., Mezava, H., Hieda, K., Okada,
S. and Ito, T. (1985), J. Radial. Res., 26, 334-338.
20. Coll, M., Aymami, J., van der Marel, G.A., van Boom, J.H., Rich, A.
and Wang, A.H.-J. (1989), Biochemistry, 28, 310-320.
21. Dickerson, R.E. and Kopka, M.L. (1985), J. of Biomol. Structure &
Dynamics, 3, 423-431.
22. Kopka, M.L., Joon, C , Goodsell, D., Pjura, P., Dickerson, R.E. Proc.
Nat. Acad. Sci. U.S.A. (1985), 82, 1376-1380.
23. Klevit, R.E., Wemmer, D.E. and Reid, B.R. (1986), Biochemistry, 25,
3296-3303.
24. Kuwabara, M., Yoon, Ch., Goyne, T., Hederahn, T. and Sigman, D.S.
(1986), Biochemistry, 25, 7401-7408.