of 6
JOURNAL
OF
VIROLOGY,
Mar.
1971,
p.
389-394
Copyright
©
1971
American
Society
for
Microbiology
Vol.
7,
No.
3
Printted
in
U.S.A.
Ribonucleic
Acid
Polymerase
in
Virions
of
Newcastle
Disease
Virus:
Comparison
with
the
Vesicular
Stomatitis
Virus
Polymerase
ALICE
S.
HUANG,'
DAVID
BALTIMORE,
AND
MICHAEL
A.
BRATT
Department
of
Biology,
Massachusetts
Institute
of
Technology,
Cambridge,
Massachusetts
02139,
and
Departmenit
of
Microbiology
and
Molecular
Genetics,
Harvard
Medical
School,
Boston,
Massachusetts
02115
Received
for
publication
4
January
1971
The
virions
of
Newcastle
disease
virus
(NDV)
contained
an
enzyme
that
catalyzed
the
incorporation
of
ribonucleotides
into
ribonucleic
acid
(RNA).
Optimal
condi-
tions
for
this
polymerase
activity
were
identical
to
the
conditions
for
the
vesicular
stomatitis
virus
(VSV)
polymerase,
and
both
enzymes
were
active
for
longer
times
at
32
C
than
at
37
C.
However,
the
specific
activity
of
the
NDV
polymerase
was
less
than
3
%l,
that
of
the
VSV
polymerase.
Product
RNA
species
from
the
NDV
and
VSV
polymerase
reactions
annealed
specifically
to
the
homologous
virion
RNA
species.
Transcriptive
intermediates
containing
product
RNA
attached
to
the
respective
virion
RNA
could
be
identified
in
both
systems.
Ribonucleic
acid
(RNA)
extracted
from
virions
of
Newcastle
disease
virus
(NDV)
lacks
infectivity
(14),
and
during
NDV
replication
there
is
exten-
sive
synthesis
of
RNA
complementary
in
base
sequence
to
the
major
species
of
virion
RNA
(6,
15).
This
complementary
RNA
appears
to
be
attached
to
polyribosomes
and
may
serve
as
messenger
RNA
(6).
Similar
findings
with
vesicu-
lar
stomatitis
virus
(VSV)
led
to
the
characteriza-
tion
of
an
RNA
polymerase
in
virions
of
VSV
(1).
Therefore,
NDV
is
also
likely
to
have
such
an
enzyme.
A
previous
attempt
which
failed
to
detect
polymerase
activity
in
purified
preparations
of
NDV
led
to
the
conclusion
that
if
the
NDV
polymerase
exists
it
must
have
less
than
2%
of
the
specific
activity
of
VSV
preparations
(1).
By
ob-
taining
a
more
concentrated
preparation
of
virions
and
by
increasing
the
sensitivity
of
the
assay,
an
RNA-dependent
RNA
polymerase
has
now
been
detected
in
virions
of
NDV.
MATERIALS
AND
METHODS
Viruses.
NDV
strains
used
were
NDV-HP
(Israel,
HP,
1935),
NDV-IM
(Milano,
Italy,
1945),
NDV-N
(NJ,
La
Sota,
1946),
and
NDV-RO
(California
RO,
1944).
Each
strain
was
cloned,
grown,
and
concen-
trated
as
previously
described
(5,
7)
and
then
partially
purified
by
sedimentation
through
207'O
sucrose
(Cla-
vell
and
Bratt,
in
preparation).
The
pellet
of
virus
was
resuspended
and
further
purified
through
an
isopycnic
1
Present
address:
Channing
Laboratory,
774
Albany
St.,
Boston,
Mass.
02118.
sucrose-D20
gradient
as
described
in
Fig.
3.
Prepara-
tions
of
NDV-HP,
containing
1.2
to
2.0
mg
of
NDV
protein
(17)
per
ml
and
5
X
1010
to
10
X
1010
plaque-
forming
units
per
ml,
were
used
in
all
experiments
unless
otherwise
indicated.
A
preparation
of
the
Indiana
serotype
of
VSV,
similar
to
the
stocks
previously
described
(1,
24)
and
containing
200
,ug
of
VSV
protein
per
ml,
was
used
throughout
these
experiments.
Procedures
and
reagents.
All
of
the
reagents
and
procedures
were
as
previously
described
(1),
except
for
modifications
noted
in
the
figure
legends.
Virion
RNA.
For
the
annealing
experiment,
virion
RNA
was
extracted
and
purified
as
previously
de-
scribed
for
NDV
(10)
and
for
VSV
(23).
A
pool
of
virion
RNA
from
several
NDV
strains
was
used.
This
is
justified
by
the
high
degree
of
cross-annealing
be-
tween
virion
RNA
and
complementary
RNA
from
cells
infected
with
different
NDV
strains
(14;
Bratt,
unpublished
data).
RESULTS
General
properties
of
the
NDV
RNA
polym-
erase.
The
polymerase
activity
found
in
virions
of
NDV
had
properties
very
similar
to
those
of
the
VSV
polymerase
(Tables
1
and
2,
Fig.
1
and
2),
except
that
the
specific
activity
was
1
to
3
%
of
the
specific
activity
of
VSV
preparations
(Fig.
1
and
2).
The
amount
of
activity
in
NDV
preparations
was
approximately
proportional
to
the
concentra-
tion
of
viral
protein
in
the
reaction
mixture.
The
activity
was
dependent
on
activation
by
a
non-
ionic
detergent,
Triton
N-101
or
Nonidet
P-40,
and
on
the
presence
of
a
sulfhydryl
reagent,
dithiothreitol.
The
omission
of
uridine
triphos-
389
HUANG,
BALTIMORE,
AND
BRATT
TABLE
1.
General
properties
of
the
Newcastle
disease
virus
polymerasea
Amt
(pmoles)
of
GMIP
Reaction
mixture
incorporated
per
mg
of
protein
Complete.............................
Minus
dithiothreitol................
Minus
UTP........................
Minus
CTP
..................
Minus
CTP,
plus
dCTP.............
Minus
Triton
N-101................
Minus
Triton,
plus
Nonidet
P-40....
215
<20
28
<20
22
<20
224
a
The
complete
reaction
mixture
consisted
of
the
following
components,
in
a
total
volume
of
0.1
ml:
5
4moles
of
tris(hydroxymethyl)amino-
methane-hydrochloride,
pH
7.3;
0.4
,mole
of
mag-
nesium
acetate;
0.3
,mole
of
dithiothreitol;
10
,umoles
of
NaCl;
0.07
,umole
each
of
adenosine
tri-
phosphate,
uridine
triphosphate
(UTP),
and
cyti-
dinetriphosphate
(CTP);
0.001
,umole
of
3H-guano-
sine
triphosphate
(1,160
counts
per
min
per
pmole);
80
,.g
of
Triton
N-101;
and
10
,ug
of
viral
protein.
The
reaction
mixture
was
incubated
at
32
C
for
1
hr.
The
reaction
was
terminated
and
each
sample
was
assayed
for
acid-precipitable
3H-guanosine
mono-
phosphate
(GMP)
as
previously
described
(1).
An
incubated
sample
without
virions
contained
16%,7
of
the
radioactivity
of
the
complete
system,
and
this
value
was
subtracted
from
all
of
the
experi-
mental
values.
dCTP,
deoxycytidine
triphosphate.
TABLE
2.
Susceptibility
of
the
Newcastle
disease
virus
(ND
V)
polymerase
reaction
to
inhibitors,,
Amt
(pmoles)
of
3H-GNIP
Reaction
mixture
incorporated
per
mg
of
protein
Complete
.............................
198
Plus
ribonuclease
(50,/g,/ml)
<10
Plus
trypsin
(5,4g
/ml)
20
Plus
deoxycholate
(1
mg/ml)
........
<10
Plus
actinomycin
D
(20
,g/ml)
......
188
Plus
rifampin
(20
pg/mI)
...........
190
Plus
deoxyribonuclease
(100l
g/ml).
186
t
Standard
0.1-ml
reaction
mixtures,
as
in
Table
1,
containing
20
pg
of
NDV
protein
and
each
of
the
inhibitors,
were
incubated
at
32
C
for
1
hr.
Mix-
tures
lacking
virions
but
incubated
at
32
C,
or
con-
taining
virions
but
unincubated,
contained
approximately
10%
of
the
radioactivity
of
the
complete
system.
This
value
was
subtracted
from
all
of
the
experimental
values.
GMP,
guanosine
monophosphate.
phate
or
cytidine
triphosphate
or
the
substitution
of
deoxycytidine
triphosphate
for
cytidine
tri-
phosphate
in
the
reaction
mixture
virtually
elim-
inated
the
activity
(Table
1).
The
enzymatic
ac-
tivity
was
sensitive
to
trypsin,
ribonuclease,
and
deoxycholate
and
insensitive
to
inhibitors
of
de-
oxyribonucleic
acid
(DNA)-dependent
RNA
syn-
thesis
(Table
2).
The
magnesium
ion
concentra-
tion
required
for
optimal
activity
was
4
to
6
mM,
and
substitution
with
manganese
at
concentrations
from
2
to
26
mm
did
not
result
in
any
detectable
activity.
A
salt
concentration
of
0.05
to
0.1
M
was
necessary
for
the
detection
of
optimal
activity.
Effects
of
temperature
and
pH.
For
both
NDV
and
VSV
polymerases,
the
optimal
activity
was
observed
at
pH
7.3
(Fig.
1).
Also,
for
both
viruses
incorporation
of
guanosine
monophosphate
(GMP)
continued
for
longer
times
at
32
C
than
at
37
C
(Fig.
2).
The
kinetics
of
synthesis
by
NDV
were
complicated
and
require
further
study.
At
temperatures
higher
than
37
C,
polymerase
activ-
ity
for
NDV
and
VSV
was
markedly
reduced.
Coincidence
of
polymerase
activity
in
an
iso-
pycnic
gradient
with
NDV
infectivity
and
hemag-
glutinin.
To
investigate
whether
the
polymerase
activity
was
part
of
the
virion,
a
partially
purified
preparation
of
NDV
was
centrifuged
to
equilib-
rium
in
a
sucrose
density
gradient.
Polymerase
activity
coincided
with
both
hemagglutinating
activity
and
infectivity
at
a
buoyant
density
of
1.19
g/ml
(Fig.
3).
Polymerase
activity
in
other
NDV
strains.
To
test
whether
strains
of
NDV
other
than
NDV-HP
c
I
o
15
0.
C
0
~
O
0
C.
U
Z
cn
°E
E5
O
-
a
Q
NDV
N
o
0
\
\\,
-vs
v
KS
\
\
"I-
*0
0
I;
;
o
L
6.5
0
7.0
7.5
8.0
8.5
p
H
C
ro
.9
15
o
ca.
X
O
C
U
-
c
(U
0
a-
102
ca
tD>
w
>
7O
C31
5E
E
0
a0.0
FIG.
1.
Dependenice
ont
pH
of
the
Newcastle
disease
virus
(NDV)
and
vesicular
stomatitis
virus
(VSV)
polymerase
activities.
Standard
reaction
mixtures
of
0.1
ml
for
ND
V
with
5
lrnoles
of
tris(hydroxymethyl)-
aminomethane
(Tris)
buffer
at
the
different
pH
values
contained
12
pg
of
viral
protein
and
were
incuibated
at
37
C
for
30
min.
Standard
reactioni
mixtures
for
VSV
(1)
in
0.3
ml
withl
15
,umoles
of
Tris
buffer
at
the
different
pH
values
conztained
13
pug
of
viral
protein
and
were
incubated
at
37
C
for
20
min.
GMP,
guanosine
mono-
phosphate.
390
J.
VIROL.
NDV
RNA
POLYMERASE
391
12
I
._
C'
o
O
C
-
8
u
'
c
°-
E-
X
Qn.
0
2
a
6o
>
4
E
E
o
_
0
a.
CL
2
M
INUTES
FIG.
2.
Time
course
of
guanosine
monophosphate
(GMP)
incorporation
at
32
and
37
C
for
Newcastle
disease
virus
(ND
V)
anid
vesicular
stomatitis
virus
(VSV)
polymerase
reactions.
Standard
0.1-ml
reaction
mixtures
were
used
conitaining
either
16
,ug
of
ND
V
protein
or
2
jig
of
VSV
protein
and
incubated
for
the
inidicated
times
at
37
C
or
32
C.
o
_
o
0
C,
2-
0
a)
_
O
Iv
.-0
E-a
0.
10
5
o
x
A
>
x
X
X
"?
E
E
"I
11
<
QI
-
5
10
1
5
FRACTION
NUMBER
FIG.
3.
Location
of
the
Newcastle
disease
virus
(ND
V)
polymerase
activity
by
isopycnic
sedimentation
in
a
sucrose
gradient.
Partially
purified
ND
V
was
layered
onto
a
linear
gradient
made
from
15%
sucrose
in
water
and
65%
sucrose
in
D20.
The
buffer
was
0.01
m
tris(hydroxymethyl)aminomethane
(pH
7.4)
with
0.03
M
NaCI.
Centrifugation
was
in
a
Beckman
SW41
rotor
at
-170,000
X
g
for
3
hr
at
4
C.
Fractions
of
0.7
ml
were
collected
from
the
bottom.
Plaque
assay
(5)
and
hemagglutination
(HA)
titration
of
ND
V
(7)
were
done
as
previously
described,
except
for
some
modifica-
tions
of
the
hemagglutination
assay
(Clavell
and
Bratt,
in
preparation).
GMP,
guanosine
monophosphate;
PFU,
plaque-forming
unit.
also
contained
virion-associated
RNA
polymer-
ase
activity,
strains
NDV-IM,
NDV-RO,
and
NDV-N
were
tested
for
enzymatic
activity.
All
three
strains
contained
polymerase
activity
but
at
a
slightly
lower
specific
activity
(-50
pmoles
per
mg
of
protein)
than
NDV-HP
(-200
pmoles
per
mg
of
protein)
when
incubated
for
1
hr
at
32
C.
The
presence
of
an
RNA
polymerase
for
four
independently
purified
strains
of
NDV
indicates
that
the
enzymatic
activity
is
not
due
to
contami-
nation
of
NDV-HP
with
another
virus.
Annealing
of
the
products
to
virion
RNA.
To
demonstrate
that
the
product
made
by
the
NDV
polymerase
is
a
transcript
of
NDV
RNA,
polym-
erase
products
of
NDV
and
VSV
were
tested
for
annealing
to
either
NDV
or
VSV
RNA.
Table
3
shows
that
the
NDV
product
annealed
to
only
NDV
RNA
and
not
to
VSV
RNA,
whereas
the
VSV
product
annealed
only
to
its
homologous
RNA.
The
high
degree
of
self-annealing
in
the
absence
of
added
virion
RNA
probably
resulted
from
the
large
amount
of
virion
RNA
templates
already
present
in
the
reaction
mixture.
However,
until
further
experiments
are
done,
we
cannot
rule
out
the
possibility
of
self-annealing
of
the
prod-
uct,
especially
because
there
are
reports
of
self-annealing
of
RNA
from
virions
of
NDV
(19,
20).
Initial
transcriptive
intermediates.
If
virion
RNA
serves
as
a
template
for
the
NDV
and
VSV
polym-
erases,
labeled
products
made
during
the
polym-
erase
assay
should
be
initially
associated
with
virion
RNA
in
a
ribonuclease-resistant
form
and
VOL.
7,1971
c
0._
C),
o
°
CL>
C
._
-
z
a)
a.
-
CD
O
1.19
gm
/cc
:i
i
I
I
lj
HUANG,
BALTIMORE,
AND
BRATT
TABLE
3.
Specificity
of
Newcastle
disease
virus
(NDV)
and
vesicular
stomatitis
virus
(VSV)
polymerase
products
by
cross-annealing
to
NDV
and
VSV
ribonucleic
acid
(RNA)a
Annealed
with
Per
cent
Polymerase
product
Annealed
Nith
ribonuclease-
added
RNA
~resistance
NDV
None
39.5
NDV
VSV
42.2
NDV
NDV
107.0
VSV
None
5.3
VSV
VSV
101.2
VSV
NDV
6.2
a
Standard
polymerase
reaction
mixtures
con-
taining
virions
of
NDV
or
VSV
were
terminated
after
1
hr
at
32
C
by
the
addition
of
0.1%
sodium
dodecyl
sulfate.
Details
of
annealing
were
as
pre-
viously
described
(12),
except
that
product
at
5,000
counts
per
min
per
tube
and
NDV
or
VSV
RNA
at
4
,g
per
tube
were
used,
and
all
the
sam-
ples
were
boiled
and
then
annealed.
should
sediment
at
approximately
the
same
rate
as
virion
RNA.
Such
a
complex
will
be
called
a
transcriptive
intermediate.
As
more
product
is
synthesized,
the
transcriptive
intermediate
should
sediment
more
slowly
than
virion
RNA
alone
and
product
RNA
should
be
released
from
the
com-
plex.
Because
virion
RNA
species
of
NDV
and
VSV
sediment
at
approximately
50
to
57S
(10,
14,
22;
W.
R.
Adams,
Fed.
Proc.
24:
159,
1965)
and
40
to
45S
(13,
23),
respectively,
examination
of
the
initially
labeled
products
in
sucrose
gradients
should
show
a
difference
in
size
between
the
two
viral
systems.
When
NDV
or
VSV
products
which
were
made
during
the
first
10
min
of
the
polymerase
reaction
were
deproteinized
by
sodium
dodecyl
sulfate
and
then
separated
in
sucrose
density
gradients,
a
peak
of
radioactivity
was
observed
at
48S
for
NDV
and
at
38S
for
VSV,
which
coincided
with
the
optical
density
at
260
nm
of
the
virion
RNA
species
and
contained
partially
ribonuclease-resistant
prod-
ucts
(Fig.
4).
The
different
rates
of
sedimenta-
tion
for
the
initial
transcriptive
intermediates
of
NDV
and
VSV
indicate
the
individual
specificity
of
the
two
polymerase
reactions.
These
results
also
indicate
that
the
template
is
not
degraded
during
transcription
for
10
min.
The
data
are
consistent
with
the
model
of
a
large-template
RNA
asso-
ciated
by
base
pairing
with
small
amounts
of
labeled
product.
Bishop
and
Roy
(2),
with
poly-
acrylamide
gel
electrophoresis,
have
also
observed
that
the
initial
VSV
polymerase
products
are
associated
with
the
viral
RNA.
With
increased
time
of
reaction,
the
transcrip-
0
0x
o
x
0
(0
0
0
10
20
TOP
FRACTION
NUMBER
FIG.
4.
Sucrose
gradient
analyses
of
Newcastle
disease
virus
(ND
V)
and
vesicular
stomatitis
virus
(VSV)
polymerase
products
labeledfor
10
min
at
32
C.
Standard
reaction
mixtures
were
made
in
a
volume
of
I
ml,
containing
a
total
of
either
460
jAg
of
ND
V
protein
or
30
jAg
of
VSV
protein.
After
incubation
at
32
C
for
10
min,
sodium
dodecyl
sulfate
was
added
to
a
final
concentration
of
1%-7,,
and
marker,
HeLa
cell
ribosomal
ribonucleic
acid
was
also
added.
Each
reaction
mixture
was
layered
onto
a
linear
15
to
30c%c
sucrose
gradient
in
0.01
M
tris(hydroxymethyl)aminomethane
buffer
(pH
7.4),
0.1
im
NaCl,
0.001
im
ethylenediaminetetraacetic
acid,
and
0.5%O
sodium
dodecyl
sulfate.
Centrifugation
in
a
Beckman
SW27
rotor
with
large
buckets
was
at
42,000
X
gfor
18
hr
at
22
C.
Collection
of
the
gradient
into
individual
fractions,
measurement
of
ultraviolet-
absorbing
material,
and
assay
of
the
acid-insoluble
radioactivity
in
each
fraction
have
been
previously
described
(11).
Prior
to
acid
precipitation,
half
of
each
fraction
was
ribonuclease-digested
as
previously
de-
scribed
(12).
Symbols:
solid
line,
total
absorbancy
at
260
nm;
(0)
acid-precipitable
radioactivity
in
half
of
each
fraction;
(0)
acid-precipitable
radioactivity
in
half
of
each
fraction
after
digestion
by
ribonuclease.
OD260,
optical
density
at
260
,zm.
392
J.
VIROL.