Abstract The spontaneously hypertensive rat (SHR) is a well studied animal model of genetic hypertension and heart disease of unknown cause. With the use of differential display, a transcript was found in SHR myocardium that on sequence analysis was identified as an endogenous retrovirus (ERV). ERV gene expression was greater than an order of magnitude increased in adult SHR hearts relative to age-matched normotensive Wistar-Kyoto rats and was further increased in hearts from SHR with heart failure. In situ hybridization studies demonstrated that increased ERV gene expression was localized to myocardial cells. Increases in ERV transcripts in SHR suggest a possible link between inherited proviral elements and genetic hypertensive heart disease.
The spontaneously hypertensive rat (SHR) is a genetic model of hypertension1 that exhibits many of the cardiovascular features observed in humans with essential hypertension.2 Studies of SHR and normotensive rat strains provide an opportunity to make comparisons that may further elucidate fundamental mechanisms underlying hypertension and its pathological effects on the heart. Differential display has proved to be a useful tool for comparing differences in gene expression.3 Using this technique, we identified increased expression of an endogenous retrovirus (ERV) in SHR hearts. ERVs belong to a large family of mobile DNA elements that have elicited considerable interest primarily because of their relation to cancer and autoimmune disease. The family of retroviruses identified in the present study resides in the germ line of rats. Abundant expression has been reported to be associated with liver4 5 and prostate6 malignancies and is now observed in hearts from the SHR.
Male SHR and normotensive Wistar-Kyoto rats (WKY) were purchased from Taconic Farms (Germantown, NY). Rats were housed two per container and fed regular rat chow and water. All experiments were conducted in accordance with institutional guidelines and the Guide for the Care and Use of Laboratory Animals (US Department of Health and Human Services, NIH publication 86-23). After age 18 months, SHR were observed several times a week and studied when they developed tachypnea and labored respirations.7 8 Age-matched SHR without respiratory disturbances and WKY were also studied. One- and 12-month-old SHR were also studied, as were 24-month-old SHR administered the angiotensin-converting enzyme inhibitor captopril (2.0 g/L drinking water) from age 12 to 24 months. Animals were killed and hearts quickly removed and placed in oxygenated Krebs-Henseleit solution9 at 28°C, in which the right and left ventricles were carefully dissected. Tissues, including samples of liver, lung, kidney, and smooth and skeletal muscles, were gently blotted, weighed, and then immediately frozen in liquid nitrogen for subsequent analysis. Left and right ventricular wet weights normalized by body weight were used as indexes of ventricular hypertrophy.
RNA samples for differential display and Northern blot analysis were isolated from the left ventricles of study animals using the method of Chomczynski and Sacchi10 with modifications. Samples (100 to 200 mg) of myocardium, immediately frozen in liquid nitrogen at the time of animal death, were pulverized to a fine powder in liquid nitrogen with a mortar and pestle. The frozen sample was transferred to a glass-polytetrafluoroethylene homogenizer and extracted with 1 mL of a 4 mol/L solution of guanidinium thiocyanate containing 25 mmol/L citrate buffer, pH 7.0, 0.5% sarcosyl, and 0.1 mol/L 2-mercaptoethanol. Proteins and DNA were then removed by phenol-chloroform extraction, and the RNA was precipitated by adding isopropanol and cooling to −20°C. Finally, the RNA was redissolved and precipitated again.
RNA for differential display was further purified by treatment with RNAase-free DNAase for 30 minutes at 37°C. DNAase was removed by phenol/chloroform extraction, and then purified RNA was recovered by precipitation in ethanol. The precipitate was dissolved in water to make a stock solution, aliquoted in 1- to 2-μg quantities, and kept frozen until needed.
Differential Display and Sequence Analysis
The cDNA templates for polymerase chain reaction (PCR) amplification of mRNA samples were obtained by reverse transcription of total RNA from differing hearts using 5′-AAGCTTTTTTTTTTTG-3′, 5′-AAGCTTTTTTTTTTTC-3′, or 5′-AAGCTTTTTTTTTTTA-3′ oligonucleotides as anchored primers and Moloney murine leukemia virus reverse transcriptase (GeneHunter Co) with the RNA Image kits.
Each reverse transcription mix was amplified by random priming PCR containing [α-35S]dATP with the appropriate anchored oligonucleotide and 24 different arbitrary 13-mer primers, as provided by GeneHunter Co. The number of oligo pairs (72) was chosen so at least 95% of RNA transcripts would be displayed.11 These cDNAs were separated by 6 mol/L urea polyacrylamide sequencing gels. Autoradiographs of these gels were evaluated by visual comparison of individual band intensities of differing heart samples run side by side (see Fig 1⇓). cDNAs were recovered from bands that appeared to be differentially expressed by excision of the appropriate section of the dried gel and extraction with hot water. The recovered DNA was then reamplified and cloned into pCRTrap cloning vector (GeneHunter Co). Plasmid DNA was purified from cultures of transformed Escherichia coli cells for subsequent sequencing. DNA sequencing was carried out manually using the dideoxy chain termination method with Sequenase 2.0 (Amersham) and [α-35S]dATP (DuPont–New England Nuclear). Sequences were compared with the currently available sequence database using the Geneworks program (Intelligenetics).
Heart total RNA (10 μg) from control and treated animals was separated on 1.3% agarose formaldehyde gels and blotted onto nylon membranes (Duralon). The blots were hybridized with random primed 32P-labeled DNA previously prepared by PCR amplification of differentially expressed cDNAs. Hybridization was carried out at 68°C for 1 hour using 20×106 cpm probe per 15 mL QuikHyb hybridization solution. Blots were washed twice for 15 minutes in 2× SSC and 0.1% SDS at 25°C followed by a single stringency wash in 0.1× SSC and 0.1% SDS at 60°C for 30 minutes. The relative amount of mRNA per lane was determined by exposing the blots to Fujifilm RX (Fuji Photo Film Co) with an intensifying screen at −70°C and measuring the density of the exposed band with a laser densitometer (Bio-Rad GS-700). mRNA levels were normalized using 18S rRNA or cardiac α-actin mRNA, which we have previously demonstrated does not change in hypertrophied and failing hearts relative to controls.12
Genomic DNA was isolated by standard methods.13 Tissue samples (100 to 400 mg) were powdered with a mortar and pestle under liquid nitrogen. After homogenization, samples were digested with proteinase K at 50°C for 16 hours. The samples were then extracted with a mixture of chloroform, isoamyl alcohol, and phenol (49:1:50) and precipitated with ethanol. Genomic DNA (12 μg) from each sample was digested with Bgl II and HindIII at 37°C14 ; 10 μg of the DNA was then loaded on a 0.7% agarose gel and separated by electrophoresis in 0.5× Tris/boric acid/EDTA buffer. After denaturation and neutralization, the DNA was transferred to Duralon membranes. Blots were prehybridized, probed, and washed as described for Northern blotting.
Localization of Transcripts: Studies of In Situ Hybridization
Hearts from SHR and WKY were briefly perfused at 28°C with oxygenated Krebs’ solution at a pressure of 100 mm Hg. Hearts were then perfused with freshly made 4% paraformaldehyde/phosphate-buffered saline buffer and removed from the perfusion apparatus. Parallel transverse slices approximately 1 to 2 mm thick, midway between the apex and base and encompassing both right and left ventricles, were obtained. After sectioning, tissue slices continued to be fixed in paraformaldehyde. The tissues were then dehydrated and embedded in Paraplast Plus embedding medium. Serial sections 4 μm thick were prepared for in situ hybridization as described by Sassoon and Rosenthal.15 Single-stranded sense and antisense RNA probes were transcribed from a linearized pCRTrap (GeneHunter Co) containing the ERV cDNA fragment and radiolabeled with [35S]UTP. After incubation, the DNA templates were digested. Sense and antisense probes were subjected to alkaline hydrolysis to decrease fragment lengths to approximately 150 bases. Probes were then extracted with phenol/chloroform and precipitated in ethanol. Antisense and sense sections were obtained from adjacent tissue slices.
Data are presented as mean±SEM. Statistical analysis was performed with one-way ANOVA to evaluate differences among the three groups, and post hoc comparisons between groups were performed by Tukey’s procedure. Differences were considered significant at a value of P<.05.
Heart failure was considered present when animals exhibited pleuropericardial effusions, atrial thrombi, and right ventricular hypertrophy (Table⇓). Data were compared with those from age-matched SHR and WKY without these findings. The degree of hypertrophy and pathological findings consistent with the presence of heart failure were similar to those observed in previous studies of these animals.7 8 16 SHR treated with captopril from 12 to 24 months of age did not demonstrate effusions, atrial thrombi, or right ventricular hypertrophy.
Isolation and Identification of Differentially Expressed ERV
Autoradiographs of complementary DNA obtained by reverse transcription–PCR of RNA samples from nine left ventricles—three left ventricles each from failing SHR (SHR-F), nonfailing SHR (SHR-NF), and age-matched WKY—are shown in Fig 1⇑. The differentially expressed band extracted from the gel was a 249-bp sequence that was approximately 95% homologous to an ERV (RATRLTR) isolated from rat tumors (see Reference 55 ; Fig 2⇓). Differential expression was then tested on Northern blots made with total RNA of the original left ventricular samples.
Northern Blot Analysis
ERV gene expression in SHR and WKY was compared on Northern blots prepared with RNA pooled from the hearts of age-matched SHR-F, SHR-NF, and WKY (n=5 per group; Fig 3⇓). Hybridization was carried out with the cloned cDNA fragment as a probe. The approximately 7.5-kb ERV transcript was found to be abundant in SHR samples but only marginally detected in the normal WKY. Although there appeared to be a difference in ERV gene expression between SHR-NF and SHR-F hearts on average (Fig 3⇓), Northern blot analysis of RNA from three to four individual hearts per group revealed considerable variation among SHR. Therefore, we directly compared transcripts from nine additional SHR-NF and six age-matched SHR-F and found an approximately twofold increase in ERV gene expression in SHR-F relative to SHR-NF (P<.05).
ERV transcripts were compared in 1-, 12-, and 20-month-old SHR (three to five samples per age group) and in captopril-treated relative to untreated SHR (four samples per group). Relatively little difference in ERV gene expression was observed in the SHR hearts with age (Fig 4⇓) or with captopril treatment (data not shown). Comparison of cardiac ERV gene expression relative to other tissues in failing SHR is also shown in Fig 4⇓. ERV transcripts, although detectable in liver, lung, kidney, and skeletal muscle, were considerably increased in SHR myocardium relative to the other tissues examined.
Analysis of Genomic ERV by Southern Blot
We isolated genomic DNA from both WKY and SHR to further identify and analyze the proviruses present. For Southern blot analysis, two different six-cutter restriction enzymes were chosen: Bgl II, which cuts the provirus twice, yielding a 4830-bp fragment; and HindIII, which cuts only once within the retroviral sequence at 5111 of RATRLTR (accession No. D9300055 ). The ERV DNA fragment hybridized to an approximately 5-kb fragment of the Bgl II digest, in good agreement with data reported for tumor ERV (see Reference 44 ; Fig 5⇓). HindIII digestion, as shown on the same blot (Fig 5⇓), yielded several different fragments. The bands, all of them weaker than the band obtained with Bgl II digestion, suggest the presence of multiple genomic copies of ERV or related proviruses. WKY and SHR contained similar banding patterns.
Localization of Transcripts: Studies of In Situ Hybridization
In situ hybridization studies were carried out in cross-sections of SHR and WKY hearts. Antisense sections demonstrated localization of ERV transcripts to myocytes or portions of myocytes in SHR hearts (Fig 6⇓). Sense sections demonstrated no focal collections of grains, and ERV transcripts could not be detected in WKY hearts. ERV gene expression in the SHR myocytes was nonuniform; expression was marked in some cells (Fig 6⇓) and absent in others. In some SHR sections, areas of the myocardium appeared to show increased ERV gene expression, but transcripts could not be identified as originating from specific myocardial cell locations such as interstitium, perivascular regions, epicardium, or endocardium. Transcripts were observed in both right and left ventricles.
The present study demonstrates for the first time a marked increased in the gene expression of an ERV in association with genetic hypertension in the SHR. Martin et al17 recently identified a similar ERV from a heart cDNA library obtained from a Wistar rat. In their study, RNA blot analysis showed that the ERV transcript was expressed in heart and liver and that a 10-fold decrease in cardiac expression occurred between birth and 250 days. The present study demonstrates an increase of nearly two orders of magnitude in ERV gene expression in adult SHR myocardium relative to that in age-matched normotensive WKY myocardium. Although variability in expression was observed, transcripts were significantly further increased in failing SHR hearts.
Suzuki et al4 first identified a similar ERV (RATRLTR) expressing transcripts in rat liver tumors. Nakamuta et al5 presented the complete nucleotide sequence of the 7.5-kb RATRLTR. This ERV family has been shown to be expressed in a variety of tumors and tissues.4 5 6 17 In the present study, differential display of RNA isolated from SHR myocardium yielded a 249-bp cDNA with approximately 95% homology to bases 1556 through 1805 of RATRLTR (accession No. D9300055 ; Fig 2⇑). The cDNA fragment isolated by differential display (ERV) hybridized to a 5-kb Bgl II–Bgl II fragment on a genomic Southern blot similar to that observed for RATRLTR,4 demonstrating that ERV expressed in SHR hearts is similar to RATRLTR. However, because of incomplete sequence identity and the small fragment sequenced, ERV identified in the present study cannot be determined as being RATRLTR. Regional variations in homology (ranging from <30% to 98%) in portions of the U3′ and R domains of the long terminal repeats of this ERV family have been described.17 Variations in lengths of the U3′ domains have also been reported. Thus, the retrovirus identified in the present study cannot be assumed to be identical to RATRLTR.
Most ERVs have termination codons or deletions interrupting potential open reading frames and are nonpathogenic. Gene expression, when it occurs, has been associated with neoplastic disease and enhanced mitotic activity. Suzuki et al4 demonstrated a smear pattern on Northern blot that may be interpreted as suggesting a general upregulation of this ERV family. The present study demonstrates a single discrete band that may be consistent with expression at a single genomic site. Although it is possible that augmented ERV gene expression may simply be a marker associated with the SHR, Nakamuta et al5 point out that ERV sequences are expressed in a very limited number of states. Integrated ERVs may result in the activation or inactivation of neighboring genes. Long terminal repeats contain enhancer-promoter elements that may modify gene expression and tissue phenotype. Mechanisms by which ERVs may exert their effects are reviewed by Krieg et al.18 In addition to integration and insertional mutagenesis, retroviruses may have direct actions via produced proteins and indirect effects due to the stimulation of immune responses. Recombination with infectious retroviruses may result in a recombinant retrovirus.
Studies to map genetic loci involved in blood pressure regulation have been outlined by Kurtz and St Lezin.19 A number of investigators have suggested that a blood pressure–regulatory sequence may be located in or near the renin gene in the SHR.20 21 Both cardiac decompensation and particularly angiotensin-converting enzyme inhibition stimulate renin production in the SHR (Brooks WW et al, unpublished data, 1996). Increased ERV gene expression associated with cardiac decompensation in the SHR at first suggested a possible relationship between ERV and the renin gene. However, treatment of SHR with the angiotensin-converting enzyme inhibitor captopril resulted in no significant change in ERV mRNA. Therefore, no clear relation between ERV transcript and renin production is suggested from the present studies. Another possibility is that increased ERV transcript is secondary to an elevation in blood pressure. However, blood pressure is reduced after chronic captopril administration with no change in ERV gene expression. In addition, enhanced ERV gene expression is present in the 1-month-old SHR before hypertension is established; thus, it appears that increased ERV gene expression can be dissociated from high blood pressure.
An intriguing possibility is that enhanced ERV gene repression is associated with events underlying hypertension. In the 1-month-old SHR, events preceding hypertension may be in process. In the captopril-treated SHR, the factor or factors underlying the hypertension may continue to be produced even though the end product (eg, angiotensin II) of a sequence of events resulting in hypertension is inhibited. How might augmented ERV gene expression be specifically related to hypertension? One hypothesis might be that enhanced calcium transients22 and contractile state, as has been reported in SHR myocytes,23 24 result from the effects of a protein or proteins produced by ERV. Changes in vascular wall stress associated with an enhanced contractile state have been shown to alter vascular smooth muscle and be a factor contributing to hypertension.25 Studies to determine possible proteins produced by ERV and their effect on cardiac contractile function are among those that will be required to further elucidate the relation- ship between ERV and hypertension in the SHR. In summary, the present study is the first to demonstrate a marked increase in the gene expression of an in- herited provirus in myocytes from rats with genetic hypertension and raises the possibility of a link between inherited proviral structures and genetic hypertensive heart disease.
This work was supported by Merit Review funds from the Department of Veterans Affairs Medical Research Program. The authors would like to thank Kathleen G. Robinson for her technical assistance, Dr Edgar G. Lucey for his assistance with in situ photography, Dr Joel Maslow for his valuable discussions, and Robyn Squire for her help in the preparation of this manuscript.
Reprint requests to Oscar H.L. Bing, MD (151), Boston VA Medical Center, 150 S Huntington Ave, Boston, MA 02130.
- Received October 9, 1996.
- Revision received November 11, 1996.
- Accepted December 9, 1996.
Trippodo NC, Frohlich ED. Similarities of genetic (spontaneous) hypertension: man and rat. Circ Res. 1981;48:309-319.
Liang P, Pardee AB. Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction. Science. 1992;257:967-270.
Suzuki N, Fujiyoshi T, Maehara Y, Takahashi M, Yamamoto M, Endo H. A new family of LTR-like sequences abundantly expressed in rat tumors. Nucleic Acids Res. 1986;14:9271-9289.
Conrad CH, Brooks WW, Robinson KG, Bing OHL. Impaired myocardial function in spontaneously hypertensive rats with heart failure. Am J Physiol. 1991;260:H136-H145.
Bauer D, Miller H, Reich J, Riedel H, Ahrenkiel V, Warthoe P, Strauss M. Identification of differentially expressed mRNA species by an improved display technique (DDRT-PCR). Nucleic Acids Res. 1993;21:4272-4280.
Boluyt MO, O’Neill L, Meredith AL, Bing OHL, Brooks WW, Conrad CH, Crow MT, Lakatta EG. Alterations in gene cardiac expression during the transition from stable hypertrophy to heart failure. Circ Res. 1994;75:23-32.
Strauss WM. Preparation of genomic DNA from mammalian tissues. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K, eds. Current Protocols in Molecular Biology. New York, NY: John Wiley & Sons Inc; 1995:Unit 2.2.
Sambrook J, Fritch EF, Maniatis T. Molecular Cloning. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1989:Section 9.24.
Conrad CH, Brooks WW, Hayes JA, Sen S, Robinson KG, Bing OHL. Increased myocardial stiffness in the spontaneously hypertensive rat with heart failure. Circulation. 1995;91:161-170.
Krieg AM, Gourley MF, Perl A. Endogenous retroviruses: potential etiologic agents in autoimmunity. FASEB J. 1992;6:2537-2544.
Kurtz TW, St Lezin EM. Gene mapping in essential hypertension. J Am Soc Nephrol. 1992;3:28-34.
Kurtz TW, Simonet L, Kabra PM, Wolfe S, Chan L, Hjelle BL. Cosegregation of the renin allele of the spontaneously hypertensive rat with an increase in blood pressure. J Clin Invest. 1990;85:1328-1332.
Brooksby P, Levi AJ, Jones JV. Investigation of the mechanisms underlying the increased contraction of hypertrophied ventricular myocytes isolated from the spontaneously hypertensive rat. Cardiovasc Res. 1993;27:1268-1277.
Oparil S, Chen YF, Naftilan AJ, Wyss JM. Pathogenesis of hypertension. In: Fozzard HA, Haber E, Jennings RB, Kate AM, Morgan HE, eds. The Heart and Cardiovascular System. 2nd ed. New York, NY: Raven Press Publishers; 1992:295-331.