BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Vol. 180, No. 2, 1991
Pages ] ] 0 3 - ] ] 0 9
October 31, 1991
Expression of the Human Angiotensinogen Gene in Transgenic Mice and Transfected Cells Shigeru Takahashi, Akiyoshi Fukamizu, Takanori Hasegawa*, Minesuke Yokoyama*, Tatsujt N o m u r a , M o t o y a K a t s u k i , , and K a z u o M u r a k a m i Institute of Applied Biochemistry, University of Tsukuba, lbaraki 305, Japan *Central Institute for Experimental Animals, Kawasaki 213, Japan ¶Department of DNA Biology, School of Medicine, Tohkai University, Kanagawa 259-11, Japan Received
September
24,
1991
Summary:
We have generated two lines of transgenic mice with integrated copies of a 14-kilobase pair (kb) haman DNA fragment containing the angiotensinogen gene, which includes 1.3 kb of 5'- and 3'flanking regions. In both transgenic lines, a considerable quantity of the correctly initiated and processed angiotensinogen mRNA was detected in the liver and it was detectable in heart. Unexpectedly, mRNA for the transgene was accumulated in the kidney, where is normally the minor source of angiotensinogen, to levels comparable to that in the liver. In addition, an in vitro transfection analysis suggested that the 1.3kb 5'-flanking sequences are essential for expression of the angiotensinogen gene in hepatic and renal cells and that neither DNA segment within the 14-kb construct contributes significantly to repression of the gene expression in renal cells. ~ 1991 Academic Press, Inc.
Human angiotensinogen, a glycoprotein consisting of 452 amino acids (1), is synthesized predominantly in the liver and secreted into the circulation, where the enzyme renin (EC3.4.23.15) cleaves angiotensinogen to generates angiotensin I. Subsequently, angiotensin I is processed to octapeptide angiotensin II, a most potent vasoconstrictor, by angiotensin-converting enzyme. Since the cleavage of angiotensinogen by renin is thought to be the rate-limiting step in the physiological cascade, the reninangiotensin system plays an important role in the regulation of blood pressure and thus has been implicated in the pathogenesis of hypertension. The human angiotensinogen gene consists of five exons interrupted by four introns (2,3) and maps to chromosome 1 close to the human renin gene within the region of lq4 (4). The structure of the angiotensinogen gene, as well as that of the rat gene, with respect to the number and location of the introns is similar to that found in genes for human cq-antitrypsin and eq-antichymotrypsin, which are classified as serine proteinase inhibitor family (5). We have previously shown that a considerable amount of angiotensinogen mRNA is detected in human liver and human hepatoma HepG2 cells (2) and its mRNA is also detectable but very low in human kidney (6). In a transient expression assay, we demonstrated that the 1.3-kb 5'-flanking region of the gene is a functional promoter (2). However, little is known about an in
1103
0006-291X/91 $1.50 Copyright © 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol. 180, No. 2, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
vivo regulatory mechanism of the gene expression and for the tissue distribution of human angiotensinogen mRNA because of the limited availability of haman materials. In order to further investigate the angiotensinogen expression, we have generated and analyzed two lines of transgenic mice carrying the 14-kb human angiotensinogen gene and examined transcriptional activity of the gene segments by in vitro transfection analysis.
MATERIALS AND METHODS Generation of transgenic mice: The 14-kb NheI DNA fragment containing the human angiotensinogen gene was purified from phAGTM14 by sucrose gradient velocity centrifugation. Microinjection and oviduct transfer procedures were done as described (7). Embryos for microinjection were derived from C57BL/6. Offspring were analyzed for the presence of the injected fragment by Southern blot analysis of tail DNA. Transgenic lines were established by mating to either male and female mice. Preparation of probes: A 298-bp ApaI/EcoRI fragment from pHag3 (1) and a 229-bp NcoI/Cfrl3I fragment from pRagl6 (8) were used for detecting the transgenic human and endogenous mouse mRNAs, respectively, on Northern blot analysis. Preparation of RNA and Northern blot analysis: Total RNAs isolated from HepG2, 293, and T98G cells or various tissues of transgenic and nontransgenic mice were denatured, electrophoresed on a 1.2% agarose gel, and transferred to GeneScreen Plus membrane. Hybridization was carried out using 32p. labelled DNA probe as described previously (2). Primer extension analysis: Primer extension analysis was performed using a 20-residue oligonucleotide (5'-dGTACTCTCATrGTGGATGAC-3') synthesized complementary to residues 169 to 188 of human angiotensinogen mRNA (1). After 5'-end labeling, the oligonucleotide was used for primer extension with 4.5 I.tg of poly(A)+RNA, Plasmid constructions: pUChAGcatl3 was constructed by inserting the 135-bp HpaI-BamHI DNA fragment containing SV40 early polyadenylation signals immediately 5' to the 1.3-kb 5'-flanking sequences of pUChAG(-1222)cat (2). The unique BamHI site 3' of the chloramphenicol acetyltransferase (CAT)-coding domain of pUChAGcat13 was used as an insertion site for the various fragments. The 5.5-kb BamHI DNA fragment (fragment A) was inserted directly in this site, whereas BamHI linkers were used for the insertion of the 6.0-kb BglII-ClaI (fragment B) and 3.8-kb ClaI-BgllI (fragment C) fragments with non-complementary restriction sites. Using the resulting fragments, we constructed pUChAGcatl3A, pUChAGcatl 3B and pUChAGcatl3C. Cell culture, DNA transfection, and CAT assay: Human hepatoma (HepG2) and human glioma (T98G) cells were maintained in minimum essential medium (SIGMA) containing 10% fetal bovine serum and nonessential amino acids. Haman embryonic kidney 293 cells was maintained in minimum essential medium (SIGMA) containing 10% heat-inactivated horse serum. Approximately 5x105 cells were transfected with 5.2 pmol of plasmid DNA/dish by the calcium phosphate precipitation technique. The total amount of DNA was maintained constant at 4 Ixg by the addition of pUC19. Cells were harvested after 36 h, and extracts were prepared and assayed for CAT activity as described oreviously (2).
RESULTS Generation of transgenic mice
Transgenic mice were produced by injecting the 14-kb genomic DNA fragment encompassing the human angiotensinogen gene with the 1.3-kb 5'- and 3'-flanking regions into the male pronuclei of mouse (C57BL/6) fertilized eggs. The DNA from tails of potentially transgenic mice were screened by Southern blot analysis using a human angiotensinogen probe. Of 29 live offspring, one female (hAG2-5) and one male (hAG3-2) were shown to carry the transgene (data not shown). The intensity of bands corresponding to the transgene compared with those corresponding to the cloned gene for copy number suggested that in the founder animals, 100 copies or more of the transgene were present. 1104
Vol. 180, No. 2, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
A
Human Probe hAG2-5 -1- ~
;~
"" 1[
hAG3-2 "
"
N.M. "
"
"
'
"
2 8 S Ib18S D,.-
B
Rat Probe hAG2-5 -I-
~/
~
hAG3-2 •
~.
._.
-" =[ ~
N.M. "
¢i.
..~ ..~
,:
=[
,;
28S l~ 18S
Fig. 1.
Tissue-specific expression of human angiotensinogen gene. Total RNA (4 p.~lane) from various tissues of transgenic and nontransgenic mice or HepG2 cells were denatured and electrophoresed on a 1.2% agarose gel. After transfer to GeneScreen Plus membrane, the angiotensinogen mRNA was detected using human (A) and rat (B) angiotensinogen probes. The tissues analyzed were as follows: liver (Li), kidney (Ki), brain (Br), submandibular gland (SMG), heart (He), lung (Lu), and spleen (Sp). The internal markers were 18S and 28S rRNA stained with ethidium bromide.
Expression of human angiotensinogen gene in transgenic mice To determine the tissue specificity of the human angiotensinogen gene expression in hAG2-5 and hAG3-2 mouse lines, the level of RNA in various tissues was analyzed by Northern hybridization with the human-specific angiotensinogen probe. As judged by comparison with human HepG2 RNA, the expected 2kb RNA for the human transgene was highest in liver of both mouse lines (Fig. 1A) and generated from the normal transcription start site (Fig. 2), suggesting that initiation of the transgene transcription, processing o f the nuclear precursor mRNA and polyadenylation are performed correctly in the mice. Surprisingly, a significant quantity of human angiotensinogen mRNA were also found in transgenic kidney, although levels of mouse angiotensinogen were very low in kidney (Fig. 1B) as well as in human (6). In addition, mRNA for the transgene was detectable in heart. Lower levels of the human transcripts were also seen in every tissue of the two transgenic lines upon prolonged exposure of these autoradiograms (data not shown). To compare expression of the endogenous mouse angiotensinogen gene between transgenic and nontransgenic mice, we probed the RNA from these mice with a murine specific-angiotensinogen DNA. As shown in Fig. 1B, the pattern of expression was identical in these three mice. This suggested that the transfer of new genes into mice did not disturb the normal regulation of the mouse angiotensinogen gene. 1105
Vol. 180, No. 2, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
A -I~r
Vol. 180, No. 2, 1991
Pages ] ] 0 3 - ] ] 0 9
October 31, 1991
Expression of the Human Angiotensinogen Gene in Transgenic Mice and Transfected Cells Shigeru Takahashi, Akiyoshi Fukamizu, Takanori Hasegawa*, Minesuke Yokoyama*, Tatsujt N o m u r a , M o t o y a K a t s u k i , , and K a z u o M u r a k a m i Institute of Applied Biochemistry, University of Tsukuba, lbaraki 305, Japan *Central Institute for Experimental Animals, Kawasaki 213, Japan ¶Department of DNA Biology, School of Medicine, Tohkai University, Kanagawa 259-11, Japan Received
September
24,
1991
Summary:
We have generated two lines of transgenic mice with integrated copies of a 14-kilobase pair (kb) haman DNA fragment containing the angiotensinogen gene, which includes 1.3 kb of 5'- and 3'flanking regions. In both transgenic lines, a considerable quantity of the correctly initiated and processed angiotensinogen mRNA was detected in the liver and it was detectable in heart. Unexpectedly, mRNA for the transgene was accumulated in the kidney, where is normally the minor source of angiotensinogen, to levels comparable to that in the liver. In addition, an in vitro transfection analysis suggested that the 1.3kb 5'-flanking sequences are essential for expression of the angiotensinogen gene in hepatic and renal cells and that neither DNA segment within the 14-kb construct contributes significantly to repression of the gene expression in renal cells. ~ 1991 Academic Press, Inc.
Human angiotensinogen, a glycoprotein consisting of 452 amino acids (1), is synthesized predominantly in the liver and secreted into the circulation, where the enzyme renin (EC3.4.23.15) cleaves angiotensinogen to generates angiotensin I. Subsequently, angiotensin I is processed to octapeptide angiotensin II, a most potent vasoconstrictor, by angiotensin-converting enzyme. Since the cleavage of angiotensinogen by renin is thought to be the rate-limiting step in the physiological cascade, the reninangiotensin system plays an important role in the regulation of blood pressure and thus has been implicated in the pathogenesis of hypertension. The human angiotensinogen gene consists of five exons interrupted by four introns (2,3) and maps to chromosome 1 close to the human renin gene within the region of lq4 (4). The structure of the angiotensinogen gene, as well as that of the rat gene, with respect to the number and location of the introns is similar to that found in genes for human cq-antitrypsin and eq-antichymotrypsin, which are classified as serine proteinase inhibitor family (5). We have previously shown that a considerable amount of angiotensinogen mRNA is detected in human liver and human hepatoma HepG2 cells (2) and its mRNA is also detectable but very low in human kidney (6). In a transient expression assay, we demonstrated that the 1.3-kb 5'-flanking region of the gene is a functional promoter (2). However, little is known about an in
1103
0006-291X/91 $1.50 Copyright © 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol. 180, No. 2, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
vivo regulatory mechanism of the gene expression and for the tissue distribution of human angiotensinogen mRNA because of the limited availability of haman materials. In order to further investigate the angiotensinogen expression, we have generated and analyzed two lines of transgenic mice carrying the 14-kb human angiotensinogen gene and examined transcriptional activity of the gene segments by in vitro transfection analysis.
MATERIALS AND METHODS Generation of transgenic mice: The 14-kb NheI DNA fragment containing the human angiotensinogen gene was purified from phAGTM14 by sucrose gradient velocity centrifugation. Microinjection and oviduct transfer procedures were done as described (7). Embryos for microinjection were derived from C57BL/6. Offspring were analyzed for the presence of the injected fragment by Southern blot analysis of tail DNA. Transgenic lines were established by mating to either male and female mice. Preparation of probes: A 298-bp ApaI/EcoRI fragment from pHag3 (1) and a 229-bp NcoI/Cfrl3I fragment from pRagl6 (8) were used for detecting the transgenic human and endogenous mouse mRNAs, respectively, on Northern blot analysis. Preparation of RNA and Northern blot analysis: Total RNAs isolated from HepG2, 293, and T98G cells or various tissues of transgenic and nontransgenic mice were denatured, electrophoresed on a 1.2% agarose gel, and transferred to GeneScreen Plus membrane. Hybridization was carried out using 32p. labelled DNA probe as described previously (2). Primer extension analysis: Primer extension analysis was performed using a 20-residue oligonucleotide (5'-dGTACTCTCATrGTGGATGAC-3') synthesized complementary to residues 169 to 188 of human angiotensinogen mRNA (1). After 5'-end labeling, the oligonucleotide was used for primer extension with 4.5 I.tg of poly(A)+RNA, Plasmid constructions: pUChAGcatl3 was constructed by inserting the 135-bp HpaI-BamHI DNA fragment containing SV40 early polyadenylation signals immediately 5' to the 1.3-kb 5'-flanking sequences of pUChAG(-1222)cat (2). The unique BamHI site 3' of the chloramphenicol acetyltransferase (CAT)-coding domain of pUChAGcat13 was used as an insertion site for the various fragments. The 5.5-kb BamHI DNA fragment (fragment A) was inserted directly in this site, whereas BamHI linkers were used for the insertion of the 6.0-kb BglII-ClaI (fragment B) and 3.8-kb ClaI-BgllI (fragment C) fragments with non-complementary restriction sites. Using the resulting fragments, we constructed pUChAGcatl3A, pUChAGcatl 3B and pUChAGcatl3C. Cell culture, DNA transfection, and CAT assay: Human hepatoma (HepG2) and human glioma (T98G) cells were maintained in minimum essential medium (SIGMA) containing 10% fetal bovine serum and nonessential amino acids. Haman embryonic kidney 293 cells was maintained in minimum essential medium (SIGMA) containing 10% heat-inactivated horse serum. Approximately 5x105 cells were transfected with 5.2 pmol of plasmid DNA/dish by the calcium phosphate precipitation technique. The total amount of DNA was maintained constant at 4 Ixg by the addition of pUC19. Cells were harvested after 36 h, and extracts were prepared and assayed for CAT activity as described oreviously (2).
RESULTS Generation of transgenic mice
Transgenic mice were produced by injecting the 14-kb genomic DNA fragment encompassing the human angiotensinogen gene with the 1.3-kb 5'- and 3'-flanking regions into the male pronuclei of mouse (C57BL/6) fertilized eggs. The DNA from tails of potentially transgenic mice were screened by Southern blot analysis using a human angiotensinogen probe. Of 29 live offspring, one female (hAG2-5) and one male (hAG3-2) were shown to carry the transgene (data not shown). The intensity of bands corresponding to the transgene compared with those corresponding to the cloned gene for copy number suggested that in the founder animals, 100 copies or more of the transgene were present. 1104
Vol. 180, No. 2, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
A
Human Probe hAG2-5 -1- ~
;~
"" 1[
hAG3-2 "
"
N.M. "
"
"
'
"
2 8 S Ib18S D,.-
B
Rat Probe hAG2-5 -I-
~/
~
hAG3-2 •
~.
._.
-" =[ ~
N.M. "
¢i.
..~ ..~
,:
=[
,;
28S l~ 18S
Fig. 1.
Tissue-specific expression of human angiotensinogen gene. Total RNA (4 p.~lane) from various tissues of transgenic and nontransgenic mice or HepG2 cells were denatured and electrophoresed on a 1.2% agarose gel. After transfer to GeneScreen Plus membrane, the angiotensinogen mRNA was detected using human (A) and rat (B) angiotensinogen probes. The tissues analyzed were as follows: liver (Li), kidney (Ki), brain (Br), submandibular gland (SMG), heart (He), lung (Lu), and spleen (Sp). The internal markers were 18S and 28S rRNA stained with ethidium bromide.
Expression of human angiotensinogen gene in transgenic mice To determine the tissue specificity of the human angiotensinogen gene expression in hAG2-5 and hAG3-2 mouse lines, the level of RNA in various tissues was analyzed by Northern hybridization with the human-specific angiotensinogen probe. As judged by comparison with human HepG2 RNA, the expected 2kb RNA for the human transgene was highest in liver of both mouse lines (Fig. 1A) and generated from the normal transcription start site (Fig. 2), suggesting that initiation of the transgene transcription, processing o f the nuclear precursor mRNA and polyadenylation are performed correctly in the mice. Surprisingly, a significant quantity of human angiotensinogen mRNA were also found in transgenic kidney, although levels of mouse angiotensinogen were very low in kidney (Fig. 1B) as well as in human (6). In addition, mRNA for the transgene was detectable in heart. Lower levels of the human transcripts were also seen in every tissue of the two transgenic lines upon prolonged exposure of these autoradiograms (data not shown). To compare expression of the endogenous mouse angiotensinogen gene between transgenic and nontransgenic mice, we probed the RNA from these mice with a murine specific-angiotensinogen DNA. As shown in Fig. 1B, the pattern of expression was identical in these three mice. This suggested that the transfer of new genes into mice did not disturb the normal regulation of the mouse angiotensinogen gene. 1105
Vol. 180, No. 2, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
A -I~r
