Processing of Chimeric Antisense Oligonucleotides by Human Vascular Smooth Muscle Cells and Human Atherosclerotic Plaque
Implications for Antisense Therapy of Restenosis After Angioplasty
Background Antisense oligonucleotides have been used in animals to inhibit the accumulation of vascular smooth muscle cells (VSMCs) after arterial injury. This has raised prospects for an oligonucleotide-mediated approach to prevent restenosis in patients undergoing angioplasty. However, little is known about the processing of oligonucleotides by human VSMCs or their bioavailability in human atherosclerotic tissue.
Methods and Results Oligonucleotides were synthesized with a mixture of unmodified and sulfur-modified linkages (S-chimeric oligonucleotides). These were more stable than unmodified oligonucleotides and could be recovered from within human VSMCs after 36 hours. Oligonucleotide antisense to human proliferating cell nuclear antigen mRNA specifically reduced DNA synthesis (P<.01) and proliferating cell nuclear antigen protein content (P<.05) in human VSMCs. Confocal microscopy of both live and fixed cells showed modest oligonucleotide uptake that was primarily nuclear. Surprisingly, cationic liposomes did not enhance nuclear uptake but led to extensive, punctated cytoplasmic loading without an enhanced antisense effect. Oligonucleotides incubated with human coronary atherosclerosis fragments associated with cells within 1 hour, despite the presence of abundant extracellular matrix.
Conclusions S-chimeric oligonucleotides are stable and can specifically inhibit gene expression in human VSMCs. Nuclear transport is a feature of oligonucleotide processing by human VSMCs, indicating a potential influence at the nuclear level rather than with cytoplasmic mRNA. Cationic liposomes increased oligonucleotide uptake but not intracellular bioavailability, and S-chimeric oligonucleotides can be incorporated into cells within human atherosclerotic plaque, despite the presence of a dense extracellular matrix.
The ability of antisense oligonucleotides to bind to complementary sequences in mRNA or its precursor constitutes a valuable tool for probing gene function.1 2 Oligonucleotide-mediated inhibition of gene expression also is being evaluated as a new paradigm for therapy of human disease.3 4 The potential impact of this technology on the treatment of vascular disease was illustrated recently by reports demonstrating that antisense oligonucleotides can be used to inhibit VSMC accumulation both in vitro5 6 7 8 and in vivo.9 10 11 12 13 14 The persistent problem of restenosis after angioplasty might therefore be addressed with antisense oligonucleotides.
Limitations to antisense oligonucleotide technology, however, have been recognized for many years,2 and it has become increasingly appreciated that despite perfect complementarity, potent and specific inhibition of gene expression by antisense oligonucleotides cannot be ensured.15 16 17 Evidence also shows that the efficiency and mechanism of cellular incorporation of oligonucleotides can vary considerably between cell types.18 Successful translation of the findings in animals to human vascular applications may therefore require a better understanding of the precise cellular fate and molecular interactions of oligonucleotides specifically in human VSMCs. Furthermore, if the therapeutic aim is to prevent restenosis after balloon angioplasty, oligonucleotides must be able to enter cells within or subjacent to atherosclerotic plaque. In patients undergoing angioplasty, this plaque typically is characterized by large regions of relative hypocellularity and abundant, dense extracellular matrix.19 These features are not found in animal models of balloon injury, such as those used to date to study antisense inhibition of gene expression, but could potentially limit cellular bioavailability of oligonucleotide in patients.
The present study was designed to address these issues by examining the stability, efficacy, and cellular processing of antisense oligonucleotides in human arterial smooth muscle cells and fragments of human atherosclerotic plaque. Cellular processing of oligonucleotides in vitro was evaluated with and without cationic liposomes. The use of cationic liposomes has been used widely to augment the delivery of DNA20 to cells; however, no data are available on the effect of these liposomes on the intracellular trafficking of oligonucleotides. We specifically studied oligonucleotides that were made up of a mixture of unmodified and sulfur-modified linkages, which we termed S-chimeric oligonucleotides. These oligonucleotides contain sulfur substitutions on internucleoside linkages at the 5′ and 3′ termini and normal phosphodiester linkages between the intervening nucleosides. The rationale for this design was that the reduced number of sulfur molecules may impart fewer nonspecific effects,21 but their presence at each terminus may still confer a degree of nuclease resistance.16 22 The cellular target for antisense inhibition was the PCNA. This target was chosen because it is a necessary protein for cell proliferation,23 it has been identified in human restenosis lesions,24 and previous antisense studies of nonhuman cells have shown an inhibitory effect on cell growth.6 23
The goals of the present study were (1) to evaluate the biological effect of S-chimeric antisense oligonucleotides against PCNA in cultured human VSMCs, (2) to define the extent of uptake and determine the intracellular fate of S-chimeric oligonucleotides in human VSMCs with confocal microscopy, (3) to evaluate the influence of cationic liposomes on the biological effect and intracellular handling of oligonucleotides, and (4) to determine whether S-chimeric oligonucleotides can be incorporated into cells of human fibrous atherosclerotic plaque.
Human VSMCs were cultivated from fragments of internal mammary artery retrieved from patients undergoing bypass surgery, as previously described.25 Cells were grown on a fibronectin-coated substrate (10 μg/cm2), and the identity of smooth muscle cells was confirmed by their growth pattern and positive immunostaining for smooth muscle α-actin.25
18-Base S-chimeric oligonucleotides were synthesized on an Applied Biosystems DNA synthesizer (model 380B). The standard oxidation cycle was changed at specific times so that sulfur-modified linkages were generated as the first four and last four internucleoside linkages. In some instances, standard phosphodiester oligonucleotides also were synthesized so that comparisons of biological stability could be performed. The antisense oligonucleotide was complementary to a sequence in the 5′ untranslated region (nucleotides −101 to −83) of human PCNA mRNA. Control sequences consisted of the sense oligonucleotide, a scrambled oligonucleotide (identical bases in a scrambled order), and an irrelevant oligonucleotide antisense to human α1(I) collagen mRNA (Fig 1⇓). The uniqueness of all oligonucleotides was confirmed by comparison with all sequences published in Genbank with GCG Sequence Analysis Software Package (Genetics Computer, Inc).
Oligonucleotides were 5′ end-labeled with [γ32P]ATP (3000 Ci/mmol, Du Pont/NEN) and T4 polynucleotide kinase (Promega). Unincorporated label was removed by exclusion chromatography with Bio-Spin 6 columns (Biorad). Specific activity was 1 to 5×108 cpm/μg oligonucleotide. All antisense experiments involved incubating cultures with oligonucleotide initially in serum-free medium and subsequently in the presence of heat-inactivated FBS (see below). 32P–end-labeled oligonucleotides were incubated at 37°C, and aliquots of the medium were removed at designated intervals. Samples were then concentrated by lyophilization, and aliquots of equal radioactivity were electrophoresed on a denaturing (8 mol/L urea) 19% polyacrylamide gel with subsequent autoradiography. Quantification of band density was performed by densitometry (LKB Ultrascan XL).
To assess intracellular incorporation of oligonucleotides, cell monolayers were washed extensively with PBS, trypsinized, pelleted, and further washed with an acid-salt solution (1 mol/L NaCl, 0.4 mol/L sodium acetate, pH 2.3). This procedure removes both trypsin-sensitive and trypsin-insensitive cell surface binding of oligonucleotide.26 Cells were then lysed with 1% SDS, and volumes equivalent to 1×105 cells were electrophoresed as described above.
DNA Synthesis Assays
Oligonucleotides were added to cultured cells growing in serum-free media (Endothelial SFM, GIBCO/BRL) supplemented with 10 ng/mL basic fibroblast growth factor (GIBCO/BRL). After 3 hours, heat-inactivated (56°C for 30 minutes) FBS was added directly to the culture well to a final concentration of 10%. After 72 hours, cultures were washed and incubated for 4 hours with [3H]thymidine (6 μCi/mL, 6.7 Ci/mmol, New England Nuclear). Incorporation of labeled thymidine into cell nuclei was assessed by autoradiography of acetone-fixed cells by use of previously described methods.24 The DNA synthesis rate was expressed as the labeling index, denoting the proportion of VSMCs with labeled nuclei (>5 silver grains per nucleus).
Cells in 100-mm dishes were trypsinized, counted, and frozen in liquid nitrogen. The cell pellet was resuspended in ice-cold suspension buffer (0.1 mol/L NaCl, 0.01 mol/L Tris·Cl, pH 7.6, and 0.001 mol/L NaEDTA) containing leupeptin (10 μg/mL) and phenylmethylsulfonyl fluoride (100 μg/mL) and subsequently lysed in an equal volume of 2× SDS gel loading buffer (100 mmol/L Tris·Cl, pH 6.8, 4% SDS, 20% glycerol, 200 mmol/L dithiothriotol, and 0.2% bromphenol blue). Sample volumes equivalent to 5×105 cells were electrophoresed on a 12% polyacrylamide gel, and separated proteins were transferred to nitrocellulose (Biorad). The membrane was incubated overnight with blocking buffer (PBS, 5% nonfat milk, 0.05% Tween 20, and 0.02% sodium azide) and subsequently probed with monoclonal antibody to PCNA (PC10, 1:200 dilution, Signet Laboratories). Bound antibody was detected by autoradiography of the transfer membrane incubated with 125I protein A (5 μCi, New England Nuclear). To confirm equivalent loading of protein on each gel lane and to ensure specificity of an antisense effect, the membranes also were probed with a monoclonal antibody against smooth muscle α-actin (clone 1A4, Dako Corp) and a monoclonal antibody against the human EGF receptor (Upstate Biotechnology). The latter is an unrelated gene product with a half-life shorter than that of PCNA.
Uptake and Intracellular Distribution of S-Chimeric Oligonucleotides
Fluorescence microscopy was used to assess the proportion of cells that incorporate chimeric oligonucleotides and to evaluate their intracellular distribution. Oligonucleotides were synthesized with a biotin residue linked to the 3′ end through a 12-carbon linkage (Glenn Research). Cultured VSMCs were incubated with various concentrations of biotinylated S-chimeric oligonucleotide for 1, 8, and 24 hours; then they were washed extensively with PBS and fixed and permeabilized in acetone at −20°C. Cells were incubated for 45 minutes with streptavidin-FITC (2.5 μg/mL, GIBCO/BRL), washed 3 times with PBS, mounted with glycerol/PBS (9:1 vol:vol, pH 8.7) medium containing 2.5% diazabicylooctane (Sigma Chemical Co), and visualized by epifluorescence microscopy (Olympus B×50) and laser scanning confocal microscopy (Biorad MRC600). The confocal microscope was equipped with an argon ion laser, and apertures were adjusted to minimize autofluorescence. Cells incubated without oligonucleotide or with an equimolar concentration of free d-biotin (Sigma Chemical Co) were studied in an identical fashion and served as controls.
The influence of cationic liposomes on oligonucleotide uptake was evaluated by use of Lipofectamine (GIBCO/BRL). Cells were incubated in serum-free media with biotinylated oligonucleotide in the presence or absence of Lipofectamine. To ensure optimal conditions of lipofection, eight different oligonucleotide-liposome mixtures were tested, each in duplicate. The concentrations of oligonucleotide and Lipofectamine ranged from 0.2 to 5.0 μmol/L and 1.3 to 20 μg/mL, respectively. The cationic lipid component of Lipofectamine is DOSPA, and the range of liposome concentrations used is equivalent to a molar concentration of DOSPA of 0.5 to 10 μmol/L. Thus, the molar ratio of oligonucleotide to DOSPA in the eight mixtures used ranged from 1:1 to 1:50. This wide range encompasses the various compositions recommended by the manufacturer and that used by others.20 27 After incubation for 1 to 8 hours, cells were washed and processed for evaluation by confocal microscopy as described above.
Uptake of S-Chimeric Oligonucleotides Into Human Plaque Fragments
Nine fragments of atherosclerotic plaque were retrieved from coronary arteries of two patients undergoing percutaneous directional atherectomy. Fresh plaque fragments were placed in serum-free culture medium immediately after retrieval and subsequently incubated at 37°C for 1 hour with either biotinylated oligonucleotide (5 μg/mL) or an equal volume of saline. Fragments were washed three times with PBS and frozen in liquid nitrogen, and sections were cut onto glass slides. Sections were then incubated with streptavidin-FITC as described above. The bis-benzimide nuclear dye Hoechst 33258 (2.5 μg/mL, Sigma Chemical Co) was included in the mounting media to allow visualization of cell nuclei. Total cell number was ascertained by counting of fluorescent nuclei, and the proportion of cells that had incorporated oligonucleotide was determined.
Results are expressed as mean±SEM. Comparisons were made by Student’s t test or ANOVA with Sheffé’s post hoc test. Statistical significance was set at P=.05.
Stability of S-Chimeric Oligonucleotides in Biological Medium
Unmodified oligonucleotides rapidly degraded in culture containing 10% heat-inactivated serum, and no intact oligonucleotide was evident after 21 hours. In contrast, undegraded S-chimeric oligonucleotide was abundant after 21 hours and still detectable after 72 hours. The half-life in serum-supplemented medium was 30.5 hours as determined by exponential curve fitting of band densities (Fig 2A⇓). Undegraded S-chimeric oligonucleotide could be recovered from VSMC lysates after 24 and 36 hours (Fig 2B⇓), providing an indication of the stability of S-chimeric oligonucleotides in the intracellular environment.
Effect of S-Chimeric Oligonucleotides Complementary to PCNA mRNA on DNA Synthesis and PCNA Content
A trend of dose-dependent inhibition in thymidine incorporation for the control oligonucleotides was suggested (Fig 3⇓). At a 30-μmol/L concentration of the sense, scramble, and α1(I) collagen oligonucleotides, thymidine labeling was 79.1±3.1%, 82.7±4.4%, and 80.3±1.4%, respectively, of that of saline-treated cultures. However, a much greater effect was noted for the PCNA antisense oligonucleotide, in which case thymidine labeling fell to 53.7±9.2% of that of saline-treated cells (at 30 μmol/L). This represented a significantly greater effect compared with that of the three control oligonucleotides (P<.01).
Concomitant with the reduction in DNA synthesis induced by the antisense oligonucleotide was a specific reduction in total PCNA content as assessed by Western blot analysis (Fig 4⇓). PCNA content of VSMCs was quantified by densitometry, and the signal was expressed relative to that of smooth muscle α-actin. The ratio of PCNA to smooth muscle α-actin in VSMCs incubated with 30 μmol/L antisense oligonucleotide was 53.1±10.6% of that of saline-treated cultures and 55.8±7.3% of that of cells incubated with scramble control oligonucleotide (P<.05). In contrast to the reduction in PCNA level, there was no reduction in the level of the EGF receptor in VSMCs exposed to the PCNA antisense oligonucleotide (Fig 4⇓). The half-life of the EGF receptor protein is shorter than that of PCNA (≈4 hours for the EGF receptor28 versus 20 hours for PCNA29 ); thus, the stable level of EGF receptor protein confirms a specific effect of the oligonucleotide and does not reflect a relatively slow turnover rate of the control protein.
Uptake and Intracellular Fate of S-Chimeric Oligonucleotides
Fig 5⇓ illustrates the uptake of S-chimeric oligonucleotides into human VSMCs. Virtually all cells incorporated the oligonucleotide, and this was evident among cells incubated with as little as 0.2 μmol/L oligonucleotide for only 1 hour. A consistent pattern of both nuclear and cytoplasmic distribution was found (Fig 5A⇓ and 5B⇓). The nuclear signal was much stronger than the cytoplasmic signal, and this relationship did not vary appreciably with incubation times up to 24 hours. Confocal microscopy was used to discriminate among signals originating within the cytoplasm, the nucleus, and the surrounding membranes (Fig 5C⇓). This revealed a faint, relatively homogeneous cytoplasmic distribution with no evidence of compartmentalization into endosomes or other vesicles. The nuclear concentration was also diffuse but with a more heterogeneous distribution. Cells incubated with biotin without oligonucleotide did not show a specific signal, suggesting that the findings do not reflect an effect of the biotin label on the oligonucleotide (Fig 5A⇓). Similarly, studies with oligonucleotides in which biotin was linked to two internal nucleosides, both situated in the sulfur-modified region (nucleosides 3 and 16), showed a similar uptake pattern (data not shown) to that in Fig 5B⇓ and 5C⇓, further suggesting that the findings do not reflect localization of the biotin label degraded from the oligonucleotide.
We further considered the possibility that the fluorescence distribution may have been influenced by cell fixation. To assess this, oligonucleotides were synthesized with FITC directly conjugated to the 3′ end. Cells were incubated with the fluorescent oligonucleotide (5 μmol/L) for 1 hour and washed 3 times; then live, unfixed cells incubated in Hanks’ salt solution supplemented with 20 mmol/L HEPES were studied by confocal microscopy. Compared with the findings of the biotinylated oligonucleotides, the intensity of fluorescence was reduced. This probably reflects the lack of fluorescence enhancement that is afforded by the biotin-streptavidin-FITC complexing approach in which one molecule of biotin will bind three to four molecules of streptavidin-FITC. Nevertheless, both nuclear and cytoplasmic fluorescence was evident. When the same cells were then postfixed with acetone and revisualized, the fluorescence pattern did not change (data not shown). Thus, the finding of nuclear localization does not appear to be a function of cell fixation.
Influence of Liposomes
The pattern of fluorescence from cells incubated with biotinylated oligonucleotide in the presence of Lipofectamine was distinctly different from that of cells incubated with oligonucleotide alone. Nuclear fluorescence was still observed and had an intensity similar to that seen with oligonucleotide alone. However, there were also extremely bright punctated structures within the cytoplasm (Fig 5D⇑ through 5F). These structures were present regardless of the amount of liposome or oligonucleotide used. By phase-contrast microscopy, the structures appeared as phase-dense vesicles 2 to 4 μm in diameter. The cells themselves sometimes had ill-defined cell borders, most notably when the oligonucleotide concentration was >1.0 μmol/L and/or the Lipofectamine concentration was >15 μg/mL, suggesting a cytotoxic effect of the liposome-DNA complexes. Compared with VSMCs exposed to naked oligonucleotide, a nonspecific effect (ie, inhibition of both antisense and control oligonucleotide-treated cultures) on DNA synthesis assessed by autoradiography was more prominent. The magnitude of specific inhibition with the oligonucleotide-liposome mixtures studied was in no instance greater than that achieved with oligonucleotide alone (data not shown).
Oligonucleotide Uptake in Human Atherosclerotic Plaque
As Fig 6⇓ illustrates, incubation of freshly retrieved fragments of human atherosclerotic plaque resulted in specific association of oligonucleotide with the cells. All nine lesion fragments were relatively hypocellular (Fig 6A⇓). Fragments incubated with saline or d-biotin did not show cellular fluorescence (Fig 6B⇓). In contrast, fragments incubated with 5 μmol/L labeled S-chimeric oligonucleotide showed evidence of cell association in a majority of cells (Fig 6C⇓ and 6D⇓), and no significant signal was evident in the extracellular matrix. Quantitative assessment confirmed that the fragments were relatively hypocellular, with a total of 33±18 cells per fragment. Of the total cells, 88±16% were associated with oligonucleotide, and the oligonucleotide fluorescence signal localized primarily to the cell nucleus. This was confirmed by confocal microscopy. Optical sections (1 μm) imaged serially in the z axis revealed a diffuse, intranuclear distribution of fluorescence, with no evidence of oligonucleotide bound to the nuclear membrane. Immunostaining of adjacent tissue sections with anti–smooth muscle α-actin (Clone 1A4, Sigma Chemical Co) revealed that approximately half (51±16%) of the cells expressed this α-actin isoform.
Antisense oligodeoxynucleotide technology is a relatively recent approach to modifying gene expression and has generated enthusiasm for its potential therapeutic utility. The potential use of antisense oligonucleotides for therapy of vascular restenosis has received attention, with reports demonstrating inhibition of VSMC proliferation in vitro5 6 7 8 and after balloon injury to arteries in the rat9 10 11 and the pig.13 Recently, however, Villa and coworkers17 observed a high degree of nonspecific inhibition of VSMC proliferation using phosphorothioate oligonucleotides and did not observe in vivo inhibition of VSMC proliferation with oligonucleotides antisense to c-myb, as had been previously noted by others.5 These discrepant findings highlight, as previously emphasized by Epstein et al,15 that the “conceptual simplicity” of antisense approaches may belie the difficulties that needed to be overcome before they can be translated into reliable treatment tools for patients. In this context, it is noteworthy that the mechanism of antisense inhibition is incompletely understood and several criteria must be met before a reliable, potent, and specific inhibition of gene expression in patients can be accomplished with oligonucleotides. These include adequate oligonucleotide stability in biological fluids, adequate intracellular uptake, sequence specificity, and an intracellular processing path that enables interaction with the cellular target.16 Application to human disease may be further complicated by differences in oligonucleotide uptake between different cell types.18 Moreover, the influence that the complex, fibrous extracellular matrix of human atherosclerotic plaque has on the bioavailability of oligonucleotides is unknown. The present study therefore advances our understanding of antisense oligonucleotides as potential therapeutic agents by demonstrating that partially S-modified oligonucleotides are stable and can effect a partial, specific inhibition of gene expression in human VSMCs over a background of nonspecific changes. Oligonucleotides designed in this manner are rapidly transported to the nucleus of human VSMCs, indicating a potential influence at the nuclear level rather than with cytoplasmic mRNA. Adjunctive use of cationic liposomes markedly enhances cellular uptake of oligonucleotides but with extensive cytoplasmic compartmentalization, which probably limits their bioavailability. Finally, we established that the presence of a dense extracellular matrix does not preclude uptake of S-chimeric oligonucleotides into cells within human atherosclerotic plaque.
Stability and Efficacy of S-Chimeric Antisense Oligonucleotides in Human VSMCs
Ironically, advances in rendering oligonucleotides stable in biological fluids may have added new complexities by introducing further non–sequence-specific effects. Phosphorothioate oligonucleotides are relatively resistant to nucleolytic degradation and have been used widely. Compared with unmodified oligonucleotides, however, they appear to have a greater propensity for protein binding with consequent effect on cellular function.30 For example, phosphorothioate oligonucleotides have been shown to directly inhibit protein kinase C,31 DNA polymerase, and RNase H.21 Inhibition of protein kinase C or DNA polymerase can induce nonspecific effects, whereas inhibition of RNase H could lead to unpredictable therapeutic end points because activation of RNase H is one means by which antisense oligonucleotide inhibition of gene expression is thought to be mediated.
We observed that S-chimeric oligonucleotides, synthesized as a hybrid of phosphorothioate and unmodified phosphodiester oligonucleotides, were stable both extracellularly and intracellularly and that nonspecific effects were not prominent. The half-life in serum-supplemented media (30.5 hours) is similar to that reported for phosphorothioate oligonucleotides32 33 and suggests that stability is not appreciably compromised by modifying only the termini with sulfur.
S-chimeric oligonucleotides to PCNA led to a dose-dependent inhibition of DNA synthesis and PCNA content in human VSMCs. The effect of the antisense oligonucleotide was significantly greater than background nonspecific inhibition by the control oligonucleotides but apparently much less than the 100% inhibition of mitotic activity in NIH 3T3 cells observed by Jaskulski and coworkers,23 even when the identical oligonucleotide was used (data not shown). These variable findings suggest that differences in cell type may be important influences on the efficacy of antisense oligonucleotides. As noted below, cellular uptake of oligonucleotide into human VSMCs was relatively low, which may, in part, explain the lower efficacy.
Intracellular Fate of S-Chimeric Oligonucleotides in Human VSMCs
An understanding of the intracellular fate of oligonucleotides is critical to the design of optimal antisense strategies; however, descriptions of intracellular translocation patterns have varied. Some investigators have described a punctated pattern of cytoplasmic incorporation, which has been attributed to endosomal localization.26 34 35 Others, however, have observed a preferential localization in the nucleus.36 37 Moreover, marked differences in uptake patterns between continuous cell lines and cells in primary culture have been noted.18 Using confocal microscopy, we found that S-chimeric oligonucleotides entered both the cytoplasmic and nuclear compartments of human VSMCs, with preferential accumulation in the nucleus. This was observed in fixed cells after incubation with biotin-labeled oligonucleotide and in cells incubated with FITC-labeled oligonucleotide and visualized live without fixation. The concordant results from these two approaches appear to rule out any significant influence of the biotin molecule or the fixation process on the nuclear translocation process. It is also unlikely that the nuclear signal represents degraded labeled nucleotides because (1) the biotin (or FITC) molecule was linked to a base within the sulfur-modified, nuclease-resistant region; (2) the findings were evident within 1 hour of incubation, well within the observed period of oligonucleotide stability; and (3) no signal was present in cells incubated with d-biotin alone.
Rapid translocation of oligonucleotide to the nucleus has been a consistent finding in cells in which the cytoplasm is microinjected with oligonucleotide.38 39 The rapid nuclear incorporation process demonstrated by these investigators was diffusion mediated and would appear to represent the unimpeded flux of oligonucleotides not bound to or incorporated into cytoplasmic structures. Thus, the present studies suggest that a significant proportion of S-chimeric oligonucleotides entering human VSMCs are “translocatable” and not trapped in cytoplasmic structures. The finding of nuclear localization also raises the possibility that the nucleus could be exploited as a site of action for oligonucleotides. Previous studies have shown biological activity of oligonucleotides complementary to splice junctions40 and introns.41 VSMCs may therefore be particularly well suited for strategies targeting precursor mRNA or for DNA triplex-based approaches. It should be noted, however, that the heterogeneous pattern of nuclear incorporation that we and others18 38 have observed probably reflects oligonucleotide bound to nuclear proteins,39 and the strength of this binding may be critical in determining the antisense effect.
Influence of Liposomes on Subcellular Processing of S-Chimeric Oligonucleotides in Human VSMCs
Coincubation of VSMCs with oligonucleotide and Lipofectamine resulted in a distinctly different subcellular distribution of oligonucleotides. Specifically, an intense punctate fluorescence pattern in the cytoplasm was present, suggesting that considerably more oligonucleotide had entered the cell than with oligonucleotide alone but that it was associated with cytoplasmic vesicles. Thus, although the liposome-treated VSMCs become “loaded” with oligonucleotide, a large proportion of oligonucleotide may not be available to generate an antisense effect. This was further suggested by the observation that the intensity of nuclear fluorescence was not enhanced by lipofection and the lack of augmentation of a specific antisense effect when Lipofectamine was used. Although it is possible that other liposome formulations or DNA or liposome concentrations could yield a greater antisense effect, the current findings demonstrated that relative to oligonucleotide alone, a clear dissociation between the extent of oligonucleotide uptake and the degree of specific antisense effect became evident when liposomes were used. Given the high total cellular uptake afforded by the liposomes, strategies to promote efflux of oligonucleotides from endosomes, such as mutant adenovirus,42 in combination with cationic liposomes might lead to much greater efficacy.
Can Oligonucleotides Be Taken Up by Cells Within Human Fibrous Atherosclerotic Tissue?
An important step in evaluating the potential therapeutic value of an oligonucleotide-based approach for vascular disease is establishing whether or not oligonucleotides can in fact associate with cells within human atherosclerotic plaque. To date, a biological effect of oligonucleotides has been observed in vivo in the carotid artery of rats9 11 12 14 and the porcine coronary artery.13 The latter study documented the presence of oligonucleotide in the artery after intraluminal delivery, although the location within the vessel wall (cellular versus extracellular) was not determined. Morishita and coworkers11 recently demonstrated uptake of oligonucleotide into medial cells of the rat carotid artery using a liposome-virus complexing technique. All of these animal models differ, however, from the circumstances of balloon angioplasty in humans in that they do not have a substrate of advanced atherosclerotic plaque with dense fibrous material. Among patients referred for percutaneous revascularization, the native lesion is typically hypocellular with dense collagenous tissue.19 The current findings suggest that these features do not preclude cellular uptake of oligonucleotide. Cellular association of oligonucleotides was observed after only 1 hour of incubation of oligonucleotide with human plaque specimens, and confocal microscopy indicated that during this time the oligonucleotides had entered the cells and translocated to the nucleus. There was little evidence of accumulation of oligonucleotide in the extracellular space, and it is possible that a net negative charge of the matrix may favor the observed cellular uptake. To the best of our knowledge, this represents the first demonstration of incorporation of oligonucleotides into atherosclerotic tissue from humans.
Selected Abbreviations and Acronyms
|EGF||=||epidermal growth factor|
|PCNA||=||proliferating cell nuclear antigen|
|VSMC||=||vascular smooth muscle cell|
This work was supported by grants from the Medical Research Council of Canada (MRC) (MT-11715 [Dr Pickering]) and the NHLBI (HL-40518-05 and HL-02824 [Dr Isner]) and an MRC scholarship to Dr Pickering. We gratefully acknowledge Drs D. Almond and B. Foley, London, Canada, for providing the atherectomy samples.
- Received June 29, 1995.
- Revision received August 21, 1995.
- Accepted September 25, 1995.
- Copyright © 1996 by American Heart Association
Stein CA, Cohen JS. Oligonucleotides as inhibitors of gene expression. Cancer Res. 1988;48:2659-2668.
Reynolds T. First antisense drug trials planned in leukemia. J Natl Cancer Inst. 1992;84:288-290.
Simons M, Rosenberg RD. Antisense nonmuscle myosin heavy chain and c-myb oligonucleotides suppress smooth muscle cell proliferation in vitro. Circ Res. 1992;70:835-843.
Speir E, Epstein SE. Inhibition of smooth muscle cell proliferation by antisense oligodeoxynucleotide targeting by the messenger RNA encoding proliferating cell nuclear antigen. Circulation. 1992;86:538-547.
Shi Y, Hutchison HG, Hall DJ, Zalewski A. Downregulation of c-myc expression by antisense oligonucleotides inhibits proliferation of human smooth muscle cells. Circulation. 1993;88:1190-1195.
Biro S, Fu Y-M, Yu Z-X, Epstein S. Inhibitory effects of oligodeoxynucleotides targeting c-myc mRNA on smooth muscle cell proliferation and migration. Proc Natl Acad Sci U S A. 1993;90:654-658.
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.
Morishita R, Gibbons GH, Ellison KE, Nakajima M, von der Leyen H, Zhang L, Kaneda Y, Ogihara T, Dzau VJ. Intimal hyperplasia after vascular injury is inhibited by antisense cdk 2 kinase oligonucleotide. J Clin Invest. 1994;93:1458-1464.
Bennet MR, Anglin S, McEwan JR, Jagoe R, Newby AC, Evan GI. Inhibition of vascular smooth muscle cell proliferation in vitro and in vivo by c-myc antisense oligodeoxynucleotides. J Clin Invest. 1994;93:820-828.
Shi Y, Fard A, Galeo A, Hutchinson HG, Vermani P, Dodge GR, Hall DJ, Shaheen F, Zalewski A. Transcatheter delivery of c-myc antisense oligomers reduces neointimal formation in a porcine model of coronary artery balloon injury. Circulation. 1994;90:944-951.
Simons M, Edelman E, Rosenberg R. Antisense proliferating cell nuclear antigen oligonucleotides inhibit intimal hyperplasia in a rat carotid artery injury model. J Clin Invest. 1994;93:2351-2356.
Epstein SE, Speir E, Finkel TF. Do antisense approaches to the problem of restenosis make sense? Circulation. 1993;88:1351-1353.
Stein CA, Cheng Y-C. Antisense oligonucleotides as therapeutic agents: is the bullet really magical? Science. 1993;261:1004-1012.
Villa AE, Guzman LA, Poptic EP, Labhasetwar V, D’Souza S, Farrell CL, Plow EF, Levy RJ, DiCorleto PE, Topol EJ. Effects of antisense c-myb oligonucleotides on vascular smooth muscle cell proliferation and response to vessel wall injury. Circ Res. 1995;76:505-513.
Felgner PL, Gadak TR, Holm M, Roman R, Chan HW, Wenz M, Northrop JP, Ringold GM, Danielson M. Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure. Proc Natl Acad Sci U S A. 1987;84:7413-7417.
Gao W-C, Han F-S, Storm C, Egan W, Cheng Y-C. Phosphorothioate oligonucleotides are inhibitors of human DNA polymerase and RNase H: implications for antisense technology. Mol Pharmacol. 1992;42:223-229.
Shaw JP, Kent K, Bird J, Fishback J, Froehler B. Modified deoxyoligonucleotides stable to exonuclease degradation in serum. Nucleic Acids Res. 1991;19:747-750.
Jaskulski D, DeRiel J, Mercer E, Calabretta B, Baserga R. Inhibition of cellular proliferation by antisense oligodeoxynucleotides to PCNA cyclin. Science. 1988;240:1544-1546.
Pickering JG, Weir L, Jekanowski J, Kearney M, Isner JM. Proliferative activity in peripheral and coronary atherosclerotic plaque among patients undergoing percutaneous revascularization. J Clin Invest. 1993;91:1469-1480.
Shoji Y, Akhtar S, Periasamy A, Herman B, Juliano RJ. Mechanism of cellular uptake of modified oligodeoxynucleotides containing methylphosphonate linkages. Nucleic Acids Res. 1991;19:5543-5550.
Renfrew CA, Hubbard AL. Degradation of epidermal growth factor receptor in rat liver: membrane topology through the lysosomal pathway. J Biol Chem. 1991;266:21265-21273.
Bravo R, Macdonald-Bravo H. Existence of two populations of cyclin/proliferating cell nuclear antigen during the cell cycle: association with DNA replication sites. J Cell Biol. 1987;105:1549-1554.
Cazenave C, Stein CA, Loreau N, Thuong NT, Neckers LM, Subasinghe C, Hélene C, Cohen JS, Toulmé JJ. Comparative inhibition of rabbit globin mRNA translation by modified antisense oligodeoxynucleotides. Nucleic Acids Res. 1989;19:4255-4273.
Ho PTC, Bacon TA, Wickstrom E, Sartorelli AC. Efficacy of antisense phosphodiester and phosphorothioate oligodeoxyribonucleotides to the transferrin receptor correlates with serum stability and cellular uptake. J Cell Biol. 1989;107:329a. Abstract.
Loke SL, Stein CA, Zhang XH, Mori K, Nakanishi M, Subasinghe C, Cohen JS, Neckers LM. Characterization of oligonucleotide transport into living cells. Proc Natl Acad Sci U S A. 1989;86:3474-3478.
Yakubov LA, Deeva EA, Zarytova VF, Ivanova EM, Ryte AS, Yurchenko LV, Vlassov VV. Mechanism of oligonucleotide uptake by cells: involvement of specific receptors? Proc Natl Acad Sci U S A. 1989;86:6454-6458.
Boutorin AS, Guslova LV, Ivanova EM, Kobetz ND, Zarytova VF, Ryte AS, Yurchenko LV, Vlassov VV. Synthesis of alkylating oligonucleotide derivatives containing cholesterol or phenazium residues at their 3′ terminus and their interaction with DNA within mammalian cells. FEBS Lett. 1989;254:129-132.
Goodchild J, Lestsinger RL, Sarin PS, Zamecnik M, Zamecnik PC, eds. Human Retroviruses, Cancer and AIDS: Approaches to Prevention and Therapy. New York, NY: Alan R Liss Inc; 1988:423-438.
Leonetti JP, Mechti N, Degols G, Gagnor C, Lebleu B. Intracellular distribution of microinjected antisense oligonucleotides. Proc Natl Acad Sci U S A.. 1991;88:2702-2706.
Kulka M, Smith CC, Aurelian L, Fishelevich R, Meade K, Miller P, Ts’o POP. Site specificity of the inhibitory effects of oligo(nucleoside methylphosphonate)s complementary to the acceptor splice junction of herpes simplex virus 1 immediate early mRNA4. Proc Natl Acad Sci U S A. 1989;86:6868-6872.
McManaway ME, Neckers LM, Loke LS, Al-Naeser AA, Redner RI, Shirmizo BT, Goldschmidts WI, Huber BE, Bhatia K, Magrath IT. Tumor-specific inhibition of lymphoma growth by antisense oligodeoxynucleotide. Lancet. 1990;1:808-810.
Wagner EL, Zatloukal K, Cotten M, Kirlappos H, Mechtler K, Curiel DT, Birnsteil ML. Coupling of adenovirus to transferrin-polylysine/DNA complexes greatly enhances receptor-mediated gene delivery and expression of transfected genes. Proc Natl Acad Sci U S A. 1992;89:6099-6103.