In 1960, Moloney published the isolation of a strain of murine leukemia virus that now carries his name (Moloney, 1960). Since then, the Moloney murine leukemia virus (MoMLV) has been widely used in research and applied biotechnology.

The MoMLV belongs to the large Retroviridae family of enveloped RNA viruses. The hallmark of this family is their ability to reverse transcribe their genome from RNA to double stranded DNA and integrate into the genome of the host cell. With the discovery of reverse transcription, the first exemption to the unidirectional and presumably irreversible flow of genetic information from DNA to RNA to protein was found. The finding of an RNA-dependent DNA polymerase or reverse transcriptase in purified virions proved that the RNA could be transcribed into DNA and its discoverers, Baltimore and Temin, shared a Nobel Prize in Physiology and Medicine with their former mentor Dulbecco in 1975 (Baltimore, 1970; Temin and Mizutani, 1970). Today, the reverse transcriptase of the MoMLV is a standard enzyme used in molecular biology laboratories to generate complementary DNA (cDNA) copies of RNA. Unlike mRNA which is unstable, cDNA can be manipulated with relative ease and thus is used in a variety of molecular techniques including cloning, sequencing, RT-PCR, microarrays and serves as a specific hybridization probe (Skalka and Goff, 1993). The cDNA copies of cellular messenger RNA (mRNA) extracts represent the genes that are being expressed in a given cell.

The study of retroviral oncogenesis started at the turn of the century with the first evidence of retrovirus existence (Ellermann and Bang, 1908; Rous, 1911) and led to the discovery of viral oncogenes (v-oncogenes) (Huebner and Todaro, 1969). These were found to be cellular mutated genes picked up by tranducing oncogenic retroviruses which transfer them into new hosts inducing tumor development (Bishop, 1983; Varmus, 1984). These naturally occurring retroviruses not only helped lay the foundation for our current understanding of cancer but also inspired scientists to use genetically modified retroviruses to transfer a gene of choice to cells. The retrovirus genome can be divided functionally into cis -acting sequences that are required for encapsidation, reverse-transcription and integration and trans -acting sequences that code for the products of the 3 viral genes ( gag, pol and env ). After the retroviral core is formed, no further protein synthesis is required for the events leading to retrovirus integration in the host cell genome. In fact all viral genes can be removed from the genome and replaced with a gene of interest without affecting its ability to be encapsidated, reverse-transcribed and integrated. This is the principle behind the retroviral vector system (Miller, 1997). In nature, most retroviruses carrying oncogenes are replication-defective and require the presence of non-defective “helper” viruses which provide virus proteins in trans to replicate. The development of packaging systems capable of producing replication-defective retroviruses started by mimicking this natural strategy (Shimotohno and Temin, 1981; Wei et al., 1981). However, the presence of helper virus was undesired for many applications. In 1983, Baltimore and Temin’s research groups reported the development of the first packaging cells that supplied in trans all viral proteins supporting the replication of defective retroviruses in the absence of helper viruses (Mann et al., 1983; Watanabe and Temin, 1983). This represented a major advance in retroviral vector design.

The ability of retroviruses to stably integrate into the host cell genome, thereby providing the possibility of long-term expression in the transduced cells and their progeny, provided additional incentive for the development of retroviral gene transfer vectors. Retroviral vectors were the first vectors used for efficient and stable gene transfer into mammalian cells (Cone and Mulligan, 1984). The most extensively used retrovirus for the generation of vectors is the MoMLV partly because of the simplicity of its well-studied genome. For the last 20 years, advances in retroviral vector design were accompanied by the development of novel applications (Barquinero et al., 2004). Experimental applications now include among others the construction of cDNA libraries, the generation of transgenic animals, gene silencing by stable RNA interference, chromosome tagging and cell tagging for cell lineage and clonality studies (Barquinero et al., 2004; Miller, 1997).

Perhaps the most exciting application of retroviral vectors is human gene therapy. Gene therapy is a new therapeutic approach that involves the transfer of genes into a patient’s cells. Retroviral vectors have long been recognized as ideal delivery vehicles for gene transfer and thus were among the first viral systems to be developed. In 1990, they were the first viral vector to be approved by the government of the United States for gene therapy (Anderson et al., 1990). The aim of this first clinical protocol was to treat a 4-year old patient suffering from adenosine deaminase (ADA)-deficient severe combined immunodeficiency (SCID) by transducing her own T-lymphocytes with a retroviral vector expressing a normal ADA gene ex vivo and then re-infusing these modified blood cells into her circulation. Although results show relatively long-term persistence of modified T cells, ideally, one would like to transduce blood stem cells to provide a permanent solution for ADA deficiency since stem cells can continuously give rise to new modified lymphocytes. In further trials, successful retroviral-mediated transfer to bone marrow CD34+ cells was reported (Aiuti et al., 2002).

In 2000, scientists put theory into practice using MoMLV-based vectors in the first clinical trial that cured a disease demonstrating proof of principle for gene therapy. In this clinical protocol, conducted by Fischer (Paris), 9 out of 11 children suffering from X-linked severe combined immunodeficiency (X-SCID) showed a clear clinical improvement (Cavazzana-Calvo et al., 2000). The protocol consisted in transducing the patient's hematopoietic stem cells ex vivo with a retroviral vector carrying a normal cytokine-receptor gamma chain gene. Unfortunately, a few months following this treatment 3 out of the 11 patients developed T-cell leukemia likely due to retrovirus insertional mutagenesis near the LMO2 locus. After considerable research, it has now been determined that this adverse event appears to be due to a uniquely high-risk situation using that specific vector construct in patients suffering from that particular disease (Barquinero et al., 2004; Berns, 2004; Sinn et al., 2005). In a further ongoing trial for patients with X-SCID, significant clinical benefit with no adverse effects was reported for the 4 patients treated (Gaspar et al., 2004). In addition to the treatment of genetic disorders, retroviral vectors are also attractive candidates for the treatment of cancer ex vivo and in vivo , because they can selectively target rapidly dividing cells (Gunzburg, 2003; Rainov and Ren, 2003). Cancer is the most frequently targeted disease by gene therapy (www.wiley.co.uk). With the initiation of more than 272 clinical trials since 1990, retroviral vectors have been the most widely used vectors in gene therapy to date.

Interest in the use of retroviral vectors for gene therapy applications continues to grow, although most scientists agree that progress toward controlled integration for increased biosafety must be done and further developments in large-scale manufacturing are needed (Barquinero et al., 2004; Gunzburg, 2003; Sinn et al., 2005). In spite of the extensive work done with retroviral vectors, advances in vector design and production systems did not parallel the lagging development of purification methods (Merten, 2004). Retrovirus purification still largely relies on the traditional method of sucrose equilibrium density gradient ultracentrifugation which typically results in poor recovery of infective particles and significant contamination with cell membrane vesicles (Vogt, 1997). Highly purified retrovirus preparations are required not only for gene transfer purposes but for characterization studies, immunological studies, and as “gold standards” for downstream processing quality control. These preparations are difficult to obtain, particularly in cases where viruses occur at low titers and the viral particles are unstable, which is the case for retroviruses. For retroviruses to be used as gene transfer vectors they need to be active, that is the ability to transduce target cells must be preserved during downstream processing. Furthermore, the development of scaleable purification strategies is necessary to ensure that sufficient quantities of high purity vector stocks are available for pre-clinical and clinical trials (Andreadis et al., 1999).

The main objective of this thesis was to develop purification procedures to obtain highly purified MoMLV-derived vector preparations for use in clinical and experimental applications. The finding that retroviruses can efficiently bind heparin ligands led us to the development of a novel affinity chromatography technique and to study retrovirus-heparin interactions. The lack of a purified standard to evaluate the purity of the vectors led us to the development of a powerful alternative purification strategy based on rate zonal ultracentrifugation. Finally, a detailed analysis of the purity of the preparations was performed.