Vol.
14,
No.
1
MOLECULAR
AND
CELLULAR
BIOLOGY,
Jan.
1994,
p.
207-213
0270-7306/94/$04.00+0
Copyright
X
1994,
American
Society
for
Microbiology
Identification
of
a
Thymidylate
Synthase
Ribonucleoprotein
Complex
in
Human
Colon
Cancer
Cells
EDWARD
CHU,`*
DONNA
M.
VOELLER,1
KRISTEN
L.
JONES,'
TEIJI
TAKECHI,1
GLADYS
F.
MALEY,2
FRANK
MALEY,2
SHOSHANA
SEGAL,'
AND
CARMEN
J.
ALLEGRA'
NCI-Navy
Medical
Oncology
Branch,
National
Cancer
Institute,
Bethesda,
Maryland
20892,1
and
Wadsworth
Center
for
Laboratories
and
Research,
New
York
State
Department
of
Health,
Albany,
New
York
122012
Received
12
May
1993/Returned
for
modification
28
June
1993/Accepted
28
September
1993
Translation
of
thymidylate
synthase
(TS)
mRNA
is
controlled
by
its
own
protein
product,
TS,
in
an
autoregulatory
manner.
Direct
binding
of
TS
protein
to
two
different
cis-acting
elements
on
the
TS
mRNA
is
associated
with
this
translational
regulation.
In
this
study,
an
immunoprecipitation-reverse
transcription-PCR
technique
was
used
to
identify
a
TS
ribonucleoprotein
(RNP)
complex
in
cultured
human
colon
cancer
cells.
Using
antibodies
specific
for
TS
protein,
we
show
that
TS
is
complexed
in
vivo
with
its
own
TS
RNA.
Furthermore,
evidence
demonstrating
a
direct
interaction
between
the
mRNA
of
the
nuclear
oncogene
c-myc
and
TS
protein
is
presented.
Recently,
there
has
been
an
increased
interest
in
the
characterization
of
translational
regulatory
mechanisms.
There
are
a
number
of
eukaryotic
mRNAs
whose
expression
is
controlled
at
the
level
of
translation
(18).
The
regulated
synthesis
of
the
iron
storage
protein
ferritin
by
iron
repre-
sents
one
of
the
best-studied
examples
of
this
form
of
regulation
(25).
A
stem-loop
structure
located
within
the
5'
untranslated
region
(UTR)
of
ferritin
mRNA,
termed
the
iron-responsive
element
(IRE),
represents
the
cis-acting
element
to
which
the
IRE-binding
protein
binds
(16,
23).
Recent
studies
have
demonstrated
that
the
redox
state
in
the
cell
is
an
important
determinant
of
the
binding
affinity
of
this
protein
to
the
IRE
(15,
17).
Using
an
in
vitro
translation
system,
we
showed
that
translation
of
human
thymidylate
synthase
(TS)
mRNA
is
regulated
by
its
own
protein
product,
TS,
in
a
negative
autoregulatory
manner
(7).
Although
translational
autoregu-
lation
has
been
described
in
prokaryotic
systems
(2,
5),
TS
mRNA
represents
the
first
eukaryotic
mRNA
whose
regula-
tion
is
controlled
in
such
a
fashion.
Furthermore,
we
dem-
onstrated
that
incubation
of
TS
protein
with
either
the
nucleotide
substrate
dUMP
or
the
inhibitor
5-fluoro-dUMP
repressed
its
inhibitory
effect
on
TS
mRNA
translation.
These
results
suggest
that
either
the
native
conformational
state
or
direct
occupancy
of
the
TS
enzyme
active
site
is
the
critical
factor
determining
the
TS
protein-TS
mRNA
inter-
action.
The
TS
protein-inhibitory
effect
on
TS
mRNA
trans-
lation
is
associated
with
direct
binding
of
TS
protein
to
at
least
two
specific
regions
on
its
corresponding
mRNA.
One
binding
site
contains
a
putative
stem-loop
structure
that
incorporates
the
translational
start
site,
while
the
second
site
is
contained
within
a
200-nucleotide
(nt)
sequence
of
the
protein-coding
region
(10).
Recent
in
vitro
studies
from
this
laboratory
demonstrated
that
short-term
exposure
of
human
colon
cancer
H630
cells
to
the
antineoplastic
agent
5-flu-
orouracil
(5-FU)
results
in
an
acute
increase
in
the
expres-
sion
of
TS
protein
that
is
due
to
an
enhanced
translational
efficiency
of
TS
mRNA
(8).
While
the
expression
of
TS
*
Corresponding
author.
Mailing
address:
National
Cancer
Insti-
tute,
NCI-Navy
Medical
Oncology
Branch,
Bethesda
Naval
Hospi-
tal,
Bldg.
8,
Rm.
5101,
Bethesda,
MD
20889.
Phone:
(301)
402-1841.
Fax:
(301)
496-0047.
during
the
cell
cycle
is
primarily
regulated
at
the
transcrip-
tional
level
(1,
19,
30),
there
is
now
recent
evidence
suggest-
ing
control
at
the
level
of
translation
(22).
These
findings,
taken
together,
offer
supportive
evidence
for
the
model
of
TS
translational
autoregulation.
The
purpose
of
the
present
study
was
to
identify
a
TS
ribonucleoprotein
(RNP)
complex
in
a
cultured
cell
system.
With
the
use
of
specific
antisera
to
TS,
we
show
that
TS
protein
is
complexed
with
its
corresponding
TS
RNA
in
human
colon
cancer
cells.
In
addition,
we
present
evidence
demonstrating
a
specific
interaction
between
the
mRNA
of
the
nuclear
oncogene
c-myc
and
TS
protein.
MATERIALS
AND
METHODS
Cell
culture.
The
characteristics
of
the
human
colon
can-
cer
cell
line
H630
have
been
previously
described
(32).
The
resistant
H630-R1O
subline
was
selected
in
vitro
for
resis-
tance
to
5-FU
by
exposure
of
the
parent
H630
cell
line
to
stepwise
increases
in
5-FU
and
was
maintained
in
medium
containing
10
p,M
5-FU
(20).
Cell
lines
were
grown
in
75-cm2
plastic
tissue
culture
flasks
(Falcon
Labware,
Oxnard,
Calif.)
in
growth
medium
containing
RPMI
1640
with
10%
dialyzed
fetal
bovine
serum
and
2
mM
glutamine.
Dialyzed
fetal
bovine
serum
and
all
other
medium
components
were
ob-
tained
from
GIBCO
(Grand
Island,
N.Y.).
Whole-cell
extraction.
Whole-cell
extracts
were
prepared
as
described
by
Lerner
and
Steitz
(27)
and
Steitz
(34).
In
brief,
human
colon
cancer
cells
were
washed
three
times
with
ice-cold
phosphate-buffered
saline
and
harvested
from
150-cm2
plastic
tissue
culture
flasks
with
a
rubber
policeman.
Cells
were
resuspended
in
1
ml
of
NET-2
buffer
(50
mM
Tris-HCl
[pH
7.4],
150
mM
NaCl,
0.05%
[vol/vol]
Nonidet
P-40)
(32)
and
ruptured
by
sonication
with
three
5-s
bursts
with
a
Branson
Sonifier
(setting
2).
The
homogenate
was
centrifuged
at
14,000
x
g
for
10
min,
and
the
whole-cell
supernatant
was
used
as
the
source
of
antigen.
Immunoprecipitation
of
RNP
complex.
Immunoprecipita-
tion
of
RNPs
was
performed
as
described
by
Steitz
(34).
The
whole-cell
extract
(1.5
mg)
was
first
cleared
of
nonspecific
binding
material
by
incubation
with
300
,ul
of
Pansorbin
(GIBCO
BRL,
Gaithersburg,
Md.)
for
30
min
on
ice
followed
by
centrifugation
to
remove
the
Pansorbin.
The
cleared
207
208
CHU
ET
AL.
extract
was
then
incubated
with
the
appropriate
antibody
for
30
min,
to
which
Pansorbin
(300
RI),
yeast
tRNA
(35
p,g;
U.S.
Biochemical,
Cleveland,
Ohio),
and
Inhibitase
(20
U;
5
Prime-3
Prime,
Boulder,
Colo.)
were
added
for
an
additional
30
min.
The
Pansorbin-immune
complex
precipitates
were
centrifuged
at
14,000
x
g,
4°C
for
3
min,
and
then
washed
four
times
with
350
plI
of
NET-2
buffer.
After
addition
of
300
pI
of
NET-2
buffer,
the
immunoprecipitate
pellets
were
subjected
to
phenol-chloroform
extraction.
The
RNA
frac-
tion
of
the
RNP
complex
was
isolated
by
ethanol
precipita-
tion
and
was
immediately
used
in
the
reverse
transcription
(RT)
reaction.
RT
of
RNA.
Using
the
Stratagene
first-strand
cDNA
synthesis
protocol
(Stratagene,
San
Diego,
Calif.),
we
incu-
bated
the
entire
immunoprecipitated
RNA
sample
with
ran-
dom
primers
(300
ng)
for
5
min
at
65°C.
The
mixture
was
allowed
to
cool
at
room
temperature
for
15
min,
and
5
pl
of
transcription
buffer,
5
,ul
of
0.1
mM
dithiothreitol,
1
pI
of
RNase
block,
3
pl
of
a
25
mM
deoxynucleotide
triphosphate
solution,
and
1
,ul
of
Moloney
murine
leukemia
virus
reverse
transcriptase
(20
U/pI)
were
then
added.
The
reaction
was
incubated
for
1
h
at
37°C,
and
the
solution
was
then
stored
at
-200C.
PCR
conditions.
The
primers
were
synthesized
on
an
Applied
Biosystems
model
391
DNA
synthesizer.
Their
sequences
are
as
follows
(the
underlined
bases
are
not
part
of
the
target
gene
sequence
and
represent
unique
restriction
enzyme
digestion
sites):
DHFR-1,
GGATCCCGCTGCTGT
CATGGTTGGTT
(sense);
DHFR-2,
AAGCTTACTT`17TCT
AATGTAAAAAT
(antisense);
TS-1,
GAGCTCCCGAGAC
ITI'-TIGGACAGCC
(sense);
TS-2,
AAGCTTAAGAATCC
TGAGCTTTGGGA
(antisense);
c-myc-1,
GGCGAACACA
CAACGTCTTGGAG
(sense);
c-myc-2,
GCTCAGGACAT
TCTGTTAGAAGG
(antisense);
max-1,
CCGTAGGAAAT
GAGCGATAA
(sense);
and
max-2,
AGTGGCTFTAGCTGG
CCTCCAT
(antisense).
The
single-stranded
cDNA
obtained
from
the
first-strand
synthesis
reaction
was
used
as
the
template
for
amplifica-
tion.
The
reaction
conditions
were
those
outlined
by
the
Perkin-Elmer
protocol
(Perkin-Elmer
Cetus,
Emeryville,
Calif.),
and
the
total
volume
of
the
reaction
was
100
pA.
Mineral
oil
(40
IlI)
was
placed
on
top
of
the
aqueous
solution.
Reactions
were
cycled
in
a
Perkin-Elmer
Cetus
thermal
cycler,
and
samples
were
incubated
at
97°C
for
1
min,
620C
for
1
min,
and
72°C
for
1
min
for
40
cycles.
At
the
end
of
the
40th
cycle,
the
reactions
were
incubated
for
an
additional
10
min
at
72°C
then
cooled
to
4°C.
DNA
analysis.
RT-PCR
products
were
analyzed
by
frac-
tionation
on
a
1%
nondenaturing
agarose
gel
and
then
transferred
to
a
Nytran
filter
membrane
(Schleicher
&
Schuell,
Keene,
N.H.).
The
membrane
was
UV
cross-linked
for
2
min
at
254
nm,
using
the
Stratagene
UV
cross-linker.
The
membrane
was
prehybridized
for
4
h
at
42°C
in
50%
formamide-5x
SSC
(lx
SSC
is
0.15
M
NaCl
plus
0.015
M
sodium
citrate)-5x
Denhardt's
solution-20
mM
Na2HPO4-
NaH2PO4
(pH
7.4)-200
p,g
of
salmon
sperm
DNA
per
ml-0.1%
sodium
dodecyl
sulfate
and
then
hybridized
for
24
h
with,
per
ml,
106
cpm
of
32P-labeled
TS
cDNA
probe
that
was
synthesized
according
to
the
random
primer
method
of
Feinberg
and
Vogelstein
(14).
Hybridized
filters
were
washed
as
previously
described
(6)
and
then
subjected
to
autoradiography
using
Kodak
XAR-5
film.
Isolation
and
analysis
of
total
RNA.
Total
RNA
was
iso-
lated
from
human
colon
cancer
H630
and
H630-R1O
cells
as
previously
described
(6).
For
Northern
(RNA)
blot
hybrid-
ization
analysis,
total
cellular
RNA
(20
p,g
per
sample)
was
o
Ir
CI~)
CVI)
(
D
I
I
28S
-
18S-
1J
0
/--Acti
n
FIG.
1.
Northern
blot
analysis
of
TS
mRNA
in
parent
H630
and
resistant
H630-R1O
cells.
Total
RNA
(20
p,g)
from
each
cell
line
was
fractionated
on
a
1%
formaldehyde-agarose
gel,
transferred
to
a
Nytran
filter
membrane,
and
hybridized
with
a
32P-radiolabeled
1.5-kb
TS
cDNA
insert
as
previously
described
(6).
The
filters
were
then
stripped
of
the
TS
probe
and
rehybridized
with
a
human
0-actin
probe
to
control
for
loading
and
integrity
of
mRNA.
Quantitation
of
signal
intensities
was
performed
by
densitometry
on
a
ScanJet
Plus
scanner
(Hewlett-Packard)
and
analyzed
by
using
NIH
Image
1.36
software
(Wayne
Rasband,
National
Institute
of
Mental
Health,
Bethesda,
Md.).
denatured,
fractionated
on
a
1%
formaldehyde-agarose
gel,
and
then
transferred
to
a
Nytran
filter
membrane
(Schleicher
&
Schuell).
The
1.5-kb
TS
cDNA,
0.65-kb
DHFR
cDNA,
1.4-kb
c-myc
insert,
and
0.5-kb
max
insert
were
used
as
hybridization
probes
after
each
was
labeled
with
[32P]dCTP
(3,000
Ci/mmol)
by
the
random
primer
labeling
method
of
Feinberg
and
Vogelstein
(14).
In
vitro
mRNA
transcription.
Full-length
c-myc
RNA
transcript
was
synthesized
with
SP6
RNA
polymerase
after
linearization
with
HpaI
according
to
the
Promega
protocol
(9).
All
mRNA
transcripts
were
evaluated
on
a
1%
formal-
dehyde-agarose
gel
to
verify
their
integrity
and
size.
The
concentration
of
unlabeled
RNA
was
determined
by
spec-
trophotometry.
Labeled
RNA
transcript
was
made
by
inclu-
sion
of
[a-32P]CTP
at
200
Ci/mmol.
RNA-protein
binding
assay.
Gel
mobility
shift
assays
were
performed
as
previously
described
(10,
25).
Samples
were
electrophoresed
in
a
nondenaturing
4%
polyacrylamide
gel
(acrylamide-N,N'-methylenebisacrylamide;
weight
ratio,
60/1),
dried,
and
then
visualized
by
autoradiography.
Com-
petition
experiments
were
performed
with
human
recombi-
nant
TS
protein
(300
pmol)
and
32P-labeled
c-myc
RNA
(2.2
fmol,
100,000
cpm).
These
conditions
were
selected
on
the
basis
of
control
experiments
using
a
fixed
amount
of
c-myc
RNA
with
increasing
TS
protein
concentrations
to
determine
linearity
of
binding.
Unlabeled
competitor
RNAs
were
mixed
with
labeled
probe
prior
to
addition
of
TS
protein.
RESULTS
AND
DISCUSSION
The
human
colon
cancer
H630-R1O
cell
line
overexpresses
TS
protein
by
32-fold
compared
with
the
parental
H630
cell
line
(20).
As
determined
by
Northern
blot
and
scanning
densitometric
analysis,
the
level
of
TS
mRNA
expression
was
approximately
30-fold
greater
in
the
H630-R1O
line
than
in
the
parental
H630
line
(Fig.
1).
In
our
initial
attempts
to
isolate
TS
RNP
complexes
from
the
human
colon
cancer
H630-R1O
cell
line,
we
used
the
immunoprecipitation
tech-
nique
described
by
Lerner
and
Steitz
(27).
H630-R1O
cells
MOL.
CELL.
BIOL.
TS
RNP
COMPLEX
IN
HUMAN
COLON
CANCER
CELLS
209
A
BP
872
-
603
-
310
-
1
2
3
4
5
6
7 8
9
B
,
C
434-454
ATG
II
l
V
VI
TAG
1052
9159539
505
bu
FIG.
2.
Analysis
of
TS
RNA
immunoprecipitated
from
human
colon
cancer
H630-R10
cells.
(A)
Ethidium
bromide
stain.
(B)
Hybridization
analysis.
Whole-cell
extracts
were
immunoprecipi-
tated
with
either
TS
polyclonal
antibody
(lanes
1,
2,
and
4),
mouse
monoclonal
TS
antibody
(lane
3),
no
antibody
(lane
6),
rabbit
immunoglobulin
G
(lane
7),
or
anti-a-tubulin
monoclonal
antibody
(lane
8),
treated
with
(lanes
2
to
8)
or
without
(lane
1)
reverse
transcriptase,
and
amplified
by
using
the
TS-specific
primers
shown
in
panel
C.
Lane
4
represents
immunoprecipitated
RNA
that
was
treated
with
RNase
A
prior
to
reverse
transcription;
lane
5
repre-
sents
whole-cell
extract
deproteinized
by
phenol-chloroform
extrac-
tion
prior
to
immunoprecipitation
with
TS
polyclonal
antibody.
Full-length
TS
cDNA
was
subject
to
PCR
amplification
using
the
same
TS-specific
primers
(lane
9).
The
PCR-amplified
products
resolve
at
a
position
corresponding
to
505
nt.
(C)
TS
gene,
primer
locations,
and
PCR
amplification
product.
Annealing
sites
for
prim-
ers
used
for
PCR
are
shown
in
relation
to
their
position
on
the
TS
gene.
Translational
start
and
stop
sites
are
identified,
as
are
the
positions
of
the
intervening
intron
sequences
(indicated
by
roman
numerals)
(21).
were
32P
radiolabeled
for
24
h,
whole-cell
extracts
were
prepared,
and
immunoprecipitation
with
TS
polyclonal
anti-
body
(8)
was
then
performed.
Fractionation
of
immunopre-
cipitates
on
a
10%
polyacrylamide-7
M
urea
gel
failed
to
detect
discrete
RNA
complexes
(data
not
shown).
As
a
result,
this
immunoprecipitation
technique
was
modified
in
the
present
study
by
coupling
it
to
an
RT-PCR-based
method
to
enhance
the
detection
of
potential
RNP
complexes.
Immunoprecipitation
of
H630-R1O
cell
extracts
with
either
a
polyclonal
antibody
or
a
murine
monoclonal
antibody
to
human
TS
followed
by
reverse
transcription
and
PCR
am-
plification
with
TS-specific
primers
gave
rise
to
a
505-nt
fragment
(Fig.
2A,
lanes
2
and
3).
This
band
resolved
at
the
same
position
as
control
TS
DNA
obtained
from
an
identical
PCR
amplification
using
the
full-length
TS
cDNA
as
the
template
DNA
(7)
(Fig.
2A
and
B,
lane
9).
Hybridization
of
the
transferred
gel
onto
a
nitrocellulose
filter
with
full-length
(1,524-nt)
32P-labeled
TS
cDNA
confirmed
that
the
band
resolving
at
505
nt
was
specific
for
TS
mRNA
(Fig.
2B,
lanes
2,
3,
and
9).
Since
the
primers
used
for
PCR
amplification
anneal
to
exon
regions
that
are
separated
by
four
intervening
intronic
sequences
corresponding
to
9,692
nt
(Fig.
2C),
the
TS
amplified
products
are
not
the
result
of
DNA
contamina-
tion
of
the
immunoprecipitated
samples.
As
further
proof
that
the
origin
of
this
band
was
RNA
bound
to
TS
protein,
we
performed
experiments
in
which
RNAs
immunoprecipi-
od
cc
CD
X
CY)
CY)
4TS
FIG.
3.
Analysis
of
immunoprecipitated
TS
RNA
from
parent
H630
and
resistant
H630-R10
human
colon
cancer
cells.
TS
RNA
was
precipitated
by
TS
polyclonal
antibody
from
parent
H630
and
H630-R1O
human
colon
cancer
cell
extracts,
reverse
transcribed,
and
PCR
amplified
with
TS-specific
primers.
tated
with
TS
polyclonal
antibody
either
were
directly
PCR
amplified
without
RT
(Fig.
2A
and
B,
lane
1)
or
were
treated
with
RNase
A
prior
to
RT
(Fig.
2A
and
B,
lane
4).
In
each
case,
there
was
complete
absence
of
the 505-nt
product.
In
contrast,
this
PCR
product
was
observed
when
immunopre-
cipitated
samples
were
treated
with
DNase
prior to
RT-PCR
(data
not
shown).
These
findings,
taken
together,
provide
further
evidence
that
RNA
represents
the
nucleic
acid
of
origin
for
these
PCR
products.
When
whole-cell
supematant
extracts
were
subjected
to
deproteinization
by
phenol-chlo-
roform
extraction
prior
to
immunoprecipitation
with
TS
polyclonal
antibody,
no
PCR
product
was
observed
(Fig.
2A
and
B,
lane
5).
This
result
demonstrates
that
purified
RNA
was
not
recognized
by
anti-TS
antibody,
suggesting
that
an
intact
TS
RNP
complex
was
a
critical
requirement
for
the
immunoprecipitation
procedure.
To
further
examine
the
specificity
of
antibody
recognition
in
the
immunoprecipita-
tion
reaction,
either
no
antibody
(Fig.
2A
and
B,
lane
6)
or
an
unrelated
antibody
(rabbit
immunoglobulin
[Fig.
2A
and
B,
lane
7]
or
anti-a-tubulin
mouse
monoclonal
antibody
[Fig.
2A
and
B,
lane
8])
was
used.
None
of
these
controls
gave
rise
to
the
505-nt
TS
mRNA
band.
In
these
initial
studies,
the
TS-overexpressing
H630-R1O
human
colon
cancer
cell
line
was
used.
We
subsequently
isolated
whole-cell
extracts
from
the
parental
H630
cell
line
and
performed
an
identical
immunoprecipitation-RT-PCR
analysis.
As
shown
in
Fig.
3,
TS
RNA
was
precipitated
by
TS
polyclonal
antibody
from
H630
whole-cell
extracts,
a
finding
that
confirmed
the
presence
of
the
TS
RNP
complex
in
the
parental
cell
line.
However,
the
amount
of
complexed
TS
RNA
in
the
H630-R1O
line
was
significantly
greater
(approximately
40-fold)
than
that
observed
in
the
parent
H630
line
(Fig.
3).
This
observation
is
consistent
with
our
studies
demonstrating
an
approximately
30-fold
increase
in
levels
of
both
TS
mRNA
(Fig.
1)
and
TS
protein
(20)
in
the
resistant
H630-R1O
cell
line
compared
with
the
parental
H630
cell
line.
To
investigate
whether
the
TS
RNP
complex
was
the
result
of
immunoprecipitation
of
polysomes
containing
na-
scent
polypeptide
chains,
we
examined
the
effect
of
the
protein
synthesis
inhibitor
puromycin
on
the
ability
of
TS
antibody
to
immunoprecipitate
TS
RNP
from
H630-R1O
cells.
Puromycin
treatment
inhibits
protein
synthesis
by
causing
the
premature
release
of
mRNAs
from
ribosomes
(31,
33).
When
[35S]methionine
was
used
to
label
newly
synthesized
proteins,
protein
synthesis
was
>99%
inhibited
when
H630-R1O
cells
were
exposed
to
2
mM
puromycin
for
2
h
prior
to
whole-cell
extraction.
Under
these
conditions,
TS
RNP
complexes
were
precipitated
by
TS
polyclonal
antibody
to
essentially
the
same
degree
in
puromycin-treated
VOL.
14,
1994
210
CHU
ET
AL.
2.37-
t
w
1.35
-
0
0
40
-
f3-Actin
_
(-
c-myc
FIG.
4.
Analysis
of
immunoprecipitated
TS
RNA
from
human
colon
cancer
cells.
(A)
H630-R1O
cells
were
treated
in
either
the
absence
(-)
or
presence
(+)
of
2
mM
puromycin
for
2
h,
immuno-
precipitated
with
TS
polyclonal
antibody,
and
then
subjected
to
RT-PCR.
(B)
H630-R1O
cells
were
treated
in
either
the
absence
(-)
or
presence
(+)
of
2
mM
puromycin
for
2
h,
immunoprecipitated
with
c-myc
monoclonal
antibody,
and
then
subjected
to
RT-PCR
using
c-myc-specific
primers.
and
untreated
(control)
cells
(Fig.
4).
This
result
suggests
that
the
interaction
between
TS
protein
and
its
target
TS
mRNA
does
not
require
an
association
with
polysomes
for
complex
formation.
A
band
resolving
at
a
slightly
higher
position
than
505
nt
was
also
observed.
While
this
higher-
molecular-weight
band
was
not
a
consistent
finding
in
the
PCRs,
it
was
also
observed
when
TS
RNA
was
immunopre-
cipitated
from
whole-cell
extracts
of
H630-R1O
cells
as
shown
in
Fig.
3.
The
precise
nature
of
this
band
is
unclear;
however,
it
is
specific
for
TS,
given
its
hybridization
with
a
TS-specific
DNA
probe,
and
may
represent
an
alternative
PCR-amplified
product
that
results
from
annealing
of
TS
primers
to
regions
flanking
their
predicted
sequences.
To
further
demonstrate
that
the
TS
RNA-TS
protein
complex
is
not
polysome
associated,
additional
experiments
were
performed
to
determine
the
effect
of
puromycin
treat-
ment
on
a
separate
RNA-protein
complex.
Immunoprecipi-
tation
of
H630-R1O
cell
extracts
with
a
c-myc
monoclonal
antibody
followed
by
RT-PCR
with
c-myc-specific
primers
gave
rise
to
a
294-nt
product
(Fig.
4).
However,
following
treatment
with
puromycin
under
the
same
conditions,
the
level
of
c-myc
RNA
immunoprecipitated
from
these
cells
was
not
detectable
(Fig.
4),
suggesting
that
any
interaction
between
the
Myc
protein
product
and
its
corresponding
c-myc
RNA
is
polysome
associated.
This
control
experiment
offers
additional
support
that
the
TS
RNA-TS
protein
in
vivo
interaction
is
not
mediated
by
a
polysome-bound
complex.
A
given
protein
has
the
potential
to
specifically
interact
with
more
than
one
RNA
species
(26,
27).
To
determine
whether
TS-specific
antibody
was
capable
of
immunoprecip-
itating
RNAs
other
than
TS
RNA,
we
used
the
same
RT-PCR
technique,
using
primer
sets
specific
for
unrelated
genes
such
as
c-myc,
max,
and
DHFR.
These
genes
were
selected
since
they
are
all
involved
in
cellular
proliferation
and
their
level
of
mRNA
expression
in
the
H630-R1O
cell
line
was
comparable
to
that
of
TS
mRNA
(Fig.
5A).
To
confer
a
greater
degree
of
specificity
to
these
experiments,
we
used
a
mouse
monoclonal
TS
antibody
(20).
The
expression
of
the
nuclear
oncogene
c-myc
is
correlated
with
cellular
prolifer-
ation,
and
its
function
appears
to
be
required
in
growing
cells
(4,
12,
28).
The
c-myc
gene
is
expressed
at
constitutively
high
levels
in
a
number
of
tumors,
particularly
gastrointes-
tinal
malignancies,
suggesting
an
important
role
in
carcino-
genesis
(.2).
As
seen
in
Fig.
SB,
mouse
monoclonal
TS
B
C
BP
878
-
605
310
-
2
3
4
1
2
3
4
5
6
7
8
9
~,
*
~
,
*
0
RNase
Digested
l
s
Products
~~~~~~~~~~~~f
;_
TS
Protein
-
+
+
+
+
+
+
FIG.
5.
(A)
Northern
blot
analysis
of
c-myc,
max,
DHFR,
and
TS
mRNAs
in
H630-R1O
cells.
Total
RNA
was
isolated
from
H630-R1O
cells
as
previously
described
(6).
The
1.5-kb
TS
cDNA,
0.65-kb
DHFR
cDNA,
1.4-kb
c-myc
insert,
and
0.5-kb
max
insert
were
used
as
hybridization
probes
after
each
was
labeled
with
[32P]dCTP
(3,000
Ci/mmol)
by
the
random
primer
labeling
method
of
Feinberg
and
Vogelstein
(14).
The
filters
were
then
stripped
and
rehybridized
with
a
human
0-actin
probe
to
control
for
loading
and
integrity
of
mRNA.
(B)
Analysis
of
immunoprecipitated
RNAs
from
human
colon
cancer
cells.
RNAs
isolated
by
immunoprecipitation
of
H630-R1O
cell
extracts
with
TS
monoclonal
antibody
were
reverse
transcribed
and
PCR
amplified
by
using
either
the
c-myc-specific
(lane
1),
DHFR-specific
(lane
2),
max-specific
(lane
3),
or
TS-
specific
(lane
4)
primer
(13).
RT-PCR
products
were
resolved
on
a
1%
nondenaturing
agarose
gel
and
analyzed
by
staining
with
ethid-
ium
bromide.
(C)
Specific
binding
of
TS
protein
to
c-myc
RNA.
Gel
mobility
shift
assays
were
performed
with
full-length
c-myc
RNA
as
the
probe.
The
specific
complex
is
indicated
by
the
arrow.
Labeled
RNA
probe
(2.2
fmol,
100,000
cpm)
was
incubated
in
either
the
absence
(lane
1)
or
presence
(lane
2)
of
human
recombinant
TS
protein
(300
pmol).
TS
protein
was
included
in
reaction
mixtures
shown
in
lanes
2
to
8.
Competition
studies
were
performed
with
1-fold
(lane
3),
10-fold
(lane
4),
100-fold
(lane
5),
and
1,000-fold
(lane
6)
molar
excess
of
unlabeled
c-myc
RNA
and
1,000-fold
molar
excess
of
either
human
preplacental
lactogen
RNA
(lane
7)
or
yeast
mRNA
(lane
8).
Labeled
c-myc
RNA
was
incubated
with
bovine
serum
albumin
(300
pmol;
lane
9).
antibody
precipitated
c-myc
RNA,
identifying
this
RNA
species
as
a
member
of
a
TS
RNP
complex
(Fig.
SB,
lane
1).
Immunoprecipitation
experiments
using
a
TS
polyclonal
antibody
revealed
identical
findings.
Hybridization
analysis
with
a
3P-labeled
full-length
c-myc
cDNA
probe
confirmed
that
this
band
was
c-myc
(data
not
shown).
The
finding
that
c-myc
RNA
is
part
of
a
TS
RNP
complex
lends
further
support
to
the
data
presented
in
Fig.
4
that
suggest
that
the
TS
RNP
complex
is
not
polysome
associated.
In
addition,
0
E
0
11
(-)
(+)
CLH
A
A
U
U-
C
1
<-
TS
kb
4.40
-
B
A
x
CID
(J)
H
I
MOL.
CELL.
BIOL.
TS
RNP
COMPLEX
IN
HUMAN
COLON
CANCER
CELLS
211
these
results
demonstrate
that
TS
protein
is
capable
of
binding
to
an
unrelated
mRNA.
Recently,
several
groups
have
shown
that
the
association
between
the
Max
and
Myc
proteins
gives
rise
to
a
heterodimeric
complex
that
specifi-
cally
binds
to
the
consensus
sequence
CACGTG,
with
regulatory
effects
on
both
cell
growth
and
differentiation
(3,
11,
24,
29).
Given
the
close
relationship
between
Myc
and
Max
proteins,
we
investigated
the
presence
of
max
RNA
as
a
potential
member
of
a
TS
RNP
family.
In
contrast
to
c-myc
RNA,
however,
max
RNA
was
not
detected
within
a
TS
RNP
(Fig.
5B,
lane
2).
A
third
unrelated
gene,
DHFR,
was
examined
since
it
encodes
for
a
folate-dependent
protein
product
that
is
critically
involved
in
DNA
biosynthesis
and,
thus,
is
functionally
related
to
TS
protein.
The
absence
of
DHFR
RNA
in
a
TS
RNP
is
consistent
with
recent
work
that
demonstrated
no
direct
interaction
between
human
recom-
binant
TS
protein
and
human
DHFR
mRNA
(Fig.
5B,
lane
3)
(9).
Control
experiments
revealed
that
both
DHFR
and
max
oligonucleotide
primer
sets
were
able
to
PCR
amplify
their
respective
genes
from
total
RNA
isolated
from
the
human
colon
cancer
H630-R1O
cell
line
(data
not
shown).
Moreover,
each
of
these
primer
sets
has
been
previously
used
to
isolate
DHFR
and
max
cDNA
sequences
from
total
cellular
RNA
(9,
11).
The
inability
to
detect
either
DHFR
or
max
RNA
in
the
immunoprecipitated
samples
provides
further
support
for
the
specificity
of
the
c-myc
mRNA-TS
protein
in
vivo
complex.
To
confirm
the
presence
of
a
direct
interaction
between
c-myc
RNA
and
TS
protein,
we
used
the
RNA
electro-
phoretic
gel
mobility
shift
assay
system.
When
full-length
3P-labeled
c-myc
RNA
probe
was
incubated
with
pure
human
recombinant
TS
protein,
a
complex
was
formed
(Fig.
5C,
lane
2).
In
contrast,
no
complex
was
formed
when
the
c-myc
RNA
probe
was
incubated
with
bovine
serum
albumin
(Fig.
SC,
lane
9).
To
further
support
the
specific
nature
of
this
RNA-protein
interaction,
competition
studies
were
per-
formed.
Addition
of
unlabeled
c-myc
RNA
in
molar
excess
of
1-fold
(Fig.
5C,
lane
3),
10-fold
(Fig.
SC,
lane
4),
100-fold
(Fig.
5C,
lane
5),
and
1,000-fold
(Fig.
5C,
lane
6)
effectively
blocked
complex
formation
in
a
dose-dependent
manner.
In
contrast,
unrelated
RNAs
such
as
human
preplacental
lac-
togen
mRNA
(Fig.
5C,
lane
7)
and
yeast
mRNA
(Fig.
5C,
lane
8)
at
1,000-fold
molar
excess
did
not
compete
for
TS
protein
binding.
To
compare
the
relative
affinities
of
TS
RNA
and
c-myc
RNA
for
TS
protein,
we
performed
competition
experiments
using
the
RNA
gel
mobility
shift
assay.
Using
32P-radiola-
beled
full-length
c-myc
RNA
as
the
probe
and
human
recom-
binant
TS
protein,
a
1-
to
1,000-fold
molar
excess
of
either
unlabeled
c-myc
RNA
or
TS
RNA
was
included
in
each
reaction
sample.
As
seen
in
Fig.
6,
c-myc
RNA
and
TS
RNA
effectively
inhibited
RNA-protein
complex
formation
to
the
same
degree,
suggesting
that
their
relative
binding
affinities
to
TS
protein
were
similar
(Kd
3
nM).
In
this
report,
we
have
established
the
identity
of
a
TS
RNP
complex
within
an
intact
biological
system.
With
the
use
of
specific
antibody
to
TS,
we
were
able
to
immunopre-
cipitate
and
isolate
TS
RNA
from
two
different
human
colon
cancer
cell
lines.
Earlier
studies
in
this
laboratory
character-
ized
a
direct
interaction
between
human
recombinant
TS
protein
and
its
corresponding
human
TS
mRNA
(7,
10),
using
cell-free
systems.
To
perform
the
in
vivo
experiments
presented
in this
study,
a
modification
of
the
radioimmuno-
precipitation
technique
that
was
originally
described
by
Lerner
and
Steitz
(27)
was
used.
In
contrast
to
the
small
nuclear
RNAs
that
are
relatively
abundant
within
a
cell
1
2
3
4
5
6
7
8
9
10
RNA
(-)
TS
c-myc
RNase
Digested
Products
TS
Protein
-
+
+
+
+
+
+
+
+
+
FIG.
6.
Competition
analysis
to
determine
relative
binding
of
c-myc
RNA
and
TS
RNA
to
TS
protein.
32P-radiolabeled
c-myc
RNA
(105
cpm,
2.2
fmol)
was
incubated
in
the
absence
(lane
1)
or
presence
(lanes
2
to
10)
of
human
recombinant
TS
protein
(300
pmol).
Competition
studies
were
performed
with
either
1-fold
(lane
3),
10-fold
(lane
4),
100-fold
(lane
5),
or
1,000-fold
(lane
6)
molar
excess
of
unlabeled
TS
RNA
or
1-fold
(lane
7),
10-fold
(lane
8),
100-fold
(lane
9),
or
1,000-fold
(lane
10)
excess
of
c-myc
RNA.
RNA-protein
complexes
were
resolved
on
a
nondenaturing
4%
polyacrylamide
gel
as
described
in
Materials
and
Methods.
The
specific
complex
is
indicated
by
the
arrow.
(approximately
105
to
106
molecules
per
cell),
the
cellular
expression
of
TS
mRNA
is
low
(approximately
103
mole-
cules
per
cell)
(13).
With
the
addition
of
a
sensitive
RT-PCR
technique
to
the
immunoprecipitation
method,
the
direct
interaction
between
TS
protein
and
its
corresponding
mRNA
within
human
colon
cancer
cells
was
established.
Using
this
method,
we
have
further
attempted
to
quanti-
tate
the
percentage
of
total
cellular
TS
mRNA
in
complex
form
with
TS
protein
in
the
H630-R1O
cell
line.
We
deter-
mined
that
the
amount
of
TS
mRNA
in
the
TS
RNP
form
represents
approximately
14%
of
the
total
TS
mRNA
within
the
cell.
However,
this
determination
certainly
represents
an
underestimation
of
the
true
value.
The
immunoprecipita-
tion-RT-PCR
method
used
is
associated
with
a
number
of
technical
limitations,
which
include
intrinsic
RNase
activity
present
in
cells,
that
are
problematic
during
the
isolation
of
RNP
complexes
and
the
relative
inefficiencies
of
immuno-
precipitation
between
complexed
and
uncomplexed
TS
mRNA.
Moreover,
there
are
several
intracellular
factors
that
determine
the
interaction
between
TS
protein
and
its
corresponding
TS
mRNA,
such
as
the
levels
of
the
nucle-
otide
(dUMP)
and
the
reduced
folate
(5,10-methylenetet-
rahydrofolate)
ligands
and
the
redox
state
of
the
TS
protein
(lOa).
For
these
reasons,
the
level
of
intracellular
TS
RNP
complex,
at
a
given
time,
may
vary
significantly,
making
interpretation
of
the
data
for
complexed
and
total
TS
mRNA
problematic.
Previous
studies
identified
two
cis-acting
elements
on
human
TS
mRNA
to
which
TS
protein
specifically
binds
(10).
The
first
site
is
located
within
the
first
188
nt
on
the
TS
mRNA
and
includes
the
translational
start
site,
while
the
second
site
is
located
between
nt
434
and
634
within
the
protein-coding
region.
In
this
study,
the
505-nt
TS
sequence
that
is
PCR
amplified
corresponds
to
the
second
TS
mRNA
binding
site.
Using
a
TS-specific
primer
set
corresponding
to
the
first
188
nt
of
TS
mRNA,
we
were
also
able
to
PCR
amplify
this
TS-specific
region.
In
contrast,
when
primer
sets
defined
by
either
the
full-length
3'
UTR
(nt
939
to
1524)
or
the
5'
UTR
and
the
protein-coding
region
together
(nt
1
to
VOL.
14,
1994
212
CHU
ET
AL.
939)
were
used,
no
PCR
product
was
obtained
(data
not
shown).
These
results
suggest
the
presence
of
at
least
two
discrete
regions
on
TS
mRNA
with
which
TS
protein
spe-
cifically
interacts
and
protects
from
cellular
RNase
degrada-
tion.
While
the
inability
to
RT-PCR
amplify
either
the
5'
or
3'
UTR
suggests
that
these
two
regions
do
not
specifically
interact
with
TS
protein,
other
possibilities
exist
that
might
explain
these
findings.
Complex
secondary
structure(s)
in
these
regions
of
TS
mRNA,
an
especially
high
content
of
GC
ribonucleotides
as
is
observed
in
the
5'
UTR
of
TS
mRNA,
and
regions
in
TS
mRNA
that
may
be
particularly
sensitive
to
nuclease
digestion
represent
alternative
possibilities.
However,
the
identification
of
two
binding
regions
by
using
this
immunoprecipitation-RT-PCR
method
is
consistent
with
the
results
of
the
RNA
binding
experiments
previously
obtained
in
a
cell-free
electrophoretic
gel
mobility
shift
assay
(10).
The
results
of
this
study
also
demonstrate
that
TS
protein
binds
in
vivo
to
c-myc
RNA.
At
this
time,
it
is
not
clear
whether
c-myc
and
TS
RNA
are
each
contained
in
a
distinct
RNP
complex
or
are
part
of
the
same
complex.
As
in
the
case
of the
small
nuclear
RNAs
(26,
27),
it
is
conceivable
that
each
RNA
species
is
part
of
a
unique
complex.
Further
work
to
analyze
each
TS
RNP
molecule
is
required
to
address
this
issue.
This
is
the
first
report
of
a
direct
interaction
between
TS,
an
enzyme
involved
in
DNA
biosynthesis,
and
the
mRNA
of
a
nuclear
oncogene.
Our
studies
using
an
electrophoretic
gel
mobility
shift
assay
provide
further
evidence
for
the
specific
interaction
between
human
c-myc
mRNA
and
human
recom-
binant
TS
protein.
The
observation
that
TS
protein
and
c-myc
RNA
represent
members
of
an
RNP
complex
suggests
that
TS
may
be
involved
in
the
regulation
of
expression
and/or
function
of
c-myc
RNA.
An
alternative
possibility
is
that
binding
of
TS
protein
to
c-myc
RNA
disrupts
one
of
the
normal
regulatory
functions
of
TS,
namely,
inhibition
of
TS
mRNA
translation.
Preliminary
work
suggests
that
the
cis-
acting
elements
on
c-myc
RNA
involved
in
TS
protein
binding
are
different
from
the
two
binding
sites
previously
identified
for
human
TS
mRNA.
Further
studies
are
required
to
elucidate
the
molecular
basis
for
the
TS
protein-c-myc
RNA
interaction.
In
addition,
the
functional
significance
of
this
particular
mRNA-protein
interaction
will
need
to
be
determined.
The
observation
that
TS
RNP
complexes
contain
TS
RNA
as
well
as
c-myc
RNA
suggests
that
TS
protein
may
be
involved
in
the
coordinate
regulation
of
a
number
of
other
cellular
genes.
The
immunoprecipitation-RT-PCR
method
described
in
this
report
should
allow
for
the
identification
of
those
genes
whose
expression
may
be
under
some
level
of
control
by
TS.
Finally,
given
the
increased
role
of
RNA-
protein
interactions
in
determining
translational
regulation
of
gene
expression,
this
technique
may
also
be
applied
to
the
study
of
other
RNA-binding
proteins.
ACKNOWLEDGMENTS
We
thank
Bruce
Chabner
and
Frederick
Kaye
for
valuable
discussions
and
review
of
the
manuscript
and
Kathy
Moore
for
editorial
assistance
in
the
preparation
of
the
manuscript.
This
work
was
supported
in
part
by
grants
from
the
National
Cancer
Institute
(CA44355
to
F.M.)
and
the
National
Science
Foundation
(DMB90-03737
to
G.F.M.).
REFERENCES
1.
Ayusawa,
D.,
K.
Shimizu,
H.
Koyama,
S.
Kaneda,
K.
Takeishi,
and
T.
Seno.
1986.
Cell-cycle-directed
regulation
of
thymidylate
synthase
messenger
RNA
in
human
diploid
fibroblasts
stimu-
lated
to
proliferate.
J.
Mol.
Biol.
190:559-567.
2.
Bernardi,
A.,
and
P.-F.
Spahr.
1972.
Nucleotide
sequence
at
the
binding
site
for
coat
protein
on
RNA
of
bacteriophage
R17.
Proc.
Natl.
Acad.
Sci.
USA
69:3033-3037.
3.
Blackwood,
E.
M.,
and
R.
N.
Eisenman.
1991.
Max:
a
helix-loop-
helix
zipper
protein
that
forms
a
sequence-specific
DNA-bind-
ing
complex
with
Myc.
Science
251:1211-1217.
4.
Blackwood,
E.
M.,
L.
Kretzner,
and
R.
N.
Eisenman.
1992.
Myc
and
Max
function
as
a
nucleoprotein
complex.
Curr.
Opin.
Genet.
Dev.
2:227-235.
5.
Carey,
J.,
and
0.
C.
Uhlenbeck.
1983.
Kinetic
and
thermody-
namic
characterization
of
the
R17
coat
protein-ribonucleic
acid
interaction.
Biochemistry
22:2610-2615.
6.
Chu,
E.,
J.
C.
Drake,
D.
M.
Koeller,
S.
Zinn,
C.
A.
Jamis-Dow,
G.
C.
Yeh,
and
C.
J.
Allegra.
1991.
Induction
of
thymidylate
synthase
associated
with
multidrug
resistance
in
human
breast
and
colon
cancer
cell
lines.
Mol.
Pharmacol.
39:136-143.
7.
Chu,
E.,
D.
M.
Koeller,
J.
L.
Casey,
J.
C.
Drake,
B.
A.
Chabner,
P.
C.
Elwood,
S.
Zinn,
and
C.
J.
Allegra.
1991.
Autoregulation
of
human
thymidylate
synthase
messenger
RNA
translation
by
thymidylate
synthase.
Proc.
Natl.
Acad.
Sci.
USA
88:8977-
8981.
8.
Chu,
E.,
D.
M.
Koeller,
P.
G.
Johnston,
S.
Zinn,
and
C.
J.
Allegra.
1993.
Regulation
of
thymidylate
synthase
in
human
colon
cancer
cells
treated
with
5-fluorouracil
and
interferon-
gamma.
Mol.
Pharmacol.
43:527-533.
9.
Chu,
E.,
C.
H.
Takimoto,
D.
Voeller,
J.
L.
Grem,
and
C.
J.
Allegra.
1993.
Specific
binding
of
human
dihydrofolate
reduc-
tase
protein
to
dihydrofolate
reductase.
Biochemistry
32:4756-
4760.
10.
Chu,
E.,
D.
Voeller,
D.
M.
Koeller,
J.
C.
Drake,
C.
H.
Takimoto,
G.
F.
Maley,
F.
Maley,
and
C.
J.
Allegra.
1993.
Identification
of
an
RNA
binding
site
for
human
thymidylate
synthase.
Proc.
Natl.
Acad.
Sci.
USA
90:517-521.
10a.Chu,
E.,
D.
Voeller,
G.
Maley,
F.
Maley,
and
C.
Allegra.
1993.
Role
of
the
redox
state
in
the
interaction
of
thymidylate
syn-
thase
protein
with
thymidylate
synthase
mRNA.
Proc.
Am.
Assoc.
Cancer
Res.
34:413.
11.
Cogliati,
T.,
B.
K.
Dunn,
M.
Bar-Ner,
C.
M.
Cultraro,
and
S.
Segal.
1993.
Transfected
wild-type
and
mutant
max
regulate
cell
growth
and
differentiation
of
murine
erythroleukemia
cells.
Oncogene
8:1263-1268.
12.
DePinho,
R.
A.,
N.
Schreiber-Agus,
and
F.
W.
Alt.
1991.
myc
family
oncogenes
in
the
development
of
normal
and
neoplastic
cells.
Adv.
Cancer
Res.
57:1-46.
13.
Dolnick,
B.
J.,
Z.-G.
Zhang,
J.
D.
Hines,
and
Y.
M.
Rustum.
1992.
Quantitation
of
dihydrofolate
reductase
and
thymidylate
synthase
mRNAs
in
vivo
and
in
vitro
by
polymerase
chain
reaction.
Oncol.
Res.
4:65-72.
14.
Feinberg,
A.,
and
B.
Vogelstein.
1983.
A
technique
for
radiola-
beling
DNA
restriction
endonuclease
fragments
to
high
specific
activity.
Anal.
Biochem.
137:6-13.
15.
Haile,
D.
J.,
M.
W.
Hentze,
T.
A.
Rouault,
J.
B.
Harford,
and
R.
D.
Klausner.
1989.
Regulation
of
interaction
of
the
iron-
responsive
element
binding
protein
with
iron-responsive
RNA
elements.
Mol.
Cell.
Biol.
9:5055-5061.
16.
Hentze,
M.
W.,
S.
Caughman,
T.
Rouault,
J.
Barriocanal,
A.
Dancis,
J.
B.
Harford,
and
R.
D.
Klausner.
1987.
Identification
of
the
iron-responsive
element
for
the
translational
regulation
of
human
ferritin
mRNA.
Science
238:1570-1573.
17.
Hentze,
M.
W.,
T.
A.
Rouault,
J.
B.
Harford,
and
R.
D.
Klausner.
1989.
Oxidation-reduction
as a
molecular
mechanism
of
a
regulatory
RNA-protein
interaction.
Science
244:357-359.
18.
Hershey,
J.
W.
B.
1991.
Translational
control
in
mammalian
cells.
Annu.
Rev.
Biochem.
60:717-755.
19.
Jenh,
C.-H.,
P.
K.
Geyer,
and
L.
F.
Johnson.
1985.
Control
of
thymidylate
synthase
mRNA
content
and
gene
transcription
in
an
overproducing
mouse
cell
line.
Mol.
Cell.
Biol.
5:2527-
2532.
20.
Johnston,
P.
G.,
C.-M.
Liang,
S.
Henry,
B.
A.
Chabner,
and
C.
J.
Allegra.
1991.
Production
and
characterization
of
mono-
clonal
antibodies
that
localize
human
thymidylate
synthase
in
MOL.
CELL.
BIOL.
TS
RNP
COMPLEX
IN
HUMAN
COLON
CANCER
CELLS
213
the
cytoplasm
of
human
cells
and
tissue.
Cancer
Res.
51:6668-
6676.
21.
Kaneda,
S.,
J.
Nalbantoglu,
K.
Takeishi,
K.
Shimizu,
0.
Gotoh,
T.
Seno,
and
D.
Ayusawa.
1990.
Structural
and
functional
analysis
of
the
human
thymidylate
synthase
gene.
J.
Biol.
Chem.
265:20277-20284.
22.
Keyomarsi,
K.,
and
A.
B.
Pardee.
1991.
Differential
expression
of
thymidylate
synthase
message
versus
protein
in
synchronized
human
MCF-7
breast
cancer
cells.
Proc.
Am.
Assoc.
Cancer
Res.
32:330.
23.
Klausner,
R.
D.,
and
J.
B.
Harford.
1989.
Cis-trans
models
for
post-transcriptional
gene
regulation.
Science
246:870-872.
24.
Kretzner,
L.,
E.
M.
Blackwood,
and
R.
N.
Eisenman.
1992.
Myc
and
Max
proteins
possess
distinct
transcriptional
activities.
Nature
(London)
359:426-429.
25.
Leibold,
E.
A.,
and
H.
N.
Munro.
1988.
Cytoplasmic
protein
binds
in
vitro
to
a
highly
conserved
sequence
in
the
5'
untrans-
lated
region
of
ferritin
heavy-
and
light-subunit
mRNAs.
Proc.
Natl.
Acad.
Sci.
USA
85:2171-2175.
26.
Lerner,
M.
R.,
J.
A.
Boyle,
J.
A.
Harden,
and
J.
A.
Steitz.
1981.
Two
novel
classes
of
small
ribonucleoproteins
detected
by
antibodies
associated
with
lupus
erythematosus.
Science
211:
400-402.
27.
Lerner,
M.
R.,
and
J.
A.
Steitz.
1979.
Antibodies
to
small
nuclear
RNAs
complexed
with
proteins
are
produced
by
pa-
tients
with
systemic
lupus
erythematosus.
Proc.
Natl.
Acad.
Sci.
USA
76:5495-5499.
28.
Luscher,
B.,
and
R.
N.
Eisenman.
1990.
New
light
on
Myc
and
Myb.
Part
I.
Myc.
Genes
Dev.
4:2025-2035.
29.
Makela,
T.
P.,
P.
J.
Koskinen,
I.
Vastrik,
and
K.
Alitalo.
1992.
Alternative
forms
of
Max
as
enhancers
or
suppressors
of
Myc-Ras
cotransformation.
Science
256:373-377.
30.
Navalgund,
L.
G.,
C.
Rossana,
A.
J.
Muench,
and
L.
F.
Johnson.
1980.
Cell
cycle
regulation
of
thymidylate
synthetase
gene
expression
in
cultured
mouse
fibroblasts.
J.
Biol.
Chem.
255:
7386-7390.
31.
Pachter,
J.
S.,
T.
J.
Yen,
and
D.
W.
Cleveland.
1987.
Autoreg-
ulation
of
tubulin
expression
is
achieved
through
specific
deg-
radation
of
polysomal
tubulin
mRNAs.
Cell
51:283-292.
32.
Park,
J.
G.,
H.
K.
Oie,
P.
H.
Sugarbaker,
J.
G.
Henslee,
T.
R.
Chen,
B.
E.
Johnson,
and
A.
Gazdar.
1987.
Characteristics
of
cell
lines
established
from
human
colorectal
carcinoma.
Cancer
Res.
47:6710-6718.
33.
Pestka,
S.
1971.
The
use
of
inhibitors
in
studies
of
protein
synthesis.
Methods
Enzymol.
30:261-282.
34.
Steitz,
J.
A.
1989.
Immunoprecipitation
of
ribonucleoproteins
using
autoantibiotics.
Methods
Enzymol.
180:468-481.
VOL.
14,
1994
