J. gen. Virol. (1989), 70, 2409-2419.
2409
Printed in Great Britain
Key words: CD V/nucleocapsid
Isolation and Characterization of Canine Distemper Virus Nucleocapsid
Variants
By M. O G L E S B E E , 1. L. T A T A L I C K , 1 J. R I C E 2
AND
S. K R A K O W K A
1
The Department of Veterinary Pathobiology, College of Veterinary Medicine, 1925 Coffey Road,
Columbus, Ohio 43210 and 2The Cellular and Molecular Biology Section, Battelle Columbus
Laboratories, 505 King Avenue, Columbus, Ohio 43201, U.S.A.
(Accepted II May 1989)
SUMMARY
Nucleocapsid (NC) variants expressed by the Onderstepoort strain of canine
distemper virus (CDV) were ultrastructurally and biochemically characterized. Three
isolated variants were defined which corresponded to the three variants observed
within the cytoplasm of infected cells. Dense NC (D-NC), isolated on discontinuous
caesium chloride (CsCI) isopycnic gradients, had an average density of
1.2971 + 0.0042 g/ml. Ultrastructurally, D-NC were 1620.0 + 112.1 nm in length with
a 20.1 _+ 1-3 nm outer and a 5.8 + 0.7 nm inner core diameter. The D-NC protein
composition was 89-7 ~ of a 61K protein (N), 8-4 ~ of a 75K protein (P) and 1-9 ~o of a
160K to 200K protein (L). A single species of nucleic acid, 15 kb in length, was isolated
from D-NC. Light NC (L-NC), similarly isolated, had an average density of
1-2894+0-0040g/ml. L-NC differed ultrastructurally from D-NC in that poor
resolution of NC subunits, a larger outer diameter (32.0 + 2.8 nm), and a greater inner
core diameter (10.4 + 0.6 nm) were observed. The average L-NC strand length was
1574.4 + 115.8 nm. The protein composition was the same as D-NC with the exception
of an additional 70K protein, representing 4.0 to 7.7 ~ of the total L-NC protein mass.
A 15 kb nucleic acid was also identified in L-NC, although heightened sensitivity of
encapsidated L-NC nucleic acid to non-specific nuclease degradation was observed.
The ratio of D-NC to L-NC isolated from individual virus preparations varied and was
independent of viral infectivity. A third N C variant, defective-NC (Df-NC), was also
identified. This had the lowest density on CsC1 gradients (1.2460 + 0.0046 g/ml). The
Df-NC structures were truncated to a uniform length of 87.0 + 5.8 nm. Diameter
measurements were between those of D-NC and L-NC, being 24.4 + 1-4 nm (outer)
and 6.9 + 0.4 nm (inner core). Like L-NC, the 70K protein was present but in greater
amounts, representing as much as 43-7~ of the total Df-NC protein mass. RNase
A-sensitive nucleic acid was isolated from Df-NC which ranged in size from 1.16 to
0-67 kb with a majority of the material being 0.86 kb in length. For both L-NC and
Df-NC, canine CDV convalescent serum reacted with viral N and P proteins in
Western blot analyses but not with the 70K protein, suggesting a host cellular origin for
the latter.
INTRODUCTION
The nucleocapsid (NC) of non-segmented negative-strand R N A viruses represents not only
the repository of viral genetic information, but also the self-contained means for regulated
expression of that information (Seifried et al., 1978). This is exemplified by measles virus (MV),
a paramyxovirus within the morbillivirus genus, in which infectivity of isolated NC structures
has been demonstrated (Rozenblatt et al., 1979). For this family of viruses, the major NC
structural element is the N protein (Robbins & BusseU, 1979; Lamb et al., 1976; Bussell et al.,
1974). By analogy to rhabdoviruses, N functions in the regulation of genome transcription
(Kolakofsky & Blumberg, 1982), as well as imparting nuclease resistance to the encapsidated
0000-8916 © 1989 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34
2410
M. OGLESBEE AND OTHERS
genome (Kingsbury & Darlington, 1968). The phosphorylated P protein is the second most
abundant core protein species (Lamb et al., 1976; Cattaneo et al., 1987) and is more randomly
distributed along the length of the NC than the N protein (Portner & Murti, 1986). This protein
plays a direct role in the viral transcriptional complex (Deshpande & Portner, 1985; Chinchar &
Portner, 1981), along with an additional structural element, the L protein (Pringle, 1987;
Schubert et al., 1984). The latter is the largest (160K to 200K) of the paramyxovirus proteins
(Rima, 1983) and is produced in the lowest amounts (Cattaneo et al., 1987). For canine
distemper virus (CDV), a morbillivirus closely related to MV, the protein composition by mass
of isolated nucleocapsids is 89.79/o N, 8-49/oP and 1.9% L protein (Oglesbee et al., 1989) with the
RNA genome representing 5% of the total NC mass (Waters & Bussell, 1974).
Numerous studies have focused on ultrastructural characteristics of paramyxovirus NCs and
the functional significance of structural variants (Robbins et al., 1980; Matsumoto, 1966;
Oyanagi et al., 1971 ; Dubois-Dalcq et al., 1974). The smooth NC variant is a flexiform linear
structure, approximately 1-1 ~tm in length with an 18 nm outer diameter (Finch & Gibbs, 1970;
Kingsbury, 1972; Bussell et al., 1974). Well delineated caps id subunits consisting of N protein
monomers are arranged in a single-start helical configuration, angled relative to the long axis
(Finch, 1970), imparting a herringbone appearance to the NC strands. A second granular NC
form has been described for MV and CDV (Koestner & Long, 1970; Robbins et al., 1980;
Narang, 1981). The granular coat increases the NC diameter to 25 to 30 nm and obscures the
definition of capsid subunits (Matsumoto, 1966; Oyanagi et al., 1971; Dubois-Dalcq et al.,
1974). For MV, it has been suggested that granular NC is the biologically active form involved
in productive infection (Robbins et al., 1980). The granular form predominates at the outer cell
membrane where budding virus structures are observed. The smooth variant, on the other hand,
has an intranuclear and perinuclear distribution (Robbins et al., 1980; Chui et al., 1986) and is
associated with chronic infection and virus persistence (Minagawa et al., 1976; Fournier et al.,
1981; Robbins, 1983; Uchiyama et al., 1985). Thus, it has been suggested that smooth NC is
either biologically inert or mediates the suppressed transcriptional levels characteristic of
persistence (Robbins et aL, 1980).
The nature of the substance responsible for imparting the granular nature to biologically
active NC is unknown. Investigators working with MV have suggested that the viral M protein
may be responsible for the granularity, on the basis of a correlation between the presence of
smooth NC and defective M protein synthesis in persistence (Oyanagi et al., 1970; DuboisDalcq et al., 1974). However, M protein has never been specifically copurified with granular NC
and it has been shown that the association of M protein with paramyxovirus NC is a transitory
and late event (Portner & Kingsbury, 1976). Other investigators identified a 52K and 45K
protein which copurified with granular NC (Robbins et al., 1980). Although a host origin was
suggested, Western blot analysis to disprove a viral origin was not included. Furthermore, the
size of these proteins corresponds to major N protein degradation products (Croce et al., 1980).
Cellular actin (Mr 43K) has been associated with MV NC budding (Bohn et al., 1986) and has
been isolated from intact MV virions (Tyrrell & Norrby, 1978). However, it is not a good
candidate for the element imparting the granular coat because the association between actin and
budding NC is mediated by binding to the viral M protein (Giuffre et al., 1982). As previously
indicated, M protein association with NC is a transitory and late event.
The present work describes the isolation of smooth and granular CDV NC, in addition to a
defective NC variant (Df-NC), from lytically infected Vero cells using a recently perfected
isolation procedure for NC (Oglesbee et al., 1989). The technique utilizes protease inhibitors to
maintain the integrity of isolated NC and to prevent the formation of core protein degradation
products which confuse the analysis of NC protein constitution. This has resulted in the
identification of a unique 70K non-viral protein associated with granular and defective NC.
METHODS
Infection of cells. For NC isolation, log-phase Veto cells were infected in suspension at an m.o.i, of 0.01 with
virus stocks derived from plaque-purified Onderstepoort CDV (Ond-CDV). Infected cells, seeded into tissue
cultureflasks,weremaintained at 37 °C in Dulbecco'sminimumessentialmediumcontaining 10%(v/v)foetalcalf
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34
C D V nucleocapsid variants
2411
serum, until a pronounced cytopathic effect was seen (approx. 22-5 h). An identical number of uninfected control
Vero cells were processed in the manner outlined for the infected cells. At the time of harvest, the medium was
made 1.0 mM PMSF and 0.23 trypsin inhibitor units/ml (TIU/ml) Aprotinin (protease inhibitor from bovine
lung; Sigma). Cytoplasmic NC was released by exposing infected celts, released from the flasks by scraping, to two
successive freeze-thaw cycles. Following low-speed centrifugation, samples were collected for virus titration on
log-phase Vero cells.
For ultrastructural evaluation of NC in situ, Vero cells were infected as above and maintained in Transwell
(Costar) cell culture chamber inserts under the conditions described above. Inserts were processed for
ultrastructural examination of cells 22 h after infection.
Nucleocapsid isolation. Clarified medium from infected and uninfected cells was made 2.0~ Triton X-100 and
an additional clarification by centrifugation at 10000 g (20 min at 4 °C) was performed. NC was recovered by
centrifugation at 100000 g. Recovery was also achieved by ultrafiltration followed by pelleting NC, concentrated
in the retentate, against a 4 0 ~ (w/w) CsC1 cushion at lO0000g. The latter approach employed ultrafiltration
membranes with a 300K exclusion limit. Crude NC preparations were then incubated in a small volume of
gradient buffer (25 mM-Tris-HC1, 50mM-NaC1, 2mr,i-EDTA disodium salt, 0.2~ w/v sodium dodecyl
sarcosinate) for 12 h at 4 °C, before isolation of purified NC on discontinuous CsC1 gradients. These steps
consisted of 5% (w/w) sucrose, 2 5 ~ (w/w) CsCI, 3 0 ~ (w/w) CsCI, and 40% (w/w) CsCI in gradient buffer.
Isopycnic banding of N C was achieved by centrifugation at 170 000 g at 4 °C for 2 h. Samples were harvested by
needle aspiration and refractive indices were measured at 25 °C. Gradients of uninfected cellular material were
similarly processed. Samples in the latter represented bands visualized in the 25 or 3 0 ~ CsCI step, or regions
corresponding to variant NC band positions. Sample density was computed from refractive indices using the
relationship described by Bruner & Vinograd (1965).
Ultrastructural evaluation of nucleoeapsid. Fractions of gradient samples were dialysed extensively against
distilled deionized water using dialysis filters (MiUipore) with a 0.05 ~tm pore size. Samples were then negatively
stained with phosphotungstic acid, spotted onto 300-mesh Formvar-coated copper grids (Ladd Research
Industries) and examined using a Philips 300 transmission electron microscope.
Tissue culture chamber inserts were washed in phosphate-buffered saline (PBS) and cells were fixed in 3 ~
glutaraldehyde, buffered with 0.l M-cacodylate pH 7.4, for 1.0 h at 4 °C. Post-fixation for an additional 1.0 h at
4 °C was performed using 3 5 ~ osmium tetroxide in S-cotlidine buffer (2,4,6-trimethylpyridine; Kodak) pH 7-4.
Following dehydration in a graded series of alcohols, insert strips were embedded in Medcast resin (Ted Pella) and
stained with uranium acetate and lead citrate. Selected areas were mounted on copper grids for examination with a
Philips 300 transmission electron microscope. NC outer diameters were established for multiple cytoplasmic foci
of NC replication by direct measurements taken from photographic images. Each recorded diameter represented
an average of three measurements per NC segment. Histograms of the frequency of specific outer diameters were
then constructed for each focus.
Protein and RNA analysis. Nucleocapsid fractions were desalted by dilution with TNE buffer (0.01 M-Tris-HC1
pH 8.0, 0.01 M-NaCI, 0-01 M-EDTA disodium salt) and pelleting at 100000g. Uninfected cellular material was
desalted by extensive dialysis against 100 mM-ammonium bicarbonate and concentrated by lyophilization. Pellets
in both instances were resuspended by heating to 37 °C in a small volume of TNE buffer containing 2 ~ SDS. For
protein analyses, samples were resolved electrophoretically through 12~ polyacrylamide gels under reducing
conditions according to the method of Laemmli (t 970). Quantification of individual proteins was performed by
laser densitometric evaluation of the Coomassie Brilliant Blue staining intensity of NC proteins relative to that of
known dilutions of bovine serum albumin (Oglesbee et al., 1989). Silver staining of gels was subsequently
performed by the method of Merril et al. (1981) for the detection of minor protein species.
Virus specificity of protein bands was determined by Western blot analysis of core proteins, using CDV
convalescent gnotobiotic canine serum as the source of primary antibody. Samples were resolved electrophoretically as above and electroblotted onto nitrocellulose. The nitrocellulose strips were blocked by incubation with
20% porcine serum in PBS (pH 7.4) for 12 h at 25 °C, and then incubated with primary antibody diluted 1:50 in
PBS, 1 ~ bovine serum albumin, 1 ~ Tween-20, for 1.5 h at 25 °C. Bound primary antibody was detected with a
rabbit anti-canine IgG-alkaline phosphatase conjugate (Miles Laboratories) and a 5-bromo-4-chloro-3-indolylphosphate (BCIP)/nitroblue tetrazolium (NBT) substrate detection system (Kirkegaard & Perry Laboratories).
RNA was extracted from desalted gradient fractions using three successive phenol : chloroform : isoamyl alcohol
(25:24:1) extractions, followed by an additional extraction with chloroform:isoamyl alcohol (24:1).
Electrophoretic analysis was performed using 1.0 and 1-5~ agarose gel electrophoresis under formamideformaldehyde denaturation conditions (Miller, 1987). Bands were visualized following ethidium bromide staining.
Assignments of fragment lengths in banded nucleic acid were based on linear regression equations relating
logl0(relative size) to gel migratory distance, by comparison to the migratory distances of RNA size standards.
Sample digestion with RNase A was used to establish specific nuclease sensitivity.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34
2412
M. OGLESBEE AND OTHERS
RESULTS
Nucleocapsid isolation
Nucleocapsid was recovered from 5-5 × 107 infected ceils per isolation experiment. O n the
gradients, a prominent double band was present in the 30 ~ CsC1 step (Fig. 1). The lower portion
of the double band, subsequently referred to as dense N C (D-NC), consisted of a broad zone of
dense finely globular particles. The average refractive index of material sampled from this level
was 1.3618 +__0.0008 (n = 22), and corresponded to a density of 1.2971 + 0-0042 g/ml. The upper
portion of the double band, referred to as light-NC (L-NC), consisted of a well defined dense
narrow zone of coarsely globular material. The average refractive index from this level was
1.3611 + 0.0008 (n = 11), and corresponded to a density of 1.2894 + 0.0040 g/ml. The ratio of
D - N C to L - N C varied from one isolation to the next, ranging from sole representation o f D - N C
to the simultaneous presence of prominent D - N C and L - N C bands. Recovery of L - N C alone
was not encountered. Viral titres measured from crude cytoplasmic extracts were independent of
D - N C to L - N C ratios and averaged 2.87 x 105 infectious units per 1 x 106 infected cells.
Fig. 1. Discontinuous CsCI gradient indicating isopycnic banding positions of CDV NC variants.
D-NC resolved as the lower portion of a double NC band within the 30 ~ (w/w) CsC1 step, equilibrating
at an average density of 1.2971 + 0-0002 g/ml (25 °C), L-NC position, corresponding to an average
density at 25 °C of 1.2894 + 0.0040g/ml, and Df-NC position within the 25~ (w/w) CsCI step,
corresponding to an average density at 25 °C of 1-2460 + 0.0046 g/ml are shown. The two faint bands
between Df-NC and L-NC correspond to bands in gradients of uninfected cellular material.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34
CD V nucleocapsid variants
2413
Fig. 2. Transmission electron micrograph of negatively stained CDV NC structural variants. (a) Intact
D-NC with a 20.1 + 1.3 nm outer diameter, 5.8 + 0.7 nm inner core diameter and a 1620.0 + 112.1 nm
strand length. Minimal strand breakage is observed. Bar marker represents 123-6nm. (b) Intact L-NC
exhibiting a lower resolution of capsid subunits, greater outer diameter (32.0 + 2-8 nm) and a greater
inner core diameter (10-4+ 0.6 nm). Fragmentation into short segments allowing an end-on view is
more frequent (arrow). Mean strand length is 1574.4 + 115.8 nm. Bar marker represents 123-6nm. (c)
Truncated Df-NC, with a 24-4+ 1-4 nm outer diameter, 6-9 + 0-4 nm inner core diameter and an
87.0 + 5.8 nm mean strand length. Fragmentation into short segments is observed as with L-NC
(arrow). Bar marker represents 42.0 nm.
Gradients of uninfected cellular material exhibited two faint bands in the upper portion of the
3 0 ~ CsCI step (data not shown). Refractive indices at this level averaged 1-3591 + 0-0005
( n = 12), corresponding to an average calculated density of 1.2692+ 0.0026g/ml. This
represents a density greater than that of D f - N C (see below) and less than that of L-NC. The
position of these bands corresponds to those present between L-NC and D f - N C in Fig. 1.
Additional material was not observed in regions corresponding to the b a n d i n g positions of the
N C variants.
A poorly defined homogeneous opaque band was also observed within the 25 ~ CsCI step of
the discontinuous gradients of infected cellular material. Refractive indices from this level
averaged 1-3568 + 0-0005 (n = 7), which corresponded to a density of 1-2460 _+ 0.0046 g/ml.
Increasing centrifugation times did not alter band positions or the measured refractive indices,
demonstrating the isopycnic nature of the separation.
Ultrastructural evaluation of band material
D - N C was characteristic of paramyxovirus nucleocapsids. Well delineated capsid subunits,
angled relative to the long axis of the strands, were seen (Fig. 2). The m e a n N C dimensions were
20-1 __+1 . 3 n m diameter with a 1620-0+ 112.1 n m strand length (Table 1). These capsids
exhibited m i n i m a l fragmentation, most of which was the result of air-drying on the grid surface
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34
2414
M. O G L E S B E E A N D O T H E R S
Table 1. Ultrastructural dimensions of CDVC NC variants
Width (nm)
Core diameter (nm)
Length (nm)
D-NC
20-1 + 1.3
5-8 + 0.7
1620-0 + 112.1
Df-NC
24.4 + 1.4
6.9 + 0.4
87.0 + 5.8
L-NC
32.0 + 2.8
10.4 + 0.6
1574.4 + 115.8
(M. Oglesbee, unpublished results). Where separation of short segments did occur, allowing an
end-on evaluation, an inner core diameter of 5.8 + 0.7 nm was measured.
L-NC differed from D-NC in that strand fragmentation was more frequent, capsid subunits
were less well defined, and the inner and outer core diameters were larger. The dimensions of
L-NC were 3 2 . 0 + 2 . 8 n m outer diameter, lO.4+__0.6nm inner core diameter, and
1574.4 + 115.8 nm mean strand length. Isolates from the 2 5 ~ CsCI step appeared as truncated
paramyxovirus capsids with an 87.0 + 5-8 nm length. These structures, subsequently referred to
as Df-NC, were intermediate between D-NC and L-NC in size, with a 24.4 + 1.4 nm outer
diameter and a 6-9 + 0.4 nm inner core diameter. The frequency of strand fragmentation was
also intermediate between that of D-NC and L-NC.
Careful ultrastructural examination of virus-infected whole ceils revealed that cytoplasmic
NC was pleomorphic. Histograms relating the frequency of occurrence of NC outer diameters
(n = 255) yielded three peaks at 24, 29 and 36 nm. Within a given cytoplasmic N C focus,
histograms were either unimodal or bimodal. Unimodal distributions were obtained which
corresponded to each of the aforementioned peaks. For bimodal distributions, peaks occurred
predominantly at 29 and 36 nm and represented the most frequently encountered N C
aggregates. NC morphology, apart from variance in diameter, was similar within and between
cytoplasmic loci of NC aggregates; NC strands were electron-dense with granular margins and
capsid subunits were not discernible.
Protein analysis
Gel electrophoresis of D-NC revealed the presence of two major and one minor protein
species (Fig. 3). Based on a linear regression equation relating loglo(relative mass) to gel
migratory distance (correlation coefficient > 0.99), the Mr assignments corresponding to visible
bands were calculated as 61K, 75K and 160K to 200K. These values are in agreement with
published assignments for paramyxovirus N, P and L capsid proteins, respectively (Rima,
1983). An additional band was detected in L-NC and Df-NC samples, with a calculated Mr of
70K. Western blot analysis of Df-NC and L-NC protein demonstrated reactivity of only the 61K
(N) and 75K (P) major core species with CDV convalescent gnotobiotic canine serum. In
addition, a number of lower Mr bands, predominantly between 55K and 44K, were detected.
These correspond to major N protein degradation products based on mass (Croce et al., 1980)
and reactivity to viral N protein-specific antiserum (Oglesbee et al., 1989).
Protein an~tlysis of the double band sample from gradients of uninfected cellular material
demonstrated a heterogeneous population of proteins, with prominent representation of a 70K
species (data not shown). Protein was not detectable in gradient samples harvested from regions
corresponding to D-NC or L-NC, with trace amounts occasionally recovered from regions
corresponding to the density of Df-NC. The protein content of the latter was as described for the
double band sample.
The N, P and L protein mass composition, from Coomassie Brilliant Blue staining intensity,
was consistent between NC isolates; the previously determined averages are 89.7 ~o N, 8.4~ P
and 1.9~ L (Oglesbee et aL, 1989). The 70K protein of L-NC was present in amounts ranging
from 4-0 to 7-7 ~ of the total NC protein mass. These proportions are not reflected in combined
Coomassie Brilliant Blue-silver-stained polyacrylamide gels containing low amounts (e.g.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34
C D V nucleocapsid variants
4
2415
5
b -
cd-
--68K
--55K
--40K
--29K
/:
•
• v•~•• • •
••ii ~•
•
.'
' • ••i
Fig. 3. SDS-PAGE (12~o) analysis of CDV NC variant proteins and Western blot analysis of L-NC
protein. Corresponding positions of Mr standards are indicated to the left of lanes 1 to 5 and to the right
of lane 6. Lanes 1 and 2, combined Coomassie Brilliant Blue-silver stain of 0-6 and 1"2lagof total D-NC
protein, respectively, resolved by 12~ SDS-PAGE. The L protein (band a) is present in low
proportions and is visualized only when larger amounts of total protein are examined. The P (band b)
and N (band d) proteins are readily evident. Lane 3, silver stain of 0.6 lagof total L-NC protein resolved
by 12~ SDS-PAGE. An additional 70K species (band c) is evident. Lanes 4 and 5, Western blot
analysis of 1.0 I~gof L-NC protein, resolved by 12~ SDS-PAGE, using uninfected canine serum (lane
4) and CDV convalescent canine serum (lane 5) as the source of primary antibody. Although prominent
reactivity to the N (d) and P (b) protein bands by the convalescent serum is evident, reactivity to the
70K protein is lacking. Reactivity to a number of lower Mr species corresponding to major N protein
degradation products is also apparent. Lane 6, Coomassie Brilliant Blue-stained gel of a Df-NC
preparation, demonstrating the prominent representation of the 70K protein (band c) relative to the
other major core protein species P (band b) and N (band d).
1.6 ~tg) of total N C protein (Fig. 3, lanes 1 and 3) due to the non-stoichiometric nature of the
protein-silver interaction. The amount of 70K protein present in D f - N C samples was more
variable than in L-NC, occasionally representing as much as 47.3 ~ of the total D f - N C protein.
The amount structurally associated with D f - N C was obscured by the fact that protein of cellular
origin, similarly derived from uninfected cells, could occasionally be isolated from CsC1
gradients at levels corresponding to the density of Df-NC.
R N A analysis
A single high Mr species of nucleic acid extracted from D - N C was visualized by ethidium
bromide staining of the agarose gel (Fig. 4). Some smearing below the band was present,
indicative of degradation products. Based on a regression equation relating logl0(fragment
length) to gel migratory distance (correlation coefficient > 0-99), it was calculated that the band
represented a fragment 15 kb in length. This correlates with published estimates of the length of
paramyxovirus genomic R N A (Udem & Cook, 1984). Similar results were obtained from L-NC,
except that the effects of non-specific nuclease degradation were more pronounced. That is, a
single high Mr band, comigrating with that derived from D - N C , was observed with lower Mr
smearing more evident. Analysis of D f - N C extracts yielded products ranging in size from 1.16 to
0-67 kb, with the most intense ethidium bromide staining corresponding to an estimated
fragment length of 0.86 kb. Small amounts of a well defined band, having a gel migratory
distance corresponding to 5.23 kb, were occasionally observed.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34
2416
M. O G L E S B E E A N D O T H E R S
1
2
3
a~
b~
-9.5
-7.5
-4-4
-1.4
-0-2
Fig. 4. Ethidium bromide staining of CDV NC variant RNA separated by agarose gel electrophoresis
under formamide-formaldehyde denaturing conditions. Size standards are presented to the right of
each sample lane(s), with the corresponding fragment length indicated in kb. Lanes 1 and 2, 1-0~
agarose gel electrophoresis of nucleic acid extracted from D-NC (lane l) and L-NC (lane 2). A single
prominent species (band a) is observed in the D-NC sample lane with a corresponding length calculated
as 15.0kb. Some smearing of lower Mr products is present, reflecting non-specific nuclease
degradation. Similar results are observed in the L-NC sample lane, though the effects of non-specific
nuclease degradation are more pronounced. Lane 3, 1.5% agarose gel electrophoresis of RNase
A-sensitive Df-NC nucleic acid extracts. Most extracted material is represented by a small range of
sizes, between 1-16 and 0.67 kb, with the most intensely stained sample corresponding to a fragment
length of 0.86 kb (band b).
DISCUSSION
We have noted the tendency of CDV NC to form visible aggregates at low CsCI gradient
concentrations and for this tendency to be particularly pronounced for L-NC (unpublished
observation). This is interpreted to be a concentration-independent property of NC in high salt
environments. The degree of gradient loading employed in these experiments was based on that
amount which most clearly demonstrated differential gradient band position and morphology.
Since variable particle aggregation can affect sedimentation velocity, it is necessary to ensure
that NC bands are present at a position of equivalent density at the time of harvest. Increased
centrifugation times have previously been shown to have no effect on the resultant gradient
banding positions (Oglesbee et al., 1989), indicating that an isopycnic density had been
achieved. This pattern was unchanged at different levels of gradient loading. Thus, where
particle rate-zonal migration may be affected by aggregation, isopycnic banding is not.
Three ultrastructural NC variants were identified in Ond-CDV-infected Vero cells.
Distributions of NC diameters yielded high frequencies at 24, 29 and 36 nm. These correspond
to the diameters of the subsequently isolated D-NC (20.1 nm), Df-NC (24.4 nm) and L-NC
(32.0 nm), being increased by 3.9 nm, 4.6 nm and 4.0 nm respectively. The size discrepancy
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34
C D V nucleocapsM variants
2417
between the NC diameters measured in situ and after isolation of each species of NC was
consistent (4.1 nm average), indicating that the isolated NC represent naturally occurring
structural variants and not artefacts generated during the isolation procedure.
The observed differences in NC diameters are important in terms of understanding potential
mechanisms by which NC variants could arise. We found the outer and inner core diameters of
D-NC to be 20.1 and 5.8 nm, respectively. Published values for paramyxovirus NC generally
report an outer diameter of 18 nm and an inner core diameter of 5-2 nm (Waters & Bussell,
1974). Thus, the ratio of our measurements to those in the literature may be expressed as 1.12:1
and 1.12:1 respectively. The increased inner diameter proportionate to the increased outer
diameter suggest an expansion of the NC single-start helix (Finch & Gibbs, 1970). When similar
ratio analyses were performed with L-NC and DfoNC, significant helical expansion was noted.
That is, the ratio of Df-NC to D-NC measurements is 1.21 (outer diameter) and 1.19 (inner core
diameter). Similarly, the ratios of L-NC to D-NC measurements are 1.59 and 1.79 for outer and
inner core diameters, respectively. In the case of L-NC, helical expansion coincides with an
increased susceptibility of the encapsidated genome to non-specific nuclease degradation,
probably the result of decreased NC subunit overlapping, which is thought to protect the
underlying genome (Finch & Gibbs, 1970). The increased helical expansion noted for Df-NC
and L-NC, when compared to D-NC, correlates with the presence of the unique 70K protein.
Protein binding can also directly affect diameters by simple association with N protein NC
structural subunits (Finch & Gibbs, 1970), thereby increasing the external projection of these
elements. Such binding would explain the lower resolution of N C subunits observed with L-NC.
Overall, the contribution to external diameter would necessarily be small in order to maintain
the proportion of inner and outer core diameters.
The average NC lengths for L-NC and D-NC were 1.57 _+ 0-12 ~tm and 1.62_+ 0-12~tm,
respectively. These are greater than those generally reported in the literature which have an
average of 1.1 ~tm (Kingsbury, 1972; Finch & Gibbs, 1970). We attribute this discrepancy to our
method for NC isolation, in which N C fragmentation is minimized. Finch & Gibbs (1970) have
described MV NC lengths of 0-5 to 1"75 ~tm. Their lower reported mean length is probably the
result of strand fragmentation biasing their sample.
Functional defective interfering (DI) particles are commonly produced in paramyxovirus
systems (Leppert et al., 1977). Specific genomic deletion mutations are produced which permit
the survival of self-replicating truncated genomes possessing highly variable genetic
information (Lazzarini et al., 1981). In the Sendai virus system, DI particle genome lengths
range between 4.3 and 1.19 kb (Re et al., 1983). These values parallel those determined for
nucleic acid extracted from Df-NC, which ranged in size between 1.16 and 0.67 kb, with an
occasional minor species detected at 5.23 kb. The low density of these structures may be
explained by an increased protein : R N A ratio. The self-replicating nature of these structures is
supported by electron microscopic observations. Here, foci of cytosolic NC replication were
observed which consisted of a unimodal distribution of capsid diameters, where the mode was
correlated to the diameter of isolated Df-NC. Bimodal distributions of capsid diameters within
specific cytosolic foci, with modes correlated to Df-NC and L-NC diameters, suggests that
defective particles arise from L-NC.
The nature of the 70K protein recovered from L-NC and Df-NC, but not D-NC, is unknown.
Failure to react with CDV convalescent antiserum on Western blot analysis strongly suggests
that it is of Vero cell (host) origin. That the protein is specifically associated with one form of NC
but not another suggests more than a casual association. It is tempting to speculate that this
protein imparts the decreased subunit resolution and mediates the helical expansion of L-NC. If
the latter truly represents the biologically active form of paramyxovirus NC, then this protein
may well function as a cellular trans-activator of virus transcriptional activity. Investigations
into the identity of the CDV NC-associated 70K protein, viral and cellular determinants
affecting the association, and the biological significance of such an association are in progress.
This work was supported by the State of Ohio Canine Research Fund and grant RO1-14821, NIH, PHS. Stipend
support for the first author was provided by the DuPont Corporation and an NIH, PHS, National Research
Service Award F32-107811.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34
2418
M. O G L E S B E E AND OTHERS
REFERENCES
BOHN, W., RUTrER, G., HOHENBERG, H., MANNWEILER, K. & NOBIS, P. (1986). Involvement of actin filaments in
budding of measles virus: studies on cytoskeletons of infected cells. Virology 149, 91-106.
BRUNER, R. & VINOGKAD, J. 0965). The evaluation of standard sedimentation coefficients of sodium R N A and
sodium D N A from sedimentation velocity data in concentrated NaC1 and CsCl solutions. Biochimica et
biophysica acta 108, 18-29.
BUSSELL,t . H., WATERS,D. J. & SEALS,M. K. (1974). Measles, canine distemper and respiratory syncytial virions and
nucleocapsids. A comparative study of their structure, polypeptide and nucleic acid composition. Medical
Microbiology and Immunology 160, 105-124.
CATTANEO, R., REBMANN, G., SCHMIDT, A., BACZKO, K., TER MEULEN, V. & BILLETER, M. A. (1987). Altered
transcription of a defective measles virus genome derived from a diseased human brain. EMBO Journal 6,
681-688.
CHINCI-IAR,V. G. & PORTNER, A. (1981). Inhibition of R N A synthesis following proteolytic cleavage of Newcastle
disease virus P protein. Virology 115, 192-202.
CHUI, L. W-L., VAINIONP~f,,R., MARUSYK,R., SALMI,A. & NORRBY, E. (1986). Nuclear accumulation of measles virus
nucleoprotein associated with a temperature-sensitive mutant. Journal of General Virology 67, 2153-2161.
CROCE, C. M., LINNENBACH,A., HALL,W., STEPLEWSKI,Z. & KOPROWSKI,H. (1980). Production of human hybridomas
secreting antibodies to measles virus. Nature, London 288, 488-489.
DESHPANDE,K. L. & PORTNER, A. (1985). Monoclonal antibodies to the P protein of Sendal virus define its structure
and role in transcription. Virology 140, 125-134.
DUBOIS-DALCQ,M., BARBOSA,L. H., HAMILTON,R. & SEVER, J. L. (1974). Comparison between productive and latent
subacute sclerosing panencephalitis viral infection in vitro: an electron microscopic and immunoperoxidase
study. Laboratory Investigation 30, 241-250.
FINCH, I. T. & Gram, A. J. (1970). Observations on the structure of the nucleocapsids of some paramyxoviruses.
Journal of General Virology 6, 141-150.
FOURNIER, J., PRIUVAT, A. & BOUTEILLE, M. (1981). Structural changes in the cell nucleus during measles virus
infection of cerebellar explants. Journal of Ultrastructure Research 77, 319-328.
GIUFFRE, R. M., TOVELL, D. R., KAY, C. M. & TYRRELL, D. L. J. (1982). Evidence for the interaction between the
membrane protein of a paramyxovirus and actin. Journal of Virology 42, 963-968.
KINGSBURY, D. W. (1972). Paramyxovirus replication. Current Topics in Microbiology and Immunology 59, 1-33.
KINGSBURY, D. W. & DARLINGTON, R. W. (1968). Isolation and properties of Newcastle disease virus nucleocapsid.
Journal of Virology 2, 248-255.
KOESTNER, A. & LONG, J. (1970). Ultrastructure of canine distemper virus in explant tissue cultures of canine
cerebellum. Laboratory Investigation 23, 196-201.
KOLAKOFSKY,D. & BLUMBERG,B. M. (1982). A model for the control of non-segmented negative strand virus genome
replication. In Virus Persistence. Edited by B. W. J. Mahy, A. C. Minson & G. K. Darby. Cambridge:
Cambridge University Press.
LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
Nature, London 227, 680-685.
LAMa, R. A., MXh~, B. W. J. & CHOPPII% e. w. (1976). The synthesis of Sendai virus polypeptides in infected cells.
Virology 69, 116-131.
LAZZARINI, R. A., KEENE, J. D. & SCHUBERT,M. (1981). The origins of defective interfering particles of the negative
strand R N A viruses. Cell 26, 145-154.
LEPPERT, M., tORT, L. & KOLAKOFSKY,D. (1977). Further characterization of Sendai virus DI-RNAs: a model for
their generation. Cell 12, 539-552.
MATSUMOTO,N. (1966). Studies on measles virus in tissue culture. IL Electronmicroscopic study of measles virus
infected cells and localization of virus antigen examined by ferritin-conjugated antibody method. Bulletin of
the Yamaguchi Medical School 13, 167-189.
MERRIL,C. R., GOLDMAN,D., SEDMAN,S. A. & EBERT,M. H. (1981). Ultrasensitive stain for proteins in polyacrylamide
gels shows regional variation in cerebrospinal fluid proteins. Science 211, 1437.
MILLER, K. (1987). Gel electrophoresis of RNA. Focus 9 (3), 14. Gaithersburg: Bethesda Research Laboratories.
MINAGAWA,T., SAKUMA,T., KUWmIMA,S., YAMAMOTO,T. K. & IIDA, H. (1976). Characterization of measles viruses in
establishment of persistent infections in human lymphoid cell line. Journal of General Virology33, 361-379.
NARANG, H. K. (1981). Comparative morphology of measles virus and paramyxovirus-like tubules in multiple
sclerosis using ruthenium red stain. Neuropathology and Applied Neurobiology 7, 411-420.
OGLESBEE, i . , TATALICK, L., RINGLER, S., RICE, J. & KRAKOWKA, S. (1989). Rapid isolation of morbillivirus
nucleocapsid for genomic R N A e D N A cloning and the production of specific core protein antisera. Journalof
Virological Methods (in press).
OYANAGI, S., TER MEULEN, V. & MULLER, D. (1970). Electron microscope observations in subacute sclerosing
panencephalitis brain cell cultures: their correlation with cytochemical and immunocytological findings.
Journal of Virology 6, 370-379.
OYANAGI,S., TER MEULEN,V., KATZ,M. & KOPROWSKI,H. ( 1971 ). Comparison of subacute sclerosing panencephalitis
and measles virus: an electron microscope study. Journal of Virology 7, 176-182.
PORTNER, A. & KINGSBURY,D. W. (1976). Regulatory events in the synthesis of Sendal virus polypeptides and their
assembly into virions. Virology 73, 79-88.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34
C D V nucleocapsid variants
2419
r~aR'ta~S, A. & ~LmYI, K. G. (1986). Localization of P, NP, and M proteins on Sendal virus nucleocapsid using
immunogold labeling. Virology 150, 469-478.
PRINGLE, ¢. t . (1987). Rhabdovirus genetics. In The Rhabdoviruses, pp. 167-243. Edited by R. R. Wagner. New
York: Plenum Press.
RE, G. G., GUPTA,K. C. & KINGSBURY,D. W. (1983). Genomic and copy-back 3' termini in Sendal virus defective
interfering RNA species. Journal of Virology 45, 659-664.
RIMA, 13. K. (1983). The proteins of morbilliviruses. Journal of General Virology 64, 1205-1219.
ROBBINS, S. I. (1983). Progressive invasion of cell nuclei by measles virus in persistently infected human cells.
Journal of General Virology 64, 2335-2338.
ROnBINS, S. J. a BUSSELL,R. H. (1979). Structural phosphoproteins associated with purified measles virions and
cytoplasmic nucleocapsids. Intervirology 12, 96-102.
ROBBIN$, S. J., BUSSELL,R. H. & RAPP, F. (1980). Isolation and partial characterization of two forms of cytoplasmic
nucleocapsid from measles virus-infected cells. Journal of General Virology 47, 301-310.
ROZENBLATI',S., KOCH,T., PlNI{ASI,O. a BRATOSIN,S. (1979). Infective substructures of measles ,titus isolated from
acutely and persistently infected cells. Journal of Virology 32, 329-333.
SCHUBERT,M., HARMISON,G. G. & MEIER,E. (1984). Primary structure of the vesicular stomatitis virus polymerase
(L) gene: evidence for a high frequency of mutations. Journal of Virology 51, 505-514.
SEIFRIED, A. S., ALBRECHT,P. & MILSTEIN,J. B. (1978). Characterization of an RNA-dependent RNA polymerase
activity associated with measles virus. Journal of Virology 25, 781-787.
TYRRELL,D. L. J. &NORRBY,E. (1978). Structural polypeptides of measles virus. Journal of General Virology 39, 219229.
UCHIYAMA,T., TAKEDA,K. & YAMANOUCm,K. (1985). Replication of measles virus and localization of the viral
antigens in long term infection of organotypic cultures of hamster dorsal root ganglion. Japanese Journal of
Experimental Medicine 55, 109-121.
tmEM, S. A. a COOK,K. A. (1984). Isolation and characterization of measles virus intracellular nucleocapsid RNA.
Journal of Virology 49, 57-65.
WATERS,D. J. & BUSSELL,R. H. (1974). Isolation and comparative study of the nucleocapsids of measles and canine
distemper viruses from infected cells. Virology 61, 64-79.
(Received 14 February 1989)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 16:26:34