Imaging immediate-early and strict-late promoter activity during oncolytic herpes simplex virus type 1 infection and replication in tumors

An increasing number of oncolytic viruses have been developed and studied for cancer therapy. In response to needs for non-invasive monitoring and imaging of oncolytic virotherapy, several different approaches, including a positron emission tomography-based method, a method using secreted marker peptides, and optical imaging-based methods, have been reported. Among these modalities, we utilized the luciferase-based bioluminescent assay/imaging systems to determine the kinetics and dynamics of a productive viral infection. The replication cycle of herpes simplex virus type 1 (HSV-1) is punctuated by a temporal cascade of three classes of viral genes: immediate-early (IE), early (E) and late (L) genes. U(L)39- and gamma(1)34.5-deleted, replication-conditional HSV-1 mutants that express firefly luciferase under the control of the IE4/5 or strict-late gC promoters were generated. These oncolytic viruses were examined in cultured cells and a mouse tumor model. IE promoter- and strict-late promoter-mediated luciferase expression was confirmed to indicate viral infection and replication, respectively. Incorporation of a strict-late promoter-driven luciferase cassette into oncolytic HSV-1 vectors would be useful for assessing tumor oncolysis in preclinical tumor treatment studies.     

Identification of novel insertion sites in the Ad5 genome that utilize the Ad splicing machinery for therapeutic gene expression.

Therapeutic transgene expression from oncolytic viruses represents one approach to increasing the effectiveness of these agents as cancer therapeutics. In the case of the oncolytic adenovirus (Ad), however, the genomic packaging capacity is constrained. To address this, we explored whether a transposon-based system could identify sites in the viral genome where endogenous Ad promoters could drive transgene expression via splicing and still maintain the replication capacity of the virus. Using GFP as a reporter gene and an E3-deleted Ad genome as a target, we tested three splicing signals. RACE analysis confirmed that gene expression from the GFP-expressing Ads occurs via splicing and traced expression to the Ad major late promoter (MLP). Replacement of the GFP transposon by an equivalent splice acceptor-luciferase expression cassette in the same orientation confirmed that substitute transgenes are also expressed via splicing from the MLP. Interestingly, insertion of the substitute transgene in the opposite orientation also resulted in expression that, in some cases, originated from within the ITR region of the viral genome. In summary, splice acceptor sequences can be used to control transgene expression from endogenous Ad promoters and this represents a genomically economical approach to arming oncolytic Ads.     

Valproic acid, a histone deacetylase inhibitor, is an antagonist for oncolytic adenoviral gene therapy.

Oncolytic adenoviruses preferentially replicate in and lyse tumor cells. However, their application to cancer gene therapy has been complicated by the low levels of coxsackie and adenovirus receptor (CAR) expressed in many solid tumors. Histone deacetylase inhibitors (HDACIs) significantly up-regulate CAR expression in tumor cells and have additional antineoplastic activities. Therefore, there is a clear rationale for the combination of HDACIs and oncolytic adenoviral gene therapy. We present evidence that HDACI treatment significantly inhibits adenoviral replication, viral burst, and tumor cell kill. Valproic acid (VPA), a well-established HDACI, inhibits adenoviral replication late in the viral life cycle. We hypothesized that VPA induction of the cell-cycle-regulating protein p21(WAF1/CIP1) may be partly responsible for this activity. We demonstrate that p21(WAF1/CIP1) expression alone limits viral replication and decreases viral titers in different cancer cell models. We also demonstrate that VPA and replicating adenovirus mutually inhibit each other's ability to kill cells, independent of p21(WAF1/CIP1) expression. These results not only identify the importance of p21(WAF1/CIP1) in the biology of adenoviral replication, but also suggest that oncolytic adenoviral gene therapy will be inhibited rather than enhanced by VPA (HDACI) treatment.     

Development of a transposon-based approach for identifying novel transgene insertion sites within the replicating adenovirus.

Therapeutic gene delivery from an oncolytic adenovirus (Ad) is one approach to enhancing the potency of Ad-based virotherapies for cancer. To identify therapeutic transgene insertion sites compatible with the replicating virus, a methodology that broadly scans the viral genome is needed. To address this we modified a transposon (Tn7)-based in vitro transposition system to take advantage of its nonprejudiced scanning ability to identify insertion sites compatible with viral replication. Using this system with a plasmid containing an E3-deleted Ad5, we identified several unique sites for promoter-based expression cassette insertions within the Ad genome. The transposon-based expression cassette is bounded by PmeI restriction endonuclease sites unique to the transposon, making expression cassette substitutions easy to perform. Additional expression cassettes containing different promoters and reporter genes were substituted into two of the newly identified transgene insertion sites. The results suggest that the ease and orientation of expression cassette substitution depend on both the insertion site location and the promoter and gene of the replacement expression cassette. These studies establish the transposon-based system as an efficient approach to scanning the Ad genome and identifying insertion sites compatible with viral replication and represents a powerful tool for the development of armed therapeutic viruses for cancer.     

Lobucavir is phosphorylated in human cytomegalovirus-infected and -uninfected cells and inhibits the viral DNA polymerase.

Lobucavir (LBV) is a deoxyguanine nucleoside analog with broad-spectrum antiviral activity. LBV was previously shown to inhibit herpes simplex virus (HSV) DNA polymerase after phosphorylation by the HSV thymidine kinase. Here we determined the mechanism of action of LBV against human cytomegalovirus (HCMV). LBV inhibited HCMV DNA synthesis to a degree comparable to that of ganciclovir (GCV), a drug known to target the viral DNA polymerase. The expression of late proteins and RNA, dependent on viral DNA synthesis, was also inhibited by LBV. Immediate-early and early HCMV gene expression was unaffected, suggesting that LBV acts temporally coincident with HCMV DNA synthesis and not through cytotoxicity. In vitro, the triphosphate of LBV was a potent inhibitor of HCMV DNA polymerase with a Ki of 5 nM. LBV was phosphorylated to its triphosphate form intracellularly in both infected and uninfected cells, with phosphorylated metabolite levels two- to threefold higher in infected cells. GCV-resistant HCMV isolates, with deficient GCV phosphorylation due to mutations in the UL97 protein kinase, remained sensitive to LBV. Overall, these results suggest that LBV-triphosphate halts HCMV DNA replication by inhibiting the viral DNA polymerase and that LBV phosphorylation can occur in the absence of viral factors including the UL97 protein kinase. Furthermore, LBV may be effective in the treatment of GCV-resistant HCMV.     

Oncolytic HSV armed with platelet factor 4, an antiangiogenic agent, shows enhanced efficacy.

Oncolytic herpes simplex viruses (HSV) have emerged as a promising platform for cancer therapy. However, efficacy as single agents has thus far been unsatisfactory. Tumor vasculature is critical in supporting tumor growth, but successful antiangiogenic approaches often require maintaining constant levels of antiangiogenic products. We hypothesized that oncolytic HSV has the potential to destroy tumor vasculature and that this effect can be enhanced by combination with antiangiogenic gene transfer. We examined the strategy of arming oncolytic HSV with an antiangiogenic transgene, platelet factor 4 (PF4). The PF4 transgene was inserted into oncolytic HSV G47Delta utilizing a bacterial artificial chromosome construction system. Whereas bG47Delta-empty showed robust cell killing and migration inhibition of proliferating endothelial cells (HUVEC and Py-4-1), the effect was further enhanced by PF4 expression. Importantly, enhanced potency did not impede viral replication. In vivo, bG47Delta-PF4 was more efficacious than its nonexpressing parent bG47Delta-empty at inhibiting tumor growth and angiogenesis in both human U87 glioma and mouse 37-3-18-4 malignant peripheral nerve sheath tumor models. Enhancing the antiangiogenic properties of oncolytic HSV through the expression of antiangiogenic factors such as PF4 is a powerful new strategy that targets both the tumor cells and tumor vasculature.     

Functional analysis of various promoters in lentiviral vectors at different stages of in vitro differentiation of mouse embryonic stem cells.

Given the therapeutic potential offered by embryonic stem (ES) cells, it is critical to optimize stable gene delivery and expression at different developmental stages of ES cell differentiation. Here, we systematically analyzed lentiviral vectors containing the following promoters: the human elongation factor 1alpha (EF1alpha) promoter, the human cytomegalovirus (CMV) immediate early region enhancer-promoter, the composite CAG promoter (consisting of the CMV immediate early enhancer and the chicken beta-actin promoter), the human phosphoglycerate kinase 1 (PGK) promoter, the murine stem cell virus (MSCV) long terminal repeat (LTR), or the gibbon ape leukemia virus (GALV) LTR. Our results show that the EF1alpha promoter directed robust transgene expression at every stage of mouse ES cell differentiation, whereas the CMV promoter drove transgene expression only during late stages. Similarly, the CAG and PGK promoters drove transgene expression at a significant level only during late stages. The MSCV LTR and the GALV LTR exhibited much lower promoter activities at all stages. Interestingly, mouse ES cells transduced with the EF1alpha promoter-containing lentiviral vector lost most of their transgene expression during in vitro differentiation to neural precursors and neuronal cells. Our results demonstrate that different cellular and viral promoters exhibit very distinct and dynamic properties not only in terms of promoter strength but also with respect to differentiation stage-specific activity.     

Highly efficient and carcinoma-specific adenoviral replication restricted by the EGP-2 promoter.

Although some successes have been reported using adenoviral vectors for the treatment of cancer, adenoviral cancer gene therapy is still hampered by the lack of sufficient tumor cell killing. To increase the efficiency, adenoviruses have been modified to replicate specifically in tumor tissues by using tumor specific promoters controlling genes essential for adenoviral replication. However, many conditionally replicating adenoviral vectors replicate in one tumor type only, which limits their application. The epithelial glycoprotein-2 (EGP-2) promoter is active in a broad variety of carcinomas, the most common type of cancer. We utilized this promoter to restrict adenoviral replication. In this report we demonstrate that the potency of the replication-competent adenovirus AdEGP-2-E1 to specifically lyse EGP-2 positive cells is comparable to wild-type adenovirus (AdWT). In addition, we show that in vivo AdEGP-2-E1 replicates as efficient as AdWT in EGP-2 positive tumor cells. On the contrary, in EGP-2 negative cell lines as well as in primary human liver samples, the replication was attenuated up to 4-log in comparison to wild-type virus. This report clearly shows the potency of the EGP-2 promoter to mediate highly efficient and specific adenoviral replication for carcinoma gene therapy.