Journal qf General Virology (1994), 75, 3477 3484. Printed in Great Britain
3477
Functional reconstitution in lipid vesicles of influenza virus M2 protein
expressed by baculovirus: evidence for proton transfer activity
Cornelia Schroeder,*t Chris M. Ford,:~ Stephen A. Wharton and Alan J. Hay
Division o f Virology, National Institute for Medical Research, The Ridgeway, Mill Hill, London N W 7 1AA, U.K.
The influenza virus M2 protein was expressed from
a recombinant baculovirus in Spodoptera frugiperda
Sf9 cells, purified and reconstituted into artificial
membrane vesicles. The specific inhibitor amantadine
overcame the toxic activity of the protein and boosted
the rate of M2 synthesis by a factor of 10, allowing yields
of about 1 mg of purified M2 protein per g of Sf9 cells.
M2 protein expressed in this system was phosphorylated
and palmitoylated and displayed properties similar to
the authentic virus protein. Purified wild-type M2
protein and an amantadine-resistant mutant M2 (M26)
with a deletion in the trans-membrane domain (amino
acids 28 to 31) were incorporated into lipid vesicles,
which were loaded with the fluorescent pH indicator
pyranine. On imposition of an ionic gradient, M2 caused
a decrease in intravesicular pH, which was susceptible to
inhibition by 0.1 to 1 gM-rimantadine or N-ethylrimantadine. M26 behaved similarly but exhibited the
expected drug resistance. These experiments indicate
that isolated M2 functions as an ion channel and
demonstrates in vitro M2-mediated proton translocation.
Introduction
transport to the plasma membrane (Sugrue et al., 1990a;
Ciampor et al., 1992a; Takeuchi & Lamb, 1994; Ohuchi
et al., 1994).
Using the conformation of HA as a probe to monitor
the pH modulating activity of M2 in infected cells it was
estimated that M2 elevates the pH of trans Golgi vesicles
by up to 0.8 pH units (Grambas & Hay, 1992). Mutations
in M2 that confer resistance to amantadine may either
enhance or reduce the pH-modulating activity of the
protein and M2 proteins with a higher intrinsic activity
can complement less active M2 proteins in supporting
the maturation of acid-labile HAs (Grambas et al., 1992;
Grambas & Hay, 1992). The correlation between
differences in the pH-modulating activity of M2 and
differences in the pH stabilities of the respective HAs of
certain influenza virus strains indicates the importance of
compatibility between the two proteins. Furthermore,
since the amantadine sensitivity of HA mutants increases
with the acid lability of their HAs (Grambas & Hay,
1992) it has conversely been possible to select in the
presence of amantadine a mutant with an acid-stable HA
(Steinhauer et al., 1991). The M2 protein in the plasma
membrane can also counteract a trans-membrane pH
gradient, reducing the intracellular pH to that of the
outside medium (Ciampor et al., 1992b).
Direct measurements of ion channel activity in voltageclamp experiments demonstrated that M2 expressed in
Xenopus laevis oocytes possesses a pH-activated, Na +selective conductance (Pinto et al., 1992; Wang et al.,
1993). Under the conditions used in these experiments it
The influenza virus M2 protein is a minor virus envelope
protein expressed from a spliced mRNA encoded by the
M gene (Lamb & Choppin, 1981). The 96 amino acid
polypeptides associate to form disulphide-linked homotetramers (Sugrue & Hay, 1991; Holsinger & Lamb,
1991) which are modified by the addition of palmitate
and phosphate (Sugrue et al., 1990b; Veit et al., 1991).
As the target of the antiviral compounds amantadine
and rimantadine (Hay et al., 1985) M2 was implicated in
virus uncoating (Kato & Eggers, 1969; Bukrinskaya et
al,, 1982; Wharton et al., 1990; Martin & Helenins,
1991). More recently, it has been proposed that during
endocytosis the M2 ion channel is responsible for
allowing the passage of protons from the acid endosome
into the virion, thereby promoting a low pH-induced
dissociation of the matrix protein from the nucleocapsid
(Hay, 1989; Wharton et al., 1990). Such a role of M2
would be analogous to its ability to reduce the acidity of
the trans Golgi network and thereby protect intracellularly cleaved haemagglutinin (HA) against a premature low pH-induced conformational alteration during
t Present address: Institut ffir Virologie, Medizinische Fakultfit
(Charit6)der Humboldt-Universit/itzu Berlin,Schumannstrasse20/21,
10098 Berlin, Germany,
:~ Present address: Waite Agricultural Research Institute, The
Universityof Adelaide,Glen Osmond,South Australia 5064.
0001-2621 © 1994SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:13:27
3478
C. Schroeder and others
was not possible to resolve proton currents. However, a
peptide corresponding to the trans-membrane domain of
M2 incorporated into planar lipid bilayers has been
shown to promote amantadine-sensitive proton translocation at mitlimolar proton concentrations (Duff &
Ashley, 1992).
We report here the use of a fluorescent dye assay to
demonstrate, under mild conditions, the proton translocation activity of purified M2 reconstituted into lipid
vesicles. The M2 protein was expressed under the control
of the baculovirus polyhedron promoter in Spodoptera
frugiperda Sf9 cells. The proton translocation activity of
the wild-type M2 protein was specifically inhibited by
rimantadine whereas the activity of a drug-resistant
mutant protein was unaffected.
Methods
Virus and cells. Wild-type Autographa cal!fornica multinucleocapsid
nuclear polyhedrosis virus (AcMPNV; polyhedrin-positive) was obtained from Dr M. D. Summers (Texas A & M University, Tex.,
U.S.A.). BacPAK6 virus was a product of Clontech Laboratories. The
St9 cell subline RE was provided by Dr H. Reil/inder (Max-PlanckInstitut f~r Membranbiophysik, Frankfurt, Germany). Trichoplusia ni
cells (Hi5) were obtained from JRH Biosciences.
Sf9 cells were grown in TC100 medium (purchased from Gibco or
Flow Laboratories) supplemented with L-glutamine (292 mg/l), fetal
calf serum (FCS; 10%), penicillin (60 mg/l), streptomycin (100 mg/1)
and in some cases, amphotericin (Flow Laboratories, 2.5 mg/1). Hi5
cells were grown in SF-900 II serum-free medium (Gibco). Cell
cultivation, virus propagation and titration were carried out as
described by Luckow & Summers (1989).
Antisera. Rabbit antisera were raised against peptides corresponding
to the N-terminal 24 amino acids (R7 and R53), to the C-terminal 16
amino acids (R54) and to amino acids 60 to 76 (R15) of the M2 protein
of influenza virus strain A/chicken/Rostock/34 (H7N1).
lnhibitors. All compounds were in the form of hydrochlorides.
Amantadine was a product of Serva, rimantadine was a gift from Dr I.
Sire (Hoffmann-La Roche) and N-ethylrimantadine was provided by
Dr M. K. Indulen (August Kirchenstein Institute of Microbiology,
Latvian Academy of Sciences, Riga, Latvia).
Construction o f baculovirus recombinants. The plasmid pVL941
(Luckow & Summers, 1989) was obtained from Dr M. D. Summers.
An EcoRI-Sall fragment encompassing the M2 sequence of influenza
virus A/Weybridge (H7N7) was converted to a BamHI fragment by
attaching appropriate linkers and inserted into the BamHI site of
pVL941, generating the recombinant pVLM2. This plasmid was
cotransfected with AcMNPV wild-type DNA into Sf9 cells using
Lipofectin from BRL (Felgner et al., 1987). Polyhedrin-negative
plaques were picked and M2-producing clones were plaque-purified.
The M2-expressing recombinant baculovirus used was designated
CFM2.
A deletion mutant of M2 (lacking amino acids 28 to 31) was
constructed by PCR mutagenesis of pVLM2. Primers complementary
to the sequences flanking the intended deletion were extended with Vent
polymerase (New England Biolabs). The product was phosphorylated
with T4 polynucleotide kinase (Boehringer Mannheim), ligated with T4
DNA ligase (New England Biolabs) and used to transform competent
Escheriehia coli DH5 cells. Mutant plasmid DNA and Bsu36I-linearized
BacPAK6 DNA (Kitts et al., 1990) were cotransfected into St9 cells by
lipofection. Plaques were screened for typical M2-specific morphological alterations of the SO RE cells. One plaque was purified and the
recombinant designated CSM26.
Metabolic labelling and immune precipitation. All incubations were
done at 27 °C. SO cells were seeded onto 24-well plates and infected
with CFM2 virus at a m.o.i, of 2 p.f.u./ml. At different times postinfection (p.i.) the medium was exchanged for the appropriate depletion
medium supplemented with 10% dialysed FCS: TC100 without
cysteine and methionine (Gibco-BRL) for [3SS]cysteine incorporation;
TC100 supplemented with 5 laM-sodium pyruvate for labelling with
[3H]palmitic acid, and phosphate-free TCI00 for labelling with
[a2p]phosphate. After 1 h this medium was replaced by 50 or 100 gl of
depletion medium containing the appropriate radiolabelled compound
[2.2 MBq [aSS]cysteine (Amersham); 2 MBq [aH]palmitic acid (Amersham); 2 M B q [32P]orthophosphate (Amersham)] and cells were
incubated for 30rain or l h. Monolayers were then washed and
incubated at 27 °C with chase medium (1 ml TC100 plus 10% FCS).
Finally, monolayers were washed with ice-cold PBS pH 6-6, and
extracted with 100 lal of 1% NP40 in 150 mM-NaCl, 20 mM-Tris HCI
pH 7-5 for 10 min at 4 °C. For immune precipitation the extract was
diluted into 12 volumes of binding buffer (0.5 % NP40, 150 mM-NaC1,
1 mM-EDTA, 0-25 % BSA, 20 mM-Tris-HCt pH 8.0) containing antiM2 antiserum (1:100) and precipitated with protein A Sepharose
(Pharmacia), as described by Sugrue et al. (1990b).
Purification o f M2 protein. Suspension cultures of exponentially
growing Sf9 cells (10n/ml) were concentrated by centrifugation and
infected with baculovirus CFM2 at a m.o.i, of 2 p.f.u./cell. After a 1 h
adsorption period the cells were taken up in fresh medium containing
1 gg/ml amantadine and cultured for 48 h at 27 °C. The cells (2 to 5 g
per 1 of cell culture) were harvested, washed three times in HEPESbuffered saline (HBS; 10mM-HEPES, 150mM-NaC1 pH7-8), and
taken up in 10 volumes of ice-cold 10 mM-HEPES pH 7-8 supplemented
with 1 mM-PMSF, 0.02% trypsin inhibitor (soybean) and 10 p,g/ml
aprotinin (all from Sigma). This protease inhibitor combination was
used until the KCI extraction step. The cells were disrupted in a Dounce
homogenizer and the nuclei and remaining intact cells were pelleted.
The supernatant was made 70% (w/v) in sucrose, overlaid with HBS
pH 7.8 and centrifuged for 12 h at 20000 r.p.m, in a Beckman SW-28
rotor. The membrane fraction at the interface was pelleted at
26000 r.p.m, for 90 min, washed once in HBS pH 7.8, and extracted
with 1 M-KC1 in HBS pH 7.8 for 1 h at 4 °C. The membrane pellet was
washed once in HBS pH 7.8 containing a mixture of low M r protease
inhibitors [1 mM-PMSF, 10 gM-E-64 (trans-epoxysuccinyl-L-leucylamido(4-guanidino)-butane), 1 gM-bestatin, 1 laM-pepstatin A (Sigma)],
which was used throughout the following purification steps except
immunoaffinity chromatography. The membranes were extracted for
l h at 4°C with 4mra-n-octyl glucoside (OG; 1-O-n-octyl fl-Dglucopyranoside; Sigma), pelleted once more as above, and finally M2
was extracted with 40 mM-OG for 1 h at 4 °C.
M2 protein was purified either directly using an immunoaffinity
column, or by Phosphogel (Pierce) followed by immunoaffinity
chromatography. After application to activated Phosphogel at pH 4.0
the column was washed with 100 raM-sodium acetate pH 4.0 containing
40 mM-OG and eluted with 20 mM-NaH2PO 4 or Na2HPO 4 pH 7-0 and
40 mM-OG. M2-containing fractions were pooled and the buffer was
exchanged for HBS pH 7.0 containing 40 mM-OG. The concentrate was
applied to an immunoaNnity column [Affi-Gel Hz (Bio-Rad) to which
was coupled immunoglobulin purified from R53 serum]. Th~ colt~mn
was washed with HBS pH 7.0 and 40 mM-OG, and eluted in 1 ml
fractions with 100mM-glycine-HCl pH2.8 and 4 0 m r ~ 0 G . The
fractions were immediately neutralized, M2-containing fractfons were
pooled and concentrated and the buffer was exchanged for HBS pH 7.0
with 40 mM-OG.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:13:27
P r o t o n transfer by influenza virus M 2
The M2 deletion protein (M2J) was purified from 1 1 of Hi5 cells
infected with CSM26. The 40 mM-OG extract was applied directly to a
freshly prepared immunoaffinity column.
Protein detection. The concentration of M2 in semipure and purified
preparations was estimated from the A280 measurement using the
formula of Beaven & Holiday (1952). Concentration estimates took
into account the level of impurities as detected by silver staining.
Estimates of M2 concentrations in membrane and cell extracts were
made on the basis of Western blots of purified and crude samples run
in parallel on polyacrylamide gels.
Gel electrophoresis and Western blotting. Samples in 2 x Laemmli
buffer were boiled for 2 min in the presence or absence of 50 mM-DTT
and run on 12% or 10% polyacrylamide gels with [14C]methylated
Rainbow markers (Amersham) as M r standards. Gels were processed
for fluorography by immersing in Amplify (Amersham) or Westernblotted on Immobilon P membranes (Millipore). Blots were blocked
overnight in 1% dried skimmed milk in PBS, treated with primary
antibodies (diluted 1 : 500) for 2 h, washed five times, then treated for
1 h with iodinated donkey anti-rabbit antibody (Amersham) diluted
1 : 1000 into 1% skimmed milk in PBS. Blots and Amplify-treated gels
were dried and exposed to pre-flashed X-ray film at - 70 °C.
Reconstitution into lipid vesicles. M2 was reconstituted into liposomes
based on procedures of Ruigrok et al. (1986) and Dencher et al. (1986)•
A solution containing 0-85 mg of L-c~-dimyristoyl phosphatidylcholine
(DMPC; Sigma), 0•15mg L-c~-phosphatidyl e-serine (Sigma) and
0.7 mol % (with respect to lipids) valinomycin in chloroform methanol
(95 : 5 v/v) in a glass tube was evaporated to a thin film under a stream
of argon, dried under vacuum overnight and taken up in 250 lal o f
potassium phosphate buffer (10 mM-K2HPO4, 50 mM-K~SO4) pH 7.4,
supplemented with 40 mM-OG, with or without 50 lag of M2 or M26
protein. The molar ratio of lipid to protein was 250. Vesicles were
formed by sequential dialysis against increasing volumes of the same
buffer (Ruigrok et al., 1986) containing 0.04 % sodium azide. The azide
was omitted and amberlite XAD-2 was included in the final dialysis
buffer.
Fluorescence assay of intravesieular pH. Vesicles were loaded with
2 mM-pyranine (Molecular Probes) by sonication (3 x 10 s in a bath
sonicator). Excess dye was removed by dialysis or by passage through
a column of superfine Sephadex G50 (Pharmacia). 5 mg/ml DPX (pxylene-bis-pyridinium bromide; Molecular Probes) was included in the
assay buffer to quench fluorescence of residual surface-bound dye.
Intravesicular pH was calculated from the ratio of the emission
intensities at 510 nm for the two excitation wavelengths 403 and
460 nm, as described by Dencher et al. (1986). A calibration plot was
derived from the emission ratios of pyranine in buffers of varying pH.
Vesicles and reaction buffers were preincubated at 12 °C. Reactions
were run at 12 °C in a Shimadzu PC5001 fluorimeter using the double
wavelength recording mode. Readings were taken every 10 s for 290 to
360 s. The reaction was started by adding a 10 lal vesicle suspension to
600 lal of sodium buffer (10 mM-Na2HPO4, 50 mM-Na2SO 4 pH 7.4) in a
fluorescence cuvette. A mixture of valinomycin (5 laM) and nigericin
(10 laM) was added to equilibrate the pH gradient (Pressman, 1976).
Initial slopes of ratio plots were calculated from the difference in the
intensity ratios at 60 s and 10 s after starting the reaction, divided by
50.
Results
Effects of amantadine on M2 expression in SJ9 cells
During initial experiments it became clear that laboratory variants of the Sf9 cell line differed in their
sensitivity to toxicity caused by the M2 protein expressed
3479
by the recombinant baculovirus CFM2. The most
resilient Sf9 line, RE, was selected for the production of
M2 (Fig. 1). Later in the course of this work a 'high
producer' insect cell line, T. ni Hi5, which was superior
to Sf9 RE (Fig. 1 a, b) became available and was used to
produce the amantadine-resistant mutant M2 protein,
M2J (Fig. 1 c). Expression of the wild-type M2 protein
was not enhanced appreciably by increasing the m.o.i.
above 2. Incubation in the presence of 1 gg/ml amantadine, however, increased synthesis of M2 by about 10fold whether assessed by [3SS]cysteine incorporation or
by Western blot analysis (Fig. 1a, b). The expression of
the amantadine-resistant deletion (amino acids 28 to 31)
mutant M26 (Hay et al., 1985) was not susceptible to
amantadine stimulation (data not shown)• However, in
Hi5 cells the level of M26 expression was comparable to
that of wild-type M2 in the presence of amantadine (Fig.
1 c).
Characteristics o f M 2
Influenza virus M2 protein is a palmitoylated phosphoprotein (Sugrue et al., 1990b; Sugrue & Hay, 1991).
Metabolic labelling of CFM2-infected Sf9 cells with
[32p]phosphate or with [3H]palmitate demonstrated that
the M2 protein expressed in insect cells was phosphorylated (Fig. 2 a, lanes 6 to 8) and palmitoylated (Fig. 2 a,
lanes 2 and 3). The M2 monomer appeared as a doublet
(c)
(a)
1
2
3
4
5
6
7
8
1
69K 46K
97K
69K
30K
46K
30K
21K
2
21K
14K
(b)
i~ ~ ~ i ~
• ~i~
~
~ ,
97K
69K - ~!!'!~
46K
30K
21K
Fig. 1. Expression of M2 and M2J in Sf9 and Hi5 cells. Multiplicitydependent expression o f M2 from CFM2-infected Sf9 RE cells (a) or
Hi5 cells (b) in the presence (lanes 2, 4, 6, 8) or absence (lanes 1, 3, 5,
7) of amantadine at 1 lag/ml. Cultures were harvested 48 h p•i. and
processed for PAGE. Western blots were developed with anti-M2
rabbit serum R54. The m.o.i, were: 10 (lanes 1, 2); 2 (lanes 3, 4); 0.4
(lanes 5, 6); 0-08 (lanes 7, 8). (c) Expression of M2J (m.o.i. of
2 p.f.u./cell) in Sf9 RE cells (lane 1) and Hi5 cells (lane 2), 48 h p.i• with
CSM26~. Positions of M r markers are indicated on the left.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:13:27
3480
C. Schroeder and others
(a)
(a)
1
2
3
4
5
6
7
8
1
2
3
3*
4
9
200K
97K 69K
46K 30K 21K (b)
14K
1
2
3
(b) i
u
1
2
3
4
5
6
7
3*
97K
(3)
69K
46K
(2) ~ ....
Fig. 2. Palmitoylation, phosphorylation and oligomerization of the M2
protein expressed in Sf9 cells. (a) Palmitate radiolabelling: non-infected
Sf9 ceils (lane 1) and CFM2-infected cells in the absence (lane 2) or
presence (lane 3) of amantadine (1 lag/ml) were labelled with
[~H]palmitate. Phosphate radiolabelling: non-infected (lane 5) and
infected (lane 6) amantadine-treated cells were labelled with [z2P]phosphate at 24 h p.i. Infected cells without (lane 7) or with (lane 8)
amantadine (1 gg/ml) were labelled at 48 h p.i. Lane 9, infected
amantadine-treated Sf9 cells were labelled with [35S]cysteine. Lane 4,
35S-labelled influenza virus Weybridge strain M2, immune-precipitated
with R7 from influenza virus-infected chick cells. Boiled, reduced
samples were run on 12 % polyacrylamide gels. M~ standards are on the
left. (b) M2 oligomers: boiled, unreduced samples were run on a 10 %
polyacrylamide gel. Weybridge strain M2 was immune-precipitated
with anti-M2 rabbit sera R7 (lane 1) and R15 (lane 2) from pulse-chase
35S-labelled CFM2-infected S19 cells or from influenza virus-infected
chick embryo cells (lane 3). The positions of M2 dimers (2), trimers (3)
and tetramers (4) are indicated.
(Fig. 2 a , l a n e 4), since a p r o p o r t i o n o f the p a l m i t a t e is
lost d u r i n g b o i l i n g a n d r e d u c t i o n . T h e b a n d o f l o w e r
m o b i l i t y c o n t a i n s the p a l m i t o y l a t e d species ( S u g r u e et
al., 1990b).
A n a l y s e s o f M 2 s y n t h e s i z e d in r e c o m b i n a n t b a c u l o v i r u s - i n f e c t e d Sf9 cells s h o w e d the p r e s e n c e o f m o n o m e r s
(Fig. 2a), d i m e r s , t r i m e r s a n d t e t r a m e r s (Fig. 2b), as for
the M 2 s y n t h e s i z e d i n i n f l u e n z a v i r u s - i n f e c t e d cells. I n
p u r e a n d relatively c o n c e n t r a t e d M 2 p r e p a r a t i o n s
(200 g g / m l ) h i g h M r species, o f a p p r o x i m a t e l y 2 0 0 K ,
p r e d o m i n a t e d (Fig. 3 c ) ; s i m i l a r b a n d s h a v e b e e n o b served i n a n a l y s e s o f i n f l u e n z a v i r u s v i r i o n p r o t e i n s . By
several criteria, therefore, M 2 expressed f r o m rec o m b i n a n t b a c u l o v i r u s in Sf9 cells r e s e m b l e d the a u t h entic i n f l u e n z a virus p r o t e i n .
30K
21K
14K
(c)
200K
97K 69K 46K 30K
21K 14K
Fig. 3. Immunoaffinitypurification of M2 protein. A detergent extract
(60 ml) from CFM2-infected Sf9 cells was applied to an immunoaffinity
column. After washing, 1 ml fractions were eluted as described in
Methods and 10 gl samples were applied to 12% polyacrylamide gels.
One gel was silver-stained (a) and the other blotted for 30 rain (b) and
reblotted for another 90 rain (c). Lane i, detergent extract input; lane
u, unbound material; lanes 1 to 7, fraction numbers; lane 3",
unreduced fraction 3. By OD measurements, sample 3 was found to
contain 0.9 ~g protein.
Purification of M2 protein
M 2 p r o t e i n was p u r i f i e d f r o m Sf9 R E cells as d e s c r i b e d
in Methods. The membrane fraction was treated with
1 M-KC1 a n d 4 m M - O G to r e m o v e p e r i p h e r a l p r o t e i n s
a n d the M 2 p r o t e i n was e x t r a c t e d q u a n t i t a t i v e l y w i t h
40 m M - O G . I m m u n o a f f i n i t y c h r o m a t o g r a p h y gave 95 %
o r g r e a t e r p u r i f i c a t i o n , as j u d g e d b y silver s t a i n i n g o f the
p r o t e i n , a n a l y s e d o n a p o l y a c r y l a m i d e gel. T h e assignment of silver-stained bands of the samples from
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:13:27
Proton transfer by influenza virus M2
1-7
6.6
1.7
6.6
(a)
3481
1-5
%
1-5
6-8
6"8
1.3
1.3
7"0
0
1.1
0
7.0
O
~z
O
0-9
7.2
0.7
0.9
i
0.7
7"4
0.5
0,3
1.7
0.5
I
I
0
100
I
200
Time (s)
I
300
17.6
400
6.6
(c)
1.5
6.8
1.3
7.0
1-1
e~
0
e~
E 0-9
7.2
0.3
0
100
200
Time (s)
300
t
7.2
7-4
7-6
400
Fig. 4. Ion translocation activity of M2 and M26
proteins reconstituted into lipid vesicles. (a) Proteinfree control vesicles; (b) M2-containing vesicles; (c)
M26-containing vesicles. Reactions were initiated at
30 s by introducing liposomes into sodium buffer
(12 °C) and terminated by adding a mixture of
valinomycin (5 laM) and nigericin (10 lau) at 330 s
except (a), the control, terminated at 340 s, (b) plus
1 ~M-rimantadine at 340 s or 10 gM at 300 s; plus
0"1 gM-N-ethylrimantadine at 290s or 101aM at
360s. Symbols: (11), control; (V), plus 0.1 gUrimantadine; (2x), plus 1 ~tM-rimantadine; (~),
plus 10 laM-rimantadine; (IN), plus 0.1 laM-N-ethylrimantadine; ([]), plus 0.67 laM-N-ethylrimantadine;
(0), plus 10 laM-N-ethylrimantadine.
0-7
7.4
0-5
0-3
0
I
100
I
200
Time (s)
I
300
7.6
4OO
c o l u m n fractions 2, 3 and 4 (Fig. 3a) as m o n o m e r s ,
dimers and tetramers is m a d e on the basis o f bands at
equivalent positions in a parallel gel (Fig. 3 b, c). M o s t o f
the m o n o m e r transferred within 30 min (Fig. 3 b) while
the tetramer and the high M r ( > 200K) species in the
unreduced, unboiled sample (3") o f fraction 3 transferred
m o r e slowly (Fig. 3 c ) .
In larger scale preparations the detergent extract was
enriched for p h o s p h o p r o t e i n s using a c o l u m n o f immobilized Fe 3+ ions (Phosphogel) before immunoaffinity
chromatography.
Proton translocation activity of M2 protein in lipid
vesicles
We examined the suitability o f various lipid mixtures and
reconstitution techniques for producing M 2 vesicles in
which ion translocation activity could be assayed. The
following m e t h o d was established: detergent solutions o f
M 2 (purified by Phosphogel followed by immunoaffinity
c h r o m a t o g r a p h y ) and M2O (immunoaffinity-purified)
were reconstituted with a mixture o f D M P C , phosphatidyl serine and valinomycin (0-7 tool % with respect
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:13:27
3482
C. Schroeder and others
Table 1. Initial acidification rates of vesicles containing
M 2 and M2c~
Compound
None
Rimantadine
N-ethylrimantadine
Initial acidificationrate*
Concentration
(i/M)
Control
M2
M2O
0
0.1
l
10
O"1
0"67
10
0-15
0.28
0-67
1.5
ND'~
0"78
ND
69
4-5
3-9
8.2
4"2
3"0
8"2
7.0
7.0
6"2
6.2
ND
7'0
NO
* Expressedas 103timesthe emissionintensityper second,calculated
from the data presented in Fig. 4 as described in Methods.
t NO, Not done.
to lipids) in a potassium phosphate buffer containing K +
as the only cation. OG was removed by extensive
dialysis. Control vesicles without protein were prepared
in parallel. The vesicles were loaded with 2 mu-pyranine
(a fluorescent pH indicator; Dencher et al., 1986) and
fluorescence caused by residual externally bound dye was
quenched by DPX.
The driving force for proton flux across a membrane
can either be a pH gradient or a membrane potential
(Garcia et al., 1984). The incorporation of valinomycin
in the membrane of K+-containing vesicles allowed the
creation of a diffusion potential on exposure of these
vesicles to the sodium buffer. The internal pH was
monitored by recording, every 10 s, emission intensities
at 510nm for two excitation wavelengths, 403 and
460 nm (Fig. 4; Dencher et al., 1986). Reactions were
terminated by adding an excess of the K+/H + ionophore
nigericin together with the K + ionophore valinomycin to
collapse the proton gradient. This elicited an unexplained
transient (10 s) drop in pH in the control vesicles before
pH equilibration.
After introducing the vesicles into the sodium buffer,
proton influx into M2- and M2g-containing vesicles was
initially rapid and the pH reached an equilibrium within
4 min (Fig. 4b, c). The plot resembles a saturation curve.
Initial acidification rates, represented by the slopes, were
calculated as described in Methods and are shown in
Table 1. The internal pH of M2 and M2g vesicles,
determined using a calibration plot, decreased by about
0.5 units (Fig. 4b, c). Under the same conditions the pH
in the control vesicles (Fig. 4a) exhibited a slow linear
decrease of 0.06 units over 5 min and did not reach
equilibrium. In the presence of rimantadine the pH
decreased more rapidly, at 1 jam by 0"19 units and at
10 JaM by 0"33 units, indicating that the drug itself
promoted detectable vesicle acidification (Fig. 4a).
Both the initial acidification rate and the equilibrium
pH of M2-containing vesicles were influenced by inhibitors of M2. However, only the initial rates were
obtained under identical ionic conditions. Rimantadine
at 0"1 and 1 jaM inhibited the acidification of M2 vesicles
to a similar extent whereas 10 jaM-rimantadine enhanced
acidification (Fig. 4b; Table 1). Exactly the same effects
of rimantadine (inhibitory at 1 jaM but stimulatory at
10 jaM) were observed when the reaction was not
terminated with ionophores but instead was re-initiated
by addition of H3PO4, lowering the external pH by about
1 unit and thereby causing a further decrease in the
internal pH until a new equilibrium was reached (data
not shown).
N-ethylrimantadine causes a similar degree of inhibition of influenza virus infection as rimantadine
(Indulen et al., 1979). 0"67jaM-N-ethylrimantadine
caused a somewhat greater reduction (2.3-fold) in the
initial rate of the pH change in M2-containing vesicles
than did 1 jaM-rimantadine (l'8-fold). Stimulatory concentrations (10 jaM) of both compounds increased the
rate 1.2-fold (Fig. 4b; Table 1).
M26-containing vesicles responded differently to these
inhibitors. Acidification proceeded similarly in the
absence and in the presence of rimantadine or Nethylrimantadine (Fig. 4c; Table 1), consistent with the
drug resistance of the protein. However, 1 jaM-rimantadine, but not 0-67 jaM-N-ethylrimantadine, clearly elevated the equilibrium pH (Fig. 4c), although the initial
rate was only slightly reduced; 1.1-fold as compared to
1.8-fold in the case of wild-type M2-containing vesicles
(Table 1). The reason for this is not clear.
Discussion
The baculovirus-Sf9 cell expression system was chosen
to produce the M2 protein since it has been successfully
exploited for the other influenza virus envelope proteins
(Kuroda et al., 1986; Weyer & Possee, 1991) as well as
for ion channel proteins (e.g. Kamb et al., 1992; Carter
et al., 1992). M2 expressed by the recombinant baculovirus resembled the authentic virus protein in that it was
phosphorylated and palmitoylated and formed homotetramers (Sugrue & Hay, 1991; Sugrue et al., 1990b;
Veit et al., 1991 ; Holsinger & Lamb, 1991). In pure and
relatively concentrated preparations very large M2
complexes were observed, the functional significance of
which is not known. Under optimum conditions the yield
per I of Sf9 cell suspension was 2 to 4 mg of M2 protein,
of which about 50 % was recovered in purified form.
Two- to threefold higher yields were obtained using Hi5
cells.
The alleviation of M2 toxicity to the insect cells by
amantadine was due to inhibition of M2 function, since
the drug enhanced M2 synthesis about 10-fold. Black et
al. (1993) reported a similar drug-induced enhancement
of the expression of the M2 protein of influenza virus
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:13:27
Proton transfer by influenza virus M 2
strain A/Ann Arbor/6/60, which differs from the
Weybridge strain M2 by 15 amino acids (Ito et al., 1991).
In contrast, amantadine treatment has no effect on M2
synthesis in virus-infected cells or M2 expression in
mammalian cells or E. coli (F. Geraghty, A. Hayhurst &
A. Hay, unpublished results). Thus, the effect appears to
result from M2 activity on the one hand and properties
of the baculovirus-insect cell system on the other. M2 in
the plasma membrane of mammalian cells can dissipate
a pH gradient between extra- and intracellular environments, causing a decrease in cytoplasmic pH (Ciampor et
al., 1992b). The pH of the insect cell medium is
particularly low (pH 6.2 to 6-4); thus changes in the
internal ionic environment may well underlie the distinctive M2-related phenomena in these cells. The
resemblance between M2-induced changes in cell shape
to changes produced by 10 jaM-monensin (unpublished
data), an ionophore known to substitute for M2 in
regulating pH within the transport pathway (Sugrue et
al., 1990a), is consistent with this interpretation.
The assay of M2 activity in the reconstituted membrane
system employed a K + gradient rather than a proton
gradient, emulating the relationship between the cytoplasm (high [K+]) and the trans Golgi compartment (high
[Na+]). M2- and M2d-mediated proton fluxes (initial
rates) were 46 times faster than fluxes into control
vesicles. In view of the estimated purity of M2 and M2d
proteins of 95 %, the presence of traces of other channelforming proteins cannot be excluded. However, the
specific inhibition by low concentrations of rimantadine
(0.1 to 1 jaM) and the drug resistance of the mutant
protein M2d support the interpretation that the proton
translocation activity was due to the viral protein.
In response to the K + diffusion potential and against
the proton gradient, rimantadine and N-ethylrimantadine lowered the pH in protein-free control vesicles,
which is consistent with the ability of the hydrochloride,
not only the free base, to penetrate lipid bilayers (Duff et
al., 1993). The concentration ranges for the specific antiM2 and non-specific effects of rimantadine were similar
to those observed in virus-infected cells, where rimantadine, at 5 jaM and above, counteracts its own specific M2mediated effect (maximum at 0.5 jaM) on trans Golgi pH
(Grambas & Hay, 1992). Other amphiphilic primary
amines and weak bases have also been shown to act this
way (Sugrue et al., 1990).
Though proton influx is driven by the K + gradient, it
makes only a minor contribution to the dissipation of
this gradient. Whether, like M2 expressed in X. laevis
oocytes (Pinto et al., 1992), the M2 protein in the in vitro
system is also capable of forming Na+-permeable
channels is not known, since Na + influx was not
monitored. Recent voltage-clamp studies have, however,
indicated that the M2 protein expressed in mouse
3483
erythroleukaemia cells does form proton-selective channels (I. Chizmachov, F. Geraghty, D. Ogden & A. Hay,
unpublished results).
The data presented here are consistent with the ability
of the M2 protein to promote proton translocation in
vivo and with its two proposed roles in virus infection,
acidification of the virion interior during uncoating and
protection of HA against premature acid-induced conformational change during protein transport.
We thank Dr Barry K. Ely and Dr Ian Jones (NERC Institute of
Virology, Oxford, U.K.) for generous help with all aspects of
baculovirus-insect cell expression systems. We are obliged to Setareh
Grambas for support in various experiments, and are grateful to A.
Colin Young for large-scale cultivation of insect cells and to Dr Helmut
Reil~inder (Max-Planck-Institut ffir Biophysik, Frankfurt, Germany)
for the gift of the Sf9 subline RE. Prof. Norbert A. Dencher (HahnMeitner-lnstitut, Berlin, Germany) and Dr Petra A. Burghaus provided
expertise on the membrane reconstitution of channel proteins and the
fluorescent dye assay. Dr Javier Martin-Gonzfiles gave valuable advice
on site-directed mutagenesis. C.S. gratefully acknowledges the support
of a Wellcome Research Fellowship (M/90/2078).
References
BEAVEN,G. H. & HOLIDAY, E. R. (1952). Ultraviolet absorption spectra
of proteins and amino acids. Advances in Protein Chemistry 7,
319-368.
BLACK, R. A., ROTA, P. A., GORODKOVA,N., CRAMER,A., KLENK, H.D. & KENDAL, A.P. (1993). Production of the M2 protein of
influenza A virus in insect cells is enhanced in the presence of
amantadine. Journal of General Virology 74, 1673-1677.
BUKRINSKAYA,A.G., VORKUNOVA,N.K. & PUSHKARSKAYA,N.L.
(1982). Uncoating of a rimantadine-resistant variant of influenze
virus in the presence of rimantadine. Journal of General Virology 60,
61-66.
CARTER, D.B., THOMSEN, D.R., IM, W.B., LENNON, D.J., NGO,
D. M., GALE, W., IM, H. K., SEEBURG,P. H. & SMITH, M. W. (1992).
Functional expression of GABA A chloride channels and benzodiazepine binding sites in baculovirus infected insect cells. BioTechnology 10, 679-681.
CIAMPOR, F., BAYLEY, P.M., NERMUT, M.V., HIRST, E . M . A . ,
SUGRtrE, R. J. & HAY, A. J. (1992a). Evidence that the amantadineinduced, M2-mediated conversion of influenza A virus haemagglutinin to the low pH conformation occurs in an acidic trans Golgi
compartment. Virology 188, 14~24.
CIAMPOR, F., THOMPSON,C. A. & HAY, A. J. (1992b). Regulation ofpH
by the M2 protein of influenza A viruses. Virus Research 22, 247-258.
DENCHER, N.A., BURGHAUS, P.A. & GRZESIEK, S. (1986). Determination of the net proto~hydroxide ion permeability across
vesicular lipid bilayers and membrane proteins by optical probes.
Methods in Enzymology 127, 746-760.
DUFF, K. C. & ASHLEY, R. H. (1992). The transmembrane domain of
influenza A M2 protein forms amantadine-sensitive proton channels
in planar lipid bilayers. Virology 190, 485489.
DuvF, K. C., CUDMORE, A. J. & BRADSHAW,J. P. (1993). The location
of amantadine hydrochloride and free base within phospholipid
multilayers: a neutron and X-ray diffraction study. Biochimica et
Biophysica Acta 1145, 149-156.
FELGNER, P. L., GADEK, T. R., HOLM, M., ROMAN, R., CHAN, H. W.,
WENZ, M., NORTHROP, J.P., RINGOLD, G. M. & DANIELSEN, M.
(1987). Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure. Proceedings of the National Academy of Sciences,
U.S.A. 84, 7413-7417.
GARCIA, M. L., KITADA, M., EISENSTEIN, H. C. & KRULWICH, T. A.
(1984). Voltage-dependent proton fluxes in liposomes. Biochimica et
Biophysica Acta 766, 109-115.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:13:27
3484
C. Schroeder and others
GRAMBAS, S. & HAY, A.J. (1992). Maturation of influenza A virus
haemagglutinin - estimates of the pH encountered during transport
and its regulation by the M2 protein. Virology 190, 11 18.
GRAMBAS, S., BENNET, M.S. & HAY, A.J. (1992). Influence of
amantadine-resistance mutations on the pH regulatory function of
the M2 protein of influenza A viruses. Virology 191, 541-549.
HAY, A, J, (1989). The mechanism of action of amantadine and
rimantadine against influenza viruses. In Concepts in Virus Pathogenesis vol. III, pp. 361-367. Edited by A. L. Notkins & M. B. A.
Oldstone. New York: Springer Verlag.
HAY, A. J., WOLSTENHOLME,A. J., SKEH~,L,J. J. & SMITH, M. H. (1985).
The molecular basis of the specific anti-influenza action of
amantadine. EMBO Journal 4, 3021 3024.
HOLSINGER, L. J. & LAMB, R. A. (1991). Influenza virus M2 integral
membrane protein is a homotetramer stabilized by formation of
disulphide bonds. Virology 183, 32-43.
INDULEN, M.K., POLIS, J.J., KANEL, I.A., RYASANTSEVA,G.M.,
DSEGUDSE, D. R., KALNINA, V. A. & GRAVE, I. J. (1979). Antiviral
activity of a series of rimantadine derivatives. In Antiviral Activity
and Mode ofAction of Various Chemical Compounds (in Russian), pp.
3440. Edited by R. A. Kukaine. Riga: Sinatne.
ITO, T., GORMAN,O. T., KAWAOKA,Y., BEAN,W. J. & WEBSTER, R. G.
(1991). Evolutionary analysis of the influenza A virus M gene with
comparison of the M 1 and M2 proteins. Journal of Virology 65,
5491 5498.
KAMB, A., KORENBROT,J. I. & KITAJEWSKI,J. (1992). Expression of ion
channels in cultured cells using baculovirus. Methods in Enzymology
207, 423-431.
KATO, N. & EGGERS, H. J. (1969). Inhibition of uncoating of fowl
plague virus by 1-adamantanamine hydrochloride. Virology 37,
63~641.
KITTS, P. A., AYRES, M. D. & POSSEE, R. D. (1990). Linearization of
baculovirus DNA enhances the recovery of recombinant virus
expression vectors. Nucleic Acids" Research 18, 5667-5672.
KURODA, K., HAUSER, C., ROTT, R., KLENK, H-D. & DOERELER, W.
(1986). Expression of the influenza virus haemagglutinin in insect
cells by a baculovirus vector. EMBO Journal 5, 1359 1365.
LAMB, R.A. & CHOPPIN, P.W. (1981). Identification of a second
protein (M2) encoded by RNA segment 7 of influenza virus. Virology
112, 729 737.
LUeKOW, V. A. & SUMMERS, M. D. (1989). High level expression of
nonfused foreign genes with Autographa californiea nuclear polyhedrosis virus expression vectors. Virology 170, 31-39.
MARTIN, K. & HELENIUS, A. (1991). Nuclear transport of influenza
virus ribonucleoproteins: the viral matrix protein (MI) promotes
export and inhibits import. Cell 67, 112130.
-OHUCHI, M., CRAMER, A., VEY, M., OHUCHI, R., GARTEN, W. &
KLENK, H.-D. (1994). Rescue of vector-expressed fowl plague virus
hemagglutinin in biologically active form by acidotropic agents and
coexpressed M2 protein. Journal of Virology 68, 920-926.
PINTO, L. H., HOLSINGER, L. J. & LAMB, R. A. (1992). Influenza virus
M2 protein has ion channel activity. Cell 69, 517 528.
PRESSMAN, B. C. (1976). Biological applications of ionophores. Annual
Review of Biochemistry 45, 501-530.
RUIGROK, R . W . H . , WRIGLEY, N.G., CALDER, L.J., CUSACK, S.,
W~tARTON, S, A., B~OWN, E.B. & SKEHEL, J.J. (1986). Electron
microscopy of the low pH structure of influenza virus haemagglutinin. EMBO Journal 5, 41-49.
STEINHAUER,D.A., WHARTON, S.A., SKEHEL,J. J., WILEY,D. C. &
HAY, A. J. (1991). Amantadine selection of a mutant influenza virus
containing an acid-stable hemagg]utinin glycoprotein: evidence for
virus-specific regulation of the pH of glycoprotein transport vesicles.
Proceedings of the National Academy of Sciences, U.S.A. 88,
11525 11529.
SUGRUE, R. J. & HAY, A. J. (1991). Structural characteristics of the M2
protein of influenza A viruses : evidence that it forms a tetrameric
channel. Virology 180, 617-624.
SUGRUE, R.J., BAHADUR, G., ZAMBON, M.C., HALL-SMITH, M.,
DOUGLAS, A. R. & HAY, A. J. (1990a). Specific structural alteration
of the influenza haemagglutinin by amantadine. EMBO Journal 9,
3469-3476.
SUGRUE, R. J., BELSHE, R. B. & HAY, A. J. (1990b). Palmitoylation of
the influenza A virus M2 protein. Virology 179, 51-56.
TAKEUCHI, K. & LAMB, R. A. (1994). Influenza virus M2 protein ion
channel activity stabilizes the native form of fowl plague virus
hemagglutinin during intracellular transport. Journal of Virology 68,
911-919.
VEIT, M., KLENK, H.-D., KENDAL, A. & ROTT, R. (1991). The M2
protein of influenza A virus is acylated. Journal of General Virology
72, 1461 1465.
WANG, C. K., TAKEUCHI, K., PINTO, L. H. & LAMB, R. A. (I993). The
ion channel activity of influenza A virus M2 protein : characterization
of the amantadine block. Journal of Virology 67, 5585 5594.
WrYEr,, U. & POSSEE, R.D. (1991). A baculovirus dual expression
vector derived from the Autographa californica nuclear polyhedrosis
virus polyhedrin and pl0 promoters: co-expression of two influenza
virus genes in insect cells. Journal of General Virology 72, 29622974,
WHARTON, S. A., HAY, A. J., SUGRUE, R. J., SKEHEL, J. J., WEIS, W. I.
& WILEY, D. C. (1990). Membrane fusion by influenza viruses and
the mechanism of action of amantadine. In Use of X-ray Crystallography in the Design of Antiviral Agents, pp. 1 12. Edited by
W. G. Laver & G. M. Air. New York: Academic Press.
(Received 9 May 1994; Accepted 29 July 1994)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:13:27