- Email: [email protected]
Myeloma Xenograft Destruction by a Nonviral Vector Delivering Oncolytic Infectious Nucleic Acid
original article
© The American Society of Gene & Cell Therapy
Myeloma Xenograft Destruction by a Nonviral Vector Delivering Oncolytic Infectious Nucleic Acid Elizabeth M Hadac1, Elizabeth J Kelly1 and Stephen J Russell1 Department of Molecular Medicine, Mayo Clinic, Rochester, Minnesota, USA
1
The feasibility of using a nonviral vector formulation to initiate an oncolytic viral infection has not been previously demonstrated. We therefore sought to determine whether infectious nucleic acid (INA) could be used in place of virus particles to initiate an oncolytic picornavirus infection in vivo. Infectious RNA encoding coxsackievirus A21 (CVA21) was transcribed from plasmid DNA using T7 polymerase. Within 48 hours of injecting this RNA into KAS6/1 myeloma xenografts, high titers of infectious CVA21 virions were detected in the bloodstream. Tumors regressed rapidly thereafter and mice developed signs of myositis. At euthanasia, CVA21 was recovered from regressing tumors and from skeletal muscles. Treatment outcomes were comparable following intratumoral injection of naked RNA or fully infectious CVA21 virus. Dose–response studies showed that an effective oncolytic infection could be established by intratumoral injection of 1 µg of infectious RNA. The oncolytic infection could also be initiated by intravenous injection of infectious RNA. Our study demonstrates that INA is a highly promising alternative drug formulation for oncolytic virotherapy. Received 12 October 2010; accepted 14 March 2011; published online 19 April 2011. doi:10.1038/mt.2011.68
Introduction Gene therapy using viral or nonviral vectors is a promising approach to cancer therapy but is currently beset with practical limitations. Viral gene transfer vectors are sequestered by phagocytic cells in the liver and spleen1–3 neutralized by complement proteins and/or antiviral antibodies,4–6 extravasate inefficiently from tumor blood vessels,7,8 and are difficult to manufacture.9,10 Nonviral vectors, by comparison, are easily manufactured, can be formulated in numerous ways and have superior pharmacokinetic properties.11–13 However, despite these significant advantages, the usefulness of nonviral vectors is greatly limited by the low levels of transduction they achieve. The idea of using infectious nucleic acid (INA) for cancer therapy to combine the favorable pharmacokinetic properties of nonviral vectors with the potent bystander killing activity of replication competent viruses was originally proposed as long ago as 1991.14 Potential advantages include the ease and simplicity of producing and purifying large quantities of clinical grade nucleic acid,
its compatibility with a variety of nonviral gene delivery platforms and its relative lack of immunogenicity compared to intact viral particles. But despite the obvious appeal of the approach, feasibility has not yet been demonstrated, and we are aware of only one published study addressing the subject. Thus, Seymour and colleagues15 used synthetic INA vectors for direct intratumoral delivery of a conditionally replicating oncolytic adenovirus genome but the injected tumors did not regress. Also, the adenovirus genomes did not generate functional viruses unless covalently linked to the virally encoded terminal protein. In contrast to conditionally replicating adenoviruses, picornaviruses can be rescued with high efficiency from RNA transcripts corresponding to the full length viral genome.16–18 Another attractive feature is that several oncolytic picornaviruses have been shown to cause complete regression of susceptible mouse tumors, even when administered at a very low dose because they spread rapidly from initially infected tumor cells. This has been demonstrated for poliovirus RIPO,19 seneca valley virus,20 bovine enterovirus,21 encephalomyocarditis virus,22 and coxsackievirus A21 (CVA21).23–25 Moreover, CVA21 is currently being evaluated in phase 1/2 clinical trials. We therefore sought to determine in the currently reported study whether infectious ribonucleic acid encoding the full length genome of CVA21 could be used to initiate a spreading intratumoral infection that would mediate regression of human myeloma xenografts in laboratory mice.
Results Preparation and characterization of infectious RNA CVA21 is a picornavirus with a single-stranded positive-sense RNA genome. In order to increase the efficiency of viral rescue from nucleic acid, we used the immediate precursor to viral protein production, single strand RNA corresponding to the CVA21 genome.25 Full length CVA21 run-off transcripts were generated by in vitro transcription of linearized plasmid DNA (pGEM-EZ CVA21) using T7 polymerase, and their integrity (i.e., absence of RNA degradation) was confirmed on RNA Flash gels (Figure 1a). Typical CVA21 cytopathology (cell rounding) was observed in H1-HeLa cells as early as 8 hours after transfecting them with 1 µg in vitro transcribed CVA21 RNA. Similar cytopathology was induced in fresh HeLa cell monolayers by exposing them to filtered supernatant from the RNA-transfected HeLa cells and virus recovery was further confirmed by serial passage of the supernatants
Correspondence: Stephen J Russell, Department of Molecular Medicine, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA. E-mail: [email protected] Molecular Therapy vol. 19 no. 6, 1041–1047 june 2011
1041
© The American Society of Gene & Cell Therapy
Infectious Nucleic Acids for Cancer Therapy
CVA21 transcript
b −7.4 kb
10
10 109 108 107 106 105 104 103 102 101 100
b 1.0
0.4
0.8 0.6
CVA21 virus
0.2
0
12
24
36
48
60
72
Figure 1 Characterization of coxsackievirus A21 (CVA21) RNA by electrophoresis and assessment of viral production. (a) RNA Flash Gel with molecular weight (kb) markers (left) and CVA21 transcript (right). (b) Time-course of virus production in H1-HeLa cells transfected with CVA21 RNA. TCID50, tissue culture infectious dose 50.
onto new monolayers of H1-HeLa cells. CVA21 virus titration studies demonstrate the increasing abundance of CVA21 progeny in the transfected H1-HeLa cell supernatant (Figure 1b). Full length sequencing of virus genomes extracted from the rescued viral particles confirmed their identity with CVA21.
Initiating CVA21 infection by intratumoral injection of infectious RNA We next sought to determine whether INA could be used to initiate an oncolytic virus infection in vivo. We25 and others23 have previously shown that CVA21 has potent oncolytic activity in preclinical models of myeloma. Immune deficient (SCID) mice bearing established subcutaneous KAS6/1 myeloma xenografts tumors were treated by intratumoral injection of 20 µg of the in vitro transcribed CVA21 RNA. Control animals were given CVA21 viral particles (106 TCID50) by the intratumoral route, or RNA that had been pretreated with RNAse A. As previously reported,25 within 7 days of virus injection, tumor regression was apparent in animals treated with CVA21 viral particles, while animals injected with control carrier had unencumbered tumor progression (Figure 2a,b). Animals treated with CVA21 RNA were also cured of established myeloma tumors (Figure 2c), though tumor regression was delayed ~1–3 days compared to virus-treated animals. By contrast, animals treated intratumorally with CVA21 RNA that had been exposed to RNase A exhibited no tumor regression, and had to be sacrificed due to tumor size (Figure 2d). RNase A does not harm the genomes of intact CVA21 virions and does not reduce their infectivity or integrity, but completely degrades the naked in vitro transcribed RNA destroying its activity. Hence this experiment proves that the RNA is the active agent that initiates the oncolytic infection. In addition to this proof, the RNA is not prepared by processing virus particles. It is transcribed in vitro from plasmid DNA, in a reaction mix that lacks the translational machinery required for generation of viral proteins. The RNA is then purified from the in vitro transcription reaction. Thus there cannot be any contaminating virions. Because of its susceptibility to nucleases and its short, variable half-life in body fluids, we were not able to perform meaningful biodistribution studies of the administered RNA. However, by overlaying homogenized tissues onto HeLa cell monolayers we
Opti-MEM
0.4
0.1
Hours postinfection
1042
a 0.5 0.3
Tumor volume (cm3)
RNA marker (kb)
Viral titer, TCID50
a
0.2
0.0
0.0 −5
0
5
10
−5
15
0
Days
c 0.5
15
10
15
0.4
0.3
0.3
0.2
0.2
0.1
0.1
0.0
10
d 0.5
CVA21 RNA
0.4
5 Days
CVA21 RNA + RNase
0.0 −5
0
5 Days
10
15
−5
0
5 Days
Figure 2 In vivo demonstration of tumor response to intratumorally injected coxsackievirus A21 (CVA21) RNA. Tumor volumes of SCID mice carrying subcutaneous Kas6/1 human myeloma xenografts were treated with (a) 106 TCID50 CVA21 virus (n = 3), (b) Opti-MEM control (n = 8), (c) 20 µg CVA21 RNA (n = 7) or (d) 20 µg CVA21 RNA treated with 60 µg RNaseA (n = 2). Based on the segmented model, there was no overall group effect (P = 0.16) or group effect prior to day 6 (P = 0.49). There was, however, a significant effect after day 6 (P < 0.0001) with reductions of tumor volume in the RNA (P < 0.0001) and virus (P < 0.0001) groups compared to the control group. There was no after day 6 difference in the RNA + Rnase group compared to the control group (P = 0.29). MEM, minimal essential medium.
did detect replication competent CVA21 in the tumor as early as 2 days after intratumoral administration, at which time point was no virus detectable in liver, spleen, lung, brain, muscle, or spinal cord (data not shown). Virus biodistribution studies were not conducted at later time-points since we previously reported them following CVA21 virus administration.25 Mouse cells are known to be resistant to CVA21 infection because they lack the main viral receptor (human intracellular adhesion molecule 1). We therefore performed additional experiments to determine the relative efficiencies with which mouse and human cells can convert the transfected CVA21 RNA to infectious virions. We transfected a mini cell panel of mouse and human tumor cell lines derived from hepatocellular carcinoma, lung cancer, rhabdomyosarcoma, and melanoma with CVA21 RNA. Viral recovery was assessed by harvesting and filtering the cell supernatants and overlaying them onto H1-HeLa cells. Virus was released into the supernatants of all the human cell lineages tested, but of the mouse lineages only the melanoma cells were positive (Table 1). These data give further support to the conclusion that, in our in vivo studies, the CVA21 is rescued from RNA and amplified primarily in the tumor xenografts. We have previously shown that CVA21 treatment causes highlevel viremia, as well as the development of a severe myositis that induces paralysis in SCID mice. We therefore collected blood, as well as tissue biopsies from virus and RNA treated mice in order to observe if virus was produced, and at what sites it was found. At time of euthanasia, animals treated with either CVA21 virus or CVA21 RNA had serum viremia in excess of 106 TCID50 per ml (Table 2), and in addition had virus present in biopsies from both tumor and skeletal muscle of the hind limbs. Creatine www.moleculartherapy.org vol. 19 no. 6 june 2011
© The American Society of Gene & Cell Therapy
Infectious Nucleic Acids for Cancer Therapy
Table 1 Percent killing of H1-HeLa cells by CVA21 released into supernatant of mouse or human tumor cells transfected with CVA21 RNA Mouse cell line
Percent cell death
Human cell line
a
b
c
d
Percent cell death
Hepa 1-6
−
Hep3B
++
LLC-1
−
A549
+
C2C12
−
TE671
+
B16
+
Mel 624
++
−, <30%; +, 30–75%; ++, >75%.
Table 2 Serum virema, creatine phosphokinase levels, and clinical indications in treated mice at time of euthanasia Treatment
RNA + RNase CVA21 RNA
CVA21 virus
Mouse Serum CPK ID level (U/l)
A 026
Serum TCID50
85
Tumor regression Paralysis
<3.16
−
−
A 776
70
<3.16
−
−
B 016
1,105
5.62 × 106
+
+
B 368
3,250
3.16 × 106
+
+
C 330
240
3.16 × 106
+
+
C 290
960
10
7
+
+
C 264
355
106
+
+
D 815
805
1.78 × 10
7
+
+
Discussion
D 035
400
5.62 × 10
6
+
+
We have shown that intratumoral injection of infectious RNA corresponding to the full length genome of CVA21 can mediate the complete regression of myeloma xenografts in a mouse model. After being internalized into host cells the injected RNA is able to initiate a productive CVA21 virus infection that spreads rapidly through the tumor, and spills into the bloodstream causing sustained viremia. Eventually the viremic animals succumb to generalized myositis, although this side effect is probably an artifact of the SCID mouse model since immunocompetent animals are known to be resistant to CVA21 myositis,26 experimental CVA21 infections in humans cause little more than a short lived upper respiratory tract infection,27 and only one human case of localized CVA21 myositis has been reported.28 It will be interesting also to test the approach in an immunocompetent syngeneic mouse model but to date we have not been able to identify a mouse tumor that responds to CVA21 oncolytic virotherapy. The idea that INA coding for an oncolytic virus could be used for cancer therapy was formally proposed by Sutton in a letter to the Lancet in 1991.14 But despite the obvious appeal of the approach, its feasibility has not previously been demonstrated. One reason for this apparent lack of progress may be that the viruses employed most widely as oncolytic agents do not lend themselves to the approach. Negative-strand RNA viruses which have been shown to be highly oncolytic both in preclinical and early stage clinical trials29,30 cannot be generated solely by transfection of nucleic acid. Concomitant expression of several viral proteins is required.31,32 Thus, it is not currently feasible to initiate oncolytic infections with vesicular stomatitis virus, measles virus, or newcastle disease virus by transfecting a single nucleic acid construct. The problems are even greater for oncolytic reoviruses which have a segmented double stranded RNA genome.33 Similarly, the double stranded DNA genomes of oncolytic HSV (~150 kb)34 and vaccinia viruses (~200 kb)35,36 are simply too large
Abbreviations: CPK, creatine phosphokinase; CVA21, coxsackievirus A21; TCID50, tissue culture infectious dose 50.
phosphokinase, a serum marker for myositis, was elevated to a similar degree in animals treated with either CVA21 virus or RNA (Table 2), and both groups exhibited myositis both clinically and histologically (Figure 3).
Determination of minimum dose of infectious RNA required to mediate tumor regression After confirming that delivery of 20 µg INA coding for CVA21 could indeed initiate a potent oncolytic infection, we next endeavored to determine if lower doses of in vitro transcribed CVA21 could also initiate a therapeutically effective oncolytic CVA21 infection. We therefore conducted a dose–response study in which seven groups of animals with established Kas6/1 human myeloma tumors of ~5 mm diameter were intratumorally injected with varying doses of naked RNA and tumor progression was monitored daily thereafter. Tumor-bearing animals were treated with doses of 1, 2, 4, 8, 16, or 32 µg of RNA intratumorally in carrier-free buffer, while control animals received buffer only (Figure 4a–g). At RNA doses of 4, 8, 16 ,or 32 µg, all tumors were rapidly controlled although the kinetics of regression were delayed in one of the animals treated at the lower dose of 4 µg. At the lowest doses of RNA (1 or 2 µg), CVA21-mediated tumor regression was observed in only one third of treated animals, and these nonresponding animals did not become viremic. As shown before, although CVA21 proves a curative treatment for established Kas6/1 tumors, all mice that efficiently converted CVA21 RNA to virus later developed severe myositis that induced paralysis and necessitated euthanasia, with kinetics that did not differ significantly across dose levels. (Figure 4h). Molecular Therapy vol. 19 no. 6 june 2011
Figure 3 Formalin-fixed paraffin-embedded tumor sections stained by hematoxylin and eosin. Microscopic images shows inflammation and necrosis in (a) coxsackievirus A21 (CVA21) virus and (b) CVA21 RNA treated tumors while the (c) Opti-MEM control and (d) CVA21RNA + RNaseA tumors are within normal limits. Bar = 0.1 mm. MEM, minimal essential medium.
1043
© The American Society of Gene & Cell Therapy
Infectious Nucleic Acids for Cancer Therapy
a
2.0
b
Control N=3
2.0
1.5
1.0
1.0
0.5
0.5 0.0
0.0 −7 0 7 14 21 28 35 42 49 56 63 70 77
c
1 µg N=6
1.5
Days 2.5
2 µg N=6
2.0
d
−7 0 7 14 21 28 35 42 49 56 63 70 77 Days
0.6
4 µg N=6
0.5 0.4
1.5
0.3 1.0
0.2 0.1
0.0
0.0 −7 0 7 14 21 28 35 42 49 56 63 70 77
−7 0 7 14 21 28 35 42 49 56 63 70 77
e
Days 0.3
Days 0.3 16 µg N=6
0.2
0.2
0.1
0.1
0.0
g
8 µg N=5
f
0.0 −7 0 7 14 21 28 35 42 49 56 63 70 77 Days
0.4
32 µg N=6
0.3 0.2 0.1 0.0
−7 0 7 14 21 28 35 42 49 56 63 70 77 Days
h 100 Percent survival
Tumor volume, cm3
0.5
−7 0 7 14 21 28 35 42 49 56 63 70 77 Days
Control 1 µg 2 µg 4 µg 8 µg 16 µg 32 µg
75 50 25 0 0
7 14 21 28 35 42 49 56 63 70 77 Time, days
Figure 4 In vivo dose–response of tumor regression following intratumoral treatment with coxsackievirus A21 (CVA21) RNA. Tumor volumes of SCID mice carrying subcutaneous Kas6/1 human myeloma xenografts were treated with (a) Opti-MEM control, (b) 1 µg, (c) 2 µg, (d) 4 µg, (e) 8 µg, (f) 16 µg, or (g) 32 µg CVA21 RNA. Mice with tumor progression (gray lines), mice with tumor regression (black lines). (h) Kaplan–Meier survival curves of control and RNA treated mice. Based on the segmented model, there was no overall dose effect (P = 0.19) or dose effect prior to day 10 (P = 0.20). There was, however, a significant dose effect after day 10 (P = 0.0008) showing increased dose was associated with decrease in tumor volume. This was confirmed when evaluating the after day 10 results in a separate model (P < 0.0001). MEM, minimal essential medium.
and their rescue efficiencies from INA too low to allow them to be used for this approach. Even oncolytic adenoviruses which have a smaller genome size of 35 kb could not be effectively exploited in this way because they are rescued only inefficiently from transfected DNA.15 There are, however, several oncolytic viruses with smaller DNA or positive-sense RNA genomes that can be rescued reasonably efficiently simply by transfecting cells with a nucleic acid construct. Examples include the H1 autonomously replicating parvovirus which has a single stranded DNA genome,37 Sindbis, an alphavirus with a positive-sense RNA genome,38 the recently developed oncolytic C-type retroviruses,39 and several oncolytic picornaviruses including poliovirus RIPO,19 seneca valley virus,20 and CVA21.23–25 After considering these possibilities, we concluded that the picornavirus CVA21 offered perhaps the highest chance of success because (i) RNA transcripts corresponding to its full length viral genome have very high specific 1044
infectivity25 and (ii) injection of as few as 1,000 infectious CVA21 viral particles was sufficient to mediate complete eradication of susceptible tumors in mice.24,25 The current study focused on the use of naked RNA as the therapeutic formulation. Advantages of RNA are that it provides an immediate precursor for the generation of progeny picornaviruses and does not need to gain access to the nucleus before it is decoded. But RNA is rapidly degraded by nucleases ubiquitously present in body fluids.40,41 Also, while short siRNAs have been shown to be compatible with liposomal vectors,42,43 longer messenger RNAs have not, and have yet to be tested for compatibility with many of the better developed nonviral gene delivery vehicles. In light of the poor stability of RNA, it was somewhat unexpected to see tumor responses after intravenous therapy with CVA21 RNA. While virus was recovered from both tumor and skeletal muscle of these mice upon euthanasia, it is still unclear www.moleculartherapy.org vol. 19 no. 6 june 2011
© The American Society of Gene & Cell Therapy
where the oncolytic infection initiated. One possibility is that the RNA may have been taken up and converted to virus in the liver or spleen, which are highly accessible to blood-borne agents, and that the released virus then trafficked to the tumor where it mediated tumor regression. In support of this hypothesis, our group has previously determined that cell lineages lacking CVA21’s natural cellular receptor, intracellular adhesion molecule 1,44 transfected with CVA21 RNA can become viral producer cells despite the fact that they are resistant to infection with CVA21 virus (data not shown). It is possible that DNA might provide a more stable and versatile material than RNA for incorporation into nonviral vector formulations, but we were not yet able to rescue CVA21 by plasmid transfection and, to our knowledge, the feasibility of CVA21 virus rescue by transfection of plasmid DNA has not been established by others. In the only paper describing rescue of infectious poliovirus from transfected plasmid DNA,45 the specific infectivity was extremely low. To facilitate a future comparison between infectious RNA and infectious DNA, we are currently evaluating the specific infectivities of DNA plasmids coding for a number of oncolytic picornaviruses. Regarding the question of relative doses required of INA versus intact virus particles to mediate tumor regression, 1 µg of RNA contains ~3 × 1011 viral genomes. The particle to infectivity ratio for CVA21 is unknown but, based on other viruses studied, a reasonable estimate is 100:1. The minimum effective dose of CVA21 administered intratumorally in this myeloma model is ~103 infectous units, containing an estimated 105 CVA21 genomes (data not shown). Hence in this feasibility study 106 copies of infectious RNA was therapeutically equivalent to one infectious virion. We anticipate that this number will be enhanced by optimizing the specific infectivity of the RNA. Compared to many other nonviral vectors, several of which have already entered clinical trials, naked CVA21 RNA is an extremely potent anticancer therapeutic. However, to maximize efficacy in surgically inaccessible tumors, as well as those that have metastasized, a cancer therapeutic would be formulated such that it could be delivered intravenously. Bolstered by our data that showed that as little as 1 µg of CVA21 RNA administered by intratumoral injection could efficiently initiate an oncolytic infection, we did perform preliminary studies to determine if intravenous administration of infectious RNA could elicit similar results as intratumoral RNA. We therefore delivered a high dose (50 µg) of unconjugated CVA21 RNA by intravenous injection in carrier-free buffer to Kas6/1 tumor-bearing SCID mice, and monitored tumor progression daily. Surprisingly, even unconjugated, naked CVA21 RNA was able to initiate an oncolytic infection that induced complete tumor regression (data not shown). However, compared to the intratumoral route of administration, much higher doses of RNA were required to realize anticancer activity after intravenous delivery. There is therefore a strong impetus to test the compatibility of CVA21 RNA with various lipid and nanoparticle vector formulations, both to target its delivery to sites of tumor growth and to enhance its resistance to circulating nucleases after intravenous administration. These experiments are currently underway. In conclusion, we have shown that infectious RNA encoding the full length genome of CVA21 can be used to initiate a Molecular Therapy vol. 19 no. 6 june 2011
Infectious Nucleic Acids for Cancer Therapy
spreading intratumoral infection that causes the rapid and complete regression of human myeloma xenografts in laboratory mice. This novel approach to cancer therapy combines the considerable strengths of nonviral gene therapy vectors with the complementary strengths of oncolytic viral vectors and is highly deserving of further development.
Materials and Methods Cell culture. H1-HeLa cells were obtained from American Type Culture Collection and were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum in 5% CO2. Kas6/1 cells were obtained from Dr Diane Jelinek, Department of Immunology, Mayo Clinic and maintained in RPMI + 10% fetal bovine serum + interleukin-6 at 37 °C in 5% CO2. Infectious RNA preparation, virus rescue, CVA21 sequencing. Infectious RNA was prepared as previously described25 by initial linearization of 1 µg pGEM-CVA21 plasmid with restriction endonuclease MluI (New England Biolabs, Ipswich, MA), followed by ethanol precipitation and resuspension in RNase free water. In vitro transcription reaction was then assembled using the Ambion MEGAscript T7 polymerase kit. RNA was purified thereafter using MEGAclear Kit (Ambion, Austin, TX) and in addition ethanol precipitated. This precipitate was pelleted, washed, and quantified on a spectrophotometer. To insure the transcript was the correct size and retained integrity, in vitro transcribed RNA was run on an RNA Flash Gel (Lonza, Basel, Switzerland). H1-HeLa cells were plated in a 12 well plate 24 hours prior to transfection. To obtain live virus, 1 µg CVA21 RNA was transfected into H1-HeLa cells using Mirus TransIT-mRNA Transfection Kit (Mirus Biosciences, Madison, WI). Forty-eight hours post-transfection, cleared lysate was collected and titrated on H1-HeLa cells. In order to assure that virus rescued from transfected RNA retained sequence integrity, we sequenced rescued virus from virus collected from the supernatant of cells transfected with CVA21 RNA. Briefly, virus was collected from cleared supernatant of transfected cells. Viral RNA was then extracted using a Viral RNA Extraction Kit (Qiagen, Valencia, CA). and cDNA prepared using Superscript one step RT-PCR Kit (Invitrogen, Carlsbad, CA) using the below primers. Overlapping fragments were then sent for sequencing.
Primer ID
Sequence
Nucleotide
CVA-F1
TTAAAACAGCTCTGGGGTTGTTCC
CVA-F2
TGAGTAGGTTGTCGTAATGCG
499–519
CVA-F3
AGCAGCCAATGCTATTGTTGC
1001–1021
CVA-F4
GCACCAGATCATCAACCTACG
1502–1522
CVA-F5
GCTTCTGATAAACGACTGAGC
2001–2021
CVA-F6
AGAGTGTCCCAACCACCCTCG
2499–2519
CVA-F7
TATGGAAATGCACCTCCACGG
3000–3020
CVA-F8
ACCATCGCCAGATGTAGTTGC
3501–3521
CVA-F9
ACTCGGGTGTGATATATCTCC
4007–4027
CVA-F10
TACTTCATTAATAGGTAGAGC
4502–4522
CVA-F11
TTCACTGTAGATGAGATTACC
4989–5009
CVA-F12
AGCAATATTGCCAACCCATGC
5504–5524
CVA-F13
ATCCAAGACAAAGTTAGAACC
6011–6031
CVA-F14
GCAATGAGAATGGCCTTTGGC
6507–6527
CVA-F15
TGGACTAACCATGACTCCAGC
7001–7021
CVA-F16
AATTTTCCTTTAATTTCGGAG
7386–7406
1–24
1045
© The American Society of Gene & Cell Therapy
Infectious Nucleic Acids for Cancer Therapy
CVA-F17
GCTCGTTGTGCCTATTAGC
CVA21-R1
GGGTGGGAACAACCCCAGAGC
7195–7213 29–9
CVA21-R2
GGGTTGTGAGCGGTTGCTCCG
496–476
CVA21-R3
GTGTAGTAATAGTCGAGTTGC
996–976
CVA21-R4
GTACACAAAGGCGTTTCCTGC
1499–1479
CVA21-R5
GGATAAACAGAATATAGGTGC
1991–1971
CVA21-R6
TTTATCGCTGTGTCAATGAGG
2488–2468
CVA21-R7
AGATGGAAGGGTTTGAAGAGG
2991–2971
CVA21-R8
CCCTGGGCTCTAGATTCCACC
3493–3473
CVA21-R9
CTAGTGTTGCAAGGATCGTGG
4002–3982
CVA21-R10
TATAGATTTACCCGTCCCTGG
4499–4479
CVA21-R11
CTACAGTGAACCTTTGTCTGG
4998–4978
CVA21-R12
ACATTATCATGCACCCCTAGC
5503–5483
CVA21-R13
GGGCATTTATAATGGGGTACC
6009–5989
CVA21-R14
TGAATCATTAAGGCTGGAAGC
6503–6483
CVA21-R15
TCTTTTCCTGATTGGGCTAGG
6997–6977
CVA21-R16
GAAATTAAAGGAAAATTTACC
7402–7382
Virus and RNA growth curves. H1-HeLa cells were infected with CVA21 at an multiplicity of infection of 1.0. Two hours after initial infection, cells were washed and media replaced. At this time, alternate cells were transfected in parallel with 1 µg CVA21 RNA in 24 well plates. Cell lysates from H1-HeLa cells were collected at specific times postinfection or post-transfection (12, 24, 36, 48, 60, 72 hours). At the completion of all time-points, lysates were thawed, and cell pellets were cleared from the samples by centrifugation providing a cleared cell lysate fraction. The cleared lysates were then titrated and virus titer was determined by the Spearman–Karber method. In vivo experiments. All animal protocols were reviewed and approved
by Mayo Clinic Institutional Care and Use Committee. CB17 ICR-SCID mice were obtained from Harlan Sprague Dawley (Indianapolis, IN). Mice were irradiated and implanted with 5 × 106 Kas 6/1 in the right flank. When tumors reached an average of 0.5 cm × 0.5 cm, tumors were treated with varying amounts of RNA, or 106 TCID50 CVA21 in 100 µl of OptiMEM. Intravenous experiments proceeded in the same fashion, but with Opti-MEM diluted RNA or virus being delivered via the tail vein. Tumor volume was measured using a hand held caliper and blood was collected by retro-orbital bleeds. Serum samples from retro-orbital bleeds were sent to Marshfield laboratories (Marshfield, WI) for creatine phosphokinase analysis.
Statistical analysis. Tumor volume was modeled on the log2 scale. Due
sharp change in the response curve in mice with tumor regression, a segmented repeated measures model was used to assess overall group or dose effects on tumor volume. Based on graphical observation, the breakpoint chosen for segmented models was 6 days for the experiment shown in Figure 2 and 10 days for the experiment shown in Figure 4. In addition to the segmented model, standard repeated measures models for both pre and postbreakpoint observations were fit to confirm the results from the segmented model.
ACKNOWLEDGMENTS This work was funded in part by R01CA129966 from NIH/NCI and Mayo Clinic.
REFERENCES
1. Bernt, KM, Ni, S, Gaggar, A, Li, ZY, Shayakhmetov, DM and Lieber, A (2003). The effect of sequestration by nontarget tissues on anti-tumor efficacy of systemically applied, conditionally replicating adenovirus vectors. Mol Ther 8: 746–755.
1046
2. Russell, SJ and Peng, KW (2007). Viruses as anticancer drugs. Trends Pharmacol Sci 28: 326–333. 3. Tao, N, Gao, GP, Parr, M, Johnston, J, Baradet, T, Wilson, JM et al. (2001). Sequestration of adenoviral vector by Kupffer cells leads to a nonlinear dose–response of transduction in liver. Mol Ther 3: 28–35. 4. Liu, C, Russell, SJ and Peng, KW (2010). Systemic therapy of disseminated myeloma in passively immunized mice using measles virus-infected cell carriers. Mol Ther 18: 1155–1164. 5. Mader, EK, Maeyama, Y, Lin, Y, Butler, GW, Russell, HM, Galanis, E et al. (2009). Mesenchymal stem cell carriers protect oncolytic measles viruses from antibody neutralization in an orthotopic ovarian cancer therapy model. Clin Cancer Res 15: 7246–7255. 6. Russell, SJ and Peng, KW (2008). The utility of cells as vehicles for oncolytic virus therapies. Curr Opin Mol Ther 10: 380–386. 7. Tseng, JC, Granot, T, DiGiacomo, V, Levin, B and Meruelo, D (2010). Enhanced specific delivery and targeting of oncolytic Sindbis viral vectors by modulating vascular leakiness in tumor. Cancer Gene Ther 17: 244–255. 8. Wojton, J and Kaur, B (2010). Impact of tumor microenvironment on oncolytic viral therapy. Cytokine Growth Factor Rev 21: 127–134. 9. Working, PK, Lin, A and Borellini, F (2005). Meeting product development challenges in manufacturing clinical grade oncolytic adenoviruses. Oncogene 24: 7792–7801. 10. Yamaguchi, T and Uchida, E (2007). Regulatory aspects of oncolytic virus products. Curr Cancer Drug Targets 7: 203–208. 11. Al-Dosari, MS and Gao, X (2009). Nonviral gene delivery: principle, limitations, and recent progress. AAPS J 11: 671–681. 12. Li, S and Ma, Z (2001). Nonviral gene therapy. Curr Gene Ther 1: 201–226. 13. Ohlfest, JR, Freese, AB and Largaespada, DA (2005). Nonviral vectors for cancer gene therapy: prospects for integrating vectors and combination therapies. Curr Gene Ther 5: 629–641. 14. Sutton, PM (1991). Treatment of cancer by infectious nucleic acid. Lancet 337: 1553. 15. Carlisle, RC, Briggs, SS, Hale, AB, Green, NK, Fisher, KD, Etrych, T et al. (2006). Use of synthetic vectors for neutralising antibody resistant delivery of replicating adenovirus DNA. Gene Ther 13: 1579–1586. 16. Agol, VI, Pilipenko, EV and Slobodskaya, OR (1996). Modification of translational control elements as a new approach to design of attenuated picornavirus strains. J Biotechnol 44: 119–128. 17. Girard, M, Altmeyer, R, van der Werf, S, Wychowski, C and Martin, A (1995). The use of picornaviruses as vectors for the engineering of live recombinant vaccines. Biologicals 23: 165–169. 18. Nomoto, A, Iizuka, N, Kohara, M and Arita, M (1988). Strategy for construction of live picornavirus vaccines. Vaccine 6: 134–137. 19. Ochiai, H, Campbell, SA, Archer, GE, Chewning, TA, Dragunsky, E, Ivanov, A et al. (2006). Targeted therapy for glioblastoma multiforme neoplastic meningitis with intrathecal delivery of an oncolytic recombinant poliovirus. Clin Cancer Res 12: 1349–1354. 20. Reddy, PS, Burroughs, KD, Hales, LM, Ganesh, S, Jones, BH, Idamakanti, N et al. (2007). Seneca Valley virus, a systemically deliverable oncolytic picornavirus, and the treatment of neuroendocrine cancers. J Natl Cancer Inst 99: 1623–1633. 21. Sedmak, GV, Taylor, MW, Mealey, J Jr and Chen, TT (1972). Oncolytic effect of bovine enterovirus on mouse and human tumours. Nature New Biol 238: 7–9. 22. Roos, FC, Roberts, AM, Hwang, II, Moriyama, EH, Evans, AJ, Sybingco, S et al. (2010). Oncolytic targeting of renal cell carcinoma via encephalomyocarditis virus. EMBO Mol Med 2: 275–288. 23. Au, GG, Lincz, LF, Enno, A and Shafren, DR (2007). Oncolytic Coxsackievirus A21 as a novel therapy for multiple myeloma. Br J Haematol 137: 133–141. 24. Au, GG, Lindberg, AM, Barry, RD and Shafren, DR (2005). Oncolysis of vascular malignant human melanoma tumors by Coxsackievirus A21. Int J Oncol 26: 1471–1476. 25. Kelly, EJ, Hadac, EM, Greiner, S and Russell, SJ (2008). Engineering microRNA responsiveness to decrease virus pathogenicity. Nat Med 14: 1278–1283. 26. Hyypiä, T, Kallajoki, M, Maaronen, M, Stanway, G, Kandolf, R, Auvinen, P et al. (1993). Pathogenetic differences between coxsackie A and B virus infections in newborn mice. Virus Res 27: 71–78. 27. Buckland, FE, Bynoe, ML and Tyrrell, DA (1965). Experiments on the spread of colds. II. Studies in volunteers with coxsackievirus A21. J Hyg (Lond) 63: 327–343. 28. Dekel, B, Yoeli, R, Shulman, L, Padeh, S and Passwell, JH (2002). Localized thigh swelling mimicking a neoplastic process: involvement of coxsackie virus type A21. Acta Paediatr 91: 357–359. 29. Liu, TC, Galanis, E and Kirn, D (2007). Clinical trial results with oncolytic virotherapy: a century of promise, a decade of progress. Nat Clin Pract Oncol 4: 101–117. 30. Russell, SJ and Peng, KW (2009). Measles virus for cancer therapy. Curr Top Microbiol Immunol 330: 213–241. 31. García-Sastre, A and Palese, P (1993). Genetic manipulation of negative-strand RNA virus genomes. Annu Rev Microbiol 47: 765–790. 32. Palese, P, Zheng, H, Engelhardt, OG, Pleschka, S and García-Sastre, A (1996). Negative-strand RNA viruses: genetic engineering and applications. Proc Natl Acad Sci USA 93: 11354–11358. 33. Van Den Wollenberg, DJ, Van Den Hengel, SK, Dautzenberg, IJ, Kranenburg, O and Hoeben, RC (2009). Modification of mammalian reoviruses for use as oncolytic agents. Expert Opin Biol Ther 9: 1509–1520. 34. Marconi, P, Argnani, R, Epstein, AL and Manservigi, R (2009). HSV as a vector in vaccine development and gene therapy. Adv Exp Med Biol 655: 118–144. 35. Guo, ZS and Bartlett, DL (2004). Vaccinia as a vector for gene delivery. Expert Opin Biol Ther 4: 901–917. 36. Thorne, SH, Hwang, TH and Kirn, DH (2005). Vaccinia virus and oncolytic virotherapy of cancer. Curr Opin Mol Ther 7: 359–365. 37. Ponnazhagan, S (2004). Parvovirus vectors for cancer gene therapy. Expert Opin Biol Ther 4: 53–64.
www.moleculartherapy.org vol. 19 no. 6 june 2011
© The American Society of Gene & Cell Therapy
38. Atkins, GJ, Smyth, JW, Fleeton, MN, Galbraith, SE and Sheahan, BJ (2004). Alphaviruses and their derived vectors as anti-tumor agents. Curr Cancer Drug Targets 4: 597–607. 39. Tai, CK and Kasahara, N (2008). Replication-competent retrovirus vectors for cancer gene therapy. Front Biosci 13: 3083–3095. 40. Alvarez-Salas, LM (2008). Nucleic acids as therapeutic agents. Curr Top Med Chem 8: 1379–1404. 41. Yamamoto, A, Kormann, M, Rosenecker, J and Rudolph, C (2009). Current prospects for mRNA gene delivery. Eur J Pharm Biopharm 71: 484–489.
Molecular Therapy vol. 19 no. 6 june 2011
Infectious Nucleic Acids for Cancer Therapy
42. Higuchi, Y, Kawakami, S and Hashida, M (2010). Strategies for in vivo delivery of siRNAs: recent progress. BioDrugs 24: 195–205. 43. Schroeder, A, Levins, CG, Cortez, C, Langer, R and Anderson, DG (2010). Lipid-based nanotherapeutics for siRNA delivery. J Intern Med 267: 9–21. 44. Shafren, DR, Dorahy, DJ, Greive, SJ, Burns, GF and Barry, RD (1997). Mouse cells expressing human intercellular adhesion molecule-1 are susceptible to infection by coxsackievirus A21. J Virol 71: 785–789. 45. Racaniello, VR and Baltimore, D (1981). Cloned poliovirus complementary DNA is infectious in mammalian cells. Science 214: 916–919.
1047
© The American Society of Gene & Cell Therapy
Myeloma Xenograft Destruction by a Nonviral Vector Delivering Oncolytic Infectious Nucleic Acid Elizabeth M Hadac1, Elizabeth J Kelly1 and Stephen J Russell1 Department of Molecular Medicine, Mayo Clinic, Rochester, Minnesota, USA
1
The feasibility of using a nonviral vector formulation to initiate an oncolytic viral infection has not been previously demonstrated. We therefore sought to determine whether infectious nucleic acid (INA) could be used in place of virus particles to initiate an oncolytic picornavirus infection in vivo. Infectious RNA encoding coxsackievirus A21 (CVA21) was transcribed from plasmid DNA using T7 polymerase. Within 48 hours of injecting this RNA into KAS6/1 myeloma xenografts, high titers of infectious CVA21 virions were detected in the bloodstream. Tumors regressed rapidly thereafter and mice developed signs of myositis. At euthanasia, CVA21 was recovered from regressing tumors and from skeletal muscles. Treatment outcomes were comparable following intratumoral injection of naked RNA or fully infectious CVA21 virus. Dose–response studies showed that an effective oncolytic infection could be established by intratumoral injection of 1 µg of infectious RNA. The oncolytic infection could also be initiated by intravenous injection of infectious RNA. Our study demonstrates that INA is a highly promising alternative drug formulation for oncolytic virotherapy. Received 12 October 2010; accepted 14 March 2011; published online 19 April 2011. doi:10.1038/mt.2011.68
Introduction Gene therapy using viral or nonviral vectors is a promising approach to cancer therapy but is currently beset with practical limitations. Viral gene transfer vectors are sequestered by phagocytic cells in the liver and spleen1–3 neutralized by complement proteins and/or antiviral antibodies,4–6 extravasate inefficiently from tumor blood vessels,7,8 and are difficult to manufacture.9,10 Nonviral vectors, by comparison, are easily manufactured, can be formulated in numerous ways and have superior pharmacokinetic properties.11–13 However, despite these significant advantages, the usefulness of nonviral vectors is greatly limited by the low levels of transduction they achieve. The idea of using infectious nucleic acid (INA) for cancer therapy to combine the favorable pharmacokinetic properties of nonviral vectors with the potent bystander killing activity of replication competent viruses was originally proposed as long ago as 1991.14 Potential advantages include the ease and simplicity of producing and purifying large quantities of clinical grade nucleic acid,
its compatibility with a variety of nonviral gene delivery platforms and its relative lack of immunogenicity compared to intact viral particles. But despite the obvious appeal of the approach, feasibility has not yet been demonstrated, and we are aware of only one published study addressing the subject. Thus, Seymour and colleagues15 used synthetic INA vectors for direct intratumoral delivery of a conditionally replicating oncolytic adenovirus genome but the injected tumors did not regress. Also, the adenovirus genomes did not generate functional viruses unless covalently linked to the virally encoded terminal protein. In contrast to conditionally replicating adenoviruses, picornaviruses can be rescued with high efficiency from RNA transcripts corresponding to the full length viral genome.16–18 Another attractive feature is that several oncolytic picornaviruses have been shown to cause complete regression of susceptible mouse tumors, even when administered at a very low dose because they spread rapidly from initially infected tumor cells. This has been demonstrated for poliovirus RIPO,19 seneca valley virus,20 bovine enterovirus,21 encephalomyocarditis virus,22 and coxsackievirus A21 (CVA21).23–25 Moreover, CVA21 is currently being evaluated in phase 1/2 clinical trials. We therefore sought to determine in the currently reported study whether infectious ribonucleic acid encoding the full length genome of CVA21 could be used to initiate a spreading intratumoral infection that would mediate regression of human myeloma xenografts in laboratory mice.
Results Preparation and characterization of infectious RNA CVA21 is a picornavirus with a single-stranded positive-sense RNA genome. In order to increase the efficiency of viral rescue from nucleic acid, we used the immediate precursor to viral protein production, single strand RNA corresponding to the CVA21 genome.25 Full length CVA21 run-off transcripts were generated by in vitro transcription of linearized plasmid DNA (pGEM-EZ CVA21) using T7 polymerase, and their integrity (i.e., absence of RNA degradation) was confirmed on RNA Flash gels (Figure 1a). Typical CVA21 cytopathology (cell rounding) was observed in H1-HeLa cells as early as 8 hours after transfecting them with 1 µg in vitro transcribed CVA21 RNA. Similar cytopathology was induced in fresh HeLa cell monolayers by exposing them to filtered supernatant from the RNA-transfected HeLa cells and virus recovery was further confirmed by serial passage of the supernatants
Correspondence: Stephen J Russell, Department of Molecular Medicine, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA. E-mail: [email protected] Molecular Therapy vol. 19 no. 6, 1041–1047 june 2011
1041
© The American Society of Gene & Cell Therapy
Infectious Nucleic Acids for Cancer Therapy
CVA21 transcript
b −7.4 kb
10
10 109 108 107 106 105 104 103 102 101 100
b 1.0
0.4
0.8 0.6
CVA21 virus
0.2
0
12
24
36
48
60
72
Figure 1 Characterization of coxsackievirus A21 (CVA21) RNA by electrophoresis and assessment of viral production. (a) RNA Flash Gel with molecular weight (kb) markers (left) and CVA21 transcript (right). (b) Time-course of virus production in H1-HeLa cells transfected with CVA21 RNA. TCID50, tissue culture infectious dose 50.
onto new monolayers of H1-HeLa cells. CVA21 virus titration studies demonstrate the increasing abundance of CVA21 progeny in the transfected H1-HeLa cell supernatant (Figure 1b). Full length sequencing of virus genomes extracted from the rescued viral particles confirmed their identity with CVA21.
Initiating CVA21 infection by intratumoral injection of infectious RNA We next sought to determine whether INA could be used to initiate an oncolytic virus infection in vivo. We25 and others23 have previously shown that CVA21 has potent oncolytic activity in preclinical models of myeloma. Immune deficient (SCID) mice bearing established subcutaneous KAS6/1 myeloma xenografts tumors were treated by intratumoral injection of 20 µg of the in vitro transcribed CVA21 RNA. Control animals were given CVA21 viral particles (106 TCID50) by the intratumoral route, or RNA that had been pretreated with RNAse A. As previously reported,25 within 7 days of virus injection, tumor regression was apparent in animals treated with CVA21 viral particles, while animals injected with control carrier had unencumbered tumor progression (Figure 2a,b). Animals treated with CVA21 RNA were also cured of established myeloma tumors (Figure 2c), though tumor regression was delayed ~1–3 days compared to virus-treated animals. By contrast, animals treated intratumorally with CVA21 RNA that had been exposed to RNase A exhibited no tumor regression, and had to be sacrificed due to tumor size (Figure 2d). RNase A does not harm the genomes of intact CVA21 virions and does not reduce their infectivity or integrity, but completely degrades the naked in vitro transcribed RNA destroying its activity. Hence this experiment proves that the RNA is the active agent that initiates the oncolytic infection. In addition to this proof, the RNA is not prepared by processing virus particles. It is transcribed in vitro from plasmid DNA, in a reaction mix that lacks the translational machinery required for generation of viral proteins. The RNA is then purified from the in vitro transcription reaction. Thus there cannot be any contaminating virions. Because of its susceptibility to nucleases and its short, variable half-life in body fluids, we were not able to perform meaningful biodistribution studies of the administered RNA. However, by overlaying homogenized tissues onto HeLa cell monolayers we
Opti-MEM
0.4
0.1
Hours postinfection
1042
a 0.5 0.3
Tumor volume (cm3)
RNA marker (kb)
Viral titer, TCID50
a
0.2
0.0
0.0 −5
0
5
10
−5
15
0
Days
c 0.5
15
10
15
0.4
0.3
0.3
0.2
0.2
0.1
0.1
0.0
10
d 0.5
CVA21 RNA
0.4
5 Days
CVA21 RNA + RNase
0.0 −5
0
5 Days
10
15
−5
0
5 Days
Figure 2 In vivo demonstration of tumor response to intratumorally injected coxsackievirus A21 (CVA21) RNA. Tumor volumes of SCID mice carrying subcutaneous Kas6/1 human myeloma xenografts were treated with (a) 106 TCID50 CVA21 virus (n = 3), (b) Opti-MEM control (n = 8), (c) 20 µg CVA21 RNA (n = 7) or (d) 20 µg CVA21 RNA treated with 60 µg RNaseA (n = 2). Based on the segmented model, there was no overall group effect (P = 0.16) or group effect prior to day 6 (P = 0.49). There was, however, a significant effect after day 6 (P < 0.0001) with reductions of tumor volume in the RNA (P < 0.0001) and virus (P < 0.0001) groups compared to the control group. There was no after day 6 difference in the RNA + Rnase group compared to the control group (P = 0.29). MEM, minimal essential medium.
did detect replication competent CVA21 in the tumor as early as 2 days after intratumoral administration, at which time point was no virus detectable in liver, spleen, lung, brain, muscle, or spinal cord (data not shown). Virus biodistribution studies were not conducted at later time-points since we previously reported them following CVA21 virus administration.25 Mouse cells are known to be resistant to CVA21 infection because they lack the main viral receptor (human intracellular adhesion molecule 1). We therefore performed additional experiments to determine the relative efficiencies with which mouse and human cells can convert the transfected CVA21 RNA to infectious virions. We transfected a mini cell panel of mouse and human tumor cell lines derived from hepatocellular carcinoma, lung cancer, rhabdomyosarcoma, and melanoma with CVA21 RNA. Viral recovery was assessed by harvesting and filtering the cell supernatants and overlaying them onto H1-HeLa cells. Virus was released into the supernatants of all the human cell lineages tested, but of the mouse lineages only the melanoma cells were positive (Table 1). These data give further support to the conclusion that, in our in vivo studies, the CVA21 is rescued from RNA and amplified primarily in the tumor xenografts. We have previously shown that CVA21 treatment causes highlevel viremia, as well as the development of a severe myositis that induces paralysis in SCID mice. We therefore collected blood, as well as tissue biopsies from virus and RNA treated mice in order to observe if virus was produced, and at what sites it was found. At time of euthanasia, animals treated with either CVA21 virus or CVA21 RNA had serum viremia in excess of 106 TCID50 per ml (Table 2), and in addition had virus present in biopsies from both tumor and skeletal muscle of the hind limbs. Creatine www.moleculartherapy.org vol. 19 no. 6 june 2011
© The American Society of Gene & Cell Therapy
Infectious Nucleic Acids for Cancer Therapy
Table 1 Percent killing of H1-HeLa cells by CVA21 released into supernatant of mouse or human tumor cells transfected with CVA21 RNA Mouse cell line
Percent cell death
Human cell line
a
b
c
d
Percent cell death
Hepa 1-6
−
Hep3B
++
LLC-1
−
A549
+
C2C12
−
TE671
+
B16
+
Mel 624
++
−, <30%; +, 30–75%; ++, >75%.
Table 2 Serum virema, creatine phosphokinase levels, and clinical indications in treated mice at time of euthanasia Treatment
RNA + RNase CVA21 RNA
CVA21 virus
Mouse Serum CPK ID level (U/l)
A 026
Serum TCID50
85
Tumor regression Paralysis
<3.16
−
−
A 776
70
<3.16
−
−
B 016
1,105
5.62 × 106
+
+
B 368
3,250
3.16 × 106
+
+
C 330
240
3.16 × 106
+
+
C 290
960
10
7
+
+
C 264
355
106
+
+
D 815
805
1.78 × 10
7
+
+
Discussion
D 035
400
5.62 × 10
6
+
+
We have shown that intratumoral injection of infectious RNA corresponding to the full length genome of CVA21 can mediate the complete regression of myeloma xenografts in a mouse model. After being internalized into host cells the injected RNA is able to initiate a productive CVA21 virus infection that spreads rapidly through the tumor, and spills into the bloodstream causing sustained viremia. Eventually the viremic animals succumb to generalized myositis, although this side effect is probably an artifact of the SCID mouse model since immunocompetent animals are known to be resistant to CVA21 myositis,26 experimental CVA21 infections in humans cause little more than a short lived upper respiratory tract infection,27 and only one human case of localized CVA21 myositis has been reported.28 It will be interesting also to test the approach in an immunocompetent syngeneic mouse model but to date we have not been able to identify a mouse tumor that responds to CVA21 oncolytic virotherapy. The idea that INA coding for an oncolytic virus could be used for cancer therapy was formally proposed by Sutton in a letter to the Lancet in 1991.14 But despite the obvious appeal of the approach, its feasibility has not previously been demonstrated. One reason for this apparent lack of progress may be that the viruses employed most widely as oncolytic agents do not lend themselves to the approach. Negative-strand RNA viruses which have been shown to be highly oncolytic both in preclinical and early stage clinical trials29,30 cannot be generated solely by transfection of nucleic acid. Concomitant expression of several viral proteins is required.31,32 Thus, it is not currently feasible to initiate oncolytic infections with vesicular stomatitis virus, measles virus, or newcastle disease virus by transfecting a single nucleic acid construct. The problems are even greater for oncolytic reoviruses which have a segmented double stranded RNA genome.33 Similarly, the double stranded DNA genomes of oncolytic HSV (~150 kb)34 and vaccinia viruses (~200 kb)35,36 are simply too large
Abbreviations: CPK, creatine phosphokinase; CVA21, coxsackievirus A21; TCID50, tissue culture infectious dose 50.
phosphokinase, a serum marker for myositis, was elevated to a similar degree in animals treated with either CVA21 virus or RNA (Table 2), and both groups exhibited myositis both clinically and histologically (Figure 3).
Determination of minimum dose of infectious RNA required to mediate tumor regression After confirming that delivery of 20 µg INA coding for CVA21 could indeed initiate a potent oncolytic infection, we next endeavored to determine if lower doses of in vitro transcribed CVA21 could also initiate a therapeutically effective oncolytic CVA21 infection. We therefore conducted a dose–response study in which seven groups of animals with established Kas6/1 human myeloma tumors of ~5 mm diameter were intratumorally injected with varying doses of naked RNA and tumor progression was monitored daily thereafter. Tumor-bearing animals were treated with doses of 1, 2, 4, 8, 16, or 32 µg of RNA intratumorally in carrier-free buffer, while control animals received buffer only (Figure 4a–g). At RNA doses of 4, 8, 16 ,or 32 µg, all tumors were rapidly controlled although the kinetics of regression were delayed in one of the animals treated at the lower dose of 4 µg. At the lowest doses of RNA (1 or 2 µg), CVA21-mediated tumor regression was observed in only one third of treated animals, and these nonresponding animals did not become viremic. As shown before, although CVA21 proves a curative treatment for established Kas6/1 tumors, all mice that efficiently converted CVA21 RNA to virus later developed severe myositis that induced paralysis and necessitated euthanasia, with kinetics that did not differ significantly across dose levels. (Figure 4h). Molecular Therapy vol. 19 no. 6 june 2011
Figure 3 Formalin-fixed paraffin-embedded tumor sections stained by hematoxylin and eosin. Microscopic images shows inflammation and necrosis in (a) coxsackievirus A21 (CVA21) virus and (b) CVA21 RNA treated tumors while the (c) Opti-MEM control and (d) CVA21RNA + RNaseA tumors are within normal limits. Bar = 0.1 mm. MEM, minimal essential medium.
1043
© The American Society of Gene & Cell Therapy
Infectious Nucleic Acids for Cancer Therapy
a
2.0
b
Control N=3
2.0
1.5
1.0
1.0
0.5
0.5 0.0
0.0 −7 0 7 14 21 28 35 42 49 56 63 70 77
c
1 µg N=6
1.5
Days 2.5
2 µg N=6
2.0
d
−7 0 7 14 21 28 35 42 49 56 63 70 77 Days
0.6
4 µg N=6
0.5 0.4
1.5
0.3 1.0
0.2 0.1
0.0
0.0 −7 0 7 14 21 28 35 42 49 56 63 70 77
−7 0 7 14 21 28 35 42 49 56 63 70 77
e
Days 0.3
Days 0.3 16 µg N=6
0.2
0.2
0.1
0.1
0.0
g
8 µg N=5
f
0.0 −7 0 7 14 21 28 35 42 49 56 63 70 77 Days
0.4
32 µg N=6
0.3 0.2 0.1 0.0
−7 0 7 14 21 28 35 42 49 56 63 70 77 Days
h 100 Percent survival
Tumor volume, cm3
0.5
−7 0 7 14 21 28 35 42 49 56 63 70 77 Days
Control 1 µg 2 µg 4 µg 8 µg 16 µg 32 µg
75 50 25 0 0
7 14 21 28 35 42 49 56 63 70 77 Time, days
Figure 4 In vivo dose–response of tumor regression following intratumoral treatment with coxsackievirus A21 (CVA21) RNA. Tumor volumes of SCID mice carrying subcutaneous Kas6/1 human myeloma xenografts were treated with (a) Opti-MEM control, (b) 1 µg, (c) 2 µg, (d) 4 µg, (e) 8 µg, (f) 16 µg, or (g) 32 µg CVA21 RNA. Mice with tumor progression (gray lines), mice with tumor regression (black lines). (h) Kaplan–Meier survival curves of control and RNA treated mice. Based on the segmented model, there was no overall dose effect (P = 0.19) or dose effect prior to day 10 (P = 0.20). There was, however, a significant dose effect after day 10 (P = 0.0008) showing increased dose was associated with decrease in tumor volume. This was confirmed when evaluating the after day 10 results in a separate model (P < 0.0001). MEM, minimal essential medium.
and their rescue efficiencies from INA too low to allow them to be used for this approach. Even oncolytic adenoviruses which have a smaller genome size of 35 kb could not be effectively exploited in this way because they are rescued only inefficiently from transfected DNA.15 There are, however, several oncolytic viruses with smaller DNA or positive-sense RNA genomes that can be rescued reasonably efficiently simply by transfecting cells with a nucleic acid construct. Examples include the H1 autonomously replicating parvovirus which has a single stranded DNA genome,37 Sindbis, an alphavirus with a positive-sense RNA genome,38 the recently developed oncolytic C-type retroviruses,39 and several oncolytic picornaviruses including poliovirus RIPO,19 seneca valley virus,20 and CVA21.23–25 After considering these possibilities, we concluded that the picornavirus CVA21 offered perhaps the highest chance of success because (i) RNA transcripts corresponding to its full length viral genome have very high specific 1044
infectivity25 and (ii) injection of as few as 1,000 infectious CVA21 viral particles was sufficient to mediate complete eradication of susceptible tumors in mice.24,25 The current study focused on the use of naked RNA as the therapeutic formulation. Advantages of RNA are that it provides an immediate precursor for the generation of progeny picornaviruses and does not need to gain access to the nucleus before it is decoded. But RNA is rapidly degraded by nucleases ubiquitously present in body fluids.40,41 Also, while short siRNAs have been shown to be compatible with liposomal vectors,42,43 longer messenger RNAs have not, and have yet to be tested for compatibility with many of the better developed nonviral gene delivery vehicles. In light of the poor stability of RNA, it was somewhat unexpected to see tumor responses after intravenous therapy with CVA21 RNA. While virus was recovered from both tumor and skeletal muscle of these mice upon euthanasia, it is still unclear www.moleculartherapy.org vol. 19 no. 6 june 2011
© The American Society of Gene & Cell Therapy
where the oncolytic infection initiated. One possibility is that the RNA may have been taken up and converted to virus in the liver or spleen, which are highly accessible to blood-borne agents, and that the released virus then trafficked to the tumor where it mediated tumor regression. In support of this hypothesis, our group has previously determined that cell lineages lacking CVA21’s natural cellular receptor, intracellular adhesion molecule 1,44 transfected with CVA21 RNA can become viral producer cells despite the fact that they are resistant to infection with CVA21 virus (data not shown). It is possible that DNA might provide a more stable and versatile material than RNA for incorporation into nonviral vector formulations, but we were not yet able to rescue CVA21 by plasmid transfection and, to our knowledge, the feasibility of CVA21 virus rescue by transfection of plasmid DNA has not been established by others. In the only paper describing rescue of infectious poliovirus from transfected plasmid DNA,45 the specific infectivity was extremely low. To facilitate a future comparison between infectious RNA and infectious DNA, we are currently evaluating the specific infectivities of DNA plasmids coding for a number of oncolytic picornaviruses. Regarding the question of relative doses required of INA versus intact virus particles to mediate tumor regression, 1 µg of RNA contains ~3 × 1011 viral genomes. The particle to infectivity ratio for CVA21 is unknown but, based on other viruses studied, a reasonable estimate is 100:1. The minimum effective dose of CVA21 administered intratumorally in this myeloma model is ~103 infectous units, containing an estimated 105 CVA21 genomes (data not shown). Hence in this feasibility study 106 copies of infectious RNA was therapeutically equivalent to one infectious virion. We anticipate that this number will be enhanced by optimizing the specific infectivity of the RNA. Compared to many other nonviral vectors, several of which have already entered clinical trials, naked CVA21 RNA is an extremely potent anticancer therapeutic. However, to maximize efficacy in surgically inaccessible tumors, as well as those that have metastasized, a cancer therapeutic would be formulated such that it could be delivered intravenously. Bolstered by our data that showed that as little as 1 µg of CVA21 RNA administered by intratumoral injection could efficiently initiate an oncolytic infection, we did perform preliminary studies to determine if intravenous administration of infectious RNA could elicit similar results as intratumoral RNA. We therefore delivered a high dose (50 µg) of unconjugated CVA21 RNA by intravenous injection in carrier-free buffer to Kas6/1 tumor-bearing SCID mice, and monitored tumor progression daily. Surprisingly, even unconjugated, naked CVA21 RNA was able to initiate an oncolytic infection that induced complete tumor regression (data not shown). However, compared to the intratumoral route of administration, much higher doses of RNA were required to realize anticancer activity after intravenous delivery. There is therefore a strong impetus to test the compatibility of CVA21 RNA with various lipid and nanoparticle vector formulations, both to target its delivery to sites of tumor growth and to enhance its resistance to circulating nucleases after intravenous administration. These experiments are currently underway. In conclusion, we have shown that infectious RNA encoding the full length genome of CVA21 can be used to initiate a Molecular Therapy vol. 19 no. 6 june 2011
Infectious Nucleic Acids for Cancer Therapy
spreading intratumoral infection that causes the rapid and complete regression of human myeloma xenografts in laboratory mice. This novel approach to cancer therapy combines the considerable strengths of nonviral gene therapy vectors with the complementary strengths of oncolytic viral vectors and is highly deserving of further development.
Materials and Methods Cell culture. H1-HeLa cells were obtained from American Type Culture Collection and were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum in 5% CO2. Kas6/1 cells were obtained from Dr Diane Jelinek, Department of Immunology, Mayo Clinic and maintained in RPMI + 10% fetal bovine serum + interleukin-6 at 37 °C in 5% CO2. Infectious RNA preparation, virus rescue, CVA21 sequencing. Infectious RNA was prepared as previously described25 by initial linearization of 1 µg pGEM-CVA21 plasmid with restriction endonuclease MluI (New England Biolabs, Ipswich, MA), followed by ethanol precipitation and resuspension in RNase free water. In vitro transcription reaction was then assembled using the Ambion MEGAscript T7 polymerase kit. RNA was purified thereafter using MEGAclear Kit (Ambion, Austin, TX) and in addition ethanol precipitated. This precipitate was pelleted, washed, and quantified on a spectrophotometer. To insure the transcript was the correct size and retained integrity, in vitro transcribed RNA was run on an RNA Flash Gel (Lonza, Basel, Switzerland). H1-HeLa cells were plated in a 12 well plate 24 hours prior to transfection. To obtain live virus, 1 µg CVA21 RNA was transfected into H1-HeLa cells using Mirus TransIT-mRNA Transfection Kit (Mirus Biosciences, Madison, WI). Forty-eight hours post-transfection, cleared lysate was collected and titrated on H1-HeLa cells. In order to assure that virus rescued from transfected RNA retained sequence integrity, we sequenced rescued virus from virus collected from the supernatant of cells transfected with CVA21 RNA. Briefly, virus was collected from cleared supernatant of transfected cells. Viral RNA was then extracted using a Viral RNA Extraction Kit (Qiagen, Valencia, CA). and cDNA prepared using Superscript one step RT-PCR Kit (Invitrogen, Carlsbad, CA) using the below primers. Overlapping fragments were then sent for sequencing.
Primer ID
Sequence
Nucleotide
CVA-F1
TTAAAACAGCTCTGGGGTTGTTCC
CVA-F2
TGAGTAGGTTGTCGTAATGCG
499–519
CVA-F3
AGCAGCCAATGCTATTGTTGC
1001–1021
CVA-F4
GCACCAGATCATCAACCTACG
1502–1522
CVA-F5
GCTTCTGATAAACGACTGAGC
2001–2021
CVA-F6
AGAGTGTCCCAACCACCCTCG
2499–2519
CVA-F7
TATGGAAATGCACCTCCACGG
3000–3020
CVA-F8
ACCATCGCCAGATGTAGTTGC
3501–3521
CVA-F9
ACTCGGGTGTGATATATCTCC
4007–4027
CVA-F10
TACTTCATTAATAGGTAGAGC
4502–4522
CVA-F11
TTCACTGTAGATGAGATTACC
4989–5009
CVA-F12
AGCAATATTGCCAACCCATGC
5504–5524
CVA-F13
ATCCAAGACAAAGTTAGAACC
6011–6031
CVA-F14
GCAATGAGAATGGCCTTTGGC
6507–6527
CVA-F15
TGGACTAACCATGACTCCAGC
7001–7021
CVA-F16
AATTTTCCTTTAATTTCGGAG
7386–7406
1–24
1045
© The American Society of Gene & Cell Therapy
Infectious Nucleic Acids for Cancer Therapy
CVA-F17
GCTCGTTGTGCCTATTAGC
CVA21-R1
GGGTGGGAACAACCCCAGAGC
7195–7213 29–9
CVA21-R2
GGGTTGTGAGCGGTTGCTCCG
496–476
CVA21-R3
GTGTAGTAATAGTCGAGTTGC
996–976
CVA21-R4
GTACACAAAGGCGTTTCCTGC
1499–1479
CVA21-R5
GGATAAACAGAATATAGGTGC
1991–1971
CVA21-R6
TTTATCGCTGTGTCAATGAGG
2488–2468
CVA21-R7
AGATGGAAGGGTTTGAAGAGG
2991–2971
CVA21-R8
CCCTGGGCTCTAGATTCCACC
3493–3473
CVA21-R9
CTAGTGTTGCAAGGATCGTGG
4002–3982
CVA21-R10
TATAGATTTACCCGTCCCTGG
4499–4479
CVA21-R11
CTACAGTGAACCTTTGTCTGG
4998–4978
CVA21-R12
ACATTATCATGCACCCCTAGC
5503–5483
CVA21-R13
GGGCATTTATAATGGGGTACC
6009–5989
CVA21-R14
TGAATCATTAAGGCTGGAAGC
6503–6483
CVA21-R15
TCTTTTCCTGATTGGGCTAGG
6997–6977
CVA21-R16
GAAATTAAAGGAAAATTTACC
7402–7382
Virus and RNA growth curves. H1-HeLa cells were infected with CVA21 at an multiplicity of infection of 1.0. Two hours after initial infection, cells were washed and media replaced. At this time, alternate cells were transfected in parallel with 1 µg CVA21 RNA in 24 well plates. Cell lysates from H1-HeLa cells were collected at specific times postinfection or post-transfection (12, 24, 36, 48, 60, 72 hours). At the completion of all time-points, lysates were thawed, and cell pellets were cleared from the samples by centrifugation providing a cleared cell lysate fraction. The cleared lysates were then titrated and virus titer was determined by the Spearman–Karber method. In vivo experiments. All animal protocols were reviewed and approved
by Mayo Clinic Institutional Care and Use Committee. CB17 ICR-SCID mice were obtained from Harlan Sprague Dawley (Indianapolis, IN). Mice were irradiated and implanted with 5 × 106 Kas 6/1 in the right flank. When tumors reached an average of 0.5 cm × 0.5 cm, tumors were treated with varying amounts of RNA, or 106 TCID50 CVA21 in 100 µl of OptiMEM. Intravenous experiments proceeded in the same fashion, but with Opti-MEM diluted RNA or virus being delivered via the tail vein. Tumor volume was measured using a hand held caliper and blood was collected by retro-orbital bleeds. Serum samples from retro-orbital bleeds were sent to Marshfield laboratories (Marshfield, WI) for creatine phosphokinase analysis.
Statistical analysis. Tumor volume was modeled on the log2 scale. Due
sharp change in the response curve in mice with tumor regression, a segmented repeated measures model was used to assess overall group or dose effects on tumor volume. Based on graphical observation, the breakpoint chosen for segmented models was 6 days for the experiment shown in Figure 2 and 10 days for the experiment shown in Figure 4. In addition to the segmented model, standard repeated measures models for both pre and postbreakpoint observations were fit to confirm the results from the segmented model.
ACKNOWLEDGMENTS This work was funded in part by R01CA129966 from NIH/NCI and Mayo Clinic.
REFERENCES
1. Bernt, KM, Ni, S, Gaggar, A, Li, ZY, Shayakhmetov, DM and Lieber, A (2003). The effect of sequestration by nontarget tissues on anti-tumor efficacy of systemically applied, conditionally replicating adenovirus vectors. Mol Ther 8: 746–755.
1046
2. Russell, SJ and Peng, KW (2007). Viruses as anticancer drugs. Trends Pharmacol Sci 28: 326–333. 3. Tao, N, Gao, GP, Parr, M, Johnston, J, Baradet, T, Wilson, JM et al. (2001). Sequestration of adenoviral vector by Kupffer cells leads to a nonlinear dose–response of transduction in liver. Mol Ther 3: 28–35. 4. Liu, C, Russell, SJ and Peng, KW (2010). Systemic therapy of disseminated myeloma in passively immunized mice using measles virus-infected cell carriers. Mol Ther 18: 1155–1164. 5. Mader, EK, Maeyama, Y, Lin, Y, Butler, GW, Russell, HM, Galanis, E et al. (2009). Mesenchymal stem cell carriers protect oncolytic measles viruses from antibody neutralization in an orthotopic ovarian cancer therapy model. Clin Cancer Res 15: 7246–7255. 6. Russell, SJ and Peng, KW (2008). The utility of cells as vehicles for oncolytic virus therapies. Curr Opin Mol Ther 10: 380–386. 7. Tseng, JC, Granot, T, DiGiacomo, V, Levin, B and Meruelo, D (2010). Enhanced specific delivery and targeting of oncolytic Sindbis viral vectors by modulating vascular leakiness in tumor. Cancer Gene Ther 17: 244–255. 8. Wojton, J and Kaur, B (2010). Impact of tumor microenvironment on oncolytic viral therapy. Cytokine Growth Factor Rev 21: 127–134. 9. Working, PK, Lin, A and Borellini, F (2005). Meeting product development challenges in manufacturing clinical grade oncolytic adenoviruses. Oncogene 24: 7792–7801. 10. Yamaguchi, T and Uchida, E (2007). Regulatory aspects of oncolytic virus products. Curr Cancer Drug Targets 7: 203–208. 11. Al-Dosari, MS and Gao, X (2009). Nonviral gene delivery: principle, limitations, and recent progress. AAPS J 11: 671–681. 12. Li, S and Ma, Z (2001). Nonviral gene therapy. Curr Gene Ther 1: 201–226. 13. Ohlfest, JR, Freese, AB and Largaespada, DA (2005). Nonviral vectors for cancer gene therapy: prospects for integrating vectors and combination therapies. Curr Gene Ther 5: 629–641. 14. Sutton, PM (1991). Treatment of cancer by infectious nucleic acid. Lancet 337: 1553. 15. Carlisle, RC, Briggs, SS, Hale, AB, Green, NK, Fisher, KD, Etrych, T et al. (2006). Use of synthetic vectors for neutralising antibody resistant delivery of replicating adenovirus DNA. Gene Ther 13: 1579–1586. 16. Agol, VI, Pilipenko, EV and Slobodskaya, OR (1996). Modification of translational control elements as a new approach to design of attenuated picornavirus strains. J Biotechnol 44: 119–128. 17. Girard, M, Altmeyer, R, van der Werf, S, Wychowski, C and Martin, A (1995). The use of picornaviruses as vectors for the engineering of live recombinant vaccines. Biologicals 23: 165–169. 18. Nomoto, A, Iizuka, N, Kohara, M and Arita, M (1988). Strategy for construction of live picornavirus vaccines. Vaccine 6: 134–137. 19. Ochiai, H, Campbell, SA, Archer, GE, Chewning, TA, Dragunsky, E, Ivanov, A et al. (2006). Targeted therapy for glioblastoma multiforme neoplastic meningitis with intrathecal delivery of an oncolytic recombinant poliovirus. Clin Cancer Res 12: 1349–1354. 20. Reddy, PS, Burroughs, KD, Hales, LM, Ganesh, S, Jones, BH, Idamakanti, N et al. (2007). Seneca Valley virus, a systemically deliverable oncolytic picornavirus, and the treatment of neuroendocrine cancers. J Natl Cancer Inst 99: 1623–1633. 21. Sedmak, GV, Taylor, MW, Mealey, J Jr and Chen, TT (1972). Oncolytic effect of bovine enterovirus on mouse and human tumours. Nature New Biol 238: 7–9. 22. Roos, FC, Roberts, AM, Hwang, II, Moriyama, EH, Evans, AJ, Sybingco, S et al. (2010). Oncolytic targeting of renal cell carcinoma via encephalomyocarditis virus. EMBO Mol Med 2: 275–288. 23. Au, GG, Lincz, LF, Enno, A and Shafren, DR (2007). Oncolytic Coxsackievirus A21 as a novel therapy for multiple myeloma. Br J Haematol 137: 133–141. 24. Au, GG, Lindberg, AM, Barry, RD and Shafren, DR (2005). Oncolysis of vascular malignant human melanoma tumors by Coxsackievirus A21. Int J Oncol 26: 1471–1476. 25. Kelly, EJ, Hadac, EM, Greiner, S and Russell, SJ (2008). Engineering microRNA responsiveness to decrease virus pathogenicity. Nat Med 14: 1278–1283. 26. Hyypiä, T, Kallajoki, M, Maaronen, M, Stanway, G, Kandolf, R, Auvinen, P et al. (1993). Pathogenetic differences between coxsackie A and B virus infections in newborn mice. Virus Res 27: 71–78. 27. Buckland, FE, Bynoe, ML and Tyrrell, DA (1965). Experiments on the spread of colds. II. Studies in volunteers with coxsackievirus A21. J Hyg (Lond) 63: 327–343. 28. Dekel, B, Yoeli, R, Shulman, L, Padeh, S and Passwell, JH (2002). Localized thigh swelling mimicking a neoplastic process: involvement of coxsackie virus type A21. Acta Paediatr 91: 357–359. 29. Liu, TC, Galanis, E and Kirn, D (2007). Clinical trial results with oncolytic virotherapy: a century of promise, a decade of progress. Nat Clin Pract Oncol 4: 101–117. 30. Russell, SJ and Peng, KW (2009). Measles virus for cancer therapy. Curr Top Microbiol Immunol 330: 213–241. 31. García-Sastre, A and Palese, P (1993). Genetic manipulation of negative-strand RNA virus genomes. Annu Rev Microbiol 47: 765–790. 32. Palese, P, Zheng, H, Engelhardt, OG, Pleschka, S and García-Sastre, A (1996). Negative-strand RNA viruses: genetic engineering and applications. Proc Natl Acad Sci USA 93: 11354–11358. 33. Van Den Wollenberg, DJ, Van Den Hengel, SK, Dautzenberg, IJ, Kranenburg, O and Hoeben, RC (2009). Modification of mammalian reoviruses for use as oncolytic agents. Expert Opin Biol Ther 9: 1509–1520. 34. Marconi, P, Argnani, R, Epstein, AL and Manservigi, R (2009). HSV as a vector in vaccine development and gene therapy. Adv Exp Med Biol 655: 118–144. 35. Guo, ZS and Bartlett, DL (2004). Vaccinia as a vector for gene delivery. Expert Opin Biol Ther 4: 901–917. 36. Thorne, SH, Hwang, TH and Kirn, DH (2005). Vaccinia virus and oncolytic virotherapy of cancer. Curr Opin Mol Ther 7: 359–365. 37. Ponnazhagan, S (2004). Parvovirus vectors for cancer gene therapy. Expert Opin Biol Ther 4: 53–64.
www.moleculartherapy.org vol. 19 no. 6 june 2011
© The American Society of Gene & Cell Therapy
38. Atkins, GJ, Smyth, JW, Fleeton, MN, Galbraith, SE and Sheahan, BJ (2004). Alphaviruses and their derived vectors as anti-tumor agents. Curr Cancer Drug Targets 4: 597–607. 39. Tai, CK and Kasahara, N (2008). Replication-competent retrovirus vectors for cancer gene therapy. Front Biosci 13: 3083–3095. 40. Alvarez-Salas, LM (2008). Nucleic acids as therapeutic agents. Curr Top Med Chem 8: 1379–1404. 41. Yamamoto, A, Kormann, M, Rosenecker, J and Rudolph, C (2009). Current prospects for mRNA gene delivery. Eur J Pharm Biopharm 71: 484–489.
Molecular Therapy vol. 19 no. 6 june 2011
Infectious Nucleic Acids for Cancer Therapy
42. Higuchi, Y, Kawakami, S and Hashida, M (2010). Strategies for in vivo delivery of siRNAs: recent progress. BioDrugs 24: 195–205. 43. Schroeder, A, Levins, CG, Cortez, C, Langer, R and Anderson, DG (2010). Lipid-based nanotherapeutics for siRNA delivery. J Intern Med 267: 9–21. 44. Shafren, DR, Dorahy, DJ, Greive, SJ, Burns, GF and Barry, RD (1997). Mouse cells expressing human intercellular adhesion molecule-1 are susceptible to infection by coxsackievirus A21. J Virol 71: 785–789. 45. Racaniello, VR and Baltimore, D (1981). Cloned poliovirus complementary DNA is infectious in mammalian cells. Science 214: 916–919.
1047