Helga Petznek

Specific gene transfer to neurons, endothelial cells and hematopoietic progenitors with lentiviral vectors

We present a flexible and highly specific targeting method for lentiviral vectors based on single-chain antibodies recognizing cell-surface antigens. We generated lentiviral vectors specific for human CD105+ endothelial cells, human CD133+ hematopoietic progenitors and mouse GluA-expressing neurons. Lentiviral vectors specific for CD105 or for CD20 transduced their target cells as efficiently as VSV-G pseudotyped vectors but discriminated between endothelial cells and lymphocytes in mixed cultures. CD133-targeted vectors transduced CD133+ cultured hematopoietic progenitor cells more efficiently than VSV-G pseudotyped vectors, resulting in stable long-term transduction. Lentiviral vectors targeted to the glutamate receptor subunits GluA2 and GluA4 exhibited more than 94% specificity for neurons in cerebellar cultures and when injected into the adult mouse brain. We observed neuron-specific gene modification upon transfer of the Cre recombinase gene into the hippocampus of reporter mice. This approach allowed targeted gene transfer to many cell types of interest with an unprecedented degree of specificity.

Toca 511 gene transfer and 5-fluorocytosine in combination with temozolomide demonstrates synergistic therapeutic efficacy in a temozolomide-sensitive glioblastoma model

Toca 511 (vocimagene amiretrorepvec), an amphotropic retroviral replicating vector (RRV), can successfully and safely deliver a functional, optimized cytosine deaminase (CD) gene to tumors in orthotopic glioma models. This agent, in conjunction with subsequent oral extended-release 5-fluorocytosine (5-FC) (Toca FC), is currently under investigation in patients with recurrent high-grade glioma . Temozolomide (TMZ) with radiation is the most frequently used first-line treatment for patients with glioblastoma, the most common and aggressive form of primary brain cancer in adults. However, subsets of patients with certain genetic alterations do not respond well to TMZ treatment and the overall median survival for patients who respond remains modest, suggesting that combinatorial approaches may be necessary to significantly improve outcomes. We show that in vitro TMZ delays but does not prevent RRV spread, nor interfere with Toca 511+5-FC-mediated cell killing in glioma tumor cells, and in vivo there is no significant hematologic effect from the combination of 5-FC and the clinically relevant dose of TMZ. A synergistic long-term survival advantage is observed in mice bearing an orthotopic TMZ-sensitive glioma after Toca 511 administration followed by coadministration of TMZ and 5-FC. These results provide support for the investigation of this novel combination treatment strategy in patients with newly diagnosed malignant glioma.

Comparative evaluation of preclinical in vivo models for the assessment of replicating retroviral vectors for the treatment of glioblastoma

Despite impressive improvements in neurosurgical techniques, radiation and chemotherapy during the past few years, little progress has been made in the treatment of malignant gliomas. Recently, the efficacy of suicide gene therapy based on replication-competent retroviral (RCR) vectors as delivery vehicles for the therapeutic gene has been described in the treatment of experimental cancer, including gliomas. In this study, we have thus critically evaluated a panel of human and rodent glioma/glioblastoma cell lines (U-87MG, U-118MG, LN-18, LN-229, 8-MG-BA, 42-MG-BA, A-172, T-98G, UVW, C6, 9L, G-26, GL-261, Tu-2449, Tu-9648) with respect to RCR virus vector spread, sensitivity towards the cytosine deaminase (CD)/5-flurocytosine (5-FC)/5-flurouracil (5-FU) suicide system, and orthotopic growth characteristics in mice to identify suitable preclinical animal models for the development of a glioblastoma gene therapy. Rapid virus spread was observed in eight out of nine human cell lines tested in vitro. As expected, only CD-expressing cells became sensitive to 5-FC, due to their ability to convert the prodrug in its toxic form, 5-FU. All LD50 values were within the range of concentrations obtained in human body fluids after conventional antifungal 5-FC administration. In addition, a significant bystander effect was observed in all human glioma cell lines tested. Injection of the RCR vector into pre-established orthotopic mouse tumor xenografts revealed substantial infection and virus spread of tumor tissue from most cell types.

Molecularly Characterised Xenograft Tumour Mouse Models: Valuable Tools for Evaluation of New Therapeutic Strategies for Secondary Liver Cancers

To develop and evaluate new therapeutic strategies for the treatment of human cancers, well-characterised preclinical model systems are a prerequisite. To this aim, we have established xenotransplantation mouse models and corresponding cell cultures from surgically obtained secondary human liver tumours. Established xenograft tumours were patho- and immunohistologically characterised, and expression levels of cancer-relevant genes were quantified in paired original and xenograft tumours and the derivative cell cultures applying RT-PCR-based array technology. Most of the characteristic morphological and immunohistochemical features of the original tumours were shown to be maintained. No differences were found concerning expression of genes involved in cell cycle regulation and oncogenesis. Interestingly, cytokine and matrix metalloproteinase encoding genes appeared to be expressed differentially. Thus, the established models are closely reflecting pathohistological and molecular characteristics of the selected human tumours and may therefore provide useful tools for preclinical analyses of new antitumour strategies in vivo.

Evaluation of a Gene-Directed Enzyme-Product Therapy (GDEPT) in Human Pancreatic Tumor Cells and Their Use as In Vivo Models for Pancreatic Cancer

Background Gene-directed enzyme prodrug therapy (GDEPT) is a two-step treatment protocol for solid tumors that involves the transfer of a gene encoding a prodrug-activating enzyme followed by administration of the inactive prodrug that is subsequently activated by the enzyme to its tumor toxic form. However, the establishment of such novel treatment regimes to combat pancreatic cancer requires defined and robust animal model systems. Methods Here, we comprehensively compared six human pancreatic cancer cell lines (PaCa-44, PANC-1, MIA PaCa-2, Hs-766T, Capan-2, and BxPc-3) in subcutaneous and orthotopical mouse models as well as in their susceptibility to different GDEPTs. Results Tumor uptake was 83% to 100% in the subcutaneous model and 60% to 100% in the orthotopical mouse model, except for Hs-766T cells, which did not grow orthotopically. Pathohistological analyses of the orthotopical models revealed an infiltrative growth of almost all tumors into the pancreas; however, the different cell lines gave rise to tumors with different morphological characteristics. All of the resultant tumors were positive for MUC-1 staining indicating their origin from glandular or ductal epithelium, but revealed scattered pan-cytokeratin staining. Transfer of the cytochrome P450 and cytosine deaminase suicide gene, respectively, into the pancreatic cancer cell lines using retroviral vector technology revealed high level infectibility of these cell lines and allowed the analysis of the sensitivity of these cells to the chemotherapeutic drugs ifosfamide and 5-fluorocytosine, respectively. Conclusion These data qualify the cell lines as part of valuable in vitro and in vivo models for the use in defined preclinical studies for pancreas tumor therapy.

Tropism, intracerebral distribution, and transduction efficiency of HIV- and SIV-based lentiviral vectors after injection into the mouse brain: a qualitative and quantitative in vivo study

Lentiviruses are suitable to transfer potential therapeutic genes into non-replicating cells such as neurons, but systematic in vivo studies on transduction of neural cells within the complete brain are missing. We analysed the distribution of transduced cells with respect to brain structure, virus tropism, numbers of transduced neurons per brain, and influence of the Vpx or Vpr accessory proteins after injection of vectors based on SIVsmmPBj, HIV-2, and HIV-1 lentiviruses into the right striatum of the mouse brain. Transduced cells were found ipsilaterally around the injection canal, in corpus striatum and along corpus callosum, irrespective of the vector type. All vectors except HIV-2SEW transduced also single cells in the olfactory bulb, hippocampus, and cerebellum. Vector HIV-2SEW was the most neuron specific. However, vectors PBjSEW and HIV-1SEW transduced more neurons per brain (means 41,299 and 32,309) than HIV-2SEW (16,102). In the presence of Vpx/Vpr proteins, HIV-2SEW(Vpx) and HIV-1SEW(Vpr) showed higher overall transduction efficiencies (30,696 and 27,947 neurons per brain) than PBjSEW(Vpx) (6636). The distances of transduced cells from the injection canal did not differ among the viruses but correlated positively with the numbers of transduced neurons. The presence of Vpx/Vpr did not increase the numbers of transduced neurons. Parental virus type and the vector equipment seem to influence cellular tropism and transduction efficiency. Thus, precision of injection and choice of virus pseudotype are not sufficient when targeted lentiviral vector transduction of a defined brain cell population is required.