Permanent Inhibition of Angiotensinogen Synthesis by Antisense RNA Expression
Abstract The renin-angiotensin system plays a pivotal role in blood pressure regulation. Recent molecular biological findings led to the new concept that in addition to the classic endocrine system, local tissue systems may also play an important role in cardiovascular diseases such as hypertension. In particular, the brain renin-angiotensin system was shown to influence the central control of blood pressure and is thought to contribute to the hypertensive phenotype of genetically hypertensive rat models. To identify the physiological role of these local systems, we established an antisense strategy to downregulate the expression of the precursor hormone angiotensinogen (AOGEN) in cell culture, which can also be used to establish transgenic rat lines. Plasmids encoding an RNA sequence complementary to the rat AOGEN mRNA under control of different viral and tissue-specific promoters were constructed and transfected into an AOGEN-expressing cell line. A competitive reverse transcription–polymerase chain reaction method was established for the quantification of AOGEN mRNA. Depending on the level of antisense RNA, the expression of the AOGEN gene was reduced down to 22% of control levels. Furthermore, the secretion of AOGEN protein was totally abolished. These results clearly demonstrate that the antisense constructs used are functional in reducing the AOGEN gene expression in vivo and can be used for the production of transgenic rats.
The regulation of blood pressure, fluid, and sodium homeostasis by the RAS is a classic example of an endocrine mechanism. Renin produced and secreted by the kidney generates the inactive peptide angiotensin I by cleaving AOGEN, which is produced and secreted to the bloodstream by the liver. Further cleavage of angiotensin I by the angiotensin-converting enzyme produces the biologically active hormone angiotensin II. However, more recent studies using molecular biological techniques have extended this traditional paradigm.1 2 In particular, the discovery of various components of the RAS and their mRNAs in other organs and the involvement of these local systems in physiological and pathophysiological alterations by autocrine and paracrine mechanisms have contributed to the change of the original concept. The local RAS in the brain has attracted much interest because it was shown to influence the central control of blood pressure and to be involved in the pathogenesis of hypertension in genetically hypertensive rat models, including the transgenic rats TGR(mREN2)27.3 4 5 6
To elucidate the physiological role of the local RAS, it is necessary to modulate various components of this system in a tissue-specific manner. One of the promising experimental approaches is based on the inhibition of gene expression by use of AS techniques. There are two different ways to inhibit gene expression by an AS approach: AS oligodeoxynucleotides are short DNA fragments (15 to 25 bp) that are complementary to a defined sequence on the target mRNA.7 8 They are chemically synthesized and directly applied. A tissue-specific inhibition of gene expression, however, can only be achieved by organ-specific application. The other approach is based on a construct containing the cDNA of a specific gene in opposite orientation under the control of an active promoter. Depending on the characteristics of this promoter, it can be expressed tissue-specifically or ubiquitously in a transgenic animal. After introduction of such an AS-coding plasmid into cultured cells or the germ line of animals, the AS RNA is produced within the cells themselves. The complementary sequence on the target mRNA forms a stable RNA-RNA duplex, with the AS RNA leading to the suppression of the gene product.9 Several mechanisms underlying this AS inhibition have been suggested. The binding of the AS RNA to the pre-mRNA could lead to an interruption of transcription, processing,7 or transport from the nucleus to the cytoplasm of the mRNA.10 11 Alternatively, the binding of ribosomes to the mRNA and, therefore, translation could be inhibited in the cytoplasm.12
To inhibit the RAS by AS RNA expression, AOGEN is the target of choice because it is the only precursor of the active peptide angiotensin II, and such an approach has already been successful using AS oligodeoxynucleotides in vivo13 14 and AS RNA in cell culture15 and transgenic mice.16
Using transgenic rats, which express AOGEN-specific AS RNA, we wanted to analyze the functional role of the local RAS in the brain, where AOGEN is expressed primarily in astrocytes. Since GFAP has been shown to be colocalized with the AOGEN in this cell type17 and its promoter restricts the expression of a reporter gene in transgenic mice predominantly to the brain,18 we chose the GFAP promoter to drive AS expression in vivo.
In this study, we established an AS RNA–based approach targeting the rat AOGEN mRNA for the production of transgenic animals with a tissue-specifically reduced RAS activity. We constructed plasmids coding for a sequence complementary to the 5′ end of the AOGEN cDNA and tested the inhibitory effect of these constructs on the AOGEN gene expression in cell culture. The rat hepatoma cell line FTO-2B was best suited as a model system, because these cells synthesize and secrete AOGEN constitutively into the cell medium.19 The effects on mRNA levels were measured by a competitive RT-PCR assay with an insertional DNA mutant as internal standard. In this way, we could demonstrate that the chosen AS sequence was able to decrease AOGEN mRNA and protein expression and therefore may be useful for the production of transgenic animals.
DNA Constructs and Oligonucleotides
The second intron of the rabbit β-globin gene was obtained by PCR amplification of the plasmid pSG5 (Stratagene) with a β-globin–specific upstream primer GATCCTGAGAACTTCAGGG (+377 to +395) and downstream primer CCCAGGAGCTGTAGGAAA (+958 to +975). This PCR product (599 bp) was ligated into the Stu I site of the rat AOGEN cDNA in pRag 16-2.20 The AOGEN-specific AS fragment was obtained by PCR using upstream primer 5′-KAO: ATAGCTGTGCTTGTCTGG (+1 to +18) and downstream primer 3′-KAO: TCTCCAGCTGGGCGCAGG (+183 to +200), both containing, in addition, an Xba I restriction site at the 5′ end. This fragment was ligated into the Xba I site of the pRc/RSV and pRc/CMV (Invitrogen). The resulting plasmids contain the strong viral promoters of the RSV long-terminal repeat and of the CMV immediate early genes, respectively, and the 200-bp AS sequence with the internal β-globin intron. These constructs were designated pRag220AβSCMV and pRag220AβSRSV, respectively. To obtain a construct with a tissue-specific promoter, the CMV fragment in pRag220AβSCMV was exchanged by the GFAP promoter. Briefly, the CMV promoter was excised by BglII digestion. The GFAP promoter fragment was obtained as a BglII/BamHI fragment of the plasmid pGfaCAT-2, kindly provided by Dr Michael Brenner, Bethesda, Md18 and ligated into the BglII site of the pRag220AβSCMV. The resulting construct was designated pRag220AβSGFAP.
For the quantification of the AOGEN RNA by RT-PCR, a 320-bp insertion mutant was cloned replacing the β-globin intron by a 120-bp Rsa I fragment of another part of the AOGEN cDNA into the Stu I site of pRag220AβSCMV. All constructs and PCR products were verified by sequencing by the Sanger dideoxy method.21
The RT-PCR assays were performed with the primers 5′-KAO and 3′-KAO, lacking the Xba I restriction sites.
FTO-2B19 cells were grown at 37°C and 5% CO2 in Dulbecco’s modified Eagle’s medium (Gibco BRL)/Nutrient Mix F12 (1:1) containing 10% fetal calf serum, penicillin (100 U/mL), streptomycin (100 μg/mL), and 2 mmol/L glutamine. The cells were seeded at approximately 1×106 cells per 60-mm dish, allowed to grow for 12 hours, and transfected by use of the calcium phosphate coprecipitation method as described previously22 with 5 μg AS plasmids. To obtain stably transfected cell lines, a selection with 400 μg/mL G418 (Gentamycin, Gibco BRL) in the culture medium was started 48 hours after transfection and carried on for 12 days.
All of the following measurements were performed in nonclonal cell lines in the subconfluential stage.
The AOGEN concentration in the medium was measured by a radioimmunoassay for angiotensin I after treatment of the samples with renin.23
Extraction of total cell RNA was performed according to the method described by Wilkinson.24 The RNA concentration was determined by spectrophotometric measurement at 260 nm, and 1 μg of total RNA was checked for integrity by gel electrophoresis in a 1% agarose gel stained with ethidium bromide. AOGEN mRNA was measured by RT-PCR (see below).
For RT, 10 ng RNA was dissolved in 20 μL of a reaction mixture containing 2.5 mmol/L of dATP, dCTP, dTTP, and dGTP; 40 U of RNasin (Promega); 25 pmol of primer 3′-KAO or 100 pmol of random hexamers; 50 mmol/L KCl; 20 mmol/L Tris-HCl (pH 8.4); 2.5 mmol/L MgCl2; 1 μg/μL nuclease-free bovine serum albumin, and 200 U murine leukemia virus reverse transcriptase (MULV-RT; Gibco BRL). The samples were incubated for 60 minutes at 42°C, boiled for 5 minutes at 95°C, and then quickly chilled on ice. For amplification of the resulting cDNA, the sample volume was increased to 100 μL by addition of a solution containing 50 mmol/L KCl, 20 mmol/L Tris-HCl (pH 8.4), and 2.5 mmol/L MgCl2 and 25 pmol of either primer 5′-KAO alone or, in addition, 25 pmol 3′-KAO as well as 3 U Taq polymerase (Gibco BRL). After 5 minutes of denaturation at 95°C, the following thermal profile was used in the Thermal Cycler 480 (Perkin-Elmer Cetus): denaturation for 30 seconds at 95°C, annealing for 30 seconds at 60°C, and extension for 60 seconds at 72°C. After 30 cycles, the samples were loaded on an ethidium bromide–stained 3% agarose gel, and the fluorescence picture taken by an Image VideoScan (Appligene) was directly analyzed with the program NIH Image.25
Statistical significance of the data was determined by the Mann-Whitney U test.
For quantification of the AOGEN mRNA, we used a competitive RT-PCR method with an insertion mutant of the rat AOGEN cDNA as internal standard, which was coamplified in the same vial. The mutant cDNA gave an amplification product of 320 bp, compared with 200 bp for native AOGEN cDNA. Both products were isolated and verified by sequencing (data not shown).
To confirm that the measurement of the AOGEN mRNA concentration takes place in the exponential phase of the PCR, aliquots of the same sample containing the RT products of 10 ng total FTO-2B RNA were removed after the 25th, 28th, 30th, 32nd, and 35th cycles. Fig 1⇓ shows that the linearity of the amplification is ensured for the entire range and that no plateau phase is reached, indicating that the amplification is not limited by one of the components. Therefore, the measurement of AOGEN mRNA was carried out with 30 cycles in the PCR.
Equal amplification efficiency for both templates, the AOGEN cDNA and the mutant DNA, is a prerequisite for the reliable quantification of AOGEN mRNA. Therefore, sequential dilutions of 1 μg total RNA of FTO-2B cells were prepared and used for RT. Before the PCR was performed, 1×108 molecules of the mutant DNA were diluted in the same way and added to the reaction mixture. The results shown in Fig 2⇓ confirm that the mutant and the native AOGEN PCR products are amplified with the same efficiency.
In the next experiment, the linearity of the AOGEN mRNA measurement was tested. A constant amount of mutant DNA (2.2×106 molecules) was added to reverse-transcribed sequential dilutions of total FTO-2B RNA. Fig 3⇓ shows the ratios of native AOGEN to mutant PCR product calculated with the intensities of the corresponding bands. Despite the competition for the primers between the two templates in the PCR, the calculated ratios show a linear relation to the amount of added RNA. According to these results, the measurement of AOGEN mRNA was performed with 10 ng of total cellular RNA and 2.2×106 molecules of the DNA mutant.
FTO-2B cells are rat hepatoma cells that express considerable amounts of AOGEN mRNA and secrete the protein into the medium.19 Therefore, these cells were used to test the effect of AS RNA expression. Assaying the activity of different promoters in this cell culture model with luciferase as reporter gene, the viral CMV and RSV promoters showed the highest expression levels (data not shown). Surprisingly, a low amount of luciferase was also produced by FTO-2B cells transiently transfected with a plasmid containing the astrocyte-specific GFAP promoter. Therefore, pRag220AβSGFAP encoding the AOGEN AS under control of this promoter could be tested in FTO-2B cells for AS efficiency at a relatively low expression level.
FTO-2B cells were stably transfected with AS plasmids (Fig 4A⇓), total RNA was isolated, and the expression of AS RNA was detected with RT-PCR using the primer 5′-KAO for AS-specific RT. Correct splicing of the internal β-globin intron (Fig 4B⇓) was confirmed by Northern blotting and sequencing of the PCR fragments (data not shown). To estimate the relative levels of AS RNA and AOGEN mRNA, total RNA was reverse-transcribed with random hexamer primers, allowing the production of both the AS and the AOGEN cDNA, and was used for PCR with the primers 5′-KAO and 3′-KAO. AOGEN and AS RNA can be differentiated, because the AS RNA contains 37 additional nucleotides representing the residual flanking sequences of the β-globin intron (Fig 4B⇓). Fig 5⇓ shows that AS RNA expression, driven by RSV or CMV promoters, is about 20-fold higher than the AOGEN mRNA expression. Conversely, the AS RNA levels produced by the GFAP promoter are equal to AOGEN mRNA levels in FTO-2B cells.
The AOGEN mRNA content in untransfected FTO-2B cells and cells transfected with pRag220AβSRSV, pRag220AβSCMV, or pRag220AβSGFAP was measured by RT-PCR (Fig 6⇓). The amount of AOGEN mRNA was expressed as the ratio of the intensity of the band of the native and the mutant PCR fragments. The level of AOGEN mRNA in cells transfected with pRag220AβSGFAP was not reduced compared with untransfected control cells (pRag220AβSGFAP, 2.59±0.57; FTO-2B, 2.22±0.33; P>.5). However, AOGEN expression in cells transfected with pRag220AβSCMV and pRag220AβSRSV was reduced down to 30% and 22%, respectively, of the AOGEN mRNA level in untransfected cells (pRag220AβSCMV, 0.67±0.12; P<.01; pRag220AβSRSV, 0.49±0.07; P<.0002). Taken together, the extent of the AOGEN mRNA decrease was linearly dependent on the amount of AS RNA present in the differently transfected cell lines (Fig 5B⇑).
In addition, the AOGEN protein was determined by radioimmunoassay in the culture medium of pRag220AβSRSV-transfected and -untransfected FTO-2B cells. The untransfected cells produced significant amounts of AOGEN (402.6±19.4 pg angiotensin I/108 cells). In contrast, the culture medium of cells transfected with pRag220AβSRSV did not contain more AOGEN than the medium without cells (173.0±2.3 versus 179.3±2.6 pg angiotensin I/108 cells; Fig 7⇓). These results were confirmed by an ELISA with an AOGEN-specific antibody (data not shown).
This study clearly demonstrates that rat AOGEN mRNA and protein can be reduced by the permanent expression of an AS RNA in cell culture. The mechanism of this decrease is speculative at the present. The AS effect could be due to an inhibition of splicing of the pre-mRNA,7 blocking of ribosome binding,12 or modification of the target mRNA in the RNA-RNA duplex by the enzyme unwindase.26 The AS RNA used in the present study could function via all of these mechanisms, since it covers the transcription start site; the exon/intron borders of exon 1, intron 1, and exon 2; and the translation start site. In addition, it overlaps the coding sequence for angiotensin II, the effector hormone of the RAS.
To increase the expression, an intron was introduced into the plasmids coding for the AS RNA.27 In addition, this introduction allows us to distinguish the PCR product of the AS RNA (200 bp) from the DNA of the transfected vectors (800 bp) possibly contaminating the RNA preparations. After splicing, 37 bp of the exon/intron boundaries were still present in the AS sequence, obviously not disturbing the inhibitory effect.
The ratio of AS RNA to target mRNA is one of the important factors in the efficiency of gene inhibition by AS nucleic acids.28 Three different promoters were used in this study; as shown in cell culture and transgenic animals (unpublished observations),29 30 viral promoters like CMV and RSV seem to be the best candidates to achieve a high and permanent expression of a linked gene with the disadvantage of lacking tissue specificity. The GFAP promoter of human origin has also been used to drive the expression of a reporter gene in a transgenic mouse model,18 in which it was shown to be active only in glia cells and not in the liver. The observed expression in the rat hepatoma cell line FTO-2B was therefore unexpected and possibly reveals a partial dedifferentiation of this cell line.
In this study, cells transfected with pRag220AβSRSV as well as pRag220AβSCMV exhibited a high expression of the AS RNA, even exceeding the AOGEN mRNA production in untransfected cells. As a result, both cell lines stably transfected with these vectors exhibited a clear threefold to fourfold reduction of the AOGEN mRNA as measured by RT-PCR. On the other hand, the AOGEN expression in cells transfected with pRag220AβSGFAP was unchanged, which is probably due to the relatively low AS expression. Since AS RNA levels in these cells equaled AOGEN mRNA levels, we conclude that the AS expression has to exceed the target gene expression significantly to achieve an inhibitory effect. By correlating the amounts of AOGEN mRNA and AS RNA in a cell line, we could show an inverse proportionality of the two RNA levels.
Technically, the target of the AS effect is the mRNA. On the other hand, biological effects are mediated by proteins. Therefore, the AS inhibition of the AOGEN mRNA should also lead to a reduction of AOGEN protein concentration in the culture medium. Indeed, FTO-2B cells transfected with pRag220AβSRSV did not secrete detectable amounts of AOGEN protein into the medium. The stronger effect of the AS RNA on the protein compared with the mRNA level may be due to a translational block in addition to a pretranslational inhibition.
This inhibition of gene expression seems not to be a general effect on the protein-synthesizing machinery of the cells caused by the double-stranded RNA, which could activate a kinase leading to the inactivation of an essential translation initiation factor,31 32 since there was no difference in cell proliferation between normal and transfected FTO-2B cells for all plasmids used (data not shown).
In this study, a DNA mutant was used in the PCR to measure AOGEN mRNA levels. This does not allow the absolute quantification of the mRNA, because the RT efficiency cannot be controlled in this assay. To achieve this, a defined amount of an RNA mutant would have to be added to the RT reaction.33 However, for the detection of the AS effect in cell culture, the determination of relative amounts of AOGEN mRNA was sufficient.
The results of this study are in good agreement with the data of Clouston et al,15 who used a mouse AOGEN–specific AS RNA–expressing plasmid driven by the inducible metallothionein promoter in another rat hepatoma cell line (H4IIEC3) and could show a comparable reduction in AOGEN mRNA. However, this group did not measure changes in AOGEN protein secreted from the cells, which is the functionally most important parameter for the evaluation of an AS strategy.
In summary, this study demonstrates that the AOGEN expression was reduced on mRNA as well as on protein level in cell lines stably transfected with vectors expressing an AS RNA under the control of strong viral promoters. Therefore, the chosen AS sequence can also be used to reduce the AOGEN content in tissues of transgenic animals. However, a high expression of AS seems to be important to reduce AOGEN expression. Indeed, transgenic rats expressing high amounts of AS RNA for AOGEN specifically in the brain have been produced with pRag220AβSGFAP, and their altered physiology is currently analyzed.
Selected Abbreviations and Acronyms
|GFAP||=||glial fibrillary acidic protein|
|RSV||=||Rous sarcoma virus|
|RT-PCR||=||reverse transcription–polymerase chain reaction|
The excellent technical assistance of Adelheid Böttger is gratefully acknowledged.
Reprint requests to Manfred Böhm, Max Delbrück Center for Molecular Medicine (MDC), Hypertension Research, 134D, Wiltbergstr 50, D-13125 Berlin, FRG.
Bader M, Wagner J, Lee M, Ganten D. The role of the renin-angiotensin system in cardiovascular disease. Hypertens Res. 1994;17:1-16.
Griendling KK, Murphy TJ, Alexander RW. Molecular biology of the renin-angiotensin system. Circulation. 1993;87:1816-1828.
Ganten D, Hermann K, Bayer C, Unger T, Lang RE. Angiotensin synthesis in the brain and increased turnover in hypertensive rats. Science. 1983;221:869-871.
Ghosh MK, Cohen JS. Oligonucleotides as antisense inhibitors of gene expression. Prog Nucleic Acids Res Mol Biol. 1991;42:79-126.
Wormington M. Stable repression of ribosomal protein L1 synthesis in Xenopus oocytes by microinjection of antisense RNA. Proc Natl Acad Sci U S A. 1986;83:8639-8643.
Ch’ng JL, Mulligan RC, Schimmel P, Holmes EW. Antisense RNA complementary to 3′ coding and noncoding sequences of creatine kinase is a potent inhibitor of translation in vivo. Proc Natl Acad Sci U S A. 1989;86:10006-10010.
Wielbo D, Sernia C, Gyurko R, Phillips MI. Antisense inhibition of hypertension in the spontaneously hypertensive rat. Hypertension. 1995;25:314-319.
Tomita N, Morishita R, Higaki J, Aoki M, Nakamura Y, Mikami H, Fukamizu A, Murakami K, Kaneda Y, Ogihara T. Transient decrease in high blood pressure by in vivo transfer of antisense oligonucleotides against rat angiotensinogen. Hypertension. 1995;26:131-136.
Clouston WM, Lloyd CJ, Richards RI. Inducible anti-sense RNA for angiotensinogen in stably transformed hepatoma cell lines. J Mol Endocrinol. 1990;4:107-117.
Brenner M, Kisseberth WC, Su Y, Besnard F, Messing A. GFAP promoter directs astrocyte-specific expression in transgenic mice. J Neurosci. 1994;14:1030-1037.
Ohkubo H, Kageyama R, Ujihara M, Hirose T, Inayama S, Nakanishi S. Cloning and sequence analysis of cDNA for rat angiotensinogen. Proc Natl Acad Sci U S A. 1983;80:2196-2200.
Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A. 1977;74:5463-5467.
Pesquero JB, Lindsey CJ, Zeh K, Paiva AC, Ganten D, Bader M. Molecular structure and expression of the rat bradykinin B2 receptor gene: evidence for alternative splicing. J Biol Chem. 1994;269:26920-26925.
Hilgenfeldt U, Schott R. Differences in pattern of plasma angiotensinogen in native and nephrectomized rats. Biochim Biophys Acta. 1987;708:335-342.
Wilkinson M. RNA isolation: a mini-prep method. Nucleic Acids Res. 1988;16:10933.
Studer R, Reinecke H, Müller B, Holtz J, Just H, Drexler H. Increased angiotensin I-converting enzyme gene expression in the failing human heart: quantification by competitive RNA polymerase chain reaction. J Clin Invest. 1994;94:301-310.
Wagner RW, Yoo C, Wrabetz L, Kamholz J, Buchhalter J, Hassan NF, Khalili K, Kim SU, Perussia B, McMorris FA, Nishikura K. Double-stranded RNA unwinding and modifying activity is detected ubiquitously in primary tissues and cell lines. Mol Cell Biol. 1990;10:5586-5590.
Brinster RL, Allen JM, Behringer RR, Gelinas RE, Palmiter RD. Introns increase transcriptional efficiency in transgenic mice. Proc Natl Acad Sci U S A. 1988;85:836-840.
Colman A. Antisense strategies in cell and developmental biology. J Cell Sci. 1990;97:399-409.
Overbeek PA, Lai SP, van Quill KR, Westphal H. Tissue-specific expression in transgenic mice of a fused gene containing RSV terminal sequences. Science. 1986;231:1574-1577.
Furth PA, Hennighausen L, Baker C, Beatty B, Woychick R. The variability in activity of the universally expressed human cytomegalovirus immediate early gene 1 enhancer/promoter in transgenic mice. Nucleic Acids Res. 1991;19:6205-6208.