Biochimica et Biophysica Acta, 1161 (1993)291-294
291
© 1993 Elsevier Science Publishers B.V. All rights reserved 0167-4838/93/$06.00
BBAPRO 34389
The differential lysis of phosphoester bonds by nuclease P1
Harold C. Box a, Edwin E. Budzinski a, Marianne S. Evans a, John B. French a
and Alexander E. Maccubbin b
a Biophysics Department, Roswell Park Cancer Institute, Buffalo, N Y (USA)
and b Experimental Therapeutics Department, Roswell Park Cancer Institute, Buffalo, N Y (USA)
(Received 28 February 1992)
(Revised manuscript recieved 5 August 1992)
Key words: Phosphodiesterbond; Nuclease P1; Hydrolysis;DNA damage
The hydrolysis by nuclease P1 of the 16 common deoxydinucleoside monophosphates was examined. The rates of hydrolysis of
phosphodiester bond differ by more than two orders of magnitude; dinucleotide monophosphates of the type d(TpN) being most
resistant and d(GpN) being next most resistant. The profiles of a mixture of the 16 common dinucleoside monophosphates and of
DNA after partial hydrolysis by nuclease P1 and simultaneous treatment with acid phosphatase were compared. The resultant
profiles are very similar, except for the appearance of 5-methyldeoxycytidine in the latter. Similar profiles are also obtained from
a mixture of dinucleoside monophosphates and from DNA exposed to ionizing radiation beforehand. The 8-hydroxyguanine
lesion and a formamido remnant of thymine appear in both profiles as a modified nucleoside and as modified dinucleoside
monophosphate respectively. These results suggest that certain radiation induced DNA lesions can be selectively postlabeled
based on their resistance to hydrolysis by nuclease P1. The nature of the nuclease Pl-substrate interaction is discussed.
Introduction
Nuclease P1 is an endonuclease capable of hydrolyzing D N A completely to the level of mononucleoside
5'-monophosphates. It digests single-stranded D N A
about 200-times faster than double-stranded DNA. The
enzyme seems to recognize particular conformations of
the phosphodiester bonds rather than specific sequences as in the case of D N A binding proteins. Nevertheless, there is a significant variation of the enzyme's
effectiveness, depending on which nucleosides are coupled by the phosphodiester bond. The nature of this
dependence was the subject of our investigation. Fujimoto and co-workers have previously investigated nuclease Pl-substrate interactions [1].
One of the motives for this study was the prospect
that analyses of D N A damages might be better carried
out at the dinucleoside monophosphate level rather
than at the monomer level. The basis for this approach
to D N A damage analysis is the fact that certain lesions
cause particular dinucleoside monophosphate sequences to resist hydrolysis by nuclease P1 [2,3].
Correspondence to: H:C. Box, Biophysics Department, Roswell Park
Cancer Institute, Elm and Carlton Streets, Buffalo,NY 14263,USA.
Materials and M e t h o d s
Dinucleoside monophosphates were obtained from
Sigma (St. Louis, MO, USA) with the exception of
d(CpC) which was synthesized by ChemGenes (Needham, MA, USA). Calf thymus DNA was purchased
from Cooper Biochemicals (Malvern, PA, USA). The
chromatographic behavior on our HPLC system (see
below) of the individual dinucleoside monophosphates
was determined beforehand.
Mixtures of dinucleoside monophosphates were subjected to enzymatic hydrolysis and dephosphorylation.
The mixture was incubated in a 1 ml solution containing 1.44/~l of 30 mM ZnCl2, 240/zl of 0.25 M sodium
acetate (pH 5.0), 0.48/zg of nuclease P1 (BoehringerMannheim) and 3.3/zl of a solution containing 100 mU
per /.tl of prostatic acid phosphatase (Sigma). The
amount of substrate used in an experiment is indicated
in captions.
The mixture was incubated at 37°C for a duration
also indicated in captions. Reactions were terminated
by adding 23.3 /xl of 1.0 M 2-[N-cyclohexylamino]ethanesulfonic acid (CHES) buffer (pH 9.5). The
foregoing procedures were used in the treatment of
D N A samples also. Some samples were irradiated prior
to enzymatic digestion. The irradiation procedure has
been described previously [4].
292
The progress of enzymatic digestions was monitored
by HPLC. Approx. 100 /xg of material was injected
onto a Ultremex 3/x C18 RP column and eluted at 0.2
ml/min for 30 min with 0.1 ammonium acetate, followed by a 60 min 0-10% acetonitrile gradient in 0.1
M ammonium acetate at a flow rate of 2 ml/min,
followed by isocratic elution at 2 ml/min with 10%
acetonitrile in 0.1 M ammonium acetate.
Results
Quantities equal in A254 absorbtion units of the 16
dinucleoside monophosphates were combined in aqueous solution and chromatographed by HPLC. The result is shown in Fig. 1, where all 16 dinucleoside
monophosphates are distinguished and identified. The
order of elution of the individual nucleosides under
our HPLC conditions is deoxycytidine, deoxyguanosine,
thymidine and deoxyadenosine (not shown). A regularity appears in the order of elution of the dinucleoside
monophosphates. The dinucleoside monophosphates,
for a given 3' nucleoside, elute according to the 5'
nucleoside in the same order as the individual nucleosides. For example, for d(NpN')when N' is deoxycytidine, the order is d(CpC), d(GpC), d(TpC) and d(ApC).
The same elution order holds when N' is deoxyguanosine, thymidine or deoxyadenosine.
Various combinations of the 16 dinucleoside
monophosphates were partially digested by the action
of nuclease P1. Acid phosphatase was added to the
solution to remove terminal 5'-phosphates and maintain conditions comparable to a following experiment
using DNA (see Materials and Methods). The fraction
of each dinucleoside remaining after hydrolysis was
d(CpC ) ~
.
I
d(ApT)
°,ooo). \ l
.1¢,,,°c,
I
30
d (T p G ) ~ , . ~
~
d(TpC ) ~ , , , ~
~
TI ME (min.) ~
/
))
I
90
Fig. 1. The HPLC elution profile of a mixture of the 16 dinucleoside
monophosphates. Elution conditions are described in Materials and
Methods. Detection was by absorption at 254 mm.
TABLE I
The fraction of dinucleoside monophosphate d(NpN ') remaining after
digestion by nuclease P1 plus acid phosphatase
The data are taken from experiments on different mixtures of
dinucleoside monophosphates. In each experiment the total amount
of starting dincleoside monophosphates was 1 absorbance unit which
was incubated under the conditions described in Materials and
Methods for 10 min. In experiment 1, the mixture contained equiabsorptive amounts of all 16 dinucleoside monophosphates. In experiment 2 the mixture contained 0.2 absorbance units each of d(CpC),
d(CpA), d(TpC), d(ApG) and d(TpT). In experiment 3, the mixture
contained 0.2 absorbance units each of d(CpG), d(TpG), d(GpA),
d(ApC) and d(TpT).
d(NpN ')
Ratio
Experiment No.
d(TpT)
d(TpA)
d(TpG)
d(TpC)
d(GpA)
d(GpT)
d(GpG)
d(GpC)
d(CpC)
d(CpG)
d(CpT)
d(ApA)
d(ApT)
d(ApG)
d(CpA)
d(ApC)
0.96
0.71
0.69
0.64
0.53
0.53
0.42
0.35
0.28
0.19
0.14
0.11
0.07
0.05
0.01
0.004
3
1
3
2
3
1
1
1
2
3
1
1
1
2
2
3
determined. Some results are given in Table I. An
interesting regularity appears in these results also. The
four most resistant dinucleoside monophosphates are
those having thymidine as the 5'-nucleoside. The next
four most resistant dinucleoside monophosphates have
deoxyguanosine as the 5'-nucleoside. Dinucleosides
having either deoxycytidine or deoxyadenosine as the
5' nucleoside are the most susceptible to hydrolysis.
The HPLC elution profile after partial hydrolysis of
a mixture of all 16 dinucleoside monophosphates by
nuclease P1 plus acid phosphatase is shown in Fig. 2a.
Fig. 2b shows the HPLC profile of calf thymus DNA
after similar digestion by nuclease P1 plus acid phosphatase. The digestions were allowed to progress far
enough such that only the more resistant phosphoester
bonds (see Table I) remain intact. Consequently, in
Fig. 2 the elution products are mainly nucleosides
together with several persistent dinucleoside
monophosphates. The general correspondence between the residual dinucleoside monophosphates in
Fig. 2a and 2b is evident. One difference is the appearence of a peak due to 5-methyldeoxycytidine in the
DNA digest due to its natural abundance in the polymer.
The investigation was extended to the enzymatic
digestion of dinucleotide monophosphates and of D N A
after exposure to ionizing radiation. Fig. 3a shows the
293
HPLC elution profile from a mixture of the 16 dinucleotide monophosphates following exposure to Xirradiation (600 Gy) in an oxygenated aqueous solution.
Fig. 3b shows the results obtained from the same
weight of polymer D N A after analogous treatment.
Damaged entities give rise to additional product peaks
in Fig. 3 compared with Fig. 2. The largest of the
additional peaks is the 8-hydroxyguanine modification,
a well-known oxidatively-induced lesion in DNA. The
second most prominent additional peak in Fig. 3 is due
to undigested d(TpA) in which the thymine base is
degraded to a formamido remnant. The identities of
these two damaged products could be established be-
ca, dcI
ilk.it
II
III
JlII
Ill
llt
II .'''('r'T)
III ...,,(~°T)
II
III ~.o~iLll
II JiLl IILi:nl
II
IIIlll nn
(a)
dG
dC\
I/dA
/
.d(TpT )
.d(TpA)
dC
(b)
~G
/ dA
5
"
/d(GPA)
d(GpC)
d(GpG)
J
dtrpG)
/80HdG
/d(TpT)
/ d(TpA)
.d(GpT)
/d(GpA)
%
(b)
~T
,,c
It
/d(TpT)
/d(TpA)
i
i/
I
~,0
/d(GpA)
II
d(GpG ~"11
d(GpC)X I
\
cause reference materials were available from prior
studies at the dimer level [4,5]. Overall, the profiles in
Fig. 3a and 3b are very similar.
i
I
30
I
I10
Fig. 3. HPLC elution profiles of nuclease P1 (plus acid phosphatase)
digests. (a), mixture of 16 dinucleoside monophosphates totalling 100
/.~g, after irradiation in oxygenated aqueous solution, was incubated
for 24 h under conditions described in Materials and Methods; (b),
calf thymus D N A (100 /zg) irradiated in oxygenated aqueous solution, after similar treatment. Note the appearance of same
radiation-damaged products, 8-hydroxydeoxyguanosine and the formamido modification of d(TpA) (arrows), in both profiles.
~(GpT)
d(TpG)
5Me-de
T I M E (min.) ~
T I ME (min.)
~
I
IIO
Fig. 2. HPLC elution profiles of nuclease P1 (plus acid phosphatase)
digests. (a), mixture of 16 dinucleoside monophosphates totalling 100
/zg was incubated for 10 rain under conditions described in Materials
and Methods; (b), calf thymus D N A (100/zl0 after similar treatment.
Note the appearance of 5-methyldeoxycytidine (arrow) in the polymer digest.
Discussion
Several insights were gained from this study of nuclease P1 activity. HPLC has been used successfully to
distinguish each dinucleoside monophosphate in a mixture of the 16 normal dinucleosides. Under our conditions, the four compounds in each set of compounds
294
dN, d(NpA), d(NpC), d(NpG) and d(NpT) eluted in
the same order according to N, namely deoxycytidine,
deoxyguanosine, thymidine and deoxyadenosine.
Although nuclease P1 accepts all 16 dinucleoside
monophosphates as substrates for phosphoester bond
cleavage, the cleavage efficiency is considerably better
when deoxyadenosine or deoxycytidine is the 5'nucleoside. The efficiency is less and least, respectively, when deoxyguanosine or thymidine are the 5'nucleoside. Our studies were carried out on mixtures
of dinucleosides since the aim was to assist in the
interpretation of nuclease P1 digests of DNA, particularly damaged DNA. For this reason also, acid phosphatase was added to the enzyme preparation. Previously Fujimoto et al. [1] measured the rate of cleavage
of phosphoester bonds by nuclease P1 in eight individual dinucleoside monophosphates. A detailed comparison of results is inappropriate since competitive inhibition comes into play when several substrates are present. However, from the work of Fujimoto et al. it can
also be concluded that d(TpN') dinucleosides are most
resistant whereas d(CpN') and d(ApN') are most compliant. Dinucleosides d(GpN') were not tested by Fujimoto et al. [1].
Recently, the crystal structure of nuclease P1 has
been determined [6]. Two nucleotide binding sites were
identified. Studies of a complex between nuclease P1
and the R stereoisomer of the thiophosphorylated dinucleotide dA. p(S). dA, an uncleavable substrate analog, suggest hydrogen bonding between the protein and
the 5'-deoxyadenosine involving adenine N-6 and N-1
[7]. Perhaps this hydrogen-bonding arrangement facilitates the cleavage by the protein of a normal dinucleoside monophosphate having a 5'-deoxyadenosine, as we
have observed. Moreover, since cytosine offers the
possibility of a similar hydrogen-bonding arrangement,
cleavage of a dinucleoside monophosphate having a
5'-deoxycytidine may also be facilitated. This point of
view coincides with our experimental results. The expected availability of a more refined crystal will allow
these ideas to be probed in greater detail.
A partial nuclease P1 (plus acid phosphatase) digest
of DNA and of a mixture of dinucleoside monophosphates were compared. There is little difference in the
profiles of the two digests. Nucleosides and resistant
dinucleoside monophosphates dominate the profile. Of
interest is the effect of differential hydrolysis on the
profile of a digest of damaged DNA. Dinucleoside
monophosphates d(TpN) are relatively poor substrates
compared with other dinucleoside monophosphates.
However d(T*pN'), where T* is a damaged thymidine, are usually even poorer substrates. Examples
include thymine modified to a glycol or to a formamido
remnant [8,9]. Thymine is the base most often damaged
by ionizing radiation or oxidative stress. Consequently,
because of their prominence and resistance to hydrolysis it would seem that these lesions could be effectively
isolated from DNA digests as modified dimers.
Acknowledgement
This work was supported by Grant CA44808 from
the National Cancer Institute.
References
1 Fujimoto, M., Fijiyawa, K., Kummaka, A. and Yoshino, H. (1974)
Agr. Biol. Chem. 38, 2142-2147.
2 Randerath, K., Randerath, E., Danna, T.F., Van Golan, K.L. and
Putman, K.L. (1989) Carcinogenesis 10, 1231-1239.
3 Weinfeld, M., Liuzzi, M. and Patterson, M.C. (1989) Nucleic Acids
Res. 17, 3735-3765.
4 Paul, C.R., Belfi, C.A., Arakali, A.V. and Box, H.C. (1987) Int. J.
Radiat. Biol. 51, 103-114.
5 Maccubbin, A.E., Evans, M.S., Budzinski, E.E., Wallace, J.C. and
Box, H.C. (1992) Int. J. Radiat. Biol. 61,729-736.
6 Lahn, A., Vabeda, A. and Suck, D. (1990) J. Mol. Biol. 215,
207-210.
7 Volbeda, A., Lahn, A., Sakiyama, F. and Suck, D. (1991) EMBO J.
10, 1607-1618.
8 Paul, C.R., Budzinski, E.E., Maccubbin, A.E., Wallace, J.C. and
Box, H.C. (1990) Int. J. Radiat. Biol. 58, 759-768.
9 Maccubbin, A.E., Evans, M.S., Paul, C.R., Budzinski, E.E., Przybyszewski, J. and Box, H.C. (1991) Radiat. Res. 126, 21-26.