2627
Journal of Cell Science 109, 2627-2636 (1996)
Printed in Great Britain © The Company of Biologists Limited 1996
JCS4281
Identification of the nuclear localization signal of mouse DNA primase:
nuclear transport of p46 subunit is facilitated by interaction with p54 subunit
Takeshi Mizuno1, Tomoko Okamoto2,3, Masayuki Yokoi2, Masako Izumi1, Akio Kobayashi4,
Takahisa Hachiya4, Katsuyuki Tamai4, Tadashi Inoue3 and Fumio Hanaoka1,2,*
1The Institute of Physical and Chemical Research (RIKEN), Wako, Saitama 351-01, Japan
2The Institute for Molecular and Cellular Biology, Osaka University, Suita, Osaka 565, Japan
3College of Agriculture and Veterinary Medicine, Nihon University, Fujisawa, Kanagawa 252,
4Medical and Biological Laboratories Co., Ltd, Ina, Nagano 396, Japan
Japan
*Author for correspondence
SUMMARY
DNA polymerase α-primase is a replication enzyme
necessary for DNA replication in all eukaryotes. Mouse
DNA primase is composed of two subunits: a 46 kDa
protein (p46), which is the catalytic subunit capable of
RNA primer synthesis, and a 54 kDa protein (p54), whose
physiological role is not clear. To understand the structurefunction relationship of DNA primase, we set out to characterize these two subunits individually or in combination
using a cDNA expression system in mammalian cultured
cells, and determined the subcellular distribution of ectopically expressed DNA primase. The p54 expressed in COS1 cells after transfection was predominantly localized in the
nucleus, whereas p46 was retained in the cytoplasm as
shown by indirect immunofluorescence analysis. Using
several mutant proteins with deletions or substitutions as
well as chimeric constructs, we identified the nuclear local-
ization signal of p54 as RIRKKLR, encoded near the amino
terminus (residues 6-12). Furthermore, co-expression of
both p46 and p54 subunits markedly altered the subcellular distribution of p46; co-expressed p46 was transported
into the nucleus as efficiently as p54. These results demonstrate that p54 has a nuclear localization signal and is able
to be translocated into the nucleus independently of DNA
polymerase α subunits. In contrast, p46 lacks a nuclear
localization signal, and its nuclear translocation is facilitated by interaction with p54. We present here first
evidence for a novel role of p54 in the nuclear translocation process, and a piggy-back binding transport
mechanism of mouse DNA primase.
INTRODUCTION
system (Copeland and Wang, 1991). p68 has no known
catalytic activity. However, p68 was shown to contribute to
binding of the SV40 large T antigen with DNA polymerase αprimase in human cells (Collins et al., 1993). The others, p54
and p46, form a tightly bound complex and can be resolved
from p180 and p68 only after treatment with 50% ethylene
glycol (Suzuki et al., 1985, 1989). This resolved complex of
p54 and p46 exhibits primase activity to synthesize unit length
RNA primers, showing that DNA primase is composed of
these two subunits (Kaguni et al., 1983; Suzuki et al., 1985,
1989; Nasheuer and Grosse, 1988; Wang, 1991). From recent
studies in yeast (Santocanale et al., 1993), bovine (Nasheuer
and Grosse, 1988), Drosophila (Bakkenist and Cotterill, 1994),
and mouse cells (Stadlbauer et al., 1994), it has been shown
that p46 is a catalytic subunit of DNA primase which is capable
of initiation and elongation of primer RNA. However, studies
of mouse DNA primase expressed in bacteria demonstrated
that while only p46 is necessary for elongation, both subunits
are required for initiation (Copeland and Wang, 1993;
Copeland and Tan, 1995). Whatever the requirements for the
initiation stage, it is certain that p46 is a catalytic subunit of
RNA primer synthesis. In contrast to p46, limited information
In eukaryotic cells, chromosomal DNA replication requires a
large number of replication factors including three types of
DNA polymerases, α, δ, and ε. Among these factors, DNA
polymerase α associates tightly with DNA primase, which synthesizes short oligoribonucleotide primers which are then
further elongated by polymerase α. Therefore, the DNA polymerase α-primase complex is the only enzyme capable of initiating both leading strand synthesis as well as Okazaki
fragment synthesis on the lagging strand (Kaguni et al., 1983;
Wang, 1991; Stillman, 1994).
DNA polymerase α-primase consists of four subunits and its
subunit composition is highly conserved among eukaryotes
including S. cerevisiae, Drosophila, mouse, and humans. The
complex of mouse DNA polymerase α-primase contains four
polypeptides whose apparent molecular masses are 180, 68, 54,
and 46 kDa (Takada-Takayama et al., 1990a,b). According to
recent studies using cDNA cloning and characterization,
functions for each subunit have been suggested. p180, the
largest subunit, is a catalytic subunit and possesses intrinsic
polymerase activity as revealed in a baculovirus expression
Key words: DNA polymerase α-primase, DNA primase, Nuclear
localization signal, Piggy-back binding transport
2628 T. Mizuno and others
is available concerning the function of p54, and enzymatic
activity, if any, has not been found. To date, several roles of
p54 have been suggested: (i) p54 may stabilize the thermolabile nature of the catalytic activity of p46 (Copeland and
Wang, 1993; Santocanale et al., 1993; Stadlbauer et al., 1994);
(ii) p54 may contribute to the interaction of p46 with p180
(Copeland and Wang, 1993; Longhese et al., 1993); and (iii)
p54 may have affinity for ribonucleotide triphosphate and be
intimately associated with the NTP binding region (Foiani et
al., 1989; Copeland and Wang, 1993). However, the precise
physiological as well as cell biological roles of p54 remain to
be clarified.
To understand the structure-function relationship of DNA
polymerase α-primase, we purified DNA polymerase αprimase from mouse FM3A cells, cloned cDNAs for all
subunits, and determined their primary structure (TakadaTakayama et al., 1990a,b; Miyazawa et al., 1993). In the course
of analyzing a temperature-sensitive mutant which has a
defective DNA polymerase α, we constructed a cDNA overexpression system in mammalian cultured cell lines (Izumi et
al., 1994). To characterize DNA primase subunits, we overexpressed these cDNAs individually or in combination. Unexpectedly, we found that the subcellular distributions of the two
subunits of ectopically expressed DNA primase were quite
different. Here we demonstrated by indirect immunofluorescence analysis that overexpressed p54 in COS-1 cells is
translocated to the nucleus exclusively, whereas overexpressed
p46 was retained in the cytoplasm. Using deletion and point
mutation analysis, we identified the nuclear localization signal
(NLS) of p54 and showed that this sequence, RIRKKLR, corresponding to residues 6-12 is necessary and sufficient for
targeting of p54 to the nucleus. Finally, when both subunits of
primase were co-expressed in COS-1 cells, p46 was transported to the nucleus as efficiently as p54, indicating that p54
facilitates nuclear translocation of p46. These results represent
the first evidence for a novel cellular role of p54 in the nuclear
translocation pathway.
MATERIALS AND METHODS
Materials
All restriction enzymes and Klenow fragment were purchased from
Takara (Japan); phenylmethylsulphonyl fluoride was from Sigma; Pfu
DNA polymerase was from Stratagene; horseradish peroxidase-conjugated anti-rabbit and anti-mouse IgG antibodies were from MBL
(Japan); FITC- or Texas red-conjugated anti-rabbit or anti-mouse IgG
antibodies were from Vector Inc; anti-β-galactosidase and anti-HA
(12CA5) monoclonal antibodies were from Boehringer Mannheim;
fetal bovine serum was from Nipro (Japan); calf serum was from
Hyclone. The expression vector pcDEB∆ was a gift from Dr Y.
Nakabeppu (Nakabeppu et al., 1993). Unless otherwise noted, all
other chemicals and reagents were obtained from Wako Chemicals
(Japan).
Construction of expression vector
Mouse p46 and p54 cDNAs were introduced into pcDEB∆ which
contains the SRα promoter (Takebe et al., 1988; Nakabeppu et al.,
1993). The 1.9 kb HindIII fragment of p54 (Miyazawa et al., 1993)
was inserted into the HindIII site of pcDEB∆. The 1.3 kb EcoRIEcoT14I fragment of p46 cDNA (Miyazawa et al., 1993) was filled
in, ligated to HindIII linkers, and subcloned into the HindIII site of
pcDEB∆. The resultant plasmids were designated pSRα54 and
pSRα46, respectively.
For construction of carboxyl-terminal truncation mutants, a stop
codon and restriction enzyme sites were introduced by polymerase
chain reaction (PCR) with the amino-terminal primer 5′-CAGGTGGTCTAGATGCAGTTC-3′ and different carboxyl-terminal primers,
5′-CATGATATCCTATTCCCGAAGG-3′ for p54∆164-505 and 5′CTCCTGCTTCTAGAACTACAATGCCTGC-3′ for p54∆325-505.
The PCR products for p54∆164-505 was then digested with XbaI and
subcloned into XbaI-digested pcDEB∆. The PCR product for
p54∆325-505 was digested with XbaI and EcoRV and subcloned into
XbaI-EcoRV-digested pcDEB∆. Amino-terminal deletion mutants
were constructed by cleaving with naturally occurring unique restriction enzyme sites, BglII, PvuII, and DraI. p54∆1-306 was constructed
by subcloning the 855 bp PvuII-DraI fragment of p54 cDNA into the
blunt-ended HindIII site of pcDEB∆. p54∆1-124 and p54∆1124∆325-505 were constructed by removing the 400 bp BglII
fragment from pSRα54 or p54∆325-505, respectively, followed by
self-ligation. Mutants with substitutions were constructed by PCR
using the overlap extension technique (Ito et al., 1991). The following
primer sets were used to introduce mutations; p54RKK(8-10)QNN:
5′-GGAAGGATCCAGAACAATCTGCGATTG-3′ and 5′-CTTTGGTAAAATACTGAA-3′; p54RRR(96-98)QQQ: 5′-TATGAGCCACAGCAACAGGACCACATC-3′ and 5′-AGACTTTCCTTCCTCTGA-3′. The p54:p46 fusion constructs were produced by
introducing either the 130 bp HindIII-PstI fragment or 380 bp
HindIII-BglII fragment of p54 cDNA into the HindIII site of
pSRα46. The p54:β-galactosidase fusion clones were constructed
using the vector pCH110 (Pharmacia) and the fragments amplified by
PCR. The 1.0 kb EcoRI-EcoRV fragment of pcDEB∆ containing the
SRα promoter was digested, blunt-ended and subcloned into the
blunt-ended HindIII site of pCH110 vector to produce pSRα-β-galactosidase. The fragments encoding the amino-terminal region of p54
were synthesized by PCR to introduce a KpnI site. The PCR fragments
were then digested with HindIII and KpnI, and inserted into
HindIII-KpnI-digested pSRα-β-galactosidase. Oligonucleotides used
for β-galactosidase fusion constructs were as follows; SRα sense:
5′-TCCAAGCTCCTCGAGGAACT-3′; p541-14:βgal antisense:
5′-CTCTGGTCAGGTACCAATCGCAG-3′; p541-35:βgal antisense: 5′-AACGATATGGGTACCGTAGGCGGCT-3′; p541-91:βgal
antisense: 5′-GGCTCATACGGTACCTCCAGGTT-3′; p541-115:βgal
antisense:
5′-CATCGTCTAGGTACCTCCGACTGGC-3′.
For
double staining, a hemagglutinin (HA) tag was introduced into the
amino terminus of pSRα54 and p54RKK(8-10)QNN by PCR using
the primers 5′-GAAGATCTAGAATGTA-CCCATACGACGTGCCTGACTATGCCGGCCAGTTCTCAGGA-AGG-3′ and 5′-CTCCTGCTTCTAGAACTACAATGCCTGC-3′. The PCR products were
digested with BglII and then subcloned into BglII-digested pSRα54
and p54RKK(8-10)QNN, respectively. The identity of all constructs
was confirmed by double-stranded sequencing using ALF DNA
Sequencer (Pharmacia).
Cell culture and transfection
COS-1 cells were grown in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum in a 5% CO2 incubator.
NIH3T3 cells were grown with 10% calf serum. Transfection was
performed by electroporation as described (Sorimachi et al., 1993).
The state of protein expression was analyzed 24 hours after transfection unless otherwise indicated.
Antibodies
The anti-p54 polyclonal antibody was generated in a rabbit against a
GST fusion protein expressed in Escherichia coli. The 1.8 kb BamHIHindIII fragment of p54 cDNA was inserted into the pGEX2T
expression vector (Pharmacia). The pGEX vector encoding the fusion
protein was introduced into E. coli BL21 cells, and after induction by
0.5 mM isopropyl-1-thio-β-D-galactopyranoside for 4 hours, GST
Nuclear localization signal of mouse DNA primase 2629
fusion protein precipitated as inclusion bodies was purified and used
as antigen.
The anti-p46 polyclonal antibody was raised in a rabbit against p46
protein derived from insect cells infected with the recombinant baculovirus. The 1.5 kb EcoRI-EcoT14I fragment of p46 cDNA was
blunted by Klenow fragment and subcloned into the SmaI site of
pVL1393 transfer vector (Pharmingen). A recombinant baculovirus
was generated using the Baculogold system (Pharmingen). p46 was
expressed at high levels in Sf9 insect cells at 72 hours after infection
and extracted in the lysis buffer (20 mM Tris-HCl, pH 8.0, 300 mM
NaCl, 1 mM EDTA, 10% glycerol, 0.5% NP-40, 1 mM DTT, and 0.25
mM phenylmethylsulfonyl fluoride) using a Dounce homogenizer.
Insoluble material was removed by centrifugation of the lysate at
10,000 g for 30 minutes at 4°C, then the soluble fraction was applied
to a phosphocellulose column (P11, Whatman) equilibrated with P1
buffer (20 mM Tris-HCl, pH 8.0, 1 mM EDTA, 10% glycerol, and
0.1% NP-40) containing 100 mM NaCl. After the column was washed
with P1 buffer containing 200 mM NaCl, the sample was eluted by P1
buffer containing 300 mM NaCl, then applied directly to a Q
Sepharose column (Pharmacia) equilibrated with P1 buffer containing
300 mM NaCl. Flow-through fractions were collected, dialyzed, and
subjected to SDS-PAGE. Proteins in the gel were detected by staining
with Coomassie brilliant blue R-250, and the protein band corresponding to p46 was excised, homogenized and used as antigen.
For immunizations, materials were injected into rabbits firstly with
complete Freund’s adjuvant, and subsequently with incomplete
Freund’s adjuvant. Rabbit anti-p46 and anti-p54 antibodies were
prepared by subcutaneous injection of p46 protein purified from insect
cells and GST-p54 fusion protein purified from E. coli, respectively,
as described above. Both anti-p46 and anti-p54 antibodies were
purified using the antigen-immobilized columns as follows.
Immunoglobulin fractions were precipitated from 30 ml of each
antiserum with 50% saturation of ammonium sulfate, followed by
centrifugation, and dissolved into 10 ml of PBS/0.1% NaN3. After
dialysis against the same solution, each protein fraction was applied
on the recombinant p46-conjugated Sepharose or the recombinant
GST-p54-conjugated Sepharose columns. The column volumes were
3 ml each. After extensive washing of the columns with PBS, antibodies were eluted with 0.17 M glycine-HCl, pH 2.3, and neutralized.
In the case of anti-p54 antibodies, the antibody fraction was passed
over a GST affinity column to deplete anti-GST antibodies, and then
an E. coli whole-protein affinity column to remove anti-E. coli protein
antibodies. These antibody fractions were dialyzed against PBS/0.1%
NaN3 and stored at −80°C.
Indirect immunofluorescence staining
Cells grown in chamber slides (Nunc) coated with poly-L-lysine were
washed with PBS and fixed with 3.7% formaldehyde in PBS for 30
minutes on ice. Cells were then washed with PBS and permeabilized
sequentially with 50%, 75%, and 95% EtOH on ice for 5 minutes
each. The slides were then blocked with PBS containing 3% dried
milk (blocking buffer) for 30 minutes at room temperature, incubated
with anti-p46 or anti-p54 antibodies (1.3 µg/ml in blocking buffer)
for 1 hour at room temperature, and washed three times with PBS for
5 minutes each time. Cells were then incubated with FITC-conjugated
secondary antibodies for 1 hour at room temperature, washed three
times with PBS, and preserved in PBS containing 80% glycerol and
0.5% triethylenediamine. DNA staining was performed by adding 1
µg/ml of bis-benzimide (Hoechst 33258) into the final PBS wash. The
samples were examined using an Olympus PROVIS AX70 fluorescence microscope. For the detection of fusion proteins linked with
either β-galactosidase or HA epitope, anti-β-galactosidase antibody,
anti-HA antibody and Texas red-conjugated secondary antibody were
used at 5 µg/ml, 5 µg/ml and 1.5 µg/ml, respectively.
Preparation of cell extracts and western blot analysis
After transfection, COS-1 cells were washed with PBS, scraped from
Fig. 1. Immunofluorescence localization of p46 and p54.
(A) Endogenous p46 and p54 in NIH3T3 cells were examined by
indirect immunofluorescence analysis using affinity-purified anti-p46
and anti-p54 antibodies and FITC-conjugated anti-rabbit IgG
antibody (upper row). The lower row shows nuclear staining by
Hoechst 33258. (B) pSRα46 and pSRα54 were transiently
transfected into COS-1 cells, and expression of p46 and p54 was
detected by immunofluorescence with anti-p46 and anti-p54
antibodies and FITC-conjugated anti-rabbit IgG antibody (upper
row). The lower row shows nuclear staining by Hoechst 33258. Bar,
40 µm.
plates in PBS, centrifuged for 5 minutes, and resuspended in either
Laemmli sample buffer (Laemmli, 1970) for whole-cell extracts or
extraction buffer as described previously (Takada-Takayama et al.,
1990a). The samples were resolved by SDS-PAGE and transferred
electrophoretically onto 0.45 µm PVDF membranes (Millipore). After
incubation of the membranes with either 0.3 µg/ml of anti-p54 or antip46, or 0.5 µg/ml of anti-β-galactosidase antibody in TBS (Trisbuffered saline; 50 mM Tris-HCl, pH 7.5 and 150 mM NaCl) containing 5% (w/v) dried milk for 1 hour at room temperature, the
membranes were washed four times with TBS containing 0.1%
Tween-20. The membranes were then incubated for 1 hour at room
temperature with horseradish peroxidase-conjugated goat anti-rabbit
or anti-mouse IgG in TBS containing 5% dried milk and washed
2630 T. Mizuno and others
again. Detection of the protein bands was performed using an
enhanced chemiluminescent kit (Amersham) following the manufacturer’s instructions.
RESULTS
Expression of mouse p46 and p54 subunits in COS1 cells
To characterize p46 and p54 subunits of DNA primase individually, we constructed a cDNA expression system using
COS-1 cells as host cells, SRα as a eukaryotic promoter, and
electroporation as a method for DNA transfection. To
determine the subcellular distributions of overexpressed p46
and p54 in COS-1 cells, immunohistochemical analysis was
performed. We prepared affinity-purified polyclonal antibodies
raised against either p54 expressed in E. coli or p46 derived
from insect cells infected with recombinant baculovirus. Specificity of these antibodies was demonstrated by western blot
analysis with purified primase subunits of mouse FM3A cells
(data not shown) and with exogenously overexpressed mouse
DNA primase subunits (see below, Fig. 2). These antibodies
reacted with endogenous primase subunits of mouse NIH3T3
cells as revealed by indirect immunofluorescence analysis and
showed blight nuclear staining (Fig. 1A). Using these antibodies, the subcellular distribution of exogenously overexpressed mouse DNA primase subunits were determined. The
transiently transfected COS-1 cells were fixed 24 hours after
transfection, permeabilized, and incubated with anti-p46 or
anti-p54 antibody. We found that cells transfected with
pSRα54 exhibited intense fluorescence in the nucleus except
for the nucleolus. In contrast to p54, p46 showed diffused cytoplasmic staining (Fig. 1B). Endogenous monkey DNA primase
in COS-1 cells was scarcely detected under these conditions
(compare the Hoechst staining with p46 or p54 staining), and
was observed in longer exposures (not shown). These observation suggest that (i) p54 has a NLS and transported into the
nucleus, whereas p46 lacks the NLS and is not able to enter
the nucleus by itself, and (ii) this overexpression system was
suitable to study expression of mouse DNA primase exogenously in monkey COS-1 cells since the signals of monkey
primase were far weaker than those of overexpressed primase.
To evaluate the level of proteins expressed in transfected
COS-1 cells, proteins were extracted from the cells, electrophoresed on SDS-PAGE, and subjected to western blot
analysis. As shown in Fig. 2, p46 and p54 expressed exogenously were detected as intense signals, although endogenous
p46 and p54 in COS-1 cells were undetectable. Insoluble
proteins accumulated more in cells incubated for 48 hours than
for 24 hours. Therefore, these insoluble forms could be aggregated proteins caused by an abnormal increase in protein level
due to the strong promoter activity of SRα. Indeed, the
expression levels of p46 and p54 in each cell were varied and
about 10% of the cells showed an intense punctate immunoreactivity pattern resembling condensed aggregated clots as
revealed by immunohistochemical analysis (data not shown).
It was reported that overexpressed protein occasionally
become inclusion bodies in COS-1 cells (Sorimachi et al.,
1993).
Several minor bands around the p54 and p46 were observed
after longer incubation (Fig. 2). Although we have not yet
Fig. 2. Western blot analysis of transfected p46 and p54 in COS-1
cells. COS-1 cells transfected with pSRα46 (A) and pSRα54 (B)
were incubated for 24 hours and 48 hours. Then, cells were
harvested, separated into soluble (S) and insoluble (P) fractions, and
subjected to western blot analysis with either anti-p46 (A) or antip54 (B) antibody; 20 µg of soluble proteins (lane 1, 3, 5, 7, 9, and
11) and corresponding insoluble fractions (lane 2, 4, 6, 8, 10, and 12)
were loaded in each lane. Mock indicates control experiment in the
absence of plasmid.
defined these protein bands completely, they might be
degraded or modified products, since they are specific for both
cDNA transfected and antibodies used. These results show that
affinity-purified polyclonal antibodies used here were
confirmed to be sufficiently specific to detect expressed
proteins without cross-reactivity with endogenous unrelated
COS-1 proteins.
Identification of the NLS in p54 using deletion and
point mutants
The finding that overproduced p54 is localized in the nucleus
prompted us to identify the NLS of p54. Several motifs
composed of clusters of basic residues have been defined for
targeting of proteins to the nucleus such as -KKKRK- for
SV40 large T antigen (Kalderon et al., 1984; Lanford and
Butel, 1984) and KR---------KKKK for nucleoplasmin
(Robbins et al., 1991). However, a search for sequences in p54
homologous to a canonical NLS was not fruitful. Therefore,
to determine which region of p54 is required for its nuclear
transport, we first constructed a set of deletion mutants (Fig.
3A) and evaluated the ability of the resultant mutant proteins
to enter the nucleus in transiently transfected COS-1 cells by
indirect immunofluorescence analysis. Expression of each
mutant in COS-1 cells was detected by western blot analysis
as shown in Fig. 3B. We found that the cells expressing
p54∆325-505 in which the carboxyl-terminal region was
truncated showed intense nuclear fluorescence similarly to
that seen with the wild-type protein (Fig. 4G,A). In contrast,
cells expressing two kinds of mutant proteins with truncation
of the amino-terminal region (p54∆1-124 and p54∆1-306)
exhibited diffuse cytoplasmic staining (Fig. 4B,C). In
addition, cells expressing mutant proteins in which the
carboxyl-terminal region was truncated (p54∆164-505 and
p54∆1-124∆325-505) showed intense staining of aggregated
clots (Fig. 4E,F). In the case of p54∆164-505, 10% of the cells
expressing the protein showed nuclear staining patterns
Nuclear localization signal of mouse DNA primase 2631
A.
Wild Type
p54∆325-505
RIRKKLR
1
p54∆1-124∆325-505
p54
505
324
1
p54∆1-124
p54∆164-505
RRR
505
125
163
1
324
125
p54∆1-306
RIQNNLR
1
p54RKK(8-10)QNN
QQQ
p54RRR(96-98)QQQ 1
30
p54 1-30 :p46
1
114
p54 1-114 :p46
1
RIQNNLR
p54RKK(8-10)QNN:βgal
QQQ
p54RRR(96-98)QQQ:βgal
307
505
505
505
p46
β-galactosidase
p541-114 :βgal
1-90
p54
:βgal
p541-34 :βgal
p541-13 :βgal
Fig. 3. Mutant proteins used in this study.
(A) Schematic representation of the mutant
constructs. Open boxes, cross-hatched boxes, and
slashed and interrupted boxes represent cDNA
encoding p54, p46, and β-galactosidase, respectively.
Thick lines indicate the clusters of the basic residues.
The numbers indicate amino acid positions of p54.
(B) Western blot analysis of mutant proteins
expressed transiently in COS-1 cells. COS-1 cells
were transfected with the plasmids encoding mutant
proteins, incubated for 24 hours, and lysed with
Laemmli sample buffer to generate whole-cell
extract. Samples were subjected to SDS-PAGE,
followed by western blotting with anti-p54, anti-p46,
and anti-β-galactosidase antibody. The asterisks
indicate the positions of proteolytic fragments.
similar to those seen with the wild-type p54 and p54∆325-505
(Fig. 4D). These findings suggest that: (i) the amino-terminal
region of p54 is necessary for its translocation to the nucleus;
and (ii) the carboxyl-terminal region and overall structure are
important to maintain the conformation of the protein.
Therefore, although the mutant protein p54∆164-505 retained
its amino-terminal region, 90% of the cells expressing this
mutant protein showed aggregated clots probably due to lack
of the carboxyl-terminal region.
To further define the region necessary for nuclear translocation and to eliminate marked changes in the overall structure
caused by deletion mutations, we next constructed point
mutants. In the amino-terminal region of p54, we noted that
there are two clusters of basic residues. These sequences,
RIRKKLR (residues 6-12) and RRR (residues 96-98), were
substituted with neutral amino acid residues resulting in
RIQNNLR and QQQ, respectively. Interestingly, both mutant
proteins were found predominantly in the cytoplasm (Fig.
4H,I), indicating that the amino-terminal region of p54 is really
important for its translocation to the nucleus.
Subcellular distribution of chimeric proteins of p54
and either p46 or β-galactosidase
Deletion and point mutant analyses suggested that the aminoterminal region could be necessary for the NLS. To confirm
that these sequences actually function as the NLS, we constructed chimeric proteins in which the amino-terminal
regions of p54 were fused with cytoplasmic reporter proteins.
We first used the p46 subunit as a cytoplasmic reporter
protein, since p46 is exclusively retained in the cytoplasm and
can be detected clearly in our system. Plasmids containing the
chimeric gene encoding residues 1-30 and 1-114 of p54 fused
with p46 were transfected into COS-1 cells. At 24 hours after
transfection, the subcellular distribution of chimeric proteins
was determined with anti-p46 antibody. We found that
residues 1-30 of p54 targeted p46 to the nucleus as efficiently
as the wild-type p54 (Fig. 5). This suggests that only the
amino-terminal cluster of basic residues is sufficient for
protein targeting to the nucleus. On the other hand, a fusion
protein containing amino acids 1-114 which contained both
basic clusters of p54 with p46 was retained in the cytoplasm
2632 T. Mizuno and others
Fig. 4. Subcellular distribution of deletion and point mutant p54 by
indirect immunofluorescence analysis. Deletion and point mutant
proteins were detected at 24 hours after transfection by indirect
immunofluorescence analysis using anti-p54 antibody and FITCconjugated anti-rabbit IgG antibody. Bar, 40 µm.
similarly to the wild-type p46. In addition, point mutant
p54RRR(96-98)QQQ, which contained basic residues 6-12,
was also retained in the cytoplasm (Fig. 4H), suggesting that
residues 6-12 are not sufficient for targeting the protein to the
nucleus. To clarify this apparent discrepancy and to determine
whether both clusters of basic residues are necessary for the
NLS, further chimeric proteins using β-galactosidase of E.
coli as a cytoplasmic reporter were examined. We made constructs for expression of fusion proteins in which β-galactosidase was fused with a set of deletion proteins with progressively truncated amino-terminal regions of p54 (Fig. 3A). The
subcellular distribution of these fusion proteins was detected
using an anti-β-galactosidase antibody. In this case, while
wild-type β-galactosidase was localized exclusively in the
cytoplasm, the fusion proteins linked with residues either 113, 1-34 or 1-90 were located predominantly in the nucleus
(Fig. 6). All of these fusion proteins contained the basic
residues 6-12. In contrast, the fusion protein p541-114:βgal and
p54RRR(96-98)QQQ:βgal, which contained amino-terminal
residues 1-114 including basic residues 6-12 were localized
diffusely in both cytoplasm and the nucleus (Fig. 6). Finally,
the fusion protein p54RKK(8-10)QNN:βgal, which also
contained amino-terminal residues 1-114 including basic
residues 96-98 but which lacked the basic charge of residues
6-12, was localized exclusively in the cytoplasm. Taken
together, we concluded that: (i) residues 6-12, RIRKKLR, are
not only essential but also sufficient to ensure the exclusive
nuclear location of p54; (ii) residues 96-98, RRR, are
important for the amino-terminal residues 6-12 to function as
an NLS, but not sufficient to work as an NLS themselves; and
(iii) transport by this NLS was easily affected by conformational changes in an artificial construct, especially those containing residues 1-114. Therefore, the cytoplasmic localization of mutants such as p541-114:p46, p541-114:βgal, and
p54RRR(96-98)QQQ, all of which contain residues 1-114,
may be due to inadequate conformation of the fusion
construct.
Fig. 5. Immunofluorescence localization of
chimeric proteins of p54 and p46. Fusion
proteins were detected at 24 hours after
transfection by indirect immunofluorescence
analysis using anti-p54 antibody (upper row,
left panel) and anti-p46 antibody (upper row,
center and right panels) and FITC-conjugated
anti-rabbit IgG antibody. Nuclei
counterstained by Hoechst 33258 are shown in
the lower row. Bar, 40 µm.
Nuclear localization signal of mouse DNA primase 2633
Fig. 6. Immunofluorescence localization of a
chimeric protein of β-galactosidase and the aminoterminal region of p54. Fusion proteins were
detected at 48 hours after transfection by indirect
immunofluorescence analysis using anti-βgalactosidase antibody and Texas red-conjugated
anti-mouse IgG antibody. Bar, 40 µm.
Fig. 7. Double-staining of COS-1 cells coexpressed with p46 and HA-tagged p54 using
anti-p46 antibody and anti-HA monoclonal
antibody. pSRα46, HA-tagged pSRα54, and
their mutants were co-transfected into COS-1
cells, and expressed proteins were detected
simultaneously by indirect immunofluorescence
using anti-p46 polyclonal antibody and FITCconjugated anti-rabbit IgG antibody (upper
panels) and anti-HA monoclonal antibody and
Texas red-conjugated anti-mouse IgG antibody
to detect HA-tagged p54 (middle panels). DNA
was stained by Hoechst 33258 (lower panels).
Bar, 40 µm.
Effect of co-expression of p46 and p54 on
subcellular distribution in COS-1 cells
Since p46 overexpressed in COS-1 cells remained predominantly in the cytoplasm, we analyzed the effect of coexpression with p54 on subcellular distribution of p46.
pSRα46 and HA-tagged pSRα54 were co-transfected into
COS-1 cells and these proteins in resultant transformants were
detected by indirect immunofluorescence. The HA tag allowed
us to detect both subunits of DNA primase at the same time in
the same cells using Texas red-conjugated anti-mouse IgG
antibody for detection of HA-tagged p54 and FITC-conjugated
anti-rabbit IgG antibody for detection of p46. When both
subunits were co-expressed, the subcellular distribution of p46
changed dramatically. As shown in Fig. 7, p46 co-expressed
with p54 was localized predominantly in the nucleus similarly
to p54. In contrast, when p46 was co-expressed with the
nuclear-targeting deficient mutant p54RKK(8-10)QNN,
targeting of p46 to the nucleus was completely abolished.
These results suggested that the nuclear translocation-deficient
p46 is carried into the nucleus by the nuclear translocation-pro-
2634 T. Mizuno and others
ficient p54. However, there is the alternative possibility that
p46 might have a cryptic NLS on its own sequence and that
conformational change of p46 caused by binding to p54 makes
the cryptic NLS functional. In addition, cytoplasmic staining
of p54RKK(8-10)QNN and p46 as shown in Fig. 7 may be
caused by the inability of complex formation of p54RKK(810)QNN with p46 because of the mutation. To eliminate such
possibilities, further co-expression studies were performed. We
used two constructs characterized above: HA-tagged
p54RKK(8-10)QNN as a nuclear translocation-deficient
subunit, and the fusion construct p541-30:p46 as a nuclear
translocation-proficient subunit as described in Fig. 5. When
both mutants were co-expressed in COS-1 cells, predominant
nuclear staining of both subunits was observed as shown in Fig.
7. This suggests that p54RKK(8-10)QNN can bind p46 efficiently. Therefore, it was confirmed that p54RKK(8-10)QNN
and p46 formed a complex in the cytoplasm and there is no
NLS in DNA primase besides RIRKKLR at the amino
terminus of p54. In addition, the NLS of p54 can function in
the amino terminus of p46, indicating that the domain encoding
the NLS of DNA primase is exchangeable at the amino
terminus of both subunits.
DISCUSSION
In this report, using a cDNA overexpression system with the
SRα promoter in COS-1 cells, we identified the NLS of mouse
DNA primase which is located from residues 6-12 near the
amino-terminal region of p54. This sequence was not only
necessary for nuclear targeting of the parent p54 protein but
also sufficient to direct non-nuclear proteins such as the p46
subunit of primase and β-galactosidase to the nucleus.
Therefore, this sequence fulfills both criteria for the NLS, indicating that p54 has an authentic NLS. Furthermore, we demonstrated that p46, which is deficient in nuclear translocation by
itself, can be transported into the nucleus by interaction with
p54, indicating that DNA primase translocates to the nucleus
in a ‘piggy-back’ binding transport system.
The transport of large proteins into the nucleus is an ATPdependent active process. Extensive studies on the nuclear
import of viral and cellular proteins have revealed that the
process of transport to the nucleus generally falls into two categories (Chelsky et al., 1989; Dingwall and Laskey, 1991;
Garcia-Bustos et al., 1991; Silver, 1991); one mechanism is the
recognition of the specific NLS, comparatively short sequences
rich in basic amino acids residues, and the other is dependent
on the piggy-back binding of a protein which lacks the NLS
on its own sequence to another protein which has an NLS
(Zhao and Padmanabhan, 1988). Although selective and facilitated transport occur through the NLS-mediated or piggy-back
binding-dependent passive pathway, diffusion of some small
molecules (<40~60 kDa) also occurs. Therefore, it is possible
that p46, which has an affinity for ssDNA (Copeland and
Wang, 1993), may diffuse into and be retained in the nucleus.
However, our results showed that p46 expressed in COS-1
cells was retained predominantly in the cytoplasm, indicating
that transport of p46 to the nucleus is facilitated efficiently by
interaction with p54 and may be the main pathway for DNA
primase import into the nucleus. An alternative, though less
likely, interpretation of these data is that p46 may be able to
passively diffuse into the nucleus but cannot be retained there
unless there is free, nuclear, p54 with which to bind. Although
this possibility may not be ruled out, co-expressed nucleartranslocation deficient p54 mutant and nuclear-translocation
proficient p46 mutant were translocated into the nucleus as
shown in Fig. 7, suggesting that translocation of these subunits
would be dependent on piggy-back binding rather than the
presence of free p54 in the nucleus.
The observation that overexpressed p54 was transported into
the nucleus efficiently and possessed the NLS was unexpected.
As no canonical NLS was found by cDNA sequence analysis
of subunits of the DNA polymerase α-primase complex except
p180 (Miyazawa et al., 1993), we considered that translocation
of the DNA polymerase α-primase complex to the nucleus
should occur after complex formation of the four subunits in
the cytoplasm, and that the NLS on p180 should play a major
role in carrying the complex into the nucleus. However, when
p54 was overexpressed in COS-1 cells, predominantly cells
expressing p54 showed intense nuclear staining, suggesting
that p54 has an NLS on its own sequence. On the other hand,
from the results of a recent study using yeast strains, it was
shown that overexpressed p68 or p180 were localized in the
nucleus (Foiani et al., 1994; Bouvier and Baldacci, 1995).
Therefore, although biochemical and cell physiological characterization remain to be performed, p68 and p180 are likely
to have NLS on their own sequences. Taken together, these
results indicate that only p46 lacks an NLS in the DNA polymerase α-primase complex and the nuclear translocationdeficient p46 is carried into the nucleus by the nuclear translocation-proficient p54. In addition, DNA primase can enter the
nucleus independently of DNA polymerase α, since p46 can
interact only with p54 (Copeland and Wang, 1993; Longhese
et al., 1993).
Functions for each subunit of DNA primase have been
suggested based on recent advances in cDNA cloning and characterization. The smaller subunit, p46, has a zinc finger motif
in its amino-terminal region, and is a catalytic subunit capable
of initiation and elongation of RNA primers (Santocanale et
al., 1993; Bakkenist and Cotterill, 1994; Stadlbauer et al.,
1994). In addition, p46 has been suggested to be able to bind
ssDNA (Copeland and Wang, 1993), ribonucleotide triphosphate (Foiani et al., 1989; Copeland and Wang, 1993),
polyoma large T antigen (Brückner et al., 1995), and p54
(Copeland and Wang, 1993; Longhese et al., 1993). In contrast
to p46, no known catalytic activity of p54 has been found, and
the function of p54 is still under investigation. Since considerable homology was found among human, mouse, and yeast
p54 (mouse p54 is 89% identical to human p54 and 32%
identical to yeast p54 at the amino acid level), p54 is considered to play a pivotal role conserved during evolution. Our
results presented here represent the first evidence that p54
works as a molecular guide of p46 to the nucleus. Comparison
of amino acid sequences around the mouse NLS indicated that
basic residues are well-conserved among these species (data
not shown). Therefore, the role of p54 to carry p46 into the
nucleus may be conserved in all eukaryotes.
In yeast cells, free p46, whose function remains uncertain,
was detected by immunopurification (Santocanale et al., 1993).
Our observation that p46 lacks the NLS and is not able to enter
the nucleus in the absence of p54 suggests that free p46 might
be a reservoir in the cytoplasm awaiting binding to p54 for
Nuclear localization signal of mouse DNA primase 2635
translocation to the nucleus. Although no information has yet
been reported regarding the existence of free p46 in other
species, it is tempting to speculate that the level of p54 protein
expression is a limiting step in the biosynthesis and processing pathway of DNA primase, and the transport of p46 by p54
to the nucleus maintains the correct level of catalytic p46
protein in the nucleus. Since free p46, which possesses primase
activity and will be regulated through other subunits, might be
extremely toxic for the cell, transport of DNA primase to the
nucleus may become highly dependent upon p54 to avoid entry
of free p46 into the nucleus. Indeed, we observed previously
that the DNA polymerase α-primase level in mouse cells was
highly coordinated at the mRNA level during cell cycle and
proliferation (Miyazawa et al., 1993), suggesting that the ratio
of each subunit for the complex is a precisely maintained
constant. To determine whether the p46 level in the nucleus is
actually controlled by nuclear translocation level, it will be
necessary to characterize promoter activity, mRNA stability,
translation efficacy, and protein stability of both p46 and p54,
and such studies are now underway in our laboratory.
In conclusion, using a cDNA expression system in cultured
mammalian cells, we identified the NLS of DNA primase and
demonstrated that p46 is transported to the nucleus by piggyback binding to p54 which has an NLS. Efficient co-expression
of subunits now provides a system in which the role of each
subunit in DNA primase activity, i.e. initiation and elongation,
can be assessed, and other cellular factors which influence
DNA primase functions can be identified.
We thank Dr Yusaku Nakabeppu for providing the pcDEB∆
expression vector, Dr Tetsu Akiyama for providing NIH3T3 cells, Dr
Hiroshi Miyazawa for critical reading of the manuscript and helpful
discussions, and Dr Shigehisa Hirose for encouragement. This work
was supported by grants from the Ministry of Education, Science,
Sports, and Culture of Japan and a special grant for the promotion of
research from the Institute of Physical and Chemical Research
(RIKEN) and the Biodesign Research Program of RIKEN. T.M. is a
special postdoctoral researcher of RIKEN.
REFERENCES
Bakkenist, C. J. and Cotterill, S. (1994). The 50-kDa primase subunit of
Drosophila melanogaster DNA polymerase α: molecular characterization of
the gene and functional analysis of the overexpressed protein. J. Biol. Chem.
269, 26759-26766.
Bouvier, D. and Baldacci, G. (1995). The N-terminus of fission yeast DNA
polymerase α contains a basic pentapeptide that acts in vivo as a nuclear
localization signal. Mol. Biol. Cell 6, 1697-1705.
Brückner, A., Stadlbauer, F., Guarino, L. A., Brunahl, A., Schneider, C.,
Rehfuess, C., Prives, C., Fanning, E. and Nasheuer, H.-P. (1995). The
mouse DNA polymerase α-primase subunit p48 mediates species-specific
replication of polyomavirus DNA in vitro. Mol. Cell. Biol. 15, 1716-1724.
Chelsky, D., Ralph, R. and Jonak, G. (1989). Sequence requirements for
synthetic peptide-mediated translocation to the nucleus. Mol. Cell. Biol. 9,
2487-2492.
Collins, K. L., Russo, A. A. R., Tseng, B. Y. and Kelly, T. J. (1993). The role
of the 70 kDa subunit of human DNA polymerase α in DNA replication.
EMBO J. 12, 4555-4566.
Copeland, W. C. and Wang, T. S.-F. (1991). Catalytic subunit of human DNA
polymerase α overproduced from baculovirus-infected insect cells:
structural and enzymological characterization. J. Biol. Chem. 266, 2273922748.
Copeland, W. C. and Wang, T. S.-F. (1993). Enzymatic characterization of
the individual mammalian primase subunits reveals a biphasic mechanism
for initiation of DNA replication. J. Biol. Chem. 268, 26179-26189.
Copeland, W. C. and Tan, X. (1995). Active site mapping of the catalytic
mouse primase subunit by alanine scanning mutagenesis. J. Biol. Chem. 270,
3905-3913.
Dingwall, C. and Laskey, R. A. (1991). Nuclear targeting sequences – a
consensus? Trends Biochem. Sci. 16, 478-481.
Foiani, M., Lindner, A. J., Hartmann, G. R., Lucchini, G. and Plevani, P.
(1989). Affinity labeling of the active center and ribonucleoside triphosphate
binding site of yeast DNA primase. J. Biol. Chem. 264, 2189-2194.
Foiani, M., Marini, F., Gamba, D., Lucchini, G. and Plevani, P. (1994). The
B subunit of the DNA polymerase α-primase complex in Saccharomyces
cerevisiae executes an essential function at the initial stage of DNA
replication. Mol. Cell. Biol. 14, 923-933.
Garcia-Bustos, J., Heitman, J. and Hall, M. N. (1991). Nuclear protein
localization. Biochim. Biophys. Acta 1071, 83-101.
Ito, W., Ishiguro, H. and Kurosawa, Y. (1991). A general method for
introducing a series of mutations into cloned DNA using the polymerase
chain reaction. Gene 102, 67-70.
Izumi, M., Miyazawa, H., Harakawa, S., Yatagai, F. and Hanaoka, F.
(1994). Identification of a point mutation in the cDNA of the catalytic subunit
of DNA polymerase α from a temperature-sensitive mouse FM3A cell line.
J. Biol. Chem. 269, 7639-7644.
Kaguni, L. S., Rossignol, J.-M., Conaway, R. C., Banks, G. R. and Lehman,
I. R. (1983). Association of DNA primase with the β/γ subunits of DNA
polymerase α from Drosophila melanogaster embryos. J. Biol. Chem. 258,
9037-9039.
Kalderon, D., Richardson, W. D., Markham, A. F. and Smith, A. E. (1984).
Sequence requirements for nuclear location of simian virus 40 large-T
antigen. Nature 311, 33-38.
Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of
the head of bacteriophage T4. Nature 227, 680-685.
Lanford, R. E. and Butel, J. S. (1984). Construction and characterization of an
SV40 mutant defective in nuclear transport of T antigen. Cell 37, 801-813.
Longhese, M. P., Jovine, L., Plevani, P. and Lucchini, G. (1993).
Conditional mutations in the yeast DNA primase genes affect different
aspects of DNA metabolism and interactions in the DNA polymerase αprimase complex. Genetics 133, 183-191.
Miyazawa, H., Izumi, M., Tada, S., Takada, R., Masutani, M., Ui, M. and
Hanaoka, F. (1993). Molecular cloning of the cDNAs for the four subunits
of mouse DNA polymerase α-primase complex and their gene expression
during cell proliferation and the cell cycle. J. Biol. Chem. 268, 8111-8122.
Nakabeppu, Y., Oda, S. and Sekiguchi, M. (1993). Proliferative activation of
quiescent Rat-1A cells by ∆FosB. Mol. Cell. Biol. 13, 4157-4166.
Nasheuer, H.-P. and Grosse, F. (1988). DNA polymerase α-primase from calf
thymus: determination of the polypeptide responsible for primase activity. J.
Biol. Chem. 263, 8981-8988.
Robbins, J., Dilworth, S. M., Laskey, R. A. and Dingwall, C. (1991). Two
independent basic domains in nucleoplasmin nuclear targeting sequence:
identification of a class of bipartite nuclear targeting sequence. Cell 64, 615623.
Santocanale, C., Foiani, M., Lucchini, G. and Plevani, P. (1993). The
isolated 48,000-dalton subunit of yeast DNA primase is sufficient for RNA
primer synthesis. J. Biol. Chem. 268, 1343-1348.
Silver, P. A. (1991). How proteins enter the nucleus. Cell 64, 489-497.
Sorimachi, H., Toyama-Sorimachi, N., Saido, T. C., Kawasaki, H., Sugita,
H., Miyasaka, M., Arahata, K., Ishiura, S. and Suzuki, K. (1993).
Muscle-specific calpain, p94, is degraded by autolysis immediately after
translation, resulting in disappearance from muscle. J. Biol. Chem. 268,
10593-10605.
Stadlbauer, F., Brueckner, A., Rehfuess, C., Eckerskorn, C., Lottspeich,
F., Förster, V., Tseng, B. Y. and Nasheuer, H.-P. (1994). DNA replication
in vitro by recombinant DNA-polymerase-α-primase. Eur. J. Biochem. 222,
781-793.
Stillman, B. (1994). Smart machines at the DNA replication fork. Cell 78, 725728.
Suzuki, M., Enomoto, T., Hanaoka, F. and Yamada, M. (1985). Dissociation
and reconstitution of a DNA polymerase α-primase complex. J. Biochem. 98,
581-584.
Suzuki, M., Enomoto, T., Masutani, C., Hanaoka, F., Yamada, M. and Ui,
M. (1989). DNA primase-DNA polymerase α assembly from mouse FM3A
cells: purification of constituting enzymes, reconstitution, and analysis of
RNA priming as coupled to DNA synthesis. J. Biol. Chem. 264, 10065-10071.
Takada-Takayama, R., Tada, S., Hanaoka, F. and Ui, M. (1990a). Peptide
mapping of the four subunits of the mouse DNA polymerase α-primase
complex. Biochem. Biophys. Res. Commun. 170, 589-595.
2636 T. Mizuno and others
Takada-Takayama, R., Suzuki, M., Enomoto, T., Hanaoka, F. and Ui, M.
(1990b). Purification and characterization of mouse DNA polymerase α
devoid of primase activity. FEBS Lett. 273, 27-30.
Takebe, Y., Seiki, M., Fujisawa, J., Hoy, P., Yokota, K., Arai, K., Yosida,
M. and Arai, N. (1988). SRα promoter: an efficient and versatile
mammalian cDNA expression system composed of the simian virus 40 early
promoter and the R-U5 segment of human T-cell leukemia virus type 1 long
terminal repeat. Mol. Cell. Biol. 8, 466-472.
Wang, T. S.-F. (1991). Eukaryotic DNA polymerases. Annu. Rev. Biochem. 60,
513-552.
Zhao, L.-J. and Padmanabhan, R. (1988). Nuclear transport of adenovirus
DNA polymerase is facilitated by interaction with preterminal protein. Cell
55, 1005-1015.
(Received 6 June 1996 – Accepted 6 August 1996)