High efficiencies of gene transfer with immobilized recombinant retrovirus: kinetics and optimization

We used a combination of mathematical modeling and experiments to investigate the rate-limiting steps of retroviral transduction on surface-bound fibronectin (FN) and identify the conditions that maximize the efficiency of gene transfer. Our results show that fibronectin-assisted gene transfer (FAGT) is a strong function of the time and temperature of virus incubation in FN-coated plates. Gene transfer increases sharply at short times, reaches a maximum at intermediate times, and eventually declines as a result of loss of retroviral activity. The maximum transduction efficiency and the time at which this is attained increase with decreasing temperature of virus incubation. Depending on the temperature and the type of target cells, the initial rate of gene transfer increases by 3- to 10-fold and the maximum transduction efficiency increases by 2- to 4-fold as compared to traditional transduction (TT). Interestingly, Polybrene (PB) inhibits FAGT in a dose-dependent manner by inhibiting binding of retrovirus to FN. In contrast to traditional transduction, FAGT yields higher than 10-fold transduction efficiencies with concentrated retrovirus stocks. Gene transfer is directly proportional to the concentration of the virus-containing medium with no sign of saturation for the range of concentrations tested. These results suggest that immobilization of recombinant retrovirus can be rationally optimized to yield high efficiency of gene transfer to primary cells and improve the prospect of gene therapy for the treatment of human disease.     

Polybrene increases retrovirus gene transfer efficiency by enhancing receptor-independent virus adsorption on target cell membranes

Cationic polymers, such as polybrene and protamine sulfate, are typically used to increase the efficiency of retrovirus-mediated gene transfer, however, the mechanism of their enhancement of transduction has remained unclear. As retrovirus transduction is fundamentally limited by the slow diffusion of virus to the target cell surface, we investigated the ability of polybrene to modulate this initial transport step. We compared the ability of both envelope (gp70) and capsid (p30) protein based assays to quantitate virus adsorption and found that p30 based assays were more reliable due to their ability to distinguish virus binding from free gp70 binding. Using the p30 based assay, we established that polybrene concentrations, which yielded 10-fold increases in transduction also, yielded a significant increase in virus adsorption rates on murine fibroblasts. Surprisingly, this enhancement, and adsorption in general, were receptor and envelope independent, as adsorption occurred equivalently on receptor positive and negative Chinese hamster ovary cells, as well as with envelope positive and negative virus particles. These findings suggest that the currently accepted physical model for early steps in retrovirus transduction may need to be reformulated to accommodate an initial adsorption step whose driving force does not include the retrovirus concentration, and the reclassification of currently designated 'receptor' molecules as fusion triggers. The implication of these findings with respect to the development of targeted retrovirus-mediated gene therapy protocols is discussed.     

Complexation with chondroitin sulfate C and Polybrene rapidly purifies retrovirus from inhibitors of transduction and substantially enhances gene transfer

Using amphotropic retrovirus stocks produced by TELCeB6-A cells that encode the Escherichia coli lacZ gene, we found that complexation with chondroitin sulfate C (CSC) and Polybrene (PB) is an effective means to purify retrovirus. Virus stocks contained high levels of inhibitory activity that blocked amphotropic, but not ecotropic, retrovirus transduction. When virus stocks were brought to 80 microg/mL each of CSC and PB, complexes of CSC and PB formed. These complexes incorporated more than 70% of the virus particles but less than 0.4% of all other proteins and no detectable inhibitory activity. Purified virus transduced NIH 3T3 murine fibroblasts 21 to 186-fold more efficiently than virus that was not purified. In addition, virus purification significantly altered the dose response of transduction. When virus that had not been purified was used to transduce cells, the relationship between transduction and virus concentration was highly non-linear. In contrast, when purified virus was used, transduction increased monotonically and was linearly proportional to virus concentration, except when high doses of virus were used. Interestingly, when high doses of virus were used gene transfer reached a maximum plateau level, most likely because particle-associated amphotropic envelope proteins had saturated the cellular receptors for the virus. Our findings illustrate that retrovirus purification increases the maximum number of genes that can be transferred, reduces the amount of virus required to achieve a given level of gene transfer, and reduces uncertainties about the relationship between the amount of virus used and the number of genes transferred.     

Membrane adsorption characteristics determine the kinetics of flow-through transductions

Retrovirus-mediated gene transfer is currently limited by random Brownian motion of the retrovirus. This limitation can be overcome by flowing the retrovirus solution through a porous membrane that supports the target cells, leading to a significant increase in the transduction efficiency. This procedure is termed "flow-through transduction." In this study, we characterized the effects of the fluid flowrate and the influence that membrane characteristics have on the flow-through transduction procedure. The transduction efficiencies increased with flowrate until a plateau was reached at average flow velocities exceeding 0.3 cm/h for flow times of 3 to 4 h, using a collagen-coated depth (COL) membrane. A correlation between the optimal time for maximal gene transfer using flow-through transductions and the optimal time for maximal virus activity on the membrane was found, suggesting that the membrane adsorption capacity for virus determined the amount of gene transfer that could occur.Membrane adsorption characteristics were further investigated using two different membrane types: a tracketched polyester screen (PE) membrane and the COL membrane. Flow-through transductions using the PE and COL membranes showed that a high level of gene transfer could be attained using the COL membrane while the PE membrane gave much lower transduction efficiencies. The addition of the polycation polybrene (PB) changed these results markedly, making transductions achieved on the PE membrane similar in number to those obtained on the COL membrane. Since PB is believed to influence the virus adsorption to PE membrane, these results further support the conclusion that the increase in gene transfer achieved by the flow-through transduction procedure is due to virus adsorption to the membrane. The flow-through transduction procedure thus leads to co-localization of the viral vector and the target cell that in turn leads to a high transduction efficiency. (c) 1996 John Wiley & Sons, Inc.     

Entry of Amphotropic and 10A1 Pseudotyped Murine Retroviruses Is Restricted in Hematopoietic Stem Cell Lines

Although transduction with amphotropic murine leukemia virus (MLV) vectors has been optimized successfully for hematopoietic differentiated progenitors, gene transfer to early hematopoietic cells (stem cells) is still highly restricted. A similar restriction to gene transfer was observed in the mouse stem cell line FDC-Pmix compared with transfer in the more mature myeloid precursor cell line FDC-P1 and the human erythroleukemia cell line K562. Gene transfer was not improved when the vector was pseudotyped with gp70^SU of the 10A1 strain of MLV, which uses the receptor of the gibbon ape leukemia virus (Pit1), in addition to the amphotropic receptor (Pit2). Although 10A1 and amphotropic gp70^SU bound to FDC-P1, K562, and fibroblasts, no binding to FDC-Pmix cells was detected. This indicates that FDC-Pmix cells lack functional Pit2 and Pit1 receptors. Pseudotyping with the vesicular stomatitis virus G protein improved transduction efficiency in FDC-Pmix stem cells by 2 orders of magnitude, to fibroblast levels, confirming a block to retroviral infection at the receptor level.     

Rapid concentration and purification of retrovirus by flocculation with Polybrene

We have previously shown that the combined addition of Polybrene (PB) and chondroitin sulfate C (CSC) to retrovirus stocks leads to the formation of retrovirus-polymer complexes (i.e., flocs) that rapidly sediment onto cells, increases the efficiency of gene transfer, and can be used to rapidly concentrate and purify retrovirus stocks. The viruses remain associated with the polyelectrolyte complexes, however, which may complicate their use in downstream applications. In this study we determined if retrovirus could be flocculated using only one polymer (PB). We found that when retrovirus stocks were incubated with 320 microg/ml of PB, more than 70% of the viruses, and fewer than 0.3% of all other proteins, were pelleted by low-speed centrifugation. In contrast to retrovirus complexes formed with two polymers, retrovirus flocculated with PB disaggregated when they were resuspended in fresh medium. We conclude that flocculation of retroviruses with a single cationic polymer (PB) is a useful method for rapidly concentrating and purifying retroviruses, and may prove particularly useful when it is desirable to generate purified virus that is not part of a polymer complex.     

Heparan sulfate targets the HIV-1 envelope glycoprotein gp120 coreceptor binding site

Human immunodeficiency virus (HIV) attachment to host cells is a multi-step process that involves interaction of the viral envelope gp120 with the primary receptor CD4 and coreceptors. HIV gp120 also binds to other cell surface components, including heparan sulfate (HS), a sulfated polysaccharide whose wide interactive properties are exploited by many pathogens for attachment and concentration at the cell surface. To analyze the structural features of gp120 binding to HS, we used soluble CD4 to constrain gp120 in a specific conformation. We first found that CD4 induced conformational change of gp120, dramatically increasing binding to HS. We then showed that HS binding interface on gp120 comprised, in addition to the well characterized V3 loop, a CD4-induced epitope. This epitope is efficiently targeted by nanomolar concentrations of size-defined heparin/HS-derived oligosaccharides. Because this domain of the protein also constitutes the binding site for the viral coreceptors, these results support an implication of HS at late stages of the virus-cell attachment process and suggest potential therapeutic applications.     

The synthetic peptide P-197 inhibits human T-cell leukemia virus type 1 envelope-mediated syncytium formation by a mechanism that is independent of Hsc70

Entry of human T-cell leukemia virus type 1 (HTLV-1) into cells is mediated by the viral envelope glycoproteins gp46 and gp21. The gp46 surface glycoprotein binds to a poorly characterized cell surface receptor, thereby promoting the gp21-dependent fusion of the viral and cellular membranes. Interestingly, a synthetic peptide (P-197) simulating amino acids 197 to 216 of gp46 strongly inhibits envelope-dependent membrane fusion with Molt-4 target cells. It has been suggested that this peptide acts by competitively binding to Hsc70, a putative cellular receptor for HTLV-1. We now demonstrate that P-197 inhibits membrane fusion among diverse HTLV-1-permissive target cells. Importantly, most of these cells lack detectable levels of Hsc70, indicating that P-197 inhibits membrane fusion by a mechanism that is Hsc70 independent. We now suggest that competition for primary receptor binding is unlikely to account for the inhibitory activity of P-197. Understanding the mechanism by which P-197 functions may reveal concepts of general relevance to antiretroviral chemotherapy.     

Differential inhibition of retrovirus transduction by proteoglycans and free glycosaminoglycans.

We have previously shown that the efficiency of retrovirus-mediated gene transfer is limited in part due to the presence of chondroitin sulfate proteoglycans in virus stocks. In this study, we have used a model recombinant retrovirus encoding the Escherichia coli lacZ gene, bovine aorta chondroitin sulfate proteoglycan (CSPG), various free glycosaminoglycan chains (GAGs), and quantitative assays for retrovirus transduction to explore the mechanism by which proteoglycans and glycosaminoglycans inhibit retroviruses. We found that CSPG and GAGs block an early step in virus-cell interactions but do not act by inactivating viruses or by reducing the growth rate of the target cells. CSPG and most of the GAGs tested (chondroitin sulfate A, chondroitin sulfate B, heparin, heparan sulfate, and hyaluronic acid) inhibited transduction, but with widely varying degrees of activity. The chemical structure of GAGs was found to be an important determinant of their inhibitory activity, which suggests that GAGs do not inhibit transduction simply because they are highly negatively charged polymers. When GAGs were used in combination with a cationic polymer (Polybrene), however, their inhibitory activity was neutralized, and interestingly, at optimal doses of GAG and Polybrene, transduction efficiency was actually enhanced by as much as 72%. In contrast, the inhibitory activity of CSPG, due to the influence of its core protein, was not substantially reduced by Polybrene. The importance of these findings to our understanding of retrovirus-cell interactions and to the development of more efficient retrovirus gene transfer protocols is discussed.     

The murid herpesvirus-4 gH/gL binds to glycosaminoglycans

The first contact a virus makes with cells is an important determinant of its tropism. Murid Herpesvirus-4 (MuHV-4) is highly dependent on glycosaminoglycans (GAGs) for cell binding. Its first contact is therefore likely to involve a GAG-binding virion glycoprotein. We have previously identified two such proteins, gp70 and gp150. Gp70 binds strongly to GAGs. However, deleting it makes little difference to MuHV-4 cell binding or GAG-dependence. Deleting gp150, by contrast, frees MuHV-4 from GAG dependence. This implies that GAGs normally displace gp150 to allow GAG-independent cell binding. But the gp150 GAG interaction is weak, and so would seem unlikely to make an effective first contact. Since neither gp70 nor gp150 matches the expected profile of a first contact glycoprotein, our understanding of MuHV-4 GAG interactions must be incomplete. Here we relate the seemingly disconnected gp70 and gp150 GAG interactions by showing that the MuHV-4 gH/gL also binds to GAGs. gH/gL-blocking and gp70-blocking antibodies individually had little effect on cell binding, but together were strongly inhibitory. Thus, there was redundancy in GAG binding between gp70 and gH/gL. Gp150-deficient MuHV-4 largely resisted blocks to gp70 and gH/gL binding, consistent with its GAG independence. The failure of wild-type MuHV-4 to do the same argues that gp150 is normally engaged only down-stream of gp70 or gH/gL. MuHV-4 GAG dependence is consequently two-fold: gp70 or gH/gL binding provides virions with a vital first foothold, and gp150 is then engaged to reveal GAG-independent binding.     

Amphotropic retrovirus transduction is inhibited by high doses of particle-associated envelope proteins

Using a panel of amphotropic murine leukemia virus packaging cell lines that differed only in their levels of envelope protein (gp70) expression, we examined the relationship between transduction and the number of envelope proteins per virus. We generated virus stocks that contained different levels of virus-associated envelope proteins, purified them from gp70 that was not associated with the viruses, quantified their titers, and measured the efficiency with which they transduced NIH 3T3, TE671, and HeLa cells. As expected, titers increased monotonically with viral envelope protein number. Titers are measured using highly dilute virus, however, and are often not predictive of gene transfer when high doses of virus are used, as is done in gene therapy protocols. Interestingly, when we used high doses of virus, we observed significantly different trends: gene transfer increased, reached a maximum, and then declined sharply as the number of envelope proteins per virus increased. The highest levels of gene transfer occurred when cells were transduced with a moderate dose of virus that contained low levels of envelope protein. Our results indicate that transduction is inhibited when viruses that contain large numbers of envelope proteins are used. This is most likely because each virus, when it binds to a cell, delivers a large payload of envelope proteins that occupy or inactivate multiple virus receptors, reducing or eliminating the susceptibility of the cell to being transduced by additional viruses. The implications of our findings for the design of improved retroviral vectors for human gene therapy are discussed.     

Simultaneous infection with retroviruses pseudotyped with different envelope proteins bypasses viral receptor interference associated with colocalization of gp70 and target cells on fibronectin CH-296

Several factors are thought to limit the efficiency of retroviral transduction in clinical gene therapy protocols that target hematopoietic stem cells. For example, the level of expression of the amphotropic receptor Pit-2, a phosphate symporter, appears to be low in human and murine hematopoietic stem cells. We have previously demonstrated that transduction of hematopoietic cells in the presence of the fibronectin (FN) fragment CH-296 is extremely efficient (H. Hanenberg, X. L. Xiao, D. Dilloo, K. Hashino, I. Kato, and D. A. Williams, Nat. Med. 2:876-882, 1996). To examine functionally whether the retrovirus receptor is a limiting factor in transduction of hematopoietic cells, we performed competition experiments in the presence of FN CH-296 with retrovirus vectors pseudotyped with the same or a different envelope protein. We demonstrate in both human erythroleukemia (HEL) cells and primary human CD34(+) hematopoietic cells inhibition of efficient infection due to receptor interference when two vectors targeting the amphotropic receptor are used simultaneously. Receptor interference lasted up to 24 h. No interference was demonstrated when vectors targeting the amphotropic receptor and the gibbon ape leukemia virus (GALV) receptor Pit-1 were used concurrently. In contrast, simultaneous infection with vectors targeting both Pit-1 and Pit-2 yielded transduction efficiencies consistently higher than with either vector alone in both HEL cells and human CD34(+) hematopoietic cells. These data demonstrate that the use of FN CH-296 leads to amphotropic receptor saturation in these cells. Simultaneous infection with vectors targeting both amphotropic and GALV receptors may prove to be of additional benefit in the design of gene therapy protocols.     

Heparin inhibits retrovirus binding to fibronectin as well as retrovirus gene transfer on fibronectin fragments

Fibronectin fragments have been shown to improve retrovirus gene transfer efficiency by binding retrovirus and target cells. Using a novel virus adhesion assay, we confirmed binding of type C oncoretrovirus vectors to the heparin II domain of fibronectin and demonstrated inhibition of viral binding and gene transfer by heparin.     

Expression of PiT1 and PiT2 retroviral receptors and transduction efficiency of tumor cells

Recombinant retroviral vectors are still the most common gene delivery vehicles for gene therapy purposes, especially for construction of genetically modified tumor vaccines (GMTV). However, these vehicles are characterized by relatively low titre and in the case of many tumor cell lines, low transduction efficiency. We constructed bicistronic retroviral vector pseudotypes of amphotropic murine leukemia virus (A-MuLV) and gibbon ape leukemia virus (GaLV), encoding enhanced green fluorescent protein (EGFP) as a rapid and easily detectable reporter gene. Transduction of five different human melanoma and four renal carcinoma cell lines by these two virus pseudotypes revealed differences in transduction efficiency, which wase markedly lower for the renal carcinoma cell lines. Stimulation of retroviral receptor expression (PiT1 and PiT2) by phosphate depletion induced a limited increase of receptor mRNA levels, but did not improve the gene transfer efficiency. In contrast, simultaneous transduction with both vector pseudotypes markedly increased the transduction efficiency, compared to GaLV or A-MuLV alone. The same effect could be achieved by several repeated exposures of target cells to fresh vector preparation. Overexpression of GaLV receptor (PiT1) in target cells significantly increased the transduction rate and enabled retrovirus mediated gene transfer into the cells which normally are not transducible by GaLV pseudotypes. We demonstrated that, using different transduction strategies, the relatively inefficient, widely used retroviral vector systems could be significantly improved.     

High-Efficiency Gene Transfer into Normal and Adenosine Deaminase-Deficient T Lymphocytes Is Mediated by Transduction on Recombinant Fibronectin Fragments

Primary human T lymphocytes are powerful targets for genetic modification, although the use of these targets in human gene therapy protocols has been hampered by low levels of transduction. We have shown previously that significant increases in the transduction of hematopoietic stem and progenitor cells with retroviral vectors can be obtained by the colocalization of the retrovirus and target cells on specific fibronectin (FN) adhesion domains (H. Hanenberg, X. L. Xiao, D. Dilloo, K. Hashino, I. Kato, and D. A. Williams, Nat. Med. 2:876–882, 1996). We studied the transfer of genes into primary T lymphocytes by using FN-assisted retroviral gene transfer. Activated T lymphocytes were infected for three consecutive days on the recombinant FN fragment CH-296 with a retroviral vector encoding the murine B7-1 protein. Transduced lymphocytes were analyzed for murine B7-1 expression, and it was found that under optimal conditions, 80 to 89% of the CD3⁺ lymphocytes were transduced. Gene transfer was predominantly augmented by the interaction between VLA-4 on the T lymphocytes and the FN adhesion site CS-1. Adenosine deaminase (ADA)-deficient primary T lymphocytes transduced on CH-296 with a retrovirus encoding murine ADA (mADA) exhibited levels of mADA activity severalfold higher than the levels of the endogenous human ADA protein observed in normal human T lymphocytes. Strikingly, the long-term expression of the transgene was dependent on the activation status of the lymphocytes. This approach will have important applications in human gene therapy protocols targeting primary T lymphocytes.     

Ligand-directed gene targeting to mammalian cells by pseudotype baculoviruses

The baculovirus Autographa californica multiple nucleopolyhedrovirus (AcMNPV) can infect a variety of mammalian cells, as well as insect cells, facilitating its use as a viral vector for gene delivery into mammalian cells. Glycoprotein gp64, a major component of the budded AcMNPV envelope, is involved in viral entry into cells by receptor-mediated endocytosis and subsequent membrane fusion. We examined the potential production of pseudotype baculovirus particles transiently carrying ligands of interest in place of gp64 as a method of ligand-directed gene delivery into target cells. During amplification of a gp64-null pseudotype baculovirus carrying a green fluorescent protein gene in gp64-expressing insect cells, however, we observed the high-frequency appearance of a replication-competent virus incorporating the gp64 gene into the viral genome. To avoid generation of replication-competent revertants, we prepared pseudotype baculoviruses by transfection with recombinant bacmids without further amplification in the gp64-expressing cells. We constructed gp64-null recombinant bacmids carrying cDNAs encoding either vesicular stomatitis virus G protein (VSVG) or measles virus receptors (CD46 or SLAM). The VSVG pseudotype baculovirus efficiently transduced a reporter gene into a variety of mammalian cell lines, while CD46 and SLAM pseudotype baculoviruses allowed ligand-receptor-directed reporter gene transduction into target cells expressing measles virus envelope glycoproteins. Gene transduction mediated by the pseudotype baculoviruses could be inhibited by pretreatment with specific antibodies. These results indicate the possible application of pseudotype baculoviruses in ligand-directed gene delivery into target cells.     

Interaction of human plasma fibronectin with viral proteins of human immunodeficiency virus

Abstract Fibronectin (FN) is present in soluble and matrix forms in various body fluids and tissues, and has been shown to bind to several pathogens, including viruses. The interaction of FN with viral proteins of human immunodeficiency virus (HIV-1) was investigated by immunofluorescence technique using a cell line chronically infected with HIV-1 (H9-V). The results of this study showed that FN binds to HIV-1 infected cells. especially at FN concentration of 5 μg/ml. In addition, FN-pentapeptide has shown the ability to bind to HIV-1 infected cells. On the other hand, preincubation with antibodies against FN abolished the binding of FN to HIV-1 infected cells. Finally, FN has shown to bind to HIV-1 glycoproteins, including gp41 and pg120. In contrast, no binding to HIV-1 core proteins, including p15 and p24, was noted. We suggest that FN, in binding HIV-1 particles, may reduce viremia and thus may be involved in the clearance of viral proteins from the cells.     

An amino-terminal fragment of the Friend murine leukemia virus envelope glycoprotein binds the ecotropic receptor.

Retrovirus entry into cells is mediated by specific binding of the envelope glycoprotein to a cell membrane receptor. Constitutive envelope gene expression prevents infection by interfering with the binding of viruses which recognize the same receptor. We have used this property to investigate the receptor binding capacities of deleted or truncated murine leukemia virus ecotropic envelope glycoproteins. Friend murine leukemia virus envelope glycoproteins bearing internal amino-terminal deletions, or a soluble 245-amino-acid gp70 amino-terminal fragment, were expressed in NIH 3T3 cells. The susceptibility of these cells to ecotropic and amphotropic virus infection was determined. We observed that both membrane-bound and soluble forms of the gp70 245-amino-acid amino-terminal domain induced resistance to ecotropic virus, indicating that this fragment binds the ecotropic receptor. Binding occurs both at the cell surface and in the endoplasmic reticulum, as shown by the use of soluble envelope fragments either secreted in the culture supernatants or retained in the endoplasmic reticulum lumen by a KDEL sequence. These results suggest that the gp70 amino-terminal domain folds into a structure which recognizes the ecotropic receptor regardless of the carboxy-terminal part of the molecule.