Targeted Inactivation of the Ren-2 Gene in Mice
Several recent studies have demonstrated that ablation of genes of the renin-angiotensin system can have wide-ranging and sometimes unexpected effects. Renin is directly involved in blood pressure regulation and is encoded by a single gene in most mammals. Wild mouse strains and some inbred laboratory strains have a duplicated renin gene (Ren-2), the physiological significance of which is unclear. Significant differences exist in the structure and expression of these renin genes, but as yet, no distinct biological function that distinguishes these genes has been defined. We have used gene targeting to discover the effects of inactivating the duplicated (Ren-2) gene in strain 129 mice, and we show that mice lacking the Ren-2 gene are viable and healthy. There appear to be no histopathological differences in renin-expressing tissues between Ren-2–null mice and their controls. Studies of our Ren-2–null mice allow, for the first time, a direct evaluation of the ability of the Ren-1d gene to regulate blood pressure in the absence of expression of the Ren-2 enzyme. We observed no alteration to blood pressure in adult mice homozygous for the mutated Ren-2 gene, even though the concentration of active renin is increased and of prorenin is decreased in plasma of these mice. Ren-1d is therefore capable of regulating normal blood pressure and despite a different tissue expression profile, is functionally equivalent to Ren-1c.
The renin-angiotensin system (RAS) is involved in the regulation of blood pressure (BP) and electrolyte balance as well as in the pathogenesis of several diseases, including hypertension. The aspartyl protease renin catalyses the cleavage of the plasma glycoprotein angiotensinogen, its only known substrate, thereby initiating the first and rate-limiting step in the generation of the potent vasoactive octapeptide hormone angiotensin II (Ang II). Renin is secreted primarily from modified smooth muscle cells of the afferent arteriole of the kidney, which constitute a major element of the juxtaglomerular apparatus.
Molecular cloning has demonstrated that renin is encoded by a single gene in humans, rats, and some strains of laboratory mice. However, wild strains and some inbred laboratory mouse strains have a duplication of approximately 30 kb of DNA on chromosome 1, encompassing the whole of the original renin locus (designated Ren-1d) and including a second functional renin gene designated Ren-2. The single gene present in some inbred strains is designated Ren-1c. The rat and mouse renin genes all span approximately 13 kb and are split into nine exons. The Ren-1c and Ren-1d genes are 99% identical, and the Ren-1d and Ren-2 genes have 97% sequence identity in their respective coding regions; thus, all three murine genes give rise to highly conserved proteins, which are approximately 97% similar at the amino acid level.1 The gene duplication event is thought to have occurred 3 to 10 million years ago, after the speciation of the mouse.2 The most notable differences in the renin-2 protein are the changes at three potential asparagine-linked glycosylation sites present in the renin-1 protein and reduced thermostability compared with renin-1.3
The Ren-1d and Ren-2 genes are expressed at approximately equal levels in the juxtaglomerular cells of the kidney, but these genes are differentially expressed in a number of tissues. The Ren-1c gene is expressed in the developing adrenal gland, but it is developmentally restricted and the mRNA becomes undetectable near birth. The Ren-1d and Ren-2 genes are both expressed at similar levels in the adult adrenal gland of DBA/2 mice, and the expression level varies through the estrus cycle.4 The granular convoluted tubule cells of the submandibular gland show a profound differential expression of renin genes, such that Ren-2 is expressed in 100-fold excess over Ren-1c and is regulated by androgens5 and Ren-1d expression is detectable by ribonuclease protection only.6 On a per-cell basis, the relative amount of mRNA from the Ren-2 gene in the submandibular gland is approximately equal to the level in the kidney. The Ren-1d and Ren-2 genes are also differentially expressed in the interstitial Leydig cells of the testis and a population of subcutaneous fibroblasts in the developing fetus. The renin-2 protein has biochemical properties similar to those of the renin-1 protein,7 but it has not yet been possible to attribute a specific role to the product of the Ren-2 gene. It is not known whether this gene is involved in any aspect of cardiovascular homeostasis or indeed whether it is functionally equivalent to the Ren-1d gene. The duplicated mouse renin genes have long been used as a model system for the analysis of differential gene expression and of the evolutionary development of gene families.
The introduction of the mouse Ren-2 gene into transgenic rats illustrates that in these circumstances, the renin gene can profoundly affect BP regulation, as this genetic modification results in severe hypertension.8 It is thought that the rate of Ang II generation is limited by renin activity in all species except the mouse, in which the reaction is substrate limited. In support of this, expression of the rat angiotensinogen gene in transgenic mice causes hypertension,9 whereas introduction of extra mouse renin genes does not alter BP.4 The powerful studies of Kim and coworkers10 have shown that titration of gene copy number for the mouse angiotensinogen gene not only increases plasma angiotensinogen levels in line with gene copy number but also results in a near-linear increase in BP.
The physiological role of individual genes can be investigated by specifically removing or altering the function of the gene in the absence of any other genetic changes. This is possible by use of cellular mechanisms of homologous recombination in embryonic stem (ES) cells, which can then be used to derive a genetically modified mouse strain.11 12 This approach has been applied recently to some of the genes of the RAS and has demonstrated unequivocally that the genes encoding angiotensinogen,10 13 14 angiotensin-converting enzyme,15 and angiotensin type 1A receptor16 17 are all important in the maintenance of normal BP. Common findings of the studies on angiotensinogen and angiotensin-converting enzyme gene knockouts are histopathological changes in the adult kidney, including medial hyperplasia of the interlobular arteries and afferent arterioles, interstitial fibrosis, and cortical thinning.10 13 14 15 Because of the precise nature of the genetic change introduced in targeting experiments, particularly when performed in an inbred mouse strain such that normal littermates are genetically identical except for the introduced mutation, the data generated provide the proof of causation of a particular phenotype by a defined mutation in a candidate gene.18 As part of a strategy to define pathological effects of ablating all renin gene expression, we attempt here to define separately the physiological functions of the Ren-1d and Ren-2 genes.
Plasmid pPGKneo, ES cell line E14Tg2a, and differentiation inhibiting activity were gifts from Austin Smith.
Amplification of Genomic Regions and Building of Targeting Construct
Two regions of the Ren-2 gene, extending from exon 1 to exon 3 (4.32 kb) and from exon 5 to exon 9 (3.76 kb), were amplified from a partially characterized 129/Ola genomic clone with the use of the following primers: region 1 (4.32 kb): forward: 5′-GGACAGGAGGAGGATGCCTC-3′; reverse: 5′-AAGGTCTGGGGTGGGGTACC-3′; and region 2 (3.76 kb): forward: 5′-CGGGATCCAGTTTGACGGGGTTCTAGG-3′; reverse: 5′-CGGGATCCGGCGCGCCTTGCGGATGAAGGTGGCAC-3′. DNA was amplified for 40 cycles with Pfu DNA polymerase (Stratagene). The amplification conditions were denaturation for 1 minute at 95°C; annealing for 1 minute at 70°C (region 1) or 64°C (region 2) for 6.5 minutes at 74°C; and a single final extension period of 10 minutes at 74°C. Amplifications were performed in 20 mmol/L Tris-HCl (pH 8.75), 10 mmol/L KCl, 10 mmol/L (NH4)2SO4, 100 μg/mL bovine serum albumin, 0.1% Triton X-100, 2 mmol/L MgCl2, 0.2 mmol/L (each) dNTP, 0.5 μmol/L (each) primer, and 50 to 100 ng template DNA. Two factors contribute to the high number of cycles required to produce sufficient material for cloning: (1) the relative inefficiency of Pfu polymerase compared with Taq DNA polymerase, and (2) the fact that the large polymerase chain reaction products shown here are at the limit of amplification for pure proofreading enzymes using genomic clone template DNA.
These amplified products were digested with the restriction endonucleases Xba I and Kpn I (5′ arm of homology) or Kpn I and HindIII (3′ arm of homology) to yield larger fragments that were cloned into a pSP72-based plasmid and smaller fragments that were used as genomic probes flanking the targeting vector sequences during the screening of recombinant ES cell clones. The selectable marker cassette PGKneo, which has the neomycin phosphotransferase gene driven by the phosphoglycerokinase promoter, was inserted between the regions of genomic DNA to yield the plasmid pR2 neoKO (Fig 1⇓). This plasmid was linearized before transfection into ES cells. The entire 3′ homology arm was sequenced, and these data were compared with the 3′ arm polymerase chain reaction product sequenced directly with the use of an automated sequencer (ABI Prism 377, Perkin-Elmer). No differences were found between the targeting construct and the genomic sequence, and so no mutations were introduced during polymerase chain reaction amplification.
The ES cell line E14Tg2a19 was grown on gelatin-coated plastic in Glasgow Minimal Essential Medium supplemented with 1× nonessential amino acids, 0.25% (wt/vol) sodium bicarbonate, 0.4 mmol/L β-mercaptoethanol, 2 mmol/L glutamine, 1 mmol/L pyruvate, 10% fetal calf serum, and mouse DIA/LIF. Electroporation of 1×108 cells with 150 μg linearized DNA was followed 24 hours later by 9 days of selection in G418 (175 μg/mL). Individual clones were picked and expanded and then screened by Southern blot hybridization of PvuII-digested DNA to the 3′ hybridization probe (Fig 1⇑). DNAs from clones that appeared to be targeted at the Ren-2 locus were then digested with Sac I, and the filters were hybridized with the 5′ flanking probe to confirm that homologous recombination had occurred in both arms of the targeting construct (unpublished observations, 1995). These combinations of restriction enzymes and probes were chosen because they distinguish between the two closely linked renin genes and also between the predicted results of homologous recombination at either gene.
Animal Handling and Breeding
Correctly targeted ES cells were thawed, expanded, and injected into blastocysts derived from C57/Bl6 mice and offspring containing a good proportion of ES cell–derived tissue identified by coat color chimerism. These chimeric mice were bred with 129/Ola female mice; thus, transmission of the disrupted Ren-2 gene from ES cells to 129/Ola mice maintains a pure, inbred genetic background. All mice were bred in-house, fed standard chow and tap water ad libitum, and maintained in accordance with the Animals (Scientific Procedures) Act, 1986.
Gene Expression Analysis
RNA was prepared with the guanidinium isothiocyanate/phenol method,20 quantified by UV spectroscopy, and visually assessed by agarose gel electrophoresis. Primer extension with a 38-mer oligonucleotide specific for exon 8 of all mouse renin genes was by the method of Field and Gross.5 RNA expression was quantified by a PhosphorImager (Molecular Dynamics) and was adjusted for sample recovery by comparison of the signal from the primer. Polyacrylamide gels (12%) were exposed for 7 days to Kodak X-Omat film and then for a further 7 days for PhosphorImager analysis.
Mean BP was measured by direct cannulation of the abdominal aorta in adult (12- to 20-week-old) mice. A fine cannula made from drawn polyethylene tubing (Portex) was inserted into the vessel, while the aorta and vena cava were supported under tension with a suture to prevent blood flow and hence blood loss. The aortic flow was stopped for a maximum of 30 seconds, which is sufficient time to insert the catheter into the aorta and fix it in place with tissue glue. After at least 24 hours of recovery, the catheter was connected to a pressure transducer (Viggo-Spectralab), and BP was measured on a chart recorder. The catheter was prefilled with heparin-saline and flushed daily. Measurements were made in conscious mice in restraining tubes for at least 15 minutes for each mouse. Mean BP was calculated as 60% of mean diastolic pressure plus 40% of mean systolic pressure. Each mouse had undergone 10 days of training over the preceding 2 weeks involving at least 10 minutes of restraint per day. All data were collected between 10 am and 3 pm, and the data gathered from mice measured on more than 1 day were averaged.
Plasma Renin and Prorenin Measurements
Blood was collected by cardiac puncture immediately after death into fresh Ωphenanthroline (0.05 mol/L) and EDTA (0.1 mol/L) on ice in a ratio of 10 μL per 100 μL whole blood. Blood was spun immediately for 6 minutes at 4000g; plasma was snap-frozen in liquid nitrogen and stored at −70°C until assayed. Plasma renin and prorenin concentrations were determined as previously described.21 Briefly, inactive renin in 20 μL plasma was activated with 40 μL trypsin (400 U/mL, dissolved in TES buffer: 0.1 mol/L N-tris(hydroxymethyl)methyl-2-aminoethane-sulfonic acid [pH 7.2], 0.01% neomycin, 10 mmol/L EDTA). Samples were incubated on ice for 10 minutes, and the reactions were stopped by the addition of 40 μL soybean trypsin inhibitor (600 U/mL, in TES buffer). Active renin was measured after addition of 80 μL TES buffer without trypsin, as described below.
Pretreated samples were incubated with lyophilized renin substrate isolated from nephrectomized rat plasma (final concentration: 80 mg/mL; 0.11% 2,3-dimercapto-1-propanol, 1.15 mg/mL 8-hydroxychinolin in TES buffer). The reaction was stopped with radioimmunoassay buffer (0.1 mol/L Tris-acetate, pH 7.4) (1) immediately before the incubation and (2) 1 to 3 hours after incubation at 37°C. Generated Ang I was measured by radioimmunoassay.22 23
Mice were killed by exposure to 100% CO2, and tissues were fixed in 10% phosphate-buffered formal saline for 24 hours before dehydration and embedding in paraffin wax. Sections of 6 μm were mounted on slides and stained with hematoxylin and eosin. Multiple sections from kidneys, adrenal, and submandibular glands were examined in a blinded fashion by two experimenters using standard light microscopy.
Screening of G418-resistant ES cells transfected with pR2 neoKO with the 3′ flanking probe (Fig 1⇑) identified 15 of 228 clones that possessed a restriction fragment size diagnostic of homologous recombination in the Ren-2 gene. The 5′ flanking probe was used to show that 13 of these clones were correctly targeted also in the 5′ arm of homology, which is an overall targeting frequency of 5.7%. We used two of these ES cell clones to make chimeric mice that transmitted the disrupted Ren-2 gene to their offspring. None of the targeted cells screened in this way had evidence of the recombination at the highly homologous Ren-1d locus.
Analysis of Renin Gene Expression
The transcripts of the Ren-1d and Ren-2 genes can be distinguished by a dideoxynucleotide primer extension assay5 as well as by ribonuclease protection.6 As shown by primer extension (Fig 2⇓), Ren-2–derived transcripts were not detectable in adult Ren-2–null mice in either kidney or submandibular gland RNA at this level of sensitivity. The relative level of Ren-2–derived RNA, which has been studied in a limited number of mice and in heterozygous mouse kidney, is 55% (range, 54.7% to 55.4%; n=2) compared with wild-type littermates and is 63.3% (range, 44.5% to 79.3%; n=3) in the male heterozygote submandibular gland. The level of Ren-1d mRNA detected in the kidney is slightly increased to 107.6% (range, 101.1% to 114.1%; n=2) and 117.3% (range, 107.2% to 127.3%; n=2) in heterozygous and Ren-2–null mice, respectively.
Analysis of Ren-2 Knockout Mice
Homozygous Ren-2–null mice are healthy and viable. Intercrossing of F1 heterozygotes produced the expected numbers of wild-type, heterozygous, and homozygous mice. The adult Ren-2–null mice had no gross abnormalities postmortem and had normal histomorphology of the kidney, adrenal gland, and submandibular gland compared with wild-type littermates (n=2 for males and females, unpublished observations, 1996).
Plasma Renin and Prorenin Concentrations
The prorenin level in the plasma of heterozygous mice was lower than that of wild-type mice (P<.05, t test), as shown in Fig 3⇓. Homozygous Ren-2–null mice also had decreased plasma prorenin, but this difference did not reach the 5% level of significance (.05<P<.10). Interestingly, plasma renin concentration was higher in Ren-2–null mice than in wild-type mice (P<.05, t test) and was at an intermediate level in heterozygous mice (P=NS). This increase in plasma renin concentration in the absence of two copies of the Ren-2 gene may be due to increased synthesis and release of renin-1 by a feedback mechanism linked either to the lower levels of circulating prorenin or to expression of renin-2 itself.
Does Loss of Renin-2 Affect Resting BP?
To determine whether the lack of a functional Ren-2 gene has any effect on the resting BP of 129/Ola mice, we measured mean BP by direct cannulation of the abdominal aorta (Fig 4⇓). Measured in this way, BP values in wild-type, heterozygous, and Ren-2–null mice did not differ (P>.1, t test). Mixed genetic backgrounds could affect BP, irrespective of the presence or absence of an intact Ren-2 gene. For this reason, we chose to breed purely within the 129/Ola inbred mouse strain. Because we did not backcross to a different strain, there was no segregation of unlinked polymorphic genes in subsequent generations and no false association of a phenotype with the introduced mutation due to selection for 129 alleles linked to the targeted locus.18 All offspring were genetically identical except for the mutated gene.
It is presently unclear what the exact role of the duplicated Ren-2 gene is in the regulation of BP and electrolyte and fluid balance in the mouse strains that carry this gene. The aim of this study was to inactivate the Ren-2 gene by homologous recombination and thus (1) determine whether the remaining renin gene, Ren-1d, is sufficient to maintain normal BP, and (2) reveal unrecognized functions of this gene. The Ren-2 renin gene was inactivated, and mice carrying the null allele were analyzed for changes in resting BP, expression of the Ren-1d gene, levels of active and inactive renin in the circulation, and histopathological changes in renin-expressing organs.
The Ren-1d gene is the orthologue of the Ren-1c gene of “single-gene” mouse strains, and it may be reasonable to expect, on the basis of the high degree of similarity, that the Ren-1d gene product is capable of performing all of the functions of the renin proteins from strains and species that contain only a single renin gene. Previously, studies comparing the Ren-1c and Ren-1d genes have had to take into account the presence of the Ren-2 gene in “two-gene” mice. Inactivation of the Ren-2 gene has generated a new mouse strain that is better suited to the direct comparison of the orthologous Ren-1c and Ren-1d genes. The physiological functions of Ren-1d gene expression can now be assessed in the absence of the Ren-2 gene and so without the confounding effects of the overlapping expression of this similar activity. The effect of genetic background in these comparisons could be taken into account by parallel studies on the wild-type mice of each strain (ie, C57/Bl6 and 129) and by sampling a large number of F2 mice derived from an intercross between Ren-2–null (129) and C57/Bl6 mice after the offspring are typed and sorted according to which Ren-1 alleles (Ren-1c or Ren-1d) are present.
In this study, we aimed to show whether the Ren-1d gene product alone is sufficient for normal BP regulation and also hoped to uncover any critical function of the renin-2 protein not previously recognized. The major sites of expression of the Ren-2 gene are the juxtaglomerular cells of the kidney, the X-zone of the adrenal gland, and the granular convoluted tubule cells of the submandibular gland; however, the specific physiological properties of this protein are not defined. The high level of expression in the secretory epithelial cells of the submandibular gland may be the incidental result of the disruption of a negative regulatory DNA element, located in the 5′ region of the Ren-2 gene, by the insertion of a repetitive sequence termed M2.24 Renin-2 is also expressed and secreted from the juxtaglomerular cells of the kidney and so may play a role in Ang II generation in the classic circulating RAS. We have used a generally applicable, rapid strategy for the construction of vectors for use in gene targeting experiments based on amplification and cloning of long regions of chromosomal homology by the polymerase chain reaction. The result of the homologous recombination reported here is predicted to completely remove enzyme activity. The absence of renin-2 has no effect on the resting BP of young adult mice. However, the normal function of the renin-2 protein may become manifest only after an environmental or physiological stimulus or insult. Studies are ongoing in this respect. The mutation at Ren-2 does affect the level of prorenin in the circulation, and it could be that circulating prorenin is predominantly derived from the Ren-2 gene in normal “two-gene” mice. There is a concomitant increase in the plasma renin concentration in animals with depressed levels of prorenin, suggestive of a compensatory response. The signals that might mediate a feedback response may involve the activation of prorenin in tissues (or in the circulation) or some direct signal passed through the prorenin molecule itself.
Overall, we have shown that the mouse Ren-2 gene is not essential and that mutation of this gene does not affect kidney, submandibular gland, or adrenal gland histomorphology. The levels of the inactive zymogen prorenin are reduced in homozygous Ren-2–null mice, and yet resting BP does not change in these mice. An activity other than that encoded by the Ren-2 gene is sufficient for all the functions of renin that have been described to date, and that activity is probably the product of the Ren-1d gene. More detailed analysis of these mice, under different physiological conditions, is being used to distinguish any functions that may be ascribed to the Ren-2 gene only.
Genetic ablation of the renin substrate angiotensinogen illustrates profound alterations in BP and kidney vascular morphology.10 13 The question of whether the Ren-2 gene is active in the regulation of the classic circulating RAS and is able to participate in the normal regulation of BP will come from the targeted inactivation of the Ren-1d gene in mice that retain the normal Ren-2 gene. The Ren-1d gene has been targeted in mouse ES cells,25 but mice derived from these targeted cells have not been reported to date. We have generated mice that carry a mutation of the Ren-1d gene, and homozygous null mice are healthy and viable (unpublished observations, 1996). Comparison of these strains, which are isogenic except for the renin gene defects, will allow various functions of hemodynamic regulation and physiology of the RAS to be associated with one or both of the forms (nonglycosylated and potentially glycosylated) of renin present in “two-gene” mice. The mice reported here will be used for further study of the role of the renin-2 protein in mouse physiology and to characterize which functions of renin are performed by the products of each gene. This may aid in the elucidation of pathologically important sites or forms of renin expression in human populations.
We are grateful to the Biotechnology and Biological Sciences Research Council, Deutsche Forschungsgemeinschaft (DFG No. Pe366/3-1) and European Commission (Concerted Action Transgeneur) for financial support. We are grateful to Morag Meikle, Jan Ure, Steve Morley, Craig Watt, Hans-Joseph Wrede, and Louise Anderson for help and advice.
Reprint requests to Dr M. Sharp, Centre for Genome Research, Edinburgh University, West Mains Road, Edinburgh EH9 3JQ, UK. E-mail email@example.com.
- Received June 20, 1996.
- Revision received July 11, 1996.
- Revision received August 14, 1996.
Sigmund CD, Gross KW. Structure, expression, and regulation of the murine renin genes. Hypertension. 1991;18:446-457.
Dickinson DP, Gross KW, Piccini N, Wilson CM. Evolution and variation of renin genes in mice. Genetics. 1984;108:651-667.
Wilson CM, Taylor BA. Genetic regulation of thermostability of mouse submaxillary gland renin. J Biol Chem. 1982;257:217-223.
Field LJ, Gross KW. Ren-1 and Ren-2 loci are expressed in mouse kidney. Proc Natl Acad Sci U S A. 1985;82:6196-6200.
Miller CC, Carter AT, Brooks JI, Lovell-Badge RH, Brammar WJ. Differential extra-renal expression of the mouse renin genes. Nucleic Acids Res. 1989;17:3117-3128.
Bing J, Poulsen K, Hackenthal E, Rix E, Taugner R. Renin in the submaxillary gland: a review. J Histochem Cytochem. 1980;28:874-880.
Kim H-S, Krege JH, Kluckman KD, Hagaman JR, Hodgin JB, Best CF, Jennette JC, Coffman TM, Maeda N, Smithies O. Genetic control of blood pressure and the angiotensinogen locus. Proc Natl Acad Sci U S A. 1995;92:2735-2739.
Tanimoto K, Sugiyama F, Goto Y, Ishida J, Takimoto E, Yagami K, Fukamizu A, Murakami K. Angiotensinogen-deficient mice with hypotension. J Biol Chem. 1994;269:31334-31337.
Niimura F, Labosky PA, Kakuchi J, Okubo S, Yoshida H, Oikawa T, Ichiki T, Naftilan AJ, Fogo A, Inagami T, Hogan BLM, Ichikawa I. Gene targeting in mice reveals a requirement for angiotensin in the development and maintenance of kidney morphology and growth factor regulation. J Clin Invest. 1995;96:2947-2954.
Ito M, Oliverio MI, Mannon PJ, Best CF, Maeda N, Smithies O, Coffman TM. Regulation of blood-pressure by the type 1A angiotensin II receptor gene. Proc Natl Acad Sci U S A. 1995;92:3521-3525.
Sugaya T, Nishimatsu S-I, Tanimoto K, Takimoto E, Yamagishi T, Imamura K, Goto S, Imaizumi K, Hisada Y, Otsuka A, Uchida H, Sugiura M, Fukuta K, Fukamizu A, Murakami K. Angiotensin-II type 1a receptor-deficient mice with hypotension and hyperreninemia. J Biol Chem. 1995;270:18719-18722.
Smithies O, Maeda N. Gene targeting approaches to complex genetic diseases: atherosclerosis and essential hypertension. Proc Natl Acad Sci U S A. 1995;92:5266-5272.
Peters J, Münter K, Bader M, Hackenthal E, Mullins JJ, Ganten D. Increased adrenal renin in transgenic hypertensive rats, TGR(mREN2)27, and its regulation by cAMP, angiotensin-II, and calcium. J Clin Invest. 1993;91:742-747.
Hermann K, Ganten D, Unger T, Bayer C, Lang RE. Measurement and characterization of angiotensin peptides in plasma. Clin Chem. 1988;34:1046-1051.
Nakamura N, Burt DW, Paul M, Dzau VJ. Negative control elements and cAMP responsive sequences in the tissue-specific expression of mouse renin genes. Proc Natl Acad Sci U S A.. 1989;86:56-59.
Miller CCJ, McPheat JC, Potts WJ. Targeted integration of the Ren1D locus in mouse embryonic stem cells. Proc Natl Acad Sci U S A. 1992;89:5020-5024.