of 7
1112
MICROBIOLOGY:
WAGNER
AND
HUANG
PROC.
N.
A.
S.
19
Brock,
T.
D.,
Science,
129,
960
(1959).
20
Clamp,
J.
R.,
and
L.
Hough,
Biochem.
J.,
94,
17
(1965).
21
Vidaver,
A.
M.
K.,
Ph.D.
dissertation,
Indiana
University,
Bloomington
(1965).
22
Stahl,
E.,
and
U.
Kaltenbach,
J.
Chromatog.,
5,
351
(1961).
23Falcone,
G.,
and
W.
J.
Nickerson,
Science,
124,
272
(1956).
24
Nickerson,
W.
J.,
and
G.
Falcone,
Science,
124,
318
(1956).
25
Ibid.,
p.
722.
26
Lampen,
J.
O.,
in
Function
and
Structure
in
Microorganisms
(Cambridge:
Society
for
General
Microbiology,
1965),
p.
120.
REVERSIBLE
INHIBITION
OF
INTERFERON
SYNTHESIS
BY
PUROMYCIN:
EVIDENCE
FOR
AN
INTERFERON-SPECIFIC
MESSENGER
RNA
*
BY
ROBERT
R.
WAGNER
AND
ALICE
S.
HUANG
DEPARTMENT
OF
MICROBIOLOGY,
JOHNS
HOPKINS
UNIVERSITY
SCHOOL
OF
MEDICINE,
BALTIMORE
Communicated
by
W.
Barry
Wood,
Jr.,
August
24,
1965
The
capacity
to
synthesize
interferon
appears
to
be
a
latent
cellular
function
which
is
induced
by
viral
infection
or
other
stimuli.
The
sequence
of
events
can
be
conveniently
studied
in
suspended
cultures
of
Krebs-2
mouse
ascites
cells
infected
with
Newcastle
disease
virus
(NDV).
Interferon
synthesis
in
this
model
system
begins
3-4
hr
after
infection
and
reaches
maximal
levels
within
20
hr
after
viral
infection.'1
2
Previous
studies
have
shown
that
interferon
formation
requires
unimpaired
cellular
RNA
synthesis
during
the
first
4
hr
after
viral
induction."
2
Following
this
time,
interferon
continues
to
be
formed
even
if
RNA
synthesis
is
inhibited
by
actinomycin.
The
present
experiments
provide
further
support
for
the
hypothesis
that
a
virus
can
switch
on
the
cellular
interferon
gene
by
inducing
the
transcription
of
a
messenger
RNA
that
specifies
the
synthesis
of
the
interferon
protein.
These
two
events
in
the
cycle
of
interferon
production
can
be
dissected
by
suppression
of
protein
synthesis
with
puromycin
and
of
RNA
synthesis
with
actino-
mycin.
Materials
and
Methods.-Cells
and
media:
As
previously
described,
interferon
production
was
studied
in
suspensions
of
Krebs-2
carcinoma
cells
maintained
in
the
ascitic
form
by
weekly
intra-
peritoneal
passage
at
10-s
to
10-6
dilutions
in
adult
Swiss
Webster
mice.
Washed
and
packed
cells
were
prepared
from
ascitic
fluid
harvested
just
prior
to
each
experiment.
Cells
were
sus-
pended
in
medium
199
supplemented
with
2%
calf
serum
to
give
a
final
concentration
of
2
X
106
cells/ml.
Interferon
content
of
cell-free
media
was
assayed
on
monolayer
cultures
of
the
yL
line'
of
L-929
mouse
fibroblasts
which
were
grown
in
medium
199
with
20%
calf
serum.
Interferon-
treated
and
infected
monolayers
of
yL
cells
were
overlayered
with
medium
199
containing
1%
agar
and
no
serum.
Viruses:
Interferon
synthesis
was
induced
by
infecting
packed
Krebs-2
cells
with
a
vaccine
strain
of
Newcastle
disease
virus
(NDV).1
2
The
virus
stock
consisted
of
chick-embryo
allantoic
fluid
with
a
hemagglutinin
titer
of
128
and
an
infectivity
titer
of
approximately
101
plaque-forming
units/ml
when
plated
on
chick
embryo
cells.
In
all
experiments
1
ml
of
NDV
stock
was
mixed
with
2
X
107
packed
Krebs-2
cells
at
370
for
30
min
with
constant
shaking.
This
procedure
re-
sulted
in
greater
than
90%
attachment
of
virus
and
virtually
complete
penetration
as
determined
by
lack
of
elution
after
addition
of
neuraminidase
(Behringwerke,
Marburg,
Germany).
Pre-
treatment
of
cells
with
puromycin
or
actinomycin
did
not
prevent
NDV
attachment
and
penetra-
tion.
provided
that
the
reaction
was
carried
out
at
37°.
Virus
attached
at
40
but
did
not
penetrate.
VOL.
54,
1965
MICROBIOLOGY:
WAGNER
AND
HUANG
1113
Encephalomyocarditis
(EMC)
virus
was
used
as
the
test
virus
for
interferon
assays.
The
plaque-inhibition
method
for
measuring
interferon
content
has
been
described
in
detail.'
2
All
preparations
tested
for
interferon
were
subjected
to
three
1-hr
cycles
of
centrifugation
at
100,000
g
and
dialyzed
overnight
against
0.1
M
HCl.
Neutralized
samples
were
diluted
serially
and
0.2-ml
aliquots
adsorbed
on
duplicate
b-cell
monolayers
for
2
hr
prior
to
challenge
with
100
plaque-
forming
units
(PFU)
of
EMC
virus.
Interferon
titers
were
recorded
as
the
reciprocal
of
the
dilu-
tion
that
reduced
the
number
of
EMC
plaques
to
50%
of
control
counts.
Chemicals:
Puromycin-dihydrochloride
and
the
aminonucleoside
of
puromycin
were
purchased
from
Nutritional
Biochemicals
Corp.,
Cleveland.
Actinomycin
D
(Cl)
was
a
gift
of
Dr.
F.
D.
Lawrason,
Merck,
Sharpe
and
Dohme,
West
Point,
Pennsylvania.
H'-uridine
(specific
activity,
21.5
c/mmole)
and
H3-dl-leucine
(specific
activity,
10.9
c/mmole)
were
obtained
from
Nuclear-
Chicago
Corp.,
Des
Plaines,
Illinois.
Determinations
of
RNA
and
protein
synthesis:
The
procedure
for
measuring
uptake
of
H3-
uridine
into
Krebs-2
cells
by
a
modification
of
the
Baltimore-Franklin
method4
has
been
described
elsewhere.2
After
a
30-min
pulse
of
H3-uridine,
cells
were
chilled,
washed
with
unlabeled
uridine,
disrupted
by
ultrasonic
vibration,
and
their
contents
precipitated
with
0.5
M
cold
perchloric
acid
(PCA).
The
results
were
comparable2
to
those
obtained
by
the
procedure
of
Scherrer
and
Dar-
nell5
for
extracting
RNA
with
phenol
and
sodium
lauryl
sulfate.
Protein
synthesis
was
studied
by
incubating
Krebs-2
cells
for
1
hr
at
370
with
H3-leucine.
The
cells
were
then
washed
with
excess
unlabeled
leucine,
disrupted
by
sonication,
their
contents
precipitated
with
0.5
M
PCA
at
40,
and
the
mixtures
heated
to
900
for
15
min.
The
precipitates
were
washed
twice
with
PCA
and
once
with
ethanol-ether
(1:
1),
dissolved
in
hyamine,
and
radioactivity
was
measured
in
a
Nuclear-
Chicago
scintillation
counter.
Cellular
incorporation
of
both
H3-uridine
and
H3-leucine
was
ex-
pressed
as
cpm/,gg
of
protein
as
determined
by
the
Lowry
method.
The
values
for
H3-leucine
incorporation
were
corrected
for
considerable
nonspecific
absorption.
Results
and
Discussion.-Inhibition
by
puromycin
of
cellular
RNA,
protein,
and
interferon
synthesis:
Nathans6
has
recently
reviewed
current
concepts
of
the
site
and
mechanism
of
puromycin
action.
This
antibiotic
is
a
structural
analogue
of
the
adenosine
end
of
aminoacyl-sRNA
and
is
incorporated
into
growing
peptide
chains.
It
also
inhibits
the
transfer
of
amino
acids
from
sRNA
into
acid-precipitable
polypep-
tides
and
causes
premature
release
from
polyribosomes
of
incomplete
protein
mole-
cules.
Biologically
inactive
proteins
may
be
formed
in
the
presence
of
low
doses
of
pu-
romycin
which
may
not
appreciably
influence
total
amino
acid
incorporation.
Ho
and
Breinig7
have
demonstrated
slight
inhibition
of
interferon
formation
in
chick
embyro
cells
exposed
to
10
iAg/ml
of
puromycin,
a
dose
that
was
found
to
reduce
amino
acid
incorporation
into
protein
by
about
67
per
cent.
We
have
repeated
these
studies
with
Krebs-2
cells
in
order
to
compare
their
susceptibility
and
to
establish
a
baseline
for
later
experiments.
Figure
1
shows
the
effect
of
various
concentrations
of
puromycin
on
protein,
RNA,
and
interferon
synthesis
by
Krebs-2
cells.
Inhibition
of
H'-leucine
incorporation
into
acid-precipitable
protein
of
uninfected
cells
required
more
than
20
ug/ml
of
puromycin.
Much
smaller
doses
caused
marked
inhibition
of
interferon
formation.
As
little
as
1
ug/ml
(1.84
X
10-6
M)
reduced
the
interferon
yield
to
50
per
cent
and
6
ug/ml
to
2
per
cent
of
controls.
We
were
surprised
to
find
that
relatively
low
doses
of
puromycin
inhibited
total
RNA
synthesis
in
Krebs-2
ascites
cells
(Fig.
1).
Rabinovitz
and
Fisher"
reported
that
puromycin
does
not
impair
incorporation
of
adenine
into
ribosomal
or
sRNA
of
Ehrlich
ascites
cells;
doses
of
puromycin
similar
to
those
used
in
our
experiments
were
required
for
90
per
cent
inhibition
of
protein
synthesis.
On
the
other
hand,
Holland9
has
demonstrated
considerable
reduction
in
ribosomal
RNA
synthesis
in
HeLa
cells
exposed
to
100
,ug/ml
of
puromycin.
It
seems
evident
that
different
1114
MICROBIOLOGY:
WAGNER
AND
HUANG
PROC.
N.
A.
S.
>
Ho
FIG.
1.-Inhibition
of
interferon,
total
protein,
100
loo
'"and
RNA
synthesis
by
varying
concentrations
of
5;
_
FLUX
puromycin.
For
the
interferon
studies
Krebs-2
cells
were
infected
with
NDV
(multiplicity
-
5),
<
80
\PROTEIN
washed,
suspended
in
medium-199
to
a
concentra-
z
s\
tion
of
2
X
106
cells/ml,
and
distributed
in
10-ml
N
volumes
in
siliconized
flasks
containing
puromycin
,,
60
-
RNA
at
the
concentrations
shown.
The
medium
was
withdrawn
after
incubation
at
370
for
8
hr,
proc-
z
40-
essed
to
remove
cells,
residual
virus,
and
puromycin,
and
assayed
by
the
plaque-reduction
method.
The
interferon
titers
of
two
control
cultures
were
each
z
20-
INTERFERON
320
units/ml.
Protein
synthesis
was
determined
\A
\
'a
by
measuring
1-hr
incorporation
of
Hs-leucine
(10
pic/ml)
into
the
hot
PCA-precipitable
fraction
of
L
07
'
a
|5
X
106
uninfected
Krebs-2
cells
incubated
in
puro-
5
K)
20
40
l00
mycin-containing
Eagle's
medium
free
of
unlabeled
PUROMYCIN
x
10-6M
leucine.
The
results
are
recorded
as
the
%
specific
activity
of
puromycin-treated
cells
based
on
the
specific
activity
(14.9
cpm/Iug
of
protein
and
2393
cpm/ml)
of
control
cells
incubated
in
the
absence
of
puromycin.
RNA
synthesis
was
deter-
mined
by
measuring
30-min
incorporation
of
H3-uridine
(1.29
,uc/ml)
into
the
cold
PCA-pre-
cipitable
fraction
of
Krebs-2
cells
that
had
been
incubated
for
3
hr
in
puromycin-containing
MM-
S
medium.
The
RNA
results
are
recorded
as
%
uridine
uptake
based
on
control
levels
in
cells
incubated
without
puromycin
of
60.7
cpm/hg
of
protein
and
4682
cpm/ml.
In
a
sepa-
rate
experiment,
9.2
X
10-5
M
puromycin
reduced
RNA
synthesis
in
Krebs-2
cells
by
78%.
vertebrate
cells
vary
considerably
in
their
susceptibility
to
inhibition
by
puromycin
of
protein,
RNA,
and
interferon
synthesis.
Further
studies
with
Krebs-2
cells
were
performed
with
puromycin
doses
of
50
,ug/ml,
9.2
X
10-5
M.
Effect
of
aminonucleoside
of
puromycin:
One
possible
explanation
for
the
results
of
the
foregoing
experiments
is
that
puromycin
inhibits
interferon
synthesis
by
blocking
transcription
of
cellular
RNA
in
a
manner
similar
to
that
of
actinomycin.
Such
an
effect
might
be
anticipated
if
a
cellular
enzyme
can
hydrolyze
off
the
3'
amino-linked
L-O-methyltyrosine
group
of
puromycin
leaving
the
aminonucleoside
residue.
Farnham10
has
shown
that
the
aminonucleoside
of
puromycin
inhibits
RNA
synthesis
in
L
cells,
although
the
effect
is
primarily
on
the
ribosomal
fraction
rather
than
on
messenger
RNA.
Total
RNA
synthesis
in
Krebs-2
cell
was
reduced
by
about
45
per
cent
after
incubation
for
3
hr
with
0.5-2.0
X
10-4
M
aminonucleo-
side.
To
test
the
possibility
that
the
aminonucleoside
is
the
active
puromycin
component
that
inhibits
interferon
synthesis,
NDV-infected
Krebs-2
cells
were
incubated
for
8
hr
in
the
presence
of
10-4
M
concentrations
of
either
the
aminonu-
cleoside
or
of
intact
puromycin.
No
interferon
was
produced
in
the
presence
of
puromycin,
whereas
the
aminonucleoside-treated
cells
produced
80
units/ml
as
did
control
cells.
It
seems
unlikely,
therefore,
that
the
effect
of
puromycin
on
RNA
synthesis
in
Krebs-2
cells
is
responsible
for
its
capacity
to
inhibit
interferon
forma-
tion.
Comparative
effects
of
puromycin
and
actinomycin
on
interferon
synthesis:
If
puro-
mycin
inhibits
interferon
synthesis
primarily
by
virtue
of
its
capacity
to
interfere
with
completion
of
polypeptide
chains,
it
should
be
active
throughout
the
cycle
of
interferon
formation.
If,
on
the
other
hand,
the
action
of
puromycin
resembles
that
of
actinomycin,
it
should
be
less
effective
in
the
late
stages
of
interferon
formation.
To
test
these
alternatives
Krebs-2
cells
were
exposed
to
puromycin
(50
gug/ml)
or
actinomycin
(2
Ag/ml)
at
intervals
after
induction
of
interferon
synthesis
with
NDV.
This
dose
of
actinomycin
reduces
total
cellular
RNA
synthesis
to
about
1
per
cent
of
that
in
control
cells.2
VOL.
54,
1965
MICROBIOLOGY:
WAGNER
AND
HUANG
1115
Figure
2
shows
the
previously
described2
delay
in
appearance
of
interferon
and
the
linear
increase
in
titer
in
control
cultures
not
exposed
to
puromycin
or
actinomycin.
Interferon
synthesis
was
shut
off
when
either
antibiotic
was
added
to
infected
cells
at
2
or
4
hr.
Introduction
of
actinomycin
at
6
or
8
hr
did
not
appreciably
reduce
the
final
yields
of
interferon.
However,
puromycin
inhibited
continuing
interferon
synthesis
regardless
of
when
it
was
added
to
the
cells.
The
final
interferon
titers
in
puromycin-treated
cultures
were
identical
to
the
titers
at
the
time
that
the
antibiotic
was
introduced.
These
data
indicate
that
puromycin
acts
in
this
system
primarily
as
an
inhibitor
of
protein
synthesis,
although
multiple
metabolic
effects
cannot
be
ruled
out
by
these
experiments.
As
shown
by
the
following
studies,
the
interferon-inhibiting
action
of
puromycin
is
reversible,
whereas
the
effect
of
actinomycin
is
not.
Reversal
of
puromycin
inhibition:
Previous
studies2
and
the
preceding
experi-
ments
(Fig.
2)
suggest
that
the
interferon-specific
messenger
RNA
is
stable as
indi-
160
-
140
-
-J
FIG.
2.-Effect
on
final
interferon
2
120
-
yields
of
puromycin
(----0)
and
I
n
actinomycin
(
-)
added
to
cultures
z
l00
l
at
varying
times
after
viral
induction.
v
/..
The
fluids
from
untreated
control
z
o
80-
cultures
(.-*)
of
2
X
107
Krebs-2
W
l
/.
cells
were
withdrawn
at
intervals
L
60-
after
NDV
infection
and
assayed
for
w
II
interferon
content.
At
times
indicated
z
40
by
the
arrows,
puromy
cin
(50
,g/ml)
or
/
~~<
actinomycin
(2
gg/ml)
was
introduced
20
-
into
a
series
of
replicate
flasks
and
their
4
interferon
content
determined
at
20
hr.
<5
-
-
0
2
4
6
8
20
HOURS
AFTER
NDV
INFECTION
cated
by
continuing
synthesis
of
interferon
when
actinomycin
is
added
to
cultures
6-8
hr
after
viral
induction.
If,
as
has
been
assumed,
puromycin
does
not
affect
transcription
of
the
messenger,
the
prediction
can
be
made
that
interferon
synthesis
will
resume
promptly
after
puromycin
is
washed
out
of
inhibited
cells.
Figure
3
shows
the
rate
of
interferon
synthesis
by
NDV-induced
Krebs-2
cells
that
had
been
incubated
for
8
hr
in
media
containing
50
,g/ml
of
puromycin,
then
washed
and
resuspended
in
puromycin-free
media.
After
a
lag
period
of
about
1
hr,
inter-
feron
could
be
detected
in
the
media,
and
increased
in
titer
to
a
peak
at
6
hr.
More-
over,
cells
incubated
with
puromycin
in
the
same
way
and
resuspended
in
media
containing
2
jhg/ml
of
actinomycin
exhibited
equal
capacity
to
resume
interferon
synthesis
on
being
released
from
puromycin
inhibition.
These
data
indicate
that
messenger
RNA
specific
for
interferon
synthesis
accumulates
in
puromycin-blocked
cells
and
is
expressed
when
the
puromycin
is
removed.
The
actinomycin
studies
also
show
that
most
of
the
messenger
was
made
during
incubation
with
puromycin
rather
than
after
the
puromycin
is
washed
out.
It
is
of
additional
interest
that
the
lag
period
for
resumption
of
interferon
syn-
thesis
is
shorter
for
cells
released
from
puromycin
inhibition
than
is
the
lag
period
after
primary
viral
induction
(Fig.
3).
This
finding
may
mean
that
more
copies
of