J. gen. ViroL (I98O),48, 231-236
Printed in Great Britain
23I
Catalysis of Adenovirus I ) N A Synthesis in vitro
by D N A Polymerase 7
( A c c e p t e d 26 N o v e m b e r I979)
SUMMARY
DNA replication complexes were purified from adenovirus type 5 (Ad5)infected HeLa cells, DNA synthesis by these complexes in vitro was extremely
sensitive to the inhibitors dideoxythymidine triphosphate, N-ethyl maleimide and
p-hydroxymercuribenzoate. The bound DNA polymerase was released from
the complexes by limited digestion with micrococcal nuclease. This released
polymerase preferred poly(rA):(dT)12_ls as template over activated calf thymus
DNA. These results are compatible with the major polymerase in the replication
complex being of the y class.
Although a number of virus-coded functions are required for the replication of adenovirus
DNA, there is no evidence for any of these being a DNA polymerase and it has been
assumed that host enzymes are involved in some way. DNA polymerase ~ has been implicated on the basis of the increase in activity of this enzyme in infected cells following adenovirus infection (de Jong et al. t977) and on the inhibitory effect of aphidicolin (Longiaru
et al. t979) on virus replication. In contrast, experiments examining the effect of 2'3'
dideoxythymidine 5' triphosphate (ddTTP) on adenovirus DNA synthesis in isolated
nuclei (Van der Vliet & Kwant, I978) have pointed to DNA polymerase 7 playing a crucial
role in replication. On the other hand, studies of the properties of putative replication
complexes isolated from adenovirus-infected cells have suggested that cellular polymerases
of both the cc and y classes are involved (Ito et al. I976; Arens et al. I977; Brison et al.
I977; Frenkel, r978; Abboud & Horwitz, I979).
We have previously described (Shaw et al. I978, 1979; Shaw, I979) a soluble subnuclear
fraction (the $2oo fraction) which synthesizes adenovirus DNA in vitro. This relatively crude
fraction contains adenovirus DNA replication complexes, with a template-bound DNA polymerase, and also a vast excess of non-template bound DNA polymerase which can be
effectively separated by sucrose gradient centrifugation. The nature of the DNA polymerase
responsible for adenovirus DNA replication in this system has been studied.
The $2oo fraction was prepared using the method of Wilhelm et al. (1976), by treating
nuclei isolated from adenovirus type 5 (Ad5)-infected HeLa cells at I7 to I8 h p.i. with
20o raM-ammonium sulphate. Removal of the nuclear pellet by centrifugation produced a
supernatant (the S2oo fraction) containing most of the replicating DNA and the major
portion of the in vitro DNA synthetic activity found in the infected cell. This fraction was
then layered on to a Io to 58% (w/v) sucrose gradient containing 20 mM-tris-HCI, pH 7"9,
z mM-EDTA and 2oo mM-ammonium sulphate, and centrifuged for 3 h at 4I ooo rev/min
in a Beckman SW4r rotor. Fractions were collected dropwise from the bottom of the tube
and analysed for DNA polymerase activity in the presence or absence of activated calf
thymus DNA. Fractions in the middle third of the gradient containing the Ad5 DNA
replication complex, and the non-template-bound DNA polymerase towards the top, were
pooled separately for further analysis. The DNA replication complex thus isolated synthesized full length adenovirus DNA in vitro and accounted for the major portion of the
Downloaded
oo22-1317/8o/oooo-3982 $o2.oo~)
I98ofrom
SGMwww.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:20:16
232
Short communications
T a b l e I. Template preferences of DNA polymerases*
(a) Micrococcal nuclease-treated replication complex
Incorporation with
A
Template
None
Poly(rA): (dT)x2_ls (1 : I)
Activated calf thymus DNA
Poly(rA): (dT)~2-ts (7: 3)
Poly(dA): (dT)t~-ts (7 : 3)
5 mM-Mg~+
0'8 mM-Mn2+
o 053"5)I"
28"t
2I'7
12-8
4"I
o (47"51")
38"2
2-0
ND~
ND
(b) Non-template-bound DNA polymerase
Incorporation with
A
Template
None
Poly(rA): (dT)~2-1s ([ : 0
Activated calf thymus DNA
Poly(rA): (dT)t 2-18 (7: 3)
Poly(dA) : (dT)x~-as(7 : 3)
5 mM-Mg2+
O-8 mu-Mn 2+
< o-oi
o'o ~
5"9
o'o I
o"~5
< o'o1
o'ot
3"4
o'o t
o'05
* 2o/zl samples were incubated in a reaction vol. of loo Itl which contained 4 mM-ATP, 5 mM-MgCl2,
5o mM each of dATP, dCTP, dGTP, o5 #M-3H-dTTP (sp. act. 3o Ci/mmol) 2o mM-Hepes/KOH, pH 7"9,
4omM-(NH4)2SO4, 44mM-NaCI, o.2 mM-CaCI2, o'4 mM-dithiothreitol and 20% glycerol. After 6omin
incubation at 37 °C, the reactions were stopped by adding 2 ml of cold 2o % trichloroacetic acid (TCA).
The precipitates were collected by filtration on to Whatman GF/C discs, which were washed with 2oo vol.
of cold TCA and 5o vol. of cold ethanol. The discs were air dried and counted in 5 ml of toluene scintillation
fluid containing 0'5 % PPO and o.ol % POPOP. Template concentration used in all assays was 50/~g/ml.
Figures in parentheses indicate the ratio of adenosine-containing polymer to thymine-containing polymer
by molarity of nucleotides. Incorporation is expressed in pmol ~H-dTMP incorporated in 60 min. Poly(rA):
(dT)l~-is ( t : 0 was obtained from Boehringer (Lewes, Sussex), poly(rA), poly(dA) and (dT)l~-ls were
obtained from P. L. Biochemicals (Windsor, Berks.) and the appropriate primer-template prepared by
mixing and incubating at 37 °C for 15 rain, followed by rapid chilling.
t Incorporation with intact replication complex as template.
:]: NO, Not done.
e n d o g e n o u s D N A synthetic activity o f the 82oo fraction (Shaw et al. I979b). The nont e m p l a t e - b o u n d D N A p o l y m e r a s e could also be isolated in a similar m a n n e r from uninfected cells.
T h e t e m p l a t e preferences o f purified e u k a r y o t i c D N A p o l y m e r a s e s have been used as a
means o f differentiating these enzymes. Thus, D N A p o l y m e r a s e 7 prefers d e o x y r i b o oligonucleotide-primed polyribonucleotides, whereas the a - a n d fl-polymerases prefer n a t u r a l
D N A activated by limited D N a s e digestion (Weissbach, I977)- To assess the t e m p l a t e
preference o f the D N A p o l y m e r a s e a l r e a d y b o u n d to the replication c o m p l e x the enzyme
was released by t r e a t m e n t o f the replication complex with micrococcal nuclease ( l o # g / m l )
for I5 rain at 37 °C then inhibiting the nuclease b y a d d i t i o n o f 2 m M - E G T A (Pelham &
Jackson, I976). Such micrococcal nuclease-treated replication complexes were d e p e n d e n t
on a d d e d template D N A for synthetic activity and exhibited negligible residual nuclease
activity.
Table T(a) shows the results o f assaying micrococcal nuclease-treated replication c o m plexes for D N A p o l y m e r a s e activity in the presence o f various templates, utilizing M g ~+ or
M n 2+ as the divalent cation. U s i n g M g 2+ as divalent cation, activity was greatest with
poly(rA):(dT)12 is ( I : I ) followed closely b y activated calf t h y m u s D N A and p o l y ( r A ) :
(dTh2-1s (7:3). A small a m o u n t o f synthesis was detectable using poly(dA):(dT)12 18 (7:3).
However, using M n 2+ as divalent cation, poly(rA):(dTh2_18 was greatly preferred to
activated calf thymus, activity on the former being t9-fold greater. In addition, M g 2+ was
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:20:16
Short communications
233
the preferred cation for copying activated calf thymus DNA. These results are all consistent
with the replication complexes containing mainly DNA polymerase y.
In contrast, Table I (b) shows that when assayed in the same manner, the non-templatebound DNA polymerase from Ad5-infected (or uninfected) ceils exhibited a 4o to 7o-fold
preference for activated calf thymus DNA over poly(dA) : (dT)i~_18, with negligible synthesis
occurring on poly(rA): (dT)12_18. This data is consistent with the non-template-bound DNA
polymerase being mainly DNA polymerase ~z.
The nature of the replication complex and non-template-bound DNA polymerase was
investigated further by the use of specific inhibitors of DNA synthesis. The three classes of
eukaryotic DNA p o l y m e r a s e respond differently in vitro to ddTTP. The ~-polymerase is
largely resistant to d d T T P / d T T P ratios of up to Io, whereas the fl-polymerase is 8o to 95 %
inhibited at this concentration (Edenberg et al. I978; Van der Vliet & Kwant, 1978; Waqar
et al. i978 ). DNA polymerase 7 is the most sensitive, being more than 95 % inhibited at
ratios of less than I (Edenberg et al. 1978; van der Vliet & Kwant, I978). The effect of this
inhibitor on DNA polymerase activity in the pooled fractions, in the nuclease-treated
replication complex and in the $2oo fraction was investigated, and the results plotted in
Fig. I (a). The values determined for the purified DNA polymerases e¢,/J' and y by Edenberg
et al. (I978) and Waqar et al. 0978) are also shown (solid lines).
Endogenous DNA synthesis in the replication complexes was extremely sensitive to
ddTTP, 9o o/,) inhibition occurring at a d d T T P / d T T P ratio of I i 2o. Micrococcal nucleasetreated replication complexes copying activated calf thymus DNA were slightly less sensitive, 9o% inhibition occurring at a ratio of 1:3. Both these values are lower than those
previously described for DNA polymerase 7 (Edenberg et al. 1978; Van der Vliet& Kwant,
~978; Waqar et al. 1978). In neither case did any significant residual activity remain above
a d d T T P / d T T P ratio of I :I, even when copying activated calf thymus DNA. This is the
preferred template for DNA polymerases ~ and fl and, if significant amounts of these
polymerases were present, they should have been detected by such an analysis. Since no
ddTTP-resistant DNA polymerase activity was detectable under these conditions it would
seem to preclude an involvment of DNA polymerase ~z or fl in this in vitro DNA synthesis
system (assuming that the results obtained with the purified enzymes are equally applicable
to this relatively crude system).
In contrast, the non-template-bound DNA polymerase from both Ad5-infected (and
uninfected) cells was largely resistant to ddTTP and retained 4o~o activity at ratios up to
Ioo: r. The slightly greater sensitivity to low ddTTP concentrations indicates that there
may be some fl- or y-polymerase in this fraction.
Endogenous DNA synthesis in the $2oo fraction was sensitive to low concentrations of
ddTTP. Up to a ratio of I : 1o, the activity resembles DNA polymerase 7- However, above
this value the activity remains resistant to ddTTP and retains 25 0/,, activity at a ratio of
7o: ~. Thus, DNA polymerase ~zprobably accounts for 20 to 3o °/~, of the endogenous DNA
synthetic activity of the $2oo fraction. Whether this is due to DNA polymerase ~ loosely
bound to the replication complexes which is subsequently removed by centrifugation, or
simply adventitious synthesis by non-template-bound DNA polymerase on fragments of
DNA is not known.
N-ethyl maleimide (NEM) and p-hydroxymercuribenzoate (pHMB) are sulphydryl
group blocking agents which are extremely inhibitory to DNA polymerase ~z and y, but
DNA polymerase fl is resistant to up to 5 mM-NEM and 30/~M-pHMB (Weissbach et al.
I97I ; Wang et al. I974; Matsukage et al. I975). The effect of these inhibitors on D N A
polymerase activity was thus investigated and the results expressed in Fig. l (b) (NEM) and
(c) (pHMB) in which the values determined for a purified fl-polymerase by Matsukage
et al. (1975) are shown as solid lines.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:20:16
Short communications
234
I
I
I
(a) _
100
o
"7- 50
<
f)
o
0
=t_
0
T
i
•
I
100
5
Ratio ddTTP/dTTP
I
I
I
I
I
I
I
i
I
I
(b)
(c)
100
O.
<
50
\
0
'",.,\
\
-
"c2 \
~\\~\ ~
' "")'A
%
\'O.
\ '"....
O-~
"'©
0
I
10o
i
I
I
101
102
103
NEM concn.(btu)
--o
104
10 o
101
10 2
10 3
10 4
pHMBconcn.(taM)
Fig. i. Effect of specific inhibitors on DNA polymerase activity. DNA polymerase assays were
performed as described in the legend to Table I. (a) ddTTP. 0 , Non-template-boundDNA polymerase from Ad5-infected cells, copying 5o/~g/ml activated calf thymus DNA: 1oo% activity
= I2"5 pmol 3H-dTMP incorporated/6o min. O, Non-template-bound DNA polymerase from
mock-infected cells, as above: Ioo % activity = 7"8 pmol 3H-dTMP incorporated/6o min. A, EndogenousDNA synthesis in the Szoo fraction: IOO% activity = I pmol ~H-dTMP incorporated/
60 rain. M, Endogenous DNA synthesis in the replication complexes: ioo % activity = zzo fmol
3H-dTMP incorporated/6o min. [], Micrococcal nuclease-treated replication complexes copying
5o/~,g/ml activated calf thymus DNA: ~oo % activity = 13 fmol ~H-dTMP incorporated/6o rain.
In all assays the dTTP concentration was adjusted to I /~M. (b) NEM. O, Endogenous DNA
synthesis in the replication complexes: Ioo % activity = 93 fmol 3H-dTMP incorporated/6o rnin. O,
Non-template-bound DNA polymerase copying 5o/*g/ml activated calf thymus DNA: ioo%
activity = 9'8 pmol 3H-dTMP incorporated/6o min. (c) pHMB: symbols as in (b).
Both N E M and p H M B were extremely inhibitory to the e n d o g e n o u s D N A synthetic
activity of the replication complexes, p H M B being effective at [/~,M. The n o n - t e m p l a t e b o u n d D N A polymerase was somewhat less sensitive to b o t h inhibitors, b u t still considerably more sensitive than a purified/5'-polymerase. These results indicate that it is u n likely that D N A polymerase/3' is involved in Ad5 D N A synthesis in vitro, although small
a m o u n t s of this enzyme m a y be present a m o n g the n o n - t e m p l a t e - b o u n d D N A polymerase.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:20:16
Short communications
235
However, it should also be considered that other enzymes involved in DNA synthesis
containing a functional sulphydryl group could also be inhibited under these conditions.
The results presented here amplify previous studies on this problem and indicate that
within the limits of this analysis the major DNA polymerase activity present in partiallypurified Ad5 DNA replication complexes was the 7-polymerase, although the participation
of other polymerases could not be completely excluded. It is possible that a complex could
exist containing both ~- and y-polymerases, in which each enzyme was dependent on the
other and inhibition of one enzyme would inhibit the other. If this were so, treatment with
micrococcal nuclease which could conceivably dissociate such a complex should reveal the
two activities. It is also worth noting in the light of the method used to prepare the replication complex that the y-polymerase appears to bind more strongly to DNA than the ~- or
fl-polymerase under high salt conditions (B. Otto, personal communication). Since the
replication complexes described here appear to elongate DNA chains presumably initiated
in vivo (Shaw, 1979) it is probable that certain components essential for complete adenovirus DNA replication are removed during their preparation. If one of these components
were DNA polymerase ~, this could account for the relative resistance of endogenous DNA
synthesis in the $2oo fraction to ddTTP and also the finding that aphidicolin, a specific
inhibitor of this enzyme, inhibits adenovirus DNA synthesis in intact cells (Longiaru et al.
I979).
The results presented in this report are in agreement with those of Van der Vliet& Kwant
(I978), who showed that ddTTP inhibited adenovirus DNA synthesis in isolated nuclei.
On the other hand, Abboud & Horwitz (1979), analysing replication complexes prepared
from a fraction similar to the $2oo fraction, detected more ~-polymerase than 7-polymerase.
It should be noted, however, that these workers separated the replication complexes by
sedimentation on to a sucrose cushion and thus could not exclude contamination by other
aggregates in this fraction.
The function of the 7-polymerase is not known for certain, although it does seem to be the
sole DNA polymerase detectable in mitochondria (Bolden et al. 1977). The mitochondrial
genome, although circular (Borst & Grivell, t978), has several features in common with
adenovirus DNA. The genome is not complexed with cellular histones (D. Williamson,
personal communication) and DNA replication proceeds via a displacement mechanism
(Robberson et al. I972). A further interesting observation is that adenovirus infection
inhibits host nuclear DNA synthesis (Mantyjarvi & Russell, 1969)but not mitochondrial
DNA synthesis (Fisher & Horwitz, I977). Although the results presented in this report
may not be applicable to the in vivo situation, which is certainly m o r e complex, they suggest
that DNA polymerase 7 is intimately involved in the replication of adenovirus DNA.
Division of Virology
National Institute f o r Medical Research
Mill Hill, London N W 7 I A A .
C . H . SHAW*
D . M . REKOSHt
W . C . RUSSELL
REFERENCES
ABBOUD,M. M. &HORWITZ,M. S. 0979). The DNA polymerase associated with the adenovirus type z replication complex: effect of 2"3" dideoxythymidine 5" triphosphate on viral DNA synthesis. Nacleic Acids
Research 6, lO25-1o39.
ARENs~ M., YAMASHITA, T.~ PADMANAI~HAN,R., TSURUO,T. & GREEN,M. (I977). Adenovirus deoxyribonucleic
acid replication. Characterization of the enzyme activities of a soluble replication complex. Journal
of Biological Chemistry 252, 7947-7954.
* Present address: Max-Planck-lnstitut fur Zuchtungsforschung, 5 Koln-Vogelsang, D5ooo Koln 30,
Germany.
"i" Present address: Department of Biochemistry, Schools of Medicine and Dentistry, State University of
New York at Buffalo, lo2 Cary Hall, Buffalo, New York 14214, U.S.A.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:20:16
236
Short communications
BOLDEN, A., PEDRAH NOV, G. & WEISSBACH,A. (1977)- D N A polymerase of mitochondria is y polymerase.
Journal of Biological Chemistry 252, 3351-3356.
BORST, P. & GRIVELL, L. A. (5978). The mitochondrial genome of yeast. Cell xS, 705-724.
BRISON, O., KEDINGER, C. & WILHELM, J. 0977)- Enzymatic properties o f viral replication complexes isolated
from adenovirus type 2 infected HeLa cell nuclei. Journal of Virology 24, 423-435EDENBERG, H. J., ANDERSON, S. & DEPAMPITILIS,M. L. (I978). lnvolvement of D N A polymerase ~ in simian
virus 4o D N A replication. Journal of Biological Chemistry 253, 3273-328o
EISNER, P. B. & HORWtTZ, M. S. (t977). Mitochondrial D N A synthesis in adenovirus type z infected HeLa
cells. Journal of Vb'ology 22, 340-345 ,
ERENKEL, G. D. (I978). Adenovirus D N A synthesis in vitro in an isolated complex. Journal of Virology 25,
459-463.
fro, K., ARENS, M. & GREEN, M. (1976). Characterization of D N A polymerase associated with nuclear membrane fractions from uninfected and adenovirus 2 infected KB cells. Biochimica et Biophysica Acta 447,
34o-352 .
de soY~, A., VAN der VLIET, ~'. C. & JANSZ, n. s. 0977). D N A polymerases in adenovirus type 5 infected and
uninfected KB cells. Induction o f an ~ type D N A polymerase in adenovirus type 5 infected fast growing
cells. Biochimica et Biophysica Acta 476, 156-I 65.
LONGIARU, M., IKEDA, J.-E., JARKOVSKY, Z., HORWITZ, S. B. & HORWITZ, M. S. (I979). The effect of aphidicolin
on adenovirus D N A synthesis. Nucleic Acids Research 6, 3369-3386.
MANTYJARVI, R. & RUSSELL,W. C. (I969). Studies on D N A synthesis in cells infected with adenovirus. Journal
of General Virology 5, 339-35o.
MATSUKAGE, A., BonN, E. W. & WILSON, S. H. (I975). Differential sensitivity of low molecular weight D N A
polymerase to sulphydryl blocking agents. Biochimica et Biophysica Acta 383, 338-343.
PELHAM, H. R. B. & JACKSON, R. J. 0976). A n efficient m R N A dependent translation system from reticulocyte
lysates. European Journal of Biochemistry 67, 247-256.
ROBBERSON, D. L., KASAMATSU, M. & VINOGRAD, J. (I972). Replication o f mitochondrial D N A . Circular
replication intermediates in mouse L cells. Proceedings of the National Academy of Sciences of the
United States of America 69, 737-741.
SHAW, C. H. (I979). A study of adenovirus D N A replication. Ph.D. thesis, C N A A , London.
SHAW, C. H., REKOSH, D. M., RUSSELL, W. C. & BLAIR, G. E. 0978). Adenovirus D N A synthesis in a soluble
fraction isolated from infected human cells. Biochemical Society Transactions 6, 752-753.
SHAW, C. n., RUSSELL, W. C. & REKOSH, D. M. 0979). Association of adenovirus early proteins with a nuclear
fraction that synthesises D N A in vitro. Virology 92, 436-448.
VAN der VLIET, P. C. & KWANT, M. M. (1978). Role of D N A polymerase y in adenovirus D N A replication.
Nature, London 276, 532-534.
WANG, T. S.-F., SEDWICK, t~. & KORN, O. (t974). Nuclear deoxyribonucleic acid polymerase. Purification and
properties Of the homogeneous enzyme from human KB cells. Journal of Biological Chemistry 249,
84[-85o.
WAOAR, M. A., EVANS, M.J. & HUBERMAN, J. A. (1978). Effect of 2'3' dideoxythymidine 5' triphosphate on
HeLa cells in vitro D N A synthesis: evidence that D N A polymerase c~ is the only polymerase required
for cellular D N A replication. Nncleic Acids Research 5, I933-I946WEISSBACFI, A. (I977). Eukaryotic D N A polymerases. Annual Review of Biochemistry 46, 25-47.
WEISSBACH, A., SCHLABACH, A., FRIDLENDER, B. & BOLDEN. A. (197I). D N A polymerases from human cells.
Nature New Biology 23x, 167-170.
WILHELM, J., BRISON, O., KEDINGER, C. & CHAMBON, P. 0976). Characterization of adenovirus type 2 transcriptional complexes isolated from infected HeLa cell nuclei. Journal of Virology x9, 6~-8I.
(Received 8 October r 9 7 9 )
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 13:20:16