Journal of General Virology (1993), 74, 132%1333. Printed in Great Britain
1327
Photochemical cross-linking of influenza A polymerase to its virion RNA
promoter defines a polymerase binding site at residues 9 to 12 of the
promoter
Ervin Fodor, B a i k L. S e o n g t and G e o r g e G. Brownlee*
Chemical Pathology Unit, Sir William Dunn School o f Pathology, University o f Oxford, South Parks Road,
Oxford OX1 3RE, U.K.
A previous study of the 12 nucleotide-long influenza A
virion RNA promoter has shown that three nucleotides,
residues 9 to 11, were crucial for transcription in vitro,
although other nucleotides play a significant but less
important role. A model for polymerase-promoter
recognition was proposed, according to which there
were two sites: a binding site at residues 9 to 11 and a
regulatory site at or near the site of initiation at residue
1. By studying the effect of point mutations in the
promoter on the binding efficiency of the polymerase
using a photochemical cross-linking assay, we now show
that residues 9 to 12 are crucial for binding. In addition
residues 4 to 8, though not as important, are involved
in binding, possibly by stabilizing the polymerasepromoter complex. Both PB1 and PB2 apparently play
an important role during virion RNA promoter recognition and binding.
Introduction
1989). PB2 recognizes and binds to the cap structure of
mRNAs (Ulmanen et al., 1981) and is probably
responsible for the endonucleolytic cleavage of the
capped host cell mRNAs (Galarza et al., 1991). No
specific function has been assigned to PA, although it
may be required for replication since temperaturesensitive mutations of PA affect replication, but not
transcription (Mahy et al., 1981 ; Krug et al., 1975). PA
is, however, essential for polymerase activity as it is
required for successful assembly of an active polymerase
complex along with PB1 and PB2 (Kobayashi et al.,
1992).
It is well known that there is a 12 nucleotide (nt)-long
highly conserved sequence 5" C C U G C U U U U G C U o ~ 3'
at the Y end of all vRNA segments of influenza A viruses
(Robertson, 1979; Desselberger et al., 1980). Recently,
we have shown that these 12 conserved nucleotides are
sufficient to function as a promoter for mRNA and
cRNA synthesis in vitro (Seong & Brownlee, 1992a).
Furthermore, by studying the efficiency of RNA synthesis in vitro using model RNA templates with individual point mutations in all 12 positions of the vRNA
promoter, we have shown that three nucleotides, residues
9, 10 and 11, within the promoter are crucial for activity,
although other nucleotides also play a significant but less
important role. We proposed that the promoter has two
sites: (i) a polymerase binding site at residues 9 to 11, and
possibly 12, and (ii) a regulatory site at residues 1, and
possibly 2, 3 and 4 (Seong & Brownlee, 1992b). To test
During the influenza virus life cycle three types of RNA
transcripts are synthesized. These are mRNA, complementary RNA (cRNA) and virion RNA (vRNA).
Influenza virus mRNAs are synthesized using short,
capped RNA fragments as a primer originating from
host cell mRNAs. They are polyadenylated and are
incomplete copies of vRNA because they lack sequences
complementary to 17 to 22 nucleotides at the 5' end of
vRNA. By contrast, cRNAs are full-length, non-capped,
non-polyadenylated copies of vRNA, and serve as
templates for the production of more vRNA (Ishihama
& Nagata, 1988; Lamb & Choppin, 1983; McCauley &
Mahy, 1983). The molecular mechanism and regulation
of transcription and replication are poorly understood.
All three reactions, i.e. transcription of negative sense
vRNA into mRNA, replication of vRNA into cRNA
and replication of cRNA into vRNA, are catalysed by
the influenza virus RNA polymerase complex consisting
of three virus-encoded proteins, PB1, PB2 and PA. PB1
is involved in the initiation and elongation of mRNA
chains (Braam et al., 1983; Ulmanen et al., 1981).
Consistent with this, its amino acid sequence contains
motifs found in a series of positive- and negativestranded RNA virus polymerase genes (Poch et al.,
I"Present address: Hanhyo Institute of Technology, Seoul, Korea.
0001-1550 © 1993SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:27:38
1328
E. Fodor, B. L. Seong and G. G. Brownlee
this m o d e l we n o w investigate the b i n d i n g o f the
p o l y m e r a s e c o m p l e x to the v R N A p r o m o t e r using a
p h o t o c h e m i c a l c r o s s - l i n k i n g assay. By s t u d y i n g the
effects o f m u t a t i o n s at all 12 p o s i t i o n s in the v R N A
p r o m o t e r , we p r o v i d e evidence t h a t residues 9 to 12 are
crucial for the effective b i n d i n g o f the p o l y m e r a s e
c o m p l e x to v R N A p r o m o t e r a n d t h a t they r e p r e s e n t a
b i n d i n g site. I n a d d i t i o n to its k n o w n c a p - b i n d i n g a n d
e n d o n u c l e a s e functions ( U l m a n e n et al., 1981; G a l a r z a
et al., 1991), PB2 c o o p e r a t e s with PB1 in p r o m o t e r
r e c o g n i t i o n a n d binding.
Methods
Preparation of micrococcal nuclease-treated viral core. The micrococcal nuclease-treated viral core was prepared as described previously
(Seong & Brownlee, 1992a). Briefly, the viral core isolated from X-31
(a recombinant of influenza viruses A/HK/68 × A/PR/8/34) by Triton
X-100 and lysolecithin disruption and glycerol gradient centrifugation
was treated with micrococcal nuclease (Boehringer) followed by EGTA
inactivation.
Preparation of RNA templates. Model RNA templates, wild-type
(14 nt long) or with point mutations (14 nt long) or with an insertion of
a C residue between positions 4 and 5 (15 nt long) in the 12 nt-long
vRNA promoter, were synthesized by T7 RNA polymerase transcription of a partial DNA duplex as before (Seong & Brownlee, 1992b)
in the presence of [~-a2p]CTP (800 Ci/mmol). The sequence of the wildtype model RNA template is 5' GGCCUGCUUUUGCUo M 3' (the
12 nt-long vRNA promoter is underlined). Where mutant sequences
were used, this is indicated in the Results. Full-length products of T7
polymerase transcription were purified by 18 % PAGE in 7 M-urea
followed by elution from the gel with QO1 buffer (100 mM-NaC1,
50 mM-MOPS pH 7.0, 15 % ethanol) overnight at 4 °C and purification
on a Qiagen column. The concentration of template RNA was
estimated from the radioactive yield (measured by scintillation
counting) assuming that it had the same specific activity as the starting
labelled triphosphate (800 Ci/mmol). Wild-type 12nt-long vRNA
promoter 5' CCUGCUUUUGCUoH 3' and U15 sequence were
chemically synthesized using an Applied Biosystems DNA synthesizer
(Model 380B) and after deblocking (B. Sproat, personal communication) were purified by Sephadex G-25 (NAP-10, Pharmacia) column
chromatography and 20 % PAGE in 7 M-urea, followed by elution with
H20 and purification on Qiagen column. Yields were measured
spectroscopically at A260. The synthetic oligonucleotides were 5' endlabelled (for cross-linking studies) with [~-32p]ATP and T4 polynucleotide kinase followed by phenol~zhloroform extraction and
purification on a Qiagen column.
Reconstitution and photochemical cross-linking. About 4 gl of
micrococcal nuclease treated-viral core protein was mixed with 0.01 to
0.05 pmol of 32P-labelled RNA template (see above) and 1 unit of
human placental ribonuclease inhibitor (Amersham) in transcription
buffer (50 mM-Tris-HC1 pH 7.8, 50 mM-KC1, 10 mM-NaC1, 5 mMMgC12, 1 mM-DTT) and the volume was adjusted to 10 ~1 with H20.
After 1 h incubation at 30 °C, the mixture (in a 0-5 ml open Eppendorf
tube) was carefully inverted onto the surface of a transilluminator
(Model TM-36, Ultraviolet Products) and u.v.-irradiated (302 nm) for
10 min at room temperature. The cross-linked products were analysed
immediately, or after immunoprecipitation (see below), by 8 % PAGE.
Urea (4 M) was included in polyacrylamide gels to improve the
resolution of PA and PB2 subunits of the polymerase complex (Lamb
& Choppin, 1976). Gels were dried and the cross-linked products were
revealed by autoradiography.
Immunoprecipitation of polymerase proteins. Rabbit polyclonal
antibodies specific for influenza virus polymerase proteins PB1, PB2
and PA and prepared against bacterially expressed fusion proteins were
used. The antisera directed against PB1 and PB2 were provided by
S. C. Inglis (Digard et al., 1989)and the anti-PA antiserum by I. M. Jones
(Jones et al., 1986). For immunoprecipitation of the polymerase
proteins with the specific antisera, the polymerase complex (after crosslinking to RNA) was disrupted with 1% SDS at 95 °C for 2 min,
followed by dilution in immunoprecipitation buffer A [50 mM-Tris-HC1
pH 7.4, 100 mM-NaCI, 2 mM-EDTA, 1% NP40, 0.5% sodium deoxycholate, 0-2 mM-PMSF, 1 gM-leupeptin (Boehringer), 1 laM-pepstatin
(Boehringer)] to 0-1% SDS and 5 gl of specific antiserum was added.
After incubation on ice for 1 to 2 h, 100 gl of 10 % Protein A-Sepharose
(Pharmacia) in immunoprecipitation buffer A was added to the
antibody-cross-linked product mixture, and the tubes were rotated at
4 °C for 1 h. The Sepharose-bound material was centrifuged and
washed twice in 1 ml of immunoprecipitation buffer B (50 muTris-HC1 pH 7.4, 100 mM-NaC1, 5 mM-MgCI~, 1% Triton X-100, 1%
sodium deoxycholate, 0.1% SDS). The immunoprecipitates were boiled
in 15 gl of SDS PAGE sample buffer for 5min. Proteins in the
supernatant were separated by 8 % PAGE in 4 M-urea and, after
drying, detected by autoradiography.
Results
All three subunits o f the polymerase complex are
covalently cross-linked to the v R N A promoter by
u. v.-irradiation
P h o t o c h e m i c a l c r o s s - l i n k i n g is a p o w e r f u l m e t h o d for
s t u d y i n g p r o t e i n - n u c l e i c acid i n t e r a c t i o n s ( S c h i m m e l &
Budzik, 1977). W e d e c i d e d to use this m e t h o d as an a s s a y
for the specific b i n d i n g o f the influenza A virus R N A
p o l y m e r a s e c o m p l e x to its v R N A p r o m o t e r . Before
testing the effect o f p o i n t m u t a t i o n s in the p r o m o t e r on
p o l y m e r a s e b i n d i n g efficiency, it was first necessary to
establish c o n d i t i o n s for p o l y m e r a s e b i n d i n g to the wildtype p r o m o t e r sequence. W e used s h o r t 32p-labelled
model RNA templates and micrococcal nuclease-treated
viral cores (see M e t h o d s ) as p r e v i o u s l y used in o u r in
vitro s t u d y o f t r a n s c r i p t i o n a n d r e p l i c a t i o n (Seong &
Brownlee, 1992a).
Fig. 1 (lane 1) shows the p r o t e i n s c r o s s - l i n k e d to the
14 n t - l o n g 32P-labelled v R N A p r o m o t e r - c o n t a i n i n g sequence. A t the expected p o s i t i o n s o f the PB1, PB2 a n d
P A p o l y m e r a s e p r o t e i n s (between 8 2 K a n d 86K) two
m a j o r labelled b a n d s ( m a r k e d A a n d C) a n d two m i n o r
b a n d s ( A ' a n d B) were observed. T h e b a n d s were a b s e n t
in a c o n t r o l e x p e r i m e n t in which, i n s t e a d o f the 14 ntl o n g v R N A p r o m o t e r - c o n t a i n i n g sequence, a U15 sequence was used in r e c o n s t i t u t i o n a n d c r o s s - l i n k i n g (Fig.
1, lane 2). This s t r o n g l y suggested t h a t the f o u r labelled
b a n d s (A, A ' , B a n d C) were specific for the v R N A
p r o m o t e r . T h e o t h e r m a j o r labelled p r o d u c t o b s e r v e d in
b o t h lanes was the n u c l e o p r o t e i n ( N P ) w h i c h is the m o s t
a b u n d a n t p r o t e i n o f the viral core a n d binds n o n -
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:27:38
Influenza A virion RNA promoter
1
shifts of the protein-RNA complexes to higher M r were
observed. Treatment of the a2p-labelled cross-linked
products with RNase A resulted in a significant loss
(> 80 %) of the 32p label from the RNA-protein complexes and a slight but definite shift in mobility consistent
with the loss of nucleotides (data not shown).
Competition experiments with the homologous 12 ntlong vRNA promoter and the heterologous U~5 sequence
and phenylalanine tRNA were carried out to provide
further evidence for the specificity of binding of the
presumptive polymerase complex proteins to the promoter sequence. The proteins of the micrococcal
nuclease-treated viral core were cross-linked to the 5' 32p_
labelled 12 nt-long synthetic vRNA promoter sequence
in the absence or presence of increasing amounts of
unlabelled competitors (Fig. 2). Homologous competitor
(100pmol; 12nt-long synthetic vRNA promoter sequence) out-competed the labelled template completely
(lane 6), while in the presence of 100 pmol of UI~ (lane
12) or 200 pmol of phenylalanine tRNA (lane 14) some
cross-linking between labelled RNA and the presumptive
polymerase proteins was still observed. We conclude that
the polymerase proteins preferentially cross-link to the
promoter RNA.
To identify formally the presumptive P protein crosslinked bands, immunoprecipitation with anti-PB 1, antiPB2 and anti-PA antisera was carried out (see Methods).
Before immunoprecipitation the cross-linked products
were treated with 1% SDS to disrupt the polymerase
complex. Under these conditions, anti-PB1, anti-PB2
and anti-PA antisera precipitated the PB1-RNA, PB2-
2
100K
71K
43K
Fig. 1. Comparison of the proteins cross-linked to the vRNA promoter
(lane 1) and Uxa sequence (lane 2). The polymerase complex was
reconstituted with a 14 nt-long a2P-labelled model template RNA
consisting of the 12 nt-long vRNA promoter and two additional G
residues at its 5' end (lane 1) and with a 5' end-labelled U15 sequence
(lane 2). After a 1 h reconstitution period, the resultant polymeras~
RNA complex was u.v.-irradiated for 10 min and the cross-linked
products (presumptive polymerase proteins, see text, major bands A
and C; minor bands A' and B) were analysed by 8 % PAGE in 4 Murea. The positions of protein standards are shown on the right and NP
denotes nucleoprotein.
specifically to RNA very efficiently, as expected (Kingsbury et al., 1987). Compared to the virus-specific noncross-linked protein controls (e.g. PB1, NP) separated on
the same gel and stained with Coomassie blue, minor
1
Competitor:
[
2
3
4
5
12 nt-long vRNA promoter
6
7
8
9
II
10
U15
11
12
13
II
14
tRNA
.[
NP[
I
Probe:
1329
12 nt-long vRNA promoter
Fig. 2. Competition experiment proving the specificity of binding of the presumptive polymerase proteins to the vRNA promoter.
Micrococcal nuclease treated-viral core (4 gl; about 1 pmol of polymerase complex and 20 pmol of NP) was reconstituted and crosslinked with 0'01 pmol of 5' 32P-labelled 12 nt-long synthetic vRNA promoter sequence (lanes 1 to 14) alone (lanes 1, 7 and 13) or in
the presence of increasing amounts of homologous competitor (lane 2, 0"01 pmol; lane 3, 0.1 pmol; lane 4, 1 pmol; lane 5, 10 pmol;
lane 6, 100 pmol), U15 sequence (lane 8, 0.01 pmol; lane 9, 0.1 pmol; lane 10, 1 pmol; lane 11, 10 pmol; lane 12, 100 pmol) or yeast
phenylalanine tRNA (Boehringer) (lane 14, 200 pmol). As the micrococcal nuclease treatment does not remove the endogenous RNA
associated with polymerase complex or NP completely (Seong & Brownlee, 1992a), the actual RNA concentration in the assay is higher
than the amounts give above. This explains why at least 100-fold molar excess of homologous competitor over probe is required to
achieve about 50 % competition (lane 4). The positions of the presumptive polymerase proteins (P) and the nucleoprotein (NP) are
indicated.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:27:38
1330
E. Fodor, B. L. Seong and G. G. Brownlee
1
2
(a)
3
2
1
PB1
PB 1'
/
- - ~
PA
PB1
PA
PB2
PB2 '
4
5
6
7
8
(b)
Fig. 3. Immunoprecipitation of the polymerase proteins after cross-
linking to the vRNA promoter. PB1 and PB' (lane 1), PB2 (lane 2) and
PA (lane 3) were immunoprecipitated after disruption of the polymerase complex with SDS treatment as described in Methods. Low
levels of NP were observed in al! three immunoprecipitates. In addition,
two unidentified bands were observed in the PB1 and PB2 immunoprecipitates. The high M,. band (about 200K) could be a dimer of PB 1
and PB2 cross-linked to the same molecule of RNA.
3
1
2
3
4
PB 1 - PB2 -NP--
R N A and P A - R N A complexes specifically (Fig. 3, lanes
1, 2, 3) with no a p p a r e n t cross-reaction with the other
P proteins. In the PB1 i m m u n o p r e c i p i t a t e a doublet
was observed (Fig. 3, lane 1, PB1, P B I ' ) , which was
also present in the original cross-linked products (Fig. 1,
A, A'). P B I ' p r e s u m a b l y represents a proteolytic
cleavage p r o d u c t of PB1 or the additional f o r m o f PB1
observed by A k k i n a et al. (1991) in infected cells.
T o characterize the interaction between the polymerase
subunits and the v R N A p r o m o t e r further, increasing
concentrations o f NaC1 up to 2 M were used during
reconstitution and cross-linking (Fig. 4a). A b o v e 0"6 MNaC1 (lane 5) there was nearly complete suppression
o f binding of all proteins except PB1 and PB2. The
identity of PB1 and PB2 in the cross-linked products
at high salt concentration (0.4 M-NaC1) was confirmed by
i m m u n o p r e c i p i t a t i o n with specific anti-PB1 and antiPB2 antisera (Fig. 4b, lanes 2, 3). N o labelled protein was
i m m u n o p r e c i p i t a t e d by anti-PA antiserum at 0.4 M-NaC1
(Fig. 4b, lane 4). The ability of PB1 and PB2 to bind is
not significantly affected in the range o f 0.04 to 1.5 MNaC1 (lanes 1 to 7), whereas there is no detectable
binding of P A above 0"2 M-NaC1 (lane 3). Similar results
were observed when KC1 was used instead o f NaC1 (data
not shown). These results suggest that the binding of
PB1 and PB2 subunits to the p r o m o t e r is sequencespecific, and the binding of P A m a y be sequenceindependent (but see Discussion). During complex
f o r m a t i o n between the polymerase complex and the
v R N A p r o m o t e r at low ionic strength, however, all three
subunits are in close contact with R N A allowing
photochemical cross-linking.
Fig. 4. (a) Effect of different salt concentrations on polymerase-RNA
complex formation. The salt concentration was adjusted before
reconstitution and cross-linking by adding NaC1 to 0.04 M (lane 1),
0"1 M(lane 2), 0"2 U (lane 3), 0.4 M(lane 4), 0.6 i (lane 5), 1.0 i (lane 6),
1.5 M(lane 7) or 2"0 M(lane 8). For the origin of the faint high M r band
(about 200K) cross-linked at high ionic strength, see legend to Fig. 3.
A slight increase in the mobility of the cross-linked proteins was
observed from lane 1 to lane 8 possibly caused by the increasing salt
concentration. The positions of the cross-linked PB 1, PA, PB2 and NP
products are marked. (b) Immunoprecipitation of the polymerase
proteins after cross-linking to the vRNA promoter at 0.4 M-NaC1. The
immunoprecipitation is described in Methods except that, after SDS
disruption, immunoprecipitation buffer A with reduced NaC1 concentration (50 mM) was used for dilution. Lane 1, cross-linked products
without immunoprecipitation; lane 2, PB1; lane 3, PB2; lane 4, PA.
Cross-linking of the polymerase complex proteins to the
vRNA promoter carrying mutations
T o test the hypothesis that residues 9 to 11 and possibly
12 of the v R N A p r o m o t e r represent the binding site of
the p r o m o t e r (Seong & Brownlee, 1992b) we used the
cross-linking assay described above. M u t a n t R N A s
carrying individual transversions at each o f the 12
positions of the p r o m o t e r , or an insertion o f a C residue
between positions 4 and 5 (IC4,5), were synthesized by
T7 R N A p o l y m e r a s e transcription of the corresponding
D N A duplex in the presence of [~-3~P]CTP (see
Methods). The R N A templates were reconstituted with
micrococcal nuclease-treated influenza viral cores and
u.v.-irradiated at low ionic strength (50 mM-KC1, 10 m i -
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:27:38
Influenza A virion R N A promoter
1
2
3
4
5
6
7
8
9
10 11 12 13 14
PB1
PA
PB2
NP
Fig. 5. Effect of point mutations or insertions on polymerase-RNA
complex formation. Lane 1 (wild-type), lane 2 ( U I + A 1 ) , lane 3
(C2~A2), lane 4 (G3 -+C3), lane 5 (U4~A4), lane 6 (IC4,5), lane 7
(U5~A5), lane 8 (U6-+A6), lane 9 (U7+A7), lane 10 (C8-+A8),
lane 11 (G9 ~ C9), lane 12 (U10~A10), lane 13 (Cll -+All) and lane
14 (C12 ~+A12). Mutant G3 + C3 (lane 4) shows decreased intensity of
cross-linked proteins due to underloading and not to the inhibitory
effect of this mutation. In repeated experiments (not shown) this mutant
showed a pattern comparable to the wild-type promoter. The positions
of the cross-linked PB1, PB2, PA and NP bands are indicated. NP
was expected to cross-link the wild-type and mutant RNA species (14
or 15 nt long) equally well as NP is known to bind to RNA nonspecifically (Kingsbury et al., 1987) and 14 nt-long oligoribonucleotides
are known to bind purified recombinant NP in a gel shift assay (P. G.
Murray, V. Knott & G. G. Brownlee, unpublished). Bands with
mobility faster than NP may represent proteolytic fragments of NP.
The identity of the other products is unknown.
NaC1, equivalent to conditions somewhere between those
for lanes 1 and 2, Fig. 4a), under conditions where all
three polymerase subunits were cross-linked (see
Methods and above). Because the separation of PA from
PB2 was poor and it was not possible to estimate the
binding of PA by laser densitometry scanning, we
assessed the polymerase binding by measuring the levels
of PB1 and PB2 bands in the crossqinking assay.
Fig. 5 shows that the point mutations at positions 1 to
6 (lanes 2, 3, 4, 5, 7, 8) or an insertion of a C residue
between positions 4 and 5 (lane 6) did not affect the
polymerase binding dramatically compared with the
wild-type promoter (lane 1), as determined by the levels
of PB1 and PB2 bands. Nevertheless, some changes
occurred. Although PB2 remained apparently unaffected
by these mutations, mutations U 4 + A 4 (lane 5) and
U 5 ~ A 5 (lane 7) resulted in a decreased binding of
PB1 (about 35 % of the wild-type as estimated by laser
densitometry scanning). Mutations at positions 7 and 8
(lanes 9 and 10) caused low levels of binding of PB1 and
PB2 (10 to 30 %) as compared to the wild-type. The most
1331
obvious effect was that mutations at positions 9 to 11
(lanes 11 to 13) completely suppressed the binding of
PB1 ( < 5% of the wild-type as estimated by laser
densitometry); however, very low levels of PB2 were
detected (< 10 %). The mutation at position 12 (lane 14)
resulted in suppression of PB1 binding nearly as great
as that arising from mutations at positions 9 to 11,
although some residual PB1 binding remained (6%).
The binding of PB2 with the position 12 mutant was
comparable to that achieved with mutants at positions
9 to 11 (i.e. < 10%). In addition, we tested all three
possible mutations at position 3 (G3-+A3, G 3 ~ C 3 ,
G3-~ U3) because of the strong upregulatory effect of
mutant G3 ~ C3 in in vitro transcription in the absence
of primer (Seong & Brownlee, 1992b). In our binding
assay all three mutants bind both PB1 and PB2 to the
same extent as wild-type (data not shown).
Taken together, these observations showed that (i)
mutations at only four positions (9 to 12) resulted in
severe inhibition of PB 1 and PB2 binding ( < 10 % of the
wild-type), (ii) mutations at positions 7 and 8 produced
decreased binding of both PB1 and PB2 (10 to 30% of
the wild-type) and mutations at positions 4 and 5 affected
only PB1 (35 % of the wild-type), and (iii) mutations at
the remaining positions (positions 1, 2, 3 and 6) or an
insertion of a C residue between positions 4 and 5 caused
no detectable inhibition of the binding of PB 1 and PB2.
Discussion
Recently we proposed a model for the recognition of the
vRNA promoter of influenza A virus (Seong & Brownlee,
1992b). According to this model the vRNA promoter is
bipartite consisting of a binding site and a regulatory site
separated by a U-rich spacer sequence. In this paper, by
studying the effects of mutations in the vRNA promoter
on the binding efficiency of the polymerase complex, we
provide evidence that there is a binding site at the
promoter region as proposed previously. The effect of
mutations on binding efficiency is summarized in Table
1.
Only mutations at positions 9 to 12 (group 1) inhibited
the binding of PB1 and PB2 dramatically which
confirms their importance for polymerase binding. In
our previous study, when we investigated the effects of
point mutations within the promoter on the efficiency of
in vitro transcription, a similar result was observed.
Mutations at positions 9 to 11 caused a significant
(> 90 %) inhibition in all three transcription reactions
(primer-independent, globin mRNA-primed and ApGprimed). The mutation at position 12 had an intermediate
effect on the transcription efficiency (Seong & Brownlee,
1992 b), whereas it showed a strong inhibition of binding
in the binding assay performed here (Table 1), suggesting
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:27:38
E. Fodor, B. L. Seong and G. G. Brownlee
1332
Table 1. Summary of effects of promoter mutations on polymerase binding
Group
Nucleotide
position
Binding
activity (%)*
1
9, 10, 11, 12
Strong
inhibition
< 10
PB1 and PB2
2
4, 5, 7, 8
Intermediate
inhibition
10 to 60
U 4 ~ A 4 and U 5 ~ A 5
only PB1,
U7 ~ A7 and C8 ~ A8
PB1 and PB2
3
1, 2, 3, 6
IC4,5
Weak to
insignificant
inhibition
60 to 100
Effect
Comments
PB1 and PB2
* Wild-type binding was taken as 100 %.
that the C residue at position 12 of the promoter is
involved in polymerase binding.
In addition, mutations at positions 4, 5, 7 or 8 (Table
1, group 2) resulted in reduced binding of PB1 or both
PB1 and PB2. Nevertheless, both polymerase subunits
associated with the promoter to some extent. In the
transcription assay (Seong & Brownlee, 1992b) the
individual mutations at these central residues (positions
4 to 8) showed weak or insignificant effects on transcription, but a triple mutation (U5U6U7-+ A5A6A7) in
this central region caused a significant inhibition of the
ApG-primed reaction. Taking these previous results and
the results of the cross-linking assay together, we deduce
that residues 4 to 8 are probably also important
(although not crucial) for efficient binding, possibly
because they help to stabilize the structure of the
polymerase promoter complex.
In contrast, mutations at positions 1, 2, 3 or 6 or an
insertion of a C residue between positions 4 and 5 in the
promoter (group 3) caused no detectable inhibition of
PB1 and PB2 binding. In our previous transcription
assay (Seong & Brownlee, 1992b) mutations at positions
1 and 2 inhibited transcription at intermediate level,
whereas a mutation at position 3 (G3->C3) or an
insertion of a C residue between positions 4 and 5 caused
strong upregulation of primer-independent reaction.
Thus these mutations in the proposed regulatory site
apparently do not affect binding of the polymerase.
The efficient cross-linking of PB 1 and PB2 to the wildtype vRNA promoter, even at high ionic strength,
suggests that both polymerase subunits specifically bind
to the promoter. The cross-linking of PA to the wild-type
promoter, however, was abolished if the salt concentration was increased above 0.2M-NaC1. This may
suggest that the interaction between PA and RNA lacks
the sequence specificity characteristic of PB1 and PB2.
However, we cannot exclude the alternative, but less
likely, explanation that PA binds to the promoter
specifically, but with lower affinity only at low salt levels.
Nevertheless, it is clear that PA binding properties differ
from those of PB1 and PB2. In the light of these results,
it seems likely that both PB 1 and PB2 play an important
role during promoter recognition and binding, but PA is
not involved directly in this process. We cannot exclude,
however, that an interaction of PA with PB2 or more
likely with PB1 (Digard et al., 1989) is necessary for the
proper conformational arrangement of the RNA binding
domains in PB1 and PB2. Most mutations (see Table 1)
inhibited the binding of PB 1 and PB2 equally suggesting
that both bind to approximately the same region of the
promoter. The binding of PB 1 and PB2 to the promoter
may be cooperative, positioning the RNA between PB1
and PB2.
A computer search of PB1 and PB2 for an RNAbinding consensus (Bandzilius et al., 1989), present in
many RNA-binding proteins, revealed two regions of
PB1 of influenza A viruses (residues 249 to 256
RGFVYFVE and 499 to 503 YGFVA) which closely
resembled the K / R G F / Y G / A F V X F / Y octapeptide of
the RNA-binding consensus. Experiments with mutated
PB1 would be required to test whether these candidate
regions are involved in RNA binding. No RNA-binding
consensus was found in the PB2 of influenza A viruses.
In summary, we show here in cross-linking studies
that both PB1 and PB2 components of the polymerase
complex are involved in recognition of the vRNA
promoter and specifically bind to residues 9 to 12 of the
promoter. This confirms our previously hypothesis of a
binding site in the vRNA promoter (Seong & Brownlee,
1992b).
We thank Drs Stephen C. Inglis and Ian M. Jones for antisera to
PB1, PB2 and PA proteins. This work was supported by an MRC
project grant to G.G.B.E.F. acknowledges the support by a Soros/
FCO Scholarship and the E. P. Abraham Trust.
References
AKKINA, R.K., RICHARDSON, J.C., AGUILERA, M.C. & YANG, C.
(1991). Heterogeneous forms of polymerase proteins exist in
influenza A virus-infected cells. Virus Research 19, 17-30.
BANDZILIUS, R.J., SWANSON, M. S. ~¢ DREYFUSS, CO. (1989). RNA-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:27:38
Influenza A virion R N A p r o m o t e r
binding proteins as developmental regulators. Genes and Development
3, 431-437.
BRAAM, J., ULMAr,~N, I. & KRUG, R. M. (1983). Molecular model of a
eucaryotic transcription complex: functions and movements of
influenza P proteins during capped RNA-primed transcription. Cell
34, 609-618.
DESSELBERGER, U., RACANIELLO, V.R., ZAZRA, J.J. & PALESE, P.
(1980). The 3' and 5' terminal sequences of influenza A, B and C
virus RNA segments are highly conserved and show partial inverted
complementarity. Gene 8, 315-328.
DIGARD, P., BLOK, V. C. & INGLIS, S. C. (1989). Complex formation
between influenza virus polymerase proteins expressed in Xenopus
oocytes. Virology 171, 162-169.
GALARZA, J. M., SOWA, A., HILL, V. M., SZEWCZYK, M., SKORKO, R.
& SUMMERS, D. F. (1991). Expression of influenza A virus RNA
polymerase subunits and NP by recombinant baculoviruses and the
biochemical activities of these proteins. In Eighth International
Conference on Negative Strand Viruses, Charleston, S.C., U.S.A.,
September 1991, p. 64.
ISHIHAMA, A. & NAGATA, K. (1988). Viral RNA polymerases. CRC
Critical Reviews in Biochemistry 23, 27-76.
JONES, I. M., REAY, P. A. & PmLPOTT, K. L. (1986). Nuclear location
of all three influenza polymerase proteins and a nuclear signal in
polymerase PB2. EMBO Journal 5, 2371-2376.
KINGSEURY, D. W., JONES, I. M. & MURTI, K. G. (1987). Assembly of
influenza ribonucleoprotein in vitro using recombinant nucleoprotein. Virology 156, 396403.
KOBAYASHI, M., TUCHIYA, K. NAGATA, K. & ISHIHAMA, A. (1992).
Reconstitution of influenza virus RNA polymerase from three
subunits expressed using recombinant baculovirus system. Virus
Research 22, 235-245.
KRUG, R. M., UEDA, M. tf¢..PALESE, P. (•975). Temperature-sensitive
mutants of influenza WSN virus defective in virus-specific RNA
synthesis. Journal of Virology 16, 79(~796.
LAMB, R.A. & CHOPPIN, P.W. (1976). Synthesis of influenza virus
proteins in infected cells: translation of viral polypeptides, including
1333
three P polypeptides, from RNA produced by primary transcription.
Virology 74, 504-519.
LAMB, R.A. & CHOPHN, P.W. (1983), The gene structure and
replication of influenza virus. Annual Review of Biochemistry 52,
467 506.
MAHY, B. W. J., BARRETT,T., NICHOL, S. T., PENN, C. R. & WOLSTENHOLMr, A.J. (1981). Analysis of the functions of influenza virus
genome RNA segments by use of temperature-sensitive mutants of
fowl plague virus. In The Replication of Negative Stranded Viruses,
pp. 379-387. Edited by D. H. L. Bishop & R. W. Compans. New
York : Elsevier.
MCCAULEY, J. W. & MAHY, B. W. J. (1983). Structure and function of
the influenza virus genome. Biochemical Journal 211, 281-294.
POEM, O., SAt:VAOET, I., DELARt~, M. & TOROO, N. (1989).
Identification of four conserved motifs among the RNA-dependent
polymerase encoding elements. EMBO Journal 8, 3867 3874.
ROBERTSON, J. S. (1979). 5' and 3' terminal nucleotide sequences of the
RNA genome segments of influenza virus. Nucleic Acids Research 6,
3745-3757.
SCHIM~Ca~L, P.R. & Bb~ZIK, G.P. (1977). Photo-cross-linking of
protein-nucleic acid complexes. Methods in Enzymology 46, 168 180.
SEONG, B.L. & BROWNLEE, G . G . (1992a). A new method for
reconstituting influenza polymerase and RNA in vitro: a study of
promoter elements for cRNA and vRNA synthesis in vitro and viral
rescue in vivo. Virology 186, 247-260.
SEON~, B. L. & BROWNLEE, G. G. (1992b). Nucleotides 9 to 11 of the
influenza A virion RNA promoter are crucial for activity in vitro.
Journal of General Virology 73, 3115-3124.
ULMAYrY, I., BRONI, B. A. & KRUC, R. M. (1981). Role of two of the
influenza virus core P proteins in recognizing cap 1 structures
(mVGpppNm) on RNAs and initiating viral RNA transcription.
Proceedings of the National Academy of Sciences, U.S.A. 78,
7355-7359.
(Received 31 December 1992; Accepted 1 March 1993)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:27:38