| Collection Mémoires et thèses électroniques |
|
| Accueil À propos Nous joindre |
| Chapter I Downstream processing of oncoretroviral and lentiviral gene therapy vectors | ||
|---|---|---|
|
|
|
Table des matières
María de las Mercedes Segura 1,2, Amine Kamen 2, and Alain Garnier 1*
1- Department of Chemical Engineering, Centre de Recherche sur la Fonction, la Structure et l’Ingénierie des Protéines, Université Laval, Québec, Canada, G1K 7P4; phone : 418-656-3106; fax: 418-656-5993; e-mail: alain.garnier@gch.ulaval.ca
2- Biotechnology Research Institute, NRC, 6100 Royalmount Avenue, Montreal, Quebec, Canada, H4P 2R2
Accepted by Biotechnology Advances, December 6th 2005
Les vecteurs rétroviraux d’origines oncorétrovirale et lentivirale, ont un grand potentiel pour livrer des gènes étant donné leur intégration efficace dans le génome des cellules cibles, procurant la possibilité d’une expression génique à long terme. Plusieurs groupes de recherche ont déployé des efforts considérables pour le développement de procédés à grande échelle permettant d’obtenir des quantités de vecteurs rétroviraux nécessaires aux essais pré-cliniques et cliniques. Une attention particulière a été portée à la conception de systèmes de production optimisés capables de générer sécuritairement de grands volumes de vecteurs ayant des titres viraux satisfaisants. Cependant, la production de vecteurs de grade clinique pour la thérapie génique, spécifiquement pour les applications in vivo , requière également des stratégies de purification pouvant être mises à l’échelle, pour enlever les contaminants présents dans les surnageants récoltés, tout en préservant la fonctionnalité des vecteurs. Dans cet article, nous avons fait le bilan des récents progrès dans le domaine des procédés de purification des vecteurs retroviraux. Les méthodes courantes décrites dans la littérature concernant la clarification, la concentration et la purification des vecteurs rétroviraux seront présentées, avec une emphase spéciale sur les nouvelles méthodes de chromatographie qui donnent la possibilité de purifier les rétrovirus à grande échelle et ce de façon sélective et efficace. Les problèmes associés à la stabilité et la quantification des particules rétrovirales seront soulignés et les défis futurs seront discutés.
Retroviral vectors from both oncoretroviral and lentiviral origin have a great potential as gene delivery vehicles because they integrate efficiently into the genome of the target cells providing the possibility of lifelong gene expression.A number of research groups have devoted considerable effort to the development of large-scale processing strategies for retroviral vectors to ensure that sufficient quantities of vector stocks are available for pre-clinical and clinical trials. Particular attention has been given to the design of optimized production systems able to generate large volumes of safe retroviral vector stocks in satisfactory viral titers. However, the manufacturing of clinical-grade vectors for gene therapy, especially for in vivo applications, additionally requires scaleable purification strategies to remove the contaminants present in the harvested supernatants while preserving the functionality of the vectors. In this article, we review recent advances made in the field of downstream processing of retroviral vectors. The methods currently described in the literature for clarification, concentration and purification of retroviral vectors will be presented, with special emphasis on novel chromatography methods that open up the possibility to selectively and efficiently purify retroviruses on a large-scale. Problems associated with stability and quantitation of retroviral particles will be outlined and future challenges will be discussed.
Gene therapy is defined as the administration of genetic material in order to modify or manipulate the expression of a gene product or to alter the biological properties of living cells for therapeutic use. It is a developing technology that holds great promise for the treatment of inherited metabolic disorders as well as acquired diseases such as cancer, cardiovascular and some infectious diseases, cancer being the most frequently targeted disease.
According to the mode of gene delivery to the target cells, there are two major categories of somatic cell gene therapy. In the ex vivo approach, cells are removed from the body, incubated with the vector and genetically modified cells are returned to the body. This procedure is generally limited to a few cell types, such as blood cells, that are easy to remove and return. The second is the in vivo approach, where the vector is administered directly to the patient. The vector can be delivered either locally into the affected tissue or systemically into the bloodstream of the patient. The injection of a vector directly into a tumour mass is a good example of local administration.
Retroviral vectors have attracted the attention of gene therapy researchers for their ability to stably integrate the transgene of interest into the target cell genome providing the possibility of long-term gene expression and ultimately long-term therapeutic effect. Being the most popular viral vector used in clinical trials, vectors derived from the oncovirus Moloney murine leukemia virus (MoMLV) have demonstrated great potential as gene delivery vehicles. More recently, vectors derived from another well characterized member of the retrovirus family, the lentivirus human immunodeficiency virus type 1 (HIV-1), have been developed and approved for use in human clinical studies. While oncoretroviral vectors can only transduce dividing cells, lentivirus vectors can deliver genes to dividing as well as non-dividing cells. This is very convenient since many potential target cells including neurons, hepatocytes, myocytes, retinal photoreceptors, macrophages and hematopoietic stem cells (HSC) divide infrequently in vivo . However, the inability of oncoviruses to transduce non-dividing cells can be attractive to selectively target rapidly dividing cells such as cancer cells (Rainov and Ren, 2003). For simplicity purposes, we will refer to oncoretroviral and lentiviral vectors as retroviral vectors in this review.
Major obstacles associated with retroviral vectors include the production of low-titer viral stocks and the instability of the viral particles produced. In the best cases, between 106 to 107 infective viral particles per mL of cell culture supernatant are produced by commonly used producer systems. These concentrations are high enough for certain ex vivo applications. However, concentration of vector stocks is required for most gene therapy applications in order to improve transduction efficiencies. Moreover, concentrated or not, viral stocks still contain contaminants that need to be removed to increase the potency and safety of the final product. Non-purified vector preparations contain contaminating molecules that are toxic to cells and reduce transduction efficiencies ex vivo (Yamada et al., 2003). These preparations also induce a systemic immune response and inflammation when injected in vivo (Baekelandt et al., 2003; Scherr et al., 2002). Impurities contained in the vector supernatant not only come from the supplements and reagents added to the culture (i.e. serum, plasmid DNA for transient transfection), but also are released by intact or disrupted producer cells (i.e. inhibitors of transduction, genomic DNA and host proteins).
The development of large-scale production and purification methods for the generation of high-titer clinical-grade retroviral vectors is critical to advances in gene therapy. Over the past several years considerable progress has been made in the fields of vector design and production systems. Using optimized bioreactor systems, production of large volumes of retroviral vector stocks is feasible (for review see McTaggart and Al-Rubeai, 2002; Merten, 2004; Zufferey, 2002). Less effort has been invested in developing and optimizing purification processes able to handle large volumes of vector stocks.
Candidate technologies for the downstream processing of oncoretroviral vectors were previously proposed (for review see Andreadis et al., 1999; Braas et al., 1996; Lyddiatt and O'Sullivan, 1998). Membrane separation and chromatography were deemed the most promising technologies for large-scale manufacturing of retroviral vectors. Additionally, the authors strongly encouraged the development of methods specifically tailored to the unique biochemical and physical features of retroviral particles. As a result, various affinity chromatography strategies and new chromatography technologies for the purification of retroviral vectors have emerged in the past few years (Segura et al., 2005; Slepushkin et al., 2003; Williams et al., 2005a; Williams et al., 2005b; Ye et al., 2004). This review will provide the reader with an overview of the techniques recently made available for downstream processing of oncoretroviral and lentiviral vectors. Since both types of vectors share common structural and physical properties, we anticipate that it will be possible to rapidly adapt techniques originally developed for one vector to the other. It is the authors’ hope that the information presented here benefits future developments.
Retroviruses comprise a family of RNA enveloped viruses broadly divided in two categories (simple and complex) according to their genome organization. All retroviruses contain at least 3 major coding domains: gag, pol and env. While simple retroviruses such as the MoMLV only carry these genes, complex retroviruses including lentiviruses present several accessory genes which regulate details in the virus replication cycle.
Retroviral particles are enveloped with a lipid membrane derived from the virus-producer cell (Figure 1). Embedded in this membrane is the viral encoded envelope protein that interacts with specific receptors on the cell surface. This protein is cleaved into transmembrane (TM) and surface (SU) subunits that remain attached to each other by noncovalent interactions. Retroviral vectors are usually pseudotyped; that is, they carry foreign virus envelope proteins that confer them beneficial properties for gene therapy. For instance, retrovirus pseudotypes bearing the VSV-G protein instead of the natural envelope protein have an extended host cell range and show increased physical stability (Burns et al., 1993). Gag, the most abundant protein in the virion, is cleaved during maturation into 3 individual structural proteins that form “layers” underneath the lipid membrane. The matrix (MA) forms the outer layer that surrounds the viral core. The core is delimited by a protein shell composed of capsid (CA) proteins and encloses the nucleoprotein complex that contains two identical positive strands of RNA genome complexed with nucleocapsid (NC) proteins. Infective retrovirus particles contain 3 virally encoded enzymes: reverse transcriptase (RT), integrase (IN) and protease (PR). Additionally, retroviruses incorporate several host cellular proteins on the surface and inside the virion, some of which are believed to play a role in the virus replication cycle (Ott, 2002). Overall, the retrovirus particles are composed of 60-70% protein, 30-40% lipid, 2-4% carbohydrate and 1-2% RNA (Andreadis et al., 1999).
Retroviruses share common physical characteristics. The particles are spherical and measure about 80-120 nm in diameter according to thin-section electron microscopy. They have a mass of ~ 2.5 × 108 Da (Vogt and Simon, 1999) and present a density of 1.16 g/mL in sucrose density gradients.
Retroviral particles are extremely labile. From the downstream processing point of view, retrovirus instability is translated into low overall recoveries of infective viral particles. To minimize the loss of infective particles, it is important to have a good knowledge of the stability of the vector and its susceptibility to different factors (i.e. temperature, pH, ionic strength, shear stress) prior to designing downstream processing strategies. Ideally, stability studies of the vector in question to the environmental conditions to which the virus will be exposed during purification should be performed.
Retroviral vectors rapidly lose their activity at 37°C, the temperature at which the vectors are produced and titered, with a half-life between 5 to 8 h (Andreadis et al., 1997; Higashikawa and Chang, 2001; Le Doux et al., 1999; McTaggart and Al-Rubeai, 2002; Segura et al., 2005). As temperature decreases, retrovirus half-life increases. At room temperature, retroviral vectors present half-lives between 1 and 2 days (Higashikawa and Chang, 2001; Segura et al., 2005). The vectors’ stability markedly improves at 4°C with half-lives over 8 days (Higashikawa and Chang, 2001). Retrovirus temperature stability was found to be dependent on the particular vector envelope protein and producer cell line type from which the viral lipid envelope was derived (Beer et al., 2003; Burns et al., 1993). The number of freeze-and-thaw cycles should be kept to a minimum during downstream processing. Retroviral vector stocks, both concentrated and nonconcentrated, lose half of their activity after the first 2 to 4 freeze-and-thaw cycles (Bowles et al., 1996; Burns et al., 1993). Therefore, in order to predict and correctly interpret the temperature-related inactivation that occurs during purification and rationally select the most convenient way to store vector stocks in between downstream processing operations, vector stability at room temperature, 4°C and the stability to freeze-and-thaw cycles should be determined in each case.
Studies of the effect of pH on the activity of VSV-G pseudotyped retroviral vectors revealed that the vectors are more stable at pH 7.0, 37°C, but their half-lives markedly dropped to less than 10 min at pH 6 or pH 8 (Higashikawa and Chang, 2001). Similar observations were reported by Ye and collaborators (Ye et al., 2003) who found that ecotropic MoMLV remains infectious in a narrow pH range from 5.5 to 8.0. Virus inactivation beyond these limits of pH was fast and irreversible. Electron microscopy studies showed that the viral envelope was degraded at extreme pH as revealed by the penetration of the heavy metals used for staining.
Hyperosmotic conditions lead to the loss of water from organelles, vesicles, and enveloped virions. Loss of infective retroviral particles following salt precipitation and sucrose density ultracentrifugation were partly attributed to retrovirus sensitivity to osmotic pressure (Aboud et al., 1982; Andreadis et al., 1999). VSV-G pseudotyped oncoretroviral vectors infectivity was shown to be affected by increasing NaCl concentrations (Segura et al., 2005). The biological inactivation of the vector after NaCl treatment was irreversible and happened very rapidly. Just 1 h of exposure to 1M NaCl at room temperature was enough to inactivate 50% of the virus. Morphological changes and broken particles were observed after a 3 h treatment with high salt concentration.
Chemical compounds introduced at some stage during downstream processing may also affect retroviral vectors’ ability to transduce. For example, oncoretroviral vectors were found to be sensitive to imidazole, a common desorption agent used for immobilized metal affinity chromatography (IMAC) (Ye et al., 2004). Recovery of infective particles was improved from 35% to 56% by using half the concentration of imidazole for vector elution. Similarly, oncoretroviral vectors were found to be susceptible to increasing concentrations of d-biotin which is used to elute bound proteins from streptavidin coated chromatography supports (Williams et al., 2005b). In addition, it has been demonstrated that exposure of retrovirus particles to denaturing agents (i.e. guanidine-HCl or urea), typically used to elute proteins from affinity matrices, results in 100% inactivation of the virus (Williams et al., 2005b). Susceptibility of the retroviral particles to EDTA used to re-dissolve retrovirus-calcium phosphate pellets has also been described (Pham et al., 2001).
Finally, shear forces encountered during ultracentrifugation also influence the stability of retroviral vectors. Due to the monomeric structure of the protein, VSV-G pseudotyped particles are more stable than those containing the widely used dimeric amphotropic Env-protein and thus can be effectively concentrated generating high-titer vector stocks (Burns et al., 1993).
Given the instability of retroviral particles, the factors described above should be considered at the time of selecting appropriate virus purification methods in order to maximize recovery of infective retroviral particles. Density ultracentrifugation using highly hyperosmotic media, aqueous two-phase extraction using high salt concentrations, precipitation with salts and also adsorptive chromatography procedures that require the use of harsh conditions to elute viral particles are among the methods that could potentially have an impact on the stability of the virus particle.
The availability of reliable tools to quantify retrovirus particles is critical for the development of downstream processing strategies. Although a variety of quantitation methods are being used, most suffer from known flaws. Ideally, a combination of two methods should be used to determine both active (transduction-competent) and total retrovirus particles. While direct quantitation of transduction-competent retroviral particles is carried out in assays involving the use of target cells, the total number of virus particles can be determined directly in vector supernatants (Figure 2).
The viral titer is usually defined as the number of transduction-competent retrovirus particles per mL of virus stock. Viral titers are typically quantitated by measuring transgene expression in target cells. For this purpose, most retroviral vectors used in developmental phases usually carry marker genes, such as GFP, lacZ or antibiotic resistance genes, which allow for rapid detection of transduced cells. The titration assays consist of overlaying serial dilutions of vector stocks onto target cells. Detection of transduced cells is carried out either by visual identification of marker protein expressing colonies (Chang and Zaiss, 2002; Srinivasakumar, 2002) or by flow cytometry (Dull et al., 1998; White et al., 1999). Although measuring marker transgene expression remains the most useful criterion to determine vector potency, the method has several limitations. Viral titers are influenced by specific transduction conditions used in the assay such as virus stock volume, time of virus exposure to target cells, the number and size of the target cells, polybrene concentrations and in the case of oncoretroviral vector the rate of cell growth. In addition, the assay is time consuming typically requiring 4 to 5 days for completion (Carmo et al., 2004) depending on the detection technique employed. Moreover, due to slow virus diffusivity and rapid virus decay only a small proportion of active virus particles in a stock (~10%) successfully transduces target cells (Andreadis et al., 2000). Hence, the method underestimates the number of transduction-competent retrovirus particles. Mathematical models that provide a better estimate of the initial concentration of active virus particles in a stock, independent of the specific conditions used in an assay, have been reported (Andreadis et al., 2000; Kwon and Peng, 2002).
Alternatively, viral titers can be determined by quantifying proviral DNA or transgene mRNA levels in the transduced target cells. The main advantage these methods offer over the traditional titration assay described above is that they do no rely on the presence of marker genes. Proviral DNA integration events can be determined using real time quantitative polymerase chain reaction (qPCR) (Pan et al., 2002; Sastry et al., 2002). However, since vector integration does not necessarily correlate with successful transgene expression, the method tends to overestimate viral titers (Sastry et al., 2002). A better approach is to measure transgene expression at the mRNA level by quantitative reverse transcriptase PCR (qRT-PCR) (Lizee et al., 2003). Semiquantitative Southern and Northern blotting could also be used to quantify proviral DNA and mRNA levels in transduced cells. However, these methods are time consuming, labor-intensive and have limited accuracy compared to PCR-based assays.
Several methods can be used to quantify total virus particles directly in vector stocks. These methods do not discriminate between active and inactive retrovirus particles and therefore provide little information concerning the potency of the vector preparation. Nevertheless, they are useful to study variations in total to active particle ratios and determine the quality of vector stocks at different stages of the purification process.
Negative stain electron microscopy is the gold standard for the quantitation of total retrovirus particles. Virus particles premixed with a known concentration of latex beads are typically stained with uranyl acetate or phosphotungstic acid and counted under transmission electron microscope (Alain, 1997). The method requires previous concentration and purification of vector supernatants since virus concentrations in vector supernatants are usually too low to be accurately quantified and impurities contained in the vector supernatant may prevent observation of virus particles (Kwon et al., 2003). Moreover, caution should be taken when examining samples containing high amounts of cellular membrane vesicles such as those obtained by sucrose density ultracentrifugation since these vesicles might be confused with retroviral particles (Bess et al., 1997; Gluschankof et al., 1997). High performance liquid chromatography (HPLC) also showed to be useful for the quantitation of total virus particles (Transfiguracion et al., 2004). Retroviral particles were separated from protein contaminants using anion exchange chromatography and detected by absorbance at 260 nm. This method also requires concentration of vector supernatants and Benzonase® treatment due to contaminating DNA interference.
Another possibility is to estimate the number of total particles by measuring virus components. A variety of immunoassays can be employed to detect and quantify viral proteins including quantitative determinations of p24Gag (lentivectors) or p30Gag (oncoretrovectors) capsid protein content by enzyme-linked immunosorbent assay (ELISA) or semiquantitative Western blotting (Naldini et al., 1996; Rigg et al., 1996). Additionally, enzymatic assays for reverse transcriptase activity can be performed. Some of these procedures may be conducted using commercially available kits (Logan et al., 2004; Naldini et al., 1996). A method for single retrovirus particle visualization and enumeration using indirect immunofluorescence microscopy has also been described (Pizzato et al., 1999). A sensitive method to quantify RNA genome copies directly in vector supernatants using qRT-PCR has been described (Carmo et al., 2004). Although the method allows for accurate and rapid results, it only quantifies vector particles containing RNA. Due to the presence of defective retrovirus particles without RNA, the method underestimates total particle counts. On the other hand, the number of transduction-competent particles is overestimated by this method since defective particles with RNA are also present in vector stocks.
Figure 2 Assays used for the quantitation of total retrovirus particles in vector stocks and transduction-competent particles.
It is important to note that each purification step carries the risk of virus inactivation and the potential to separate active from defective particles, thus active to total particle ratios may change during purification. As a result, these ratios are not universal and can not be used blindly to determine the concentration of active virus particles based on a total particle count and vice-versa. Ratios for each particular situation could be established for use in routine quantitation of samples at the same purification stage, but this approach has less value for developmental phases.
Additionally, the use of in-house virus standards is highly recommended to avoid inter-assay discrepancies. Moreover, to validate each laboratory’s in-house virus standards and assays and to facilitate inter-laboratory comparisons, it is necessary to normalize titer values to a common standard. As described for adenoviral and adeno-associated viral vectors, lentiviral and oncoretroviral vector reference standards are being established (Flotte et al., 2002).
Finally, a major concern for the safety of retroviral vector preparations is the presence of replication-competent viruses (RCV). Methods for detection of RCV are beyond the scope of this article and the reader is referred to the “Supplemental Guidance on Testing for Replication-Competent Retrovirus in Retroviral Vector-Based Gene Therapy Products and During Follow-up of Patients in Clinical Trials Using Retroviral Vectors” issued by the FDA’s Center for Biologics Evaluation and Research (CBER) in October 2000 for useful information about this subject (CBER, 2001).
At the end of the production phase, harvested retroviral vector supernatants undergo a series of processing steps aimed at improving the potency of the vector preparation and eliminating the impurities contained in the vector supernatant.
Contaminants eliminated in each process step are indicated.
Serum
Serum is the main source of contaminants in harvested supernatants. Serum supplementation increases the complexity, duration and cost of downstream processing operations and presents the risk of introducing biological contaminants. Naturally, the use of serum-free media for vector production facilitates downstream processing by dramatically decreasing the amount of contaminating proteins (i.e. bovine serum albumin, bovine transferrin and immunoglobulins) and lipids. Unfortunately, reports demonstrating successful production of retroviral vectors in serum-free media are scarce (McTaggart and Al-Rubeai, 2002). Alternatively, production of vectors in very low protein media helps reduce the chances of contamination (McTaggart and Al-Rubeai, 2000; Moy et al., 2000). Specific vector productivity is often higher in low protein media than at the 10% serum concentrations typically used (McTaggart and Al-Rubeai, 2002; Merten, 2004; Zufferey, 2002).
Inhibitors of transduction
The producer cell line itself could be a source of contamination. Producer cells release inhibitors of transduction such as proteoglycans, glycosaminoglycans and free envelope proteins into the supernatant that reduce the vectors’ potential for efficient gene delivery (Le Doux et al., 1996; Le Doux et al., 1998; Slingsby et al., 2000)
Host proteins
In addition, disrupted producer cells release membrane fragments and impurities derived from the cell cytoplasm including large amounts of host proteins and genomic DNA. Pre-clinical studies with lentiviral vectors have shown the significant contribution of 293T producer cell-derived components to the immune response (Baekelandt et al., 2003).
DNA contaminants
Contaminating genomic DNA is also considered potentially hazardous. Moreover, it interferes with RCV detection by PCR-based methods (Chen et al., 2001). The levels of DNA contamination were found to continuously increase during production of VSV-G oncoretroviral vectors probably due to VSV-G toxicity on producer cells (Segura et al., 2005). In the case of vector production by transient transfection, a large amount of plasmid DNA is added to cell cultures every time a vector lot is produced with the associated risk of introducing adventitious agents including endotoxins (a fever-producing byproduct of gram-negative bacteria commonly known as pyrogen). Removal of the plasmid DNA coding for the packaging functions may be desired to avoid the risk of transferring these functions to the target cells (Sastry et al., 2004).
Host cell-derived impurities and endotoxins are of particular concern for regulatory agencies and their removal beyond detectable limits is required for the production of clinical-grade vector preparations (Smith et al., 1996). Determining the optimum harvesting period is critical to avoid massive contamination with host cell impurities. In practice, sacrificing product yield for quality by discarding the last days of vector production can be worthwhile.
Both high molecular weight proteoglycans and DNA contaminants represent an important challenge for downstream processing. Due to their large size and strong negative charge they can co-purify with retroviruses when using common separation methods based on size or charge. Digestion steps using condroitinase ABC and DNase could be introduced in the downstream process to eliminate proteoglycans and DNA contaminants respectively (Le Doux et al., 1996; Le Doux et al., 1998; Sastry et al., 2004). However, subsequent removal of digested products and added enzymes would be required, increasing the duration of the process. Considering the instability of retroviral vectors, longer purification processes that may result in low overall recoveries of infective viral particles should be avoided whenever possible.
Strategic design and optimization of the procedures is critical to maximize yield and quality of the final product and ensure consistency of the manufacturing process (Figure 3). The selected methods for the clarification and concentration of retroviral particles should be amenable to handling large volumes of supernatant. These initial steps are primarily intended for removing cells, cell debris and water. Some degree of purification may also be accomplished during concentration. However, high resolution at these early stages of the process is not as important as scalability (Table I). The main purification issues are left to be resolved during the purification stage itself. During this stage, retroviral particles are separated from most contaminants contained in the vector supernatant. Often more than one purification step is required to bring the product to the desired level of purity. The polishing step is further introduced to remove remaining impurities and/or closely-related species (i.e. defective vector forms and/or cell membrane vesicles). The final product should be specially formulated for long-term storage stability.
Clarification, the removal of producer cells and cell debris from crude supernatant, is the first step of the downstream process. This step is performed immediately after vector harvest. At the laboratory scale, removal of cells and large cell debris is achieved by low speed centrifugation and microfiltration. The introduction of a centrifugation step before membrane filtration avoids membrane clogging. Microfiltration through 0.45 µm pore size filters follows to achieve greater clarification. For working volumes exceeding 1L, clarification using a single step of membrane filtration is preferred. In this case, fast clogging of the pores with cell debris may occur, depending on the initial membrane pore size and quality of the crude stock, resulting in reduction of the membrane actual pore size and consequently virus rejection. Indeed, recovery of infective particles after microfiltration through 0.45 µm membranes was found to correlate with filtration rates which are associated with the extent of pore obstruction (Reeves and Cornetta, 2000). Therefore, it is crucial to limit the volume of supernatant to be passed per filter. It is also convenient to filter crude supernatants through a series of membranes with decreasing pore size to minimize membrane clogging. This strategy avoids the need for a prior centrifugation step and results in efficient supernatant clarification with minimum loss in vectors’ titer (Moy et al., 2000; Reeves and Cornetta, 2000; Segura et al., 2005; Slepushkin et al., 2003)
One of the main limitations with retroviral mediated gene therapy is that gene delivery rates are usually too low to achieve therapeutic effect for most in vivo applications. Transduction efficiencies can be improved by using concentrated doses of retroviral vectors. Several methods have been proposed for concentration of viral particles (Table II). Introducing this step in the early stages of downstream processing facilitates subsequent operations by reducing the volume of feed and consequently the size of the equipment and infrastructure required.
Virus pelleting by centrifugation is traditionally employed to concentrate viruses. Both ultracentrifugation and long low-speed centrifugation methods (usually several hours) can efficiently pellet retroviruses. Using centrifugation, high concentration of the virus stocks (over 100-fold) can be easily attained by resuspending viral pellets in small volumes of resuspension buffer (Table II). However, transduction efficiencies usually do not increase proportionally with the concentration factor and often they do not increase at all compared to nonconcentrated virus stocks. This effect has been attributed to loss of active viral particles due to shear stress or extended processing time and to co-concentration of viral particles with high molecular weight inhibitors of transduction (Bajaj et al., 2001; Burns et al., 1993; Le Doux et al., 1996; Transfiguracion et al., 2003). In addition, susceptibility of each particular pseudotyped retroviral vector to hydrodynamic shear varies depending on the stability of the env-protein (Burns et al., 1993). Another important limitation of ultracentrifugation procedures is that ultra-high speed rotors currently in use generally have small volume capacity (Table II).
Several methods for the concentration of retroviruses by precipitation with additives have been described. The advantage of using additives to induce virus precipitation is that following the treatment, virus pellets can be obtained at low centrifugation speeds in a short time. Furthermore, using low-speed rotors larger volumes of supernatant can be processed per run (Table II). Charged polymers can be used to induce retrovirus precipitation. Cationic polymers enhance transduction efficiency and form virus-polymer complexes that can be pelleted by low-speed centrifugation. For instance, Zhang and collaborators (Zhang et al., 2001) reported a protocol for the concentration of 3 L of supernatant per round using poly-L-lysine, although recovery of active viral particles was only 26%. Curiously, while anionic polymers alone inhibit retroviral transduction, the addition of a mixture of anionic and cationic polymers to virus stocks improves transduction efficiencies and results in the formation of complexes that can easily be concentrated and purified by a rapid low speed centrifugation step (Le Doux et al., 2001). A major disadvantage with the use of these polymers is that they interact irreversibly with retrovirus particles to form a virus-polymer complex that cannot be dissociated for further processing. Alternatively, co-precipitation with calcium phosphate (CaPO4) has been used to concentrate retrovirus particles (Morling and Russell, 1995). In this case, virus pellets can be re-dissolved by chelation using EDTA and re-concentrated (Pham et al., 2001). However, high concentrations of EDTA have been shown to affect virus stability. Moreover, the use of salts for virus precipitation may contribute to the loss of active viral particles due to changes in osmotic pressure. In order to minimize virus inactivation during the procedure, immediate dialysis of the concentrated preparation was performed that resulted in satisfactory virus recoveries (50-60%) (Pham et al., 2001).
Ultrafiltration is the preferred method for large-scale processing of retroviral particles because it allows gentle processing of large volumes of supernatant in a relatively short time (Table II). In contrast to the standard concentration methods discussed above, filtration processes involve no change of phase (liquid to solid) which may be traumatic enough to cause virus inactivation (Lyddiatt and O'Sullivan, 1998). Viral particles are enriched in the retentate while water and small molecular weight molecules are removed with the permeate. It should be noted that in order to keep viral particles in the retentate 100,000 or 300,000 molecular weight cut-off (MWCO) membranes are most often employed (Table II). Membrane processes offer the possibility of washing off impurities (ultra/diafiltration), thus achieving greater levels of purity. In addition, the retentate could be diafiltered against equilibration buffer used for chromatography.
Ultrafiltration can be carried out using a variety of filtration devices. At small scale, centrifugal filtration devices usually work well (Reiser, 2000). To process small to medium volumes of vector stocks (10 mL to 2 L) stirred cell tanks are ideal (Miller et al., 1996). Stability of retroviral particles using this method was found to be strongly dependent on the ultrafiltration operational parameters including pressure, stirring rate and process time (Cruz et al., 2000). By keeping these variables low we were able to concentrate vector supernatants 20-fold with excellent recovery of active particles and simultaneously remove significant amounts of serum proteins, degraded DNA fragments and inhibitors of transduction (Segura et al., 2005). Larger volumes of retroviral vectors (8 to 10 L) were concentrated by tangential-flow filtration achieving high recovery of active vector particles (Kotani et al., 1994). Tangential-flow hollow fiber filters were successfully employed for the concentration of wild-type retroviruses as well as retroviral vectors that resulted in 10 to 40-fold concentration of supernatants with good recovery of retrovirus activity (Makino et al., 1994; Paul et al., 1993). This strategy is currently being used for the concentration of lentiviral vectors for phase I clinical trials (Slepushkin et al., 2003).
Membrane fouling is the main problem faced during ultrafiltration since it causes the flow rate to decrease over time. To keep process time within reasonable limits, without increasing operating pressures that might affect the virus stability, it is often necessary to restrict the volume reduction.
(a) Concentration factors are based on volume change, (b) Recoveries are based on infectious particle quantitation
Concentrated viral stocks need to be further purified in order to obtain high quality vector preparations that meet regulatory requirements for clinical applications. The separation of retroviral particles from the remaining contaminating substances present in clarified concentrates is generally divided in two distinct operations: purification and polishing. The former is aimed at eliminating the bulk of impurities while the latter removes remaining low or trace amounts of contaminants and closely-related virus structures.
Density gradient ultracentrifugation is a powerful method for purifying retroviruses. More importantly, this is one of the few methods that offers potential to separate viral particles from closely-related species such as defective vector forms and/or cell membrane vesicles, all of which pose a serious challenge in downstream processing. Equilibrium density ultracentrifugation in sucrose gradients is the most widely used method for the preparation of highly purified retrovirus material for characterization of viral proteins and enzyme activities (Vogt, 1997). Using this technique, retrovirus particles are isolated from a band at a density of ~1.16 g/mL, corresponding to 35% (w/w) sucrose. Unfortunately, this technique often results in poor recovery of infective particles and is not reproducible. Therefore, this technique is best suited for studies whereby the preservation of viral activity is not required. Moreover, virus preparations obtained by this procedure are usually contaminated with variable amounts of cell membrane vesicles (microvesicles or exosomes) that have a density similar to that of the virus (Bess et al., 1997; Gluschankof et al., 1997). Since these vesicles show a wider range of size (50-500 nm), higher levels of purification can be achieved by rate zonal ultracentrifugation. In this case, separation of viral particles from contaminants is based on size and density, in contrast to the standard equilibrium ultracentrifugation procedure in which irrespective of the size, particles are separated according to their buoyant density alone. Rate zonal ultracentrifugation showed promising results in studies performed in our laboratory (Chapter III).
The most widely used gradient media for virus purification are sucrose and cesium chloride (CsCl). Both media are hyperosmotic at the densities used to band retrovirus particles. Sucrose solutions are very viscous and thus require longer sedimentation times for efficient separation of virus particles. Moreover, the high viscosity of sucrose has been associated with loss of surface structures and thus loss of infectivity upon purification (Moller-Larsen and Christensen, 1998). The use of iodixanol, a relatively new gradient medium, has also been described for the purification of retrovirus particles. This gradient medium can be diluted in iso-osmotic buffers to form iso-osmotic solutions that help preserve retrovirus particle integrity and functionality (Dettenhofer and Yu, 1999; Moller-Larsen and Christensen, 1998). In addition, it is less viscous than sucrose resulting in shorter processing times. Moreover, this medium, originally designed as an X-ray contrast solution, is non-toxic to cells.
Several practical disadvantages are associated with density gradient ultracentrifugation methods. The preparation of density gradients requires technical expertise, time and patience. The method is currently not being used at large scale since it would require the use of costly equipment that has not yet been tested for the purification of viruses. Additionally, separations usually require long processing times which may be detrimental to preserving retroviral infectivity. Therefore, the adoption of adsorptive chromatographic procedures has been strongly encouraged to move away from these conventional virus purification procedures (Andreadis et al., 1999; Braas et al., 1996; Lyddiatt and O'Sullivan, 1998).
Chromatography is the method of choice for selective fractionation of bioproducts in large-scale since it enables fast, efficient and reproducible separations (Lyddiatt, 2002). In chromatography, clarified and usually concentrated retroviral stocks are passed though a column containing beads coated with functional groups that capture the viral particles while the rest of the solution containing undesired impurities passes through. Captured particles are then displaced from the column using desorption agents and collected in purified fractions. This process is currently being employed for the purification of plasmid DNA (Ferreira et al., 2000; Stadler et al., 2004) as well as viral gene therapy vectors including adenovirus and adeno-associated virus vectors (Arcand et al., 2003; Davidoff et al., 2004; Debelak et al., 2000; Smith et al., 2003; Zolotukhin et al., 2002). A number of adsorptive chromatography procedures have been described for the purification of retroviral particles (Table III). In contrast with the techniques mentioned above, that separate virus particles simply based on size and density, adsorptive chromatography can purify vectors based on chemical surface properties or the molecular composition of the viral envelope. Therefore, chromatographic purification greatly contributes to the manufacture of high-purity vector stocks for clinical applications. However, with the exemption of immunoaffinity chromatography, most chromatography methods are unlikely to remove a significant amount of closely-related species such as defective vector forms and/or cell membrane vesicles from viral preparations.
Ion-exchange chromatography
Anion exchange chromatography exploits the negatively charged surface of retroviruses for purification purposes. Retroviral particles bind strongly to anion exchangers carrying positively charged quaternary ammonium functional groups. Anion exchange chromatography, first used for the preparation of an inactivated HIV-1 vaccine (Prior et al., 1995), was more recently adapted to the purification of lentiviral vectors (Scherr et al., 2002; Yamada et al., 2003). Similarly, hydroxyapatite chromatography matrices, originally used to purify wild-type inactive MoMLV (Smith and Lee, 1978), were found to bind oncoretroviral vectors, although only moderate recoveries of infective particles (18-31%) were obtained (Kuiper et al., 2002). The mechanism of interaction of hydroxyapatite resins is not completely understood, but it appears to be a combined effect of anion exchange, cation exchange and calcium coordination (Gagnon, 1998). Ion exchange matrices show imperfect selectivity and high salt concentrations are required to elute retroviral particles. Consequently, in most cases further purification steps are required to eliminate similarly charged contaminants (i.e. DNA) and salt.
Affinity chromatography
In order to limit the number of purification steps, highly selective affinity chromatographic adsorbents would be ideal. Unfortunately, little is known about the composition of the viral membrane, which complicates the selection of suitable adsorbents. A possibility is to engineer vectors to contain affinity tags inserted on the surface of the virus to facilitate their purification. Hexahistidine affinity tags, often used for the purification of recombinant proteins by immobilized metal affinity chromatography (IMAC), have been inserted into the MoMLV ecotropic Env-protein. His6-tagged retroviruses showed high affinity for immobilized nickel ions and were succesfully purified by one-step IMAC with good recovery of infective particles (Ye et al., 2004). Additionally, Williams and collaborators (Williams et al., 2005b) have demonstrated the feasibility of purifying MoMLV particles by exploiting the interaction between streptavidin and biotin. Chemically biotinylated oncoretroviral particles strongly bound streptavidin coated adsorbents in batch experiments. However, low recoveries of infective particles were obtained after elution with optimized concentrations of d-biotin (maximum recovery ~17%).
Engineering vectors by inserting tags or chemically modifying the envelope structure without reducing or eliminating the virus ability to transduce cells has proved to be a difficult task as demonstrated by unsuccessful efforts to alter the structure of envelope proteins for targeting purposes (Palù et al., 2000). Another possibility is to explore the natural ability of retroviruses to bind commercially available affinity ligands. Heparin affinity chromatography proved to be a powerful tool for the purification of viruses that use heparan sulfate as cell surface receptor, including herpes simplex virus, foot and mouth disease virus and adeno-associated viral vectors (Navarro del Cañizo et al., 1996; O'Keeffe et al., 1999; Zolotukhin et al., 1999). More recently, heparin affinity chromatography was found to be useful for the purification of oncoretroviral vectors giving excellent results in terms of recovery of active particles (61%), reproducibility and selectivity (Segura et al., 2005). Although the method was described for the purification of VSV-G pseudotyped particles, we have recently found that other vector pseudotypes also show affinity for heparin ligands (Chapter IV).
The most selective affinity chromatography technique is immunoaffinity chromatography which relies on the specific interaction between immobilized antibodies and surface viral antigens. This technique could separate retroviral particles from cell membrane vesicles that frequently contaminate retrovirus preparations provided that a surface protein is found to be exclusively incorporated into either the virions or the vesicles. For instance, taking advantage of the differential incorporation of CD45 into HIV-1 and cell membrane vesicles derived from lymphoid cells (Esser et al., 2001), Trubey et al. developed an immunoaffinity approach to selectively deplete these vesicles from density-purified retrovirus preparations (Trubey et al., 2003). However, non-hematopoietic cells (i.e. HEK 293) are not expected to express CD45, limiting the usefulness of this technique. Moreover, the high costs associated with antibody purification and immobilization and the low stability of these ligands towards sanitizing agents do not favour the use of this method for large-scale (Andreadis et al., 1999).
Size-exclusion chromatography
Size-exclusion chromatography (SEC) was successfully used for the purification of wild-type retroviruses and retroviral vectors (McGrath et al., 1978; Slepushkin et al., 2003; Transfiguracion et al., 2004). Using this chromatographic method, retroviruses are excluded from the internal pores of the gel due to their large size and elute in the void volume of the column while low molecular weight contaminants are retarded by the column. Unfortunately, SEC is a non-adsorptive method likely to be useful only as a final polishing step due to its limited resolution and inherent low capacity (<10% bed volume for best peak resolution). Moreover, SEC using conventional matrices tends to operate at low linear flow rates (~15 cm/h) and typically results in product dilution of 2 to 4 folds.
Most currently available chromatographic matrices were designed to maximize the adsorption of protein macromolecules (diameter <5nm) rather than viruses. Consideration of the pore dimensions of most commercially available chromatography adsorbents (typically 30-80 nm) suggests that adsorption of retroviruses will be restricted to the bead surface area while most contaminating proteins have access to the area inside the pores as well (Lyddiatt and O'Sullivan, 1998). Therefore, using conventional matrices, both available binding capacity and purification efficiency are predicted to be poor.
In order to circumvent problems found with conventional chromatography supports, several new chromatography technologies have been proposed and tested. Tentacle supports offer the possibility of increased virus binding capacities. The advantage of using these supports is that they have sterically accessible ligands available for virus capture. The ligands are attached to an inert and flexible spacer arm that separates them from the bead. Therefore, tentacle ligands can access otherwise sterically hindered binding sites and compensate in part for the loss of surface area inside the pores. In addition, since they are no longer exclusively on the surface of the chromatographic bead, larger amounts of ligands are available for binding (Kaufmann, 1997). Tentacle matrices have been employed for the purification of inactivated HIV-1 for viral vaccines and oncoretroviral particles (Prior et al., 1995; Segura et al., 2005; Williams et al., 2005b). The superiority of these supports compared to non-tentacle ones was demonstrated in these reports. Membrane chromatography is another interesting alternative to traditional column chromatography since it combines the advantages of membrane technology (high flow rates) and liquid chromatography (high selectivity). Anion exchange membrane adsorbers have been tested for the purification of lentiviral vectors showing excellent results (Slepushkin et al., 2003). Another way to improve virus purification using chromatography is to use monolithic adsorbents instead of the traditional biopolymers (or so–called soft supports). The application of compact, macroporous monoliths as supports for the purification of retroviral vectors has been described (Williams et al., 2005a). Biotinylated oncoretroviral particles were recovered by elution from streptavidin coated monoliths with 8% recovery of infective particles.
Another key-point to consider for the successful recovery of active viral particles is to select adsorbents that do not require harsh conditions for virus adsorption and/or elution. Most methods developed so far maintain physiological pH throughout the whole purification process (Table III). Elution of retrovirus particles from heparin affinity columns is achieved under mild conditions (neutral pH and 0.35 M NaCl). In contrast, elution of strongly bound viral particles from anion exchange chromatography adsorbents requires the addition of high salt concentrations (~1 M NaCl) which may compromise the activity of retroviral particles. The concentration of imidazole and d-biotin for the elution of retroviral particles from Ni-NTA and streptavidin affinity matrices respectively was adjusted to minimize the detrimental effect of these desorption agents on virus activity (Williams et al., 2005b; Ye et al., 2004). Immunoaffinity chromatography usually requires stringent elution conditions to break antibody-antigen interactions including low pH, high salt or the use of denaturing agents. Therefore, using this method for virus capture, low recoveries of active particles are predictable upon elution.
(a) Recoveries are titer-based unless otherwise indicated, (b) The authors found two peaks of activity using a linear NaCl gradient, (c) Pool of fractions obtained throughout a gradient (0-0.5 M NaCl)
Abbreviation: N/A not applicable
Retrovirus supernatants are commonly aliquoted and frozen at –80°C to protect the virus from thermal inactivation. These stocks were shown to maintain their potency over months (McTaggart and Al-Rubeai, 2000; Wikstrom et al., 2004). However, there is a high cost and impracticality associated with long-term cryostorage of large volumes of supernatant which encourages the development of alternative storage strategies (Andreadis et al., 1999; McTaggart and Al-Rubeai, 2002). Retroviral particles can be preserved in a lyophilized form (Levy and Fieldsteel, 1982). The lyophilization of retroviral vector supernatants with and without additives was investigated by Kotani and collaborators (Kotani et al., 1994). The authors found that the recovery of active retroviral particles after lyophilization is more efficient in the presence of glucose or sorbitol with gelatin (64-83 %) than without these additives (21%). Conversely, Lee and collaborators observed a severe loss of virus activity during lyophilization (>60%) that could not be circumvented by addition of additives (Lee et al., 1996). The authors also reported that commonly used cryoprotectants, DMSO and glycerol, did not play an important role in preserving retroviral activity during long-term cryostorage. Little is known about the stability of purified vector stocks during storage. More research in this area will be required.
Retroviral vectors constitute a valuable tool for gene transfer technology. The wide clinical application of these vectors for gene therapy will depend on the availability of efficient large-scale manufacturing procedures. Significant advances in the downstream processing of retroviral vectors have been made in the past several years. Studies have shown that the potency of retroviral preparations can be significantly improved by concentration and removal of inhibitors of cell transduction. Various selective chromatography matrices have been identified and new chromatography technologies, better suited for virus purification purposes, are being tested with very promising results. However, most of these techniques have only been tested at laboratory scale. Very few reports show a complete purification scheme, from retrovirus supernatant to clinical grade virus, in which final overall recoveries are presented (Kuiper et al., 2002; Segura et al., 2005; Slepushkin et al., 2003). Yet, they indicate that overall recoveries of active retrovirus particles in the range of 30% should be considered satisfactory. Considering the instability of retroviral vectors and its susceptibility to several factors as discussed in this review, we would like to emphasize the importance of selecting gentle purification procedures. The yield in each step is critical in view of the fact that even small losses in each purification step will have a strong impact in the final overall recovery when several purification steps are required to reach the desired level of purity. On the other hand, in most reported cases the final purities achieved are poorly documented and the ability (or inability) of the different methods to remove closely related structures such as cell membrane vesicles or defective virus particles, or the possible implications of having these contaminants in clinical grade preparations, are rarely discussed. Moreover, studies concerning the proper conditions and procedures for formulation and long-term storage of purified material are often incomplete, or non-existent. Undoubtedly, downstream processing of retroviral vectors will continue to be an area of intensive research in the coming years.
The authors wish to thank Rowe Gerald, Normand Arcand and Gavin Whissell for the careful review of this manuscript and helpful discussions. This work was supported by a NSERC Strategic Project grant and the NCE Canadian Stem Cell Network.
Aboud M, Wolfson M, Hassan Y, Huleihel M. 1982. Rapid purification of extracellular and intracellular Moloney murine leukemia virus. Arch Virol 71(3):185-95.
Alain R. 1997. Quantitation of virus particles by negative stain electron microscopy. Microscopy Today 97(5):20.
Andreadis S, Lavery T, Davis HE, Le Doux JM, Yarmush ML, Morgan JR. 2000. Toward a more accurate quantitation of the activity of recombinant retroviruses: alternatives to titer and multiplicity of infection. J Virol 74(7):3431-9.
Andreadis ST, Brott D, Fuller AO, Palsson BO. 1997. Moloney murine leukemia virus-derived retroviral vectors decay intracellularly with a half-life in the range of 5.5 to 7.5 hours. J Virol 71(10):7541-8.
Andreadis ST, Roth CM, Le Doux JM, Morgan JR, Yarmush ML. 1999. Large-scale processing of recombinant retroviruses for gene therapy. Biotechnol Prog 15(1):1-11.
Arcand N, Bernier A, Transfiguracion J, Jacob D, Coelho H, Kamen A. 2003. Adenovirus type 5 (Ad5) chromatographic purification process at the 20L scale. Bioprocessing J 2:72-75.
Baekelandt V, Eggermont K, Michiels M, Nuttin B, Debyser Z. 2003. Optimized lentiviral vector production and purification procedure prevents immune response after transduction of mouse brain. Gene Ther 10(23):1933-40.
Bajaj B, Lei P, Andreadis ST. 2001. High efficiencies of gene transfer with immobilized recombinant retrovirus: kinetics and optimization. Biotechnol Prog 17(4):587-96.
Beer C, Meyer A, Muller K, Wirth M. 2003. The temperature stability of mouse retroviruses depends on the cholesterol levels of viral lipid shell and cellular plasma membrane. Virology 308(1):137-46.
Bess JW, Jr., Gorelick RJ, Bosche WJ, Henderson LE, Arthur LO. 1997. Microvesicles are a source of contaminating cellular proteins found in purified HIV-1 preparations. Virology 230(1):134-44.
Beyer WR, Westphal M, Ostertag W, von Laer D. 2002. Oncoretrovirus and lentivirus vectors pseudotyped with lymphocytic choriomeningitis virus glycoprotein: generation, concentration, and broad host range. J Virol 76(3):1488-95.
Bowles NE, Eisensmith RC, Mohuiddin R, Pyron M, Woo SL. 1996. A simple and efficient method for the concentration and purification of recombinant retrovirus for increased hepatocyte transduction in vivo. Hum Gene Ther 7(14):1735-42.
Braas G, Searle PF, Slater NK, Lyddiatt A. 1996. Strategies for the isolation and purification of retroviral vectors for gene therapy. Bioseparation 6(4):211-28.
Burns JC, Friedmann T, Driever W, Burrascano M, Yee JK. 1993. Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: concentration to very high titer and efficient gene transfer into mammalian and nonmammalian cells. Proc Natl Acad Sci U S A 90(17):8033-7.
Carmo M, Peixoto C, Coroadinha AS, Alves PM, Cruz PE, Carrondo MJ. 2004. Quantitation of MLV-based retroviral vectors using real-time RT-PCR. J Virol Methods 119(2):115-9.
CBER. 2001. Supplemental guidance on testing for replication-competent retrovirus in retroviral vector-based gene therapy products and during follow-up of patients in clinical trials using retroviral vectors. Hum Gene Ther 12(3):315-20.
Chang LJ, Zaiss AK. 2002. Lentiviral vectors. Preparation and use. Methods Mol Med 69:303-18.
Chen J, Reeves L, Sanburn N, Croop J, Williams DA, Cornetta K. 2001. Packaging cell line DNA contamination of vector supernatants: implication for laboratory and clinical research. Virology 282(1):186-97.
Cruz PE, Goncalves D, Almeida J, Moreira JL, Carrondo MJ. 2000. Modeling retrovirus production for gene therapy. 2. Integrated optimization of bioreaction and downstream processing. Biotechnol Prog 16(3):350-7.
Davidoff AM, Ng CY, Sleep S, Gray J, Azam S, Zhao Y, McIntosh JH, Karimipoor M, Nathwani AC. 2004. Purification of recombinant adeno-associated virus type 8 vectors by ion exchange chromatography generates clinical grade vector stock. J Virol Methods 121(2):209-15.
Debelak D, Fisher J, Iuliano S, Sesholtz D, Sloane DL, Atkinson EM. 2000. Cation-exchange high-performance liquid chromatography of recombinant adeno-associated virus type 2. J Chromatogr B Biomed Sci Appl 740(2):195-202.
Dettenhofer M, Yu XF. 1999. Highly purified human immunodeficiency virus type 1 reveals a virtual absence of Vif in virions. J Virol 73(2):1460-7.
Dull T, Zufferey R, Kelly M, Mandel RJ, Nguyen M, Trono D, Naldini L. 1998. A third-generation lentivirus vector with a conditional packaging system. J Virol 72(11):8463-71.
Esser MT, Graham DR, Coren LV, Trubey CM, Bess JW, Jr., Arthur LO, Ott DE, Lifson JD. 2001. Differential incorporation of CD45, CD80 (B7-1), CD86 (B7-2), and major histocompatibility complex class I and II molecules into human immunodeficiency virus type 1 virions and microvesicles: implications for viral pathogenesis and immune regulation. J Virol 75(13):6173-82.
Ferreira GN, Monteiro GA, Prazeres DM, Cabral JM. 2000. Downstream processing of plasmid DNA for gene therapy and DNA vaccine applications. Trends Biotechnol 18(9):380-8.
Flotte T, Burd P, Snyder R. 2002. Utility of a recombinant adeno-associated viral vector reference standard. Bioprocessing J:75-77.
Gagnon P. 1998. An Enigma Unmasked: How Hydroxyapatite Works and How to Make It Work for You Validated Biosystems Quarterly Web Newsletter.
Gatlin J, Melkus MW, Padgett A, Kelly PF, Garcia JV. 2001. Engraftment of NOD/SCID mice with human CD34(+) cells transduced by concentrated oncoretroviral vector particles pseudotyped with the feline endogenous retrovirus (RD114) envelope protein. J Virol 75(20):9995-9.
Gluschankof P, Mondor I, Gelderblom HR, Sattentau QJ. 1997. Cell membrane vesicles are a major contaminant of gradient-enriched human immunodeficiency virus type-1 preparations. Virology 230(1):125-33.
Higashikawa F, Chang L. 2001. Kinetic analyses of stability of simple and complex retroviral vectors. Virology 280(1):124-31.
Kaufmann M. 1997. Unstable proteins: how to subject them to chromatographic separations for purification procedures. J Chromatogr B Biomed Sci Appl 699(1-2):347-69.
Kotani H, Newton PB, 3rd, Zhang S, Chiang YL, Otto E, Weaver L, Blaese RM, Anderson WF, McGarrity GJ. 1994. Improved methods of retroviral vector transduction and production for gene therapy. Hum Gene Ther 5(1):19-28.
Kuiper M, Sanches RM, Walford JA, Slater NK. 2002. Purification of a functional gene therapy vector derived from Moloney murine leukaemia virus using membrane filtration and ceramic hydroxyapatite chromatography. Biotechnol Bioeng 80(4):445-53.
Kwon YJ, Hung G, Anderson WF, Peng CA, Yu H. 2003. Determination of infectious retrovirus concentration from colony-forming assay with quantitative analysis. J Virol 77(10):5712-20.
Kwon YJ, Peng CA. 2002. Transduction rate constant as more reliable index quantifying efficiency of retroviral gene delivery. Biotechnol Bioeng 77(6):668-77.
Le Doux JM, Davis HE, Morgan JR, Yarmush ML. 1999. Kinetics of retrovirus production and decay. Biotechnol Bioeng 63(6):654-62.
Le Doux JM, Landazuri N, Yarmush ML, Morgan JR. 2001. Complexation of retrovirus with cationic and anionic polymers increases the efficiency of gene transfer. Hum Gene Ther 12(13):1611-21.
Le Doux JM, Morgan JR, Snow RG, Yarmush ML. 1996. Proteoglycans secreted by packaging cell lines inhibit retrovirus infection. J Virol 70(9):6468-73.
Le Doux JM, Morgan JR, Yarmush ML. 1998. Removal of proteoglycans increases efficiency of retroviral gene transfer. Biotechnol Bioeng 58(1):23-34.
Lee SG, Kim S, Robbins PD, Kim BG. 1996. Optimization of environmental factors for the production and handling of recombinant retrovirus. Appl Microbiol Biotechnol 45(4):477-83.
Levy JA, Fieldsteel AH. 1982. Freeze-drying is an effective method for preserving infectious type C retroviruses. J Virol Methods 5(3-4):165-71.
Lizee G, Aerts JL, Gonzales MI, Chinnasamy N, Morgan RA, Topalian SL. 2003. Real-time quantitative reverse transcriptase-polymerase chain reaction as a method for determining lentiviral vector titers and measuring transgene expression. Hum Gene Ther 14(6):497-507.
Logan AC, Nightingale SJ, Haas DL, Cho GJ, Pepper KA, Kohn DB. 2004. Factors influencing the titer and infectivity of lentiviral vectors. Hum Gene Ther 15(10):976-88.
Lyddiatt A. 2002. Process chromatography: current constraints and future options for the adsorptive recovery of bioproducts. Curr Opin Biotechnol 13(2):95-103.
Lyddiatt A, O'Sullivan DA. 1998. Biochemical recovery and purification of gene therapy vectors. Curr Opin Biotechnol 9(2):177-85.
Makino M, Ishikawa G, Yamaguchi K, Okada Y, Watanabe K, Sasaki-Iwaki Y, Manabe S, Honda M, Komuro K. 1994. Concentration of live retrovirus with a regenerated cellulose hollow fiber, BMM. Arch Virol 139(1-2):87-96.
McGrath M, Witte O, Pincus T, Weissman IL. 1978. Retrovirus purification: method that conserves envelope glycoprotein and maximizes infectivity. J Virol 25(3):923-7.
McTaggart S, Al-Rubeai M. 2000. Effects of culture parameters on the production of retroviral vectors by a human packaging cell line. Biotechnol Prog 16(5):859-65.
McTaggart S, Al-Rubeai M. 2002. Retroviral vectors for human gene delivery. Biotechnol Adv 20(1):1-31.
Merten OW. 2004. State-of-the-art of the production of retroviral vectors. J Gene Med 6 Suppl 1:S105-24.
Miller DL, Meikle PJ, Anson DS. 1996. A rapid and efficient method for concentration of small volumes of retroviral supernatant. Nucleic Acids Res 24(8):1576-7.
Moller-Larsen A, Christensen T. 1998. Isolation of a retrovirus from multiple sclerosis patients in self-generated Iodixanol gradients. J Virol Methods 73(2):151-61.
Morling FJ, Russell SJ. 1995. Enhanced transduction efficiency of retroviral vectors coprecipitated with calcium phosphate. Gene Ther 2(7):504-8.
Moy N, Dillingham M, McGowan S, Rotherham M, Clayton T. 2000. Large scale GMP production of retroviral gene therapy vector in very low serum medium. Bernard A, Griffiths B, Noé W, Wurm F, editors: Kluwer Academic publishers. 441-443 p.
Naldini L, Blomer U, Gallay P, Ory D, Mulligan R, Gage FH, Verma IM, Trono D. 1996. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 272(5259):263-7.
Navarro del Cañizo AA, Mazza M, Bellinzoni R, Cascone O. 1996. Foot and mouth disease virus concentration and purification by affinity chromatography. Appl Biochem Biotechnol 61(3):399-409.
Neff T, Peterson LJ, Morris JC, Thompson J, Zhang X, Horn PA, Thomasson BM, Kiem HP. 2004. Efficient gene transfer to hematopoietic repopulating cells using concentrated RD114-pseudotype vectors produced by human packaging cells. Mol Ther 9(2):157-9.
O'Keeffe RS, Johnston MD, Slater NK. 1999. The affinity adsorptive recovery of an infectious herpes simplex virus vaccine. Biotechnol Bioeng 62(5):537-45.
Ott DE. 2002. Potential roles of cellular proteins in HIV-1. Rev Med Virol 12(6):359-74.
Palù G, Parolin C, Takeuchi Y, Pizzato M. 2000. Progress with retroviral gene vectors. Rev Med Virol 10:185-202.
Pan D, Gunther R, Duan W, Wendell S, Kaemmerer W, Kafri T, Verma IM, Whitley CB. 2002. Biodistribution and toxicity studies of VSVG-pseudotyped lentiviral vector after intravenous administration in mice with the observation of in vivo transduction of bone marrow. Mol Ther 6(1):19-29.
Paul RW, Morris D, Hess BW, Dunn J, Overell RW. 1993. Increased viral titer through concentration of viral harvests from retroviral packaging lines. Hum Gene Ther 4(5):609-15.
Pham L, Ye H, Cosset FL, Russell SJ, Peng KW. 2001. Concentration of viral vectors by co-precipitation with calcium phosphate. J Gene Med 3(2):188-94.
Pizzato M, Marlow SA, Blair ED, Takeuchi Y. 1999. Initial binding of murine leukemia virus particles to cells does not require specific Env-receptor interaction. J Virol 73(10):8599-611.
Prior C, Bay P, Ebert B, Gore R, Holt J, Irish T, Jensen F, Leone C, Mitschelen J, Stiglitz M and others. 1995. Process development for the manufacture of inactivated HIV-1. BioPharmacology 8:25-35.
Rainov NG, Ren H. 2003. Clinical trials with retrovirus mediated gene therapy--what have we learned? J Neurooncol 65(3):227-36.
Reeves L, Cornetta K. 2000. Clinical retroviral vector production: step filtration using clinically approved filters improves titers. Gene Ther 7(23):1993-8.
Reiser J. 2000. Production and concentration of pseudotyped HIV-1-based gene transfer vectors. Gene Ther 7(11):910-3.
Rigg RJ, Chen J, Dando JS, Forestell SP, Plavec I, Bohnlein E. 1996. A novel human amphotropic packaging cell line: high titer, complement resistance, and improved safety. Virology 218(1):290-5.
Sastry L, Johnson T, Hobson MJ, Smucker B, Cornetta K. 2002. Titering lentiviral vectors: comparison of DNA, RNA and marker expression methods. Gene Ther 9(17):1155-62.
Sastry L, Xu Y, Cooper R, Pollok K, Cornetta K. 2004. Evaluation of plasmid DNA removal from lentiviral vectors by benzonase treatment. Hum Gene Ther 15(2):221-6.
Scherr M, Battmer K, Eder M, Schule S, Hohenberg H, Ganser A, Grez M, Blomer U. 2002. Efficient gene transfer into the CNS by lentiviral vectors purified by anion exchange chromatography. Gene Ther 9(24):1708-14.
Segura M, Kamen A, Trudel P, Garnier A. 2005. A novel purification strategy for retrovirus gene therapy vectors using heparin affinity chromatography. Biotechnol Bioeng 90(4):391-404.
Slepushkin V, Chang N, Cohen R, Gan Y, Jiang B, Deausen E, Berlinger D, Binder G, Andre K, Humeau L and others. 2003. Large-scale purification of a lentiviral vector by size exclusion chromatography or mustang Q ion exchange capsule. Bioprocessing J 2:89-95.
Slingsby JH, Baban D, Sutton J, Esapa M, Price T, Kingsman SM, Kingsman AJ, Slade A. 2000. Analysis of 4070A envelope levels in retroviral preparations and effect on target cell transduction efficiency. Hum Gene Ther 11(10):1439-51.
Smith KT, Shepherd AJ, Boyd JE, Lees GM. 1996. Gene delivery systems for use in gene therapy: an overview of quality assurance and safety issues. Gene Ther 3(3):190-200.
Smith RG, Lee SA. 1978. Large-scale isolation and partial purification of type C RNA viruses on hydroxyapatite. 1. Biochemical characterization. Anal Biochem 86(1):252-63.
Smith RH, Ding C, Kotin RM. 2003. Serum-free production and column purification of adeno-associated virus type 5. J Virol Methods 114(2):115-24.
Srinivasakumar N. 2002. Packaging cell system for lentivirus vectors. Preparation and use. Methods Mol Med 69:275-302.
Stadler J, Lemmens R, Nyhammar T. 2004. Plasmid DNA purification. J Gene Med 6 Suppl 1:S54-66.
Transfiguracion J, Coelho H, Kamen A. 2004. High-performance liquid chromatographic total particles quantification of retroviral vectors pseudotyped with vesicular stomatitis virus-G glycoprotein. J Chromatogr B Analyt Technol Biomed Life Sci 813(1-2):167-73.
Transfiguracion J, Jaalouk DE, Ghani K, Galipeau J, Kamen A. 2003. Size-exclusion chromatography purification of high-titer vesicular stomatitis virus G glycoprotein-pseudotyped retrovectors for cell and gene therapy applications. Hum Gene Ther 14(12):1139-53.
Trubey CM, Chertova E, Coren LV, Hilburn JM, Hixson CV, Nagashima K, Lifson JD, Ott DE. 2003. Quantitation of HLA class II protein incorporated into human immunodeficiency type 1 virions purified by anti-CD45 immunoaffinity depletion of microvesicles. J Virol 77(23):12699-709.
Vogt VM. 1997. Retroviral virions and genomes. In: Coffin JM HS, Varmus HE., editor. Retroviruses. NY: Cold Spring Harbor Laboratory Press. p 27-69.
Vogt VM, Simon MN. 1999. Mass determination of rous sarcoma virus virions by scanning transmission electron microscopy. J Virol 73(8):7050-5.
White SM, Renda M, Nam NY, Klimatcheva E, Zhu Y, Fisk J, Halterman M, Rimel BJ, Federoff H, Pandya S and others. 1999. Lentivirus vectors using human and simian immunodeficiency virus elements. J Virol 73(4):2832-40.
Wikstrom K, Blomberg P, Islam KB. 2004. Clinical grade vector production: analysis of yield, stability, and storage of gmp-produced retroviral vectors for gene therapy. Biotechnol Prog 20(4):1198-203.
Williams SL, Eccleston ME, Slater NK. 2005a. Affinity capture of a biotinylated retrovirus on macroporous monolithic adsorbents: towards a rapid single-step purification process. Biotechnol Bioeng 89(7):783-7.
Williams SL, Nesbeth D, Darling DC, Farzaneh F, Slater NK. 2005b. Affinity recovery of Moloney Murine Leukaemia Virus. J Chromatogr B Analyt Technol Biomed Life Sci 820(1):111-9.
Yamada K, McCarty DM, Madden VJ, Walsh CE. 2003. Lentivirus vector purification using anion exchange HPLC leads to improved gene transfer. Biotechniques 34(5):1074-8, 1080.
Ye K, Dhiman HK, Suhan J, Schultz JS. 2003. Effect of pH on infectivity and morphology of ecotropic moloney murine leukemia virus. Biotechnol Prog 19(2):538-43.
Ye K, Jin S, Ataai MM, Schultz JS, Ibeh J. 2004. Tagging retrovirus vectors with a metal binding peptide and one-step purification by immobilized metal affinity chromatography. J Virol 78(18):9820-7.
Zhang B, Xia HQ, Cleghorn G, Gobe G, West M, Wei MQ. 2001. A highly efficient and consistent method for harvesting large volumes of high-titre lentiviral vectors. Gene Ther 8(22):1745-51.
Zolotukhin S, Byrne BJ, Mason E, Zolotukhin I, Potter M, Chesnut K, Summerford C, Samulski RJ, Muzyczka N. 1999. Recombinant adeno-associated virus purification using novel methods improves infectious titer and yield. Gene Ther 6(6):973-85.
Zolotukhin S, Potter M, Zolotukhin I, Sakai Y, Loiler S, Fraites TJ, Jr., Chiodo VA, Phillipsberg T, Muzyczka N, Hauswirth WW and others. 2002. Production and purification of serotype 1, 2, and 5 recombinant adeno-associated viral vectors. Methods 28(2):158-67.
Zufferey R. 2002. Production of lentiviral vectors. Curr Top Microbiol Immunol 261:107-21.
© Maria Delas Mercèdes Segura, 2006
|
|
|
|
| Introduction |
|
Chapter II A novel purification strategy for retrovirus gene therapy vectors using heparin affinity chromatography |
