Proenkephalin Gene Expression in
C6 Rat Glioma Cells: Potentiation of
Cyclic Adenosine 3',5'Monophosphate-Dependent
Transcription by Glucocorticoids
Jay Joshi and Steven L. Sabol
Laboratory of Biochemical Genetics
National Heart, Lung, and Blood Institute
National Institutes of Health
Bethesda, Maryland 20892
INTRODUCTION
Glucocorticoids enhance proenkephalin gene
expression in several cell types. To elucidate the
mechanism(s) involved, we analyzed the potentiation by dexamethasone of the cAMP-dependent increase in proenkephalin mRNA levels elicited by
forskolin in C6 rat glioma cells. This potentiation did
not require ongoing protein synthesis. In nuclear runon transcription assays, dexamethasone alone did
not alter proenkephalin transcription, but strongly
increased the magnitude and duration of transcriptional elevation by forskolin through a direct action
not requiring ongoing protein synthesis. Dexamethasone did not alter basal or stimulated cAMP levels.
To search for functionally cooperative glucocorticoid
and cAMP regulatory elements, we transfected C6
cells with plasmids containing the chloramphenicol
acetyltransferase (CAT) gene under the control of
rat proenkephalin sequences from bases -5800 to
+703. Maximum stimulation of transiently expressed
CAT activity by forskolin required more than 145 and
190 or fewer base pairs of 5'-flanking sequence,
implicating sequences up-stream from the previously described cAMP-inducible enhancer. Dexamethasone reduced forskolin-stimulated CAT
expression from plasmids with 190 or more basepairs of 5-flanking sequence, an effect apparently
involving multiple up-stream regions. Dexamethasone also reduced forskolin-stimulated CAT mRNA
levels in C6 cells stably transfected with proenkephalin/CAT chimeric genes in the presence or absence of protein synthesis. In summary, we demonstrate that glucocorticoids and cAMP synergize
positively in regulating transcription of the endogenous gene, but interact negatively in regulating the
chimeric constructs, which may lack the context or
distal element(s) required for positive synergism.
(Molecular Endocrinology 5:1069-1080, 1991)
Genes coding for neuropeptide precursors are regulated through mechanisms that ultimately couple the
rate of neuropeptide production to the need of the
organism. The regulation of the gene coding for
(pre)proenkephalin, the biosynthetic precursor of the
enkephalins (1 -3; reviewed in Ref. 4), is of considerable
interest in view of the diverse actions of these endogenous opioids as neuromodulators and circulating hormones. Recent studies have shown that in a variety of
tissues and cells, proenkephalin mRNA levels are under
the positive control of 1) cAMP, the action of which in
some cells is potentiated by activation of protein kinaseC (5-8), and 2) depolarization with the concomitant
activation of voltage-dependent Ca2+ channels (9-11).
Detailed analyses by Comb et al. (13,14) of the human
proenkephalin promoter region revealed that transcriptional regulation by both cAMP and depolarization is
dependent upon an enhancer sequence spanning the
region between bases - 7 0 and -114 relative to the
transcription start site (12-15). This region contains
multiple core elements that work synergistically and
can bind at least four protein factors, namely ENKTF1, AP-1, AP-2, and AP-4 (13,14).
Glucocorticoids, such as corticosterone and dexamethasone, increase the levels of proenkephalin mRNA
and potentiate, usually by a factor of 2-6, the effects
of major stimulators of proenkephalin mRNA levels in
several nonneuronal cell types. For example, glucocorticoids and adenylate cyclase activators, such as forskolin, synergistically elevate proenkephalin mRNA levels in C6 rat glioma cells (6), NG108-15 neuroblastomaglioma hybrid cells (7), and peritubular cells of the rat
testis (8). Furthermore, glucocorticoids potentiate the
depolarization-induced increases in proenkephalin
mRNA and enkephalin levels in bovine chromaffin cells
(10). Increases in proenkephalin mRNA levels in the rat
adrenal medulla upon explantation are potentiated by
glucocorticoids, an effect more pronounced in glands
0888-8809/91/1069-1080$03.00/0
Molecular Endocrinology
Copyright © 1991 by The Endocrine Society
1069
Vol 5 No. 8
MOL ENDO-1991
1070
taken from hypophysectomized rats (16, 17). Proenkephalin mRNA levels are increased by dexamethasone
and nicotine acting synergistically in cultured human
pheochromocytoma cells (18) and by glucocorticoid
alone in FR3T3 rat fibroblast cells (19). In neuronal
tissue, glucocorticoid treatment generally increases
proenkephalin mRNA levels in adrenalectomized or hypophysectomized animals, for example in the striatum
(20) and superior cervical ganglion (21) of the rat. However, the reported increases are more modest (1.5-2
times the control value) than those in nonneuronal
systems and are marginal in animals with normal circulating glucocorticoid levels. These studies indicate
that while glucocorticoid at elevated concentrations,
such as those during stress, is not a dominant regulator
of the proenkephalin gene, glucocorticoid at normal
physiological concentrations exerts a permissive or synergistic effect on other regulators of this gene.
The molecular mechanism of this effect has not been
investigated previously. In C6 glioma cells, glucocorticoid treatment potentiates the cAMP-dependent elevation of proenkephalin mRNA levels, but has little or
no effect alone. The question arises as to whether this
regulation occurs predominantly at the level of transcription of the proenkephalin gene, either indirectly by
affecting the expression of other gene(s) controlling
transcription or directly by an interaction of the steroidglucocorticoid receptor complex bound to a glucocorticoid regulatory element (GRE) (reviewed in Ref. 22)
on or near the proenkephalin gene, with factors bound
to the cAMP-inducible enhancer. Alternatively, glucocorticoids may stabilize proenkephalin transcripts or
have some other posttranscriptional action. In the present study we demonstrate cooperative interactions between the glucocorticoid- and cAMP-signalling systems
in the regulation of native proenkephalin gene transcription. In addition, we demonstrate unexpected complexities in the regulation of proenkephalin/reporter gene
constructs by glucocorticoids and cAMP.
RESULTS
Lack of Protein Synthesis Requirement for
Glucocorticoid Potentiation of cAMP-Stimulated
Expression of the Endogenous
Proenkephalin Gene
We have previously shown (6) that proenkephalin
mRNA levels in C6 cells are increased by the synergistic
action of forskolin (or /?-adrenergic agonist) and dexamethasone (or other glucocorticoid hormone) over a
period of 24 h, with a peak around 4 - 6 h. To determine
whether glucocorticoids increase proenkephalin mRNA
indirectly by inducing the synthesis of a protein(s) that
affects the level of this transcript, we examined the
effects of cycloheximide on the separate contributions
of dexamethasone and forskolin (Table 1). Dexamethasone alone increased the proenkephalin mRNA level
by 3 0 % and potentiated the effect of forskolin over 2-
Table 1. Effect of Dexamethasone on the cAMP-Dependent
Increase in Proenkephalin mRNA Abundance: Lack of a
Protein Synthesis Requirement
Relative Proenkephalin B a K r t _ , . r w i rvw
mRNAAbundance±SEM *at,o o f + D e x / - D e x
Treatment
Minus Chx
Plus Chx
None
1.0
1.0
Dex
1.3 ± 0 . 2 1.3 ± 0 . 0
Forskolin
6.7 ± 0.4 5.3 ± 0.3
Dex + forskolin 14.1 ± 1 . 3 11.6 ± 0 . 6
Minus Chx PlusChx
1.3
1.3
2.1
2.2
Cultures of C6 cells were treated with 1 ^ M dexamethasone
(Dex), 20 HM forskolin, or both for 4.5 h in the presence or
absence of 75 ^M cycloheximide (Chx). Cycloheximide was
added 45 min before dexamethasone and/or forskolin. Under
these conditions, [35S]methionine incorporation into protein
was inhibited by 96%. Proenkephalin mRNA abundances per
U total RNA, relative to the abundance in untreated cells, were
determined by RNA dot blotting. The results of two separate
experiments were averaged.
fold in both the presence and absence of a high concentration of cycloheximide, indicating that the effects
of glucocorticoid and cAMP do not require de novo
protein synthesis.
Effects of Glucocorticoid and cAMP on
Proenkephalin Gene Transcription
We determined by the nuclear run-on assay the effects
of dexamethasone and forskolin on the relative apparent rates of transcription of the endogenous proenkephalin gene in C6 cells. As shown in Fig. 1, treatment
of the cells with 1 HM dexamethasone alone had essentially no effect on proenkephalin transcription. Forskolin
(20 HM) alone elicited a rapid but transient stimulation
that was maximally 4 times the control value at 1 h, but
nearly extinguished by 4 h. The combination of dexamethasone and forskolin elicited a higher and more
sustained stimulation of proenkephalin transcription,
attaining a peak of 18 times the control at 2 h of
treatment and then slowly declining, but remaining at
least 7 times the control level over the 24-h period. The
effect of protein synthesis inhibition by cycloheximide
on these effects was examined at the 2 h point (Fig. 1).
The stimulation by forskolin and that by dexamethasone
plus forskolin in the presence of cycloheximide (2- and
13-fold, respectively) were only slightly less than those
in the absence of cycloheximide at this time point (3and 18-fold, respectively). These experiments demonstrate a substantial enhancement by glucocorticoid of
the extent and duration of forskolin-stimulated proenkephalin gene transcription by a direct mechanism, i.e.
one not involving de novo protein synthesis.
Effect of Dexamethasone on cAMP Content and
cAMP-Dependent Protein Kinase Activity
To determine whether dexamethasone potentiates the
stimulation by forskolin of endogenous proenkephalin
Regulation of Proenkephalin Gene Transcription
1071
A
Time (hr):
Treatment:
1
1
O
DF
2
DF
2
2
Cx
CxD
2
2
CxF CxDF
«Mft 0 P
ProEnk-*
4
4
F
0
«M»
4
DF
6
DF
24
0
24
D
24
F
24
DF
+ *>*m
Vector-*
GAPDH-*
B
0)
(£
0 1
4
6
8
10
12
Duration of Treatment (Hours)
Fig. 1. Regulation of Endogenous Proenkephalin Gene Transcription by Forskolin and Dexamethasone
C6 cells were treated with no drug (O), 1 M M dexamethasone (D or Dex), 20 HM forskolin (F), or 1 fiM dexamethasone plus 20
^M forskolin (DF) in the absence or presence of 75 MM cycloheximide (Chx) for 1-24 h, as indicated. Cycloheximide was added 45
min before the other drugs. A, Autoradiograms (7-day exposure) of blots on which 32P-labeled transcripts hybridized to proenkephalin
(Proenk), plBI31 (vector), and GAPDH probes. B, Time course of relative proenkephalin run-on transcriptional activity. Data are
derived from the autoradiograms shown in A. Proenkephalin transcription, normalized to GAPDH transcription, is expressed as
ratios to the control level at each time. O, No drug (equated to 1.0 at each time); • , dexamethasone (Dex); D, forskolin; • ,
dexamethasone plus forskolin. A zero time point is shown for clarity, but was not actually performed. Values obtained in the
presence of cycloheximide are not plotted in B in the interest of clarity; these values are 1.1, 2.0, and 13.4 times the control value
for dexamethasone, forskolin, and dexamethasone plus forskolin, respectively.
transcription merely by increasing or sustaining the
elevation of cAMP levels by forskolin, we measured the
cAMP concentrations in cells treated as described in
the experiment shown in Fig. 1. The cAMP content in
response to forskolin addition increased 175-fold within
30 min and then declined, stabilizing at approximately
4 times the control value (Fig. 2). An identical response
was obtained with dexamethasone plus forskolin. Furthermore, dexamethasone had no effect on the basal
cAMP level. Thus, glucocorticoids potentiate proenkephalin gene transcription by altering a parameter other
than the cAMP concentration.
Treatment of C6 cells over 1-4 h with 1 ^M dexamethasone modestly increased (by an average of 30%)
the specific activities of nuclear cAMP-dependent and independent protein kinase (data not shown). However,
these increases were fully inhibited by cycloheximide,
and thus, they cannot explain the effect of glucocorticoids on proenkephalin transcription in C6 cells.
Effects of Regulators on Transient Expression of
the Chloramphenicol Acetyttransferase (CAT) Gene
in Proenkephalin/CAT Chimeric Plasmids
To attempt to localize a putative GRE cooperating with
the cAMP-inducible enhancer of the rat proenkephalin
gene, we constructed a series of chimeric plasmids,
termed pREJCAT(-n/+53), which contain the bacterial
CAT gene linked to the rat proenkephalin promoter and
various lengths [n = 95-5800 basepairs (bp) from the
transcription start site] of the 5'-flanking DNA. An additional plasmid, pREJCAT(-2700/+703), was constructed to include exon I (untranslated) and the entire
intron A. These constructs are diagrammed in Fig. 3.
C6 cells were transfected with these plasmids, and
the effects of forskolin and/or dexamethasone treatment on CAT activity were determined under a variety
of conditions. As shown in Fig. 4A, basal CAT activity
was not markedly affected by the length of 5'-flanking
region, although constructs with 145 and 190 bp had
Vol 5 No. 8
MOL ENDO-1991
1072
3000
-
2000
-
h
1200 1000
-
800
-
NO DRUG
DEX
FORSKOLIN
DEX + FORSKOLIN
600
400
200
DURATION OF TREATMENT (HOURS)
Fig. 2. Effects of Dexamethasone (DEX) and/or Forskolin on
Intracellular cAMP Levels in C6 Cells
Confluent cells (average, 2.8 mg protein/60-mm plate) were
treated at zero time with no drug, 1 MM dexamethasone, 20
HM forskolin, or 1 MM dexamethasone plus 20 MM forskolin. At
the indicated times, cells were harvested, and intracellular
cAMP was measured, as described in Materials and Methods.
Plotted data are the mean of values obtained from triplicate
plates ± SEM. Where error bars are not shown, their positions
are within the symbol. Values for control and dexamethasonetreated cells averaged 14 pmol cAMP/mg protein.
twice the basal activity of larger constructs. For constructs having at least 190 bp of 5'-flanking DNA,
forskolin treatment elevated CAT activity to 16-23
times the control value. The construct containing 145
bp of 5'-flanking DNA was significantly and reproducibly
less responsive to forskolin, in both relative and absolute terms, than were the longer constructs, while the
shortest construct, with 95 bp of 5'-flanking region,
was nearly unresponsive to forskolin. Although dexamethasone alone had no effect on CAT expression, it
unexpectedly and consistently inhibited forskolin-stimulated CAT expression in cells transfected with constructs containing 190 bp or more of 5'-flanking DNA.
Inhibition was most significant with constructs that had
at least 437 bp of 5'-flanking DNA, but was consistently
observed at a marginal level (14-24%) with 190 bp. In
cells transfected with constructs that had 145 or 95 bp
of 5'-flanking DNA, dexamethasone either did not affect
forskolin activation or, in some experiments (not
shown), stimulated forskolin activation by up to 50%.
Similar results were obtained with NIH 3T3 fibroblasts
transfected with these constructs (not shown).
In C6 cells transfected with pREJCAT(-2700/+703),
the responses to forskolin and dexamethasone were
qualitatively identical to those with pREJCAT(-2700/
+53) (Fig. 4B). Dexamethasone inhibited not only the
stimulation by forskolin, but also that by 1 rriM
(Bu)2cAMP. Dexamethasone also inhibited the modest
stimulation elicited by a low concentration (1 ^M) of
forskolin (not shown). Thus, we found no evidence for
a positive or cooperative GRE in the down-stream
portion of exon I or the entire intron A.
Inhibition by dexamethasone was consistently observed in many experiments under a variety of conditions, including treatment durations from 3-24 h, transfection with suboptimal amounts of plasmid (1-3 n9l
100-mm dish), and addition of the protein kinase-C
activator phorbol 12-myristate 13-acetate (0.1 ^M)
along with forskolin (data not shown). The half-maximally inhibitory conmcentration of dexamethasone was
roughly 3 nM, a concentration consistent with involvement of glucocorticoid receptors (data not shown).
In control experiments (Fig. 4B), dexamethasone
stimulated, as expected, CAT expression driven by the
mouse mammary tumor virus long terminal repeat,
which contains a positive GRE, but did not affect CAT
expression driven by the simian virus-40 (SV40) promoter of pSV2CAT, which lacks a GRE. In cells transfected with CAT plasmids possessing the proenkephalin 5'-flanking and promoter sequences in the opposite orientation, i.e. pREJCAT(+53/-2700) and
pREJCAT(703/-2700), basal CAT activity was low, and
no effect of forskolin or dexamethasone was observed
(not shown).
From these results we derive the following conclusions about the proenkephalin promoter and up-stream
region in the chimeric plasmids under the conditions of
transient expression: 1) Dexamethasone alone has
either no significant effect or a slightly inhibitory effect
on CAT expression. 2) Cyclic AMP elicits strong transcriptional responses that absolutely require 5'-flanking
sequences between bases - 9 5 and -145, while additional sequences between bases -145 to -190 are
required for maximal activity. 3) The dominant effect of
glucocorticoid on cAMP-stimulated CAT expression is
inhibitory, in contrast to the stimulatory response of the
native proenkephalin gene.
Effects of Regulators on Expression of the
Proenkephalin/CAT Gene in Stably
Transfected C6 Cells
Under the conditions of the transient expression assay,
the tested episomal chimeric constructs may not behave similarly to the endogenous gene because of
differences in such variables as methylation and association with chromatin proteins. Therefore, we pre
pared C6 cell lines containing stably integrated
pREJCAT(-2700/+53) and (-2700/+703) DNA. Dexamethasone treatment of these cells reduced both
basal and forskolin-stimulated CAT specific activities
(Table 2). To determine whether this reduction is the
result of a protein induced by dexamethasone, we
quantitated the levels of CAT mRNA in cells treated
with drugs in the presence of cycloheximide (Fig. 5).
Forskolin elevated CAT mRNA levels in -2700/+53
and -2700/+703 transfectants, respectively, while
dexamethasone inhibited forskolin-stimulated CAT
Regulation of Proenkephalin Gene Transcription
1073
-lambda RENK1
- lambda RENK2
Bam HI
pBR322
pREJCAT(-5800/+53)
pBR322
pBR322
pREJCAT(-2700/+53)
pREJCAT(-2700/+703)
Fig. 3. Structures of Proenkephalin/CAT Plasmids
The rat proenkephalin genomic organization (3), boundaries of the two genomic clones used (32), and bacterial CAT and SV40
sequences (50) are indicated. Constructs with shorter lengths of 5'-flanking region and 3' terminus at +53 were also made and
are formally analogous to pREJCAT(-2700/+53). DNA sequencing of the promoter region of several of these constructs (500
bases up-stream and 53 bases down-stream from the presumed cap site) was generally confirmatory of the published sequence
(3), but 10 differences were found. The numbering of terminal bases of the constructs is based on the authors' sequence, using
the previously proposed (3) transcription start site as + 1 . The numbers of the up-stream proenkephalin termini of the smaller
constructs (-437 and lower) are 1-2 higher than the corresponding numbers of the published sequence (3). The revised sequence
is available from the authors upon request.
mRNA. In other experiments, in the absence of cycloheximide, quantitatively similar stimulations by forskolin
and inhibition by dexamethasone were found (not
shown).
The RNA preparations used for the experiment
shown in Fig. 5 were also analyzed for mature proenkephalin transcripts (1500 bases) derived from the native gene. Forskolin alone modestly increased the abundance of this transcript, and as in the case of standard
C6 cells, dexamethasone significantly enhanced this
stimulation. Thus, glucocorticoid regulates the native
proenkephalin gene positively, while in the same cells it
regulates the integrated chimeric gene negatively. The
magnitude of the stimulation by forskolin appeared to
be much higher for CAT transcripts than for proenkephalin transcripts, while the stimulations by dexamethasone plus forskolin over the control value are similar
for the two transcripts.
We conclude that the tested proenkephalin/CAT constructs, even when integrated into the cellular genome,
do not properly mimic the regulation of the native
proenkephalin gene. Furthermore, the inhibitory effect
of glucocorticoid on expression of the chimeric constructs is not mediated by the synthesis of new protein,
such as a specific ribonuclease, protease, or transcription factor.
DISCUSSION
The major finding of this study is that transcription of
the rat proenkephalin gene is regulated by the cooperative action of glucocorticoids and cAMP. A correlation
is apparent between the increases in the mRNA steady
state level (Table 1) (6) and the increases in the relative
transcription rate (Fig. 1) in C6 cells treated with dexamethasone and forskolin. Dexamethasone alone has
little or no effect on either parameter, but potentiates
the forskolin-dependent increase in both. Thus, we
conclude that the observed increases in proenkephalin
mRNA levels are due at least in part to transcriptional
activation. However, the relationship between transcriptional stimulation and steady state mRNA level appears
complex. In the nuclear run-on assays, the stimulation
by forskolin was always rapid and transient, disappearing within 4 h after its addition to the cells, while
glucocorticoid markedly extended, through 24 h, the
duration as well as the magnitude of this stimulation.
Yet, the brief transcriptional effect of forskolin leads to
a 3- to 7-fold increase in proenkephalin mRNA abundance, while the potentiation by glucocorticoid is only
2-fold, suggesting that additional posttranscriptional
mechanisms may be affected by these compounds. For
Vol 5 No. 8
MOL ENDO-1991
1074
C6 CELLS: A
l
i1
25
20
15
YECA
<
5
10
5
1
Proenk
Insert:
" ii
i •
-
iwL
T
I 1
1 ri1 iE
Bl
UJ
CC
B
T
-i-
mm
Hi
I1
H
1
• JJ
m
;
ft
^
in • mi
-
|i
-
-T"
c D F DF C D F DF C D F DF C D F DF C D F DF C D FDF C D F DF
-5800/+53 -2700/+53 -1050/+53 -437/+53 -190/+53 -145/+53
-95/+53
C D F DFA DA
-2700/+703
C D F DF
CD
PSV2CAT pMMTVCAT
Fig. 4. Effects of Dexamethasone and Forskolin on Proenkephalin/CAT Expression in C6 Cells
C6 cells were transfected with pREJCAT constructs containing the indicated proenkephalin genomic inserts or with control
plasmids. A, Forty-eight hours after transfection, cells were treated for 7 additional hours with vehicle alone (C), 0.1 HM
dexamethasone (D), 20 HM forskolin (F), or 0.1 HM dexamethasone plus 20 HM forskolin (DF). CAT activity is expressed relative to
the control activity of p R E J C A T ( - 5 8 0 0 / + 5 3 ) . Duplicate plates were tested; error bars indicate SEMS in cases in which the duplicates
were not in tight agreement. B, Conditions were identical to those in A, except that the forskolin concentration was 10 ^M, and 1
mwi A/ 6 O 2 '-(Bu) 2 cAMP without or with dexamethasone (A or DA, respectively) was also tested. CAT activity is expressed relative
to the control value for each plasmid. The pSV2CAT and pMMTVCAT contain no proe,nkephalin insert.
Table 2. Effect of Forskolin and Dexamethasone on CAT Enzyme Activity in C6 Cell Lines Containing Stably Integrated
Proenkephalin-CAT Chimeric Genes
Relative CAT Specific Activity ± SEM
Treatment
None
Dex
Forskolin
Dex + forskolin
C6[pREJCAT(-2700/+53)-neo]
C6[pREJCAT(-2700/+703)-neo]
1.0
0.55 ± 0.03
5.1 ± 0.2
2.0 ±0.1
0.61 ± 0.21
11.3 ± 1.2
1.0
6.1 ±0.3
Cultures (40-50% confluent) of pooled G418-resistant clones of C6[pREJCAT-/?eo] stable transfectants were treated with 1
dexamethasone (Dex), 20 HM forskolin, or both for 26 h. Results are the averages of duplicate plates.
example, glucocorticoid may moderately decrease the
stability of proenkephalin transcripts. However, such an
effect seems unlikely, because the effects of glucocorticoids on mRNA stability, both positive and negative,
usually require ongoing protein synthesis (23, 24).
The transcription of several other genes has been
reported to be increased by both glucocorticoids and
cAMP elevation in a variety of cell types (25-28). In
these cases, the effects of dexamethasone and cAMP,
separately and in combination, were reported to follow
a variety of patterns, each of which differs markedly
from the synergistic action reported here.
The activation of transcription by glucocorticoids generally involves the interaction of the steroid-glucocorticoid receptor complex with a positive GRE (reviewed
in Ref. 22) usually located in the 5'-flanking sequence
within 3000 bp of the promoter of the activated gene
or, occasionally (29), in an intron near the promoter. In
addition, GRE-bound glucocorticoid receptors can interact with other transcription factors that bind to
nearby elements to produce synergistic activation of
reporter genes, with the degree of synergism being
inversely related to the intrinsic strength of the GRE
(30, 31). When we searched the known sequences of
the rat proenkephalin gene and 5'-flanking region (Refs.
3 and 32 and this study) for those resembling the
positive GRE consensus GGTACAnnnTGTTCT (22),
we found no good match in the 5'-flanking region, but
did find three GRE-like sequences in intron A, namely
GTTCCTcatTGTCCT (bases 238-252), ATGGCAtcc-
Regulation of Proenkephalin Gene Transcription
1075
A. C6[pREJCAT(-2700/+53)-Neo]:
CAT mRNA
pENK mRNA
DF
28S —
18S —
Relative
Amount: 1
0.8
12 3.8
1
1.1
1.8 4.7
B. C6[pREJCAT(-2700/+703)-Neo]:
CAT mRNA
PENK mRNA
D
F DF
DF
28S —
18S —
-
Relative
Amount: 1
1
2.5
1
•
1.9
#
3.4 4.4
Fig. 5. Effect of Dexamethasone and Forskolin on the Abundance of CAT and Proenkephalin Transcripts in Cycloheximide-Treated
C6[Enkephalin-CAT-neo] Stable Transfectant Cells
Pools of stably transfected lines were grown to confluency. Cycloheximide (75 MM) was added to all cultures 40 min before the
addition of drugs. Duplicate flasks of cells were treated for 4 h with no drug (O), 1 MM dexamethasone (D), 20 MM forskolin (F), or
both (DF), as indicated. Total RNA was prepared from each culture. Northern blots of 25 ^g total RNA were separately hybridized
with CAT and proenkephalin cDNA probes, as shown. Single representative lanes are shown for each condition, although duplicate
cultures were analyzed. X-Ray exposure times are as follows: CAT, 11 h at - 8 0 C; and proenkephalin, 17 h at 20 C in A and 23
h at - 8 0 C in B. Radioactive signals were quantitated by a Betascope, and duplicates were averaged. The averaged signals,
expressed as a ratio to the control values (O), are listed below the representative lanes.
TGTCCT (bases 398-384, lower strand), and
ATGGCAtccTGTTCT (bases 441-427, lower strand).
Interestingly, the first two sequences are largely conserved in intron A of the mouse proenkephalin gene,
while the third is in a stretch of 43 bp missing in the
mouse gene (32).
The studies with proenkephalin/CAT chimeric genes
(Figs. 4 and 5 and Table 2) failed to demonstrate a
strong positive GRE, or one cooperating positively with
the cAMP-inducible enhancer, within 5800 bp of 5'flanking sequence, exon I, and intron A of the rat
proenkephalin gene. Instead, we found an unexpected
inhibition by dexamethasone of forskolin-stimulated
CAT activity of constructs with at least 190 bp of 5'flanking sequence. The only suggestion of potentiation
by dexamethasone of cAMP-induced CAT activity was
in some experiments with short constructs (145 or 95
bp of 5'-flanking sequence), in which the forskolin response was blunted. The data do not pinpoint a single
region responsible for this negative effect. The effect is
apparent, but marginal, with 190 bp of 5'-flanking sequence, but is greater with 437 bp or more, suggesting
multiple negative regulatory elements. Also, the negative effect clearly is not mediated by an element in intron
A, in spite of the putative GRE sequences discussed
above.
Vol 5 No. 8
MOL ENDO-1991
1076
Several mechanisms by which glucocorticoids can
repress transcription have been described. In some
cases, binding of the glucocorticoid receptor to a GRE
is believed to interfere with the binding of other transcription factors, such as those conferring activation by
cAMP, to their respective elements (33,34). In addition,
several groups have recently demonstrated a functional
antagonism between the glucocorticoid receptor and
components of the AP-1 transcription factor, particularly Jun, mediated by protein-protein interactions and
not necessarily requiring DNA binding (35-38). In one
study, the activity of a GRE could be made to vary from
positive to negative by manipulation of the ratio between the AP-1 components Fos and Jun (38). The
relevance of these mechanisms to the negative effect
of glucocorticoids seen with proenkephalin/CAT constructs is uncertain, because some of the negative
glucocorticoid effect maps to 5'-flanking sequences upstream from base -190, while maximal cAMP induction
is achieved with bases down-stream from -190 (Fig.
4). The ENKCRE2 element, which is centered at bases
- 8 4 to - 9 0 in the rat gene, has been reported to bind
AP-1 in vitro (13), but this element is probably not
involved in glucocorticoid inhibition, since the expression of pREJCAT(-145/+53) is not inhibited by dexamethasone. The up-stream sequence from bases -190
to -437 contains at least five sequences that resemble
the consensus AP-1 heptanucleotide core in at least
five of seven bases, but these putative elements are
not required for cAMP activation and, therefore, probably not in the inhibition thereof by glucocorticoid.
Unexplained inhibition, similar to that reported here,
by glucocorticoid of the cAMP-stimulated expression of
CAT constructs containing the 5'-flanking sequence of
a gene normally activated in vivo by glucocorticoid has
been reported previously for the GH gene of several
species (39). In that case, dexamethasone alone weakly
stimulated CAT expression, in contrast to the present
study.
Our results are consistent with several plausible explanations. One is that the proenkephalin gene may
contain negative GREs in the sequences tested, but
also an overriding positively cooperative GRE in an
atypical location outside the region tested. Another
possibility is that the proenkephalin gene contains one
or more GREs that interact negatively in the context of
the chimeric constructs, but positively in the context of
the native gene. A difference in contexts is suggested
by the fact that the stimulation by forskolin (cAMP) is
more robust for the chimeric gene than for the endogenous gene within the same cells (Fig. 5). One may
speculate that the constructs lack a negative element
that normally attenuates the action of cAMP in the
absence of glucocorticoid. Finally, if, in fact, the enkephalin gene has no positive GRE, we are left with the
possibility that glucocorticoids potentiate proenkephalin
gene expression by activating or stabilizing a required
transcription factor via a mechanism not involving protein synthesis.
An ancillary finding in this study is that maximal
activation of the rat proenkephalin gene by cAMP elevation requires 5'-flanking sequences between bases
-145 and -190, in contrast to the original study of
Comb et al. (12), in which the response of the human
proenkephalin promoter to cAMP elevation in CV-1 cells
was no greater with a 5' end point of base -193 than
with base -107. While this discrepancy may be due
merely to species or cell type differences, our data
suggest that the activity of the rat cAMP-inducible
enhancer may be potentiated at least 2-fold by an
additional factor(s) binding up-stream. The only described regulatory elements between -190 and the
ENKCRE1 element of the rat gene are two BETA
sequences, which resemble the nuclear factor KB consensus sequence (40).
We believe that the transcriptional regulation of the
proenkephalin gene reported here is physiologically relevant. Our results and those of others (described in the
Introduction) suggest that the normal control of this
gene, and most likely of other neurotransmitter genes,
by physiological regulators, such as neuronal activity
and cAMP, requires the presence of a basal physiological concentration of glucocorticoid. Consistent with this
idea is the finding that the developmental appearance
of enkephalin peptides in embryonic rat adrenal chromaffin cells is temporally correlated with the developmental onset of glucocorticoid production by adrenal
cortical cells (41). One can predict, therefore, that inadequate amounts of glucocorticoid could result in
impaired development or function of enkephalinergic
neurons.
MATERIALS AND METHODS
Cell Culture
C6 rat glioma cells (American Type Culture Collection, Rockville, MD; passages 56-63) or NIH 3T3 mouse fibroblast cells
were cultured at 37 C in 175- or 75-cm2 flasks containing 90%
Dulbecco's Modified Eagle's Medium (DME; Gibco, Grand
Island, NY) and 5-10% fetal calf serum (Armour, Kankakee, IL), which had been treated at 55 C for 1 h with 10 mg/
ml activated charcoal and 1 mg/ml Dextran-T40 (Pharmacia,
Piscataway, NJ) to remove endogenous steroids. For experiments other than those involving DNA transfection, C6 cells
were grown to a confluent density of 1-3 x 10s cells/flask
before initiation of drug treatments. Stock solutions (1000-fold
concentrated) of 20 mM forskolin (Calbiochem, La Jolla, CA)
and 1 mM dexamethasone (Sigma, St. Louis, MO) were prepared in 95% ethanol. Ethanol was added when necessary to
maintain an identical concentration (0.19-0.28%) in media of
control and experimental cultures. A 100-fold concentrated
stock solution of 7.5 mM cycloheximide (Sigma) was prepared
in PBS and sterilized by filtration.
Proenkephalin and CAT mRNA Analyses
Confluent C6 cells were lysed and sonicated in 4 M guanidinium
thiocyanate solution (42). Total RNA was purified by centrifugation through a cushion of 5.7 M CSCI and 0.1 M EDTA at 20
C [usually a 2-ml cushion in a Beckman SW55 Ti rotor (Palo
Alto.CA) at 42,000 rpm for 12-14 h], followed by phenolchloroform extraction and ethanol precipitation. RNA quantitation was determined by absorbance at 260 nm. Dot blots
Regulation of Proenkephalin Gene Transcription
were prepared by immobilization of four quantities (0.25-1.0
M9) of each tested RNA preparation onto a GeneScreen Plus
nylon membrane (DuPont-New England Nuclear, Boston, MA)
by the use of a dot blot manifold, as described previously (6).
Blots were hybridized with the 32P-labeled 930-bp Sacl-Smal
fragment of the rat proenkephalin cDNA clone pYSECI, as
previously described (7). Autoradiograms were quantitated by
densitometry.
Northern blot analyses were performed according to standard methods (43), with 6% formaldehyde-1% agarose gels
and capillary transfer to Nytran membranes (Schleicher and
Schuell, Keene, NH). The 675-bp H/ndlll-Scal fragment of
pPLQCAT (44), containing nearly all of the coding region of E.
coli CAT cDNA, was 32P labeled by random priming. Radioactive bands on the blots were quantitated directly by either
liquid scintillation counting (CAT mRNA) or a Betascope (Betagen, Cambridge, MA; proenkephalin mRNA).
Nuclear Run-On Transcription Analysis
The procedures of Greenberg ef a/. (45), with the modifications
described below, were used for the experiment reported in
Fig. 1. C6 cultures in 175-cm2 flasks (~1 x 108 cells/flask)
were treated with drugs (one flask per condition). Cells were
harvested in PBS (without Ca2+ and Mg2+) and pelleted. To
each cell pellet were added 4 ml ice-cold nuclear lysis buffer
[0.5% Nonidet P-40, 10 mM Tris-HCI (pH 7.4), 10 mM NaCI,
and 3 mM MgCI2]. After 5 min on ice, the cells were further
lysed in a Dounce homogenizer (B pestle, Kontes Co., Vineland, NJ; 10 strokes). Nuclei were pelleted by centrifugation
for 5 min at 500 x g and washed twice by resuspension and
recentrifugation in 4 ml nuclear lysis buffer. The nuclei were
then suspended in 0.12 ml 40% glycerol, 50 mM Tris-HCI (pH
8.0), 5 mM MgCI2, and 0.1 mM EDTA and stored over liquid N2
until use.
Membranes containing immobilized probes were prepared
as follows. Plasmids (twice CsCI-banded) were denatured by
treatment with 0.3 M NaOH at 65 C for 30 min, neutralized,
and filtered onto nitrocellulose (BA85, Schleicher and Schuell)
in a slot blot manifold (2 ^g DNA/slot): 1) pYSECI, containing
a 936-bp rat proenkephalin cDNA insert in the vector pSP65
(7); 2) a plasmid containing a 1200-bp insert of chicken glyceraldehyde 3-phosphate dehydrogenase (GAPDH) cDNA in the
vector pBR322 (46); and 3) the vector plBI31 (International
Biotechnologies, Inc., New Haven, CT) to measure nonspecific
hybridization. Membranes were baked at 80 C for 2 h under
vacuum and preequilibrated with 1 -fold concentrated hybridization buffer [50% formamide, 0.9 M NaCI, 50 mM sodium
phosphate (pH 7.7), 5 mM EDTA, 0.1% Ficoll, 0.1% polyvinylpyrollidone, 0.1% BSA, 0.4% sodium dodecyl sulfate (SDS),
100 jug/ml denatured herring sperm DNA, 50 MO/™' wheat
germ tRNA, 0.05% Na4P2O7, and 0.1 mM unlabeled UTP] for
6-12 hat 45 C.
Nuclear run-on transcription reactions contained, in a final
volume of 0.4 ml, 108 nuclei thawed immediately before use,
42.5 mM Tris-HCI (pH 8.0), 5.5 mM MgCI2,150 mM KCI, 1 mM
dithiothreitol, 100 U/ml RNAsin ribonuclease inhibitor (Promega, Madison, Wl), 0.5 mM each of ATP, CTP, and GTP, and
150-250 nC\ [a-32P]UTP (800 Ci/mmol; New England Nuclear).
Reaction mixtures were incubated for 30 min at 29 C. Reactions were stopped by the addition of 1.6 ml 5 M guanidinium
thiocyanate solution (42). Under these conditions, approximately 50% of the total incorporation of [a-32P]GTP into RNA
was blocked by 2.5 ^g/m\ a-amanitin, which at this concentration specifically inhibits RNA polymerase-ll. The suspensions
were sonicated for 30 sec to disperse chromatin, and [32P]
RNA was isolated exactly as described for nonradioactive
RNA above. Each [32P]RNA pellet was dissolved in 0.2 ml
0.25% SDS, 10 mM Tris-HCI (pH 7.4), and 1 mM EDTA. RNA
was slightly fragmented by treatment with 0.05 ml 1 M NaOH
for 10 min on ice, neutralized by the addition of 0.1 ml 1 M
HEPES, ethanol precipitated and redissolved as before, and
heated for 10 min at 65 C. Similar amounts of radioactivity (~1
1077
x 106 cpm) of each sample were used for subsequent hybridization. The volumes were reduced under vacuum to 0.07 ml.
The [32P]RNA solutions were heated as before and then diluted
with 0.63 ml 1.11-fold concentrated hybridization buffer. Each
RNA solution was hybridized with DNA probes immobilized on
nitrocellulose strips, prepared as described above, for 66-72
h at 42 C in small heat-sealed pouches. Membranes were
subsequently washed with several changes of 1 x SSPE [0.18
M NaCI, 10 mM NaPO4 (pH 7.7), and 1 mM EDTA]-0.1% SDS
over an 80- to 90-min period at 55-65 C and finally with 0.2
x SSPE-0.1% SDS for 60 min at 63-67 C, then exposed to
Kodak XAR film (Eastman Kodak, Rochester, NY) at - 7 0 C
with DuPont Lightning Plus intensifying screens (Wilmington, DE). Autoradiographic densities were quantitated with a
CS-930 scanning densitometer (Shimadzu, Columbia, MD) and
integration of peak areas. Each proenkephalin signal was
normalized to the GAPDH signal on the same strip to correct
for variations in hybridization efficiency. GAPDH transcription
was not reproducibly affected by the regulators tested.
In control experiments, proenkephalin and /3-actin transcription autoradiographic signals were eliminated, as expected, by
inclusion in the transcription reactions of 2.5 iiQ/m\ a-amanitin,
which specifically inhibits RNA polymerase-ll (data not shown).
Alternative procedures for this assay yielded essentially
similar results. These included more thorough nuclear purification by sedimentation through a cushion of 1.7 M sucrose,
RNA purification by proteinase-K-DNase-phenol, and hybridization of transcripts to purified cDNA insert probes (minus
vector sequences).
Assay of Intracellular cAMP
Cells were cultured and treated with drugs in 60-mm diameter
wells containing 5 ml medium. Culture medium then was
removed, and cell monolayers were washed twice with PBS.
Cyclic AMP was purified from the trichloroacetic acid-soluble
cell extract, as previously described (47), and quantitated (48).
Cell protein was determined on NaOH-solubilized trichloroacetic acid precipitates with the BCA (bicinchoninic acid) protein
assay reagent (Pierce Chemical Co., Rockford, IL), with BSA
as the standard.
Proenkephalin/CAT Plasmids
The source of proenkephalin genomic DNA was plasmid subclones (pRESSI and pRESS2) containing DNA from two genomic clones isolated by H. Higuchi and S. Sabol, as previously
described (32), from a Wistar rat liver genomic library (49),
kindly provided by Dr. G. Scherer (Albert-Ludwig University,
Freiburg, Germany). Appropriate restriction fragments containing the proenkephalin promoter were ligated, as described
below, into the polylinker regions of the promoterless
pSVoCAT derivatives pPLFCAT and pPLRCAT, constructed
(44) and kindly supplied by Dr. Mark Magnuson (Vanderbilt
University, Nashville, TN). In the resultant constructs, theEscherichia coli CAT gene is placed under the direction of the
proenkephalin promoter and up-stream enhancer sequences.
Standard recombinant DNA methodology (43) was employed
for the construction of these plasmids. E. coli HB101 was the
host strain used for plasmid production.
1) A proenkephalin genomic fragment extending from a Kpn\
site 5.8 kilobases (kb) up-stream from the transcription start
site to a Sacl site 53 bp down-stream from the start site was
ligated into Kpn\- and Sacl-cut pPLRCAT to produce the
construct designated pREJCAT(-5800/+53).
2) The fragment of the proenkephalin gene extending from
a Sa/I site (derived from the EMBL3 vector) of the genomic
clone) 2700 bp up-stream from the transcription start site to
the Sacl site at +53 was ligated into Sa/I- and Sacl-cut
pPLFCAT to produce pREJCAT(-2700/+53). The same genomic fragment was also ligated into similarly cut pPLRCAT,
containing the reverse form of the polylinker of pPLFCAT to
yield pREJCAT(+53/-2700). The pREJCAT(-2700/+53) was
Vol 5 No. 8
MOL ENDO-1991
1078
further digested with H/ndlll to excise three H/ndlll fragments
(totaling 1600 bp) of the proenkephalin gene up-stream region.
The plasmid was recircularized to yield pREJCAT(-1050/
+53).
3) The pRESS-2 (32) was linearized by digestion at a unique
A/col site at the beginning of exon II. Removal of 5' overhangs
was achieved with mung bean nuclease (New England Biolabs); this step eliminated the ATG initiation codon. Sa/I linkers
(New England Biolabs, Beverly, MA) were added to the blunt
ends with T4 DNA ligase. After digestion with Sa/I, a 3.45-kb
Sa/I-Sa/I proenkephalin gene fragment was ligated to a
Sa/l-digested dephosphorylated pPLFCAT plasmid. The resultant recombinant plasmids with the proenkephalin insert
in the forward and reverse directions were termed
pREJCAT(-2700/+703) and pREJCAT(+703/-2700), respectively. Sequencing of the former clone showed that although one base was deleted during cloning, the splice acceptor site of intron A was intact.
4) CAT constructs containing proenkephalin genomic inserts
with up-stream termini at positions -437, -190, -145, and
- 9 5 and each with a down-stream terminus at +53 were
constructed by ligation of DNA amplified by the polymerase
chain reaction (PCR) into pPLFCAT. PCR mixtures contained
10 ng EcoRI-linearized pREJCAT(-2700/+53) template, a
down-stream 26-base oligonucleotide primer including the
Sacl site of the proenkephalin gene (positions - 4 8 to -53),
and one of several up-stream oligonucleotide primers, each
consisting of a Sa/I adapter sequence joined to 18 bases
corresponding to the desired up-stream terminus of the insert. Amplified DNA, which was of the expected length for
each set of primers by agarose gel electrophoresis, was
restricted with Sa/I and Sacl and ligated into Sa/I- and Saclcut pPLFCAT, resulting in the constructs pREJCAT(-437/
+53), pREJCAT(-190/+53, pREJCAT(-145/+53), and
pREJCAT(-95/+53).
DNA Sequencing
Portions of pREJCAT(-5800/+53) and (-1050/+53) plasmids
were sequenced with the Sequenase kit (U.S. Biochemical
Corp., Cleveland, OH) according to the manufacturer's recommendations for double stranded sequencing. Oligodeoxynucleotide primers were synthesized by an Applied Biosystems 380B synthesizer (Foster City, CA) and used for chain
elongation in the up-stream direction.
DNA Transfections
Cells were seeded at 5 x 105 cells/100-mm dish 24-36 h
before transfection. Cells were cultured at 37 C in 90% DME
supplemented with 20 ITIM HEPES, pH 7.3, and 10% charcoal/
dextran-treated fetal calf serum in a 4% CO2 incubator. CaPCv
DNA precipitates, generally containing 10 ^g test plasmid and
2 ng pRSVjSgal plasmid, were prepared as previously described (50) and applied to the cells in DME. After incubation
for 5-7 h, cells were shocked for 2 min with 15% glycerol in
HSP buffer (10 mM KCI, 280 mM NaCI, 12 ITIM glucose, 1.5
mM Na2HPO4, and 50 mM HEPES, pH 7.1). Approximately 48
h after DNA addition, the cells were treated with forskolin,
dexamethasone, or both (as described above under Cell Culture) and harvested at different times for CAT determinations.
Most experiments included a positive control transfection with
pRSVCAT DNA.
All plasmids used in transfection assays were banded twice
by CsCI-ethidium bromide density gradient centrifugation. By
gel electrophoretic analysis, the final plasmid preparations
were found to consist almost entirely of supercoiled DNA.
CAT Assays
Cells were harvested in PBS and lysed by four cycles of
freezing and thawing. The lysates were centrifuged at 12,000
x g for 6 min, and the supernatants were assayed for protein
content by the BCA reagent. In many experiments, half of
each lysate was heated at 60 C for 10 min to inactivate
endogenous deacetylase activity; however, heating did not
change the total CAT activity of C6 lysates. CAT assays were
performed essentially as previously described (50), generally
with 50 fig protein and incubation of 2 h at 37 C. Acetylated
[14C]chloramphenicol resolved by TLC was quantitated by
liquid scintillation counting. The /3-galactosidase activity, measured as the hydrolysis of ort/jo-nitrophenyl-/8-D-galactopyranoside, was performed with 50 ng unheated cell lysate protein
for 4 h (C6 cells) or 15-60 min (NIH 3T3 cells) at 37 C, as
previously described (43). The percent acetylation of chloramphenicol was divided by the /3-galactosidase activity with
the identical amount of protein to obtain a relative CAT activity
normalized for variations in transfection efficiency.
Production of C6[Enkephalin-CAT/neo] Stable
Transfectant Cell Lines
C6 cells (8 x 10" cells/100-mm dish) were cotransfected with
7 fig pREJCAT plasmid and 0.7 ^g pRSVNeo (51), as described above. After 48 h, the cells were transferred to medium
supplemented with 725 /ug/ml G418 (Geneticin, Gibco) and
grown for 2 weeks under continuous G418 selection. At the
end of this period, more than 80 G418-resistant colonies were
pooled.
Acknowledgments
We thank Drs. Mark Magnuson and Daryl Granner (Vanderbilt
University, Nashville, TN) for providing the polylinker-supplemented pSVOCAT clones and Dr. James Battey (NINCDS,
Bethesda, MD) for providing the glyceraldehyde 3-phosphate
dehydrogenase probe. We thank Dr. Sadamitsu Asoh (NIH)
for preparation of pRSV/3-gal. We thank summer students
Laura Minna (University of California, Berkeley) and Stacey
Anderson (Harvard University) for excellent assistance. We
also thank Dr. Battey for a critical reading of the manuscript.
Received March 20, 1991. Rerevision received June 5,
1991. Accepted June 5,1991.
Address requests for reprints to: Dr. Steven L. Sabol,
National Institutes of Health, Building 36, Room 1C06, Bethesda, Maryland 20892.
REFERENCES
1. Noda M, Teranishi Y, Takahashi H, Toyosato M, Notake
M, Nakanishi S, Numa S 1982 Isolation and structural
organization of the human preproenkephalin gene. Nature
297:431-434
2. Comb M, Rosen H, Seeburg P, Adelman J, Herbert E
1983 Primary structure of the human proenkephalin gene.
DNA 2:213-229
3. Rosen H, Douglass J, Herbert E 1984 Isolation and characterization of the rat proenkephalin gene. J Biol Chem
259:14309-14313
4. Gubler U 1987 Enkephalin genes. In: Habener JF (ed)
Molecular Cloning of Hormone Genes. Humana Press,
Clifton, pp 229-276
5. Eiden LE, Hotchkiss AJ 1983 Cyclic adenosine monophosphate regulates vasoactive intestinal polypeptide and
enkephalin biosynthesis in cultured bovine chromaffin
cells. Neuropeptides 4:1-9
6. Yoshikawa K, Sabol SL 1986 Expression of the enkephalin precursor gene in C6 rat glioma cells: regulation by
Regulation of Proenkephalin Gene Transcription
/3-adrenergic agonists and glucocorticoids. Mol Brain Res
1:75-83
7. Yoshikawa K, Sabol SL 1986 Glucocorticoids and cyclic
AMP synergistically regulate the abundance of preproenkephalin messenger RNA in neuroblastoma x glioma hybrid cells. Biochem Biophys Res Commun 139:1-10
8. Yoshikawa K, Aizawa T, Nozawa A 1989 Phorbol ester
regulates the abundance of enkephalin precursor mRNA
but not of amyloid /3-protein precursor mRNA in rat testicular peritubular cells. Biochem Biophys Res Commun
161:568-575
9. Siegel RE, Eiden LE, Affolter HU1985 Elevated potassium
stimulates enkephalin biosynthesis in bovine chromaffin
cells. Neuropeptides 6:543-552
10. Naranjo JR, Mocchetti I, Schwartz JP, Costa E 1986
Permissive effect of dexamethasone on the increase of
proenkephalin mRNA induced by depolarization of chromaffin cells. Proc Natl Acad Sci USA 83:1513-1517
11. Kley N, Loeffler JP, Pittius CW, Hollt V 1986 Proenkephalin A gene expression in bovine adrenal chromaffin
cells is regulated by changes in electrical activity. EMBO
J 5:967-970
12. Comb M, Birnberg NC, Seasholtz A, Herbert E, Goodman
HM 1986 A cyclic AMP- and phorbol ester-inducible DNA
element. Nature 323:353-356
13. Comb M, Mermod N, Hyman SE, Pearlberg J, Ross ME,
Goodman HM 1988 Proteins bound at adjacent DNA
elements act synergistically to regulate human proenkephalin cAMP inducible transcription. EMBO J 7:37933805
14. Hyman SE, Comb M, Pearlberg J, Goodman HM 1989 An
AP-2 element acts synergistically with the cyclic AMP and
phorbol ester-inducible enhancer of the human proenkephalin gene. Mol Cell Biol 9:321-324
15. Nguyen TV, Kobierski L, Comb M, Hyman SE 1990 The
effect of depolarization on the expression of the human
proenkephalin gene is synergistic with CAMP and dependent upon a cAMP-inducible enhancer. J Neurosci
10:2825-2833
16. Inturrisi CE, Branch AD, Robertson HD, Howells RD,
Franklin SO, Shapiro JR, Calvano SE, Yoburn BC 1988
Glucocorticoid regulation of enkephalins in cultured rat
adrenal medulla. Mol Endocrinol 2:633-640
17. LaGamma EF, Adler JE 1987 Glucocorticoids regulate
adrenal opiate peptides. Mol Brain Res 2:125-130
18. Yanase T, Nawata H, Haji M, Kato K, Ibayashi H 1988
Dexamethasone increases mRNA of preproenkephalin A
in cultured human pheochromocytoma cells. Horm Metab
Res 20:187-189
19. Matrisian LM, Rautmann G, Breathnach R 1988 Differential expression of preproenkephalin and transin mRNAs
following oncogenic transformation: evidence for two
classes of oncogene-induced genes. Oncogene Res
2:251-262
20. Chao HM, McEwen BS 1990 Glucocorticoid regulation of
preproenkephalin mRNA in the rat striatum. Endocrinology 126:3124-3130
21. Stachowiak MK, Lee PHK, Rigual RJ, Viveros OH, Hong
JS 1988 Roles of the pituitary-adrenocortical axis in control of the native and cryptic enkephalin levels and proenkephalin mRNA in the sympathoadrenal system of the rat.
Mol Brain Res 3:263-273
22. Beato M 1989 Gene regulation by steroid hormones. Cell
56:335-344
23. Paek I, Axel R 1987 Glucocorticoids enhance stability of
human growth hormone mRNA. Mol Cell Biol 7:14961507
24. Lee SW, Tsou AP, Chan H, Thomas J, Petrie K, Eugui
EM, Allison AC 1988 Glucocorticoids selectively inhibition
the transcrption of the interleukin 1 /? gene and decrease
the stability of interleukin 1(8 mRNA. Proc Natl Acad Sci
USA 85:1204-1208
25. Sasaki K, Cripe TP, Koch SR, Andreone TL, Petersen DD,
Beale EG, Granner DK 1984 Multihormonal regulation of
1079
phosphenolpyruvate carboxykinase gene transcription.
The dominant role of insulin. J Biol Chem 259:1524215251
26. Meisner H, Loose DS, Hanson RW 1985 Effect of hormones on transcription of the gene for cytosolic phosphoenolpyruvate carboxykinase (GTP) in rat kidney. Biochemistry 24:421-425
27. Nakamura T, Niimi S, Nawa K, Noda C, Ichihara A, Takagi
Y, Anai M, Sakaki Y 1987 Multihormonal regulation of
transcription of the tryptophan 2,3-dioxygenase gene in
primary cultures of adult rat hepatocytes with special
reference to the presence of a transcriptional protein
mediating the action of glucocorticoids. J Biol Chem
262:727-733
28. Lewis EJ, Harrington CA, Chikaraishi DM 1987 Transcriptional regulation of the tyrosine hydroxylase gene by
glucocorticoid and cyclic AMP. Proc Natl Acad Sci USA
84:3550-3554
29. Slater EP, Rabenau O, Karin M, Baxter JD, Beato M 1985
Glucocorticoid receptor binding and activation of a heterologous promoter by dexamethasone by the first intron
of the human hormone gene. Mol Cell Biol 5:2984-2992
30. Strahle U, Schmid W, Schutz G 1988 Synergistic action
of the glucocorticoid receptor with transcription factors.
EMBO J 7:3389-3395
31. Schule R, Muller M, Kaltschmidt C, Renkawitz R 1988
Many transcription factors interact synergistically with
steroid receptors. Science 242:1418-1420
32. Kilpatrick DL, Zinn SA, Fitzgerald M, Higuchi H, Sabol SL,
Meyerhardt J 1990 Transcription of the rat and mouse
proenkephalin genes is initiated at distinct sites in spermatogenic and somatic cells. Mol Cell Biol 10:3717-3726
33. Akerblom IE, Slater EP, Beato M, Baxter JD, Mellon PL
1988 Negative regulation by glucocorticoids through interference with cAMP responsive enhancer. Science
241:350-353
34. Drouin J, Trifiro, MA Plante RK, Nemer M, Ericksson P,
Wrange o 1989 Glucocorticoid receptor binding to a
specific DNA sequence is required for hormone-dependent inhibition of proopiomelanocortin gene transcription.
Mol Cell Biol 9:5305-5314
35. Jonat C, Rahmsdorf HJ, Park KK, Cato AC, Gebel S,
Ponta H, Herrlich P 1990 Antitumor promotion and antiinflammation: down-modulation of AP-1 (FosJJun) activity
to glucocorticoid hormone. Cell 62:1189-1204
36. Yang-Yen HF, Chambard JC, Sun YL, Smeal T, Schmidt
TJ, Drouin J, Karin M 1990 Transcriptional interference
between c-jun and the glucocorticoid receptor: mutual
inhibition of DNA binding due to direct protein-protein
interaction. Cell 62:1205-1215
37. Schule R, Rangarajan P, Kliewer S, Ransone LJ, Bolado
J, Yang N, Verma IM, Evans RM 1990 Functional antagonism between oncoprotein c-jun and the glucocorticoid
receptor. Cell 62:1217-1226
38. Diamond Ml, Miner JN, Yoshinaga SK, Yamamoto KR
1990 Transcription factor interactions: selectors of positive or negative regulation from a single DNA element.
Science 249:1266-1272
39. Brent GA, Harney JW, Moore DD, Larsen PR 1988 Multihormonal regulation of the human, rat, and bovine
growth hormone promoters: differential effects of 3':5'cyclic adenosine monophosphate, thyroid hormone, and
glucocorticoids. Mol Endocrinol 2:792-798
40. Korner M, Rattner A, Mauxion F, Sen R, Citri Y 1989 A
brain-specific transcription activator. Neuron 3:563-572
41. Henion PD, Landis SC 1990 Asynchronous appearance
and topographic segregation of neuropeptide-containing
cells in the developing rat adrenal medulla. J Neurosci
10:2886-2896
42. Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ
1979 Isolation of biologically active RNA from sources
enriched in ribonuclease. Biochemistry 18:5294-5299
43. Sambrook J, Fritsch EF, Maniatis T 1989 Molecular Clon-
MOL ENDO-1991
1080
ing-A Laboratory Manual, ed 2. Cold Spring Harbor Laboratory, Cold Spring Harbor
44. Magnuson MA, Quinn PG, Granner DK 1987 Multihormonal regulation of PEPCK-CAT fusion genes: insulin's
effects oppose those of cAMP and dexamethasone. J
Biol Chem 262:14917-14920
45. Greenberg ME, Greene LA, Ziff EB 1985 Nerve growth
factor and epidermal growth factor induce rapid transient
changes in proto-oncogene transcription in PC12 cells. J
Biol Chem 260:14101-14110
46. Dugaiczyk A, Haron JA, Stone EM, Dennison OE, Rothblum KN, Schwartz RJ 1983 Cloning and sequencing of
a DNA copy of glyceraldehyde-3-phosphate dehydrogenase mRNA isolated from chicken muscle. Biochemistry
22:1605-1613
Vol 5 No. 8
47. Sharma S, Nirenberg M, Klee WA 1975 Morphine receptors as regulators of adenylate cyclase activity. Proc Natl
Acad Sci USA 72:590-594
48. Gilman AG 1970 A protein binding assay for adenosine
3':5'-cyclic monophosphate. Proc Natl Acad Sci USA
67:305-312
49. Shinomiya T, Scherer G, Schmid W, Zentgraf H, Schutz
G 1984 Isolation and characterization of the rat tyrosine
aminotransferase gene. Proc Natl Acad Sci USA 81:13461350
50. Gorman CM, Moffat LF, Howard BH 1982 Recombinant
genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol Cell Biol 2:1044-1051
51. Gorman C, Padmanabhan R, Howard BH 1983 High
efficiency DNA-mediated transformation of primate cells.
Science 221:551-553