© 1997 by European Society of Cardiology
Copyright © 1997, European Society of Cardiology
Vectors based on Semliki Forest virus for rapid and efficient gene transfer into non-endothelial cardiovascular cells: comparison to adenovirus
aDepartment of Clinical Pharmacology, University of Groningen, Groningen, The Netherlands
bDepartment of Clinical Pharmacology, Free University of Berlin, Hindenburgdamm 30, 12200 Berlin, Germany
cDepartment of Physiological Chemistry, University of Groningen, Groningen, The Netherlands
dDepartment of Pharmacology, University of Hamburg, Hamburg, Germany
* Corresponding author. Tel. +49-30-8445-2279; Fax: +49-30-84454482.
Received 18 February 1997; accepted 10 June 1997
 |
Abstract |
|---|
 Top Abstract 1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
Objective: Replication-deficient, recombinant adenovirus is used as a carrier for gene transfer, but it is unspecific and the onset of transgene expression is relatively late. Here, we evaluated the efficiency and selectivity of gene transfer mediated by recombinant Semliki Forest virus (SFV). Methods: We compared the efficiency of a SFV-based vector with an adenoviral vector, using LacZ as a reporter gene. Firstly, the affinity for vascular smooth muscle cells, endothelial cells and cardiac myocytes was assessed. Secondly, we compared the time course of LacZ expression and cytotoxicity in vascular smooth muscle cells. Results: The SFV-based vector infects vascular smooth muscle cells and cardiomyocytes as efficiently as adenovirus. In contrast to adenovirus, SFV hardly transfers LacZ to endothelial cells (2.6% or less). SFV-mediated expression was visible after 1 h, reaching a maximum after 6 h. In contrast, adenovirus-mediated expression became visible after 6 h, and reached a maximum after 48–72 h. Both vectors were cytotoxic. Conclusions: We demonstrate that SFV efficiently transfers LacZ to vascular smooth muscle cells and cardiomyocytes, but not to endothelial cells. In contrast, adenovirus causes efficient transgene expression in all cell types tested. Furthermore, SFV-mediated expression is faster than adenovirus-mediated expression. Therefore, SFV-mediated gene transfer may be a suitable alternative to adenovirus, providing a fast expression in non-endothelial cardiovascular cell types.
KEYWORDS Gene transfer; Cell specific; Vascular smooth muscle cell; Endothelial cell; Cardiac myocyte; Recombinant Semliki Forest virus; Recombinant adenovirus; Cell culture
|
1 Introduction |
|---|
|
TopAbstract1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
Somatic gene transfer is currently being evaluated as a treatment modality for various diseases. In cardiovascular disorders, viral gene transfer methods are most often targeted to inhibit restenosis after balloon angioplasty and atherosclerotic plaque formation [1]. Several viral vectors have been shown to be able to reduce smooth muscle proliferation both in vitro and in vivo [2].
The success of gene therapy will largely depend on efficiency of gene transfer, timing and amount of gene expression and specificity for cell type. Replication-deficient adenovirus is a widely used carrier for somatic gene transfer. It is more efficient than retroviral and non-viral vectors, and is regarded to be an effective system for gene transfer. Its drawbacks are lack of specificity for cell type, immunogenicity in the host, and late onset of transgene expression.
We suggest that an efficient approach to gene therapy of restenosis should be based on a viral vector that specifically infects smooth muscle cells and that expresses its genes early. In this study, we have evaluated a vector based on the Semliki Forest virus (SFV) replicon, carrying the Escherichia coli LacZ gene (pSFV3-LacZ). The genome of SFV consists of a positive RNA strand which can be translated directly after infection of the host cell. The immediately formed, non-structural, proteins include an RNA replicase, which mediates production of a high amount of total genomic and subgenomic RNA copies. As the subgenomic copies code for the structural proteins of SFV, new, infectious virus particles will be rapidly formed [3, 4]. In pSFV3-LacZ the region coding for the structural viral proteins is replaced with the LacZ gene. Therefore, pSFV3-LacZ is a highly efficient, replication-deficient vector with a known rapid maximal expression of the β-galactosidase reporter enzyme [3].
pSFV3-LacZ is produced by co-transfection of RNA derived from recombinant vector with a helper RNA encoding the structural proteins [3]. Two packaging vectors called Helper1 and Helper2 have been designed [5]. Helper1-based virus stocks are directly active but may contain some wild-type virus which can cause infection in the host [5]. Helper2-based virus stocks have to be activated in vitro with chymotrypsin and contain hardly any active wild-type virus so that host infection is avoided [5].
Gene transfer with Helper1- and Helper2-based pSFV3-LacZ was compared to gene transfer with an established vector based on adenovirus (Ad5LacZ) with respect to affinity for vascular smooth muscle cells (VSMC) and endothelial cells (EC). Additionally, the timing of expression of Helper1-based vector and the cytotoxicity of Helper2-based vector were compared to Ad5LacZ in VSMC. Finally, pSFV3-LacZ-mediated gene transfer in rat cardiomyocytes was assessed.
|
2 Methods |
|---|
|
TopAbstract1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
2.1 Isolation and culture of vascular smooth muscle cells
A7r5 rat embryonic aortic smooth muscle cell lines (ATCC, CRL-1444, Rockville, USA) were a kind gift from Dr. H. de Smedt (Laboratory for Physiology, Catholic University of Leuven, Belgium) and were grown in Dulbecco's Modified Eagle's Medium (DMEM) (ICN, Zoetermeer, The Netherlands) supplemented with 50 units/ml penicillin/streptomycin (Life Technologies, Breda, The Netherlands), 10 mM HEPES (Life Technologies) and 10% Fetal Calf Serum (FCS) (Life Technologies). HA-VSMC human aortic smooth muscle cell lines (ATCC; CRL-1999) were grown on nutrient mix F12 Kaighn's modification (Life Technologies) supplemented with 2 mM L-glutamine, 10 mM HEPES, 1xITS-X supplement (Life Technologies), 10 mM TES (Sigma, Deisenhofen, Germany), 50 µg/ml ascorbic acid (Sigma) and 30 µg/ml endothelial cell growth supplement (ICN). Primary cultured human umbilical cord vascular smooth muscle cells (hVSMC) were isolated and cultured as described before [6]. All cells were maintained in culture at 37°C and 5% CO2.
2.2 Isolation and culture of endothelial cells
Bovine aortic endothelial cells (BAEC) were isolated and cultured as previously described [7]. The human endothelial cell line ECV 304 (ECACC, Salisbury, UK) was cultured in M 199 containing 10% FCS. The human endothelial cell line EA.hy 926 (a kind gift from Dr. C.-J. Edgell, University of North Carolina, Durham, NC, USA to M. Paul) was grown in DMEM supplemented with 10% FCS. EC-RF24 [8](a kind gift from Prof. Dr. H. Pannekoek, Amsterdam Medical Centre, Amsterdam, The Netherlands to Y. Pinto) were grown on a 1:1 mixture of M 199 and RPMI 1640 (Life Technologies) supplemented with 2 mM L-glutamin, 50 units/ml penicillin/streptomycin and 20% heat-inactivated human serum. Human umbilical vein endothelial cells (HUVEC) (a kind gift of Dr. V.J.J. Bom, Dpt. of Haematology, Acadamic Hospital Groningen, The Netherlands) were isolated according to a modified protocol of Jaffe et al. [6]as described elsewhere [9]. Endothelial cells were grown on collagen A-coated (Biochrom) plastic culture wells or flasks. All cells were kept in culture at 37°C and 5% CO2.
2.3 Isolation and culture of cardiac myocytes
Cardiac myocytes (rCM) were isolated from neonatal rats and cultured as described elsewhere [10]on collagen-coated plastic slides (Nunc). After 48 h, the cells were rinsed with serum free medium and exposed to virus.
2.4 Virus production
Ad5LacZ viral stocks were produced and quantified using HEK 293 cells (Microbix Biosystems Inc., Toronto, Canada). The stock solution was determined to contain 107 pfu·ml–1. pSFV3-LacZ was produced using pSFV-Helper1, to obtain pSFV3-LacZ/h1, or Helper2, to obtain pSFV3-LacZ/h2, and quantified using BHK21 cells as described previously [3]. Titers were between 107 and 108 infectious units per ml. In all cases, the virus stocks that were used originated from the same virus batch.
2.5 Efficiency and timing of expression of pSFV3-LacZ/h1 and Ad5LacZ
24-wells clusters (Nunc, Roskilde, Denmark) were seeded with 25.000 VSMC or 50.000 EC per well. VSMC and EC were allowed to regain morphology for 24 and 48 h, respectively. After this, the growth medium was taken off, the cells were washed with 1x phosphate buffered saline (PBS) (Biochrom) and the virus was added, as described below. The virus was allowed to bind for 1 h, unless indicated otherwise. After this, the cells were washed with PBS and growth medium was added. In every experiment, negative controls were included, i.e. cells that were not exposed to the virus but to serum-free medium.
In order to estimate the number of infectious units, dilution ranges (10–1010 fold dilutions) of virus stocks were prepared using serum-free DMEM. 200 µl of virus dilutions were added to separate wells containing VSMC (both pSFV3-LacZ/h1 and Ad5LacZ) and EC (pSFV3-LacZ/h1). After binding of the virus, the cells were cultured for 24 h (pSFV-LacZ) or 72 h (Ad5LacZ).
pSFV was compared to Ad5LacZ in subconfluent cultures of EA.hy 926 cells in a six-wells cluster, using 400 µl of each virus stock per well and allowing 48 h of viral gene expression, at 37°C, 5% CO2. During viral gene expression, cells were kept in growth medium.
To estimate the time course of pSFV3-LacZ/h1 and Ad5LacZ expression, 12.500 A7r5 cells·well–1 were seeded in 48-wells clusters. 200 µl of virus dilution was added per well. After incubation with the virus, the cells were cultured for 1, 3, 6, 12, 24, 48 and 72 h, respectively. The procedures were repeated in a separate experiment for HA-VSMC.
To investigate the possibility of gene transfer to cardiovascular cells other than VSMC, neonatal cardiac myocytes were cultured. These cardiac myocytes were incubated with 50 µl of pSFV3-LacZ/h1 stock solution. After 14 h, 200 µl of growth medium was added, and 22 h after addition of the virus, cells were fixed and analysed as described below.
2.6 Efficiency and cytotoxicity of pSFV3-LacZ/h2 and Ad5LacZ
A7r5 and HUVEC were seeded in 24-wells clusters and treated with virus as described above. At 1, 2, 3, 6 and 12 days after exposure to the virus, the culture medium was collected, the cells were detached from the wells by trypsinisation and resuspended in the collected medium. At each time point, the number of viable cells was counted in suspensions obtained from one well per treatment by Trypan Blue exclusion in a Bürker counting chamber. Thus, each value represents the mean (±SEM) obtained from 6 samples from one well. Additionally, the effect of pSFV3-LacZ/h2 on survival of HUVECs after 12 days was assessed. Furthermore, affinity of pSFV3-LacZ/h2 and Ad5LacZ for A7r5 and HUVEC were compared.
2.7 Histological procedures
After the expression times indicated above, the cells were washed with PBS and fixed with 1,25% glutaric aldehyde (Sigma) for 5 min at room temperature. Thereafter, the cells were washed with PBS and incubated with X-Gal staining fluid for 1–4 h at 37°C. X-Gal staining fluid contained 1 mg·ml–1 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-Gal) (Eurogentec, Seraing, Belgium), 5 mM K3Fe(CN)6 (Merck, Amsterdam, The Netherlands), 5 mM K4Fe(CN)6 (Merck), 2mM MgCl2.6H2O (Merck) and 4% N,N-dimethylformamide (Sigma). Finally, the cells were washed with PBS, post-fixed with 1,25% glutaric aldehyde for 2 min and either analysed immediately or stored under 90% glycerol (Jansse, Beersse, Belgium) until analysis.
2.8 Comparison of virus titers and efficiency of gene transfer
To quantify the efficiency of gene transfer, the percentage of LacZ-positive cells was determined. After the staining procedure, treated and controled cultures were analysed in situ under a light microscope at random locations (magnifications: 63x and 200x). If necessary, photographs were taken from random locations within the separate wells and used for analysis. Each value represents the mean (±SEM) of five random samples in a single well. Data from a single experiment that were obtained from both direct observation under a light microscope and photographical analysis were similar (data not shown).
To estimate virus titers, the dilution step just before the one resulting in no LacZ expression was used. The number of positive cells in the wells which received this virus dilution were counted. The titers were calculated as Np·1/DF·5 and expressed in infectious units per ml (IU·ml–1), where Np=number of positive cells, DF=dilution factor. One IU corresponds to the amount of virus capable of transferring LacZ to a single cell.
2.9 Statistical procedures
All means were compared with t-tests for independent samples. Standard errors of the means are indicated in the graphs with error bars. The viability curves for A7r5 were analysed and compared with ANOVA for repeated measures.
|
3 Results |
|---|
|
TopAbstract1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
3.1 Efficiency of viral gene transfer into vascular smooth muscle cells
Table 1 shows virus titers determined in VSMC. The results show that the titer depended on the cell type and that there is no clear difference between both viruses. The comparison of maximal percentages of LacZ-positive VSMC is shown in Fig. 1. Both viruses were equally able to transfer LacZ to all three types of VSMC. After gene transfer with Ad5LacZ the percentage of LacZ-positive HA-VSMC is significantly higher than that obtained with pSFV3-LacZ/h1 (p<0.05). Negative controls did not contain blue stained cells.
|
|
3.2 Efficiency of viral gene transfer into endothelial cells
The number of LacZ-positive BAEC, ECV 304 and EA.hy 926 cells after gene transfer with pSFV3-LacZ/h1 never exceeded 2.6%, while HA-VSMC, as a positive control, always showed a high efficiency of gene transfer (Fig. 2). This is also reflected by the titration experiment shown in Table 2. As it is known that a small percentage of primary cultured BAEC are of other origin (e.g. smooth muscle cells or fibroblasts), this could result in a higher titer for BAEC as compared to the other EC. Still, the maximal percentage of positive BAEC is not higher than that of EA.hy 926. In a separate experiment EC-RF24 was tested yielding similar results (Fig. 2). In contrast to pSFV3-LacZ/h1, Ad5LacZ transferred LacZ into EA.hy 926 with moderate efficiency (pSFV3-LacZ/h1, 1.1%±0.2 vs. Ad5LacZ, 16.7%±3.1. p = 0.007) (Fig. 3). Comparison of pSFV3-LacZ/h2 with Ad5lacZ in HUVEC resulted in a similar observation (1.0%±0.4 vs. 33.6%±8.3). Medium-treated endothelial cells were negative.
|
|
|
3.3 Gene transfer into cardiac myocytes
The amount of LacZ-positive cardiac myocytes after gene transfer with pSFV3-LacZ/h1 was 79%±1. The morphology of positive myocytes after 22 h of expression showed that cells were intact (results not shown).
3.4 Time course of viral expression
In a time course experiment with A7r5 (Fig. 4), the first positive cells were seen 1 h after incubation with pSFV-LacZ/h1 (31%) and 6 h after incubation with Ad5LacZ (<1%). In addition, there was an increasing intensity of LacZ staining in time. At each time point, the differences between efficiency of gene transfer by pSFV3-LacZ/h1 and Ad5LacZ were significant (p
0.05). Medium-treated cultures were negative.
|
In HA-VSMC, initial LacZ expression was seen after 6 h for pSFV3-LacZ/h1 (22.5%±7.8) and after 12 h for Ad5LacZ (34.0%±5.6). Maximal efficiency was reached after 24 h for pSFV3-LacZ/h1 (78.4%±6.8) and 48 h for Ad5LacZ (54.5%±5.2). Due to low background staining that occured after long exposure of HA-VSMC to LacZ staining solution at 1 h and 3 h, low expression could be masked. Therefore, we cannot exclude expression earlier than 6 h for pSFV3-LacZ and 12 h for Ad5LacZ.
3.5 Cytotoxicity of pSFV3-LacZ/h2 and Ad5LacZ
The effect of pSFV3-LacZ/h2 and Ad5LacZ on growth of A7r5 is shown in Fig. 5. Both viruses cause cell death and after 12 days the amount of A7r5 was too low to be counted. In HUVEC, however, the number of cells was not decreased 12 days after exposure to pSFV3-LacZ/h2 (3.6·104±3739 cells·well–1 at day 0 vs. 5.7·104±3867 cells·well–1 at day 12) or vehicle (4.4·104±2656 cells·well–1 at day 0 vs. 4.6·104±4110 cells·well–1 at day 12). The percentage transfection efficiency reached in A7r5 was higher than 80%. However, we observed that A7r5 surviving the period of 12 days were not expressing LacZ.
|
|
4 Discussion |
|---|
|
TopAbstract1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
In the present study we compared recombinant SFV to adenovirus with respect to its use for gene transfer into cultured cardiovascular cells. The results show that SFV efficiently transfers LacZ to VSMC and rCM. In contrast to Ad5LacZ, pSFV3-LacZ does not cause efficient gene transfer into EC. Therefore, pSFV3-LacZ can be considered as a vector with relative specificity for non-endothelial cells. Sensitivity of both EC and VSMC for adenoviral vectors in vitro and in vivo has been described extensively by others [11–13]and confirmed in the present study. Selective adenovirus-mediated gene transfer to EC has been shown to be feasible [14].
We show here that both pSFV3-LacZ and Ad5LacZ are cytotoxic to VSMC, whereas HUVEC survive exposure to pSFV3-LacZ. pSFV3-LacZ will cause inhibition of target cell protein synthesis, ultimately leading to apoptosis of the transfected cells [4, 15]. Recently, it has been shown that overexpression of the proto-oncogene bcl-2 allowed survival of rat prostatic adenocarcinoma cells after exposure to SFV [16]. Others have already shown that Ad5LacZ is cytotoxic to both EC [14]and VSMC [17]in vivo. It remains unknown what causes the insensitivity of endothelial cells for pSFV3-LacZ. Further studies with EC should address bcl-2 expression but also virus binding, internalization and disassembly of virus particles, and RNA translation.
The properties of SFV to transfer genes rapidly and efficiently into non-endothelial cells could be important for treatment of vascular lesions in which the endothelium has to be spared and where VSMC are the specific target, e.g. after balloon injury and in atherosclerotic plaques. Furthermore, it would be particularly suitable to block early gene expression, thereby preventing a cascade of molecular processes that lead to deleterious processes such as (re)stenosis. The proto-oncogene c-myc could represent a possible target, since expression of c-myc reaches its peak as early as 2 h after balloon injury [18].
Toxicity of a gene-therapy vector can become evident at the level of the entire organism. Production of recombinant SFV with pSFV-Helper1 will result in the formation of low amounts of replication-proficiant virus (RPV) which might cause an infection [5]. Therefore, a packaging vector called pSFV-Helper2 (helper 2) has been designed to produce particles with a mutated p62 spike protein [5]. The wild-type p62 is normally cleaved into its active, mature form E2 by a host endoprotease. Mutant p62, however, cannot be cleaved in vivo and has to be converted in E2 in vitro by chymotrypsin. Thus, conditionally infectious recombinant particles can be produced that do not contain RPV and do not cause pathological effects nor result in immunity in BALB/c mice [5]. Therefore, helper 2-based recombinant SFV may be a suitable vector for in vivo gene transfer into VSMC. pSFV3 recombinant particles have been used in vivo for induction of cytotoxic T lymphocyte activity against the influenza nucleoprotein [19].
Vectors derived from DNA viruses, such as adenovirus, could incorporate into the host genome, hypothetically leading to a life-long expression of the introduced gene. This is not the case with SFV, as this is a positive-strand RNA virus that replicates in the cytoplasm. We found that those cells that did survive exposure to Ad5LacZ were not expressing LacZ after 12 days of culture. Therefore, expression of LacZ in vivo long after exposure to adenovirus may actually represent cells that start expressing Ad5LacZ relatively late. In this case, loss of transgene expression can be caused by exhaustion of transfected cells through cell death. As SFV expresses its genes more rapidly than adenovirus, it can be expected that SFV-mediated expression will last shorter.
We demonstrate that the SFV vector provides a more rapid and more selective gene transfer than adenovirus, at comparable cytotoxicity levels. Obviously, these in vitro data await further elaboration in vivo.
Time for primary review 21 days.
|
Acknowledgements |
|---|
This study was supported by a grant from The Netherlands Organisation for Scientific Research (NWO) to Y. Pinto (# 950-10-642), a grant from the BMBF to M. Paul (# 01 GB 9503), Inex Pharmaceuticals Corporation (Vancouver, BC, Canada) (grant to J. Wilschut supporting F. Pries) and by a travel grant from the Groningen Institute for Drug Studies (GIDS). Furthermore, we thank Hendrik Buikema and Robert H. Henning of the Dept. of Clinical Pharmacology, University of Groningen for their suggestions.
|
References |
|---|
|
TopAbstract1 Introduction2 Methods3 Results4 DiscussionReferences |
|---|
- Lucas A, Liu L-Y, Macen J, et al. Virus-encoded serine proteinase inhibitor SERP-1 inhibits atherosclerotic plaque development after balloon angioplasty. Circulation 1996;94:2890–2900.
- Feldman LJ, Tahlil O, Steg, G. Perspectives of arterial gene therapy for the prevention of restenosis. Cardiovasc Res 32:194–207.
- Liljeström P, Garoff H. A new generation of animal cell expression vectors based on the Semliki Forest virus replicon. Bio/Technol (1991) 9:1356–1361.[CrossRef]
- Frolov I, Hoffman TA, Prágai BM, et al. Alphavirus-based expression vectors: strategies and applications. Proc Natl Acad Sci USA 1996;93:11371–11377.
- Berglund P, Sjöberg M, Garoff H, et al. Semliki Forest virus expression system: production of conditionally infectious recombinant particles. Bio/Technol 1993;11:916–920.
- Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest (1973) 52:2745–2756.[Web of Science][Medline]
- Paul M, Zintz M, Bocker W, Dyer M. Characterization and functional analysis of the rat endothelin-1 promoter. Hypertension (1995) 25:683–693.
[Abstract/Free Full Text] - Fontijn R, Hop C, Brinkman H-J, et al. Maintenance of vascular endothelial cell-specific properties after immortilization with an amphoteric replication-deficient retrovirus containing human papilloma virus 16 E6/E7 DNA. Exp Cell Res 1995;216:199–207.
- Mulder AB, Blom NR, Smit JW, Ruiters MHJ, Van Der Meer J, Halie MR, Bom VJJ. Basal tissue factor expression in endothelial cell cultures is caused by contaminating smooth muscle cells: Reduction by using chymotrypsin instead of collagenase. Thromb Res;80:399–411.
- Webster KA, Discher DJ, Bishopric NH. Induction and nuclear accumulation of fos and jun proto-oncogenes in hypoxic cardiac myocytes. J Biol Chem (1993) 268:16852–16858.
[Abstract/Free Full Text] - Rome JJ, Shayani V, Flugelman MY, et al. Anatomic barriers influence the distribution of in vivo gene transfer into the arterial wall: Modelling with microscopic tracer particles and verification with a recombinant adenoviral vector. Arterioscler Thromb 1994;14:148–161.
- Steg PG, Feldman LJ, Scoazec JY, et al. Arterial gene transfer to rabbit endothelial and smooth muscle cells using percutaneous delivery of an adenoviral vector. Circulation 1994;90:1648–1656.
- Zhou H, Zeng G, Zhou A, et al. Adenovirus- mediated gene transfer of vascular smooth muscle cells and endothelial cells in vitro. Chin Med J Engl 1995;108:493–496.
- Schulick AH, Dong G, Newman KD, Virmani R, Dichek DA. Endothelium-specific in vivo gene transfer. Circ Res (1995) 77:475–485.
[Abstract/Free Full Text] - Strauss JH, Strauss EG. The alphaviruses: gene expression, replication, and evolution. Microbiol Rev (1994) 58:491–562.
[Abstract/Free Full Text] - Scallan MF, Allsopp TE, Fazakerley JK. bcl-2 acts early to restrict Semliki Forest virus replication and delays virus-induced programmed cell death. J Virol (1997) 71:1583–1590.[Abstract]
- Schulick AH, Newman KD, Virmani R, Dichek DA. In vivo gene transfer into injured carotid arteries. Optimization and evaluation of acute toxicity. Circulation (1995) 91:2407–2414.
[Abstract/Free Full Text] - Bennett MR, Anglin S, McEwan JR, et al. Inhibition of vascular smooth muscle cell proliferation in vitro and in vivo by c-myc antisense oligodeoxynucleotides. J Clin Invest 1994;93:820–828.
- Zhou X, Berglund P, Zhao H, Liljeström P, Jondal M. Generation of cytotoxic and humoral immune responses by nonreplicative recombinant Semliki Forest virus. Proc Natl Acad Sci USA (1995) 92:3009–3013.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||