ArticlePDF Available

Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei

Authors:

Abstract and Figures

We have developed a procedure for preparing extracts from nuclei of human tissue culture cells that directs accurate transcription initiation in vitro from class II promoters. Conditions of extraction and assay have been optimized for maximum activity using the major late promoter of adenovirus 2. The extract also directs accurate transcription initiation from other adenovirus promoters and cellular promoters. The extract also directs accurate transcription initiation from class III promoters (tRNA and Ad 2 VA).
Content may be subject to copyright.
Volume
1
1
Number
5
1983
Nucleic
Acids
Research
Accurate
transcription
initiation
by
RNA
polymerase
II
in
a
soluble
extract
from
isolated
mammalian
nuce
John
David
Dignamr+,
Russell
M.Lebovitz
and
Robert
G.Roeder*
Departments
of
Biological
Chemistry
and
Genetics,
Division
of
Biology
and
Biomedical
Sciences,
Washington
University
School
of
Medicine,
St.
Louis,
MO
63110,
USA
Received
8
November
1982;
Revised
and
Accepted
18
January
1983
ABSTRACT
We
have
developed
a
procedure
for
preparing
extracts
from
nuclei
of
human
tissue
culture
cells
that
directs
accurate
transcription
initiation
in
vitro
from
class
II
promoters.
Conditions
of
extraction
and
assay
have
been
optimized
for
maximum
activity
using
the
major
late
promoter
of
adenovirus
2.
The
extract
also
directs
accurate
transcription
initiation
from
other
adeno-
virus
promoters
and
cellular
promoters.
The
extract
also
directs
accurate
transcription
initiation
from
class
III
promoters
(tRNA
and
Ad
2
VA).
INTRODUCTION
In
recent
years
there
have
been
developed
soluble
cell-free
systems
which
mediate
the
accurate
transcription
of
purified
genes
by
class
I,
II,
and
III
RNA
polymerases
(reviewed
in
1).
These
systems
have
provided
the
means
for
more
definitive
investigations
of
eukaryotic
transcription
mech-
anisms
at
both
the
protein
(RNA
polymerase
and
accessory
transcription
factor)
and
DNA
level.
In
the
case
of
class
II
genes
accurate
transcription
of
puri-
fied
viral
(2)
and
cellular
(3)
genes
was
first
demonstrated
with
a
system
comprised
of
purified
RNA
polymerase
II
and
a
high
speed
supernatant
fraction
(S100)
from
cultured
human
cells.
This
system
has
been
used
to
determine
promoter
sequences
(4-6),
and
in
this
laboratory,
for
the
isolation
of
factors
that
are
necessary
(along
with
RNA
polymerase
II)
for
transcription
from
the
adenovirus
2
major
late
promoter
(7).
A
second
system
consisting
of
a
high
salt
extract
of
whole
cells
(and
containing
endogenous
RNA
polymerase
II)
has
also
been
shown
to
mediate
accurate
transcription
(8)
and
has
been
used
to
analyze
promoter
sequences
(9-12)
and
the
mechanism
of
action
of
a
negative
regulatory
factor
(13).
However,
neither
of
these
systems
takes
advantage
of
the
presumed
nuclear
localization
of
the
transcription
components
since
one
is
obtained
from
a
soluble
post-nuclear
fraction
at
low
ionic
strength
(2)
while
the
other
is
derived
from
a
high
salt
extraction
of
a
whole
cell
homog-
enate
(8).
In
addition,
there
is
no
indication
that
the
conditions
employed
©
I
R
L
Press
Limited,
Oxford,
England.
1475
Nucleic
Acids
Research
for
extraction
in
these
studies
were
optimized
for
the
selective
extraction
of
the
required
components
since
the
extracts
in
each
case
are
prepared
by
methods
which
are
only
slight
modifications
of
earlier
procedures
developed
for
different
purposes
(14,15).
In
the
present
studies,
we
have
developed
a
simple
procedure
for
the
preparation
of
extracts
from
nuclei
and
have
opti-
mized
both
the
conditions
of
extraction
and
assay.
This
method
has
potential
advantages
over
other
procedures
(2,8)
in
that
it
utilizes
the
nuclear
local-
ization
of
the
required
components
to
achieve
a
significant
initial
removal
of
contaminating
cytoplasmic
and
nuclear
components
that
are
not
required
for
the
in
vitro
initiation
of
transcription
and
it
utilizes
conditions
that
have
been
optimized
for
the
extraction
of
the
required
components.
EXPERIMENTAL
PROCEDURES
Assay
Conditions
-
In
the
standard
assay
for
specific
transcription
initia-
tion
up
to
25
pl
of
nuclear
extract
was
assayed
in
a
final
volume
of
50
Pl.
Final
concentrations
for
various
components
(including
those
contributed
by
the
extract)
were:
12
mM
hydroxyethyl
piperazineethanesulfonic
acid
(HEPES),
(pH
7.9
at
25°),
12%
(v/v)
glycerol,
0.3
mM
dithiothreitol
(DTT),
0.12
mM
ethylenediamine
tetraacetate
(EDTA),
60
mM
KCI,
12
mM
MgCl2,
600
vM
each
of
the
three
unlabelled
triphosphates
(ATP,
CTP
and
UTP),
25
VM
a-3
P-GTP
(5
Ci/mmole),
and
1
pg
of
pSmaF
DNA
cleaved
with
SmaI.
The
standard
incubation
was
for
60
min
at
30°C.
Conditions
were
varied
for
other
templates
and
in
various
experiments
as
indicated.
After
extraction
as
described
(2)
the
RNA
samples
were
dissolved
in
98%
formamide
and
electrophoresed
on
4%
acrylamide
gels
containing
7
M
urea;
the
running
buffer
was
0.09
M
borate,
0.09
M
Tris
and
0.01
M
EDTA.
Electrophoresis
was
continued
until
the
xylene
cyanol
tracking
dye
was
1
cm
from
the
bottom
of
the
gel,
after
which
the
gels
were
soaked
for
20
min
in
distilled
water,
dried,
and
subjected
to
autoradiography
as
previously
described
(2).
In
some
experiments
the
band
representing
the
specific
transcript
was
cut
from
the
dried
gel
and
counted
in
toluene
based
scintillation
fluid;
regions
of
equal
area
above
and
below
the
band
were
excised,
counted
and
the
average
of
these
values
was
subtracted
as
a
blank.
Buffers
-
Buffers
used
for
extract
preparation
are
designated
as
follows:
buffer
A
contains
10
mM
HEPES
(pH
7.9
at
4C),
1.5
mM
MgCl2,
10
mM
KC1
and
0.5
mM
DTT;
buffer
B
contains
0.3
M
HEPES
(pH
7.9),
1.4
M
KC1
and
0.03
M
MgCl2;
buffer
C
contains
20
mM
HEPES
(pH
7.9),
25%
(v/v)
glycerol,
0.42
M
NaCl,
1.5
mM
MgCl2,
0.2
mM
EDTA,
0.5
mM
phenylmethylsulfonyl
fluoride
(PMSF)
and
0.5
mM
DTT;
buffer
D
contains
20
mM
HEPES
(pH
7.9),
20%
(v/v)
glycerol,
0.1
M
KCl,
0.2
mM
EDTA,
0.5
mM
PMSF,
and
0.5
mM
DTT.
DTT
and
PMSF
were
added
1476
Nucleic
Acids
Research
fresh
to
the
buffers
just
before
use.
Cells
-
HeLa
cells
(a
line
obtained
from
G.
Attardi,
California
Institute
of
Technology)
were
grown
in
spinner
flasks
at
37°
in
Joklik's
MEM
containing
5%
calf
serum.
They
were
grown
to
4
to
6
x
105
cells
per ml
prior
to
harvesting
for
extract
preparation.
Standard
Procedure
for
Extract
Preparation
-
HeLa
cells
were
harvested
from
cell
culture
media
by
centrifugation
(at
room
temperature)
for
10
min
at
2000
rpm
in
a
Sorvall
HG4L
rotor.
Pelleted
cells
were
then
suspended
in
five
volumes
of
4C
phosphate
buffered
saline
and
collected
by
centrifugation
as
detailed
above;
subsequent
steps
were
performed
at
4°C.
The
cells
were
suspended
in
five
packed
cell
pellet
volumes
of
buffer
A
and
allowed
to
stand
for
10
min.
The
cells
were
collected
by
centrifugation
as
before
and
sus-
pended
in
two
packed
cell
pellet
volumes
(volume
prior
to
the
initial
wash
with
buffer
A)
of
buffer
A
and
lysed
by
10
strokes
of
a
Kontes
all
glass
Dounce
homogenizer
(B
type
pestle).
The
homogenate
was
checked
microscopic-
ally
for
cell
lysis
and
centrifuged
for
10
min
at
2000
rpm
in
a
Sorvall
HG4L
rotor
to
pellet
nuclei.
The
supernatant
was
carefully
decanted,
mixed
with
0.11
volumes
of
buffer
B,
and
centrifuged
for
60
min
at
100,000
g
(Beckman
Type
42
rotor).
The
high
speed
supernatant
from
this
step
was
dialyzed
five
to
eight
hours
against
20
volumes
of
buffer
D
and
is
designated
the
S100
fraction.
The
nuclear
extract
was
prepared
as
follows.
The
pellet
obtained
from
the
low
speed
centrifugation
of
the
homogenate
was
subjected
to
a
second
centrifugation
for
20
min
at
25,000
ga
(Sorvall
SS34
rotor),
to
remove
residual
cytoplasmic
material
and
this
pellet
was
designated
as
crude
nuclei.
These
crude
nuclei
were
resuspended
in
3
ml
of
buffer
C
per
109
cells
with
a
Kontes
all
glass
Dounce
homogenizer
(10
strokes
with
a
type
B
pestle).
The
resulting
suspension
was
stirred
gently
with
a
magnetic
stirring
bar
for
30
min
and
then
centrifuged
for
30
min
at
25,000
g
(Sorval
SS34
rotor).
The
resulting
clear
supernatant
was
dialyzed
against
50
volumes
of
buffer
D
for
five
hours.
The
dialysate
was
centrifuged
at
25,000
g
(Sorvall
SS34
rotor)
for
20
min
and
the
resulting
precipitate
discarded.
The
supernatant,
desig-
nated
the
nuclear
extract,
was
frozen
as
aliquots
in
liquid
nitrogen
and
stored
at
-80°.
The
protein
concentration
was
usually
6
to
8
mg
per
ml
and
15
to
20
mg
of
protein
were
obtained
from
109
cells.
RESULTS
To
assess
the
activity
of
extracts
prepared
under
various
conditions
for
1477
Nucleic
Acids
Research
accurate
transcription
initiation,
we
employed
primarily
the
major
late
promoter
of
adenovirus
2.
When
a
clone
containing
this
viral
promoter
(pSmaF)
is
cleaved
with
the
restriction
enzyme
SmaI,
accurate
transcription
initiation
generates
a
536
nucleotide
run-off
transcript
which
corresponds
in
length
to
the
distance
from
the
transcription
start
site
to
an
SmaI
restric-
tion
site
downstream
from
the
promoter
(2).
DNA
templates
containing
other
viral or
cellular
genes
were
employed
in
a
similar
fashion
by
cleaving
the
DNA
downstream
from
the
promoter
with
an
appropriate
restriction
enzyme.
Salt
Optimum
for
Extraction
-
Since
the
transcription
components
for
RNA
polymerase
II
are
presumably
concentrated
within
the
nucleus,
we
sought
to
take
advantage
of
this
in
our
isolation
procedure.
Hence,
the
first
step
after
cell
lysis
is
the
isolation
of
a
crude
nuclear
fraction.
To
achieve
a
selective
enrichment
of
the
required
transcription
components,
we
examined
several NaCl
concentrations
for
extraction
of
the
nuclei.
The
optimum
NaCl
concentration
was
determined
by
suspending
separate
aliquots
of
crude
nuclei
in
buffer
C
containing
NaCl
concentrations
ranging
from
0.2
M
to
0.5
M.
After
preparation
of
extracts
as
described
in
Methods,
specific
transcription
initiation
assays
were
performed
using
the
Ad
2
major
late
promoter
as
a
template.
As
Figure
1
demonstrates,
extracts
prepared
at
0.2
M
(lanes
1
and
2)
or
0.3
M
NaCl
(lane
3)
were
only
slightly
active
while
extracts
prepared
at
salt
concentrations
of
0.35
(lane
4),
0.42
(lane
5)
or
0.5
(lane
6)
M
NaCl
were
quite
active.
However,
extracts
prepared
at
0.6
M
NaCl
were
substan-
tially
less
active
and
those
prepared
at
0.75
M
and
1.0
M
NaCl
were
inactive
(data
not
shown).
When
nuclei
that
were
previously
used
to
prepare
the
0.2
M
NaCl
extract
(lanes
1
and
2)
were
subsequently
treated
with
0.5
M
NaCl,
the
resulting
extract
was
only
slightly
active
(lanes
7
and
8).
However,
a
combination
of
equal
amounts
of the
primary
0.2
M
and
secondary
0.5
M
NaCl
extracts
was
active
(lane
9),
indicating
that
essential
factors
show
a
differ-
ential
solubility.
Since
the
higher
salt
concentrations
(>0.6
M
NaCl)
ex-
tract
some
inhibitory
material,
we
chose
0.42
M
NaCl
for
subsequent
experi-
ments.
The
optimal
salt
concentration
for
extraction
of
factors
required
for
transcription
of
a
human
H4
histone
gene
(pH4A,
ref.
16)
and
a
mouse
a
globin
gene
(pMGS-l,
ref
3)
was
the
same
as
that
observed
for
the
major
late
promoter
of
adenovirus
2
(data
not
shown).
The
S100
fraction
derived
during
the
cellular
fractionation
used
for
the
preparation
of
the
nuclear
extracts
(see
Materials
and
Methods)
was
inactive
for
the
transcription
of
adenovirus
2
major
late
(see
below).
pH
Optimum
for
Extraction
-
Extracts
were
prepared
from
nuclei
at
the
follow-
1478
Nucleic
Acids Research
1
2
3
4
5
6
7
8
9
Fig.
1.
Transcription
at
the
Ad2
major
late
promoter
with
extracts
prepared
from
nuclei
extracted
at
different
NaCl
concentrations.
Extracts
were
prepared
using
the
standard
procedure
except
that
the
NaCl
concentration
for
extraction
was
varied.
Extracts
were
assayed
under
standard
conditions.
Lanes
1
and
2
show
assays
which
contained,
respectively,
12.5
pl
(50
'g
pro-
tein)
and
25
4l
(100
vg
protein)
of
the
0.2
M
NaCl
extract.
Lanes
3
through
6
show,
respectively,
assays
of
the
0.3
M
NaCl
extract
(25
p4,
120
u'g
pro-
tein),
the
0.42
M
NaCl
extract
(25
il,
140
'g
protein),
and
the
0.5
M
NaCl
extract
(25
4l,
150
ig
protein).
Lanes
7
and
8
show
assays
which
contained,
respectively,
25
pl
(80
vg
protein)
and
12.5
p4
(40
pg
protein)
of
the
0.5
M
NaCl
extract
of
nuclei
previously
extracted
with
0.2
M
NaCl.
Lane
9
shows
an
assay
which
contained
a
mixture
(12.5
p4
each)
of
the
0.2
M
NaCl
extract
and
the
0.5
M
NaCl
extract
of
nuclei
previously
extracted
with
0.2
M
NaCl.
ing
pH
values:
6.5,
7.0,
7.5,
8.0
and
8.5.
The
only
modification
made
was
in
buffer
C
in
which
20
mM
piperazinebis-ethanesulfonic
acid
(PIPES)
was
used
at
pH
6.5
and
7.0
and
20
mM
hydroxyethyl-piperazine-propane
sulfonic
acid
(HEPPS)
was
used
at
pH
8.5;
the
other
conditions
of
extraction
were
as
de-
scribed
in
Materials
and
Methods.
The
pH
optimum
for
extraction
appears
to
be
quite
broad
with
the
extract
prepared
at
pH
8.0
being
only
slightly
more
active
than
the
extracts
prepared
at
other
pH
values
(data
not
shown).
In
our
standard
procedure
we
have
employed
pH
7.9.
Effect
of
Protease
Inhibitors
-
Since
cellular
proteases
could
pose
a
serious
problem
during
the
preparation
of
crude
extracts
and
the
isolation
of
pro-
teins,
we
tested
the
effect
of
several
protease
inhibitors
to
determine
if
they
could
enhance
the
activity
of
the
nuclear
extracts
for
transcription.
1479
Nucleic
Acids
Research
The
compounds
phenylmethylsulfonyl
fluoride
(0.5
mM),
soybean
trypsin
in-
hibitor
(100
pg/ml),
leupeptin
(10
pM),
leustatin
(10
pM),
and
antipain
(10
pM)
were
tested
by
including
these
compounds
at
the
indicated
concentrations
in
the
buffers
used
for
cell
lysis,
extraction
of
the
nuclei
and
dialysis
of
the
extract.
None
of
the
extracts
prepared
with
these inhibitors
was
ap-
preciably
more
active
than
the
control
extract
(data
not
shown).
However,
since
proteases
could
have
effects
that
are
not
immediately
apparent
in
a
crude
extract,
but
which
become
apparent
during
protein
purification,
we
have
routinely
used
PMSF
during
the
preparation
of
the
extracts.
Effect
of
the
Addition
of S100
and
RNA
Polymerase
II
-
Since
components
might
partition
differentially
between
the
S100
and
the
nuclear
extract
we
examined
the
effect
of
combinations
of
these
two
fractions,
prepared
from
the
same
cells,
on
specific
transcription
from
the
major
adenovirus
late
promotor
(Fig.
2).
The
S100
is
inactive
when
assayed
alone
(data
not
shown)
or
in
the
presence
of
purified
calf
thymus
RNA
polymerase
II
(lanes
3
and
4).
As
shown
above,
the
nuclear
extract
is
active
without
additional
factors
(lanes
1
and
1
2
3
4
5
6
Fig.
Effect
of
the
addition
of
SlOO
to
a
nuclear
extract
on
the
transcription
initiation
at
the
Ad2
major
late
promoter.
Extracts
prepared
by
the
standard
procedure
were
assayed
under
standard
conditions.
Lanes
1
and
2
show
assays
containing, respectively,
25
p1l
(150
pg
protein)
and
12.5
p1
(75
pg
protein)
of
nuclear
extract
alone.
Lanes
3
and
4
show
assays
containing,
respectively,
25
P1
(250
pg
protein)
and
12.5
p1
(125
pg
protein)
of
S100
with
100
units
of
added
calf
thymus
RNA
polymerase
II
(ref.
17),
Lane
5
shows
an
assay
containing
25
p1
(150
pg
protein)
of
nuclear
extract
and
S
p1
(50
pg
protein)
of
SlOO.
Lane
6
shows
an
assay
containing
12.5
p1
(75
ug
protein)
of
nuclear
extract
and
12.5
p1
(125
pg
protein)
of
S100.
1480
Nucleic
Acids
Research
2)
and
the
signal
is
not
significantly
enhanced
with
the
addition
of the
S100
fraction
(lanes
5
and
6).
Although
90
percent
of
the
extracts
were
active,
the
addition
of
the
S100
to
approximately
25
percent
of
the
extracts
sup-
pressed
a
background
of
random
transcription
(data
not
shown).
The
addition
of
more
calf
thymus
RNA
polymerase
II
to
the
nuclear
extract
does
not
stimu-
late
specific
transcription
and
serves
only
to
increase
the
background
of
random
transcription
(data
not
shown).
Calf
thymus
RNA
polymerase
II
was
earlier
shown
to
function
in
conjunction
with
human
cell-derived
transcrip-
tion
factors
(2).
Optimum
KC1
and
Mg
Concentrations
-
Using
an
extract
prepared
under
stand-
ard
conditions,
the
optimum
KC1
concentration
for
transcription
from
the
adenovirus
major
late
promoter
was
found
to
be
60
mM
(data
not
shown);
this
is
the
same
as
that
observed
with
the
previously
described
S100
extract
in
the
presence
of
exogenous
RNA
polymerase
II
(2).
Similar
KC1
optima
(60
to
70
mM)
were
observed
when
mouse
a-globin
(pMGS-l)
and
human
H4
histone
(pH4A)
templates
were
employed.
However,
as
shown
in
Figure
3,
the
Mg
optimum
for
Fig.
3.
Mg
optimum
for
transcription
1
2
3
4
5
6
initiation
at
the
Ad2
major
late
-
#
e
promoter.
Extract
(25
pl,
150
iig
protein)
prepared
by
the
standard
procedure
was
assayed
in
duplicate
-~
under
standard
conditions
except
that
Mg++
concentration
was
varied.
Assays
in
lanes
1
through
6
contained,
respectively,
0,
6,
9,
12,
15
and
20
mM
Mg++.
After
autoradiography,
the
-T
lbands
were
cut
from
the
gels
and
counted
as
described
in
Methods.
40/
X-30D
1
-
go,
r
i
;
2
I
JI,
vY
,
_N
1481
Nucleic
Acids
Research
1
2
3
4
5
6
7
Fig.
4.
Temperature
optimum
for
*
transcription
initiation
at
the
Ad2
major
late
promoter.
Extract
(25
p1,
150
'g
protein)
prepared
by
the
standard
procedure
was
assayed
in
duplicate
under
standard
-
a
_
conditions
except
that
the
temperature
was
varied.
Assays
in
lanes
1-7
were
incubated,
respectively,
at
0°,
20°
25,
300
32,
36°
and
40°.
After
autoradiography,
bands
were
cut
from
the
gel
and
counted
as
X
described
in
Methods.
40-
30-
0
b20-
10-
0
20'
30'
40°
TEMPERAT
URE
the
adenovirus
major
late
promoter
(10
to
12
mM)
is
significantly
higher
than
that
previously
reported
for
the
original
S100
extract
(7.5
mM).
While
the
Mg
optima
for
different
DNA
templates
all
fall
between
8
and
12
mM,
they
appear
to
differ
reproducibly
from
one
to
another;
thus
the
human
histone
H4
(pH4A)
and
mouse
8
globin
(pMGS-l)
templates
showed
optima
of
8
mM
and
10
mM,
respectively,
which
differ
reproducibly
(albeit
slightly)
from
the
Mg
optimum
for
the
pSmaF
template.
These
results
indicate
that
the
Mg
optimum
should
be
determined
for
each
template.
Temperature
and
pH
Optima
-
The
pH
and
temperature
optima
were
determined
using
a
standard
extract
assayed
with
the
Ad2
major
late
promoter
template
as
described
in
Materials
and
Methods.
As
shown
in
Figure
4,
there
is
a
sharp
temperature
optimum
at
30°.
As
shown
in
Figure
5,
the
reaction
has
a
broad
pH
optimum
between
7.5
and
8.5.
Template
Concentration
Optima
-
Several
templates
have
been
examined
with
respect
to
optimal
concentrations
for
specific
initiation.
While
most
pro-
moters
are
active
at
DNA
concentrations
between
10
and
20
pg/ml,
different
promoters
exhibit
different
DNA
optima
for
a
given
extract.
In
addition,
the
optimum
for
a
given
template
can
vary
from
extract
to
extract.
Thus,
while
the
basis
for
this
variation
in
template
optima
is
not
entirely
clear,
ex-
1482
Nucleic
Acids
Research
1
2
65
70
75
pH
3
4
5
6
7
.
pH
optimum
for
transcription
IF
qp
ID
0
initiation
at
the
Ad2
major
late
promoter.
Aliquots
of
the
standard
extract
were
dialyzed
against
buffer
D
containing
20
mM
PIPES
(0-0),
20
mM
_
___
_
HEPES
(li-[l),
or
20
mM
HEPPS
(0-O).
The
pH
indicated
is
that
observed
for
the
complete
reaction
mixture
at
30°.
Dialyzed
extracts
(25
i1,
150
ig
protein)
were
assayed
in
duplicate
under
standard
conditions
except
that
pH
was
varied.
Assays
in
lanes
1-7
were
incubated,
respectively,
at
pH
6.5,
7.0,
7.6,
7.5,
8.0,
8.0
and
8.5.
After
autoradiography,
bands
were
cut
from
the
gels
and
counted
as
described
in
Methods.
80
85
tracts
should,
for
optimal
activity,
be
assayed
at
several
template
concen-
trations
to
establish
the
DNA
optimum
both
for
a
specific
template
and
a
given
extract.
In
addition,
when
an
equivalent
amount
of
DNA
(1
ig)
lacking
a
eukaryotic
promoter
(e.g.
PRB322)
is
added
to
the
reaction,
the
amount
of
plasmid
DNA
carrying
Ad2
major
late
promoter
can
be reduced
five-fold
(to
0.2
i'g)
without
changing
the
intensity
of
the
signal;
without
the
addition
of
the
PBR322
DNA,
the
signal
from
0.2
i'g
of
the
major
late
template
is
hardly
detectable
(data
not
shown).
Time
Course
of
Synthesis
and
Stability
of
the
Product
-
The
time
course
of
synthesis
of the
specific
transcript
from
the
Ad2
major
late
promoter
is
shown
in
Figure
6.
Under
optimal
conditions
of
salt,
pH,
and
temperature,
incorporation
of
radioactivity
into
the
specific
run-off
transcript
continues
linearly
for
at
least
50
min
after
a
short
lag.
When
a-amanitin
(at
a
con-
centration
that
specifically
inhibits
RNA
polymerase
II)
is
added
at
30
or
60
min
after
the
start
of
the
reaction
and
the
samples
incubated
an
additional
30
min,
little
or
no
loss
of
radioactivity
in
the
transcript
is
observed
(compare
lanes
8
and
9
with
6
and
7,
respectively).
Transcription
of
Cellular
and
Viral
Templates
-
To
assess
the
utility
of
the
extract
for
transcription
from
different
eukaryotic
promoters
we
examined
several
genes
whose
transcription
had
been
previously
characterized
(see
1483
5
x
u)
Nucleic
Acids
Research
1
2
3
4
5
6
-
r
60-
5-0F
/
o
30-
20
-
*c,
A
'I
7
8
9
Fig.
6.
Time
course
of
snthesis
for
*
*
transcription
initiation
at
the
Ad2
major
late
promoter.
Extract
(25
pl,
150
ug
protein)
prepared
by
the
standard
procedure
was
assayed
in
**
duplicate
under
standard
conditions
except
that
the
incubation
time
was
varied.
Assays
in
lanes
1-7
were
incubated,
respectively,
for
0,
10,
20,
30,
40,
50
and
60
min.
At
30
and
60
min
+
o
A
(lanes
8
and
9,
respectively)
a-amanitin
o
4
A
was
added
at
1
vig/ml
(a
concentration
that
inhibits
RNA
polymerase
II
*
4
completely)
to
parallel
tubes,
and
the
samples
were
incubated
an
additional
30
min.
After
autoradiography,
bands
were
cut
from
the
gel
and
counted
as
described
in
Methods.
2i
40
60O
MIN
u,TES
80
100
refs.
6,
7
and
14).
Figure
7
shows
the
results
of
these
experiments
in
which
the
transcripts
were
analyzed
in
the
standard
electrophoresis
system
(urea
gels).
In
the
cases
examined
the
templates
employed
generated
transcripts
of
the
expected
size.
Thus,
when
pSmaF
is
cleaved
with
SmaI,
the
Ad2
major
late
promoter
directs
the
synthesis
of
an
RNA
(see
arrow)
whose
estimated
size
of
550
nucleotides
is
in
good
agreement
with
the
size
of
536
predicted
from
sequence
analysis
of
this
gene.
When
cleaved
with
Hind
III
pSmaF
also
directs
the
synthesis
of
a
200
nucleotide
RNA
which
is
close
to
the
expected
size
of
197
nucleotides
(data
not
shown).
A
clone
containing
the
adenovirus
EIV
promoter
(Ad2
pEcoRlC)
directs
the
synthesis
of
two
transcripts
when
the
DNA
is
cleaved
with
Hind
III
(lane
2,
Fig.
7).
The
more
prominent
band
of
about
700
nucleotides
(arrow)
corresponds
to
the
transcript
expected
(673
nucleotides)
for
initiation
at
the
EIV
promoter. Two
transcripts
were
observed
with
this
template
in
an
earlier
study
from
this
laboratory
(18),
but
only
the
700
nucleotide
transcript
could
be
identified
as
the
EIV
transcript.
Only
one
transcript
of
approximately
300
nucleotides
(284
nucleotides
is
the
expected
1484
Nucleic
Acids
Research
2
3
4
5
6
1347-
667-
94
@4
-
358-
341-
317-
Fig.
Size
analysis
of
in
vitro
synthesized
transcripts
obtained
with
several
promoters.
The
standard
extract
(25
1il,
150
i'g
protein)
was
assayed
under
standard
conditions
except
that
the
Mg++
concentration
was
changed
as
indicated
for
each
template.
The
assays
shown
contained:
Lane
1,
Ad2
major
late
(pSmaF
cleaved
with
Hind
III),
12
mM
Mg++;
Lane
2,
Ad2
EIV
(pEcoRlC
cleaved
with
Hind
III),
10
nM
Mg++;
Lane
3,
Ad2
EIb
(pHindG,
cleaved
with
Kpn
I)
10
mM
Mg++;
Lane
4,
Ad2
EIII
(pHindH
cleaved
with
Hind
III),
10
mM
MgC12;
Lane
5,
Ad2
EIa
(pHindG
cleaved
with
SmaI),
10
mM
MgC12;
Lane
6,
Ad2
ppIX
(pHindC
cleaved
with
Hph
I),
10
mM
MgC12;
Lane
7,
human
H4
histone
(pH4a
cleaved
with
Hind
III),
8
mM
MgC12.
The
plasmids
employed
are
de-
scribed
elsewhere
(18).
The
arrows
indicate
the
a-amanitin
sensitive
tran-
scripts
discussed
in
the
text.
size)
is
observed
when
this
template
is
cleaved
with
HpaI
(data
not
shown),
an
observation
that
is
in
accord
with
earlier
work
(18).
Figure
7
also
shows
that
appropriately
cleaved
plasmids
(see
Figure
7
legend)
containing
the
adenovirus
2
Elb
(lane
3),
EIII
(lane
4),
Ela
(lane
5),
and
polypeptide
IX
(lane
6)
promoters
generate
transcripts
(indicated
by
arrows)
of
approximately
330,
1050,
540
and
600
nucleotides,
respectively.
The
sizes
of
these
RNAs
are
in
agreement
with
those
of
the
347,
1045,
510
and
604
nucleotide
transcripts
expected
(from
the
sequence
data)
for
accurate
initiation
at
the
respective
promoters
(18).
Transcription
of
a
cleaved
plasmid
containing
a
human
H4
histone
gene
(pH4A,
ref.
13)
is
shown
in
lane
7
of
Figure
7.
The
size
of
the
transcript
generated
(600
nucleotides)
corresponds
to
that
expected
for
accurate
initia-
tion
on
this
gene
(determined
by
S1
mapping
of
in
vivo
RNA,
N.
Heintz
and
R.
Roeder,
unpublished)
and
termination
at
the
downstream
restriction
site.
The
1485
Nucleic
Acids
Research
transcription
of
the
histone
gene
does
not
generate
a
transcript
that
corres-
ponds
to
the
size
of
a
transcript
that
would
result
from
correct
transcription
termination
(approximately
400
nucleotides).
In
other
experiments
not
shown,
the
transcripts
indicated
above
have
been
shown
to
be
sensitive
to
a-amanitin
concentrations
which
selectively
inhibit
RNA
polymerase
II
and
their
sizes
have
been
determined
after
treating
the
RNA
with
glyoxal.
The
observed
and
expected
sizes
were
as
follows:
for
Ad2
major
late
(pSmaF
cleaved
with
SmaI),
570
observed,
536
expected;
for
Ad2
EIV
(pEcoRlC
cleaved
with
Hind
III)
680
observed,
673
expected;
for
Ad2
EIb
(pHindG
cleaved
with
Kpn
I)
350
observed,
347
expected;
for
Ad2
EIII
(pHindH
cleaved
with
Hind
III)
1000
observed,
1050
expected;
for
Ad2
EIa
(pHindG
cleaved
with
SmaI)
520
observed,
510
expected;
for
Ad2
ppIX
(pHindC
cleaved
with
HphI)
630
observed,
600
expected;
for
human
H4
histone
(pH4A
cleaved
with
Hind
III)
600
observed,
approximately
600
expected.
Transcription
of
Class
III
Genes
-
The
nuclear
extract
is
active
for
the
transcription
of
Ad2
VA
and
tRNA
genes,
but
shows
very
little
capacity
for
5S
gene
transcription
(P.
Martin
and
R.
Roeder,
unpublished
observations).
The
absence
of 5S
transcription
apparently
results
from
the
absence
(or
reduced
levels)
of
transcription
factor
IIIA
in
the
nuclear
extract;
thus,
when
partially
purified
factor
IIIA
(23)
is
added
to
the
nuclear
extract,
5S
genes
are
actively
transcribed
(P.
Martin
and
R.
Roeder,
unpublished
observations).
The
S100
fraction
prepared
in
conjunction
with
the
nuclear
extract
is
nearly
as
active
as
the
originally
described S100
fractions
(22)
in
transcribing
the
5S,
tRNA
and
VA
RNA
genes,
indicating
that
it
contains
the
5S
specific
TFIIIA
factor,
as
well
as
substantial
levels
of
the
other
pol
III
factors
(data
not
shown).
DISCUSSION
We
have
described
a
simple
procedure
for
preparing
cultured
cell-derived
extracts
that
are
active
for
the
in
vitro
transcription
of
purified
cellular
and
viral
class
II
genes.
This
procedure
takes
advantage
of
selective
extraction
of
the
transcription
components
from
nuclei
isolated
at
low
ionic
strength.
It
is
noteworthy
that
partial
resolution
of
some
components
can
be
achieved
by
sequential
extraction
of
nuclei
with
buffers containing
increasing
NaCl
concentrations
(see
Fig.
1).
This
observation
is
in
accord
with
an
earlier
report
from
our
laboratory
(7)
that
the
transcription
of
the
adeno-
virus
major
late
promoter
requires
multiple
components
in
addition
to
RNA
polymerase
II.
The
primary
advantage
of
this
nuclear
extract
is
that
it
1486
Nucleic
Acids
Research
achieves
a
substantial
separation
of
the
required
transcription
components
from
contaminating
cytoplasmic
and
nuclear
material
but,
as
with
whole
cell
extracts
(8),
still
contains
endogenous
RNA
polymerase
II.
While
we
have not
completed
extensive
mapping
studies
of
all
the
tran-
scripts
generated
by
the
various
class
II
gene
templates
analyzed
in
this
system,
the
sizes
of
the
run-off
transcripts
are
in
each
case
strongly
indicative
of
accurate
initiation
events.
In
the
case
of
the
adenovirus
2
major
late
promoter
(pSmaF
template)
and
the
mouse
B-chain
promoter
(pMGS-1
template,
ref.
3)
the
5'
termini
of
the
in
vitro
transcripts
have been
shown
to
be
indistinguishable
from
those
of
the
corresponding
in
vivo
transcripts
when
compared
by
primer
extension
analysis
with
reverse
transcriptase
(ref.
19
and
D.
Luse,
personal
communication).
In
addition,
transcripts
generated
in
the
present
system
with
a
plasmid
containing
the
adenovirus
2
EIIA-early
pro-
moter
appear,
by
Sl
nuclease
mapping,
to
have
5'
termini
identical
to
those
of
the
corresponding
in
vivo
RNAs
(D.
H.
Huang
and
R.
G.
Roeder).
Moreover,
while
transcription
from
the
EIlA-early
promoter
is
difficult
to
detect
with
a
template
containing
both
the
EIIA-early
and
EIII
promoters
(ref.
18
and
Figure
7),
a
substantial
and
readily
detectable
level
of
accurate
initiation
is
observed
with
a
template
containing
only
the
EIIA
promoter
(D.
H.
Huang
and
R.
G.
Roeder,
unpublished
observation).
This
may
be
of
significance
in
view
of
the
fact
that
the
major
EILA-early
promoter
does
not
contain
a
canonical
TATA
box
and
that
this
promoter
is
recognized
only
at
a
very
low
level
in
the
other
polymerase
II
transcription
systems
(20,21).
Thus,
the
nuclear
system
could
be
more
active
in
recognizing
a
broader
group
of
class
II
promoters
with
differing
requirements
for
polymerase
II
factors,
although
this
important
point
remains
to
be
further
investigated.
Transcription
of
class
II
genes
in
vitro
has
proved
to
be
quite
inefficient
with
respect
to
the
number
of
transcripts
synthesized
per
DNA
template
(2,8).
Under
our
standard
conditions
of
assay,
only
about
0.03
transcripts
are
synthe-
sized
per
template
(on
average)
when
the
adenovirus
2
major
late
gene
is
employed,
a
result
that
is
in
accord
with
the
observations
of
others
(2,8).
However,
if
vector
DNA
(pBR322)
lacking
eukaryotic
promoters
is
added
at
the
standard
DNA
concentration
(20
jg/mi),
a
five-fold
reduction
in
the
concentra-
tion
of
the
major
late
promoter
gives
the
same
signal
as
when
the
major
late
template
is
used
alone
at
its
optimal
concentration;
thus,
while
the
amount
of
specific
transcript
does
not
increase,
the
efficiency
with
which
the
template
is
transcribed
can
be
increased
five-fold.
It
is
also
noteworthy
that
our
transcription
experiments
are
usually
performed
at
GTP
concentrations
1487
Nucleic
Acids
Research
that
are
considerably
below
the
Km
of
RNA
polymerase
II
for
this
nucleotide
(40
PM
to
60
PM).
In
accord
with
this
fact
we
have
observed
that
increasing
the
GTP concentration
from
25
PM
to
100
PM
increases
transcripton
of
the
adenovirus
2
major
late
promoter
about
four-fold.
Thus,
while
in
vitro
transcription
systems
for
class
II
genes
appear
to
be
inefficient,
this
problem
may
be
partly
overcome
by
altering
the
conditions
of
assay.
The
nuclear
extract
described
here
also
supports
the
accurate
transcrip-
tion
of
tRNA
and
adenovirus
VA
genes,
but
is
inactive
for
5S
gene
transcrip-
tion
unless
the
S100
fraction
or
a
partially
purified
preparation
of
TFIIIA
is
added.
While
the
S100
obtained
during
the
preparation
of
the
nuclear
extract
is
active
in
tRNA
and
VA
transcription,
it
appears
that
a
significant
fraction
of
the
activity
for
these
genes
remains
in
the
nuclear
extract.
Although
we
do
not as
yet
understand
the
basis
for
the
partitioning
of
the
5S
gene-specific
component
TFIIIA
into
the
S100
fraction,
it
should
prove
advan-
tageous
for
the
purification
of
this
protein
from
mammalian
cells.
In
conclusion,
the
nuclear-derived
transcription
system
described
here
mediates
the
accurate
transcription
of
a
broad
spectrum
of
class
II
(and
class
III)
genes
and
appears
promising
as
a
system
for
the
further
analysis
of
eukaryotic
transcription
controls,
including
the
isolation
and
characteriza-
tion
of
various
transcription
factors.
Acknowledgements
These
studies
were
supported
by
Public
Health
Grants
CA
16640
and
CA
23615
(awarded
by
the
National
Cancer
Institute)
and
by
an
American
Cancer
Society
Grant
NP-284
to
R.G.R.
J.D.D.
was
supported
by
Public
Health
Service
Fellowship
Award
GM
06981
(Awarded
by
the
National
Institute
of
General
Medical
Sciences)
and
R.L.
was
supported
in
part
by
National
Institutes
of
Health
Research
Service
Award
GM
07200,
Medical
Scientist
(awarded
by
the
National
Institute
of
General
Medical
Scientists).
We
also
thank
Rene
Kanan
for
excellent
technical
assistance.
*To
whom
correspondence
should
be
addressed.
+Permanent
address:
Department
of
Biochemistry,
University
of
Mississippi
Medical
Center,
Jackson,
MS
39216,
USA
REFERENCES
1.
Heintz,
N.
and
Roeder,
R.G.
in
Genetic
Engineering,
Vol.
4,
J.K.
Setlow
and
A.
Hollaender
Eds.
Plenum
Press,
New
York,
pp.
57-89
1488
Nucleic
Acids
Research
2.
Weil,
P.A.,
Luse,
D.S.,
Segall,
J.
and
Roeder,
R.G.
(1979)
Cell
18,
469-484
3.
Luse,
D.S.
and
Roeder,
R.G.
(1980)
Cell
20,
641-699
4.
Corden,
J.,
Wasylyk,
B.,
Buckwalder,
A.,
Sassone-Corsi,
P.,
Kedinger,
C.
and
Chambon,
P.
(1980)
Science
209,
1406-1414
5.
Wasylyk,
B.,
Kedinger,
C.,
Corden,
J.,
Brison,
0.
and
Chambon,
P.
(1980)
Nature
285,
367-373
6.
Lee,
D.C.,
Roeder,
R.G.
and
Wold,
W.S.M.
(1982)
Proc.
Natl.
Acad.
Sci.
USA
79,
41-45
7.
Matsui,
T.,
Segall,
J.,
Weil,
P.A.
and
Roeder,
R.G.
(1980)
J.
Biol.
Chem.
255,
11992-11996
8.
Manley,
J.L.,
Fire,
A.,
Cano,
A.,
Sharp,
P.A.
and
Gefter,
M.L.
(1980)
Proc.
Natl.
Acad.