BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 722-729
Vol. 176, No. 2, 1991 April 30, 1991
EXPRESSION OF THE RAT ALDOLASE B GENE : A LIVER-SPECIFIC PROXIMAL PROMOTER AND AN INTRONIC ACTIVATOR
Claudine GREGORI, Frederic GINOT, Jean-Fran(~ois DECAUX, Anne WEBER,Tsouria 8ERBAR, Axel KAHN and Anne-Lise PICHARD ICGM, INSERM U. 129 (Laboratoire de Recherches en G~netique et Pathologic Mol~culaires), CHU Cochin, 24 rue du Faubourg Saint-Jacques, 75014 PARIS, France Received
March
15,
1991
The nature and location of the cis-acting DNA sequences regulating expression of the rat aldolase B gene has been investigated. Two liver-specific DNAse I hypersensitive sites were detected, one located just upstream from the cap site, the second in the middle of the first, 4.8-kbp-long, intron. A fragment of 190 bp 5' to the cap site behaved as a tissue-specific but weak core promoter: it directed a detectable reporter gene expression in the Hep G2 cells and hepatocytes, but not in fibroblasts. The tissue-specific expression was stimulated at least 16 fold when constructs contained the entire first intron. The intronic activating sequences could be ascribed to an inner 2 kbp fragment in which the downstream liver-specific DNAse I hypersensitive site was located. ® 1991Academic R . . . . . I n c .
In vertebrates, three genes encode the different aldolase subunits which are associated into active tetramers (1). Aldolase A is ubiquitous and its gene is also strongly expressed from a tissuespecific promoter in adult skeletal muscle (2,3). Aldolase C is weakly expressed in fetal and actively proliferating malignant tissues and strongly expressed in brain (4,5). Aldolase B is specific to liver, kidney and enterocytes where its specific role seems to catalyze a key reaction of fructose metabolism, i.e. cleavage of fructose 1 phosphate. In vivo, the aldolase B gene is positively regulated by carbohydrates and insulin and partially inhibited by fasting and glucagon treatment (6). Aldolase B cDNAs and genes from different species have been isolated and studied by various groups (7-13). As expected from their different tissue specificity and regulatory properties, the promoters of the different aldolase genes share no homology, while the coding sequence is at least 70% conserved. In this paper, we describe the global organization of the aldolase B regulatory regions, namely a proximal tissue-specific promoter and the presence of an activator located in the middle of the large 4.8kbp-long 1st intron. In contrast to the L-pyruvate kinase gene (COGNET et al, personal communication) and the albumin gene (14), the activator is strongly efficient in transient transfection tests, stimulating activity of the promoter 16 to 60 fold. MATERIAL AND METHODS
Cell culture. Monolayers of human hepatoma (Hep G2) cells were maintained in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% (vol/vol) fetal calf serum, 1 p.M Ltriodothyronin,1 #M dexamethasone, 10 p.M insulin. Mouse 3T6 cells were maintained as monolayers in DMEM supplemented with 10% (vol/vol) fetal calf serum. Hepatocytes were isolated from adult rats 0006-291X/91 $1.50 Copyright © 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
722
Vol. 176, No. 2, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
fasted for 72 h, they were then maintained in chemically defined medium 199, supplemented with 10% (vol/vol) fetal calf serum, 10 mM lactate, 20 mM glucose, l pM L triodothyronin, I#M dexamethasone and 10-8 M epidermal growth factor (15). Cells were plated at a density of 3.5xl 06 per 78 cm2 dish. Transient transfection assays. Transfection was carried out by the calcium phosphate method (15). Calcium phosphate precipitates were removed 18 h after their addition. Cells were then washed twice with 5 ml physiological saline and fed with 10 ml of fresh culture medium. Hep G2 and 3T6 cells were harvested 40 h and hepatocytes in primary culture 68 h after initial exposure to DNA. Each dish received a total of 20 ~g DNA including 3 ~g of the pRSV luciferase standardization plasmid that was used to monitor variations in transfection efficiency. Pellets of 3.5x106 cells were resuspended in 100 #1 of 100 mM potassium buffer pH 7.8, 1 mM dithiothreitol, broken by three cycles of freezing and thawing, then centrifuged at 9800 g for 15 min at + 4°C. The supernatant was assayed for luciferase activity (17) and for CAT activity according to GORMAN et a1.(18). Transient expression vectors. Different restriction fragments of the aldolase B gene were ligated by conventional procedures into the promoterless CAT vector PeCAT. When possible, the fragments were directly inserted into cohesive sites of the vector. Otherwise, blunt ended fragments were either inserted into the Sma 1 site or ligated to linkers before insertion into cognate restriction sites (19) (Fig. 1). The series of mutants spanning all or parts of the first intron was constructed into the promoterless CAT vector VB 8 CAT. This VB 8 CAT vector was obtained by insertion of a 94 bp fragment, spanning the 86 last bp of the first intron and the 8 first bp of the second exon, i.e. just upstream from the CAT gene (Fig. 2). This fragment, synthesized by polymerase chain reaction, includes the full 3' acceptor splicing site of the first intron and stops before the AUG translation initiator of the second exon.
I
ALu(+33) Xba
I
I -2800
//
Sty (+44)
Eco R I I
AlUl
~J
-677
-192
+1
®
Standardized CAT activity (arbitrary units) Hepatocytes i
//
1.8 + 1.9
2808
Hep
G2
3T6
0.45 + 0.05
=0
3.7±0.7
0.9±0.1
=0
5.8±2.5
1.5±0.4
~0.1
1.3±0.8
0.5±0.1
0.2±0.1
+44
I -677
I U A I ,,,~
~
+44
-192
+33
Fiq 1 .Transient expression analysis of CAT constructs driven by 5' flankinq sequences of the aldolase Bgene A: Scheme of the 5' flanking region up to the 1st exon. The restrictionsites used for constructingthe plasmids are indicated.B: Different CAT constructs and the generated CAT activity in transfected cells. The results are expressed in arbitrary units of CAT/luciferaseactivity (see the method chapter). The means + or - 1 SD of at least three independent experiments, each of them in triplicate, are indicated. 723
BIOCHEMICAL A N D BIOPHYSICAL RESEARCH COMMUNICATIONS
V o l . 176, N o . 2, 1991
Exon 1
Exon 2 Hind l~t
Bgl ]] I
I
-2200
+1
A
[
'
El
+0.67kb
Hind ffl EcoRV II
Bgl 1] C,
~
I
+2.6kb
PCR amplified fragment
+4.Skb I I
Kpn l
/ //
\"/"
',
Sac l
~,bp "[ 8
®
CAT
bp
Standardized CAT activity (arbitrary units)
I',
1
H
El
,
°
H i A '
,
El
2200
Hepatocytes
Hep G2
3T6
24 ,'-7.6
39_+17
=0
Bgl ]3
[ ..............
-2200
I
H
]
EcoRV
[ .............
88±76
52±6
=0
1.5±2.2
Vol. 176, No. 2, 1991 April 30, 1991
EXPRESSION OF THE RAT ALDOLASE B GENE : A LIVER-SPECIFIC PROXIMAL PROMOTER AND AN INTRONIC ACTIVATOR
Claudine GREGORI, Frederic GINOT, Jean-Fran(~ois DECAUX, Anne WEBER,Tsouria 8ERBAR, Axel KAHN and Anne-Lise PICHARD ICGM, INSERM U. 129 (Laboratoire de Recherches en G~netique et Pathologic Mol~culaires), CHU Cochin, 24 rue du Faubourg Saint-Jacques, 75014 PARIS, France Received
March
15,
1991
The nature and location of the cis-acting DNA sequences regulating expression of the rat aldolase B gene has been investigated. Two liver-specific DNAse I hypersensitive sites were detected, one located just upstream from the cap site, the second in the middle of the first, 4.8-kbp-long, intron. A fragment of 190 bp 5' to the cap site behaved as a tissue-specific but weak core promoter: it directed a detectable reporter gene expression in the Hep G2 cells and hepatocytes, but not in fibroblasts. The tissue-specific expression was stimulated at least 16 fold when constructs contained the entire first intron. The intronic activating sequences could be ascribed to an inner 2 kbp fragment in which the downstream liver-specific DNAse I hypersensitive site was located. ® 1991Academic R . . . . . I n c .
In vertebrates, three genes encode the different aldolase subunits which are associated into active tetramers (1). Aldolase A is ubiquitous and its gene is also strongly expressed from a tissuespecific promoter in adult skeletal muscle (2,3). Aldolase C is weakly expressed in fetal and actively proliferating malignant tissues and strongly expressed in brain (4,5). Aldolase B is specific to liver, kidney and enterocytes where its specific role seems to catalyze a key reaction of fructose metabolism, i.e. cleavage of fructose 1 phosphate. In vivo, the aldolase B gene is positively regulated by carbohydrates and insulin and partially inhibited by fasting and glucagon treatment (6). Aldolase B cDNAs and genes from different species have been isolated and studied by various groups (7-13). As expected from their different tissue specificity and regulatory properties, the promoters of the different aldolase genes share no homology, while the coding sequence is at least 70% conserved. In this paper, we describe the global organization of the aldolase B regulatory regions, namely a proximal tissue-specific promoter and the presence of an activator located in the middle of the large 4.8kbp-long 1st intron. In contrast to the L-pyruvate kinase gene (COGNET et al, personal communication) and the albumin gene (14), the activator is strongly efficient in transient transfection tests, stimulating activity of the promoter 16 to 60 fold. MATERIAL AND METHODS
Cell culture. Monolayers of human hepatoma (Hep G2) cells were maintained in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% (vol/vol) fetal calf serum, 1 p.M Ltriodothyronin,1 #M dexamethasone, 10 p.M insulin. Mouse 3T6 cells were maintained as monolayers in DMEM supplemented with 10% (vol/vol) fetal calf serum. Hepatocytes were isolated from adult rats 0006-291X/91 $1.50 Copyright © 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
722
Vol. 176, No. 2, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
fasted for 72 h, they were then maintained in chemically defined medium 199, supplemented with 10% (vol/vol) fetal calf serum, 10 mM lactate, 20 mM glucose, l pM L triodothyronin, I#M dexamethasone and 10-8 M epidermal growth factor (15). Cells were plated at a density of 3.5xl 06 per 78 cm2 dish. Transient transfection assays. Transfection was carried out by the calcium phosphate method (15). Calcium phosphate precipitates were removed 18 h after their addition. Cells were then washed twice with 5 ml physiological saline and fed with 10 ml of fresh culture medium. Hep G2 and 3T6 cells were harvested 40 h and hepatocytes in primary culture 68 h after initial exposure to DNA. Each dish received a total of 20 ~g DNA including 3 ~g of the pRSV luciferase standardization plasmid that was used to monitor variations in transfection efficiency. Pellets of 3.5x106 cells were resuspended in 100 #1 of 100 mM potassium buffer pH 7.8, 1 mM dithiothreitol, broken by three cycles of freezing and thawing, then centrifuged at 9800 g for 15 min at + 4°C. The supernatant was assayed for luciferase activity (17) and for CAT activity according to GORMAN et a1.(18). Transient expression vectors. Different restriction fragments of the aldolase B gene were ligated by conventional procedures into the promoterless CAT vector PeCAT. When possible, the fragments were directly inserted into cohesive sites of the vector. Otherwise, blunt ended fragments were either inserted into the Sma 1 site or ligated to linkers before insertion into cognate restriction sites (19) (Fig. 1). The series of mutants spanning all or parts of the first intron was constructed into the promoterless CAT vector VB 8 CAT. This VB 8 CAT vector was obtained by insertion of a 94 bp fragment, spanning the 86 last bp of the first intron and the 8 first bp of the second exon, i.e. just upstream from the CAT gene (Fig. 2). This fragment, synthesized by polymerase chain reaction, includes the full 3' acceptor splicing site of the first intron and stops before the AUG translation initiator of the second exon.
I
ALu(+33) Xba
I
I -2800
//
Sty (+44)
Eco R I I
AlUl
~J
-677
-192
+1
®
Standardized CAT activity (arbitrary units) Hepatocytes i
//
1.8 + 1.9
2808
Hep
G2
3T6
0.45 + 0.05
=0
3.7±0.7
0.9±0.1
=0
5.8±2.5
1.5±0.4
~0.1
1.3±0.8
0.5±0.1
0.2±0.1
+44
I -677
I U A I ,,,~
~
+44
-192
+33
Fiq 1 .Transient expression analysis of CAT constructs driven by 5' flankinq sequences of the aldolase Bgene A: Scheme of the 5' flanking region up to the 1st exon. The restrictionsites used for constructingthe plasmids are indicated.B: Different CAT constructs and the generated CAT activity in transfected cells. The results are expressed in arbitrary units of CAT/luciferaseactivity (see the method chapter). The means + or - 1 SD of at least three independent experiments, each of them in triplicate, are indicated. 723
BIOCHEMICAL A N D BIOPHYSICAL RESEARCH COMMUNICATIONS
V o l . 176, N o . 2, 1991
Exon 1
Exon 2 Hind l~t
Bgl ]] I
I
-2200
+1
A
[
'
El
+0.67kb
Hind ffl EcoRV II
Bgl 1] C,
~
I
+2.6kb
PCR amplified fragment
+4.Skb I I
Kpn l
/ //
\"/"
',
Sac l
~,bp "[ 8
®
CAT
bp
Standardized CAT activity (arbitrary units)
I',
1
H
El
,
°
H i A '
,
El
2200
Hepatocytes
Hep G2
3T6
24 ,'-7.6
39_+17
=0
Bgl ]3
[ ..............
-2200
I
H
]
EcoRV
[ .............
88±76
52±6
=0
1.5±2.2
