J. gen. Virol. (198o), 50, 45t-454
451
Printed in Great Brffain
A Re-evaluation of the Structure of Narcissus Mosaic Virus and
Polymers Made from its Protein
( A c c e p t e d I9 M a y 198o)
SUMMARY
The structure of narcissus mosaic virus (NMV) particles and its protein polymers
have been re-examined. The virus particle contains 44 coat protein subunits in five
turns of its helix and has a mol. wt. of about 36 x I@. Its protein forms stacked discs
or rings at pH 5 which, with time, are replaced by two-start helices with 26 subunits
in three turns or nine in one.
Particles of viruses belonging to the potato virus X family such as papaya mosaic virus
(PMV; Tollin et al. I979) and potato virus X (PVX; Tollin et al. I98o) have between eight
and nine subunits per turn of their helices, the fraction depending on the true repeat.
Narcissus mosaic virus (NMV) is an apparent exception (ToUin et al. 1975) and we have reexamined its structure. We confirm the report that NMV protein assembles as two-start
helical particles (Robinson et al. I975) and re-describe their structure. We also report on an
additional structure.
NMV, a kind gift from Dr W. P. Mowat, was purified as for clover yellow mosaic virus
(Bancroft et al. 1979) and disassembled at pH 8 in 2 M-LiC1 after freezing (Robinson et al.
r975). The RNA was removed by centrifugation at roooo rev/min for 15 min in a Sorvall
centrifuge followed by centrifugation at 360oo rev/min for 4 h in a Beckman Model L
centrifuge to remove residual virus. The resulting supernatant solution had a 28o/25o nm
ratio of I-8z. The protein, at 2 to 4 mg/ml, was dialysed overnight at 4 °C against pH 4, 5
and 6 buffers composed of O-Ol M- or o.I M-sodium citrate containing o, o-I M-, o.2 ~- and
0"3 M-(NH~)2SO4 or against pH 8, o.1 M-tris containing 0"3 M-(NH~)2SO4 (Robinson et al.
I975) measured at 4 °C. The dialysed solutions were then kept at room temperature before
electron microscopy in I ~oo uranyl formate in a Philips EM 2oi or an AEI EM6B. Best reassembly results, determined by electron microscopy, were obtained with o.o~ M- or o.I Mcitrate at pH 5 in o.I M-(NH~)2SO4. Optical diffraction was as described by Tollin et al.
0979) and calibrations were made using catalase (Luftig, 1967).
NMV particles have a single-start helix repeating in five turns (Tollin et al. 1968). Tollin
et al. (I 975) deduced that there were 6 4/5 subunits per turn by using the ratio of the distances
from the meridian of the first intensity maxima on the fifth and first layer lines and applying
the helical selection rule (Cochrane et al. I952). We have measured the pitch of NMV to be
3"4 nm using catalase as the internal standard rather than more indirect estimates and the average unpacked diam. to be I4.1 nm with a standard deviation of o. 8 nm (6o particles), probably
partly reflecting different degrees of flattening. Using these figures, taking the diffraction
pattern in Fig. I (b) of Tollin et al. (1975) and employing a calculation less subject to error
than the ratio of distances from the meridian, we find that 2~rrR = lO.34, where r is the
radius of the particle and R the position of the maximum on the first layer line. This value
lies within the range of values obtained by us from measurements of diffraction patterns
obtained from 19 single particles which gave an average value of lO"9 with an error about
the mean of o.6 (95 ~ confidence interval). The values of 2 n r R as the arguments for the first
maxima of Bessel contributions Js and J9 for a cylindrical particle are 9"65 and IO'7I,
Downloaded
from www.microbiologyresearch.org by
OO22-I317/80/0000-4155 $02.00
© 1980 SGM
IP: 88.99.165.207
On: Sat, 17 Jun 2017 23:51:04
452
Short communications
Fig. I. Electron micrographs of NMV protein polymers after (a) I and (b) 3 days at zo °C in
oq M-citrate, pH 5, containing o'l M-(NH4)~SO4. The inset is a diffraction pattern showing
stacked rings. Electron micrographs of end-on discs or rings of PMV protein are shown in (c).
respectively (Moody, 1967), so our results correspond most closely to Jg.Jfit is assumed that
the stain penetrated to a depth of o. 5 to I.O nm, values between J8 and J9 are obtained. We
confirm that there is an approximate true repeat of the structure after five turns, (Tollin et al.
I975) since we obtained a value of 5"I2 + o-26 for the I9 particles. Since the first and fifth
layer lines usually occupied the same quadrant in single-sided images, we infer that the helix
is composed of 84 subunits per turn. A radial projection of such a helix and its corresponding transform are given in Fig. 2(c).
Knowing the above and taking the tool. wt. of the R N A and protein subunits as 2.2 x to e
and 24ooo (M. Short, unpublished data) respectively, and the particle length as 55o nm
(Mowat, I97I) we calculate that the virus is composed of about 14oo subunits, has a mol. wt.
of about 36 × [o 6 and contains five nucleotides per protein subunit. These values are consistent with those found for PMV (Tollin et al. 1979) and PVX (Tollin et al. 198o).
We also re-examined the polymers made by the protein in the absence of nucleic acid and
found that protein samples after I day at 2o °C contained circular rings as well as short
tubes (Fig. I a). The short tubes are composed of stacks of rings or discs (see inset, Fig. I a)
and have a repeat of 3.6 nm as measured from r 2 particles. We attempted to confirm the
value of between 8 and 9 subunits per turn deduced for the virus by examining end-on discs.
Fig. 1 (c) shows such particles and between 8 and 9 subunits can be directly counted. Since
the diameter of the discs and short rings is within the range found for the virus, we conclude
that 9 is the more likely value.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 23:51:04
Short communications
453
(a)
(i'O,1)
(8,1)
(i'2,3)
\ \,
~
(6,3)
(i'~.3)
•
7,
•
•
•
(2,2)
•
(b)
(~,2)
(64)~
(8,1)
(~,1)
~lOnm~
(8,1)
/////
~
~
\,
(6~4)
TT.."
11.4
nm/
•
.
J
...'"
•
j_/(2,3)
.
l i . - ' ' "
(2~3)
•
"
(1"-0,
• 2)
(8gl)
"
n
~-10nm-~
(c)
(9,1)
/ /
/
~
(8,4)
~
-
~(i"0,6)
¢
(i"O,6)
17
•
•
•
•
•
•
"
_
(1,53
(8,4)
~ ( T , 5 )
. . . • - ..• . ..• . .i• . .
(9,1)
~--10nm~
Fig. 2. Radial projections (left) and transforms (right) of (a)double helix with 9 subunits repeating
in one turn; (b) double helix with 26 subunits repeating in three turns; (c) single helix with 44
subunits repeating in five turns, as for the virus.
After incubation at room temperature for an additional 2 to 3 days, stacked ring particles
vere replaced by long flexuous tubes (Fig. I b). These are two-start helices repeating either
n every turn or third turn as evidenced from diffraction patterns which were similar to those
hown in Fig. 2 of Robinson et al. (I975). The helices have a pitch of 7"6 nm as measured
tom 11 particles and a diameter like that of the virus. Knowing this, we have generated
adial projections and transforms for the helical polymers, using the measured dimensions,
fraction between 8 and 9 subunits per turn and assuming a spherical subunit shape and no
lattening (Fig. 2).
The positions of the maxima in the present Fig. 2 are consistent with those found by us
n d in Fig. 2 of Robinson et aL (I975). We have interpreted Fig. 2(¢) of Robinson et al.
1975) to indicate that there are 83 rather than 91 subunits per turn since in the latter case,
he maxima on the first layer line would be expected to be further away from the meridian
ban those on the second layer line and this was not observed.
Protein added to NMV R N A in o.oi M-tris, pH 7, formed single-start helices like those
of the virus. This means that the formation of the discs or double helices does not result
from protein altered in the isolation process.
We conclude that the structure of NMV is not inconsistent with that of other viruses in
the PVX family and that the double-helical structures repeating in every one and three
turns have, respectively, 2 and I½ more protein subunits per turn than originally described
(Robinson et al. I975).
16
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 23:51:04
V l g 50
454
Short communications
We wish to thank Mr R. Johnston for some of the purification work and Dr P. Tollin for
helpful discussions. This work was supported in part from grants from the National
Research Council of Canada to J. B. Bancroft and N. C. Payne.
1 Department of Plant Sciences and 2 Chemistry
The University of Western Ontario
London, Ontario, Canada, and
The John Innes Institute
Norwich, Norfolk, U.K.
J. B. BANCROFT 1
G . J. HILLS a
J. F . RICHARDSON 2
REFERENCES
BANCROFT, J. B., ABOUHAIDAR, M. & ERICKSON, J. W. (I979). T h e assembly of clover yellow mosaic virus a n d
its protein. Virology 98, I2I-I3O.
COCHRANE, W., CglCK, F. H. C. & VAND, V. (I952). T h e structure o f synthetic polypeptides. I. T h e t r a n s f o r m
o f a t o m s o n a helix. Acta Crystallographica 5, 581-586.
LUFTIG, g. 0967). A n accurate m e a s u r e m e n t of the catalase crystal period a n d its use as a n internal m a r k e r
f o r electron microscopy. Journal of Ultrastructure Research ~_,o, 9 I-IO2.
MoooY, u . F. 0967). Structure o f t h e s h e a t h o f bacteriophage T4. t. Structure o f the contracted s h e a t h a n d
polysheath. Journal of Molecular Biology 25, I67-2oo.
MOWAT, W. P. (I971). Narcissus mosaic virus. C M I / A A B Descriptions of plant viruses No. 45.
ROaiNSON, D. J., HUTCrlESON, A., TOLLIN, P. & WILSON, H. R. (1975). A double-helical structure for re-aggregated
protein o f narcissus m o s a i c virus. Journal of General Virology 29, 325-33o.
TOLLIN, P., WILSON, H. R. & YOUNG, O. W. (I968). X-ray evidence o f the helical structure o f narcissus mosaic
virus. Journal of Molecular Biology 34, 189-192.
TOLLIN, P., WILSON, H. R. & MOWAT, W. P. (1975). Optical diffraction from particles o f narcissus mosaic virus.
Journal of General Virology 29, 33t-333 .
TOLLIN, P., BANCROFT, J. B., RICHARDSON, J. F., PAYNE, N. C. & BEVERIDGE, T. J. (I979). Diffraction studies o f
p a p a y a m o s a i c virus. Virology 98, IO8-I 15.
TOLLIN, P., WILSON, H. R. & BANCROFT, J. B. (I980). F u r t h e r observations on the structure of potato virus X.
Journal of General Virology 49, 407-4 IO.
(Received 19 February 198o)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 23:51:04