Recombination of replicon and helper RNAs and the emergence of propagation-competent vectors upon Sindbis virus vector production.

Sindbis vectors have shown remarkable antitumor efficacy and tumor-targeting capacity in animal models and hold promise for cancer therapy. Different packaging systems are used to produce propagation-incompetent Sindbis vectors. However, the vectors produced using either DH-BB single helper RNA or split helper RNA can spread in permissive cell cultures. We investigated the mechanisms of vector spreading and show, here, that recombination occurs between the replicon and DH-BB helper RNA, leading to formation of the full-length virus genome. Split helper RNA may not completely prevent wild-type reversion, although the frequency is greatly reduced. Contrary to propagation of Sindbis DH-BB vectors, Sindbis split helper vectors were frequently able to spread without cytopathic effect (CPE), a feature that was linked to wild-type reversion. Our results support the hypothesis that the non-cytopathic local spreading constantly observed with Sindbis split helper vector results from unspecific packaging of helper RNAs into vector particles and co-infection with particles containing replicon and helper RNAs. Several malignant cell lines with defective interferon responses were found to be permissive for non-cytopathic spreading of the Sindbis split helper vector. Interferon-α suppressed the spreading providing a possible option to control the vector.

Dexamethasone-induced apoptosis and up-regulation of Bim is dependent on glycogen synthase kinase-3.

Glucocorticoids are commonly used in the treatment of lymphoid malignancies. In this study, we show that apoptosis induced by dexamethasone (Dex), a synthetic glucocorticoid, was dependent on mitochondria, since overexpression of Bcl-X(L) prevented Dex-induced apoptotic changes. Dominant negative (DN) caspase-9 also prevented Dex-induced apoptotic changes including the loss of mitochondrial membrane potential indicating that caspase-9 controls mitochondrial changes. In addition, we evaluated the role of glycogen synthase kinase (GSK3) in Dex-induced apoptosis. Inhibition of GSK3 attenuated Dex-induced up-regulation of Bim, loss of mitochondrial membrane potential, release of cyt c and DNA fragmentation. These results indicate that GSK3 contributes to Dex-induced apoptosis by controlling up-regulation of Bim.

The involvement of mitochondria and the caspase-9 activation pathway in rituximab-induced apoptosis in FL cells.

Despite the wide use of anti-CD20 antibody rituximab in the cancer treatment of B cell malignancies, the signalling pathways of CD20-induced apoptosis are still not understood. By using dominant negative (DN)-caspase-9 overexpressing follicular lymphoma cells we demonstrated that the activation of caspase-9 was essential for rituximab-mediated apoptosis. The death receptor pathway mediated by caspase-8 activation was not involved in rituximab-mediated apoptosis since overexpression of FLIP(short) or FLIP(long) proteins, inhibitors of caspase-8 activation, could not inhibit rituximab-induced apoptosis. However, the death receptor pathway activation by anti-Fas antibodies showed an additive effect on rituximab-induced apoptosis. The stabilisation of the mitochondrial outer membrane by Bcl-x(L) overexpression inhibited cell death, showing the important role of mitochondria in rituximab-induced apoptosis. Interestingly, the rituximab-induced release of cytochrome c and collapse of mitochondrial membrane potential were regulated by caspase-9. We suggest that caspase-9 and downstream caspases may feed back to mitochondria to amplify mitochondrial disruption during intrinsic apoptosis.

PDTC enables type I TRAIL signaling in type II follicular lymphoma cells.

Based on Bcl-X(L) overexpression studies we identified type I and type II follicular lymphoma cell lines in response to TRAIL. We demonstrate here that either amount of caspase-8 activation or Bid cleavage could not define the dependence on mitochondria. Furthermore, an inhibitor of NF-kappaB, PDTC, enabled TRAIL to activate type I apoptotic pathway in type II cells. However, an inhibitor of IKK did not switch apoptosis to type I pathway in type II cells, indicating that NF-kappaB might not be responsible for the switch.

Oncolytic Semliki forest virus vector as a novel candidate against unresectable osteosarcoma.

Oncolytic viruses are a promising tool for treatment of cancer. We studied an oncolytic Semliki Forest virus (SFV) vector, VA7, carrying the enhanced green fluorescent protein gene (EGFP), as a novel virotherapy candidate against unresectable osteosarcoma. The efficiency and characteristics of the VA7-EGFP treatment were compared with a widely studied oncolytic adenovirus, Ad5Delta24, both in vitro and in vivo. VA7-EGFP resulted in more rapid oncolysis and was more efficient at low multiplicities of infection (MOI) when compared with Ad5Delta24 in vitro. Yet, in MG-63 cells, a subpopulation resistant to the VA7-EGFP vector emerged. In subcutaneous human osteosarcoma xenografts in nude mice treatment with either vector reduced tumor size, whereas tumors in control mice expanded quickly. The VA7-EGFP-treated tumors were either completely abolished or regressed to pinpoint size. The efficacy of VA7-EGFP vector was studied also in an orthotopic osteosarcoma nude mouse model characterized by highly aggressive tumor growth. Treatment with oncolytic SFV extended survival of the animals significantly (P < 0.01), yet none of the animals were finally cured. Sera from SFV-treated mice contained neutralizing antibodies, and as nude mice are not able to establish IgG response, the result points out the role of IgM class antibodies in clearance of virus from peripheral tumors. Furthermore, biodistribution analysis at the survival end point verified the presence of virus in some of the brain samples, which is in line with previous studies demonstrating that IgG is required for clearance of SFV from central nervous system.

Replication competent Semliki Forest virus prolongs survival in experimental lung cancer.

We evaluated the therapeutic potential of the replication competent vector VA7-EGFP, which is based on the avirulent Semliki Forest virus (SFV) strain A7 (74) carrying the EGFP marker gene in an orthotopic lung cancer tumor model in nude mice. We have previously shown that this oncolytic vector destroys tumor cells efficiently in vitro and in vivo (in subcutaneous tumor model). Tumor growth in animals with orthotopically implanted adenocarcinoma cells (A549) were monitored during the study with small animal CT. We show that locally administered virotherapy with VA7-EGFP increased survival rate in experimental lung cancer significantly (p < 0.001) comparable to results obtained with the second generation conditionally replicating adenoviral vector Ad5-Delta24TK-GFP, used for comparison. The limited efficacy in systemically administered oncolytic viruses is the essential problem in oncolytic virotherapy and also in this study we were not able to elicit significant response with systemic administration route. Despite the fact that tumor microenvironment in orthotopic lung cancer is more optimal, viruses failed to home to the tumors and were unable to initiate efficient intratumoral replication. Clearly, the efficacy of virotherapy is influenced by many factors such as the route of virus administration, immunological and physiological barriers and cancer cell-specific features (IFN-responsiveness).

HOX-GFP and WOX-GFP lentivirus vectors for inner ear gene transfer.

CONCLUSION: GFP transgene was expressed in the lining cells of the perilymphatic space. Lentivirus vectors are safe and cause only minimal inflammatory reaction. Transgene products can be delivered into the perilymph by utilizing lentivirus vectors. OBJECTIVES: To analyze the efficiency and safety of lentiviral vectors HOX-GFP and WOX-GFP in intracochlear gene transfer. MATERIALS AND METHODS: Lentivirus vectors were tested for their transduction efficiency in vivo in CD-1 mice. Half of the animals were pretreated with kanamycin. Lentivirus vector or saline (1 microl) was injected into the inner ear. All the animals were sacrificed 14 days after the surgery and the cochleae and selected organs were analyzed immunohistochemically. RESULTS: HOX-GFP and WOX-GFP expression was restricted to the lining cells of the scala tympani and scala vestibuli. No GFP expression was seen in the organ of Corti or the spiral ganglion. Aminoglycoside treatment had no effect on the expression of these vectors. The distant spread of lentivirus vectors was minimal; only the liver of one animal showed some GFP expression. Inflammatory reaction caused by these vectors was mild. Few inflammatory cells were found in the perilymphatic space of the cochlea and in the vestibular organ.

Evaluation of cancer virotherapy with attenuated replicative Semliki forest virus in different rodent tumor models.

Semliki Forest virus (SFV) is one of the latest candidates for a virotherapeutic agent against cancer, and recent studies have demonstrated its efficacy in tumor models. In the present study, we examined the antitumor efficacy of an avirulent SFV strain A7(74) and its derivative, a replication-competent SFV vector VA7-EGFP, in a partially immunodeficient mouse tumor model (subcutaneous A549 human lung adenocarcinoma in NMRI nu/nu mouse) and in an immunocompetent rat tumor model (intracranial BT4C glioma in BDIX rat). When subcutaneous mouse tumors were injected 3 times with VA7-EGFP, intratumorally treated animals showed almost complete inhibition of tumor growth, while systemically treated mice displayed only delayed tumor growth (intravenous injection) or no response at all (intraperitoneal injection). This was at least partially due to a strong type I interferon (IFN) response in the tumors. The animals did not display any signs of abnormal behavior or encephalitis, even though SFV-positive foci were detected in the brain after the initial blood viremia. Intracranial rat tumors were injected directly with SFV A7(74) virus and monitored with magnetic resonance imaging. Tumor growth was significantly reduced (p < 0.05) with one virus injection, but the tumor size continued to increase after a lag period and none of the treated animals survived. Three virus injections or T-cell suppression with dexamethasone did not significantly improve treatment efficacy. It appeared that the local virotherapy induced extensive production of neutralizing anti-SFV antibodies that most likely contributed to the insufficient treatment efficacy. In conclusion, we show here that SFV A7(74) is a potential oncolytic agent for cancer virotherapy, but major immunological hurdles may need to be overcome before the virus can be clinically tested.

Type I interferon response against viral and non-viral gene transfer in human tumor and primary cell lines.

BACKGROUND: Type I interferon (IFN-alpha/beta) response is one of the major host defence mechanisms against viruses. Some recent reports suggest that IFNs may interfere with the efficacy of both non-viral and virus-vector-mediated therapeutic gene transfer. METHODS: The type I IFN response upon different gene transfer methods in human tumor and primary cell lines was studied by analysing IFN-beta mRNA expression, secretion of type I IFNs and accumulation of IFN-alpha/beta-induced MxA protein (myxovirus resistance protein A). RESULTS: Infection with avirulent Semliki Forest virus A7[74] induced MxA protein accumulation and increased the IFN-beta mRNA level, whereas none of the studied virus vectors (adenovirus, CRAd, lentivirus or AAV) induced IFN response. However, plasmid DNA induced the accumulation of MxA protein when transfected with several commercial transfection reagents. RNA transfection appeared to be an efficient inducer of type I IFN response: replicating alphaviral RNA, eukaryotic total RNA, or mRNA all induced both MxA protein accumulation and IFN-beta expression. siRNA transfection failed to induce MxA response. CONCLUSIONS: The non-viral gene transfer methods have gained more interest in recent years due to their better safety profiles when compared to their viral counterparts. However, the efficiency of non-viral gene transfer is well below those reached by viral vector systems. The type I interferon response induced by non-viral methods may in part contribute to this inefficiency, while most currently used viral gene transfer vectors fail to induce or are able to suppress type I IFN response.

Transcriptional targeting of virus-mediated gene transfer by the human hexokinase II promoter.

The efficacy of the most commonly used form of suicide gene therapy, the HSV-TK/GCV method, utilizing herpes simplex virus thymidine kinase (HSV-TK) and antiviral drug ganciclovir (GCV) has been demonstrated in clinical trials. However, safer delivery of the therapeutic gene and more controlled regulation of the transgene expression, the essential prerequisites for successful therapeutic use, are still needed. We describe improved suicide gene therapy against cancer through transcripitional targeting by a strong and selective tumor-specific human hexokinase II promoter (hHKII). We examined the targeting properties of the human hHKII promoter in different human non-small cell lung cancer (NSCLC) and other human cancer cell lines using self-inactivating, VSV-G pseudotyped lentiviral vector. To confirm accurate transcriptional targeting of the hHKII promoter, the lack of transgene expression was verified in human primary bronchial epithelial and bronchial fibroblast cells. Furthermore, tissue-specific expression of the promoter was confirmed using transgenic mouse lines carrying the hHKII promoter driven luciferase reporter gene. We also tested the efficacy of the HSV-TK/GCV suicide gene therapy with the hHKII targeted lentiviral vector to NSCLC cells. Our results show that the hHKII promoter is strongly expressed in cancer cells. The targeted vector with the shortest hHKII promoter fragment (352 bp) appeared to have the best targeting properties because it efficiently governed the expression of the therapeutic gene in cancer cell lines, especially in certain non-small cell lung cancer cell lines, the transgene expression in human primary cells was virtually undetectable, and expression of the proximal hHKII promoter in transgenic mice was very low in most tissues. Also, the anti-cancer efficacy of HSV-TK/GCV therapy with the hHKII-targeted vector was comparable to that obtained with the control vector that utilized a commonly used constitutive promoter from the human elongation factor 1 alpha (hEF1alpha) gene. In conclusion, the transcriptionally targeted lentivirus vector with hHKII promoter can successfully direct HSV-TK/GCV suicide gene therapy to non-small cell lung cancer and other tumor cell types. These results warrant further studies with orthotopic animal tumor models and primary human cancer material.

A conditionally replicative adenovirus that codes for a TK-GFP fusion protein (Ad5Delta24TK-GFP) for evaluation of the potency of oncolytic virotherapy combined with molecular chemotherapy.

The utility of conditionally replicative adenoviruses (CRAds) in cancer gene therapy is based on their ability to destroy tumor cells by oncolysis. However, in order to achieve adequate therapeutic response, CRAds have to spread through the tumor tissue by replication. Thus, to study the potency of these viruses to replicate and penetrate in tumors, we created a herpes simplex virus type 1 thymidine kinase-green fluorescent protein fusion gene (TK-GFP) encompassing CRAd (Ad5Delta24TK-GFP), whose tumor selectivity is mediated by a retinoblastoma (Rb)-binding site mutation. In addition, we evaluated the oncolytic efficacy of Ad5Delta24TK-GFP in combination with the TK/ganciclovir (GCV) system. Based on our results, Ad5Delta24TK-GFP replicates in cancer cells resulting in oncolysis and can efficiently penetrate into tumors. Additionally, the combination of GCV with Ad5Delta24TK-GFP augmented cell death in vitro but this was not observed in vivo: tumor growth was significantly reduced by oncolysis when compared to non-replicative virus (p<0.001), but administration of GCV did not significantly enhance oncolysis. This suggests that in certain conditions, TK/GCV-mediated cell killing may be counterproductive to replication and oncolysis, which on the other hand might be useful feature for clinical trials in case of replication-associated toxicity.

Polyamine-regulated unproductive splicing and translation of spermidine/spermine N1-acetyltransferase.

Spermidine/spermine N1-acetyltransferase (SSAT), the rate-controlling enzyme in the interconversion of spermidine and spermine, is regulated by polyamines and their analogs at many levels of gene expression. Recently, SSAT pre-mRNA has been shown to undergo alternative splicing by inclusion of an exon that contains premature termination codons. In the present study, we show that alterations in the intracellular polyamine level resulted in a change in the relative abundance of SSAT transcripts. Addition of polyamines or their N-diethylated analogs reduced the amount of the variant transcript, whereas polyamine depletion by 2-difluoromethylornithine or MG-132 enhanced the exon inclusion. Experiments performed with protein synthesis inhibitors and siRNA-mediated down-regulation of Upf1 protein verified that the variant transcript was degraded by nonsense-mediated mRNA decay (NMD). Interestingly, several proteins have been shown to regulate their expression by alternative splicing-coupled NMD, termed regulated unproductive splicing and translation (RUST). Our present results suggest that in the case of SSAT, RUST is mediated by polyamines, and this system functions to fine-tune the polyamine metabolism.

hCAR-EGFP fusion receptor in human follicular lymphoma B cells - a model for adenoviral gene therapy for B cell malignancies.

Adenovirus-mediated gene therapy for hematopoietic malignancies, especially those derived from B cells, is difficult due to systemic nature of these diseases. More importantly, most tumor cells derived from B cell lineage express a very low level of the adenovirus receptor hCAR; thus, warranting the design of adenoviral vectors with high affinity to abundant B cell surface molecules. To mimic this approach and to test the validity of adenoviral vectors in gene therapy of disseminated malignancies, we created an hCAR-expressing follicular lymphoma B cell line. The cell line was generated with the aid of a lentivirus vector carrying a novel fusion gene with EGFP replacing the cytoplasmic domain of hCAR. After verifying that this cell line was expressing the hybrid receptor in a correct manner and enrichment of the bright EGFP positive population, the cells were transduced with adenoviruses expressing the red fluorescent protein DsRed2. It was shown that regular transduction with a low viral dose (1 pfu/cell) increased the gene transfer rate by a factor of 5. Furthermore, experiments with adenovirus vector carrying the HSV-TK-GFP transgene demonstrated that the modified follicular lymphoma B cells became sensitive to ganciclovir while the parental cells remained virtually resistant to this form of gene therapy. In summary, we show here with this simple model system that adenoviral gene therapy of B cell malignancies is possible provided that correct receptors for adenovirus attachment are present on the surface of the target cells. Thus, our results warrant further modifications of adenovirus capsid to obtain vectors with specific affinity to B cell epitopes.

Polyamine depletion and cell cycle manipulation in combination with HSV thymidine kinase/ganciclovir cancer gene therapy.

We have shown earlier that polyamine biosynthesis inhibition is accompanied by cell cycle alterations that can be utilized to enhance the efficacy of herpes simplex virus thymidine kinase - ganciclovir (HSV-TK/GCV) cancer gene therapy. In the present study, we asked 1) can the activated polyamine catabolism instead of biosynthesis inhibition be utilized to enhance the efficacy of HSV-TK/GCV gene therapy, and 2) can other known cell cycle inhibitors be used to make tumor cells more sensitive to this form of gene therapy? We show, using rat (9L) and human (U251-MG) glioma cell populations with 15% of HSV-TK-positive cells that DENSPM-induced activation of polyamine catabolism caused a profound polyamine deprivation in U251-MG cells, but there were no associated cell cycle effects in these cells. Consequently, we did not see any enhancement of the HSV-TK/GCV system. Aphidicolin, hydroxyurea, mimosine and resveratrol, but not lovastatin induced an apparent cell cycle arrest, followed by an intense but transient increase of the S phase cells after removal of the drug. This effect was shown to potentiate the HSV-TK/GCV cytotoxicity to some extent, especially in 9L cells and when the GCV treatment was started 0-24 h before the drug treatment. However, the enhancement was weaker than observed earlier with DFMO-induced cell cycle arrest and a considerable degree of the effect appeared to result from the growth-inhibitory actions of the drugs. In summary, we demonstrate that polyamine deprivation via DENSPM action is not associated with cell cycle effects and is not sufficient to cause enhancement of the HSV-TK/GCV system. Also, drugs with a rapid effect to the cell cycle are weak boosters of the HSVTK/GCV gene therapy, thus being less useful than DFMO for enhancement of this gene therapy form in animal studies and clinical trials.

In vivo enhancement of herpes simplex virus thymidine kinase/ganciclovir cancer gene therapy with polyamine biosynthesis inhibition.

We have earlier demonstrated that inhibition of polyamine biosynthesis with difluoromethylornithine (DFMO) can be used to enhance the cytotoxicity of herpes simplex virus thymidine kinase/ganciclovir (HSV-TK/GCV) gene therapy in different tumor cell lines. Here, the utility of this treatment combination was tested in vivo in a nude mouse tumor model. First, the effect of DFMO was verified by treating mice bearing subcutaneous 9L rat glioma tumors with 2% DFMO in drinking water. The drug treatment induced almost complete suppression of ornithine decarboxylase activity, and as a result, a strong decrease in intratumoral putrescine and spermidine concentrations, which were normalized 4 days after drug removal. Consequently, the tumors displayed a significant reduction in the proliferation activity that was increased to 20% higher than the normal level at day 4 and returned to normal level 7 days after DFMO removal. Next, 9L tumors with 30% of TK-GFP fusion gene positive cells were induced and the animals were given DFMO and GCV in 2 treatment schemes, with the drug administration periods overlapping either 5 or 2 days. The analysis of tumor size at the end of the treatment revealed that DFMO can enhance HSV-TK/GCV cytotoxicity when the overlap between DFMO and GCV was 5 days, but the result was not significant. However, the 2-day overlap scheme yielded a significantly (p < 0.05, ANOVA) enhanced antitumor effect. In conclusion, the data here confirms that a novel combination of 2 clinically relevant treatment modalities, polyamine deprivation and HSV-TK/GCV suicide gene therapy, can be used synergistically in vivo.

Characterization of transgenic mice with the expression of phenylalanine hydroxylase and GTP cyclohydrolase I in the skin.

Phenylketonuria (PKU) is a metabolic disease causing increased levels of phenylalanine in blood and body fluids. Circulating phenylalanine is normally cleared by phenylalanine hydroxylase (PAH) expressed in the liver. The aim of this study is to exploit the skin as a 'metabolic sink' removing phenylalanine from the blood. We have previously showed that the overexpression of PAH and GTP cyclohydrolase I (GTP-CH), the rate-limiting enzyme in the synthesis of the cofactor for PAH, leads to high levels of phenylalanine clearance in primary human keratinocytes. In this study, we have investigated the 'metabolic sink' strategy in an in vivo model by developing three lines of transgenic mice expressing PAH and GTP-CH in various layers of the skin. The promoters used were keratin 14 (K14), involucrin (INV) and a truncated variant of Keratin 1 (K1). The mice were crossbred to a mouse model of human PKU, the PAH(enu2) mouse, in order to obtain mice that do not express PAH in the liver and the kidney. Transgenic mice containing the INV and K14 promoters expressed PAH and GTP-CH in the epidermis. However, the K1 promoter did not lead to detectable gene expression. Analysis of the mice showed that no phenotypic effect was observed in mice expressing PAH and GTP-CH from the INV promoter. However, low level of phenylalanine clearance was observed in mice expressing PAH and GTP-CH from the K14 promoter, suggesting that the skin can be genetically engineered to function as a 'metabolic sink'.

HIV-1 TAT protein transduction domain mediates enhancement of enzyme prodrug cancer gene therapy in vitro: a study with TAT-TK-GFP triple fusion construct.

Protein transduction domain (PTD) from HIV-1 TAT protein has been reported to translocate across the mammalian cell membrane, also as a part of fusion proteins. However, the true nature of TAT-mediated intercellular spreading is still under debate because it has been claimed to be a fixation artifact. To study the spreading of TAT fusion proteins and their potency to enhance thymidine kinase/ ganciclovir (HSV-TK/GCV) cancer gene therapy, we constructed a novel triple fusion protein containing TAT PTD, HSV-TK and green fluorescent protein (TAT-TK-GFP). This fusion protein has three functional domains in the same polypeptide, allowing reliable determination of the relationship between transduction rate and cell killing efficiency. TAT-TK-GFP was cloned into a lentivirus vector and used for analyses of TAT-mediated protein translocation and enhancement of HSV-TK/GCV cytotoxicity. The triple fusion protein was expressed correctly in vitro, but cell-to-cell translocation was not observed in rat glioma cells (BT4C). However, TAT-TK-GFP made BT4C and SKOV3.ip1 (human ovarian carcinoma) cells significantly more sensitive to ganciclovir than TK-GFP, whereas the effect in PC-3 human prostate carcinoma cells was more subtle. It was also observed that growth in lower serum concentration (2.5-5%) abolished the enhancement in BT4C cells, suggesting that high proliferation rate is one of the factors that contribute to TAT PTD-mediated enhancement of cytotoxicity. In summary, our results indicate that TAT PTD fusion proteins do not translocate intercellularly at detectable levels, but enhancement of the HSV-TK/GCV cytotoxicity can be detected in rat and human tumor cell lines in vitro.

Properties of Sindbis virus vectors produced with a chimeric split helper system.

We have evaluated a chimeric, two-component Sindbis virus packaging system. As expected, use of this combination of two modified helper RNA species prevented formation of infection competent Sindbis viruses as analyzed by serial passaging. We observed, however, that vectors produced using this method were able to spread in BHK cell cultures and formed clusters of transgene positive cells that did not display cytopathic effects for up to 3 days post-transduction. Formation of spreading Sindbis virus vectors required only one of the helper components--the chimera with a deleted Ross River virus capsid and the Sindbis virus envelope glycoproteins. Spreading was also demonstrated in two rat glioma cell lines, 9L and BT4C, showing that this phenomenon was not limited to BHK cells. Our results warrant further characterization of split helper Sindbis virus vectors and imply their utility in gene therapy approaches where spreading of transgene expression and consequently high gene transfer rate could be beneficial.

VP22 does not significantly enhance enzyme prodrug cancer gene therapy as a part of a VP22-HSVTk-GFP triple fusion construct.

BACKGROUND: VP22 is a herpes simplex virus type 1 (HSV-1) tegument protein that has been suggested to spread from cell to cell, alone or as a part of fusion proteins. Creating controversy, some reports indicate that VP22 cannot facilitate significant intercellular spreading. To study the capacity of VP22 to cause spreading and enhance thymidine kinase/ganciclovir cancer gene therapy, we constructed a novel triple fusion protein containing VP22, HSV thymidine kinase and green fluorescent protein (VP22-Tk-GFP). This fusion protein has three functional domains in the same polypeptide, thus making it possible to reliably compare the causality between transduction rate and cell killing efficiency in vitro and in vivo. METHODS: VP22-Tk-GFP was cloned into lenti- and adenoviral vectors and used for expression studies, analyses for VP22-mediated protein spreading, and to study the effect of VP22 to thymidine kinase/ganciclovir-mediated cytotoxicity. The function of VP22-Tk-GFP was also investigated in vivo. RESULTS: The triple fusion protein was expressed correctly in vitro, but intercellular trafficking was not observed in any of the studied cell lines. However, under certain conditions, VP22-Tk-GFP sensitized cells more efficiently to ganciclovir than Tk-GFP. In vivo there was a trend for increased inhibition of tumor growth with VP22-Tk-GFP when ganciclovir was present, but the difference with Tk-GFP was not statistically significant. CONCLUSIONS: Based on our results, VP22 fusion proteins do not seem to traffic intercellularly at detectable levels in most tumor cell types. Even though VP22 enhanced cytotoxicity in one cell line in vitro, the effect in vivo was modest. Therefore, our results do not support the utility of VP22 as an enhancer of enzyme prodrug cancer gene therapy.

Cancer cells as targets for lentivirus-mediated gene transfer and gene therapy.

Lentiviruses have been used as gene transfer vectors for almost 10 years and their utility has been demonstrated in a variety of different applications. However, their value in cancer gene therapy has not been studied thoroughly. Here we show that VSV-G pseudotyped HIV-1-based lentiviruses are efficient vectors for human tumor cells in vitro and in vivo. Lentiviral gene transfer efficiency was demonstrated by transducing 42 different cell lines, representing 10 different human tumor types. It was shown that most of the cell lines were good or excellent targets for lentiviral transduction, allowing 50-95% gene transfer efficiency. These results were comparable to those obtained with an E1/E3 deleted, serotype 5 adenovirus vector. Analysis of lentivirus vector structure revealed that virus particles devoid of HIV-1 accessory proteins appeared to be more efficient, but the presence of enhancing elements cPPT and WPRE did not play a major role in transduction efficiency to four different human tumor cell lines. However, their effect on the gene expression level in these cells was apparent. To examine the impact of lentiviral gene expression level on suicide gene therapy approach, human osteosarcoma cells were transduced with lentivirus- or adenovirus vectors carrying the fusion gene HSV-TK-GFP and exposed to ganciclovir. Cell viability analysis after the treatment revealed that both vector types induced similar level of cytotoxicity, suggesting that lentiviral expression of a suicide gene is adequate for tumor cell destruction. Finally, in vivo transduction studies with subcutaneous tumors showed that lentivirus vectors can yield similar gene transfer efficiency than adenovirus vector, despite three orders of magnitude lower titer of the lentiviral preparation. In conclusion, these data show that lentiviruses are efficient gene transfer vehicles for human tumor cells and justify their use in further preclinical cancer gene therapy studies.