Antisense Inhibition of AT1 Receptor in Vascular Smooth Muscle Cells Using Adeno-Associated Virus-Based Vector
Abstract—Vascular smooth muscle cells (VSMCs) are the main peripheral target for vasoconstriction and growth-promoting activity of angiotensin II (Ang II), acting through angiotensin type 1 receptors (AT1-R). Current antihypertension treatments include daily reductions in the effects of Ang II. To decrease an effect of Ang II in a prolonged fashion, we have developed an adeno-associated virus (AAV) vector with antisense DNA for AT1-R. AAV has many advantages over other viral vectors. AAV is nonpathogenic, does not stimulate inflammation or immune reaction and enters nondividing cells, and provides stable long-term gene expression. To test AAV in VSMCs, we constructed and tested plasmid AAV (pAAV) and recombinant AAV (rAAV) with AT1-R antisense DNA. rAAV was constructed with a cassette containing a cytomegalovirus promoter and the cDNA for the AT1-R inserted in the antisense direction. The cassette was packaged into the virion. Transfection of VSMCs with the pAAV antisense to AT1-R produced a significant reduction in the amount of AT1-R (P<0.01). Transduction of VSMCs with the rAAV-AT1-R-AS at MOI of 5 also showed significant reduction of AT1-R and long-lasting expression of the transgene for at least 8 weeks. The reduction of AT1-R number in VSMCs was concomitant with a decrease in the Ang II–stimulated increase of intracellular calcium. The results show that AAV vector delivers AT1-R antisense to inhibit AT1-R in VSMCs. For the purpose of gene therapy for hypertension, it is necessary to demonstrate the effectiveness of a vector system in VSMCs. This study provides support for the potential use of AAV AT1-R antisense in VSMCs.
- recombinant adeno-associated virus vector
- receptors, angiotensin
- muscle, smooth, vascular
- protein, green fluorescent
Vascular smooth muscle cells (VSMCs) are the main peripheral target for hormone angiotensin II (Ang II) acting through angiotensin II type 1 receptors (AT1-R).1 In many models of hypertension the renin-angiotensin system (RAS) is overactive and contributes to the hypertension.2 Inhibiting the RAS by reducing Ang II production with angiotensin-converting enzyme (ACE) inhibitors3 or antagonizing AT1-R4 are current therapeutic approaches for controlling hypertension. To develop more specific, longer-lasting approaches to inhibiting RAS, we have previously designed antisense (AS) oligonucleotides to AT1-R and angiotensinogen mRNA.5 6 These oligonucleotides can be delivered in vivo and produce significant decreases in blood pressure for 3 to 7 days with a single injection.6 7 8 9 To produce longer-lasting antisense inhibition to reverse or ameliorate high blood pressure, we are delivering AAV-AT1-R-AS in viral vectors.10 Iyer et al11 showed that a single injection of retrovirus AT1-R-AS vector in neonate spontaneously hypertensive rats (SHR) prevented hypertension from developing for 90 days. Furthermore, Martens et al12 demonstrated that treatment with this recombinant retrovirus not only attenuates the development of hypertension for up to 120 days, it also prevents renovascular and cardiac pathophysiological changes in the treated SHR.
For human therapy, the challenge is to produce viral vector delivery of AT1-AS to reverse hypertension. There are several viral vectors to choose from, including retroviruses, adenovirus, herpes simplex virus, lentivirus, and adeno-associated virus (AAV).10 All have certain advantages and disadvantages, but the AAV offers several attractive advantages and few disadvantages. AAV is safe to use and does not induce any pathogenic or inflammatory response. The AAV is a defective parvovirus that cannot replicate in cells without the presence of a helper wild-type adenovirus or herpes virus.13 Recombinant AAV (rAAV) has the capability of gene transfer with stable long-term expression. The rAAV vector has a carrying capacity of about 4.7 kb, which is sufficient to load with antisense DNA, promoters, and reporter genes.
In this study we report the results of using AAV vector that we constructed to deliver AT1-R mRNA antisense with a cytomegalovirus (CMV) promoter. The objectives of the present study were to (1) demonstrate in vitro the effectiveness of the plasmid AAV (pAAV) as a vector for transgenic expression in VSMCs; (2) study the biological effect of pAAV-AT1-R-AS on AT1-R number and angiotensin signaling pathway; (3) test whether rAAV-AT1-R-AS produces long-lasting expression in nonselected VSMCs, using the neomycin resistance gene as a reporter gene; and (4) study the effect of rAAV-AT1-R-AS on AT1-R and cell physiology by measuring changes in intracellular calcium concentration in response to Ang II in VSMCs.
Construction of Recombinant AAV Vector Plasmids (pAAV)
The AAV-derived vector can be adapted for several genes and promoters between the inverted terminal repeats (ITRs) at each end (Figure 1⇓). The AAV genome is removed leaving only the ITRs. The vector has a 4.7-kb carrying capacity. Foreign genes, such as antisense DNA and marker genes, can be inserted together with foreign promoters. Figure 1⇓ shows the plasmid AAV with the A victoria green fluorescent protein gene (pAAV-gfp) designed by Zolotukhin et al14 and designated pAAV. The plasmid has a CMV promoter-driven gfp gene and also contains a neomycin resistance (aminoglycoside phosphotransferase) gene, neor, driven by a thymidine kinase promoter.
For the antisense vector, a 749-bp fragment (−183 to 566) of rat vascular AT1-R cDNA (provided by T.J. Murphy in the pKSCa18b plasmid)15 was amplified with polymerase chain reaction (PCR) and ligated to pAAV vector in the antisense orientation in place of gfp and designated pAAV-AS (Figure 1⇑).
Method to Prepare Recombinant AAV
Briefly, to prepare recombinant AAV, human embryonic kidney (HEK293) cells were transfected with plasmid vector containing AT1-R-AS and AAV terminal repeats (pAAV-AS), together with helper plasmid delivering rep and cap genes (necessary for AAV replication) in trans using the calcium phosphate method,16 and transduced with adenovirus. Adenovirus was inactivated by incubation of the cell lysate at 56°C for 15 minutes. AAV was purified on CsCl gradient.
Rat thoracic aorta VSMCs (a kind gift from Dr T. Inagami) prepared according to the method of Eguchi et al17 were grown in DMEM with l-glutamine and 4.5 g/L glucose supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin in an incubator (Quene Systems, Inc,) with a humidified atmosphere of 5% CO2 and 95% air at 37°C. Cells were used between passages 12 and 15 for transfection or transduction; they were then kept 48 hours in case of transfection or split every week for 8 weeks for transduction experiments.
Transfection of VSMCs With the Plasmid pAAV
VSMCs were plated (day 1) in 6-well plates and transfected the following day (day 2) at a confluence of ≈60% to 70% using lipofectamine reagent (Gibco BRL) according to manufacturer protocol; 2 μg of the plasmid (pAAV-AS) and 10 μL lipofectamine (2 mg/mL) were used per well. Cells were used 48 hours after transfection to measure AT1-R by a ligand-binding method.
Green Fluorescent Protein Detection
VSMCs were grown and transfected using lipofectamine reagent (Gibco BRL) on a cover slide and embedded in a 35-mm tissue culture dish. The same field of cells was observed under visual light and phase-contrast ring of Axiovert 135 (Zeiss, Germany) and then under UV light with a green fluorescent protein (GFP)–specific filter. The amount of the green epifluorescent area (UV light and GFP-specific filter) as a percentage of total area (visual light, phase-contrast ring) was estimated by 2 independent investigators. Photographs were taken 48 hours after transfection using a Nikon N6006 camera attached to Axiovert 135.
Transduction of VSMCs With rAAV
VSMCs were plated in 10-cm plates to be ≈60% confluent on the next day. On the day of the experiment, cell medium was removed, each plate was washed with 5 mL of Opti-MEM, cells were covered with 5 mL Opti-MEM, and recombinant AAV was added at MOI of 5. After 3 days the medium was changed to DMEM with 10% FBS and antibiotics. Control cells were treated in the same manner, except that no virus was added. Cells were passaged once a week. Transgene expression was studied from the third day, and AT1 binding was done up to 8 weeks.
Transgene and AT1-R Gene Expression by RT-PCR
To check the time course of the neor gene expression in the cells and the effect on AT1-R gene expression, RNA from control VSMCs and cells transfected with plasmid pAAV-AS was isolated from 2 to 48 hours using PureScript reagent (Gentra). Control consisted of cells treated with lipofectamine only during transfection. RNA from VSMCs treated with virus rAAV-AS was prepared every week after transduction for 8 consecutive weeks. RNA (1 to 3 μg) was digested with DNase I (RNase free) for 10 minutes at 37°C in the presence of 5 U of RNase inhibitor. After heat inactivation of DNase, RNA was reversed transcribed for 50 minutes at 42°C using SuperScript II Reverse Transcriptase and oligo(dT) as a primer (Gibco BRL). The reaction was stopped by heating samples for 15 minutes at 70°C. Next, 1 μL from the reverse transcription (RT) reaction was added to 49 μL final-volume PCR reaction, and amplification was performed using a set of neor primers yielding a 757-bp fragment. The 5′ primer was 5′-GGATTGCACGCAGGTTCTCCG-3′ and a 3′ primer was 5′-CGATAGAAGGCGATGCGCTGC-3′.18 Amplification for transfected cells was performed for 40 cycles with annealing at 65°C. RNA from transduced cells after RT was subjected to 35 cycles of PCR with the first set of neomycin primers as above and then nested PCR with 5′ primer of 5′-GGCTATTCGGCTATGACTGG-3′ and 3′ primer of 5′-GGCGATACCGTAAAGCACGA-3′, yielding a 688-bp product. In parallel, 1 μL from the RT reaction was added to 49 μL final-volume PCR reaction, and amplification was performed using a set of AT1-R primers yielding a 719-bp fragment. The 5′ primer was 5′-TTCCACCCGATCACCGATCA-3′ and the 3′ primer was 5′-CGGTGGATGACAGCTGGCAA-3′.15 The 5′ primer for AT1-R was complementary to the sequence outside the antisense part, so the antisense cDNA was not amplified during PCR. Amplification was performed for 40 cycles with annealing at 60°C. As a positive control, plasmid containing neomycin resistance gene or AT1-R cDNA was amplified as described above. For normalization, the sample RNA was also amplified using GAPDH primers. The 5′ primer was 5′-CCCATCACCATCTTCCAGGA-3′ based on exon 3 sequence, and the 3′ primer was 5′-CGCCTGCTTCACCACCTTCT-3′ based on exon 7 sequence. Amplification products were analyzed on 1% agarose gel stained with ethidium bromide.
125I-[Sar1,Ile8]Ang II Binding to VSMCs
At 48 hours after transfection or up to 8 weeks after transduction, VSMCs growing in 6-wells plates were washed twice with 2 mL phosphate-buffered saline (PBS) at room temperature and incubated for 60 minutes with 1 mL PBS solution containing 1.6% BSA and 0.2 nmol/L of the radiolabeled Ang II antagonist 125I-[Sar1,Ile8]Ang II in the absence and presence of 1 μmol/L losartan to determine the total and nonspecific AT1-R binding, respectively. At the end of the incubation cells were washed three times with 2 mL of ice-cold PBS and lysed with 0.5 mL 0.1 N NaOH for 1 hour. Cells were scraped, wells were rinsed with 0.5 mL of water, and the radioactivity of combined solutions was determined with a Beckman γ-counter. Protein content was measured in cell lysates.19
Intracellular Calcium Measurement
Intracellular calcium concentration ([Ca2+]i) in VSMCs was measured using imaging fluorescence microscopy. Briefly, the cells were loaded with fura 2 by incubation with 5 μmol/L membrane-permeable fura 2 acetoxymethylester (fura 2-AM dissolved in 1 mmol/L DMSO stock) in Tyrode’s solution for 30 to 60 minutes at 37°C. The culture dishes containing cells were then centered in the optical field of a ×40 oil immersion fluorescence objective of an inverted microscope (Nikon Diaphot).
Randomly selected cells were illuminated alternately with UV light of 340 and 380 nm wavelength using an IonOptix chopper-based, electronically controlled, dual-excitation imaging fluorescence system. Cell fluorescence (emitted light) was collected through a 510-nm barrier filter before acquisition by an ICCD camera (Phillips FTM800). Fluorescence signals were digitized on-line using an IBM-PC compatible computer and IonOptix fluorescence image acquisition and analysis software. The fluorescence signals, F340 and F380, were background subtracted during the experiment. The signals are reported here as changes in F340/F380. This gives a relative indication of [Ca2+]i.
Data were analyzed with the use of standard statistical methods. ANOVA and Neuman-Keuls tests were applied. Group data are expressed as mean±SEM. Significance was set at the P<0.05 value.
Expression of GFP in VSMCs
The first experiment was to assess transfection efficiency of pAAV-gfp in VSMCs. Expression of GFP in the cells was observed 48 hours after transfection by comparing the area showing green fluorescence with a total area of cells. In the control cells treated with lipofectamine, the field under UV light with a GFP-specific filter was black. In pAAV-GFP transfected cells, the amount of green epifluorescent area (UV light and GFP-specific filter) as a percentage of total area (visual light, phase-contrast ring) was estimated by 2 independent investigators as being ≈80% (Figure 2⇓).
Time Course of the Neomycin Resistance Gene Expression in Transfected VSMCs by RT-PCR
Time course of the neomycin resistance gene expression in host cells from 2 hours to 48 hours after the start of transfection was shown by RT-PCR of RNA extracted from VSMCs transfected with pAAV-AS (Figure 3⇓, top). The transgene band of the expected size of 757 bp was weak but was visible starting at 2 hours of transfection and increased at later time points. No visible band was present in control cells treated with lipofectamine only.
Time Course of the AT1-R Gene Expression in Transfected VSMCs by RT-PCR
Time course of the AT1-R gene expression from 2 hours to 48 hours after the transfection was shown by RT-PCR of RNA extracted from VSMCs transfected with pAAV-AS (Figure 3⇑, middle). Starting at 5 hours, AT1-R mRNA was decreased compared with control cells treated with lipofectamine only. After 48 hours AT1 mRNA in transfected cells was 42±7% (n=3, normalized by GAPDH expression).
Effect of pAAV-AS Transfection on Ang II Receptor Binding in VSMCs
The pAAV-AS plasmid transfected VSMCs showed a 74% decrease in AT1-R number assayed at 0.2 nmol/L 125I-[Sar1,Ile8]Ang II compared with the nontransfected cells in Opti-MEM (14.9±4.8 versus 56.7±8.4 fmol/mg, n=3; P<0.005) or 63% reduction compared with the nontransfected cells treated with lipofectamine (14.9±4.8 versus 40.2±5.1 fmol/mg, n=4; P<0.005). Figure 4⇓ shows the significant reduction in AT1-R number determined 48 hours after transfection.
Effect of pAAV and rAAV-AS on Ang II–Stimulated Increase in Intracellular Calcium in VSMCs
The effects of AT1-R-AS was tested on changes in [Ca2+]i as a measure of second-messenger activity in the Ang II signaling pathway. In both transfected by plasmid pAT1-R-AS (Figure 5A⇓) and transduced by virus rAAV-AS VSMCs (Figure 5B⇓), there was a significant decrease in the Ang II–stimulated change in [Ca2+]i. In plasmid or virus-treated cells there was a significant attenuation of the Ang II response: Ang II only increased [Ca2+]i 72±12% or 113±22% versus 308±15% or 278±18% in the control, respectively (n=157 cells for plasmid-treated cells, n=202 cells for virus-transduced VSMCs; difference between control and treated group is statistically significant at P<0.01). At the same time, AT1-R-AS as a plasmid or virus had no effect on depolarization-stimulated (80 mmol/L KCl) or α-adrenergic receptor–dependent increases (1 μmol/L phenylephrine) in [Ca2+]i. There was no difference between the Rmin and Rmax of control or treated cells.
Time Course of the Neomycin Resistance Gene Expression in Virus-Transduced VSMCs by RT-PCR
The time course of the neomycin resistance gene expression in host cells was shown by RT-PCR of RNA extracted from VSMCs transduced with virus rAAV-AS. Neomycin resistance gene was expressed from 1 week to at least 8 weeks after the transduction (Figure 6⇓). No visible band was present in control cells. After that time, however, both control and transduced cells became difficult to passage and culture was discontinued.
Effect of rAAV-AS on Ang II Receptor Binding in VSMCs
Starting from the 2nd week after treatment, the rAAV-AS virus-transduced, nonselected VSMCs showed significant reduction in AT1-R number assayed at 0.2 nmol/L 125I-[Sar1,Ile8]Ang II compared with the nontransduced cells, with a mean decrease of 21% (Table⇓). This continued for 8 weeks.
This study was carried out in 2 stages, first with AAV-based plasmid and second with recombinant AAV. The results show that AAV-based plasmid can efficiently transfect VSMCs. The reporter gene, green fluorescent protein, was present in approximately 80% of cells. In other cell lines, such as NG108-15, ATt-20, and L929 using the same plasmid and transfection method, we observed a similar high efficiency of transfection.20 AAV-based plasmid carrying AT1-R antisense and neor gene is expressed in these cells as early as 2 hours after the start of the transfection. The pAAV AT1-R-AS decreased AT1-R mRNA starting at 5 hours. AT1-R number determined by ligand binding 48 hours after transfection showed a statistically significant decrease in antisense-transfected cells compared with control cells. The decrease in receptor number correlates well with the transfection efficiency in these cells as shown by the expression of green fluorescent protein. As a test of the physiological effect of pAAV, we measured the changes in [Ca2+]i response to Ang II. The decreased number of AT1-R was concomitant with a reduction in the intracellular calcium response. This indicates that pAAV-based AT1-R antisense is effective in reducing the activity of endogenous VSMC Ang II. Also, when rAAV-AT1-R-AS was studied, the AT1-R number and [Ca2+]i were reduced. We have noticed previously that antisense reduces receptor number by a small amount, leading to a disproportionally larger physiological change. We hypothesize that there is a threshold for effect or a lack of spare receptors. Without an irreversible blocker of AT1-R we have not been able to test this hypothesis. The expression of the transgene was prolonged and present in the cells for at least 8 weeks after a single exposure to rAAV.
The goal of this research is to establish new ways of reducing blood pressure by gene manipulation to achieve long-lasting effects without toxicity or side effects. Although current ACE inhibitors and angiotensin receptor blockers (ARBs) are very effective in antihypertension therapy, they are drugs foreign to the body. As such, they have side effects. Additionally, they must be taken daily, which increases noncompliance problems. The antisense approach is a biological one in which the DNA can be integrated into the genome and constantly produces antisense mRNA. Inhibiting angiotensin activity is only one approach to gene manipulation. Other methods such as increasing gene expression of vasodilator proteins such as bradykinin and atrial natriuretic peptide or increasing nitric oxide synthase are also being studied and proving effective in animal models.21 22
We have used AAV-AT1-R-AS to reduce blood pressure in SHR for several weeks by delivery to the brain.10 23 The mechanism of action, however, is quite different from peripheral application, where the most likely target site is blood vessels. Recently, Martens et al12 showed that intracardiac injection of retrovirus AT1-R-AS in neonate SHR prevented hypertension from developing and normalized blood vessel vascular reactivity and voltage-dependent K+ current density.
The properties of recombinant AAV vectors include safety because they are nonpathogenic and do not replicate. AAV can transduce nondividing cells, whereas retrovirus cannot. AAV vectors target homologous chromosomal genes at high frequencies and introduce particular modifications without additional mutations.24 Adenoviruses produce viral genes that stimulate inflammatory responses. For treatment of cardiovascular diseases such as hypertension, it is important to maintain transgene expression for clinically significant amounts of time. Therefore, AAV has positive potential for ultimate use in gene therapy.
We constructed an rAAV with antisense to AT1-R. The RT-PCR detection of neomycin resistance gene confirmed that the rAAV-AS was being expressed for several weeks in transduced VSMCs without selecting the cells. The results indicate efficient uptake, long-lasting expression, and effective inhibition of Ang II–induced Ca2+ response. The mechanism for AS-AT1-R appears to be through integration of rAAV into the genome. There is extensive evidence that this occurs in the wild-type AAV13 and recent evidence that rAAV is also integrated.25 The transgene expression would produce specific AS mRNA targeted to AT1-R mRNA and hybridize with endogenous AT1-R mRNA, reducing translation of the AT1-R protein. The reduction in receptor number reduces the response to Ang II as shown by the reduced [Ca2+]i. Although these experiments are in vitro, the results suggest that rAAV-AT1-R-AS will be effective in vivo, in VSMCs of blood vessels, for reducing the vasoconstrictive effects of Ang II. In those forms of hypertension, where there is either increased Ang II or increased receptor responsiveness or both, the results support that the rAAV AT1-R-AS will likely produce prolonged attenuation of Ang II–induced vasoconstriction.
This study was supported by NIH grant HL-27334 (MERIT award).
- Received September 15, 1998.
- Revision received October 16, 1998.
- Accepted October 29, 1998.
Timmermans PB, Smith RD. Angiotensin II receptor subtypes: selective antagonists and functional correlates. Eur Heart J. 1994;15(suppl D):79–87.
Alderman MH, Mahavan S, Ooi WL, Cohen H, Sealey JE, Laragh JH. Association of the renin-sodium profile with the risk of myocardial infarction in patients with hypertension. New Engl J Med. 1991;324;1098–1104.
Waeber B, Nussberger J, Brunner HR. Angiotensin-converting enzyme inhibitors in hypertension. In: Laragh JH, Brenner BM, eds. Hypertension: Pathophysiology, Diagnosis, and Management. 2nd ed. New York, NY: Raven Press; 1995:2861–2872.
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 oligodeoxynucleotides against rat angiotensinogen. Hypertension. 1995;26:131–136.
Makino N, Sugano M, Ohtsuka S, Sawada S. Intravenous injection with antisense oligodeoxynucleotides against angiotensinogen decreases blood pressure in spontaneously hypertensive rats. Hypertension. 1998;31:1166–1170.
Phillips MI. Antisense inhibition and adeno-associated viral vector delivery for reducing hypertension. Hypertension. 1997;29:177–187.
Iyer SN, Lu D, Katovich M, Raizada MK. Chronic control of high blood pressure in the spontaneously hypertensive rat by delivery of angiotensin type 1 receptor antisense. Proc Natl Acad Sci U S A. 1996;93:9960–9965.
Martens JR, Reaves PY, Lu D, Katovich MJ, Berecek KH, Bishop SP, Raizada MK, Gelband CH. Prevention of renovascular and cardiac pathophysiological changes in hypertension by angiotensin II type 1 receptor antisense gene therapy Proc Natl Acad Sci U S A. 1998;95:2664–2669.
Muzyczka N, McLaughin S. Use of adeno-associated virus as a mammalian transduction vector. In: Gluzman Y, Hughes SH, eds. Current Communications in Molecular Biology: Viral Vectors. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1988:39–44.
Zolotukhin S, Potter M, Hauswirth WW, Guy J, Muzyczka NA. ‘Humanized‘ green fluorescent protein cDNA adapted for high-level expression in mammalian cells. J Virol. 1996;70:4646–4654.
Sambrook J, Fritsch E, Maniatis T. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1989.
Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein dye binding. Anal Biochem. 1976;72:514–517.
Mohuczy-Dominiak D, Phillips MI. Adeno-associated virus vector as a highly efficient transporter of exogenous DNA into cells. FASEB J. 1996;10:A447.
Lin KF, Chao J, Chao L. Human atrial natriuretic peptide gene delivery reduces blood pressure in hypertensive rats. Hypertension. 1995;26:847–853.
Lin KF, Chao L, Chao J. Prolonged reduction of blood pressure with human nitric oxide synthase gene delivery. Hypertension. 1997;30(pt 1):307–313.
Phillips MI, Mohuczy-Dominiak D, Coffey M, Galli SM, Ping W, Zelles T. Prolonged reduction of high blood pressure with an in vivo, nonpathogenic, adeno-associated viral vector delivery of AT1-R mRNA antisense. Hypertension. 1997;29:374–380.
Wu P, Phillips MI, Bui J, Terwilinger EF. Adeno-associated viral vector mediated transgene integration in nondividing cells. J Virol. 1998;72:5919–5926.