Inhibition of Hypertension by Peripheral Administration of Antisense Oligodeoxynucleotides
We administered liposome-encapsulated antisense oligodeoxynucleotide targeted to angiotensinogen mRNA peripherally to spontaneously hypertensive rats to test whether peripheral angiotensinogen reduction would lower their hypertensive blood pressures and to determine the role of peripheral angiotensinogen in the modulation of hypertension. Using in vitro translation techniques, we tested the sequence specificity of the antisense sequence. The selected antisense sequence decreased angiotensinogen production in vitro, enabling us to distinguish between specific and nonspecific effects. To increase the efficiency of peripheral and hepatic antisense delivery, oligonucleotides were liposome encapsulated for intra-arterial administration. Confocal microscopy was used for determination of the hepatic distribution of fluorescently labeled antisense. Encapsulated antisense molecules were seen to be distributed within liver tissue 1 hour after injection; however, little or no uptake was observed with the unencapsulated oligonucleotides. We also determined the physiological effects of antisense oligodeoxynucleotide targeted to liver angiotensinogen mRNA. Administration of liposome-encapsulated antisense significantly decreased hypertensive blood pressures to normotensive levels compared with scrambled control oligonucleotides, unencapsulated antisense, and empty liposomes (P=.013). These data were supported by biochemical changes elicited by the antisense treatment. Rats receiving liposome-encapsulated antisense had significantly lowered peripheral angiotensinogen and angiotensin II levels compared with control groups (P<.05). No significant heart rate changes were observed in the antisense or control groups. These results suggest that peripheral angiotensinogen plays a role in the maintenance of hypertensive blood pressure in this model of hypertension and that peripheral administration of antisense molecules is possible with organ-targeted delivery mechanisms.
Genetic variants of angiotensinogen are implicated in the development of essential and related forms of hypertension.1 Angiotensinogen, produced largely in the liver, is cleaved to the decapeptide Ang I, which is further cleaved to the octapeptide Ang II, a potent vasoconstrictor and regulator of BP and volume homeostasis.2 3 Stimulation of cardiac renin-angiotensin system components has been associated with cardiac hypertrophy,4 and localized vascular production of Ang II has been shown to stimulate vascular regrowth and thickening after balloon catheter angioplasty.5 Overactivity of the renin-angiotensin system has been implicated in the development and maintenance of hypertension in the SHR, the animal model of essential hypertension.6 7 The renin-angiotensin system can be inhibited by several mechanisms, including angiotensin converting enzyme inhibitors, renin inhibitors, and Ang II antagonists, but as yet there has been no specific inhibitor of angiotensinogen.
Antisense molecules have been used to successfully inhibit protein synthesis in a number of biological systems.8 9 10 ASODNs, small fragments of DNA usually 12 to 18 bases in length, bind to a complementary region of target mRNA and attenuate candidate gene expression.11 This paradigm of gene regulation has many potential therapeutic applications and is currently being developed as anticancer, antianxiety, and antiviral agents.12 In addition to the therapeutic potential of such agents, we and others have used this concept as a physiological tool to provide information on cardiovascular function and hypertension. We have previously shown that central administration of ASODNs targeted to angiotensinogen mRNA significantly decreases hypertensive BP in the SHR for prolonged periods of time, with corresponding decreases in hypothalamic Ang II levels and hypothalamic angiotensinogen.13 14 Sakai et al15 have shown that centrally administered Ang II type 1 receptor ASODN inhibits dipsogenic responses to Ang II, and Morishita and coworkers16 successfully used ASODN to inhibit neointima formation after balloon catheter angioplasty. Recently, Tomita et al17 were able to decrease BP in the SHR using a combination of oligonucleotides to target peripheral angiotensinogen. Although ASODNs show potential applications as physiological tools and therapeutic agents, problems arise in the delivery of molecules to target sites.18 Unmodified ASODNs are rapidly degraded by endonucleases and exonucleases in plasma and tissue and can be only moderately protected by backbone modification.19 Advances in targeted drug delivery now enable liposomal encapsulation of drug molecules to provide protection, sustained release, and efficient cellular uptake.20 21 We hypothesize that peripheral angiotensinogen plays a significant role in the modulation of hypertension and that hypertension in the SHR model can be attenuated through the use of liposome-encapsulated ASODNs targeted to peripheral angiotensinogen mRNA.
Adult male SHR (Harlan, Indianapolis, Ind) weighing 250 to 275 g (n=6 per group) were kept in a room with a 12-hour light/dark cycle and fed laboratory rat chow and tap water ad libitum. Rats were anesthetized with ketamine/xylazine (45:9 mg/100 g IP), and a heparinized (100 U/mL) catheter (PE-50 tubing, 0.58×0.965 mm) was placed in the left carotid artery, 2.5 mm toward the heart. Catheters were stoppered and exteriorized between the scapulae so that rats could not chew them. Rats were allowed to recover for 24 hours before further experimentation. All experimental procedures were approved by the Animal Care Committee of the University of Florida, Gainesville.
ASODNs and ScrODNs, synthesized as phosphorothioated 18-mers according to the base sequence of Ohkubo et al,22 were targeted to bases −5 to +13 of angiotensinogen mRNA, encompassing the AUG translation start codon. FITC-conjugated oligodeoxynucleotides were composed of phosphorothioated sequences with 5′ and 3′ end FITC conjugation.
Liposomes (80% phosphatidylcholine, 20% cholesterol; Avanti Polar Lipids) were prepared by rotary evaporation for drying and rehydration of the lipid film and multiple freeze-thaw cycles to enhance oligodeoxynucleotide entrapment. Liposomes were extruded through a 0.1-μm filter for size reduction, and size was determined by dynamic light scattering.
In Vitro Transcription/Translation
Transcription/translation reactions were carried out with a TNT coupled reticulocyte lysate transcription/translation kit (Promega Biotechnology). One microgram of plasmid angiotensinogen cDNA with SP6 promoter sequence was added to TNT lysate and incubated in a 50-μL reaction volume containing all components necessary for transcription and translation to occur for 2 hours at 30°C. Combined transcription and translation reactions were carried out in the presence of ASODNs or control ScrODNs at doses ranging from 0.3 to 30 μmol/L oligodeoxynucleotide. Translated protein was analyzed by electrophoresis on a 15% polyacrylamide-sodium dodecyl sulfate gel followed by autoradiography.
Hepatic Distribution of FITC-Conjugated ASODN
Male Sprague-Dawley rats (Harlan, n=2 per group) were anesthetized as above, and FITC-conjugated ASODN (50 μg in 300 μL phosphate-buffered saline) or liposome-encapsulated FITC-conjugated ASODN (50 μg in 25 mg lipid) was administered via the carotid catheter. One hour later, rats were transcardially perfused with saline and 40% formaldehyde. Livers were removed, and 50-μm cryostat sections of the median lobe were observed by laser scanning confocal microscopy for determination of oligodeoxynucleotide distribution.
Effects of Peripherally Administered ASODN on MAP
Male SHR (n=6 per group) were catheterized as above and allowed 24 hours to recover. Baseline MAP was measured for at least 1 hour, and then each rat received 50 μg liposome-encapsulated ASODN, liposome-encapsulated ScrODN, or empty liposomes (25 mg lipid) in 300 μL phosphate-buffered saline via the carotid catheter. MAP was monitored 24 hours after injection with a direct pressure transducer and recorded on a Digi-Med BP Analyzer (Micro-Med). Rats were then decapitated, and trunk blood was collected for determination of peripheral Ang II and angiotensinogen by radioimmunoassay.
Rats were anesthetized; trunk blood was collected in 200 μL of 0.5 mol/L EDTA and 10 μL o-phenanthroline and centrifuged; and 500 μL of plasma was collected. Samples were then lyophilized and assayed for angiotensinogen by the method of Sernia et al.23 Angiotensinogen sample content is measured from a standard curve of angiotensinogen diluted in medium corresponding to the sample. The assay sensitivity is 0.3 ng per tube, with interassay and intra-assay variabilities of 14% and 9%, respectively.
Ang II Assay
Plasma was frozen at −70°C until extraction with methanol on reversed-phase phenylsilylsilica extraction cartridges (Alpco; approximate recovery, 90%). Samples were analyzed by double-antibody Ang II radioimmunoassay (RK-A22, Alpco). The assay is sensitive to 0.7 pg/mL (0.7 pmol/L). Ang II levels were determined by a gamma counter (Beckman DP550).
Statistical analysis was performed by ANOVA for treatment effect, and Duncan's multiple range test was used for individual comparisons. A value of P<.05 was considered statistically significant.
Fig 1⇓ shows inhibition of angiotensinogen in vitro and demonstrates the specificity of the target antisense molecule to the chosen target. Data are presented as percent angiotensinogen expression compared with baseline angiotensinogen expression over a range of 0.3 to 30 μmol/L oligodeoxynucleotide. At high ASODN doses (3 to 30 μmol/L), angiotensinogen expression decreased to 5% to 10% of baseline. However, nonspecific attenuation of angiotensinogen expression was observed with control ScrODN. At lower ASODN doses (0.3 to 1 μmol/L), reactions carried out in the presence of ASODN showed profound attenuation of angiotensinogen production (45%), with no nonspecific attenuation by ScrODN.
Fig 2⇓ shows two confocal micrographs of rat liver tissue 1 hour after intra-arterial injection of 50 μg unencapsulated FITC-conjugated ASODN (A) or liposome-encapsulated FITC-conjugated ASODN (B). Panel A shows little or no distribution of fluorescent signal within the liver tissue. Panel B shows an intense fluorescent signal throughout the tissue, with highest intensities observed adjacent to tissue sinusoids.
For determination of the effects of peripherally administered ASODN on physiological parameters, baseline MAP was established in groups of rats, and then 50 μg liposome-encapsulated (25 mg lipid) ASODN (AS/L), liposome-encapsulated ScrODN (Scr/L), empty liposomes (25 mg, Lipo), or unencapsulated ASODN was administered intra-arterially. MAP was measured 24 hours later for determination of BP changes. Fig 3⇓ shows MAP changes 24 hours after treatment. MAP was significantly decreased in AS/L-treated rats (−24.66±2.43 mm Hg). No significant BP changes were observed in the Scr/L (1.34±3.98 mm Hg), Lipo (−5.34±3.71 mm Hg), or ASODN (−6.02±8.68 mm Hg) treatment groups (results are expressed as mean±SE, P=.013, n=6 per group).
For determination of Ang II changes, rats were killed 24 hours after treatment, and plasma Ang II levels were measured by radioimmunoassay. Fig 4⇓ shows the effect of AS/L treatment on plasma Ang II levels. Plasma Ang II was significantly lower in AS/L-treated rats (30.3±11.4 pg/mL, n=5) compared with the control groups (Scr/L, 103.1±33.2 pg/mL, n=3; Lipo, 233.5±71.1 pg/mL, n=3; and unencapsulated ASODN, 201.4±88.5 pg/mL, n=3; P<.05).
Fig 5⇓ shows the effect of AS/L on plasma angiotensinogen levels 24 hours after administration. Plasma angiotensinogen was significantly lower in AS/L-treated rats (58.5±3.71 pg/mL, n=5) compared with the control groups (Scr/L, 79.0±8.72 pg/mL, n=3; Lipo, 85±5.72 pg/mL, n=3; and unencapsulated ASODN, 77.8±3.25 pg/mL, n=3; P<.05).
We have shown that ASODNs targeted to angiotensinogen mRNA result in a substantial decrease in protein production in vitro and a significant decrease in Ang II and angiotensinogen in vivo. The observed decrease in plasma angiotensinogen indicates successful translational inhibition of the candidate gene product by phosphorothioated ASODN constructed to complement angiotensinogen mRNA. Although angiotensinogen mRNA levels were undetermined in this study, previous studies by this group have shown no decreases in target mRNA levels with antisense treatment (unpublished observations, 1993); therefore, we propose that the mechanism of action is via the attenuation of mRNA translation rather than triple helix formation at the gene level. High doses of ASODN and control ScrODN resulted in nonspecific protein attenuation in vitro. This phenomenon, recently described in the literature,24 may be attributed to nonspecific binding of oligodeoxynucleotides to cellular proteins and is commonly observed with phosphorothioated oligodeoxynucleotides. Lower doses of oligodeoxynucleotides resulted in profound inhibition of angiotensinogen by ASODN only, suggesting that ASODN specifically attenuates the translation of angiotensinogen in vitro. Tomita et al17 recently showed that liposomal encapsulation of three antisense molecules targeted to contiguous sequences of liver angiotensinogen mRNA transiently decreased hypertension after administration via the hepatic portal vein. To facilitate uptake, the liposomes were associated with viral surface antigens. Unlike Tomita et al, our strategy was to inject a single ASODN to liver angiotensinogen mRNA intra-arterially via simple liposome encapsulation. This proved to be very effective, requiring minimal surgery and no complicated delivery system. Therefore, we used a similar, although more simple, approach for ASODN delivery to the target organ, using a smaller ASODN sequence, as demonstrated by hepatic uptake of FITC-conjugated ASODN after peripheral administration. Our data show that ASODNs may be able to reach the parenchymal cells of the liver involved in angiotensinogen production. This finding is supported by the fact that in rats treated with liposome-encapsulated ASODNs, hypertensive BP values were profoundly reduced to normotensive levels. Additionally, significant decreases in plasma angiotensinogen and plasma Ang II levels in rats treated with liposome-encapsulated ASODNs support the concept that ASODN effects are mediated through the proposed mechanism of action, that is, via translational inhibition of angiotensinogen mRNA, with a subsequent decrease in plasma Ang II levels. Although plasma Ang II levels appeared to be decreased in the group treated with liposome-encapsulated ScrODN, the decrease was not significantly different from the other control treatments. The ensuing decrease in BP may be attributed to the significant decrease in plasma Ang II resulting from liposome-encapsulated ASODN treatment. Previously, we described similar BP responses with 50 μg centrally administered ASODN targeted to angiotensinogen.14 Originally, to determine the peripheral contribution to these responses, we administered unencapsulated ASODN intra-arterially at the same dose and observed no significant BP changes. In the present studies, because of liposome encapsulation, responses have now been achieved after peripheral administration. It is possible that liposomal delivery of oligodeoxynucleotides increases cellular uptake efficiency and protects the oligodeoxynucleotide from nuclease degradation to an extent that intracellular oligodeoxynucleotide levels are sufficient to elicit specific physiological responses as well as passively targeting ASODNs to the liver. In addition, our data demonstrate the possible existence of two regulatory components of BP maintenance in the SHR: a brain angiotensinogen component and a liver angiotensinogen component.
In conclusion, these data support the concept that peripheral angiotensinogen plays an important role in the maintenance of hypertension in the SHR; that peripherally administered, liposome-encapsulated antisense targeted to liver tissue decreases BP in this model of hypertension by specifically altering renin-angiotensin system components; and that ASODN shows potential as an antihypertensive agent.
Selected Abbreviations and Acronyms
|Ang I, II||=||angiotensin I, II|
|MAP||=||mean arterial pressure|
|SHR||=||spontaneously hypertensive rat(s)|
This work was funded by Research Development Awards from the University of Florida (D. Wielbo) and the American Heart Association, Florida Affiliate (Initial Investigators Award No. 9503004, D. Wielbo). We wish to thank Leping Shen, Department of Physiology, University of Florida, for his technical assistance, and the laboratory of Dr Conrad Sernia, University of South Queensland (Australia), for help in angiotensinogen radioimmunoassay.
- Received October 23, 1995.
- Revision received November 21, 1995.
- Revision received February 27, 1996.
Dzau VJ. Circulating versus local renin-angiotensin system in cardiovascular homeostasis. Circulation. 1988;77(suppl 19):1-4.
Dzau VJ. Local expression and pathophysiological role of renin-angiotensin in the blood vessels and heart. Basic Res Cardiol. 1993;88(suppl 1):1-14.
Viswanathan M, Stromberg C, Seltzer A, Saavedra JM. Balloon angioplasty enhances the expression of angiotensin II AT1 receptors in neointima of rat aorta. J Clin Invest. 1992;90:1707-1712.
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.
Yongue BG, Angulo JA, McEwen BS, Meyers MM. Brain and liver angiotensinogen messenger RNA in genetic hypertensive and normotensive rats. Hypertension. 1991;17:485-491.
Wahlstedt D, Pich EM, Koob GF, Yee F, Heilig M. Modulation of anxiety and neuropeptide Y-Y1 receptors by antisense oligodeoxynucleotides. Science. 1993;259:528-531.
Crooke ST. Progress toward oligonucleotide therapeutics: pharmacodynamic properties. FASEB J. 1993;7:533-539.
Wielbo D, Sernia C, Phillips MI. The distribution and antihypertensive effects of phosphorothioated antisense oligodeoxynucleotides in SHR. Hypertension. 1995;25:314-319.
Morishita R, Gibbons GH, Ellison KE, Nakajima M, Zhang L, Kaneda Y, Ogihara T, Dzau VJ. Single intraluminal delivery of antisense cdc2 kinase and proliferating-cell nuclear antigen oligonucleotides results in chronic inhibition of neointimal hyperplasia. Proc Natl Acad Sci U S A. 1993;90:8474-8478.
Tomita N, Morishita R, Higaki J, Kaneda Y, Mikami H, Ogihara T. Transient decrease in high blood pressure by in vivo transfer of antisense oligonucleotide against rat angiotensinogen. Hypertension. 1995;26:131-136.
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.