Cancer Research The Future of Cancer Research: Science and Patient Impact Translational Medicine Conference in Israel
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Simpson, G. R.
Right arrow Articles by Coffin, R. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Simpson, G. R.
Right arrow Articles by Coffin, R. S.
[Cancer Research 66, 4835-4842, May 1, 2006]
© 2006 American Association for Cancer Research

Experimental Therapeutics, Molecular Targets, and Chemical Biology

Combination of a Fusogenic Glycoprotein, Prodrug Activation, and Oncolytic Herpes Simplex Virus for Enhanced Local Tumor Control

Guy R. Simpson1, Ziqun Han1, Binlei Liu1, Yibing Wang1, Gregor Campbell2 and Robert S. Coffin1

1 Biovex Inc., Woburn, Massachusetts and 2 Department of Anatomy and Development Biology, University College London, London, United Kingdom

Requests for reprints: Robert S. Coffin, Biovex Inc., 34 Commerce Way, Woburn, MA 01801. E-mail: rcoffin{at}biovex.com.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
We have previously developed an oncolytic herpes simplex virus-1 based on a clinical virus isolate, which was deleted for ICP34.5 to provide tumor selected replication and ICP47 to increase antigen presentation as well as tumor selective virus replication. A phase I/II clinical trial using a version of this virus expressing granulocyte macrophage colony-stimulating factor has shown promising results. The work reported here aimed to develop a version of this virus in which local tumor control was further increased through the combined expression of a highly potent prodrug activating gene [yeast cytosine deaminase/uracil phospho-ribosyltransferase fusion (Fcy::Fur)] and the fusogenic glycoprotein from gibbon ape leukemia virus (GALV), which it was hoped would aid the spread of the activated prodrug through the tumor. Viruses expressing the two genes individually or in combination were constructed and tested, showing (a) GALV and/or Fcy::Fur expression did not affect virus growth; (b) GALV expression causes cell fusion and increases the tumor cell killing at least 30-fold in vitro and tumor shrinkage 5- to 10-fold in vivo; (c) additional expression of Fcy::Fur combined with 5-fluorocytosine administration improves tumor shrinkage further. These results indicate, therefore, that the combined expression of the GALV protein and Fcy::Fur provides a highly potent oncolytic virus with improved capabilities for local tumor control. It is intended to enter the GALV/Fcy::Fur expressing virus into clinical development for the treatment of tumor types, such as pancreatic or lung cancer, where local control would be anticipated to be clinically advantageous. (Cancer Res 2006; 66(9): 4835-42)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The use of replication competent viruses is an attractive strategy for tumor therapy because the virus can replicate and spread in situ, exhibiting oncolytic activity through a direct cytopathic effect (reviewed by ref. 1). Tumor selectivity is an inherent property of some viruses, such as autonomous parvoviruses and reoviruses (2), whereas for other viruses, such as herpes simplex virus (HSV), the property of tumor selectivity can be genetically engineered (3–5).

Mutants of HSV, in which both copies of the gene encoding ICP34.5 are inactive, replicate selectively in tumors (6, 7). Infection of cells with wild-type HSV-1 reduces antigen expression on the cell surface through the expression of ICP47, which inhibits the transporters associated with antigen presentation (8, 9). Therefore, deletion of ICP47 would be expected to increase the antitumor immune response in the presence of HSV. Deletion of ICP47 also places the Us11 gene under immediate-early promoter control, which enhances growth of HSV ICP34.5 mutants in tumor cells (10–12). We have produced an oncolytic HSV-1 based on a low passage clinical isolate of the virus (strain JS1), in which ICP34.5 and ICP47 are deleted and showing improved tumor shrinkage properties compared with previous viruses (13). A version of this virus expressing granulocyte macrophage colony-stimulating factor (GM-CSF; OncoVEXGM-CSF) has been tested in a phase I clinical trial by direct injection into a number of cutaneous tumor types, with promising results,3 and has now entered phase II studies.

It has been shown previously that expression of the membrane glycoproteins of measles virus and gibbon ape leukemia virus (GALV) can kill cells by fusion into large multinucleated syncytia (14, 15). Here, the GALV protein was artificially truncated by removing 16 amino acids in the transmembrane R-peptide, which normally serves to restrict fusion of the envelope until it is cleaved during viral infection. This renders the protein constitutively highly fusogenic in human tumor cells (16). Early attempts to produce viral vectors based on adenovirus, retrovirus, and lentivirus, encoding GALV env R– failed because rapid cell fusion inhibited vector replication (16), whereas expression from a first-generation oncolytic HSV was found to be possible (17).

Suicide gene therapy is the delivery of genes encoding prodrug-converting enzymes to cancer cells to locally convert a nontoxic prodrug into an active chemotherapy agent, thereby limiting systemic toxicity. A commonly used enzyme/prodrug combination is the Escherichia coli or yeast cytosine deaminase (CD) enzyme, which converts inactive 5-fluorocytosine (5-FC) to active 5-fluorouracil (5-FU). Combined expression of CD with the E. coli or yeast uracil phosphoribosyltransferase (UPRT) enzyme has been shown to markedly increase antitumor effects of 5-FC/CD by removing an otherwise rate-limiting enzymatic step (18).

Following the promising clinical results with OncoVEXGM-CSF, the current study aimed to develop a further virus based on this backbone, in which local control effects were enhanced through the expression of the GALV fusogenic glycoprotein and a CD/UPRT fusion gene (Fcy::Fur). This would be combined with the administration of 5-FC. Here, it was hoped that GALV expression and 5-FC conversion would individually increase the therapeutic effects observed, and that combined expression of GALV with CD would serve to aid the spread of the activated prodrug through the tumor due to the cell fusion effects. This, it was hoped, may be particularly useful where local tumor control is the primary aim of therapy (e.g., in the treatment of primary lung and pancreatic cancers, colorectal liver metastases, and glioma). Such a virus might also be particularly useful in combination with radiotherapy, as CD expression/5-FC administration is known to be a radiosensitizing treatment regime.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
All cell lines were obtained from either the American Type Culture Collection (Rockville, MD) or the European Collection of Animal Cell Cultures (Salisbury, United Kingdom) and were grown at 37°C in recommended media under a humidified atmosphere of 5% CO2. The following cell lines were used: A549, BHK-21 (clone 13), BXPC-3, CALU-1, CAPAN-1, Colo 205, HCT116, HT1080, HT-29, H460, MIC PACA-1, PANC-1, SW620, 9L LacZ.

Construction and characterization of viral vectors. The production of the oncolytic vector OncoVEX was described by Liu et al. (ref. 13; Fig. 1 ). The GALV env R– and Fcy::Fur genes were cloned into an ICP34.5 shuttle vector. The GALV env R– (Genbank accession no. NC_001885; 5,552-7,555 bp) was obtained by reverse transcription-PCR from a viral producer cell line (MLV 144; ref. 19) using primers gs11 (5-CTCGCGGCCGCTTAACATGCACTTATCCTATC-3) EcoRI and gs15 (5-GAGGAATTCGAGATGGTATTGCTGCCTGGGTCC-3) NotI. When sequenced, the envelope showed 99% identity to Delassus et al. (20) sequence. The envelope (GALV env R–) was cloned into pcDNA3. To aid cloning, a NotI site was removed from pcDNA3 GALV env R– by digestion using NotI, blunting with T4 DNA polymerase and followed by religation. The cytomegalovirus (CMV) GALV env R– pA cassette was subcloned into pGEM T Easy (Promega, Madison, WI) by PCR using primers gs17 (5-GAATCTGCTTAGGGTTAGGCG-3) and gs18 (5-AGCCCACCGCATCCCCAGCAT-3). Finally CMV GALV env R– pA was subcloned into the final shuttle vector p-34.5 that is based on pSP72 (Promega) and contains two flanking regions on either side of the HSV-1 ICP34.5 gene (nucleotides 123462-124958 and 125713-126790 bp from HSV-1 strain 17+ Genbank X14112). The Fcy::Fur gene was removed from pORF Fcy::Fur (Invivogen, San Diego, CA) as a NheI and NcoI fragment, blunted with T4 polymerase, and then subcloned into pRcRSV (Invitrogen, San Diego, CA), which had previously been cut with XbaI and HindIII and also blunted with T4 DNA polymerase. The Rous sarcoma virus (RSV) Fcy::Fur pA cassette was then cut out of pRcRSV Fcy::Fur as a PvuII-NruI fragment and subcloned into the shuttle vector p-34.5, which had been cut with NotI and blunted with T4 polymerase.


Figure 1
View larger version (15K):
[in this window]
[in a new window]
  		Figure 1. Schematic representation of viruses used in this study. OncoVEX was derived from HSV-1 strain JS-1 (13) and has two deletions, that of the genes encoding ICP34.5 (nucleotides 124948-125713 based on the sequence of HSV-1 strain 17+) and ICP47 (nucleotides 145570-145290). OncoVEXGALV expresses the envelope of GALV minus the R– peptide (Genbank accession no. NC_001885; 5,552-7,555 bp; refs. 14, 15) under the CMV promoter. OncoVEXCD expresses Fcy::Fur under the RSV promoter. OncoVEXGALV/CD expresses both the GALV and Fcy::Fur proteins.

 
A ICP34.5 double-cassette shuttle vector was produced to express GALV and Fcy::Fur from the same site. pcDNA3 (Invitrogen) was digested with PvuII, blunted, and ligated to a self-reverse complementing oligonucleotide [NheI-HpaI-NheI (NHN)NHN, GCTAGCGAATTCGCTAGC] to generate pcDNA3NHN. pcDNA3NHN (3.3 kb) is significantly smaller than pcDNA3 (5.4 kb) as the two PvuII DNA fragments containing SV40 promoter, neomycin gene, and the SV40 polyadenylation signal have been removed. pcDNA3NHN was digested with HindIII and XbaI and ligated to a linker (LK2) to generate pcDNA3NHNLK2. The multiple cloning sites (MCS) of pcDNA3NHN were replaced by the sequence of LK2 that consists of two complementing oligonucleotides: HBPXX2 (AGCTGGATCCATCGATCACGTGCAATTGCTCGAG) and HBPXX2B (CTAGCTCGAGCAATTGCACGTGATCGATGGATCC). pRcRSVLK1 is the pRcRSV (Invitrogen)–derived plasmid, in which the MCS were replaced by a linker (LK1, HindIII-NotI-PmeI-PstI-XbaI). LK1 consists of two oligonucleotides: linker1HPX1 (AGCTTGCGGCCGCGTTTAAACCCTGCAGGT) and linker1HPX2 (TAGACCTGCAGGGTTTAAACGCGGCCGCA). pRcRSV was digested with HindIII and XbaI and ligated with LK1 to form pRcRSVLK1. The expression cassette containing the RSV promoter, LK1, and BGHpA was cut from pRcRSVLK1 with NruI and PvuII and blunt ligated into the NruI site of pcDNA3NHNLK2. The newly generated plasmid was named pDCdoubleBGHpA. pDC was generated by replacing the BGHpA downstream of RSV promoter with a LacZpA. The LacZpA was amplified by with Taq DNA polymerase (Promega) and primers LZPAF1 (CATATCTAGATAATAACGGGCAGGCCATGTCTGC) and LZPAR1 (ACAGATCTAAGTCGACACAAACCACAACTAGAATGC). The template used for PCR was pDeltaICP4CMVRSVPR19. The PCR product was cut with BglII and XbaI and inserted in the BglII and XbaI sites of pDCdoubleBGHpA to replace the BGHpA. This plasmid was called pDC. The double cassette was inserted into the shuttle vector p-34.5 in two steps. First, the double cassette was cut out from pDC (HpaI/BglII) followed by T4 DNA polymerase treatment to facilitate the blunt-end cloning. This DC fragment was then inserted into p-34.5 NotI site treated with T4 DNA polymerase. For cloning, GALV and Fcy::Fur were cut out from pcDNA3kGALVenvR– (EcoRI/NotI) and pORFFcy::Fur (Invivogen; NcoI/NheI), respectively. The fragments were subsequently treated with T4 DNA polymerase to facilitate the blunt-end cloning. The GALV was then inserted into PmlI site under control of CMV and Fcy::Fur into PpmeI site under control of RSV.

The shuttle plasmids were recombined into an oncolytic HSV-1 vector (OncoVEX GFP) by standard calcium phosphate transfection technique. Recombinant viruses were identified by reporter gene transfer and plaque purified. The backbone vector was previously characterized by Southern blot (13), and the presence of GALV/Fcy::Fur sequences in the HSV-1 genome was detected by PCR.

Production of high-titer stock of recombinant vector. Roller bottles (10 x 850 cm2; Greiner Bio-one, Longwood, FL) were plated with 4 x 107 BHK cells in 100 mL of FGM per roller bottle. Each roller bottle was gassed with CO2 (BOC, Guildford, United Kingdom) for 1 minute, and the cells were incubated at 37°C/0.5 rpm until the BHK cells had grown to 80% confluency. The cells were infected at an multiplicity of infection (MOI) of 0.01 with a submaster stock of virus in a total volume of 50 mL per roller bottle of FGM. The cells were then incubated at 32°C/0.5 rpm for 3 to 5 days until complete CPE was observed. The infected cells were harvested by aspiration after vigorous shaking of the roller bottle and were then frozen overnight at –80°C. Following thawing at room temperature the cellular debris was then pelleted by centrifugation at 3,000 rpm (1,350 x g) for 30 minutes at 4°C. The supernatant was filtered through a sterile capsule pore size 0.65 ± 0.45 µm (Sartobran 300, Sartorius, Edgewood, NY) using a Whatson Marlow pump and Marprene tubing. The virus particles were pelleted by centrifugation at 12,000 rpm (21,700 x g) for 2 hours at 4°C, upon which the supernatant was removed, and 1 mL DMEM was added. The viral pellets were resuspended gently by shaking at 4°C overnight. The titer of the resuspended virus was determined using the standard viral plaque assay.

In vitro GALV dose response assay. Test cells were plated at 1.25 x 105 per well of a 24-well tray and incubated at 37°C/5% CO2 o/n. The FGM was removed, and 200 µL of OncoVexGALV/CD or OncoVex at MOIs of 0.0001, 0.001, 0.01, and 0.1 in DMEM were added. This was then incubated at 37°C/5% CO2 for 1 hour. The virus dilutions were removed and replaced with 1 mL FGM. This was then incubated at 37°C for 48 hours. The cells were washed twice with 1 mL of PBS and then fixed in 1 mL of 0.1% glutaldehyde (Sigma, St. Louis, MO) in PBS for 10 minutes at room temperature. The cells were washed twice again and stained with 0.1% Crystal Violet, 20% ethanol in PBS for 10 minutes at room temperature. Excess stain was removed with H2O, and the plates were allowed to dry. The cells were then digitally photographed using an inverted microscope (Nikon Eclipse TE200) and Lucia Image (MV-1500 version 4.6). Quantification was carried out by cell counting on four separate fields of view for each cell line, and the results were compared using the Student's t test.

Prodrug-activating assay. HT1080 or BHK cells were plated at 1 x 105 per well of a 24-well tray and incubated at 37°C/5% CO2 overnight. The cells were infected with OncoVEXCD and OncoVEXGALV/CD at an MOI of 0.01 (in 200 µL DMEM); controls included backbone virus and no virus control. After 30 minutes at 37°C/5% CO2, the virus was removed, and 1 mL of FGM containing 5-FC (C4H4FN2O; Sigma) at 600 µmol/L was added and incubated for 48 hours at 37°C/5% CO2. The cell supernatant was transferred into a fresh tube, and the cell debris was removed by spinning at 1,500 rpm (340 x g) for 5 minutes at 4°C. The supernatants were added to a fresh tube and then incubated at 60°C for 10 minutes, to inactivate the virus. The resulting supernatants were allowed to cool to room temperature. Test cells were plated at 1 x 104 per well of a 24-well tray and were incubated at 37°C/5% CO2 overnight. The heat-treated supernatants were added to the fresh HT1080 cells and incubated at 37°C/5% CO2 for 72 hours. Prior fixing, the cells were washed twice with 1 mL of PBS and then incubated in 1 mL of 0.1% glutaldehyde (Sigma) in PBS for 10 minutes at room temperature. The cells were washed twice and stained with of 0.1% Crystal Violet, 20% ethanol in PBS for 10 minutes at room temperature. The excess stain was removed with H2O, and the plates were allowed to dry. The cells were then digitally photographed using an inverted microscope (Nikon Eclipse TE200) and Lucia Image (MV-1500 version 4.6). Quantification was carried out by cell counting on four separate fields of view, and the results were compared using the Student's t test.

In vivo xenograft tumor models. All experimental procedures were done with the authority of the Home Office, following UK guidelines and in accordance with best animal practice. BALB/c nu/nu mice were obtained from Harlan Laboratories (Indianapolis, IN) at 6 to 8 weeks old. Human Fadu/HT1080 cells were implanted (2 x 106 in 100 µL DMEM) s.c. in the right flank of the animal. Tumors were allowed to develop to an average diameter of ~0.4 to 0.6 cm, and the animals were randomized into treatment groups (n = 5 of 7), before vector administration. Virus vectors were diluted in DMEM (Life Technologies) and were injected i.t. [50 µL, 5 x 107 plaque-forming units (pfu)/mL/1 x 107 pfu/mL]. Average tumor diameters were measured using the following formula: average diameter = (length + width + depth) / 3. Statistical analysis was carried out using the unpaired Student's t test.

In vivo toxicology assay. Adult (200-250 g) male Fischer 344 rats (Harlan Laboratories) were inoculated (10 µL of 107 pfu/mL) by stereotaxic injection of vector suspensions into the right forebrain. The animals were sacrificed and perfused with ice-cold 4% paraformaldehyde. The resulting rat brains were incubated at 4°C in 30% sucrose solution (in PBS) for 3 to 4 days and then cytosectioned (Leica SM 2000R). The 30-µm sections were then stained with H&E.

In vivo oncolytic/prodrug rat tumor models. Fischer 344 rats (Harlan Laboratories) were treated with 10 mg/kg/d of Neoral (Ciclosporin) for 10 days and were injected with 4 x 106 9L LacZ cells in the right flank on day 6. Tumors were allowed to develop to an average diameter of ~0.4 to 0.6 cm, and the animals were randomized into treatment groups (n = 7 of 8), before vector administration. The tumors were injected eight times with 50 µL of 107 pfu/mL (5 x 105 pfu) of the viral vectors. Prodrug 5-FC was administered by i.p. injections at 500 mg/kg. 5-FC was dissolved at 12.5 mg/mL in HBSS (Life Technologies). The 16 injections of 5-FC were carried out from days 18 to 47. The volume of 5-FC injected into each rat per kg of body weight is shown below.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Virus construction. The viruses constructed are shown in Fig. 1. OncoVEXGALV expresses the GALV fusogenic glycoprotein from a CMV promoter; OncoVEXCD expresses the fusion Fcy::Fur gene from an RSV promoter; and OncoVEXGALV/CD expresses both genes in a back to back orientation from CMV and RSV promoters, respectively. In each case, the genes were inserted to replace ICP34.5. Virus structures were confirmed by multiple PCRs, and expression of the GALV protein by phenotypic assay and Fcy::Fur were confirmed by western blot (data not shown).

GALV expression does not affect virus replication. Oncolytic viruses are by definition replication competent, and it is, therefore, important that this replication is not impaired if efficacy is to be maintained and virus stocks are to be effectively produced. Cell-to-cell fusion with GALV requires expression of the Pit-1 receptor (21), which is not present in rodent cells, such as BHK cells (22), which are often used to grow stocks of HSV. Thus, growth was not impaired in these cells (Fig. 2 ). To address whether GALV-related fusion inhibits HSV replication in other cells, stocks of OncoVEXGALV and control OncoVEX virus were prepared on HT1080 cells, which do express the Pit-1 receptor that allows GALV-related fusion (Fig. 2A). The resulting stocks were replated onto BHK cells, which showed that there was no inhibitory effect of GALV expression on virus replication. Indeed, at the lowest MOI (0.01), cell fusion seemed to increase virus replication up to 10-fold compared with the backbone virus (Fig. 2A).


Figure 2
View larger version (21K):
[in this window]
[in a new window]
  		Figure 2. Replication of OncoVEX and OncoVEXGALV on BHK and HT1080 cells. Cells were infected at various MOI (0.1, 0.01, 0.001) with either OncoVEX or OncoVEXGALV and incubated at 37°C for 48 hours. The resulting virus stocks were titered on BHK cells: (A) virus grown on HT1080 cells (human fibrosarcoma) and (B) virus grown on BHK cells (hamster fibroblasts).

 
GALV expression causes cell to cell fusion in vitro. A panel of human cancer cell lines was infected with OncoVEX and OncoVEXGALV to compare plaque morphology. Figure 3A shows that OncoVEX gives a classic HSV-1 cytopathic effect, as would be expected. In contrast, cells infected with OncoVEXGALV showed fusion in the form of large multinucleated syncytia, which were then surrounded with cells showing the more classic HSV-1-mediated effect. BHK cells, which do not express the Pit-1 receptor, are shown for comparison. Results were similar in a wide range of tumor cell lines, including fibrosarcoma (HT1080 cells), colon carcinoma (HCT 116 cells), colorectal adenocarcinoma (COLO 205 cells; data not shown), pancreatic adenocarcinoma (CAPAN-1 cells), pancreatic carcinoma (MIA PACA-2 cells; data not shown), and glioblastoma astrocytoma (U87-MG cells; data not shown).


Figure 3
View larger version (80K):
[in this window]
[in a new window]
  		Figure 3. A comparison of plaque morphology and cell killing of OncoVEX and OncoVEXGALV on a panel of malignant cells. A, plaque morphology on HT1080 (human fibrosarcoma), HCT 116 (human colonic carcinoma), CAPAN-1 (human pancreatic adenocarcinoma), BHK (hamster normal baby kidney) cells. B, cell killing. HT1080 cells were infected at various MOI and incubated for 48 hours at 37°C then fixed and stained. C, a panel of malignant cells were infected at various MOI and incubated for 48 hours at 37°C then fixed and stained. COLO 205 (human colorectal adenocarcinoma), HT-29 (human colorectal adenocarcinoma), HCT 116 (human colonic carcinoma), U87-MG (human glioblastoma astrocytoma), CAPAN-1 (human pancreatic adenocarcinoma), CALU-1 (human lung epidermoid carcinoma), and 9L LacZ (rat gliosarcoma). D, quantification of the infection studies (see Materials and Methods), showing that in all cases the differences were significant (P = 0.0012-0.00002).

 
GALV expression improves the dose response in vitro. To test for improved tumor cell killing effects in vitro, HT1080 (fibrosarcoma) cells were infected with OncoVEX or OncoVEXGALV at a range of MOIs. OncoVEXGALV showed considerably greater cell killing than the backbone vector alone at all MOI, estimated to be >20-fold compared with the OncoVEX control (Fig. 3B). To test this further, a panel of cancer cell lines, including colon carcinoma cells (HCT 116), colorectal adenocarcinoma cells (COLO 205, HT-29), pancreatic adenocarcinoma cells (CAPAN-1), lung epidermoid carcinoma cells (CALU-1), and glioblastoma astrocytoma cells (U87-MG), was infected in a similar manner. This showed that cell lines derived from a wide range of human cancers are killed more effectively by OncoVEXGALV compared with the backbone vector (Fig. 3C), which was further quantified in Fig. 3D, showing a 5- to 30-fold enhancement. Interestingly, although HT-29 colorectal adenocarcinoma cells did not show a fusogenic plaque morphology when infected with OncoVEXGALV (data not shown), there was still a 5- to 10-fold increase in cell killing with OncoVEXGALV compared with the empty OncoVEX backbone alone.

GALV expression improves the dose response in vivo. Human cells express the Pit-1 receptor to which the GALV glycoprotein binds (21), but a homologous receptor is not present on mouse cells (22). Therefore, in vivo studies were carried out using xenograft nude mouse models and human tumor cells. Human squamous cell carcinoma cells (Fadu) or human fibrosarcoma (HT1080) cells were implanted s.c. in the flank of BALB/c nu/nu mice. Tumors were allowed to develop, and the animals were randomized into three treatment groups: backbone vector (OncoVEX), backbone vector expressing GALV env R– (OncoVEXGALV), and no treatment control, before vector administration; 2.5 x 106 or 5 x 105 pfu of each virus vector was injected i.t. three times. A dose of 2.5 x 106 pfu of OncoVEXGALV does not increase the tumor shrinkage effects of the backbone vector (Fig. 4A and B ). However, at a 5-fold lower dose (5 x 105 pfu), OncoVEXGALV retains its ability to cause tumor shrinkage, whereas OncoVEX does not (Fig. 4C and D). Thus, expression of GALV env R– improves the dose response to virus injection in terms of tumor killing in vivo.


Figure 4
View larger version (37K):
[in this window]
[in a new window]
  		Figure 4. GALV expression increases therapeutic effects in vivo. Effects of OncoVEX, or OncoVEXGALV on Fadu (A and C; n = 5) or HT1080 (B and D; n = 7) BALB/c nu/nu xenograft tumor models injected thrice (arrows) with either 2.5 x 106 pfu of virus (A and B) or 5 x 105 pfu of virus (C and D). Points, average tumor diameters [(length + width + depth) / 3]; bars, SE. The unpaired Student's t test was used to compare virus treatment groups at the last time point. E, effects of OncoVEXGALV in normal tissue. OncoVEXGALV, OncoVEX (10 µL of 107 pfu/mL), or media control were injected into the forebrains of male Fischer 344 rats. Brains were sectioned (30 µmol/L) 21 days later and stained with H&E.

 
OncoVEXGALV does not cause cell fusion in normal tissue. The in vivo studies using OncoVEXGALV showed enhanced tumor shrinkage in the presence of GALV env R–. An important question was therefore whether this also resulted in increased toxicity to surrounding normal tissue or systemically. Nude mice are not suitable for such experiments as mouse cells are not susceptible to cell fusion as discussed previously. However, rat cells do express a Pit-1 receptor homologue (22); thus, rats were used for this study. To confirm fusion in rat cells, two rat glioma cell lines (9L and RG2 cells) were first infected with OncoVEX and OncoVEXGALV. As expected, OncoVEXGALV showed a fusogenic plaque morphology similar to that seen on human tumor cells (data not shown), combined with a similar enhancement in cell killing (Fig. 3C and D). ICP34.5– mutants of HSV are usually avirulent following i.c. injection (3, 6, 23–27). Therefore, we tested whether the expression of GALV env R– alters this in normal rat brain by stereotaxic injection of OncoVEXGALV/CD into the forebrain. The resulting H&E-stained section showed similar needle track effects in animals injected with either media alone or OncoVEXGALV/CD with no evidence of any viral- or GALV-mediated effects (Fig. 4E).

Fcy::Fur expression converts 5-FC to 5-FU in vitro. Fcy::Fur is a fusion of two genes CD and UPRT, which metabolizes 5-FC more efficiently than either gene alone (18). To study the cell killing effects of HSV-expressing Fcy::Fur gene in the presence of 5-FC, HT1080 cells were infected with OncoVEXCD in the presence or absence of 5-FC. The cell supernatants were then heat inactivated to neutralize the virus, allowing the effects of any 5-FU produced to be determined on fresh tumor cells. In the presence of supernatants from cells infected with OncoVEX, no cell death was seen with or without 5-FC (Fig. 5A ). However, in the presence of the supernatant from OncoVEXCD- or OncoVEXGALV/CD-infected cells displayed effective cell killing in the presence of 5-FC (Fig. 5A). Expression of the Fcy::Fur gene from the HSV-1 vector (with and without GALV env R–), therefore, promotes conversion of 5-FC into 5-FU, resulting in tumor cell killing.


Figure 5
View larger version (71K):
[in this window]
[in a new window]
  		Figure 5. Prodrug activation by OncoVEXCD in vitro and effects on virus replication. A, BHK cells infected for 48 hours with OncoVEX, OncoVEX CD, or OncoVEXGALV/CD (MOI = 0.01) and no virus control, with or without 600 µmol/L 5-FC. Supernatants were then heat inactivated and added to fresh BHK cells for 72 hours. B and C, supernatants from the OncoVEXGALV/CD-infected cells are active on a panel of cancer cells: HCT 116 (human colonic carcinoma), HT-29 (human colorectal adenocarcinoma), A549 (human lung carcinoma), H460 (human lung carcinoma), CAPAN-1 (human pancreatic adenocarcinoma), MIA PACA-2 (human pancreatic carcinoma), BXPC-3 (human pancreatic adenocarcinoma), SW620 (human colorectal adenocarcinoma). D, quantification of the prodrug studies (see Materials and Methods), showing that in all cases the differences were significant (P = 00089-0.0001). E, 5-FC does not inhibit the replication of OncoVEXCD. BHK cells were infected with OncoVEXCD at MOIs of 0.1, 0.01, or 0.001 with or without 600 µmol/L 5-FC for 48 hours, and infected lysates were replated on to BHK cells to obtain viral titer.

 
This was tested in a range of tumor cells including cell lines derived from lung cancer (A549 and H460), pancreatic cancer (CAPAN-1, MIA PACA-2, and BXPC-3), and colon cancer (HCT 116, HT-29, and SW620), where effective cell killing was seen in each case (Fig. 5B and C).

5-FU production does not inhibit virus replication. The result of 5-FC to 5-FU conversion is inhibition of host DNA replication, which might be expected to also inhibit HSV replication. We, therefore, investigated whether this was the case. BHK cells were infected with OncoVEXCD at various MOI with or without 600 µmol/L 5-FC. The resulting infected lysates were then titered on BHK cells (Fig. 5E). This showed no significant difference in titer with or without 5-FC, even at an MOI of 0.001 on both cell types tested (Fig. 5E). Therefore, it seems that the replication of HSV is not inhibited by Fcy::Fur expression and accompanying prodrug conversion.

Combined expression of GALV and Fcy::Fur with 5-FC administration improves tumor responses in vivo. To establish whether the combined expression of GALV and Fcy::Fur enhances tumor killing in vivo, Fischer F344 rats were implanted with 9L LacZ cells s.c. in the flank. The resulting animals were randomized into five treatment groups, before administration of 5 x 105 pfu (50 µL of 107 pfu/mL) of OncoVEX, OncoVEXGALV, OncoVEXCD, OncoVEXGALV/CD, or media alone. All rats were also administered with 5-FC (500 mg/kg) by the i.p. route. In this tumor model, which is relatively resistant to treatment with oncolytic HSV and the virus dose used, the OncoVEX-treated group in general showed a pattern of tumor growth similar to that of untreated tumors (Fig. 6A ), with only one of six tumors being cured (Fig. 6B). In contrast, treatment with OncoVEXGALV or OncoVEXCD gave both a considerable reduction in the average tumor diameters (Fig. 6A) and a considerable increase in the number of tumors cured (Fig. 6B), which was statistically significant (P = 0.1 and P = 0.008, respectively). Although this was confounded by the fact that some of the animals in the OncoVEX-treated group had to be killed before the end of the experiment due to their tumor size, which artificially reduced the degree of significance seen. By day 50, all animals in the control and OncoVEX groups had been sacrificed as tumors had reached the maximum allowed size. However, OncoVEXGALV had cured 57.0% of the tumors, OncoVEXCD had cured 83% of the tumors, and OncoVEXGALV/CD had cured 100% of the tumors by this time (Fig. 6B). Thus, expressing GALV env R– or Fcy::Fur dramatically increases the efficiency of tumor cure in vivo with oncolytic HSV. Whereas the maximum effects seem to be observed when both genes are expressed together, this difference between combined expression and the expression of each gene individually does not reach statistical significance, although it is notable that it is further only in the OncoVEXGALV/CD group where all tumors are cured.


Figure 6
View larger version (21K):
[in this window]
[in a new window]
  		Figure 6. Effects of GALV and Fcy::Fur expression in a rat tumor model in vivo. A, Fischer f344 rats were injected in the right flank with 107 9L LacZ cells. The resulting tumors were injected with 5 x 105 pfu of either OncoVEX, OncoVEXGALV, OncoVEXCD, or OncoVEXGALV/CD and no treatment control (PBS; arrows). Animals were also dosed with 500 mg/kg i.p. of 5-FC on days 18, 19, 22, 23, 25, 26, 30, 31, 32, 36, 37, 39, 40, 45, 46, and 47. B, number and % tumor cured in each treatment group at either 43 or 50 days after tumor induction.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this study, we sought to develop a versatile oncolytic HSV based on a second-generation oncolytic virus backbone (13) with improved capabilities for local tumor control. The intention was to increase cell killing by expression of the fusogenic GALV glycoprotein, which has previously been shown promising results in preclinical tumor models, and combine this with local conversion of 5-FC to 5-FU by the expression of a highly potent version of CD (Fcy::Fur; ref. 18). Here, it was hoped that GALV expression would aid the spread of the activated prodrug through the tumor, increasing therapeutic effects further. Such a virus could, thus, be used for local tumor control either alone, in combination of 5-FC administration, and/or in combination with radiotherapy, as the 5-FC/CD system is known to be a potently radiosensitizing treatment regime. Moreover, whereas not tested here, the virus also naturally expresses thymidine kinase, which when combined with gancyclovir is also known to be radiosensitizing (28), particularly together with CD/5-FC (29). Thus, multiple treatment options can be envisaged with such a virus.

To this end, three viruses were produced expressing either a truncated version of the GALV glycoprotein (GALV env R–), the yeast Fcy::Fur gene, or both genes together.

The expression of GALV env R– increased plaque size and gave a 5-fold to at least 30-fold increase in cell killing in a wide range of tumor cell lines representative of a broad range of potential clinical targets. This included lung cancer, pancreatic cancer, colorectal cancer, and glioma. Unlike when using, for example, adenovirus vectors to deliver the GALV gene (16), virus replication was not inhibited, which may be due to the faster replication time of HSV.

In vivo, in the tumor models tested, improved cell killing as evidenced by tumor shrinkage was again mediated by GALV, with ~5- to 10-fold improvement in the dose response to virus treatment. This was further improved by the expression of the Fcy::Fur gene and administration of 5-FC. Importantly, 5-FC to 5-FU conversion also did not inhibit the replication of the virus, which might have been expected to occur. The in vivo experiments were performed in xenograft, nude mouse models, as the GALV protein does not fuse murine cells, and also in immune competent rats, rat cells being susceptible to GALV-mediated fusion. These experiments showed no evidence of toxicity associated with GALV expression, including when the virus was tested by i.c. inoculation in the rat, and a statistically significant benefit to the therapeutic effect was seen with each gene.

Thus, while the GALV glycoprotein has previously been expressed from a first-generation oncolytic HSV (17), the work described here uses a second generation virus in which tumor-selective replication is greatly enhanced through its derivation from a clinical isolate of the virus and through the expression of the US11 gene as an immediate-early rather than late gene (13). The work here also combines GALV expression with prodrug conversion, which gave improved therapeutic effect compared with the use of either approach alone. However, the effectiveness of the individual approaches (giving 57% and 83% tumor cure for Fcy::Fur or GALV expression, respectively, compared with only 16.6% with the backbone vector) prevented a statistically significant improvement with the double approach being reached in vivo, which gave 100% tumor cure. We believe that this virus is highly promising for clinical development for the local control of cancers, such as pancreatic cancer, colorectal liver metastases, lung cancer, or glioma, and might also be combined with other therapies, such as radiotherapy, which would be anticipated from previous work to be synergistic with the prodrug conversion approach.


   Acknowledgments
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   Footnotes
 
3 Submitted for publication. Back

Received 12/ 6/05. Revised 2/14/06. Accepted 2/23/06.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Ring CJ. Cytolytic viruses as potential anti-cancer agents. J Gen Virol 2002;83:491–2.[Abstract/Free Full Text]
  2. Coffey MC, Strong JE, Forsyth PA, Lee PW. Reovirus therapy of tumors with activated Ras pathway. Science 1998;282:1332–4.[Abstract/Free Full Text]
  3. MacLean AR, ul-Fareed M, Robertson L, Harland J, Brown SM. Herpes simplex virus type 1 deletion variants 1714 and 1716 pinpoint neurovirulence-related sequences in Glasgow strain 17+ between immediate early gene 1 and the ‘a’ sequence. J Gen Virol 1991;72:631–9.[Abstract/Free Full Text]
  4. Bolovan CA, Sawtell NM, Thompson RL. ICP34.5 mutants of herpes simplex virus type 1 strain 17syn+ are attenuated for neurovirulence in mice and for replication in confluent primary mouse embryo cell cultures. J Virol 1994;68:48–55.[Abstract/Free Full Text]
  5. Mineta T, Rabkin SD, Yazaki T, Hunter WD, Martuza RL. Attenuated multi-mutated herpes simplex virus-1 for the treatment of malignant gliomas. Nat Med 1995;1:938–43.[CrossRef][Medline]
  6. Rampling R, Cruickshank G, Papanastassiou V, et al. Toxicity evaluation of replication-competent herpes simplex virus (ICP 34.5 null mutant 1716) in patients with recurrent malignant glioma. Gene Ther 2000;7:859–66.[CrossRef][Medline]
  7. Andreansky SS, He B, Gillespie GY, et al. The application of genetically engineered herpes simplex viruses to the treatment of experimental brain tumors. Proc Natl Acad Sci U S A 1996;93:11313–8.[Abstract/Free Full Text]
  8. Jennings SR, Rice PL, Kloszewski ED, Anderson RW, Thompson DL, Tevethia SS. Effect of herpes simplex virus types 1 and 2 on surface expression of class I major histocompatibility complex antigens on infected cells. J Virol 1985;56:757–66.[Abstract/Free Full Text]
  9. Hill AB, Barnett BC, McMichael AJ, McGeoch DJ. HLA class I molecules are not transported to the cell surface in cells infected with herpes simplex virus types 1 and 2. J Immunol 1994;152:2736–41.[Abstract]
  10. Mohr I, Gluzman Y. A herpesvirus genetic element which affects translation in the absence of the viral GADD34 function. EMBO J 1996;15:4759–66.[Medline]
  11. He B, Chou J, Brandimarti R, Mohr I, Gluzman Y, Roizman B. Suppression of the phenotype of gamma(1)34.5- herpes simplex virus 1: failure of activated RNA-dependent protein kinase to shut off protein synthesis is associated with a deletion in the domain of the alpha47 gene. J Virol 1197;71:6049–54.
  12. Cassady KA, Gross M, Roizman B. The herpes simplex virus US11 protein effectively compensates for the gamma1(34.5) gene if present before activation of protein kinase R by precluding its phosphorylation and that of the alpha subunit of eukaryotic translation initiation factor 2. J Virol 1998;72:8620–6.[Abstract/Free Full Text]
  13. Liu BL, Robinson M, Han ZQ, et al. ICP34.5 deleted herpes simplex virus with enhanced oncolytic, immune stimulating, and anti-tumour properties. Gene Ther 2003;10:292–303.[CrossRef][Medline]
  14. Bateman A, Bullough F, Murphy S, et al. Fusogenic membrane glycoproteins as a novel class of genes for the local and immune-mediated control of tumor growth. Cancer Res 2000;60:1492–7.[Abstract/Free Full Text]
  15. Galanis E, Bateman A, Johnson K, et al. Use of viral fusogenic membrane glycoproteins as novel therapeutic transgenes in gliomas. Hum Gene Ther 2001;12:811–21.[CrossRef][Medline]
  16. Diaz RM, Bateman A, Emiliusen L, et al. A lentiviral vector expressing a fusogenic glycoprotein for cancer gene therapy. Gene Ther 2000;7:1656–63.[CrossRef][Medline]
  17. Fu X, Tao L, Jin A, et al. Expression of a fusogenic membrane glycoprotein by an oncolytic herpes simplex virus potentiates the viral antitumor effect. Mol Ther 2003;7:748–54.[CrossRef][Medline]
  18. Tiraby M, Cazaux C, Baron M, Drocourt D, Reynes JP, Tiraby G. Concomitant expression of E. coli cytosine deaminase and uracil phosphoribosyltransferase improves the cytotoxicity of 5-fluorocytosine. FEMS Microbiol Lett 1998;167:41–9.[CrossRef][Medline]
  19. Rangan SR, Gallagher RE. Tumors and viruses in nonhuman primates. Adv Virus Res 1979;24:1–123.[Medline]
  20. Delassus S, Sonigo P, Wain-Hobson S. Genetic organization of gibbon ape leukemia virus. Virology 1989;173:205–13.[CrossRef][Medline]
  21. O'Hara B, Johann SV, Klinger HP, et al. Characterization of a human gene conferring sensitivity to infection by gibbon ape leukemia virus. Cell Growth Differ 1990;1:119–27.[Abstract]
  22. Sommerfelt MA, Weiss RA. Receptor interference groups of 20 retroviruses plating on human cells. Virology 1990;176:58–69.[CrossRef][Medline]
  23. Chou J, Kern ER, Whitley RJ, Roizman B. Mapping of herpes simplex virus-1 neurovirulence to gamma 134.5, a gene nonessential for growth in culture. Science 1990;250:1262–6.[Abstract/Free Full Text]
  24. Taha MY, Clements GB, Brown SM. The herpes simplex virus type 2 (HG52) variant JH2604 has a 1488 bp deletion which eliminates neurovirulence in mice. J Gen Virol 1989;70:3073–8.[Abstract/Free Full Text]
  25. Taha MY, Clements GB, Brown SM. A variant of herpes simplex virus type 2 strain HG52 with a 1.5 kb deletion in RL between 0 to 0.02 and 0.81 to 0.83 map units is non-neurovirulent for mice. J Gen Virol 1989;70:705–16.[Abstract/Free Full Text]
  26. Markert JM, Medlock MD, Rabkin SD, et al. Conditionally replicating herpes simplex virus mutant, G207 for the treatment of malignant glioma: results of a phase I trial. Gene Ther 2000;7:867–74.[CrossRef][Medline]
  27. Papanastassiou V, Rampling R, Fraser M, et al. The potential for efficacy of the modified (ICP 34.5(–)) herpes simplex virus HSV1716 following intratumoural injection into human malignant glioma: a proof of principle study. Gene Ther 2002;9:398–406.[CrossRef][Medline]
  28. Kim JH, Kim SH, Brown SL, Freytag SO. Selective enhancement by an antiviral agent of the radiation-induced cell killing of human glioma cells transduced with HSV-tk gene. Cancer Res 1994;54:6053–6.[Abstract/Free Full Text]
  29. Khil MS, Kim JH, Mullen CA, Kim SH, Freytag SO. Radiosensitization by 5-fluorocytosine of human colorectal carcinoma cells in culture transduced with cytosine deaminase gene. Clin Cancer Res 1996;2:53–7.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Simpson, G. R.
Right arrow Articles by Coffin, R. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Simpson, G. R.
Right arrow Articles by Coffin, R. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online