769
J. gen. Virol. (1984), 65, 769-779. Printedin Great Britain
Key words: VSV/Sendai virus/nucleocapsid
Preparation and Analysis of the Nucleocapsid Proteins of Vesicular
Stomatitis Virus and Sendai Virus, and Analysis of the Sendai Virus
Leader-NP Gene Region
By B E N J A M I N M. B L U M B E R G , * C O L O M B A G I O R G I , t
DANIEL KOLAKOFSKY
K E I T H ROSE~ A~D
D~partement de Microbiologie, 64 Avenue de la Roseraie, 1205 Geneva, Switzerland
(Accepted 17 January 1984)
SUMMARY
A procedure is presented for isolating the nucleocapsid proteins, N and NP from
vesicular stomatitis virus and Sendai virus respectively, in soluble form. These proteins
were suitable for the determination of their blocked amino-terminal peptide sequences
by gas-liquid chromatography/mass spectrometry at the low nanomole level. The N
protein prepared by this procedure was previously shown to retain some of its expected
biological activity. The sequence of 626 nucleotides from the 3' end of the Sendal virus
genome, which includes the first one-third of the NP gene, was determined. Using this
information, primer extension studies on intracellular Sendai virus mRNAs allowed
the determination of the structure of the leader-NP intervening sequence and the 5'
end of the NP mRNA. Comparison of the amino termini of the nucleocapsid proteins
with their respective mRNA sequences revealed that these proteins are similarly
processed in vivo.
INTRODUCTION
The non-segmented minus-strand ( - ) RNA genome of the rhabdovirus vesicular stomatitis
virus (VSV), and that of the paramyxovirus Sendal virus, are always found tightly complexed in
helical nucleocapsid structures with their respective viral nucleocapsid proteins, N and NP. The
nucleocapsid structure protects the genomes from nuclease attack, and in conjunction with two
less tightly associated viral proteins which constitute the viral RNA polymerase (NS and L of
VSV: Emerson & Yu, 1975; P and L of Sendal: Stone et al., 1972; Marx et al., 1974; Lamb &
Mahy, 1975), displays all the enzyme activities required for synthesis of the viral mRNAs
(Banerjee et al., 1977; Kingsbury, 1974).
Because the nucleocapsids are of central importance in both virus structure and replication, it
would be of interest to study the properties of the purified nucleocapsid proteins themselves.
However, previous attempts to isolate the N protein from nucleocapsids by treatment with
dimethyl urea (S. Emerson, personal communication) or with guanidinium salts (G. Wertz,
personal communication) have resulted in preparations with intractable physical properties.
Both proteins can be removed from their nucleocapsids by treatment with SDS or phenol, but
these reagents are likely to denature the protein. The viral nucleocapsids are also stable to highsalt conditions. Indeed, banding in CsC1 gradients at a buoyant density of 1.33 g/ml (2-5 M-CsC1)
is diagnostic for VSV and Sendai virus nucleocapsids, for under these conditions all other
cellular RNA-protein complexes disaggregate (Leppert et al., 1979). Thus, although these
proteins are major components of the mature virion (Wagner et al., 1969; Mountcastle et al.,
197t), they have resisted attempts at purification. In this paper, we describe a simple procedure
for the isolation of N and NP proteins from intracellular nucleocapsids, based on CsCI density
gradient centrifugation in the presence of guanidinium chloride.
t" Laboratorio di Virologia, Istituto Superiore di Sanitfi., Roma, Italy.
:~D6partement de Biochimie M6dicale C.M.U., 9 Avenue de Champel, 1211 Geneva 4, Switzerland.
Downloaded
from www.microbiologyresearch.org by
0022-1317/84/0000-5995 $02.000
1984 SGM
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41
770
~.
M. B L U M B E R G
AND
OTHERS
The VSV N protein prepared by this procedure was previously shown to self-assemble in the
absence of RNA, to assemble RNAs into nucleocapsid-like structures, and in particular to
selectively encapsidate VSV leader RNA over mRNAs (Blumberg et al., 1983). The selectivity
of this encapsidation is thought to reside in the nucleation reaction, i.e. the association of N
protein with the sequence of the first 14 nucleotides at the 5' end of the VSV leader RNA. The
subsequent elongation of nucleocapsids is thought to result from the cooperative binding of free
N protein with that already assembled in the initiated complex and to be sequence-independent.
In order to study the nucleocapsid assembly process more closely, it is essential to know the
structure of the N and NP proteins, since different forms of these proteins may play different
roles in the assembly mechanism.
Like many cytoplasmic proteins (Wold, 1981), the Sendai virus NP protein is N~-blocked and
refractory to sequencing (Heggeness et al., 1981). By adopting a gas-liquid chromatography/
mass spectrometry (GLC-MS) micromethod recently developed by Rose et al. (1983 b), we were
able to determine the blocked amino-termini peptide sequences of both the VSV N and the
Sendai virus NP proteins (Rose et al., 1983a). The basic technique (Gray & del Valle, 1970)
involves acetylation, protease digestion and permethylation of the protein to produce volatile
derivatives suitable for mass spectrometry. The GLC-MS techniques of Rose et al. (1983a, b)
should therefore be of general interest, as sequence information can be obtained directly from 2
to 5 nmol of proteins with blocked amino terminal. In this paper we report the DNA sequence of
the first one-third of the Sendai virus NP gene; the VSV N gene sequence has been published
(Gallione et al., 1981). Comparison of the peptide and mRNA gene sequences shows that both
proteins have been similarly processed, by cleavage of the initiating methionine and N ~acetylation of the following amino acid. In addition, primer extension experiments and other
data described here define the 5' end of the NP mRNA and help to define the nature of the
Sendai leader-NP gene intervening sequence.
METHODS
Preparation o f infected cell cytoplasmic extract. BHK21 cells in minimal essential medium (MEM) + 5 ~ foetal
calf serum were infected with standard Indiana serotype VSV (Mudd Summers strain) or Sendai virus (Harris
strain) at an m.o.i, of 10 and grown at 33 °C until c.p.e, was evident in the entire cell population. Cytoplasmic
extracts were prepared as previously described (Leppert et al., 1979). Buffers used were as follows. ET: 10 mMTris HCI pH 7-4, 1 mM-EDTA; TNE: 20 mM-Tris-HCI pH 7.4, 100 mM-NaC1, 1 mM-EDTA. For radiolabelling,
[3H]uridine or [3sS]methionine were added.
Preparation o f VS V Nprotein. Cytoplasmic extracts were prepared from VSV-infected cells at a concentration of
25 × 106 cells/ml (Leppert et al., 1979), and 9 ml portions of the extract were centrifuged on 20 ml, 20 to 4 0 ~ CsC1
gradients containing ET, overlaid with 7 ml of 5 ~o sucrose containing TNE, for 3 h at 25 000 rev/min in the SW28
rotor at 15 °C. During this relatively short centrifugation, the viral nucleocapsids (approx. 200S) approach their
buoyant density before other cellular or viral components. This single step thus provided a band of relatively pure
nucleocapsids containing mainly the viral N protein, since the high CsC1 concentration removes the NS and L
proteins (Szilagyi & Uryvaev, 1975).
The visible nucteocapsid band (t-33 g/m/) was removed by puncturing the side of the tube, diluted threefold
with ET, layered directly onto a second 20 ml, 20 to 4 0 ~ CsCI gradient containing 3 M-guanidiniu m chloride, and
re-centrifuged for a longer time so that the viral RNA could pellet. After centrifugation for 18 h at 27000 rev/min
in the SW28 rotor at 15 °C, two bands in the middle of the gradient were visible: a minor band at the position of
intact nucleocapsids, and a major band at a density of 1.26 to 1.28 g/ml, corresponding to the position of the 35Slabelled protein in Fig. 1. The upper band was removed through the side of the tube and vacuum-dialysed at 0 °C
against three changes of 50 mM-Tris HCI pH 7.8, 1 M-NaC1, 1 mM-EDTA. After clarification by low-speed
centrifugation, the final protein concentration was 1 to 2 mg/ml. The N protein prepared by this procedure could
be stored as a clear solution for months at 0 °C in 1 M-NaCI at pH 7.8. Warming the clear solution to room
temperature or lowering the pH to 6.5 resulted in visible opalescence due to the self-assembly of the protein into
helical structures, even in the absence of R N A (Blumberg et al., 1983). In a typical preparation approximately 1.2
x 109 VSV-infected cells (harvested 16 to 18 h post-infection) yielded 2 to 4 mg of soluble N protein.
Sequencing o f the first one-third o f the Sendai virus N P gene. The Sendai virus genome was cloned into pBR322,
and a series of overlapping plasmids was determined to represent the NP, P/C, and M genes (Dowling et al., 1983 ;
Giorgi et al., 1983b). Both strands of appropriate restriction fragments from N P gene clones SL-3 and SL-11
(Dowling et al., 1983) were sequenced by the Maxam & Gilbert (1980) procedure. The sequence was extended to
the 3' end of the genome by direct R N A sequencing (Giorgi et al., 1983a; Dowling et al., 1983).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41
V S V and Sendal virus nucleocapsid proteins
771
Primer extension. Restriction fragments from clones SL-3 and SL- 11 (see text and Fig. 2 of Dowling et al., 1983)
were isolated from polyacrylamide gels as dsDNA, and 5' end-labelled by means of T4 polynucleotide kinase and
[V-32p]ATP after pretreatment with alkaline phosphatase. For aqueous annealing, 20000 ct/min of the labelled
restriction fragment was mixed with RNA in 6 p.1 water and heated for 2 rain at 90 °C. After addition of 2 ~tl
0.8 ~t-NaC1, the solution was annealed 15 rain at 60 °C. For forrnamide annealing, 20 000 ct/min of the labelled
restriction fragment was mixed with RNA in 75/al of a solution containing 80% formamide, 0.4 M-NaCI, 10 m~PIPES pH 6-4, 1 mM-EDTA, heated 10 min at 60 °C and annealed 2 h at 49 °C. The nucleic acids were then
ethanol-precipitated and re-dissolved in 6 ttl H20. After annealing, the primers were extended in 20 ~tl reactions
containing 50 mM-Tris-HCl pH 8-3, 10 mM-MgClz, 50 mM-NaC1, 5 mM-dithiothreitol, 500 nM each of dATP,
dGTP, dCTP and TTP, and 10 units of avian myeloblastosis virus reverse transcriptase (Life Sciences, St
Petersburg, Fla., U.S.A.), by incubation for 10 rain at 37 °C then for 30 rain at 42 °C. Nucleic acids were recovered
by ethanol precipitation and analysed by polyacrylamide gel electrophoresis.
RESULTS
Preparation and analysis o f N and N P proteins
S o m e t i m e ago, we looked for possible differences b e t w e e n V S V virion and intracellular
nucleocapsids by e x a m i n i n g their stability to increasing c o n c e n t r a t i o n s of g u a n i d i n i u m chloride
during centrifugation on CsC1 gradients. A l t h o u g h little or no difference in the relative stability
o f these nucleocapsids could be found, we d e t e r m i n e d t h a t the p r e s e n c e of 3 M-guanidinium
chloride in the CsC1 gradients was sufficient to deproteinize the viral nucleocapsids. Fig. 1
d e m o n s t r a t e s this point. W h e n intracellular nucleocapsids labelled either w i t h [3 H ] u r i d i n e or
w i t h [35 S]methionine were centrifuged on such gradients, the 3 H-labelled R N A s e d i m e n t e d to
the pellet, w h e r e a s the 3sS-labelled protein n o w b a n d e d at the buoyant density of free protein
(1.28 g/ml). I n t a c t nucleocapsids, w h i c h would b a n d at 1.33 g / m l (fraction 6), were almost
entirely absent.
I
I
× 2
F,
*6
.o
1
5
10
Fraction no.
Fig. 1. Equilibrium sedimentation of viral nucleocapsids on CsCl-guanidinium chloride density
gradients. VSV nucleocapsids were isolated from cytoplasmic extracts of standard virus-infected BHK
cells labelled with either [3H]uridine or [35S]methionine by velocity sedimentation on 5 to 23 ~o sucrose
gradients containing NTE buffer plus 0.1 ~ Nonidet P40 (Leppert et al., 1979). A 250 ~1 portion of each
pool of labelled nucleocapsids was then layered onto a 3.7 ml, 20 to 40~ CsC1 gradient in ET buffer plus
3 M-guanidinium chloride. After centrifugation for 16 h at 55000 rev/min in the SW60 rotor at 15 °C,
the gradients were collected by pumping through a capillary tube so as not to disturb the pellet and the
bottom 200 ~tl, which together are considered the pellet fraction. Aliquots of each gradient fraction were
counted by liquid scintillation for 3H (O) and 3sS (O).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41
772
B. M. BLUMBERG AND OTHERS
1
2
3
4
G
N
NS
M
Fig. 2. Analysis of isolated VSV and Sendai virus nucleocapsid proteins by SDS-PAGE. Lanes 1 and
5: Sendai virus and VSV virion protein markers respectively. Lane 2: solubilized Sendai virus NP
protein. Lanes 3 and 4: solubilized VSV N protein. Proteins were prepared by centrifugation in CsC1guanidinium chloride gradients as described in the text, electrophoresed on a 10~ polyacrylamide gel as
described by Laemmli, and stained with Coomassie Brilliant Blue.
This finding formed the basis for the purification of VSV N protein (see Methods).
Subsequently, we have also prepared the analogous Sendai virus NP protein by this procedure,
as experiments of Mountcastle et al. (1970) suggested that CsC1 gradient centrifugation removes
the L and most of the P proteins from the nucleocapsid. Fig. 2 shows electropherograms of VSV
N and Sendai virus NP proteins. On occasion, the VSV N protein contained a small amount of
the viral NS protein (see Fig. 2, lane 3). Interestingly, such preparations did not contain
detectable levels of either protein kinase or nucleotide phosphohydrolase activity when assayed
in the presence of [y-32p]ATP (data not shown).
The VSV N protein prepared as above proved to be suitable for direct amino terminus
analysis by a modification of a recently developed micromethod involving GLC-MS (Rose et al.,
1983 b). The mass spectrum obtained from the VSV N protein showed the blocked tripeptide N ~acetyl-Ser-Val-Thr (Rose et al., 1983a). In the VSV N gene sequence reported by Gallione et al.
(1981), the tripeptide Ser-Val-Thr appears only immediately following the initiating
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41
773
VSV and Sendai virus nuc&ocapsid proteins
.3o
.~s
.~
.~
.~
.?~
~CC~G£C~G~G~A~RCGTGTATGG~RTRTRT~RTGR~GTTRG~CRGGRTTTTRGGGTCA~AGT~TCC~CECTS£G~
I - - - NP
vvvvvx
.105
i~O
~GC~GGTTCCAG~CCCTTTGCTTTGETGCCR~AGTTCACGRTG
GCC
M~t Rio
.gO
TTG
TTG
RGC
RCC
TTC
G~T
~CR
Gly
Leu
LeU
S~P
Thn
Phe
R~p
Th~
180
.150
TTT
.135
GGG
~GC
TCT
AG~
~GG
~GC
GA~
~GT
ATT
~T
~AG
TCG
GGA
GG~
GGT
GCT
GTT
~TC
CCC
GGC
2tQ
C~G ~G
Gin
Ly=
~GC
SSP
~C~
ThP
GTC
VOl
TC~
S~P
GT6
VeI
TTC
Phe
GT£
VoJ
CT£
Leu
~40
GGC C C £
Giy PPo
~GT
Sem
GTG
Vo]
~CT
Thr
G£T
~Ep
G~T
£sp
5CR
Rio
GRC
£sp
£~G
ky$
270
TTA TTC
~TT
GCA
£CT
~CC
TTC
CTA
GCT
CAC
3OO
TCA TTG
Set
Leu
ARC
RSm
ACA
T~
GRT
R$~
RRG
Ly~
CRG
Glm
CAC
H15
TCT
S~
CRG
GI~
~G~
GGA
GGG
TTC
CTC
GTC
TCT
CTG
CTT
360
RTG GCT
TRC
~GT
RGT
CCR
GA~
TTG
TRC
TTG
R~
GIy
GIy
P~
L~
VO2
~P
L~u
L~U
GCC
RIO
M~
TyP
S~
S~
P~O
GI~
L~u
Ty~
L~U
390
RCA A C A
A~C
GG~
GTA
RAC
GCC
GRT
GTC
£~£
.~20
T ~ T GTG
~TC
T~C
AAC
RT~
G~G
~
G~C
CCT
Th~
R$n
GIF
VOI
R~
RIo
Rsp
Vol
Ly~
Ty~
II~
Tyr
R~
II~
Glu
Ly~
R~p
PPO
T~p
RIo
Vol
480
45O
FIAG AGG ACG AFIG ACT GFIC AGFI TTC ATT GTG AAG ACG FIGFI GFIT ~TG
GAFf TAT GAG
I::IGG ACC
Lys
G1u
~r~
~Pg
Thr" L y s
ThP
Rsp
~rg
Pl~
Ile
Vol
Lys
ThP
~P9
Rsp
Me~
Ty~
G1u
Th~
510
ACFI GR~q TGG
CTG
TTT
GGA
CCT
ATG
GTC
AAC
.540
Iql~G AGC
CCIq
CTC
TTC
CAG
GGT
Cf=IR CGG
GRT
Ti~
LeU
Phe
G1y
Pr'o M ~ t
Val
Rsn
Ly~
Pr'o L e u
Ph~
Gln
GIy
G1m
~sp
G1u
T~p
GCT
GCIq
GRC
CCT
GFIC tqCI=l CTC
CTT
CAR
IqCC
.600
T A T GGG
TtqT
CCT
GCFt
TGC
CTR
GGR
GCT
R2o
RIo
~sp
P~O
Rsp
L~u
G1m
ThP
Ty~
TyP
P~O
R1O
Cy~
L~u
G1y
RIO
570
Tt~P L e u
S~
GIy
~g
Fig. 3. Nucleotidesequenceandpredictedaminoacidsequenceofthefirst thirdoftheSendaivirus NP
gene. The sequence of 626 nucleotides from the 3' end of the Sendai virus genome, shown as the
complementary (mRNA sense) DNA strand, was obtained as described in Methods. The wavy
undedine indicates the central TATA-like region of the leader gene.
methionine, consistent with the identification of the corresponding m R N A sequence as the
major ribosome-binding site (Rose, 1978). Therefore, the blocked tripeptide sequence was
formed in vivo by cleavage of the initiating methionine and N~-acetylation of the following
amino acid. The mass spectrum obtained Sendal virus NP protein still containing R N A clearly
showed the blocked tetrapeptide N~-acetyl-Ala-Gly-Ilx-Ilx (Rose et aL, 1983 a; Ilx stands for
either Leu or Ileu which are not differentiated by this technique). Inspection of the first onethird of the Sendai virus NP gene sequence shown in Fig. 3 shows that this blocked tetrapeptide
corresponds only to the tetrapeptide Ala-Gly-Leu-Leu which also immediately follows the
initiating methionine. Hence, within the limits of our sequence information, the Sendai virus
protein is processed in vivo similarly to the VSV N protein.
Analysis o f the Sendai virus leader and N P gene region
In VSV transcripti0n~ the first and most abundant product is the 47 bp 'leader' RNA,
followed by the capped mi~NAs all of which start with the sequence 5'-AACAG...-3' (Banerjee
et al., 1977). Comparison of the leader and N m R N A sequences with the ( - ) VSV genome also
revealed the presence of short genomic intervening sequences whose complements are not found
in either the leader or the N m R N A (Keene et al., 1980). The situation with Sendai virus is less
clear. N o such consensus m R N A 5' sequence has been reported, nor has the 5' end of the NP
m R N A been directly sequenced.
In order to determine the exact 5' end of the NP m R N A and to predict the primary
translation product of the Sendal virus nucleocapsid protein gene, we sequenced 600 bp of
cloned D N A representing the first one-third of the gene, and extended the sequence to the 3' end
of the ( - ) genome by direct R N A sequencing. The complement of the genome sequence [as
m R N A sense ( + ) D N A ] and the predicted partial protein sequence are presented in Fig. 3.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41
B. M. BLUMBERG AND OTHERS
774
(a)
1
2
3
4
5
6
7
8
396
344
298
220/1
154
Fig. 4. Location of the 5' end of the Sendal virus NP mRNA by primer extension studies. (a) Primer
extension using the RsaI 149 bp restriction fragment which had been 5' end-labelled and annealed with
the indicated template RNAs under formamide (lanes 2 to 4) or aqueous (lanes 5 to 8) conditions, and
primer-extended with reverse transcriptase. The products were electrophoresed on a 42 x 25 x
0.05 cm, 8 ~ polyacrylamide gel (38/2: acrylamide/bisacrylamide) containing 7 ~,l-urea. The gel was
autoradiographed at - 7 0 °C using an intensifying screen. Lanes 2 and 5 : 7 - 6 ~tg of uninfected
LLCMK2 cell CsCI pellet RNA. Lanes 3, 4 and 6:5.6 ixg, 11.2 ~tg and 11-2 lag respectively of Sendal
virus-infected LLCMK2 cell CsC1 pellet RNA. Lane 7: 0.5 ~tg of RNA from Sendal virion polymerase
reaction in vitro. Lane 8 : 4 lag of Sendal virus 50S genomic RNA. Marker lane 1: HinfI digest of
pBR322. (b) Primer extension using the HaeIII 86 bp restriction fragment which had been 5' endlabelled and annealed with the indicated RNAs under aqueous conditions, primer-extended and then
electrophoresed as above. Lanes 2 and 3:6-1 pg and 8.9 lag respectively of CsC1 pellet RNA from
uninfected and Sendal virus-infected LLCMK2 cells. Lanes 4 and 5 : 10.6 lag and 12.3 gtg respectively of
CsC1 pellet RNA from uninfected and Sendal virus-infected BHK cells. Lane 6:0.5 ~tg of RNA from
Sendal virion polymerase reaction in vitro. Lane 7 : 4 Ixg of Sendal virus 50S genomic RNA. Marker
lanes l and 9: Hinfl and MspI digests of pBR322. Lane 8 : primer alone.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41
QC
t~
QC
t~
00
C~
t~
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41
776
B. M. B L U M B E R G
AND
OTHERS
There is only one long open reading frame (ORF) which begins with the A T G at residue 120; the
other two reading frames contain multiple stop codons. Note that the 5' proximal A U G of the
NP m R N A is in the most favoured context for eukaryotic ribosomes, ACGA UGG (Kozak,
1983). While this work was in progress, Shioda et aL (1983) reported the sequence of the first
3687 nucleotides from the 3' end of the Sendai virus Z strain genome, including the complete NP
gene. The corresponding region of their sequence differs from that in Fig. 3 by four nucleotides
in the leader region, and nine in the NP gene that lead to seven amino acid changes. In their
sequence also, the tetrapeptide Ala-Gly-Leu-Leu occurs only immediately following the
initiating methionine.
The position of the 5' end of the Sendai virus NP m R N A on the viral ( - ) genome was
determined by primer extension of end-labelled restriction fragments isolated from clones of the
NP gene using CsCI pellet RNA from Sendai virus-infected cells as a source of viral m R N A .
The data using two of these fragments, the RsaI 149'mer (position 236 to 384 from the 3' end of
the genome) and the HaelII 86'mer (position 123 to 208) are shown in Fig. 4. Lanes 2 to 4, and 5
and 6 respectively of Fig. 4(a) show the results of extending the RsaI 149'met on both uninfected
and Sendai virus-infected LLCMK2 cell CsC1 pellet R N A under different annealing conditions.
In both cases, a single strong band, of approximately 325 bp, resulted only when Sendai virusinfected cell RNA was used, thus placing the end of the NP m R N A about 59 nucleotides from
the 3' end of the genome. Other smaller bands, which are present when both uninfected and
infected cell RNA was used and which vary in intensity from lane to lane, probably represent
primer self-extension since they are not present in the end-labelled primer preparation (not
shown).
In order to determine more accurately the 5' end of the NP mRNA, we next used the HaeIII
86'mer since this fragment is closer to the m R N A start. Lanes 2 and 3, and 4 and 5 respectively
of Fig. 4 (b) show the results of the extension of this primer on both uninfected and Sendai virusinfected LLCMK2 and BHK cell CsC1 pellet RNA. Again, a single band, which in this case comigrated with the 154 bp HinfI marker, was present only when infected cell R N A was used,
placing the end of the N P m R N A about 54 nucleotides from the 3' end of the genome. Fig. 4 (b)
also demonstrates that Sendai virus-infected BHK cells contain approximately three times more
NP m R N A than LLCMK2 cells, a result consistent with the enhanced ability of the BHK cell
CsC1 pellet RNA to direct the translation of the NP protein in vitro (our unpublished data).
Taken together, the results so far place the 5' end of the Sendai virus NP m R N A at
approximately position 54 to 59 from the 3' end of the ( - ) genome. Examination of the D N A
sequence in this region shows the sequence 5 ' - A G G G T C A A A G starting at position 56. The
Sendai P/C m R N A has been shown by dideoxynucleotide sequencing to start with the similar
sequence 5 ' - A G G G T G A A A G , and this decanucleotide was also found at the beginning of the
M gene (Giorgi et al., 1983 b; Dowling et al., 1983). In addition, we have determined that the vast
majority of Sendai virus m R N A s made in a virion polymerase reaction in vitro begin
7MeGpppA (our unpublished results). It therefore seems probable that the Sendai virus NP
m R N A starts on the A residue at position 56 from the 3' end of the ( - ) genome.
Fig. 4(a) (lane 7) and (b) (lane 6) also show the results of primer extension of the RsaI 149'mer
and the HaeIII 86'mer on an estimated 0.5 ~tg of the total R N A products of an in vitro Sendai
virion polymerase reaction. Note that NP m R N A made in vitro appears to start at the same
position as cytoplasmic NP mRNA, but that in each case a second larger band containing
approximately 55 additional nucleotides is found, suggestive of a read-through transcript of
leader and NP mRNA. This suggestion was confirmed by primer extension on intracellular
nucleocapsid 50S RNA, a mixture of ( - ) genomes and ( + ) antigenomes. The signals are weak
due to the presence of excess ( - ) genome RNA, but note that the putative read-through band,
which appears as a doublet in lane 7 of Fig. 4(a), is also found as an identical doublet in lane 8
(arrow). The band in lane 7 of Fig. 4(b) is very weak on this exposure, but note that the readthrough band in lane 6 is consistent with a length of 208 bp relative to the markers, i.e. it extends
to the precise 3' end of the ( - ) genome. Interestingly, the read-through transcripts, which
represent a sizeable fraction of the RNAs synthesized in vitro, cannot be detected among the
intracellular RNAs at the exposures shown. Such read-through bands can sometimes be seen
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41
V S V and Sendai virus nucleocapsid proteins
777
upon overexposure of the gels, but appear to represent less than 1% of the NP mRNA made in
vivo.
DISCUSSION
We have utilized a combination of techniques for the preparation and direct amino terminus
analysis of the nucleocapsid proteins of VSV and of Sendai virus. In conjunction with the
predicted amino-terminal structures of the N and NP proteins derived from the DNA
sequences, these techniques have enabled us to determine that the viral nucleocapsid proteins
undergo cytoplasmic processing in the course of their synthesis and assembly into nucleocapsids.
Minimally, this processing involves the removal of the initiating methionine followed by N ~acetylation of the following amino acid. It should be noted that two-dimensional gels of the VSV
nucleocapsid proteins show multiple spots, while the Sendai virus NP protein is phosphorylated
(Roux & Kolakofsky, 1974; Lamb & Choppin, 1977; Raghow et al., 1978; Perrault et al., 1983).
It remains unclear whether the amino-terminal modifications we have described are present in
all these different forms of the N and NP proteins. In addition, further work will clearly be
required to determine whether all or only some of these different forms of N and NP are
involved in the initiation, as opposed to elongation, of nucleocapsid assembly.
Besides the fact that the VSV and Sendai virus nucleocapsid proteins are processed at their
amino termini and exist in multiple forms, it should also be noted that the biological activity
imputed to N protein (Blumberg & Kolakofsky, 1983) may not be due to this protein acting by
itself. Pure N protein readily self-assembles, and protein expressed in vivo in cells transformed
with a DNA clone of the VSV N gene was found to be highly aggregated and non-functional
(Sprague et al., 1983). Recent findings of R. Peluso & S. Moyer (personal communication) have
indicated that cytoplasmic N protein exists as a complex with the VSV NS protein, which may
be required to keep the free N protein in an active soluble form.
The work presented here has also localized the 5' end of the Sendai virus NP mRNA at
position 56 from the 3' end of the ( - ) genome. Previous work has demonstrated that the major
Sendai virus ( + ) leader RNA, made both in vivo and in vitro, was 55 nucleotides long (Leppert et
al., 1979). In that study, the length of the leader RNA was determined as dsRNA, by annealing
leader RNA to a 3' end-labelled ( - ) genome RNA probe which was then digested with RNase
A. Since the AAAA sequence at positions 52 to 55 of the ( - ) genome would not have been
cleaved by this nuclease, the Sendai virus leader RNAs could have terminated anywhere within
this four-base region. Thus, like VSV, the Sendai virus leader RNA terminates only a few
nucleotides before the start of the nucleocapsid protein gene.
Recently, Keene and his coworkers have pointed out that the VSV ( + ) leader RNA appears
to be separated into three functional domains (Kurilla et al., 1982). The highly conserved 5'proximal third of the leader RNA is thought to play a role in the initiation of RNA synthesis and
in the nucleation of N protein assembly. The middle third contains a TATA-like sequence (Rose
& Iverson, 1979), and is thought to interact with virion polymerase components (Keene et aL,
1981). The 3' proximal third may be involved in leader RNA termination, and it is this region
that has been postulated to interact with cellular La proteins, which may represent a possible
host factor in viral RNA synthesis (Wilusz et al., 1983; Kurilla & Keene, 1983). Interestingly,
the Sendai virus ( + ) leader RNA also fits the above description (see Fig. 3). The interaction of
VSV leader RNAs with La protein in vitro has also focused attention on the similarities between
VSV leader RNA and cellular and viral RNA polymerase III (pol III) transcripts (Weck &
Wagner, 1979; McGowan et al., 1982). Pol III transcripts are characterized by their relatively
short length, lack of capping at their 5' end or polyadenylation at their 3' end, and by the finding
that the primary pol III transcripts end with a series of U residues (Rinke & Steitz, 1982). The
Sendai virus ( + ) leader RNA also fits all of these criteria. It will therefore be of interest to
determine whether the Sendai virus leader RNA also interacts with La protein in vivo and, if so,
whether the leader-La complex plays a role with the N and NP proteins in the assembly of their
respective nucleocapsids.
We thank Colette Pasquier and Rosette Bandelier for growing the cells and viruses. We are greatly indebted to
Hans P. Kocher, J.-C. Jaton and Robin Offord of the D6partement de Biochimie M~dicale, Universit~ de Gen~ve,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41
778
B. M. B L U M B E R G A N D O T H E R S
for their expertise and enthusiasm in analysing the blocked amino termini of the N and N P proteins, and to Marco
G. Simona for expert technical assistance in the preparation of chemical derivatives. This work was supported by
grants from the Fonds National Suisse.
REFERENCES
BANERJEE, A. K., ABRAHAM,G. & COLONNO,R. J. (1977). Vesicular stomatitis virus: mode of transcription. Journalof
General Virology 34, 1-8.
BLUMBERG, B. M. & KOLAKOFSKY,D. (1983). An analytical review of defective infections of vesicular stomatitis
virus. Journal of General Virology 64, 1839-1847.
BLUMBERG, B. M., GIORGI, C. & KOLAKOFSKY, D. (1983). N protein of vesicular stomatitis virus selectively
encapsidates leader R N A in vitro. Cell 32, 559-567.
DOWLING, P. C., GIORGI, C., ROUX, L., DETHLEFSEN, L. A., GALANTOWlCZ,M. E., BLUMBERG,B. M. & KOLAKOFSKY,D.
(1983). Molecular cloning of the Y-proximal third of Sendal virus genome. Proceedings of the National
Academy of Sciences, U.S.A. 80, 5213-5216.
EMERSON, S. U. & YU, Y.-I~. (1975). Both NS and L proteins are required t'or in vitro R N A synthesis by vesicular
stomatitis virus. Journal of Virology 15, 1348-1356.
GALLIONE,C. J., GREENE, J. R., IVERSON,L. E. & ROSE,.L K. (1981). Nucleotide sequences of the m R N A s encoding the
vesicular stomatitis virus N and NS proteins. Journal of Virology 39, 529-535.
GIORGI, e., BLUMBERG,E. M. & KOL~a~GrSKV,D. (1983 a). Sequence determination of the ( + ) leader R N A regions of
the vesicular stomatitis virus Chandipura, Cocal and Piry serotype genomes. Journalof Virology46, 125-130.
GIORGI, C., BLUMBERG,B. M. & KOLAKOFSKY,D. (1983b). Sendai virus contains overlapping genes expressed from a
single m R N A . Cell 35, 829-836.
GRAY, W. R. & DEL VALLE, U. E. (1970). Application of mass spectrometry to protein chemistry. I. Method for
amino-terminal sequence analysis of proteins. Biochemistry 9, 2134-2137.
HEGGENESS, M. H., SCHEID, A. & CHOPPIN, P. W. (1981). The relationship of conformational changes in the Sendai
virus nucleocapsid to proteolytic cleavage of the N P polypeptide. Virology 114, 555-562.
KEENE, J. D., SHUBERT, M. & LAZZARINI, R. A. (1980). Intervening sequence between the leader region and the
nucleocapsid gene of vesicular stomatitis virus R N A . Journal of Virology 33, 789-794.
KEENE, J. D., THORNTON,J. & EMERSON, S. U. (1981). Sequence specific contacts between the R N A polymerase of
VSV and the leader gene. Proceedings of the National Academy of Sciences, U.S.A. 78, 6192-6195.
KINGSBURY, D. W. (1974). The molecular biology of paramyxoviruses. Medical Microbiology and Immunology 160,
73-83.
KOZAK, M. (1983). Comparison of initiation of protein synthesis in procaryotes, eucaryotes, and organelles.
Microbiological Reviews 47, 1-45.
KURILLA, M. G. & KEENE, J. D. (1983). The leader R N A of vesicular stomatitis virus is bound by a cellular protein
reactive with anti-La lupus antibodies. Cell 34, 837-845.
KURILLA, M. G., PIWNICA-WORMS,H. & KEENE, J. D. (1982). Rapid and transient localization of the leader R N A of
VSV in the nuclei of infected cells. Proceedings of the National Academy of Sciences, U.S.A. 79, 5240-5244.
LAMB,R. A. & CHOPPIN, P. W. (1977). The synthesis of Sendai virus polypeptides in infected cells. Virology81, 382
397.
LAMB,R. A. & MA~Y, B. W. J. (1975). The polypeptides and R N A of Sendai virus. In Negative Strand Viruses, pp. 6587. Edited by B. W. J. Mahy & R. D Barry. New York & London: Academic Press.
LEPPERT, M., RITTENHOUSE,L., PERRAULT,J., SUMMERS,D. F. & KOLAKOFSKY,D. (1979). Plus and minus strand leader
R N A s in negative strand virus-infected cells. Cell 18, 735 747.
McGOWAN, J. J., EMERSON, S. U. & WAGNER, R. R. (1982). The plus-strand leader R N A of VSV inhibits D N A dependent transcription of adenovirus and SV40 genes in a soluble whole cell extract. Cell 28, 325-333.
MARX, P. A., PORTNER, A. & KINGSBURY, D. W. (1974). Sendai virion transcriptase complex: polypeptide
composition and inhibition by virion envelope proteins. Journal of Virology 13, 107-112.
MAXAM,A. M. & GILBERT, W. (1980). Sequencing end-labeled D N A with base-specific chemical cleavages. Methods
in Enzymology 65, 499-560.
MOUNTCASTLE,W. E., COMPANS,R. W., CALIGUIRI, L. A. & CHOPPIN, P. W. (1970). Nucleocapsid protein subunits of
simian virus 5, Newcastle disease virus, and Sendal virus. Journal of Virology 6, 6 7 7 ~ 8 4 .
MOUNTCASTLE,W. E., COMPANS, R. W. & CHOPPIN, P. W. (1971). Proteins and glycoproteins of paramyxovirus: a
comparison of simian virus 5, Newcastle disease virus, and Sendai virus. Journal of Virology 7, 47-52.
PERRAULT, J., CLINTON, G. M. & McCLURE, M. A. (1983). R N P template of vesicular stomatitis virus regulates
transcription and replication functions. Cell 35, 175-185.
RAGHOW,R., PORTNER, A., HSU, C. H., CLARK,S. B. & K1NGSBURY,D. W. (1978). Charge heterogeneity in polypeptides
of negative strand R N A viruses. Virology 90, 214-225.
RINKE, I. & STEITZ, J. A. (1982). Precursor molecules of both h u m a n 5S ribosomal R N A and transfer R N A s are
bound by a cellular protein reactive with anti-La lupus antibodies. Cell 29, 149-159.
ROSE, J. K. (1978). Complete sequences of the ribosome recognition sites in vesicular stomatitis virus m R N A s :
recognition by the 40S and 80S complexes. Cell 14, 345-353.
ROSE, J. & IVERSON, L. (1979). Nucleotide sequence from the 3' end of vesicular stomatitis virus m R N A s as
determined from cloned D N A . Journal of Virology 32, 404-411.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41
V S V and Sendai virus nucleocapsid proteins
779
ROSE, K., KOCHER, H. P., BLUMBERG, B. M. & KOLAKOFSKY,D. (1983a). A n improved procedure, involving mass
spectrometry, for N'-terminal amino acid sequence determination of proteins which are N'-blocked.
Biochemical Journal 217, 253-257.
ROSE, K., SIMONA,~l. G. & OFFORD, R. ~. (1983 b). A m i n o acid sequence determination by g.l.c.-mass spectrometry of
permethylated peptides. Biochemical Journal 215, 261-272.
Roux, I.. & KOLAKOFSKY,13. (1974). Protein kinase associated with Sendal virions. Journal of Virology 13, 545-547.
SHIODA, T., HIDAKA, Y., KANDA, T., SHIBUTA,H., NOMOTO,A. & IWASAKI,K. (1983). Sequence of 3,687 nucleotides
from the 3' end of Sendai virus genome R N A and the predicted a m i n o acid sequences of viral NP, P and C
proteins. Nucleic Acids Research 11, 7317-7330.
SPRAGUE, J., CONDRA, J. H., ARNHEITER, H. & LAZZARINI, R. A. (1983). Expression of a recombinant D N A gene
coding for the vesicular stomatitis virus nucleocapsid protein, Journal of Virology 45, 773-781.
STONE, H. O., KINGSBURY, D. W. & DARLINGTON, R. W. (1972). Sendal virus-induced transcriptase from infected
cells: polypeptides in the transcriptase complex. Journal of Virology 10, 1037-1043.
SZILAGYI, J. F. & URYVAEV,L. (1973). Isolation of an infectious ribonucleoprotein from vesicular stomatitis virus
containing an active R N A transcriptase. Journal of Virology 11, 279-286.
WAGNER, R. R., SCHNAITMAN,T. C., SNYDER,R. M. & SCHNAITMAN,C. A. (1969). Protein composition of the structural
components of vesicular stomatitis virus. Journal of Virology 3, 611-618.
WECK, P. K. & WAGNER, R. R. (1979). VSV infection reduces the number of active D N A - d e p e n d e n t R N A
polymerases in myeloma cells. Journal of Biological Chemistry 254, 5430-5434.
WILUSZ, J., KURmLA, M. G. & KEENE,J. D. (1983). A host protein (La) binds to a unique species of minus-sense leader
R N A during replication of vesicular stomatitis virus. Proceedingsof the National Academy of Sciences, U.S.A.
80, 5827-5831.
WOLD, F. (1981). In vivo chemical modification of proteins. Annual Review of Biochemistry 50, 783-814.
(Received 16 November 1983)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:19:41