Initiation of JC virus DNA replication
in vitro
by human and mouse
DNA polymerase a-primase
Richard W. P. Smith
1,
* and Heinz-Peter Nasheuer
1,2
1
Abteilung Biochemie, Institut fu
¨
r Molekulare Biotechnologie, Jena, Germany;
2
National University of Ireland, Galway,
Department of Biochemistry, Galway, Ireland
Host species specificity of the polyomaviruses simian virus
40 (SV40) and mouse polyomavirus (PyV) has been shown
to be determined by the host DNA polymerase a-primase
complex involved in the initiation of both viral and host
DNA replication. Here we demonstrate that DNA repli-
cation of the related human pathogenic polyomavirus JC
virus (JCV) can be supported in vitro by DNA polymerase
a-primase of either human or murine origin indicating that
the mechanism of its strict species specificity differs from
that of SV40 and PyV. Our results indicate that this may
be due to differences in the interaction of JCV and SV40
large T antigens with the DNA replication initiation
complex.
Keywords: DNA replication; initiation; DNA polymerase
a-primase; species specificity; polyomavirus.
Polyomavirus DNA replication has served as a model

DNA polymerase a-primase consists of four subunits with
apparent molecular masses of 180, 68, 58 and 48 kDa of
which the largest and smallest subunits are a DNA
polymerase and a primase, respectively [28–30]. SV40
DNA replication in vitro was recently shown to require a
functional interaction between the SV40 TAg and the
C-terminus of the p180 subunit of human DNA poly-
merase a-primase [26].
The genome of JCV is 69% homologous to that of SV40
and expresses an analogous set of proteins [31]. The core
origins of replication of the two viruses are also conserved to
such an extent that SV40 TAg, which is 72% identical to
JCV TAg, can efficiently support JCV DNA replication
in vivo and in vitro [32,33]. The high level of conservation
between these two primate specific viruses coupled with the
fact that JCV DNA replication is inhibited in a nonpermis-
sive host in vivo would imply that the restricted host range of
JCV is due to a requirement for human DNA polymerase
a-primase, as is the case with SV40 [8]. Previously we
reported the establishment of a cell-free system for JCV
DNA replication [32]. With this system we were able to
reproduce many features of JCV DNA replication found
in vivo, such as sequence requirements at the origin of
replication and the requirement for JCV or SV40 (but not
PyV) TAg for efficient replication. Therefore, we applied
this system to the question of host specificity regulation and
found that this differs from that of SV40 in that it is not
determined at the level of initiation of DNA replication by
DNA polymerase a-primase in vitro. This appears to be due
to differences in the interaction of the JCV and SV40 large

Munich). DNA polymerase a and DNA primase assays
were performed as previously described [34,42,43].
Preparation of S100 extracts and replication
of SV40 and JCV
in vitro
S100 extracts were prepared from logarithmically growing
FM3A cells as previously described [27,34]. Cells were
harvested by centrifugation, then washed twice with phos-
phate buffered saline (NaCl/P
i
) and once with hypotonic
buffer. The cells were resuspended in hypotonic buffer,
incubatedfor10minonice,andbrokenby12strokesina
Dounce homogenizer. The extracts were centrifuged at 4 °C
and 11 000 g. The supernatant was then adjusted to
100 m
M
NaCl and clarified by a second centrifugation at
100 000 g (S100 extract). Depletion of DNA polymerase
a-primase from S100 extracts was performed essentially as
previously described [22,27,32].
The replication of SV40 and JCV DNA in vitro was
performed as previously described [22,27,32]. Briefly, the
assay contained 0.6 lg SV40 or JCV TAg, 250 ng of pUC-
HS or pJC389 or pJC433 DNA (carrying the replication
origin of SV40 or JCV, respectively [21,32]), and 200 lg
S100 in 30 m
M
Hepes/NaOH (pH 7.8), 1 m
M

counting. The total radioactivity was measured after
spotting 5 lL of a 200-fold dilution of the replication assay
onto GF52 filters (Schleicher & Schu
¨
ll, Dassel, Germany).
EcoRI and DpnI digestion of product DNA was carried out
as described by Kautz et al. [23].
Initiation of replication on JCV DNA
Initiation reactions were performed essentially as previously
described [22,32,44,45]. Briefly, the JCV initiation assay
(40 lL) was assembled on ice and contained 0.25 lg
pJC389 (carrying the JCV replication origin), 0.6 lgJCV
TAg, and 0.5 lgRPA,in30m
M
Hepes/KOH (pH 7.8),
7m
M
magnesium acetate, 1 m
M
EGTA, 1 m
M
dithiothre-
itol, 0.2 m
M
UTP, 0.2 m
M
GTP, 0.01 m
M
CTP, 4 m
M

65 °C for 30 min, heated for 3 min at 95 °C, and electro-
phoresed in denaturing 20% polyacrylamide gels for 3–4 h
at 600 V as described previously [22]. The reaction products
were visualized by autoradiography and quantified with a
phosphoimager (Amersham Biosciences).
JCV and SV40 monopolymerase systems
These assays were adapted from Ishimi et al.[11].The
monopolymerase assay (40 lL) was assembled on ice and
contained 0.5 lg pUC-HS (carrying the SV40 replication
origin) or 0.5 lg pJC433 (carrying the JCV replication
origin [32]), 1 lg SV40 or JCV T antigen, 1.4 ng topoiso-
merase I and 0.5 lgRPA,in30m
M
Hepes/KOH (pH 7.8),
7m
M
magnesium acetate, 0.1 m
M
EGTA, 1 m
M
dithio-
threitol, 0.2 m
M
UTP, 0.2 m
M
GTP, 0.2 m
M
CTP, 4 m
M
ATP, 20 m

Results
Replication of JCV DNA in crude cell extracts with
recombinant human and murine DNA polymerase
a-primase
Previously we reported the establishment of in vitro systems
for the replication of JCV DNA, either using crude cell
extracts or purified proteins only [32]. Both these systems
were dependent upon recombinant JCV TAg and the
presence of a JCV origin of DNA replication. Here we
applied these systems to study the dependence of JCV DNA
replication on human replication proteins. Figure 1 repre-
sents a comparison between the SV40 (panel A) and JCV
(panel B) cell-free DNA replication systems. Figure 1B
shows a DNA replication assay using mouse FM3A S100
crude cell extracts depleted of DNA polymerase a-primase
supplemented with JCV TAg, a plasmid carrying the JCV
Ó FEBS 2003 Initiation of DNA replication (Eur. J. Biochem. 270) 2031
replication origin (pJC389) and either recombinant human
or murine DNA polymerase a-primase expressed using the
baculovirus system (Fig. 1B, columns 1–7). The replication
activity of JCV TAg in mouse cell extracts is not dependent
on the sequence of the plasmid as the incorporation of
radioactive dNMPs was the same whether the plasmid
pJC389 or pJC433 was used in the cell-free replication assay
(data not shown).
In parallel, we show an SV40 DNA replication assay
using the same cell extracts but with SV40 TAg and a
plasmid (pUC-HS) that carries the SV40 replication origin
(Fig. 1A, columns 1–7 [26,27]). As we showed previously,
SV40 DNA replication is absolutely dependent upon human

purified from Escherichia coli and is therefore fully methy-
lated. However, one or more full rounds of replication will
result in hemimethylated or unmethylated products and will
consequently lead to DpnI resistance which is indeed
observed after replication of JCV DNA either with human
or with murine DNA polymerase a-primase (Fig. 2B, lanes
4 and 6). The lack of species specificity we observed was
reproducible with various independently expressed and
purified batches of JCV TAg (data not shown) and with
various batches of the template DNAs pJC389 and pJC433
([32]; Figs 1 and 2). As expected murine DNA polymerase
a-primase did not support SV40 DNA replication (Fig. 2A,
lanes 5 and 6).
JCV DNA replication with purified proteins
We note that in the cell-free system JCV DNA replication is
markedly less efficient when driven by murine compared
with human DNA polymerase a-primase. In order to
Fig. 1. DNA replication assays in murine FM3A cell extracts depleted of DNA polymerase a-primase using the SV40 (A) and JCV (B) systems.
(A) The SV40 system makes use of SV40 TAg and pUC-HS template DNA containing the SV40 origin of DNA replication. (B) The JCV system
uses JCV TAg and pJC389 DNA with the JCV origin. The cell extracts were supplemented with 0.5 and 1.0 DNA polymerase units of the indicated
DNA polymerase a-primase complexes (H
4
, human heterotetramer; M
4
, murine heterotetramer; MH
3
, murine p180 with human p68, p58 and p48).
Enzyme activities were determined beforehand with a DNA polymerase assay on activated calf thymus DNA. (A) and (B) column 1, TAg omitted;
column 8, pJC389I-/II-, containing a disabled JCV replication origin [32], was used as template for human DNA polymerase a-primase with SV40
(A) and JCV TAg (B), respectively. (B) column 9, pUC-Py1, containing the PyV replication origin [21], used with human DNA polymerase

assay (Fig. 1), which suggests that steps in JCV DNA
replication subsequent to primer formation may be slightly
inhibited by the murine enzyme complex.
SV40 TAg confers species specificity to JCV origin
dependent DNA replication
It has been reported that SV40 TAg is capable of supporting
JCV DNA replication both in vivo and in vitro [32,33,47].
Therefore, we asked whether substitution of SV40 TAg for
JCV TAg would render JCV DNA replication species-
specific with regard to the nature of the DNA polymerase
a-primase complex catalysing the reaction, as is the case
Fig. 2. DNA synthesis products of the SV40
and JCV DNA replication systems. Murine
FM3A cell extracts depleted of DNA poly-
merase a-primase were supplemented with 1.0
DNApolymeraseunitsofH4orM4ornot
supplemented (–Pol). One-quarter of the
DNA synthesis products were analysed for
complete DNA replication by digestion with
EcoRI and DpnI (even numbered lanes). In
parallel, the products were linearized with
EcoRI (odd numbered lanes). The positions
of linearized template DNAs are indicated
by arrows.
Fig. 3. Autoradiogram of an in vitro JCV DNA replication initiation
assay with 0.2 and 0.4 units of primase of either human (H4) or murine
(M4) DNA polymerase a-primase complexes. Specific primase activities
were determined beforehand with a primase assay on poly (dT). Lanes
1 and 2, control reaction with DNA polymerase a-primase lacking
TAg or vice versa; lanes 3 and 4, 0.2 U and 0.4 U of human; lanes 5

Importantly, we show that nucleotide incorporation in
the JCV system by murine DNA polymerase a-primase is
dependent upon the JCV DNA replication origin (Fig. 1B)
and results in DpnI-resistant products (Fig. 2), indicating
that it is due to bona fide DNA replication and not a
consequence of Ôfilling inÕ of gaps or other short patch repair
events. The fact that we observe a complete round of
plasmid replication in murine cell extracts indicates that
other essential replication proteins, such as DNA poly-
merases d and e, proliferating cell nuclear antigen (PCNA),
replication factor C (RF-C), topoisomerase I and DNA
ligase are not responsible for JCV species specificity in vitro.
Our murine FM3A cell extracts contain relatively low levels
of endogenous RPA and are therefore supplemented with
human RPA. However, if this is left out we nevertheless
observe significant, albeit overall less efficient, incorporation
by both murine and human DNA polymerase a-primase
(data not shown) indicating that RPA also is not a species-
specific factor.
In apparent contradiction of our results, Feigenbaum
et al. [8] showed that cultured nonpermissive hamster glial
cells were unable to replicate transfected JCV DNA. Their
observation and our data could be reconciled if the murine
but not the hamster cellular DNA replication machinery
were permissive for JCV DNA replication. We consider this
unlikely. A more likely reason for the discrepancy between
Feigenbaum’s data and our own is the difference in the
ÔstateÕ of the DNA in the two assays. Transfected DNA will
become associated with histones to form chromatin in the
cell nucleus whereas our in vitro assays are carried out with

experiments in (A) and (B) were performed in
parallel.
2034 R. W. P. Smith and H. P. Nasheuer (Eur. J. Biochem. 270) Ó FEBS 2003
can extend the host range of JCV replication in vivo [52].
Alternatively, rodent, but not human, chromatin might
contain a nondiffusable factor that inhibits JCV TAg-
dependent DNA replication or the phosphorylation of JCV
TAg is different in human and mouse cells interfering with
the replication activity in vivo but not with the purified
baculovirus-expressed protein. This explanation is consis-
tent with findings that specific residues of SV TAg must be
phosphorylated whereas other may not be [17].
In summary, the available evidence strongly suggests
that, although DNA replication is the species-specific
process common to JCV, SV40 and PyV, different host
factors in each case ultimately determine the restriction of
virus propagation to a particular host. For SV40 and PyV
these are, respectively, the p180 and p48 subunits of DNA
polymerase a-primase in initiation of DNA replication
[22–24,26,27,53]. For JCV a different level of control
appears to be in operation although we cannot rule out
that the lower initiation efficiency of murine DNA
polymerase a-primase is at least in part involved when
compounded by other factors not present in our assays.
This view is supported by the recent finding that the JC
virus receptor is widely distributed on cells which contrasts
with the cellular restriction of virus propagation [54].
Factors involved in this regulation might be cellular
proteins such as puralpha, p53, or alternative splicing
products of viral proteins [51,55,56].

DNA replication. In Recent Research Developments in Virology,2
(Pandalai, S.G., ed.), pp. 67–92. Transworld Research Network,
Trivandrum, India.
3. Berger, J.R. & Concha, M. (1995) Progressive multifocal leu-
koencephalopathy: The evolution of a disease once considered
rare. J. Neurovirol. 1, 5–18.
4. Frisque, R.J. & White, F.A. III (1992) The molecular biology of
JC virus, causative agent of progressive multifocal leukoence-
phalopathy. In Molecular Neurovirology (Ross, R., ed.), pp. 25–
158. Humana Press, Cliffton, NJ, USA.
5. Gallia, G.L., Houff, S.A., Major, E.O. & Khalili, K. (1997)
Review: JC virus infection of lymphocytes – revisited. J. Infect Dis.
176, 1603–1609.
6. Major, E.O. & Ault, G.S. (1995) Progressive multifocal leuko-
encephalopathy: clinical and laboratory observations on a viral
induced demyelinating disease in the immunodeficient patient.
Curr. Opin. Neurol. 8, 184–190.
7. Walker, D.L. & Frisque, R.J. (1986) The biology and molecular
biology of JC virus. In The Papovaviridae, the Polyomaviruses
(Salzman, N., ed.), pp. 327–377. Plenum Press, New York, USA.
8. Feigenbaum, L., Khalili, K., Major, E. & Khoury, G. (1987)
Regulation of the host range of human papovavirus JCV. Proc.
NatlAcad.Sci.USA84, 3695–3698.
9. Tooze, J. (1981) Molecular biology of tumor virus, part 2. DNA
Tumor Viruses. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, New York, USA.
Fig. 5. Cell-free DNA replication assays in murine FM3A cell extracts
depleted of DNA polymerase a-primase. TAgs (SV40 & JCV) and DNA
polymerase a-primase complexes (human & murine) were tested in
parallel experiments for their ability to support DNA replication of a

simian virus 40 large T antigen. Annu.Rev.Biochem.61, 55–85.
18. Hassell, J.A. & Brinton, B.T. (1996) SV40 and polyomavirus
DNA replication. In DNA Replication in Eukaryotic Cells
(DePamphilis, M.L., ed.), pp. 639–677. Cold Spring Harbor
Laboratory Press, Cold Spring Harbor Laboratory, NY, USA.
19. Eki, T., Enomoto, T., Masutani, C., Miyajima, A., Takada, R.,
Murakami, Y., Ohno, T., Hanaoka, F. & Ui, M. (1991) Mouse
DNA primase plays the principal role in determination of
permissiveness for polyomavirus DNA replication. J. Virol. 65,
4874–4881.
20.Murakami,Y.,Wobbe,C.R.,Weissbach,L.,Dean,F.B.&
Hurowitz,J.(1986)RoleofDNApolymerasea and DNA pri-
mase in simian virus 40 DNA replication in vitro. Proc. Natl Acad.
Sci. USA 83, 2869–2873.
21. Schneider, C.K., Weißhart, L.A., Guarino, I. & Dornreiter &
Fanning, E. (1994) Species specific functional interactions of DNA
polymerase a-primase with SV40 T antigen require SV40 origin
DNA. Mol. Cell. Biol. 14, 3176–3185.
22. Bru
¨
ckner, A., Stadlbauer, F., Guarino, L.A., Brunahl, A.,
Schneider, C., Rehfuess, C., Prives, C., Fanning, E. & Nasheuer,
H P. (1995) The mouse DNA polymerase a-primase subunit p48
mediates species-specific replication of polyoma virus DNA
in vitro. Mol. Cell. Biol. 15, 1716–1724.
23. Kautz, A., Schneider, A., Weisshart, K., Geiger, C. & Nasheuer,
H.P. (2001) Different regions of primase subunit p48 control
mouse polyomavirus and simian virus 40 DNA replication in vitro.
J. Virol. 75, 1751–1760.
24. Kautz, A., Weisshart, K., Schneider, A., Grosse, F. & Nasheuer,

tions of eukaryotic DNA polymerases. Sci. SAGE KE 2003, re3
(26 February 2003), />sageke;2003/8/re3
31. Frisque, R.J., Bream, G.L. & Cannell, M.T. (1984) Human
polyomavirus JC virus genome. J. Virol. 51, 458–469.
32. Nesper,J.,Smith,R.W.P.,Kautz,A.R.,Sock,E.,Wegner,M.,
Grummt, F. & Nasheuer, H.P. (1997) A cell-free replication
system for human polyomavirus JC DNA. J. Virol. 71, 7421–7428.
33. Sock, E., Wegner, M., Fortunato, E.A. & Grummt, F. (1993)
Large T-antigen and sequences within the regulatory region of JC
virus both contribute to the features of JC virus DNA replication.
Virology 197, 537–548.
34. Stadlbauer, F., Brueckner, A., Rehfuess, C., Eckerskorn, C.,
Lottspeich, F., Fo
¨
rster,V.,Tseng,B.Y.&Nasheuer,H P.(1994)
DNA replication in vitro by recombinant DNA polymerase
a-primase. Eur. J. Biochem. 222, 781–793.
35. Stigger, E., Dean, F.B., Hurwitz, J. & Lee, S.H. (1994) Recon-
stitution of functional human single-stranded DNA-binding pro-
tein from individual subunits expressed by recombinant
baculoviruses. Proc. Natl Acad. Sci. USA 91, 579–583.
36. Tanaka, S., Hu, S Z., Wang, T.S F. & Korn, D. (1982) Pre-
paration and preliminary characterization of monoclonal anti-
bodies against human DNA polymerase a. J. Biol. Chem. 257,
8386–8390.
37. Henricksen, L.A., Umbricht, C.B. & Wold, M.S. (1994)
Recombinant replication protein A: expression, complex
formation, and functional characterization. J. Biol. Chem. 269,
11121–11132.
38. Nasheuer, H P., von Winkler, D., Schneider, C., Dornreiter, I.,

Cloning, 2nd Edn. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, New York, USA.
2036 R. W. P. Smith and H. P. Nasheuer (Eur. J. Biochem. 270) Ó FEBS 2003
47. Lynch, K.J. & Frisque, R.J. (1990) Identification of critical ele-
ments within the JC virus DNA replication origin. J. Virol. 64,
5812–5822.
48. Cheng, L.Z., Workman, J.L., Kingston, R.E. & Kelly, T.J. (1992)
Regulation of DNA replication in vitro by the transcriptional
activation domain of GAL4-VP16. Proc. Natl Acad. Sci. USA 89,
589–593.
49. Cheng, L. & Kelly, T.J. (1989) Transcriptional activator nuclear
factor I stimulates the replication of SV40 minichromosomes
in vivo and in vitro. Cell. 59, 541–551.
50. Sock, E., Wegner, M. & Grummt, F. (1991) DNA replication of
human polyomavirus JC is stimulated by NF-1 in vivo. Virology
182, 298–308.
51. Staib, C., Pesch, J., Gerwig, R., Gerber, J.K., Brehm, U., Stangl,
A. & Grummt, F. (1996) p53 inhibits JC virus DNA replication
in vivo and interacts with JC virus large T-antigen. Virology. 219,
237–246.
52. Vacante, D.A., Traub, R. & Major, E.O. (1989) Extension of JC
virus host range to monkey cells by insertion of a simian virus 40
enhancer into the JC virus regulatory region. Virology 170, 353–
361.
53. Smith, R. & Nasheuer, H. (2002) Control of complex formation of
DNA polymerase alpha-primase and cell-free DNA replication by
the C-terminal amino acids of the largest subunit p180. FEBS Lett.
527, 143–146.
54. Suzuki, S., Sawa, H., Komagome, R., Orba, Y., Yamada, M.,
Okada, Y., Ishida, Y., Nishihara, H., Tanaka, S. & Nagashima,