(Received for publication, September 23, 1996, and in revised form, January 29, 1997)
From the Department of Chemistry and
§ Biological Sciences, University of Nebraska,
Lincoln, Nebraska 68588-0304
A rat genomic library constructed in -EMBL3
(SP6/T7) vector (CLONTECH) was screened using
32P-labeled rat p67 cDNA. A clone containing a
segment of 5
-upstream region of p67 genomic DNA was obtained. The DNA
(about 1.7 kilobase pairs) was isolated and characterized. Sequence
analysis of this DNA fragment showed that the 898 base pairs at the
5
-end of the upstream region was identical to several long
interspersed nucleotide sequences. One hundred forty-eight base pairs
at the 3
-end contained the beginning of the first exon including the
ATG initiator codon. The remaining 652 base pairs in between contained
two AT-rich regions and several regulatory sequences. The mRNA
initiation site was identified at 89 base pairs upstream from the
translation start codon. The DNA fragment was also analyzed by
transient transfection. When linked to a firefly luciferase reporter
gene, this fragment enhanced transcription in a rat hepatoma cell line
(KRC-7). Using a series of deletions in the DNA, the minimum essential
promoter region (from
177 to
60) was identified. The promoter
activity was also enhanced by treatment with phorbol 13-myristate
12-acetate (PMA). This enhancement required an AP-1 sequence (
298 to
292; 5
-TGACTCA-3
) and a similar sequence (
97 to
88;
5
-ATGACATCAT-3
). Deletion of either of these sequences significantly
reduced PMA enhancement. Deletion of both of these sequences almost
completely eliminated PMA enhancement.
Protein synthesis in animal cells is regulated by phosphorylation
of a key peptide chain initiation factor,
eIF-2.1 Animal cells contain eIF-2 kinases
such as heme-regulated inhibitor and double-stranded RNA-activated
protein kinase. Under certain physiological conditions, these eIF-2
kinases phosphorylate specifically the -subunit of eIF-2. This
inactivates eIF-2 activity and inhibits protein synthesis (for recent
reviews, see Refs. 1-3). Animal cells also contain a 67-kDa
glycoprotein, p67 (4-8). p67 protects eIF-2 from inhibitory
phosphorylation by eIF-2 kinases. This promotes protein synthesis in
the presence of active eIF-2 kinase(s) present in animal cells.
An important characteristic of p67 is that the level of this protein varies widely under different physiological conditions, and this level correlates directly with the protein synthesis activity of the cells (6-8). There are indications that the p67 level in the cells under certain physiological conditions is regulated at the transcriptional level (8-9). The p67 transcription shuts off after serum starvation in a tumor hepatoma cell line (KRC-7). The same serum-starved cell line regains p67 transcription after addition of a mitogen phorbol 13-myristate 12-acetate (PMA).
To identify the regulatory sequences in p67 transcription, we have now
cloned a segment of the 5-upstream region of p67 genomic DNA. In the
present paper, we describe the characterization of this DNA fragment
and identification of the essential promoter region and the
PMA-responsive sequences.
Primers
The primers used in different experiments are listed in Table I. The primers were synthesized using the facilities of the DNA Synthesis Laboratory at the University of Nebraska, Lincoln, and Life Technologies, Inc. The preparation of pGEM-p67 cDNA has been described (10).
|
Preparation of 32P-Labeled Rat p67 cDNA
A 290-bp p67 cDNA fragment was prepared using PCR. The
pGEM-p67 cDNA was used as template along with two primers, A and B (Table I). The amplified DNA fragment was then purified and
random-labeled using [-32P]dATP following standard
experimental procedures.
Isolation and Sequencing of the 5-Upstream Region of the p67
Genomic Clone
A rat genomic library constructed in EMBL3 (SP6/T7) vector
(CLONTECH) was screened using the 290-bp
32P-labeled rat p67 cDNA. A plaque was identified and
later amplified. The phage DNA was isolated from the clone and was
digested with BamHI. The digested fragments were analyzed in
a Southern blot experiment using a synthetic 70-mer oligonucleotide
probe corresponding to +72 to +142 base pairs of p67 cDNA. One
1.7-kb fragment was detected after autoradiography. The DNA fragment
was subcloned into the BamHI site of pGEM 7Zf(+) vector
(Promega) and was sequenced following Sanger's dideoxynucleotide chain
termination method (11). The DNA sequence was analyzed using the
Genetic Computer Group (GCG) sequence analysis software. This sequence
is shown in Fig. 1.
Primer Extension Analysis
The transcription start site of the rat p67 gene was analyzed by
primer extension. Total RNA from KRC-7 cells was isolated using the
guanidium isothiocyanate method (12). Approximately 10 µg of RNA was
used as a template. The primer was an 18-base oligonucleotide (primer
C, Table I) corresponding to the inverse complement of
+75 to +92 nucleotides of the sequence reported in this paper. The
primer (10 pmol) was end-labeled with [-32P]ATP and T4
polynucleotide kinase (Promega). The primer extension reaction was
carried out with avian myeloblastosis virus reverse transcriptase and
Primer Extension System (Promega) following standard procedures. The
extended product was analyzed on an 8% sequencing gel and compared
with the DNA sequence ladder and
X-174 HinfI DNA
marker.
Cell Culture
The cloned cell-line KRC-7 (a rat hepatoma cell-line; a gift from Dr. J. Koontz, University of Tennessee, Knoxville) was cultured in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) supplemented with 5% (v/v) fetal calf serum, 5% (v/v) calf serum, 100 units/ml penicillin, 50 µg/ml streptomycin, and 10 mM sodium pyruvate at 37 °C with 5% CO2.
Preparation of Different Deleted p67 5-Upstream Region
Constructs
Different deleted p67 5-upstream regions were prepared using
PCR and in some cases followed by nested deletions. The DNA sequences
were purified and subsequently subcloned into the XhoI and
HindIII sites located upstream of the promoterless
luciferase gene in the pGL3-basic vector (Promega) to generate the
pGL3-deleted p67 5
-upstream region constructs. Table II
describes different DNA templates and primer combinations used. Fig. 3
describes the deleted constructs. The preparations of individual
constructs are described below.
|
5
(i) PCR, pGL31531/+24,
pGL3
652/+24, pGL3
454/+24, and pGL3
366/+24 were prepared by
PCR.
For preparation of pGL31531/+24, one forward primer (primer D; Table
I) with an XhoI site and a reverse primer (primer H; Table
I) with a HindIII site were synthesized. These two primers and the pGEM7-1.7-kb 5
-upstream region template were used to amplify
the fragment using Taq DNA polymerase. The PCR-amplified product was purified by Wizard PCR Prep (Promega) and was digested with
XhoI and HindIII. The digested fragment was
purified and subsequently subcloned into the XhoI and
HindIII sites located upstream of the promoterless
luciferase gene in the pGL3-basic vector (Promega) to generate the
pGL3
1531/+24 construct. The DNA sequence was confirmed by Sanger's
dideoxynucleotide chain termination method (11) using Sequencing II kit
(USB).
The experimental procedures for the preparation of pGL3652/+24,
pGL3
454/+24 and pGL3
366/+24 were the same as described above. The
primers used were pGL3
652/+24, primer X + primer H; pGL3
454/+24,
primer L + primer H and pGL3
366/+24, primer F + primer H.
(ii) PCR followed by nested deletion, pGL3271/+24, pGL3
177/+24 and
pGL3
60/+24. pGL3
652/+24 was prepared as described as above (i).
This DNA construct was subjected to nested deletion by using the
Erase-A-Base system (Promega) and standard procedures.
pGL3 reverse 652/+24.
Reverse orientation of the promoter in the expression vector was
obtained by creating a HindIII site at the 5
-end and
XhoI site at the 3
-end by PCR using primer S and primer T
(Table II). During ligation, the promoter was inserted in the reverse
orientation of the pGL3 basic vector.
pGL3652/
156. The 3
-deleted construct was
generated by PCR. The primers X and G were used (Table II). The
amplified fragment was digested, purified, and subcloned.
pGL3180/
60, pGL3
327/
179
and pGL3
477/
326. The PCR overlap-extension technique of Pease
and co-workers (13) was utilized to create the specific internal
deletions (Fig. 3, panel D). Initially, two different PCR
products were synthesized. For preparation of pGL3
180/
60, the
primers used were as follows: reaction 1, primer X and primer I
(deletion at
60 to
180 nucleotides in reverse orientation, Table
I); reaction 2, primer J (deletion at
180 to
60 nucleotides in the
forward orientation, Table I) and primer H. The two PCR products were
mixed and fused together. The fused products were amplified with primer
X and primer H. Details are given in Table II. The amplified DNA was
subcloned into pGL3 basic vector.
The procedures for preparation of pGL3327/
179 and
pGL3
477/
326 were the same as described above. Details are given
in Table II. In both cases, the fused products were amplified with primer X and primer H. The amplified DNA was subcloned into pGL3 basic
vector.
pGL3AP-1 (
298 to
292),
pGL3
AP-1-like (
97 to
88), and pGL3
AP-1 (
298 to
292)/
AP-1-like (
97 to
88).
Selective deletion(s) of the two putative PMA-responsive sequences were
also performed using the overlapping PCR technique as mentioned above.
The specific primers (Table I) were designed to eliminate the
AP-1 sequence at 298 to
292 or the AP-1-like sequence at
97 to
88 or both (Table II) (Fig. 3, panel E).
Transient Transfection
Approximately, 3 × 106 KRC-7 cells were
transiently transfected with different promoter construct by
lipopolyamide-mediated transfection (LipofectAMINETM, Life
Technologies, Inc.) according to standard procedure. The constitutive
expression vector (5 µg of pSV- galactosidase from Promega) for
-galactosidase was included in the DNA mixture as a marker for
transfection efficiency (14). The cells were harvested 48 h after
transfection and assayed for luciferase and
-galactosidase activities.
Treatment of Cells for PMA Induction
The cells were transiently transfected with different deletion mutants and were serum-starved by replacing the transfection medium with serum-free medium after 8 h post-transfection. The transfected cells were then treated with 1.5 µM PMA for 2 h before harvesting them.
Luciferase and -Galactosidase Assays
The luciferase activity was measured by the luciferase assay kit
(Promega) essentially according to the manufacturer's instructions. The cells were harvested after 48 h post-transfection and lysed in
reporter lysis buffer (Promega). Aliquots were used for luciferase and
-galactosidase assays. Cell extracts from untransfected cells and
from cells transfected with the pGL3 basic vector alone without the
652/+24 inserted sequence were used as negative controls. Luciferase
activities were determined by mixing lysates with luciferase assay
buffer containing luciferin, Mg2+, and ATP at room
temperature and were analyzed immediately by a scintillation counter
(15-16). The
-galactosidase activity was analyzed using a
commercial enzyme assay system (Promega) at 420 nm. The luciferase
activity was normalized with the
-galactosidase activity. The
relative luciferase values are the average of three independent
experiments. The mean of the luciferase activities relative to pGL3
basic activity ± S.D. are presented in all the figures.
Isolation and Characterization of Rat p67 Genomic Clone
A total of 2 × 105 plaques from a genomic DNA
library were screened using a 290-bp 32P-labeled rat p67
cDNA probe. One positive plaque was isolated after three rounds of
rescreening. The phage was amplified in Escherichia coli
following the standard procedure. The phage DNA was isolated. The DNA
was digested with BamHI and analyzed by Southern blotting. A
1.7-kb fragment from the 5-end of the gene was identified. This
fragment was later subcloned into pGEM 7Zf(+) and amplified.
Sequence Analysis of the 5-Upstream Region
The sequence of the 1.7-kb DNA fragment was determined following
the Sanger's dideoxy method (11) (Fig. 1). The 5-end
of the upstream region containing 898 base pairs was identical to several LINE sequences (17-19), and 148 base pairs of the 3
-end contained the beginning of the first exon including the ATG initiation codon. In between these two regions is the proximal promoter of the p67
genome (652 base pairs).
The promoter region contains two potential TATA-like sequences between
40 and
20 and several regulatory sequences (underlined in Fig. 1).
Determination of Transcriptional Initiation Site by Primer Extension
The results of a primer extension experiment using total RNA from
KRC-7 cells and a 18-base oligonucleotide primer (primer C, Table
I) are shown in Fig. 2. A DNA fragment
was detected corresponding to a position 89 bases upstream to the ATG
codon (Fig. 2, lane 2). No extended signal was observed with
yeast tRNA (Fig. 2, lane 1). From the sequence analysis, it
was determined that the transcription start site was located about 25 bases downstream from the proximal TATA-like element (Fig. 1).
Promoter Activity of the 5-Upstream Region of the p67 Gene
To determine the promoter region responsible for transcriptional
regulation, KRC-7 cells were transiently transfected with the p67
5-upstream promoter sequence (
1531 to +24 bp) and also different
deleted promoter constructs fused upstream to a luciferase reporter
gene (Fig. 3). The promoter activities were then
determined by analysis of luciferase expression in the transfected cell
extracts. The luciferase activity was normalized with the
-galactosidase activity in the same cell extracts. The normalized
luciferase expressions in cells transfected with different pGL3
constructs (+ promoter) and pGL3 basic (
promoter) were compared. The
fold increase (× fold) with pGL3 constructs (+ promoter) over the pGL3 basic (
promoter) is presented in different figures.
Transfection of
pGL31531/+24 construct into KRC-7 cells resulted in a 10-fold
increase in luciferase expression, and pGL3
652/+24 resulted in an
11-fold increase in luciferase expression over the promoterless pGL3
basic construct. When the promoter region was put in reverse
orientation (pGL3 reverse
652/+24), no luciferase activity was
observed indicating the directionality of the promoter (Fig.
4, panel A). These results suggest that the
promoter region lies between
652 and +24.
Functional Analysis of the p67 Promoter Using Sequential Deletions
To locate the region essential for transcription
activation, a series of deletion constructs of the promoter region were
prepared and were used to transfect KRC-7 cells. The results are shown in Fig. 4, panel B. The lysates of the cells transfected
with pGL3454/+24 (panel B, 1st bar) showed a slight but
significant decrease in luciferase expression compared with
pGL3
652/+24 (panel A, 4th bar). Further deletions to
366,
277, and
177 caused decrease in luciferase expression.
However, 80% maximum luciferase expression (pGL3
652/+24) was
retained even when deletion was extended to
177 (panel B, 4th
bar). Further decrease to
60 (pGL3
60/+24, panel D, 5th
bar) led to drastic reduction (8-fold) in luciferase expression.
These results suggest that the nucleotide sequence between
177 to
60 is essential for basal transcription. Consistent with the above
suggestion, it was observed that the 3
-deleted construct
pGL3
652/
156 (Fig. 3, panel C) did not promote
luciferase expression (panel B, 5th bar).
To further define the cis regulatory
regions, internal deletions were performed (Fig. 3, panel
D). The results of this experiment are shown in Fig. 4,
panel C. The luciferase expression from cell lysates
transfected with pGL3180/
60 suggests that the essential promoter
region of the p67 gene lies between
180 and
60 (panel C). This result also correlates with the previous result in
panel B. The luciferase expression was drastically reduced
when transfected with pGL3
180/
60. These results, in agreement
with the results shown in Fig. 4 (panel B, 4th and 5th
bars), suggest that the essential promoter region lies between
180 and
60.
As shown in panel C, the luciferase expression was not
significantly reduced when transfected with pGL3377/
179 and was increased when transfected with pGL3
477/
326. The results suggest that this region (
477 to
326) may not be necessary for p67
transcription. The reason for an increase in pGL3
477/
326 is not
clear and may indicate the presence of a negative regulatory
element.
PMA Induction of the p67 Promoter
The effects of PMA addition on p67 promoter were studied using
both confluent (Fig. 5, panel A) and
serum-starved (Fig. 5, panel B) KRC-7 cells. PMA did not
significantly enhance the promoter activity in confluent cells
(panel A, 5th and 6th bars). The promoter activity was significantly reduced upon serum starvation (panel B, 5th bar). However, addition of PMA to these serum-starved cells significantly enhanced (~50-fold) p67 promoter activity (panel B, 5th and 6th bars).
Identification of the PMA-responsive Sequences in the p67 Promoter
Initially, we used different p67 constructs described in Fig. 3
and determined the promoter activity with or without PMA. In several
cases examined (Fig. 6, panels A-C), the
results were qualitatively the same. However, as shown in the absence
of PMA, the luciferase expression in pGL3652/
156 (panel B,
11th bar) and pGL3
180/
60 (panel C, 1st bar) were
reduced to less than 25% maximum expression (panel A, 5th
bar). In the presence of PMA, the luciferase expression remained
essentially the same in pGL3
652/
156 (panel B, 12th bar)
but increased significantly in pGL3
180/
60 (panel C, 2nd
bar). This construct (pGL3
180/
60) contains the AT-rich
region not present in pGL3
652/
156. These results suggest that (i)
the AT-rich region in pGL3
180/
60 construct and (ii) other
possible PMA-responsive sequences in this construct (pGL3
180/
60)
are necessary for PMA induction.
To identify the PMA-responsive sequences, we prepared different
constructs deleting specific nucleotide sequences. Fig. 3 describes the
preparations of three such constructs with deletions at an AP-1 (298
to
292) and/or an AP-10-like (
97 to
88) element. Fig.
7 shows the effects of such deletions on luciferase
expression. The wild type promoter (pGL3
652/+24) was used as a
control. Upon PMA addition, this promoter enhanced luciferase
expression approximately 6-fold (5th and 6th
bars). Deletions at the AP-1 element (
298 to
292)
(7th and 8th bars) and the AP-1-like element
(
97 to
88) (9th and 10th bars) significantly
reduced luciferase expression. When both of these sequences were
deleted, PMA enhancement of luciferase expression was almost totally
eliminated (11th and 12th bars).
In this paper, we describe cloning and characterization of a
1.7-kb DNA fragment containing a segment of the 5-upstream region of
the p67 gene. Some significant observations are as follows.
The 3-end of the DNA fragment is part of the first exon. The
transcription start site was located 89 bases upstream from the
initiator codon and was marked as +1. No classical TATA element (20)
was detected at the expected position. Two AT-rich regions were present
between
40 and
20. Similar AT-rich regions are also present in
c-jun (21). In several cases reported (22-24), these
AT-rich regions function similarly like TATA sequence. The 5
-upstream
region contains LINE sequence (17-19) from
1549 to
652. The LINEs
are defined as a major family of long interspersed nucleotide elements
present in the genome of humans, primates, and rodents. The functions
of these LINEs (if any) in gene expression are not known.
The 652 bp between the end of the LINE sequence and the transcription
start site is the promoter region of the p67 gene. A DNA sequence
containing this promoter region (652 to +24) was inserted upstream of
the promoterless luciferase gene in the pGL3 basic vector. When
transfected into confluent KRC-7 cells, this construct produced an
11-fold increased luciferase expression over that observed with the
pGL3 basic vector. This promoter region contains multiple
cis-acting elements. Some of these elements that differ by
no more than 1 base pair from the consensus sequences of known
regulatory elements are listed as follows: Ets-like element (20) at
645 to
639; SP-1-like elements (20, 25) at
609 to
600 and at
486 to
478; Oct-like element (20) at
500 to
493; CArG-like
element (26) at
410 to
401, heat-shock element (27) at
326 to
310, AP-1 element (20) at
298 to
292; AP-3 element (20) at
142
to
132; and AP-1-like sequence (21) at
97 to
88.
Using different deletion mutants, we mapped the minimum essential
region located between 177 and
60. A purine-rich direct repeat was
located at
167 to
144 in the essential promoter region. This
sequence includes an 11-base pair direct repeat of 5
-AACARAAGAA -3
(R = purine). The relevance of this region, at present, is unclear. In human c-FOS gene, an 8-base pair direct repeat
is apparently important in promoting the basal level of expression of
the gene. This sequence is not required for PMA induction (28). In
addition to the direct repeat, this essential region contain one
AP-1-like sequence (
97 to
88) and an AP-3-like sequence (
142 to
132).
Although the essential promoter region is necessary for basal
transcription, other regulatory elements may also be used to induce
transcription under different physiological conditions. In this work,
we observed that addition of a mitogen, PMA, to the serum-starved KRC-7
cells increased transcription by 50-fold. Using different deletion
mutations, we provide evidence that an AP-1 sequence (298 to
292)
and one AP-1-like sequence (
97 to
88) are necessary for this
induced transcription. The functional assays indicate that the
AP-1-like element at
97 confers the strongest response to PMA and the
classical AP-1 is next. However, to achieve maximal PMA response, both
the elements must be present. These results suggest cooperation among
the PMA-responsive elements.
PMA also induces expression of a number of cellular genes (29-31). It
is not clear whether different genes use common or distinct elements.
Several cellular proteins such as the AP-1 (32-34), NF-B (35-36),
or novel nuclear factors (37) have been implicated in this induction.
In our studies, we observed high levels of PMA induction in KRC-7 cells
transfected with pGL3
652 construct only in serum-starved cells. We
did not observe significant PMA induction in confluent cells. It is not
clear whether this difference is due to the presence of a labile
inhibitor (38) or the presence of CArG element that may act as a
repressor in the presence of serum (39).
Several promoters for translational initiation factors eIF-2 (40),
eIF-4A (41), and eIF-4E (42) have been reported. eIF-4E promoter has
several c-myc regulatory elements (42) and eIF-2
has one
(40). Both these genes are regulated by c-myc (43). eIF-4E
gene also has a potential p53-binding element (42). The p67 promoter
does not contain either c-myc or p53-binding element. On the
other hand both p67 promoter (Fig. 1) and eIF-2
promoter contain
heat shock element. A past report has indicated that eIF-2
is a heat
shock protein (44). Recent work in our laboratory has indicated that
p67 level in the cell is also significantly increased upon heat
shock.2 Another interesting difference is
that whereas the p67 promoter has only 35% G + C, the other initiation
factor promoters contain at least 52% G + C (40-42, 45). Also, the
p67 promoter has two AT-rich regions. This promoter lacks the CAAT
element. The eIF-4A promoter has one classical TATA and CAAT elements.
These sequences are absent in eIF-2
or eIF-4E promoters. Another
interesting difference is that whereas p67 gene contains a single
transcription start site, eIF-2
or eIF-4A contains multiple
transcription start sites.