J. gen. Virol. (1986), 67, 1135-1147. Printedin Great Britain
1135
Key words: AIMV/synthetic peptides/RNA replication
Kinetics of Accumulation of the Three Non-structural Proteins of Alfalfa
Mosaic Virus in Tobacco Plants
By A N N E B E R N A , 1. J E A N - P A U L B R I A N D , 2
C H R I S T I A N E S T U S S I - G A R A U D 1 AND
THI~RI~SE GODEFROY-COLBURN
1
Laboratoires 1de Virologie et 2d'Immunochimie,
Institut de Biologie Molbculaire et Cellulaire du Centre National de la Recherche Scientifique,
15 rue Descartes, F-67084 Strasbourg-Cedex, France
(Accepted 21 February 1986)
SUMMARY
Antisera to synthetic peptides corresponding to the C-termini of two non-structural
proteins (NSP) of alfalfa mosaic virus (A1MV), P1 (126K) and P3 (32K), were prepared
and characterized. These antisera, together with one which had previously been made
against the C-terminus of P2 (90K), enabled us to detect the three NSPs in a crude
membrane fraction from A1MV-infected tobacco leaves. The accumulation of these
proteins at 25 °C and at 10 °C was followed as a function of time after inoculation, and
their amounts were compared with viral replicase activity. All three proteins
accumulated at the beginning of the infection period and then disappeared, P3 more
rapidly than the other two. There was a good correlation between replicase activity and
the amounts of P1 and P2, but not the amount of P3, These results are consistent with
the notion that P 1 and P2 are part of the replication complex. Although much less coat
protein was made in inoculated leaves at 10 °C than at 25 °C, the maximum amounts of
the three NSPs and the maximum replicase activity were at least as high at 10 °C as at
25 °C. Thus, 10 °C is not a restrictive temperature for the assembly of a functional
replication complex.
INTRODUCTION
The genome of alfalfa mosaic virus (A1MV) encodes four proteins, only one of which, t h e
coat protein (mol. wt. 24500, 24K) is produced in large quantity in infected plants. The other
three (P1,126K; P2, 90K; P3, 32K) are non-structural. Although they are readily obtained by in
vitro translation of the genomic RNAs, RNA 1, RNA 2 and R N A 3 (van Tol & van VlotenDoting, 1979, and references therein; Godefroy-Colburn et al., 1985), they do not accumulate to
any great extent in vivo.
The role of the non-structural proteins (NSPs) is still open to speculation. P1 and P2 are
almost certainly involved in replication because, in protoplasts, inoculation with a mixture of
RNA 1, RNA 2 and coat protein is necessary and sufficient to obtain replicative forms (RF)
(Nassuth et al., 1981 ; Nassuth & Bol, t983). The study of mutations in RNA 1 and R N A 2 also
suggests that P1 and P2 function in close interaction with each other (Sarachu et al., 1983), and it
is thought that both are part of the replicase. However, RNA 3 is required for the switch from
symmetric plus-strand and minus-strand synthesis (production of RF) to asymmetric plus-strand
synthesis (production of viral RNA) to occur (Nassuth & Bol, 1983). This function may therefore
involve P3 and/or the coat protein, both of which are encoded by R N A 3. The detection of P2
and P3 (Berna et al., 1984, 1985) in a membrane fraction of tobacco leaves, which is thought to
contain the viral replicase (Weening & Bol, 1975; Le Roy et al., 1977), is consistent with this
view. In addition, there are strong similarities between three regions of P2 and homologous
regions of the polymerases of poliovirus and cowpea mosaic virus (Kamer & Argos, 1984). It is
thus tempting to speculate that P2 is the equivalent of the picornavirus or comovirus replicase.
0000-7020
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
1136
A. BERNA AND OTHERS
The direct functional study of the NSPs of A1MV has not been possible because sensitive and
reliable detection methods have not been available. Proteins with the electrophoretic mobilities
of the full-length A1MV translation products were detected in pulse-labelled protoplasts when
host protein synthesis was inhibited by prior u.v. irradiation (Samac et al., 1983) and a protein
with the electrophoretic mobility of P3 was detected in infected tobacco plants by pulse-labelling
in the presence of actinomycin D and chloramphenicol (Joshi et al., 1984). However, positive
identification of P2 and P3 in infected plants was possible only with specific antisera (Berna et
al., 1984, 1985), although the methods did not permit a quantitative study to be undertaken.
In this paper, we report the immunodetection of P1, P2 and P3 using a much improved
method, the kinetics of accumulation of the three NSPs of AIMV in tobacco plants, and a
comparison of the amount of each of these proteins in a crude membrane fraction with the viral
replicase activity of the same fraction throughout the infection period.
Some early steps of tobacco mosaic virus (TMV) replication take place at low temperature
although no virus is synthesized (Dawson & Schlegel, 1976) and preliminary observations
(Berna et al., 1984) suggested that the same was true of A1MV. Therefore, we compared the
accumulation of NSPs and replicase activity at 25 °C, at which the infection is systemic and
yields large amounts of virus, and at 10 °C, at which little virus is made.
METHODS
Virus propagation and purification. The Strasbourg strain (S-strain) of A1MV was propagated in Nicotiana
tabacum (cv. Xanthi nc) according to Pinck & Hirth (1972). Six-week-old plants were inoculated and kept at 25 °C
or 10 °C for the indicated times under 16 h illumination.
For preparation of the coat protein, the virus was purified according to Franck & Hirth (1976).
Preparation ofsubcellular Jmctions. The 1000 g pellet (Pe-l), the 30000 g pellet (Pe-30) and the supernatant
fraction (S-30) were prepared from inoculated leaves and from similar leaves of healthy plants as described by
Berna et al. (1985). Pe-1 and Pe-30 were resuspended in extraction buffer (100 mi-Tris-HC1 pH 8.1 at 4 °C, 10 mMKCI, 5 mM-MgC12, 0.4 M-sucrose, 10~ glycerol, 10 mM-2-mercaptoethanol) and diluted with an equal volume of
glycerol. The final samples were concentrated 20-fold with respect to the original extract. In order to keep the
amount of supernatant protein to be tested within manageable limits, the S-30 was fractionated by ammonium
sulphate precipitation from 0 to 35~, from 35 to 45~, from 45 to 55% and from 55 to 7 5 ~ saturation. The
fractions were redissolved in extraction buffer (1/30 the original volume of S-30) and dialysed against the same
buffer.
Preparation of the wheat germ translation products of A l M V RNA. The [35S]methionine-labelled translation
products of AIMV RNA were synthesized in wheat germ extracts and partially purified by centrifugation through
a sucrose gradient (Berna et aL, 1985). Three preparations were made, with 60 gtg/ml, 120 gg/ml and 200 ttg/ml of
A1MV RNA, concentrations found to be optimum for the synthesis of P3, P1 and P2 respectively.
Peptide synthesis. Two peptides were synthesized by the solid phase method (Barany & Merrifield, 1980) and had
the sequences of the C-terminal 21 amino acids of PI (pepl, Fig. 1 a) and the C-terminal 26 amino acids of P3,
minus the terminal histidine (pep3, Fig. 1b). The peptides were purified by filtration through a Sephadex G-25
column (1.20 m × 1-4 cm for 150 mg of peptide) in 8% acetic acid. Their amino acid composition was verified
using a 'Durrum D500' analyser.
Preparation ofantisera to the symheticpeptides. Peptides were injected subcutaneously into rabbits at 3- to 4-week
intervals. Each injection contained 150 gtg of peptide in 0.5 ml phosphate-buffered saline (0-15 M-NaC1 in 10 mMphosphate pH 7.2; PBS) emulsified with 0.5 ml Freund's adjuvant, either complete (first injection) or incomplete
(boosters). Sera were prepared from blood collected l0 days after each booster and were kept at - 2 0 °C in 5 0 ~
glycerol.
Preparation of a n antiserum to the coat protein of AlMV. The coat protein was prepared from the purified virus by
SDS PAGE in 13~ gels. The band containing the coat protein was excised and injected into a rabbit as described
for P3 by Berna et al. (1985). The rabbit received three injections, each containing 200 gg of coat protein. The
antiserum was collected 10 days after the last injection.
Electrophoretie analysis of proteins. SDS-PAGE was in the buffer system of Laemmli (1970). The gels (1.5 mm
thick) were made with a ratio of bisacrylamide to acrylamide of 1 : 36 (w/w). They contained 6 ~ acrylamide for the
detection of PI and P2, 9 ~ for P3 and 12~ for the coat protein. The samples were prepared as in Berna et al.
(1985).
Immunoblotting procedure. After SDS-PAGE, the proteins were immediately blotted onto nitrocellulose (NC)
sheets essentially as described by Towbin et al. (1979), at a constant current of 2 mA/cm 2. The coat protein was
electroblotted for 45 rain. The transfer of the NSPs was monitored by the recovery of 35S-labelled translation
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
AIMV
e~
9
1137
n o n - s t r u c t u r a l p r o t e i n s in p l a n t s
(a)
Amino acid residue
1.5 1026
1076
(b)
1126
1.5
1.0
1,0
0.5
o.5]
200
Amino acid residue
250
l
3oo
0
-0.5
-0.5
-1.0
-1.0
- " J p e p
-1.5
1~ L ' -
OH ~ s P a C L ~ V PK ~FT Y ESF, K D EILP D
NH 2
A C R D A G N T D D S I L A R S Y N
HN
F
Fig. 1. Hydrophilicity of the C-terminal regions of (a) PI and (b) P3, and sequences of the synthetic
peptides. The sequences of P 1 and P3 were derived from those of RNA 1 (strain 425, Cornelissen et al.,
1983a) and RNA 3 (strain S, Ravelonandro et al., 1984) and the hydrophilicity diagrams of these
proteins were established according to Hopp & Woods (1981) with a scan size of four amino acids. For
each protein, P 1 or P3, the part of the diagram corresponding to 100 C-terminal residues is shown with
the sequence of the peptide which was synthesized (pepl or pep3). The C-terminal amino acid of P3
(histidine) was omitted from pep3.
products on NC sheets as compared to the original gel. Transfer was significantly improved by placing a 6 ~
polyacrylamide gel containing 0.1 ~ SDS and Tris-glycine buffer pH 8.3 on the cathode side of the gel containing
the samples in order to increase the SDS concentration locally, thereby favouring the migration of proteins without
hindering their adsorption on NC. Thus, the efficiency of transfer increased from 50~o of P1 (normal procedure,
transfer for 2.5 h) to 8 5 ~ (modified procedure, transfer for 1.5 h), from 6 0 ~ of P2 to 90~, and from 75 ~ of P3 to
95~0. To prevent partial elution of the proteins from the blots during the subsequent procedure, the blots were
heated for 30 to 40 min at 80 °C.
The immunoreactions were carried out on the NC sheets using goat anti-rabbit IgG conjugated with peroxidase
as second antibody, as previously described (procedure 2 of Berna et al., 1985) except that the sensitivity of the
method was increased by making solution B 5 ~ instead of 13~ in dehydrated skimmed milk.
Quantification ofimmunoreactions on the NC sheets. The bands of interest were scanned by reflection at 460 nm
with a Shimadzu TLC scanner (model CS-960). Generally, we used a 2 mm-wide spot moving perpendicularly to
the direction of migration in order to integrate the edges of the bands. However, when the viral band was too close
to a cellular contaminant (as sometimes happened with P3) the blot was scanned in the direction of migration.
Quantification oJ radioactivity in polyacrylamide gels and on NC sheets. The radioactive bands were excised,
dissolved in NCS (Amersham) and counted in NA cocktail (Beckman) as described by Godefroy-Colburn et al.
(1985).
R N A replicaseassay. The RNA replicase activity of the Pe-30 was measured at 30 °C without added template, as
described by Le Roy et at. (1977), using [3H]UTP in the presence of the other three ribonucleoside triphosphates.
The specific radioactivity of [3H]UTP was 400 c.p.m./pmol when measured on DEAE paper (Whatman DE52)
in NA cocktail. Under these conditions, the counting efficiency of the polymeric product was eightfold that of the
monomeric precursor (data not shown), and this was taken into account in calculating the molar amount of UMP
incorporated. The Pe-30 from healthy leaves had little RNA polymerase activity in this assay (i.e. less than 20 ~ of
the background) and this was subtracted from the results obtained with the A1MV-infected samples. With the
latter samples a large part of the label was incorporated into double-stranded RNA which had the same mobility as
the RFs of RNAs 1, 2 and 3 (Le Roy et al., 1977; Clerx & Bol, 1978; and our unpublished results).
RESULTS
Characterization of antisera to the C-terminal peptides of the AIMV non-structural proteins
The preparation and characterization of an antiserum against a synthetic peptide
corresponding to the C-terminus of P2 (pep2) have been described (Berna et aL, 1984). The new
antisera, made against the C-terminal peptides of P1 and P3, were similarly tested against the
wheat germ translation products of AIMV RNA.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
1138
A. B E R N A A N D O T H E R S
(a)
3
4
5
6
7
8
Pl
P2
(b)
2
3
4
5
6
7
9
10
11
N;N
P1
a?N,;, N,J N
P2
:N
N gN:~
:2N aAN ~:N
N,; Na~ N N
~NN
~NN~
II
P'3
P3
P4
Fig. 2. Characterization of the antisera to pep| and pep3 with A1MV translation products. Partially
purified preparations of the 35S-labelled AIMV translation products (or equivalent blank preparations
made without mRNA) were immunoblotted with the antisera to be tested and with a non-immune
serum. Lane 1 is an autoradiogram and lanes 2 to 11 are immunoblots. (a) Characterization of the antipepl serum. Products of translation of 120 tag/ml AIMV RNA, containing about 10 ng P1 (lanes 1 to 3
and 5 to 8), or an equivalent amount of the blank preparation (lane 4), were separated in a 6% gel and
immunoblotted with a non-immune serum (lane 2) or with the anti-pepl serum (lanes 3 to 8), both
diluted 5000-fold. In lanes 6 to 8, the antiserum (4 gl in 50% glycerol) had been preincubated with pepl
for 30 rain at room temperature in 200 gl PBS, then diluted to 10 ml for incubation with the blot. The
final concentrations of pepl were 0.01 gg/ml (lane 6), 0.1 gg/ml (lane 7) or 2-0 gg/ml (lane 8).
(b) Characterization of the anti-pep3 serum and comparison with the anti-P3 serum. Products of
translation of 60 gg/ml A1MV RNA, containing about 11 ng P3 (lanes 1 to 3 and 6 to 11), or equivalent
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
A I M V non-structural proteins in plants
1139
Of four rabbits injected with pep1, only two made antibodies which reacted with
electroblotted P1. Fig. 2(a) presents the reaction of the best antiserum (obtained after four
injections) with a partially purified preparation of the A1MV translation products. The
immunoblot (lane 3) showed one band which corresponded to radioactive P 1 (lane 1) and did not
appear if the translation products were reacted with a non-immune serum (lane 2). The antipepl serum did not react with a blank translation medium (lane 4). The specificity of the
reaction observed in lane 3 was further demonstrated by preabsorbing the anti-pep 1 serum with
increasing amounts of the peptide: the" reaction with P1 was progressively eliminated and
disappeared completely in the presence of 2 ~tg/ml pepl (lanes 5 to 8). The anti-pepl serum also
detected a fainter band (marked with a triangle in lane 3) which was competed out by the
peptide (lanes 5 to 8) and which was absent from the immunoblot of the blank translation
medium (lane 4). The corresponding protein therefore seemed to be related to P1. Since it
migrated with an apparent mol. wt. of 200K, it could have been a minor aggregated form of this
protein. An overexposed autoradiogram (not presented) showed radioactive material in this
region of the gel.
Pep3 was injected into one rabbit and sera obtained throughout the immunization period were
assayed on electroblots of the partially purified translation products. The best antiserum was
obtained after five injections. It reacted strongly with P3 and with P'3 (the minor wheat germ
translation product of RNA 3; Godefroy-Colburn et al., 1985 ; Berna et al., 1985), as shown by
comparison with the autoradiogram (Fig. 2b, lanes 1 and 3). No reaction was observed with a
non-immune serum (lane 2). The coloured bands corresponding to P3 and P'3 were also absent
when the anti-pep3 serum was assayed against a blank translation medium (lane 4), and they
were eliminated by preabsorption of the antiserum with the peptide, further proving their
specificity (lanes 7 to 11). Their elimination was complete at 2 ~tg/ml pep3. A faint band, marked
with a triangle in lane 3, was visible on all the immunoblots which had reacted with the antipep3 serum. However, this reaction was probably not due to the anti-pep3 antibodies since it
took place with the blank sample as well as with the translation medium, and, furthermore, was
not competed out by the peptide.
The anti-pep3 serum was compared with the anti-P3 serum which had previously been
prepared against the gel-purified wheat germ translation product of RNA 3 (Berna et al., 1985).
Although the anti-P3 serum specifically reacted with P3 and P'3 after preabsorption with the
wheat germ extract (Fig. 2b, lanes 5 and 6), it was fivefold less efficient than the anti-pep3 serum
when used at the same dilution. An additional, poorly reproducible reaction of the preabsorbed
anti-P3 serum (diamond in lane 5) may correspond to a gel artefact (Tasheva & Dessev, 1983).
Quantification of the immunoblots
As a test of the proportionality between the intensity of staining of immunoblots (measured by
reflection densitometry) and the amount of viral protein present, 35S-labelled translation
products were assayed by scintillation counting and immunoblotting. Fig. 3 shows that the
intensity of colour obtained after 2 h of peroxidase reaction was nearly proportional to the
amount of viral protein, especially when the integrated absorbance was less than 1.5 arbitrary
units. In our assay conditions, this absorbance corresponded to 0.2 pmol of P1, or 0.18 pmol of
P2, or 0.12 pmol of P3. The anti-pep3 serum detected P3 and P'3 with the same efficiency, which
suggests that the epitopes of pep3 which were accessible in blotted P3 were equally accessible in
P'3. A similar observation was made with the anti-P3 serum (Berna et al., 1985).
amounts of the blank preparation (lanes 4 and 5) were separated in 9% gels and immunoblottedwith a
non-immuneserum (lane 2), the anti-pep3 serum (lanes 3, 4, 7 to 11)or the anti-P3 serum, preincubated
with the wheat germ extract as in Berna et al. (1985)(lanes 5 and 6). Sera were diluted 1: 10000 (lanes 2
to 6) or 1: 5000(lanes7 to 11). In lanes 8 to 11, the anti-pep3serum had been preincubatedwith pep3 like
the samples in (a), lanes 6 to 8. The finalconcentrationsof peptide were 0.001 ~tg/ml(lane 8), 0-01/.tg/ml
(lane 9), 0.1 ~tg/ml (lane 10) or 2.0 ~tg/ml (lane 11).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
1140
A. B E R N A A N D O T H E R S
I
I
I
I
I
I
i
'r3
2
/
J
t
~..z~
L
I
0.5
0.2
0.3
0.4
Protein loaded (pmol/slot)
Fig. 3. Quantification of NSPs by immunoblotting. Partially purified 3sS-labelled translation products
were separated in a 9 ~ gel (P3) or in a 6 ~ gel (P 1 and P2), and immunoblotted with the anti-pep3 serum
or a mixture of anti-pepl and anti-pep2 respectively. Each serum was diluted 5000-fold. The peroxidase
reaction was for 2 h. The absorbance at 460 nm of the viral protein bands was measured by reflection
and the amount of protein was calculated from the amount of radioactivity in the band. O, P 1; A, P2;
II, P3; fl, P'3.
0.1
Time course of accumulation of the viral proteins at 25 °C
A batch of plants was inoculated with A1MV and kept at 25 °C. Symptoms characteristic of
systemic infection a p p e a r e d on the inoculated leaves after 2 days, and on the upper leaves after 3
days. Inoculated leaves from four to six plants were harvested at different times after inoculation
and extracted immediately, to obtain the S-30 and the Pc-30 fractions. S-30 contains most of the
coat protein (mainly in virus particles) which was assayed by immunoblotting with anti-coatprotein serum. As shown in Fig. 4 (a), coat protein synthesis proceeded at m a x i m u m rate after a
lag period of 20 to 30 h. After 9 days its concentration in the S-30 stabilized at 370 ~tg/ml
(measured on immunoblots, not shown, by comparison with purified virus). This was equivalent
to 740 ~tg of coat protein per g of leaves.
The Pc-30, which had previously been shown to contain P2 and P3, was immunoblotted with
the anti-pep1 serum (Fig. 4b). In infected extracts (lanes 1 to 8), the antiserum detected a
polypeptide which was absent from the healthy extract (lane H) and which had the same
mobility as P1 synthesized in vitro (lane T). The reactions, both with the Pc-30 and with the
translation medium, were abolished by preabsorbing the antiserum with p e p l (lanes 5' and T'),
proving their specificity. P1 first appeared at 45 h post-infection (lane 2). Its concentration was
m a x i m u m in the Pc-30 at 69 h post-infection (lane 3) and seemed to decrease towards the end of
the infection period (lane 8).
The anti-pep2 serum clearly revealed P2 in the same samples which also contained P1 (Fig.
4c, lanes 2 to 8). Several other bands (marked with filled triangles), which were visible in all the
samples, whether healthy or infected, were certainly due to cellular contaminants. However, this
immunoblot was much cleaner than those obtained in preliminary experiments (Berna et al.,
Fig. 4. Detection of AIMV proteins in infected plants kept at 25 °C. The S-30 and Pe-30 fractions were
prepared from inoculated leaves at different times (22 h, 45 h, 69 h, 77 h, 93 h, 118 h, 153 h and 213 h
post-infection respectively in lanes 1 to 8). Controls were equivalent samples from healthy leaves (lanes
H), purified virus (lane V) and partially purified translation products (lanes T). The gel strengths were
12~ in (a), 6 ~ in (b) and (c) and 9 ~ in (d). (a) Immunoblots of the S-30 samples (2 ~tl)with the anti-coat
protein serum diluted 50000-fold. Lane V contained 26 ng virus. (b to d) Immunoblots of the Pe-30
samples (30 lal, which was equivalent to 600 rtl S-30) with the anti-pep 1 (b), anti-pep2 (c) and anti-pep3
sera (d), each diluted 5000-fold. Lane T contained 15 ng P 1 in (b), 5 ng P2 in (c) and 1 ng P3 in (d). Lanes
4', 5', T' and H' are as lanes 4, 5, T and H except that the antisera had been preincubated with 2 ~tg/mlof
the corresponding peptide.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
~wb
t~
O0
",../
O0
VV
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
1142
A. B E R N A A N D O T H E R S
I
I
I
I
I
I
I
I
I
I
(a)
©
•
,00
E
o
I
,
/
s/
I/.,
,,//.~
so
#/
/
),,
•
hot
"~,,e
0
\l~....
.'
",
'o..
~
"-.,
,.
-"-o
- ~,~..
c.y ~ . .
i'11ill
/'
I-~
I
I
,--,
"- ..
~A
-.. 0
" ~'
~..-m
~--- o............
I
I
I
I
I
t
I
I
t
I
I
I
I
I
~-
(b)
I00
4
N
•
50
~I
[
j
40
80
120
160
200
Time after infection (h)
Fig. 5. Accumulation of the NSPs and replicase activity as a function of time at 25 °C. (a) The
experiment of Fig. 4 and another similar experiment done on the same plant extracts were quantified
and the mean results normalized to the maximum amount of each protein : 15 ng P1,9 ng P2 and 2 ng P3
in 30 gl Pe-30. 0 , P1; A, P2; O, P3. (b) The replicase activity of the Pe-30 was normalized to the
maximum activity, i.e. 0.51 pmol UMP incorporated in 10 gl Pe-30 in 1 h.
1984), which demonstrates the improvement of the immunoblotting method. Preabsorption
with the peptide abolished the reaction with P2 from the Pe-30 as well as from the translation
medium (lanes 5' and T'). The kinetics of P2 accumulation seemed to follow rather closely that of
P1.
P3 was similarly assayed with the anti-pep3 serum (Fig. 4d). In spite of a significant
background (particularly the bands m a r k e d with arrows, on either side of P3), a band with the
same mobility as P3 synthesized in vitro (lanes T) appeared in the infected extracts at 45 h postinfection (lane 2). This band was absent from the healthy samples (lanes H) and was eliminated
by serum preabsorption with the peptide (lane 4'), as was the reaction with bona fide P3 from the
translation medium (lane T'). The reactions with cellular contaminants were insensitive to
serum preincubation (lanes 4' and H'). Unlike the other two NSPs, P3 reached its m a x i m u m
concentration in the Pe-30 very early in infection (lane 2) and disappeared rapidly thereafter
(lanes 3 to 8).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
AIM V non-structural proteins in plants
1143
Table 1. Absolute amounts of viral proteins and replicase activity at 25 °C and at 10 °C*
Amounts of indicated viral protein after infection at
ak
(
Protein assayed
Replicase (unit/g)
P1 (ng/g)
(pmol/g)
P2 (ng/g)
(pmol/g)
P3 (ng/g)
(pmol/g)
CP (~tg/g)
25 °C
7.0
18
0.14
10.5
0' 12
7
0-22
440
10 °C
6.3
28
0.22
17
0.19
10
0.31
67
* Plants of uniform size were inoculated and kept either at 25 °C for 48 h or at 10 °C for 170 h. The NSPs were
assayed in 30 Isl of Pe-30 and the coat protein (CP) was assayed in 2 gl of crude extract (before 30000 g
centrifugation), essentially as in Fig. 4 to 7. Replicase activity was assayed in l0 ~tl of Pe-30 and is expressed as
pmol UMP incorporated in 1 h. The results are normalized to 1 g of leaves.
Fig. 5 shows the results of quantifying the immunoblots, and presents the kinetics of
accumulation of the NSPs c o m p a r e d with the replicase activity of the immunoblotted samples
(Fig. 5 b; see also Le Roy et al., 1977). The curves are plotted as percentage of the m a x i m u m to
facilitate comparison, and it is evident that the accumulation of both P 1 and P2 closely followed
the replicase activity whereas the accumulation of P3 did not. As late as 200 h post-infection
leaves still contained a significant amount of P1, P2 and replicase, but very little P3 remained.
P2 seemed to be synthesized slightly earlier than P1. The difference, although small, was also
observed at 10 °C, as shown later.
F r o m calibrations such as those shown in Fig. 3, the m a x i m u m amounts of P1, P2 and P3 in a
volume of Pe-30 corresponding to 1 g of leaves were about 50 ng of P 1 (0.4 pmol) and 30 ng of P2
(0.33 pmol) after 70 to 80 h of infection, and about 6 ng of P3 (0.2 pmol) after 45 h. The accuracy
of these determinations was about 15 ~ (Fig. 5) and the results obtained with different batches of
plants were within a factor of two. Variations seemed to be related to the age of the plant:
infection of very young leaves proceeded slightly faster and yielded larger amounts of the N S P s
than that of fully-grown leaves.
Time course of accumulation of the viral proteins at 10 °C
W h e n the inoculated plants were kept at 10 °C, small white lesions a p p e a r e d on the inoculated
leaves 7 days after infection but no symptoms were seen on the upper leaves at 15 days postinfection. Coat protein was undetectable in the S-30 of the upper leaves (not shown) although it
was m a d e in the inoculated leaves. It took about 10 times as long at 10 °C as at 25 °C for a given
amount of coat protein to be m a d e (Fig. 6a).
The NSPs were assayed in the Pe-30 as in the preceding experiment. As shown in Fig. 6(b, c),
all three were synthesized and their accumulation proceeded much faster than expected from the
rate of coat protein synthesis. P2 was first detectable at 66 h post-infection, whereas P1 and P3
a p p e a r e d at 96 h. They accumulated for the following 4 to 5 days and reached their m a x i m u m at
about 200 h after infection (Fig. 7a). The decay of the NSPs was much slower at 10 °C than at
25 °C, but the difference seen at 25 °C between P1 and P2 on the one hand, and P3 on the other,
was also apparent at 10 °C.
Replicase activity (Fig. 7b) followed closely the amount of P1 throughout infection, and that
of P2 after 150 h. (A significant amount of P2 was found as early as 66 h post-infection without
detectable replicase.) As at 25 °C, there was no correlation between replicase activity and the
amount of P3.
To compare the absolute amounts of replicase and NSPs at the two temperatures, plants from
one inoculated batch were kept either at 10 °C or at 25 °C. The leaves were harvested when
replicase activity was 50 to 6 0 ~ of the maximum, i.e. after 48 h at 25 °C or 170 h at 10 °C. Table
1 shows that the absolute amount of replicase was nearly the same at the two temperatures and
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
1144
A. B E R N A A N D O T H E R S
(a)
1
2
3
5
6
7
8
9
V
P4
(b)
1
2
3
4
5
6
7
8
9
H
P1
P2
(c)
1
2
3
4
5
6
7
8
9
H
P2
P3
Fig. 6. Detection of A1MV proteins in infected plants kept at 10 °C. Inoculated leaves were collected
after different times at 10 °C (66, 96, 120, 152,200,224, 272, 296 and 345 h post-infection respectively in
lanes 1 to 9), as well as comparable leaves from healthy controls (lane H). They were processed as in Fig.
4. (a) is equivalent to Fig. 4(a). In (b) the Pe-30 samples were separated in a 6 ~ gel and immunoblotted
with a mixture of the anti-pepl and anti-pep2 sera, both diluted 5000-fold. In (c) the Pe-30 samples were
separated in a 9 ~ gel and immunoblotted with a mixture of the anti-pep2 serum (diluted 10000-fold)
and anti-pep3 serum (diluted 2500-fold).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
AIMV non-structural proteins in plants
I
I
I
I
1
1,145
I
I
!
(a)
100
I \\ - - ' • ~
~__.~ t - ' ~ ' ~
\\
E
//7/.I
\\\
/Ol/
50
•
\
\'\\
//.¢
©/
~)
I
i
I
i
I
I
(b)
lOO
N
50
I
J
80
160
240
320
Time after inoculation (h)
Fig. 7. Accumulation of(a) the non-structural proteins and (b) replicase activity as a function of time at
10 °C. The data from three experiments similar to Fig. 6 (relative S.E.M.6 to 15~) were combined and
normalized to the maximum amount of viral protein, i.e. 19 ng P l, 14 ng P2 and 6 ng P3 in 30 gl Pe-30.
The replicase activity of the Pe-30 samples was also normalized to the maximum, i.e. 0.67 pmol UMP
incorporated in 10 gl Pe-30 in 1 h. O, P1; A, P2; O, P3.
that the amounts of the three NSPs were slightly higher at 10 °C than at 25 °C whereas the
amount of coat protein was sevenfold less at the lower temperature than at the higher
temperature.
Distribution of P1 and P2 among the subcellular fractions
In the preceding experiments, accumulation of the NSPs was examined in a crude m e m b r a n e
fraction (Pe-30) from which large organelles had been eliminated. We verified that this fraction,
which contains most of the viral replicase (Le Roy et al., 1977), also contains the major
quantities of P 1 and P2, by testing amounts of the low-speed pellet (Pe-1) and of the a m m o n i u m
sulphate-fractionated S-30 equivalent to 400 I,tl of S-30. Neither protein could be detected in any
of the supernatant fractions, but 20 to 3 0 ~ of P1 and P2 were consistently found in the Pe-1
throughout infection, both at 25 °C and at 10 °C. This could probably be accounted for by crosscontamination of the different types of membranes.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
1146
A. B E R N A A N D O T H E R S
DISCUSSION
This work shows that antibodies raised against synthetic C-terminal peptides can detect
minute amounts of the homologous protein containing that peptide by immunoblotting. As
pointed out by Westhof et al. (1984), the hydrophilicity of the peptide may not be very important
and indeed, of the three that we have used, only one, pep2, was strongly hydrophilic (compare
Fig. 1 ofBerna et al., 1984, and Fig. 1 of the present paper). Nevertheless, all three peptides were
strongly immunogenic and the antisera obtained detected P1, P2 and P3 with similar
efficiencies (Fig. 3). It was fortunate for us, in fact, that the antisera made against peptides
derived from the sequences of RNAs l and 2 of strain 425 (Cornelissen et al., 1983a, b)
recognized P1 and P2 of strain S.
The relative amounts of a given viral protein in the different lanes of an immunoblot were
easy to quantify. However, the absolute amounts were more difficult to evaluate. The
estimations relied on the use of radioactive translation products as standards, and it was
implicitly assumed that these products had exactly the same C-termini as the natural proteins.
We observed, however, that the reaction of the P1 standard with the anti-pep 1 serum weakened
significantly over a period of a few months without loss of radioactivity other than that due to
the decay of 35S. This could have resulted from the breakdown of a few amino acids at the Cterminus. Even though freshly prepared standards were used for calibration, the amount of P 1 in
leaves would have been overestimated if such breakdown occurred during in vitro protein
synthesis.
Our major finding is that the amounts of P1 and P2 in a crude membrane fraction containing
the replication complex were correlated with replicase activity throughout infection, both at
25 °C and 10 °C, but that the amount of P3 was not. This lends support to the idea that P1 and P2
(but not necessarily P3) are part of the replication complex. The fact that P2 was found early
after inoculation at 10 °C, when neither replicase nor P1 could be detected, does not contradict
this statement: it is conceivable that both P1 and P2 are required for replicase activity, and P1
may have been limiting. Maximum replicase activity, as well as the maximum amounts of P1
and P2, were nearly the same at the two temperatures (possibly a little higher at 10 °C). Thus,
10 °C is not a restrictive temperature for the assembly of a functional replication complex. This
step of A1MV replication may be equivalent to the 2-thiouridine-sensitive (or 10 mM-guanidinesensitive) step of TMV replication (Dawson & Schlegel, 1976; Dawson, 1976) which is known to
take place at 12 °C. The viral protein which is most affected by temperature is the coat protein :
it takes seven to ten times as long at 10 °C as at 25 °C to obtain a given amount of coat protein in
the inoculated leaves, whereas the same temperature drop only slows down the accumulation of
NSPs by three- to fourfold. This could reflect the smaller proportion of RNA 4 in the viral RNA
mixture synthesized at the lower temperature (Franck & Hirth, 1976). In addition, the lower
amount of coat protein may indirectly favour the synthesis of NSPs by slowing down
encapsidation of the genomic RNAs.
The infections that we have followed were obviously non-synchronous and therefore cannot
be interpreted in terms of a single infection cycle. What we measured was the balance between
the synthesis and the decay of viral proteins. Synthesis of the NSPs predominated only for a
limited period, during which the infection was spreading from cell to cell as evidenced by the
enlargement of the symptoms. Later (presumably after all the competent cells had been reached
by infection), decay took over and the NSPs gradually disappeared from the crude membrane
fraction, P3 more rapidly than the other two. Our finding that P3 reaches its maximum
concentration sooner than PI and P2 at 25 °C may be explained in two ways: either P3 is
synthesized very transiently (this is the case for the 30K protein of TMV; Watanabe et al., 1984)
whereas the cell keeps making P1 and P2 for several days, or P3 decays faster than the other two.
Several explanations can be proposed to account for the apparent decay of the NSPs. They
could be cleaved by proteases or otherwise become unrecognizable to our antibodies. If
proteolytic degradation were taking place, we would only be able to detect the pieces with an
intact C-terminus, and indeed some of our immunobtots show faint bands which could
correspond to a C-terminal fragment of P2 (Fig. 6c, open arrowhead). Alternatively, some NSPs
could migrate to a different subcellular compartment. Our search for P1 and P2 in the low-speed
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51
A l M V non-structural proteins in plants
1147
p e l l e t a n d in t h e h i g h - s p e e d s u p e r n a t a n t d i d n o t g i v e a n y e v i d e n c e f o r t h i s . T h e f a t e o f P 3 will b e
d e a l t w i t h in a f u t u r e r e p o r t .
We thank M.-J. Gagey for excellent technical assistance, M. Meyer for taking care of the rabbits, M. Le Ret for
amino acid analysis and C. Hubert for photography. This work was supported in part by grant 84V 0813 from the
Minist~re de la Recherche et de la Technologie (Biotechnology Program).
REFERENCES
BARANY,G. & MERRIFIELD,R. B. (1980). Solid phase peptide synthesis. In The Peptides, vol. 2, pp. 1 284. Edited by
E. Gross & J. Merenhofer. New York & London: Academic Press.
BERNA, A., BRIAND, J.-P., STUSSI-GARAUD,C., GODEFROY-COLBURN, T. & HIRTH, L. (1984). Immunodetection of a nonstructural protein of alfalfa mosaic virus (P2) in infected tobacco plants. Annales de Virologie 135E, 285-296.
BERNA, A., GODEFROY-COLBURN,T. & STUSSI-GARAUD,C. (1985). Preparation of an antiserum against an in vitro
translation product of alfalfa mosaic virus RNA 3. Journal of General Virology 66, 1669-1678.
CLERX, C. M. & BOL, J. F. (1978). Properties of the solubilized RNA-dependent RNA polymerase from alfalfa
mosaic virus. Virology 91, 453 463.
CORNELISSEN, B. J. C., BREDERODE, F. TH., MOORMANN, R. J. M. & BOL, J. F. (1983a). Complete nucleotide sequence of
alfalfa mosaic virus RNA 1. Nucleic Acids Research l l , 1253-1265.
CORNELISSEN, B. J. C., BREDERODE, F. TILt.,VEENEMAN, G. H., VAN BOOM, J. H. & BOL, J. F. (1983b). Complete nucleotide
sequence of alfalfa mosaic virus RNA 2. Nucleic Acids Research II, 3019 3025.
DAWSON, W. O. (1976). Guanidine inhibits tobacco mosaic virus R N A synthesis at two stages. Intervirology 6, 8389.
DAWSON, W. O. & SCHLEGEL, D. E. (1976). The sequence of inhibition of tobacco mosaic virus synthesis by
actinomycin D, 2-thiouracil and cycloheximide in a synchronous infection. Phytopathology 66, 177-181.
FRANCK, A. & HIRTH, L. (1976). Temperature-resistant strains of alfalfa mosaic virus. Virology 70, 283-291.
GODEFROY-COLBURN,T., THIVENT, C. & PINCK, L. (1985). Translational discrimination between the four R N A s of
alfalfa mosaic virus, a quantitative evaluation. European Journal of Biochemistry 147, 541-548.
HOPP, T. P. & WOODS, K. R. (1981). Prediction of protein antigenic determinants from amino acid sequences.
Proceedings of the National Academy of Sciences, U.S.A. 78, 3824-3828.
JOSHI, S., NEELEMAN, L., PLEIJ, C. W. A., HAENNI, A. L., CHAPEVILLE, F., BOSCH, L. & VAN VLOTEN-DOTING, L. (1984).
Nonstructural alfalfa mosaic virus RNA-coded proteins present in tobacco leaf tissue. Virology 139, 231-242.
KAMER,G. & ARGOS,P. (1984). Primary structural comparison of R N A-dependent polymerases from plant, animal
and bacterial viruses. Nucleic Acids Research 12, 7269 7282.
LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
Nature, London 227, 680 685.
LE ROY, C., STUSSI-GARAUD,C. & HIRTH, L. (1977). RNA-dependent R N A polymerases in uninfected and in alfalfa
mosaic virus-infected tobacco plants Virology 82, 48-62.
NASSUTH,A. & BOL,J. F. (1983). Altered balance of the synthesis of plus and minus strand RNAs induced by R N A 1
and 2 of alfalfa mosaic virus in absence of RNA 3, Virology 124, 75-85.
NASSUTH,A., ALBLAS,F. & BOL, J. F. (1981 ). Localization of genetic information involved in the replication of alfalfa
mosaic virus. Journal of General Virology 53, 207 214.
PINCK, L. & HZRTH, L. (1972). The replicative RNA and the viral R N A synthesis rate in tobacco infected with
alfalfa mosaic virus. Virology 49, 413-425.
RAVELONANDRO,M., PINCK, M. & PINCK, L. (1984). Complete nucleotide sequence of RNA 3 from alfalfa mosaic
virus, Strain S. Biochimie 66, 395-402.
SAMAC, D. A., NELSON, S. E. & LOESCH-FRIES, L. S. (1983). Virus protein synthesis in alfalfa-mosaic-virus-infected
alfalfa protoplasts. Virology 131,455-462.
SARACHU, A. N., NASSUTH, A., ROOSIEN, J., VAN VLOTEN-DOTING, L. & BOL, J. F. (1983). Replication of temperaturesensitive mutants of alfalfa mosaic virus in protoplasts. Virology 125, 64-74.
TASHEVA,8. & DESSEV,G. (1983). Artifacts in sodium dodecyl sulfate-polyacrylamide gel electrophoresis due to 2mercaptoethanol. Analytical Biochemistry 129, 98 102.
TOWBIN, H., STAEHELIN,T. & GORDON, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to
nitrocellulose sheets: procedure and some applications. Proceedings of the National Academy of Sciences,
U.S.A. 76, 4350 4354.
VAN mL, R, a. L. & VAN VLOTEN-DOTING, L. (1979). Translation of alfalfa mosaic virus RNA 1 in the m R N A dependent system from rabbit reticulocyte lysates. European Journal of Biochemistry 93, 461 468.
WATANABE,Y., EMORI, Y., OOSHIKA, I., MESH1, T., OHNO, T. & OKADA, Y. (1984). Synthesis of TMV-specific RNAs and
proteins at the early stage of infection in tobacco protoplasts: transient expression of the 30K protein and its
m R N A . Virology 133, 18-24.
WEENING, C. J. & BOL, J. e. (1975). Viral R N A replication in extracts of alfalfa mosaic virus-infected Viciafaba.
Virology 63, 77 83.
WESTHOF, E., ALTSCHUH, D., MORAS, D., BLOOMER, A. C., MONDRAGON, A., KLUG, A. & VAN REGENMORTEL, M. H. V.
(1984). Correlation between segmental mobility and the location of antigenic determinants in proteins.
Nature, London 311, 123 126.
(Received 23 December 1985)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 07:26:51