Jan J. Cornelis

cis Requirements for the Efficient Production of Recombinant DNA Vectors Based on Autonomous Parvoviruses

The replication of viral genomes and the production of recombinant viral vectors from infectious molecular clones of parvoviruses MVMp and H1 were greatly improved by the introduction of a consensus NS-1 nick site at the junction between the left-hand viral terminus and the plasmid DNA. Progressive deletions of up to 1600 bp in the region encoding the structural genes as well as insertions of foreign DNA in replacement of those sequences did not appreciably affect the replication ability of the recombinant H1 virus genomes. In contrast, the incorporation of these genomes into recombinant particles appeared to depend on in cis-provided structural gene sequences. Indeed, the production of H1 viral vectors by cotransfection of recombinant clones and helper plasmids providing the structural proteins (VPs) in trans, drastically decreased when more than 800 bp was removed from the VP transcription unit. Furthermore, titers of viral vectors, in which most of the VP-coding region was replaced by an equivalent-length sequence consisting of reporter cDNA and stuffer DNA, were reduced more than 50 times in comparison with recombinant vectors in which stuffer DNA was not substituted for the residual VP sequence. In addition, viral vector production was restricted by the overall size of the genome, with a mere 6% increase in DNA length leading to an approximately 10 times lower encapsidation yield. Under conditions fulfilling the above-mentioned requirements for efficient packaging, titers of virus vectors from improved recombinant molecular DNA clones amounted to 5 x 10(7) infectious units per milliliter of crude extract. These titers should allow the assessment of the therapeutic effect of recombinant parvoviruses expressing small transgenes in laboratory animals.

Highly efficient transduction and expression of cytokine genes in human tumor cells by means of autonomous parvovirus vectors; generation of antitumor responses in recipient mice.

The possible use of recombinant autonomous parvoviruses as vectors to efficiently express therapeutic cytokines in human tumor cells was evaluated in vitro and in vivo. The parvovirus H1 was used to generate recombinant viruses (rH1) that carried transgenes encoding either human interleukin 2 (IL-2) or monocyte chemotactic protein 1 (MCP-1), in replacement of part of the capsid genes. Such rH11 viruses have been shown to retain in vitro the intrinsic oncotropic properties of the parental virus. On infection with the recombinant viruses at an input multiplicity of 1 replication unit (RU) per cell, HeLa cultures were induced to release 4-10 microg of cytokine per 10(6) cells over a period of 5 days. The expression of the rH1-transduced human cytokine/chemokine could also be detected in tumor material recovered from nude mice that had been subcutaneously engrafted with in vitro-infected HeLa cells. The formation of tumors from HeLa xenografts was reduced by 90% compared with wild-type or mock-infected cells as a result of cells preinfected with IL2-expressing virus at an input multiplicity as low as 1 RU per cell. Tumors arising from HeLa cells infected with transgene-free or MCP1-expressing vectors or with wild-type H1 virus were not rejected at this virus dose. Tumors infected with rH1/IL-2 virus displayed markers indicative of their infiltration with NK cells in which the cytocidal program was activated, whereas little NK activity was detected in wild-type virus or mock-infected tumors. Altogether, these data show that the IL-2 expressing H1 vector was a more potent antineoplastic agent than the parental virus, and point to the possible application of recombinant autonomous parvoviruses toward therapy of some human tumors.

Parvovirus H-1 infection of human glioma cells leads to complete viral replication and efficient cell killing

The extremely poor prognosis of malignant gliomas requires the investigation of other than standard therapies, i.e., the application of oncolytic viruses. In our study, we evaluated the effects of the oncosuppressive parvovirus H‐1 on different established glioblastoma cell lines of rat and human origin and on short‐term/low‐passage cultures of human glioblastoma cells. We observed an efficient and dose‐dependent killing of all glioma cell cultures at low multiplicities of infectious particles (MOI) per cell. Southern blot analysis of viral DNA amplification, RT‐PCR analysis of viral RNA expression and Western blot analysis of the expression of viral structural (VP‐1/VP‐2) and nonstructural (NS‐1) proteins demonstrated the biosynthesis of these viral macromolecular components in all of the cultures. Moreover, all the glioma cells were proficient for the production of infectious H‐1 virus particles. The amount of virus production differed between a several fold increase of the input virus titer in most of the short‐term/low‐passage cultures up to 1,000‐fold in one short‐term glioma and in the rat cells. Glioma cells lines and, more importantly, short‐term/low‐passage cultures of human glioblastomas were found to be highly susceptible target cells for H‐1 virus mediated cytotoxicity. The formation of fully infectious progeny particles in infected glioma cells offers the chance for the induction of secondary rounds of infection resulting in an advanced cytotoxic effect. These advantageous characteristics of H‐1 virus infection of glioma cells, combined with the known low toxicity of H‐1 virus in nontransformed cells, make parvovirus H‐1 a promising candidate for oncolytic glioma therapy.

Transduction of human MCP-3 by a parvoviral vector induces leukocyte infiltration and reduces growth of human cervical carcinoma cell xenografts

The oncosuppressive properties of some autonomous parvoviruses such as H‐1 virus, together with their low pathogenicity, make them attractive vectors for tumor‐directed gene therapy. Indeed, it was recently shown that these viruses became endowed with an enhanced oncosuppressive activity after they had been engineered to deliver a recognized therapeutic transgene. This prompted us to use a parvoviral vector to analyse the antineoplastic capacity of MCP‐3 (monocyte chemotactic protein‐3), a CC chemokine which has a broad spectrum of target cells, and can thus be considered to be a promising candidate for cancer treatment.

Potentiation of a recombinant oncolytic parvovirus by expression of Apoptin.

The oncotropic and oncolytic behaviors of certain autonomous rodent parvoviruses make them promising vectors for anticancer gene therapies. However, these parvoviruses are often not potent enough to kill all tumor cells equally well. With the aim of enhancing the intrinsic antitumor effect and the range of natural parvoviruses, a recombinant H1 parvovirus vector was constructed that produces the Apoptin protein, a tumor cell–specific, p53-independent, Bcl-2–insensitive apoptotic effector. We compared the apoptotic activity exerted by a recombinant hH1/Apoptin virus with that of a Green Fluorescent Protein (GFP)–transducing recombinant virus, hH1/GFP, in three human tumor cell lines differing in their susceptibility to wild-type parvovirus H1–induced killing. We found that in cells that were rather resistant to the basal cytotoxic effect of wild-type H1 or the GFP recombinant virus, a parvovirus that expressed Apoptin caused a pronounced, additional cytotoxic effect. In contrast to its enhanced cytotoxicity toward tumor cells, hH1/Apoptin virus was not more toxic to normal human fibroblasts than was the wild-type H1 virus. Taken together, these data indicate that enhancing the oncotropic behavior of wild-type H1 parvoviruses with the tumor-specific apoptotic potency of Apoptin should lead to an effective replicative parvoviral vector. Cancer Gene Therapy (2001) 8, 958–965

Chimeric and pseudotyped parvoviruses minimize the contamination of recombinant stocks with replication-competent viruses and identify a DNA sequence that restricts parvovirus H-1 in mouse cells.

ABSTRACT Recent studies demonstrated the ability of the recombinant autonomous parvoviruses MVMp (fibrotropic variant of the minute virus of mice) and H-1 to transduce therapeutic genes in tumor cells. However, recombinant vector stocks are contaminated by replication-competent viruses (RCVs) generated during the production procedure. To reduce the levels of RCVs, chimeric recombinant vector genomes were designed by replacing the right-hand region of H-1 virus DNA with that of the closely related MVMp virus DNA and conversely. Recombinant H-1 and MVMp virus pseudotypes were also produced with this aim. In both cases, the levels of RCVs contaminating the virus stocks were considerably reduced (virus was not detected in pseudotyped virus stocks, even after two amplification steps), while the yields of vector viruses produced were not affected. H-1 virus could be distinguished from MVMp virus by its restriction in mouse cells at an early stage of infection prior to detectable viral DNA replication and gene expression. The analysis of the composite viruses showed that this restriction could be assigned to a specific genomic determinant(s). Unlike MVMp virus, H-1 virus capsids were found to be a major determinant of the greater permissiveness of various human cell lines for this virus.

MCP‐3 (CCL7) delivered by parvovirus MVMp reduces tumorigenicity of mouse melanoma cells through activation of T lymphocytes and NK cells

Monocyte chemotactic protein 3 (MCP‐3/CCL7), a CC chemokine able to attract and activate a large panel of leukocytes including natural killer cells and T lymphocytes, could be beneficial in antitumor therapy. Vectors were constructed based on the autonomous parvovirus minute virus of mice (MVMp), carrying the human (MCP‐3) cDNA. These vectors were subsequently evaluated in the poorly immunogenic mouse melanoma model B78/H1. The infection of the tumor cells with MCP3‐transducing vector at low virus input multiplicities, but not with wild‐type virus, strongly inhibited tumor growth after implantation in euthymic mice. In a therapeutic B78/H1 model, repeated intratumoral injections of MCP3‐tranducing virus prevented further tumor expansion as long as the treatment was pursued. The antitumor effects of the MCP‐3‐transducing vector were not restricted to this tumor model since they could also be observed in the K1735 melanoma. The depletion of CD4, CD8, NK cells and of interferon γ (IFNγ) in mice implanted with MVMp/MCP3‐infected B78/H1 cells abolished the antitumor activity of the vector. The latter data, together with tumor growth in nude mice and reverse‐transcriptase (RT)‐PCR analyses of MVMp/MCP3‐treated tumors, clearly showed that activated CD4, CD8 and NK cells were indispensable for the antineoplastic effect in the B78/H1 tumor. Altogether, our results show that MCP3‐transducing parvovirus vectors may be quite potent against poorly or nonimmunogenic tumors, even in conditions where only a fraction of the tumor cell population is efficiently infected with recombinant parvoviruses.

Oncolytic parvovirus H1 induces release of heat-shock protein HSP72 in susceptible human tumor cells but may not affect primary immune cells.

Certain autonomous parvoviruses preferentially replicate in and kill in vitro-transformed cells and may reduce the incidence of spontaneous and implanted tumors in animals. Hence, these viruses and their derivatives are currently under evaluation as antitumor vectors. However, the mechanisms underlying their tumor-suppressing properties are not yet understood. We asked whether the lytic parvovirus H1 may enhance the immunogenicity of infected tumor cells. Out of human melanoma and gastrointestinal tumor cells, we selected the cell line SK29-Mel-1 being very susceptible to H1-induced apoptotic killing. Here, no upregulation of HLA class I and costimulatory molecules could be observed following H1 infection. However, a strong release of the immunogenic signal—the inducible heat-shock protein HSP72, but not constitutive HSP73—was observed after H1 infection. The HSP72 release was higher and of longer duration than a conventional heat-shock treatment. We also explored H1 replication and cytotoxicity in human immune cells, as such cells may constitute targets for H1 virus replication. Long-term cultured lymphocytes, monocytes, immature and mature dendritic cells were not susceptible to H1 virus. Altogether, parvovirus-mediated cell killing may in vivo enhance tumor immunogenicity by HSP72 release and thus contribute to the antitumor effect of parvoviruses.

Cancer Gene Therapy through Autonomous Parvovirus - Mediated Gene Transfer

Parvoviruses are small nuclear replicating DNA viruses. The rodent parvoviruses are usually weakly pathogenic in adult animals, bind to cell surface receptors which are fairly ubiquitously expressed on cells, and do not appear to integrate into host chromosomes during either lytic or persistent infection. The closely related rodent parvoviruses MVM, H-1 and LuIII efficiently infect human cell lines. Most interesting, malignant transformation of human and rodent cells was often found to correlate with a greater susceptibility to parvovirus-induced killing (oncolysis) and with an increase in the cellular capacity for amplifying and / or expressing the incoming parvoviral DNA. These and other interesting properties make these autonomous rodent parvoviruses and recombinant derivatives promising candidate antitumor vectors. Capsid replacement vectors have been produced from MVM or H-1 virus that carry transgenes encoding either therapeutic products (cytokines/chemokines, Apoptin, herpes simplex virus thymidine kinase) or marker proteins (green fluorescent protein, chloramphenicolacetyl transferase, luciferase). This review describes the current state of the art regarding the potential application of wild-type parvoviruses and derived vectors for the treatment of cancer. In particular, recent successes with the development of replication-competent virus-free vector stocks are discussed and results from pre-clinical studies using recombinant parvoviruses transducing various cytokines/chemokines are presented.

Vectors based on autonomous parvoviruses: Novel tools to treat cancer?

Autonomous parvoviruses are small nuclear‐replicating DNA viruses. The rodent parvoviruses usually are non‐ or weakly pathogenic in adult animals, bind to surface receptors which are expressed on most cells, and do not appear to integrate into host chromosomes during either lytic or persistent infections. Interestingly, malignant transformation of the target cells was often found to correlate with an increase in their capacity for amplifying and/or expressing the incoming parvoviral DNA, and is associated with oncolysis, i.e., the selective killing of the infected tumor cells. Moreover, the closely related parvoviruses MVM, H‐1 and LuIII efficiently infect human cell lines. This finding makes these parvoviruses promising candidate vectors for therapies that require transient expression of a transduced gene. In particular, parvoviruses may be suitable to target and kill tumor cells and simultaneously deliver appropriate transgenes, e.g., genes coding for immuno‐stimulatory factors.