Download Influence of residue 44 on the activity of the M2 proton channel of

Journal of General Virology (2005), 86, 181–184
Short
Communication
DOI 10.1099/vir.0.80358-0
Influence of residue 44 on the activity of the M2
proton channel of influenza A virus
Tatiana Betakova,1 Fedor Ciampor1 and Alan J. Hay2
1
Institute of Virology, Slovak Academy of Sciences, Dubravska cesta 9, 845 05 Bratislava,
Slovakia
Correspondence
Tatiana Betakova
2
[email protected]
National Institute for Medical Research, The Ridgeway, London NW7 1AA, UK
Received 11 June 2004
Accepted 22 September 2004
The influenza A virus M2 proton channel plays a role in two stages of virus replication. The
proteins of two closely related strains of the avian H7 subtype of influenza A virus, Rostock and
Weybridge, were found to differ in their pH-modulating activities and activation characteristics. Of
three amino acid differences at residues 27, 38 and 44 within the membrane-spanning domain,
substitution at residue 44 was necessary and sufficient to account for differences in trans-Golgi
pH-modulating activity, whereas changes in all three were required to switch the activation
characteristics of the Weybridge M2 to those of the Rostock M2. These results not only separate
the two phenomena genetically, but also indicate that the ‘unique’ activation characteristics of the
Rostock M2 channel were selected specifically. In addition, they point to the importance of
functional complementarity between the activation characteristics of the M2 channel and the pH of
membrane fusion by haemagglutinin during virus entry.
The influenza A virus M2 protein forms a proton-selective
transmembrane ion channel, which is activated at acidic pH
(Pinto et al., 1992; Chizhmakov et al., 1996; Mould et al.,
2000) and is the specific target of the anti-influenza drugs
amantadine and rimantadine. The M2 channel plays a role in
the uncoating of influenza virions in endosomes (Martin &
Helenius, 1991; Wharton et al., 1994). In addition, during
infection by the highly pathogenic influenza A virus subtypes
H7 and H5, M2 is required to act as a proton-leak channel to
elevate pH within the trans-Golgi network; this is important
to prevent premature acid activation of newly synthesized
haemagglutinin (HA), which is cleaved intracellularly, and
consequent inactivation of progeny virus (Sugrue et al., 1990;
Steinhauer et al., 1991; Ciampor et al., 1992; Grambas & Hay,
1992). The M2 proteins of different viruses vary in their
ability to alter trans-Golgi pH. In particular, differences in
the activities of two closely related strains of avian H7 viruses,
Rostock and Weybridge, correlate with differences in the
fusion pH of their HAs (Grambas et al., 1992). The greater
activity of the Rostock M2 in reducing acidity in the transGolgi complements the lower acid stability of Rostock HA,
with a fusion pH of 5?9 compared with 5?3 for Weybridge
HA (Grambas & Hay, 1992).
Furthermore, electrophysiological studies have shown that
the M2 channels of the two viruses differ in two important
respects, representing mechanistic changes in the channel
(Chizhmakov et al., 2003): (i) the Rostock M2 possesses a
sevenfold greater proton conductance, corresponding to its
greater pH-modulating activity; and (ii) the two channels
0008-0358 G 2005 SGM
differ in activation characteristics. More specifically, they
differ in the direction of rectification induced by high pH
(>7). Whereas the Weybridge M2, like that of human
viruses, deactivates in response to external (but not internal)
high pH, the Rostock M2 deactivates in response to internal
(but not external) high pH. The latter difference in
activation characteristics was shown to be determined by
three amino acid differences, V27I, F38L and D44N, within
the transmembrane domain, which distinguish the
Weybridge and Rostock proteins, respectively. Substitution of all three residues was required to transform the
Weybridge phenotype into that of the Rostock M2 and,
conversely, single substitutions in Rostock M2 were
sufficient to effect the opposite phenotypic change.
However, these mutagenesis experiments did not resolve
which of the amino acid differences affected the ion flux
through the channel. To answer that question, we have used
a semi-quantitative HA–M2 co-expression assay to assess
the effects of single and double mutations at these three
positions, aa 27, 38 and 44, on the pH-modulating activity of
M2 and to determine whether the differences in ion flux and
activation are genetically linked.
DNA copies of coding sequences for the HA of influenza
virus strain A/chicken/Germany/34 (H7N1, Rostock strain)
and the M2 proteins of the Rostock strain, A/chicken/
Germany/27 (H7N7, Weybridge strain) and A/PR/8/34
(H1N1) were inserted into plasmid pVOTE.1 (kindly
provided by B. Moss, National Institutes of Health,
Bethesda, MD, USA) to generate pVOTE.1-HA and
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:05:43
Printed in Great Britain
181
T. Betakova, F. Ciampor and A. J. Hay
Fig. 1. Amino acid sequences of the M2 proteins of Rostock strain A/chicken/Germany/34 (R-M2), Weybridge strain
A/chicken/Germany/27 (W-M2), A/PR/8/34 and transmembrane mutants of R-M2 and PR8-M2. The transmembrane domain
is shown in bold.
pVOTE.1-M2, respectively. Plasmids encoding the mutant
M2 proteins I27V, L38F, 127V+L38F, N44D and D44N
(Fig. 1) were prepared by using four-primer PCR.
Sequences of oligonucleotide primers and details of cloning
are available upon request. CV-1 cells were infected for 1 h
with recombinant vaccinia virus vTF7.3 (10 p.f.u. per cell)
(Fuerst et al., 1986), transfected for 4 h with plasmids mixed
with Lipofectin (Life Technologies) and incubated for a
further 16 h in minimal essential medium containing 20 %
fetal calf serum and 40 mg arabinose C ml21, with or without
5 mM amantadine.
Proteins in 24-well plates were analysed by 12?5 % SDSPAGE and Western blot analysis was done as described by
Grambas et al. (1992), using anti-M2 serum, protein A–
horseradish peroxidase (HRP) conjugate (Bio-Rad) and
labelling with TMB stabilized substrate for HRP (Promega).
The amount of protein expressed depended on the amount
of pVOTE.1-X plasmid that was used for transfection (data
not shown) and was roughly equivalent for similar amounts
(0?5 mg) of plasmid DNA, as shown in Fig. 2. The variation
observed, less than twofold, was within the variation seen
between different experiments.
ELISA to detect HA co-expressed with M2 was performed
on CV-1 cells in 96-well plates (duplicate wells) following
transfection with 0?25 mg pVOTE.1-HA together with
increasing amounts of plasmid DNA encoding wt or
1
2
3
4
5
6
7
8
Fig. 2. Expression of wt and mutant M2 proteins. CV-1 cells in
24-well plates, transfected with 0?5 mg pVOTE.1-X, were lysed
and analysed by 12?5 % SDS-PAGE and Western blotting with
anti-M2 antibody. Lane 1, wt Rostock M2 (R-M2); lane 2, RM2 I27V; lane 3, R-M2 L38F; lane 4, R-M2 N44D; lane 5, wt
Weybridge M2; lane 6, A/PR/8/34 (PR8-M2); lane 7, PR8-M2
D44N; lane 8, non-transfected cells.
182
mutant M2 proteins, essentially as described by Grambas
& Hay (1992). Three mAbs were used: HC58, specific for the
native form of HA; H9, specific for the low-pH form; and
HC2, which recognizes both forms of HA. The effectiveness
of M2 in elevating trans-Golgi pH and protecting HA against
low pH-induced changes depended on the ratio of HA and
M2 proteins expressed and was indicated by corresponding
increases in the proportion of native HA and decreases in the
proportion of low-pH HA (shown for wt Rostock M2 in
Fig. 3a). The maximum percentage change in A450 obtained
with HC58 and H9 antibodies relative to HC2 was recorded
at the optimum HA/M2 ratio; further increases in expression
of M2 reduced expression of total HA. The optimum HA/M2
ratio was similar for all M2 proteins. Mean values from six
experiments were used to compare the pH-modulating
activities of the different M2 proteins (Fig. 3b).
Co-expression of HA with the wt Rostock M2 (R-M2)
resulted in an increase of approximately 24 % in native HA
and a corresponding decrease of approximately 23 % in the
low-pH form of HA, compared with HA expressed in the
absence of M2 protein (Fig. 3b). The changes conferred by
the wt Weybridge M2 (W-M2) were approximately half
those found with R-M2 (approx. ±13 %) and were
comparable to those obtained with M2 of the human
influenza virus A/PR/8/34 (PR8-M2; approx. ±16 %)
(Fig. 3b). The activities of R-M2 and W-M2 were inhibited
specifically by amantadine (5 mM), whereas PR8-M2 was
resistant, consistent with previously published data.
The mutant Rostock proteins containing single amino acid
substitutions of I27V or L38F possessed a pH-modulating
activity that was indistinguishable from that of the wt
protein (Fig. 3b), as did the double mutant I27V+L38F
(data not shown). Substitution of aspartic acid for arginine
44 (N44D), however, caused a significant decrease in activity
(approximately to the levels found for W-M2), indicating
that the change in this residue alone could account for the
difference in the pH-modulating activities of the two wt
channel proteins. Furthermore, substitution of asparagine
for aspartic acid 44 in the M2 PR8-M2 (D44N) caused an
increase in activity (resistant to amantadine) to a level
comparable to that of the wt Rostock protein, confirming
the more general influence of this particular substitution at
residue 44 on M2 activity.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:05:43
Journal of General Virology 86
Influenza A virus M2 proton channel activity
Fig. 3. pH-modulating activity of wt and mutant M2 proteins. Change in HA (%) recognized by HC58 (native form of HA) or
H9 (low-pH form of HA), respectively, was estimated from the ratios of A450 as follows: HC58 (%)=[HC58/HC2 (plus
M2)6100]”[HC58/HC2 (no M2)6100], or H9 (%)=[H9/HC2 (plus M2)6100]”[H9/HC2 (no M2)6100], respectively.
(a) Dependence of changes in HA on amount of wt R-M2 (ng). %, HC58; &, H9; m, HC58+amantadine;
$, H9+amantadine. (b) Maximum changes (%) obtained with HC58 and H9 antibodies relative to HC2. Results are
shown for wt Rostock M2 (R), mutant Rostock proteins (I27V, L38F, N44D), wt Weybridge M2 (W), A/PR/8/34 (PR8) and
mutant A/PR/8/34 (PR8-D44N). Filled bars, HC58; open bars, H9; dark hatched bars, HC58+amantadine; pale hatched
bars, H9+amantadine. Results are shown as means±SD of six experiments.
Thus, whereas any one of the single amino acid substitutions
– I27V, L38F or N44D – caused a switch in the activation
characteristics of Rostock M2 to those of the Weybridge
channel (Chizhmakov et al., 2003), only the change in
residue 44 between asparagine and aspartic acid was
necessary and was sufficient to account for the difference
between the pH-modulating activities (proton flux) of the
two channel proteins. It was apparent, therefore, that there
was no strict genetic correlation between the two phenotypic
properties and that they represented separable functional
characteristics.
Previous studies of mutant viruses have shown that a
number of single amino acid substitutions, at residues 26,
30, 31 and 34, as well as 27, can cause increases or decreases
in pH-modulating activity, as assayed by HA–M2 coexpression, e.g. I27T and I27S substitutions caused increases
in activity of Rostock M2, whereas I27N, like I27V in the
present study, caused no detectable difference (Grambas
et al., 1992). Furthermore, it was also observed, by passage in
cell culture, that mutations that altered the fusion pH of the
HA or the pH-modulating activity of M2 could influence
selection of changes in the corresponding properties of M2
and HA, respectively. Thus, for example, a Weybridge
mutant with an increased HA fusion pH selected a mutation
in M2, G34E, that caused a compensatory increase in pHmodulating activity.
As for the significance of the differences in phenotype, there
was a clear correlation between the greater pH-modulating
activity and proton flux of the M2 channel and the higher
HA fusion pH of the Rostock virus. However, there was no
clear indication as to the advantage conferred by the
http://vir.sgmjournals.org
‘unusual’ activation characteristics of its M2 channel
protein, especially with respect to modulation of transGolgi pH. As acquisition of the latter property by the
Rostock M2 required two unusual amino acid substitutions
in residues 27 and 44 (among known M2 sequences),
whereas the former alteration was effected by the change in
residue 44 (N44D) alone, it appeared likely that the change
in activation characteristics was selected for in addition to,
and not coincidentally with, the increase in M2 channel
activity. We do not know, however, how the Rostock virus
with these peculiar properties (in HA and M2) emerged –
whether in vivo or as a result of extensive passage in vitro –
or how the sequence in which the (complementary)
characteristics of HA and M2 were acquired. That the
change in activation characteristics is apparently not
associated with the change in modulation of trans-Golgi
pH points to its greater significance for the activity of M2 in
virus entry. It may be that removal of the ‘strict’ regulation
by pH outside the virion (in the case of Weybridge M2) is
necessary to facilitate uncoating of the Rostock virion at a
higher endosomal pH, consistent with the higher pH at
which fusion between the virus and endosomal membranes
is promoted by the Rostock HA. As the change in activation
affects the direction of the pH-induced effect (from outside,
the N terminus of M2, to inside, the C terminus) and not the
intrinsic nature of the pH-induced change in the voltage
dependence of channel conductance, this appears to provide a mechanism for ‘switching off’ the normal activation
property. Furthermore, maintenance of the intrinsic conductance characteristics of the channel emphasizes their
importance. However, the mechanistic significance of reduced
outward H+ flux at high external pH in, for example,
promoting H+ transfer into the virion or preserving the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:05:43
183
T. Betakova, F. Ciampor and A. J. Hay
integrity of the RNA genome of virus exposed to an alkaline
environment, has yet to be elucidated.
virus that synthesizes bacteriophage T7 RNA polymerase. Proc Natl
Acad Sci U S A 83, 8122–8126.
Grambas, S. & Hay, A. J. (1992). Maturation of influenza A virus
hemagglutinin – estimates of the pH encountered during transport
and its regulation by the M2 protein. Virology 190, 11–18.
Acknowledgements
We thank Igor Chizhmakov for many interesting and helpful
discussions and Seti Grambas for excellent assistance. This research
was supported by grants no. 2/3003/23 and 1/0640/03 of the VEGA –
Grant Agency for Science.
Grambas, S., Bennett, 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.
Martin, K. & Helenius, A. (1991). Nuclear transport of influenza virus
ribonucleoproteins: the viral matrix protein (M1) promotes export
and inhibits import. Cell 67, 117–130.
References
Mould, J. A., Drury, J. E., Frings, S. M., Kaupp, U. B., Pekosz, A.,
Lamb, R. A. & Pinto, L. H. (2000). Permeation and activation of the
Chizhmakov, I. V., Geraghty, F. M., Ogden, D. C., Hayhurst, A.,
Antoniou, M. & Hay, A. J. (1996). Selective proton permeability and
Pinto, L. H., Holsinger, L. J. & Lamb, R. A. (1992). Influenza virus M2
pH regulation of the influenza virus M2 channel expressed in mouse
erythroleukaemia cells. J Physiol 494, 329–336.
Chizhmakov, I. V., Ogden, D. C., Geraghty, F. M., Hayhurst, A.,
Skinner, A., Betakova, T. & Hay, A. J. (2003). Differences in
M2 ion channel of influenza A virus. J Biol Chem 275, 31038–31050.
protein has ion channel activity. Cell 69, 517–528.
Steinhauer, D. A., Wharton, S. A., Skehel, J. J., Wiley, D. C. & Hay,
A. J. (1991). Amantadine selection of a mutant influenza virus
conductance of M2 proton channels of two influenza viruses at low
and high pH. J Physiol 546, 427–438.
containing an acid-stable hemagglutinin glycoprotein: evidence for
virus-specific regulation of the pH of glycoprotein transport vesicles.
Proc Natl Acad Sci U S A 88, 11525–11529.
Ciampor, F., Bayley, P. M., Nermut, M. V., Hirst, E. M. A., Sugrue,
R. J. & Hay, A. J. (1992). Evidence that the amantadine-induced, M2-
Sugrue, R. J., Bahadur, G., Zambon, M. C., Hall-Smith, M., Douglas,
A. R. & Hay, A. J. (1990). Specific structural alteration of the
mediated conversion of influenza A virus hemagglutinin to the low
pH conformation occurs in an acidic trans Golgi compartment.
Virology 188, 14–24.
Wharton, S. A., Belshe, R. B., Skehel, J. J. & Hay, A. J. (1994). Role
Fuerst, T. R., Niles, E. G., Studier, F. W. & Moss, B. (1986).
Eukaryotic transient-expression system based on recombinant vaccinia
184
influenza haemagglutinin by amantadine. EMBO J 9, 3469–3476.
of virion M2 protein in influenza virus uncoating: specific reduction
in the rate of membrane fusion between virus and liposomes by
amantadine. J Gen Virol 75, 945–948.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:05:43
Journal of General Virology 86