Purification of retroviral vectors

Abstract

A method of purifying retroviral vector particles which comprises culturing retroviral vector producer cells which are capable of generating retroviral vector particles, and obtaining a supernatant containing retroviral vector particles from the culture of retroviral vector producer cells. The supernatant is clarified, and then concentrated. The supernatant then is contacted with a precipitation agent, and the supernatant and precipitation agent are centrifuged. A precipitate is recovered, resuspended, and subjected to a high-speed centrifugation. Retroviral particles then are recovered. The above method provides for increased recovery of retroviral vector particles which may be employed in gene therapy.

Description

This invention relates to the purification of retroviral vectors. More particularly, this invention relates to a method of large scale purification of retroviral vector particles whereby one obtains improved recovery of retroviral vector particles from a culture of retroviral vector producer cells.

The number of human trials in human gene therapy continues to increase. In addition, preclinical studies in new approaches directed to human gene transfer similarly have increased. While several gene delivery systems with adenoviral vectors and liposomes have been studied, at present, retroviral vector systems account for the overwhelming number of published reports and human trials (Miller, et al., Mol. Cell. Biol., Vol. 6, pgs. 2895-2902 (1986); Danos, et al., PNAS, Vol. 85, pgs. 6460-6464 (1988). Muenchau, et al., Virology, Vol. 176, pgs. 262-265 (1990); Miller, Curr. Top. Microbiol. Immunol., Vol. 158, pgs. 1-24 (1992)). To date, the diseases under investigation involving the use of retroviral vectors for gene therapy include various types of cancer, genetic diseases (including adenosine deaminase deficiency, cystic fibrosis, and familial hypercholesterolemia), and infectious diseases, specifically AIDS (Rosenberg et al., N. Engl. J. Med., Vol. 323, pgs. 570-578 (1990); Anderson, et al., Hum. Gene. Ther., Vol. 1, pgs. 331-362 (1990); Grossman, et al., J. Lab. Clin. Med., Vol. 119, pgs. 457-460 (1992); Riddel, et al., Hum. Gene. Ther., Vol. 3, pgs. 319-338 (1992)). At present, human trials are Phase I and/or II clinical trials to insure safety of the delivery system and the specific transgene. These early trials may also generate some meaningful information on efficacy.

In order to facilitate clinical applications of retroviral-mediated human gene transfer, infectious retroviral vectors must be of high titer and free of detectable replication-competent retroviruses. However, it has been difficult to obtain large quantities of such retroviral vectors.

It is an object of the present invention to provide a method for purifying retroviral vector particles whereby one obtains an increased amount of retroviral vector particles.

In accordance with an aspect of the present invention, there is provided a method of purifying retroviral vector particles, and in particular, infectious, non-replication competent retroviral vector particles. The method comprises culturing retroviral vector producer cells, which are capable of generating retroviral vector particles. A supernatant containing retroviral vector particles then is obtained from the culture of retroviral vector producer cells. The supernatant then is clarified, and then the clarified supernatant is concentrated after clarification. The concentrated supernatant then is contacted with a precipitation agent to form a precipitate containing the retroviral vector particles. The precipitate is recovered, and the recovered precipitate is suspended in a liquid. The retroviral vector particles are separated from the remainder of the precipitate by density gradient centrifugation. The separated retroviral vector particles then are recovered.

The above method enables one to obtain purified retroviral vector particles. The term "purified retroviral vector particles" as used herein means a preparation of retroviral particles containing at least 80% by weight, preferably at least 85% by weight, and more preferably at least 90% by weight, of retroviral vector particles.

The retroviral vector producer cells may be cultured in any of a variety of monolayer culture systems. Such producer cells may be cultured in T-flasks, roller bottles, or bioreactors. The cells may be cultured in any acceptable culture medium such as, for example, AIM-V medium (Gibco BRL, Grand Island, N.Y) containing 5% fetal bovine serum, or Dulbecco's modified Eagle medium (DMEM) with high glucose (4.5 g/l) supplemented with 10% heat-inactivated fetal bovine serum.

The producer cell line may be formed by transducing an appropriate packaging cell line with retroviral vectors. Examples of packaging cell lines which may be transduced include, but are not limited to, the PA317, PE501, .PSI.-2, .PSI.-AM, PA12, T19-14X, VT-19-17-H2, .PSI.RE, .PSI.CRIP, GPL, GP+E-86, GP+envAm12, and DAN cell lines. The vector may transduce the packaging cells through any means known in the art. Such means include, but are not limited to, electroporation, the use of liposomes, and CaPO.sub.4 precipitation.

The cells generally may be cultured at a temperature of from about 32.degree. C. to about 37.degree. C. Most preferably, the cells are cultured at a temperature of about 32.degree. C. At 32.degree. C., the vector particles have a half-life of up to about 48 hours.

The viral supernatant, which is obtained from the culture of retroviral vector producer cells, and which may have a titer of, for example, from about 10.sup.6 to about 10.sup.7 CFU/ml, may be clarified by passing the supernatant through a filter means, such as a nominal type filter unit. The filter may have a pore size of from about 0.2.mu. to about 5.0.mu., preferably from about 0.8.mu. to about 1.5.mu.. Most preferably, the filter has a pore size of about 1.2.mu..

The clarified viral supernatant then may be concentrated by passing the clarified viral supernatant through a tangential flow filtration system. Preferably, the membrane in such system has a 300,000 molecular weight cutoff. Such a system may reduce the volume of supernatant containing retroviral vector particles by about 16 to 25 times, with vector recovery being at least 90%. An example of a tangential flow filtration system which may be employed in accordance with the present invention is Millipore Pellicon tangential flow filtration system (Millipore, Bedford, Mass.) with a PLMK000C5 cassette (5 square feet, 300,000 NMWL). The system also may be equipped with a pump to exert a low membrane feed pressure of 5 psi. Such a concentrated viral supernatant may have a vector recovery of greater than 95% compared with the amount of vector present prior to concentration, and a vector titer of from about 10.sup.7 to about 10.sup.8 CFU/ml.

The concentrated viral supernatant then is contacted with a precipitation agent. In one embodiment, the precipitation agent is a polyalkylene glycol, such as, for example, polyethylene glycol. In a preferred embodiment, once the viral supernatant is contacted with the precipitation agent, the viral supernatant and the precipitation agent are centrifuged at a speed of from about 8,000 xg to about 10,000 xg to pellet precipitated material, which then is recovered. Such precipitate contains the retroviral vector particles. The recovered precipitate is suspended in a liquid, such as a buffer. The retroviral vector particles are separated from the remainder of the precipitate by density gradient centrifugation. Preferably, the suspended precipitate is subjected to sucrose gradient ultracentrifugation (discontinuous or preferably linear). More preferably, the ultracentrifugation is effected at a speed of from about 100,000 xg to about 120,000 xg. After centrifugation, the retroviral vector particles are recovered. Vector recovery may be greater than 95%, and vector titer may be from about 10.sup.9 to about 10.sup.10 CFU/ml. An example of purifying viral supernatant by polyethylene glycol precipitation and sucrose gradient centrifugation is described in Syrewicz, et al., Appl. Microbiol., Vol. 24, pgs. 488-494 (1972), incorporated herein by reference.

The above purification steps enable one to recover an increased amount of retroviral vector particles. The resulting preparation containing the retroviral vector particles is free of components which may compete with the retroviral vector particles for binding sites located on target cells to which the retroviral vector particles may bind and infect. Such components include free gp70 protein and empty viral particles.

If desired, the retroviral vector particles recovered from the centrifugation step may be lyophilized and saved for future use. Such lyophilization may be carried out by adding to the retroviral vector supernatant solutions of glucose and/or sorbitol and gelatin in phosphate buffered saline. The samples then may be frozen in a dry ice/acetone bath and then placed in a freeze dryer. When the retroviral vector particles are needed for gene therapy, the lyophilized retroviral vector particles may be reconstituted by adding an equal volume of sterile water. After reconstitution, vector recovery may be greater than 90%, and vector titer may be from about 10.sup.9 to about 10.sup.10 CFU/ml.

The purified retroviral vector particles are used to transduce cells in vivo or ex vivo for purposes of gene therapy. For ex vivo transduction, the inventors have developed a novel preferred transduction method, which they call spin transduction The particles, suspended in a pharmaceutically acceptable carrier, are contacted with the cells to be transduced, and the mixture is centrifuged for a sufficient period of time to permit transduction. The mixture is centrifuged at 1,000-4,000 rpm, preferably at 2,000-3,000 rpm, and most preferably at about 2,500 rpm for a period of time from about 30 minutes to 3 hours, preferably 1 to 2 hours, and most preferably about 90 minutes. The spin transduction is done at a temperature of 15.degree. C. to 37.degree. C., preferably 20.degree. C. to 35.degree. C., and most preferably 32.degree. C. This method leads to a 3- to 15- fold increase in transduction efficiency, depending on the type of cells transduced.

In accordance with another aspect of the present invention, there is provided a method of purifying infectious retroviral vector particles to obtain a preparation of infectious retroviral vector particles free of detectable replication-competent retroviruses. The process comprises generating retroviral vector particles by culturing retroviral vector producer cells, and obtaining a supernatant containing the retroviral vector particles from the culture of retroviral vector producer cells. The supernatant then is concentrated and diafiltered. The concentrated and diafiltered supernatant then is subjected to ion exchange chromatography. The supernatant again is concentrated and diafiltered to obtain a purified retroviral supernatant containing infectious retroviral vector particles free of detectable replication competent retroviruses.

The retroviral vector producer cells may be cultured as hereinabove described in order to obtain a supernatant containing retroviral vector particles. The supernatant then is concentrated through a tangential flow filtration system, which also may be hereinabove described.

Diafiltration may be carried out in the presence of an appropriate buffer, such as, for example 0.02M Bis-Tris+0.1M NaCl, pH5.50. Subsequent to diafiltration, the supernatant may again be concentrated by tangential flow filtration as hereinabove described.

Prior to ion-exchange chromatography, the supernatant may again be filtered, such as, for example, through a 0.22 .mu.m filter. The supernatant then is subjected to ion-exchange chromatography on an appropriate ion-exchange chromatography resin. Ion-exchange chromatography resins which may be employed include resins which include anion exchange groups or cation exchange groups. Examples of anion exchange groups which may be employed include, but are not limited to--O--L--N.sup.+ --(R).sub.3, wherein L is a linker group such as an alkyl group or an appropriate polymer or other moiety for linking the quarternary amine moiety to the resin, and R is an alkyl group having one or more carbon atoms. In one embodiment, R is methyl. In another embodiment, each of L and R is ethyl. Another anion exchange group which may be employed has the formula --O--L--N--(R).sub.2, wherein L and R are hereinabove described. In one embodiment, each of L and R is ethyl.

Cation exchange groups which may be employed include, but are not limited to, O--R--C00H and O--R--SO.sub.3.sup.-, wherein R is an alkyl group having at least one carbon atom. In one embodiment, the cation exchange group is O--R--COOH, and R is methyl. In another embodiment, the cation exchange group is O--R--SO.sub.3.sup.-, and R is propyl.

The anion or cation exchange group is attached to an appropriate chromatography resin. In one embodiment, the resin is formed of spherical beads of a copolymer of an alkylene glycol (such as, for example, ethylene glycol) and a methacrylate.

In a preferred embodiment, the ion exchange resin is comprised of an anion exchange group having the formula --O--L--N.sup.+ --(R).sub.3.sup.- as hereinabove described, wherein R is methyl (CH.sub.3), which is attached to a resin formed of spherical beads of a polymer of ethylene glycol and a methacrylate. An example of such an ion-exchange chromatography resin including a quarternary amine functional group is the Toyopearl.RTM. Super Q-650M resin, sold by Toso Haas. (Montgomeryville, Pa.)

The ion-exchange chromatography column is washed with an appropriate buffer, such as, for example, 0.02M Bis-Tris +0.1M NaCl, pH5.50, when a column including a quarternary amine is employed, prior to addition of the concentrated and diafiltered vector supernatant. Elution of the column then may be carried out with a linear salt gradient (such as, for example, a linear NaCl gradient) of appropriate buffers.

After the supernatant is subjected to ion-exchange chromatography, the supernatant is desalted through tangential flow filtration and diafiltration. Diafiltration may be carried out through buffer exchange employing a buffer suitable for the preparation of lyophilized virus. Such buffer may be, for example, 5% sucrose, 10% mannitol, 0.02% EDTA in water.

Subsequent to diafiltration, the vector particles may be lyophilized. The vector supernatant, which has been diafiltered into a lyophilization buffer, may be spiked with human serum albumin. The resulting formulation then may be filtered (such as, for example, through a 0.22 .mu.m filter), and then lyophilized. Vector recovery may be obtained in the order of about 90%. The lyophilized vector then is stored (preferably at about 4.degree. C.) until use, when the vector may be reconstituted prior to use in a gene therapy procedure.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1. Number of cells and vector titers of PA317/G1Na.40 producer cells grown in 75-cm.sup.2 flasks at different temperatures.

FIG. 2. Effect of temperature on viral particles per cell during 12 days of culture. PA317/G1Na.40 producer cells were grown in 75-cm.sup.2 flasks at 37 C., 34 C. and 32 C. Medium was changed every 24 hr. Cell number and vector titer were determined daily.

FIG. 3. Comparison of centrifugation and noncentrifugation on transduction efficiency of HUT 78 cells with PA317/G1MD.1-15. A total of 1.times.10.sup.6 cells in a 24-well plate were transduced with and without centrifugation (spun) at 2,500 rpm at 32 C. for 90 min. Cells were incubated overnight at 32 C. Three days after transduction the cells were prepared for PCR and FACS analyses.

A. PCR agarose gel profile. Ethidium bromide-stained gel of amplification products obtained using primers specific for MDR1 (1) or .beta.-actin. Products were amplified from 1 .mu.g of genomic DNA isolated from mock-transduced cells or cells transduced with PA317/G1MD1-15. The 100 bp ladder was obtained from GIBCO BRL. B. Flow cytometry (FACS analyses). Cells were harvested and suspended in basal DMEM containing 0.1 .mu.g/ml rhodamine-123. Cells were incubated at 37 C. for 20 min., then centrifuged at 1,500 rpm for 10 min. Cells were resuspended in basal medium and effluxed at 37 C. for 45 min. Flow cytometry was performed by EPICS cytometer (Coulter).

The invention now will be described with respect to the following examples; however, the scope of the present invention is not intended to be limited thereby.

EXAMPLE 1

Unless noted otherwise, all viral vector titers of retroviral vectors carrying the neomycin resistance marker are determined by a G418 selection method using NIH 3T3 TK-cells as target cells. The assay procedure is as follows:

On day 1, NIH-3T3 TK-cells were seeded at 1.times.10.sup.5 cells/well of a six-well tissue culture plate (Becton Dickinson, Lincoln Park, N.J.) and incubated at 37.degree. C. in 5% CO.sub.2. On day 2, serial ten-fold dilutions of virus vector specimen in medium containing 8 .mu.g/ml Polybrene were added to the target cells and incubated at 32.degree. C. for an additional 24 hours. On day 3, the medium was removed and replaced with medium containing 800 .mu.g/ml G418. Plates were incubated at 37.degree. C. in 5% CO.sub.2. On day 6, plates were refed with medium containing 800 ug/ml G418. On day 8, colonies were stained with methylene blue and the vector titer was calculated as the number of colony forming units (cfu) per ml.

A. Formation of Producer Cell Lines

The following producer cell lines were formed by transducing retroviral vectors including various desired genes and promoters (sometime hereinafter referred to as Gene 1, Gene 2, Promoter 1, and Promoter 2) into the PA317 amphotropic retrovirus packaging cell line (Miller, et al, Mol. Cell. Biol., Vol. 6, pgs. 2895-2902 (1986); Miller, et al., Biotechniques, Vol 7, pgs. 980-990 (1989), incorporated herein by reference), were formed by standard transduction techniques. The producer cell lines are given in Table I below:

TABLE I______________________________________Producer cells Gene 1 Promoter 1 Gene 2 Promoter 2______________________________________PA317/G1Na.40 Neo.sup.R LTR -- --PA317/G1Na.47 Neo.sup.R LTR -- --PA317/ IL-2R.alpha. LTR Neo.sup.R SV40G1IL2R.alpha.SvNa.3PA317/ IL-2R.beta. LTR Neo.sup.R SV40G1IL2R.alpha.SvNa.14PA317/G1NaCvI2G Neo.sup.R LTR IL-2 cmvPA317/G1NaBcI2G.6 Neo.sup.R LTR IL-2 .beta.-ActinPA317/G1.beta.Gsvna.29 lacZ LTR Neo.sup.R SV40PA317/ Neo.sup.R LTR lacZ SV40G1NaSvn.beta.g.2-1PA317/ Neo.sup.R LTR lacZ SV40G1NaSvn.beta.g.2-12PA317/ lacZ LTR Neo.sup.R SV40G1n.beta.Gsvna.8-1PA317/ lacZ LTR Neo.sup.R SV40G1n.beta.Gsvna.8-3PA317/G1NaSvAd.24 Neo.sup.R LTR ADA SV40PA317/LASN ADA LTR Neo.sup.R SV40PA317/ IFN-.alpha. LTR Neo.sup.R SV40G1F1SVNa.19cPA317/ Neo.sup.R LTR IFN-r CMVG1NaCvF32.20PA317/G1T2SvNa.24 TNF-.alpha. LTR Neo.sup.R SV40PA317/LT12SN.40 TNF-.alpha. LTR Neo.sup.R SV40PA317/LNCT11.8 Neo.sup.R LTR TNF-.alpha. CMVPA317/LNST11.20 Neo.sup.R LTR TNF-.alpha. SV40PA317/G1MD.1-15 MDR1 LTR -- --PA317/G1MD.S-5 MDR1 LTR -- --PA317/LIXSN Factor IX LTR Neo.sup.R SV40______________________________________ Producer cell line PA317/G1Na.40 is described in Miller, Curr. Top. Microbiol. Immunol., Vol. 158, pgs. 1-24(1992), incorporated herein by reference. Producer cell line PA317/LASN is described in Hock, et al., Blood, Vol. 74, pgs. 876-881 (1989), incorporated herein by reference. Producer cell line PA317/L1XSN is described in Palmer, et al., Blood, Vol. 73, pgs. 438-445 (1989), incorporated herein by reference.

B. Culturing of Producer Cells

PA317/G1Na.40 producer cells (2.times.10.sup.6 cells) were inoculated in 75-cm.sup.2 flasks and were incubated at 37.degree. C., 34.degree. C., and 32.degree. C. in 5% CO.sub.2 to determine the effect of incubation temperature on vector titer. The medium was changed every day.

The amphotropic retroviral vectors were assayed by a G418 selection method using NIH-3T3 TK.sup.- cells as target cells. On day 1, NIH-3T3 TK.sup.- cells were seeded at 1.times.10.sup.5 cells/well of a six-well tissue culture plate (Becton Dickinson, Lincoln Park, N.J.) and incubated at 37.degree. C. in 5% CO.sub.2. On day 2, serial ten-fold dilutions of vector specimen in medium containing 8 .mu.g/ml Polybrene were added to the target cells and incubated at 32.degree. C. for an additional 24 hr. On day 3, the medium was removed and replaced with medium containing 800 .mu.g/ml G418. Plates were incubated at 37.degree. C. in 5% CO.sub.2. On day 6, plates were refed with medium containing 800 .mu.g/ml G418. On day 8, colonies were stained with methylene blue and the vector titer was calculated as the number of colony forming units (cfu) per ml.

The results are shown in FIG. 1. Cells grown at 37.degree. C., 34.degree. C., and 32.degree. C. achieved 100% confluence on days 5, 6, and 8, respectively. Days 5, 6, and 8 also showed the highest vector titers for each of the three temperatures. Retroviral vector titers increased as temperature decreased. The highest vector titer of 1-day-old supernatant was always obtained from cells incubated at 32.degree. C., 8.5.times.10.sup.6 cfu/ml, and remained at that approximate level for 5 days. Cells incubated at 37.degree. C. showed the lowest titer over 7 days. In the experiment shown in FIG. 1, the maximum titer difference between 32.degree. C. and 37.degree. C. was 1.5 logs on day 8. Vector from cells incubated at 34.degree. C. yielded titers intermediate between 32.degree. C. and 37.degree. C. Interestingly, the highest number of cells (1.1.times.10.sup.6) was obtained at 32.degree. C. on day 12. During the experiment, viability of the cells for each temperature ranged between 92 and 98%.

The number of vector particles per cell was calculated by dividing the number of vector particles per milliliter determined in a standard titer assay by the number of cells from viability counts. As shown in FIG. 2, the 32.degree. C. and 34.degree. C. incubation temperatures showed a maximum of six vector particles per cell compared with a maximum of two vector particles per cell at 37.degree. C. under the conditions of these calculations.

To determine the effect of CO.sub.2 on production of viral vectors, PA317/G1Na.40 cells were grown in 75-cm.sup.2 flasks and roller bottles with and without 5% CO.sub.2 in air. The titer of vector supernatants cultivated under both conditions revealed similar titers at the three incubation temperatures studied (data not shown). The results of these studies and those with other vectors suggest that endogenous CO.sub.2 from the cells support the cell growth and vector production without 5% exogenous CO.sub.2.

Producer cells were easily adapted to grow in 850-cm.sup.2 roller bottles. Cells maintained maximum confluence for at least 2 weeks. PA317/G1Na.40 and PA317/LT12SN.40 producer cells were grown in roller bottles and incubated at 37.degree. C., 34.degree. C., and 32.degree. C. After the cells reached confluence, the medium was harvested daily. The vector titers shown in Table II below represent the titers reached after 100% confluence.

TABLE II______________________________________Vector Titers from Cells Grown in Roller Bottles (850 cm.sup.2)at 37.degree. C., 34.degree. C., and 32.degree. C. TemperatureVector 37.degree. C. 34.degree. C. 32.degree. C.______________________________________PA317/G1Na.40 2.9 .times. 10.sup.6b 4.7 .times. 10.sup.6 1.7 .times. 10.sup.7 (.+-.1 .times. 10.sup.6).sup.a (.+-.1.4 .times. 10.sup.5) (.+-.2.1 .times. 10.sup.6)PA317/LT12SN.40 2.3 .times. 10.sup.6 1.2 .times. 10.sup.7 1.8 .times. 10.sup.7 (.+-.1 .times. 10.sup.5) (.+-.2.0 .times. 10.sup.6) (.+-.3.3 .times. 10.sup.6)______________________________________ .sup.a Producer cells (2 .times. 10.sup.7) were seeded in roller bottles and incubated at 37.degree. C., 34.degree. C., and 32.degree. C. The supernatant from producer cells were harvested after 100% confluence. .sup.b Vector titer in 1day-old supernatant.

The titers from each temperature remained stable without significant change for the 2-week experimental time period. A significant increase of vector titer was achieved at 32.degree. C. incubation temperature. The titer of PA317/G1Na.40 increased from 2.9.times.10.sup.6 cfu/ml at 37.degree. C. to 1.7.times.10.sup.7 cfu/ml at 32.degree. C. A significant increase in titer also was achieved with vector PA317/LT12SN.40, 2.3.times.10.sup.6 cfu/ml at 37.degree. C. to 1.8.times.10.sup.7 cfu/ml at 32.degree. C. The results of increased vector titer at 32.degree. C. compared to 37.degree. C. incubation for both 75-cm.sup.2 flasks (FIG. 1) and roller bottles (Table II) are statistically significant at the p<0.01 probability level.

Supernatants from PA317/G1Na.40 and PA317/LT12SN.40 producer cells grown in roller bottles were collected at 1- and 2-day intervals at 32.degree. C. cultivation (Table III).

TABLE III______________________________________Vector Titers in 1-Day- and 2-Day-Old Supernatants in Roller BottlesVector 1-Day-Old Supernatant 2-Day-Old Supernatant______________________________________PA317/G1Na.40 1.1 .times. 10.sup.7b 2.1 .times. 10.sup.7 (.+-.4.5 .+-. 10.sup.6).sup.a (.+-.4.9 .times. 10.sup.6)PA317/LT12SN.40 3.1 .times. 10.sup.6 1.2 .times. 10.sup.7 (.+-.1.5 .times. 10.sup.5) (.+-.1.5 .times. 10.sup.6)______________________________________ .sup.a Producer cells were grown in roller bottle at 32.degree. C. The supernatants from producer cells were harvested at 100% cell confluence. .sup.b Vector titer (CFU/ml).

The vector titer of 2-day-old supernatants from PA317/G1Na.40 and PA317/LT12SN.40 were approximately two times and four times higher than that of 1-day-old supernatants, respectively. The above results from 2-day-old supernatant collection at 32.degree. C. incubation also were demonstrated using several of the other producer cell lines. Increased titers at lower temperature may be partially due to decreased vector inactivation at the lower temperature.

Roller bottles of different surface areas were compared for vector yield. As expected, the vector titer from supernatants of PA317/G1Na.40 and PA317/G1.beta.Gsvna.29 producer cell lines from a pleated roller bottle of 1,700-cm.sup.2 surface area were approximately twice that of the 850-cm.sup.2 roller bottle (Table IV).

TABLE IV______________________________________Vector Titers in Supernatants from Cells Grown in Different Types ofRoller Bottles at 32.degree. C. Type of Roller BottleVector 850 cm.sup.2 1,700 cm.sup.2______________________________________PA317/G1Na.40 1.2 .times. 10.sup.7b 2.6 .times. 10.sup.7 (.+-.3.5 .times. 10.sup.6).sup.a (.+-.4.3 .times. 10.sup.6)PA317/G1BgSvNa.29 4.2 .times. 10.sup.6 1.0 .times. 10.sup.7 (.+-.6.0 .times. 10.sup.5) (.+-.3.5 .times. 10.sup.5)______________________________________ .sup.a Producer cells were grown in roller bottles at 32.degree. C. Twoday-old supernatants were collected after 100% cell confluence. .sup.b Vector titer (CFU/ml).

The number of cells counted at the end of the 2-week culture from a 1,700-cm.sup.2 roller bottle was approximately twice that counted from the 850-cm.sup.2 roller bottle (data not shown), suggesting that the vector yield per cell did not differ in the pleated and unpleated bottles. To date, the highest virus titer achieved was from a 2-day-old PA317/G1Na.40 supernatant from the pleated-type roller bottle incubated at 32.degree. C., 1.times.10.sup.6 cfu/ml. This has been repeated two times. Twenty-one of 22 different retroviral vector producer cells increased titers by a 2-day-old supernatant at 32.degree. C. incubation. Vector LASN was the exception in this study.

C. Clarification of Viral Supernatant

The viral supernatant was harvested from the roller bottle by aspiration or pouring into a collection bottle. Immediately thereafter, the supernatant was clarified by pumping the supernatant through a 1.2.mu. nominal type polypropylene filter (Sartorius) to remove any debris and cells.

Vector supernatant from PA317/G1nBgSvNa producer cells was concentrated using the Millipore Pellicon tangential flow filtration system (Millipore, Bedford, Mass.) with a PLMK000C5 cassette (5 square feet, 300,000 NMWL). A pump was used to exert a low membrane feed pressure of 5 psi. Concentration was achieved within 30 minutes. To ensure high vector recovery, the surface area of the membrane was maximized and supernatant circulation was minimized with a low membrane feed pressure. As shown in Table V below, the volume of the supernatant was reduced from 16 to 25 times, and the vector titer increased from 15 to 24 times. Vector recovery ranged from 91% to 96%.

TABLE V__________________________________________________________________________ Volume of Titer supernatant (ml) (cfu/ml) Before After Volume Before After Vector VectorVector conc. conc. reduction conc. conc. concen. recovery__________________________________________________________________________PA317/G1Na.40 10,000 400 25 .times. 1.7 .times. 10.sup.6 4.1 .times. 10.sup.7 24 .times. 96%PA317/LT12SN.40 10,000 500 20 .times. 2.5 .times. 10.sup.6 4.5 .times. 10.sup.7 18 .times. 91%PA317/G1NaSvAd.24 8,000 500 16 .times. 1.6 .times. 10.sup.4 2.4 .times. 10.sup.7 15 .times. 94%__________________________________________________________________________

E. Centrifugation of Viral Supernatant and Recovery of Retroviral Vectors Therefrom

The concentrated viral supernatant (from about 100 ml to about 200 ml) was mixed with a 1/10 volume of 4M NaCl on ice while stirring. Polyethylene glycol was added (8% w/v) slowly and stirred in ice for 3 hours. The precipitate was collected by centrifugation at 8,000 xg for 15 minutes and resuspended in 0.01M tris (hydroxymethyl) aminomethane hydrochloride (TNE) pH7.5, 0.1M NaCl, and 0.001M ethylenediamine tetracetic acid (i.e., at from about 1/100 to 1/25 of the original supernatant volume).

The suspension was layered on a discontinuous sucrose gradient (20% and 55% w/w) in TNE and centrifuged in a Beckman SW-40Ti rotor at 100,000 xg for 2 hours at 4.degree. C.

After the centrifugation, the sharp virus band was observed. The virus band was collected by inserting a 21-gauge hypodermic needle into the centrifuge tube, and diluted ten-fold with TNE and stored at -70.degree. C. Vector recovery was greater than 95% and vector titer was at least 10.sup.9 CFU/ml.

F. Lyophilization of Retroviral Vectors

Lyophilization of retroviral vectors was performed using a Usifroid model SMJR lyophilizer (France). A vector formulation including virus, and (i) 10% mannitol, 5% sucrose, 1% human serum albumin (HSA), or (ii) 10% mannitol, 2% sucrose, 0.5% HSA, or (iii) 5% mannitol, 2% sucrose, and 0.5% HSA was 0.2 micron filtered and placed into 3 ml borosilicate glass tubing vials at 1 ml per vial. Stoppers were placed on the vials and pushed down only to the first stop for lyophilization. The vials were then loaded into the lyophilizer. The two and one-half day lyophilization cycle begins by freezing the vials to -40.degree. C. for 6 hours. During this time the vacuum is activated and the chamber is held at a pressure of approximately 200 microns of mercury for the remainder of the cycle. After the primary drying of the vials is completed the temperature is raised to -35.degree. C. and held for 24 hours, the temperature is then raised to -10.degree. C. and held for 12 hours and then to 15.degree. C. for 12 hours to complete the drying. At the end of the cycle the stoppers were completely pushed down to seal the vials. The vacuum was then released and the vials removed from the lyophilizer. This cycle yields a moisture content of less than 3% on this formulation.

The recovery of virus after lyophilization is 90-100%. The current data show 90-100% recovery of lyophilized virus after five days at 30.degree. C.

As shown in Table VI the formulation (i) used above best preserves the virus during the lyophilization cycle and during storage.

TABLE VI______________________________________ Recovery on Recovery RecoveryFormula Lyophilization 4 wks 4.degree. C. 5D 30.degree. C.______________________________________54 Mannitol 0.5% HSA 90% 3% --2% Sucrose10% Mannitol 0.5% HSA 60% 10%2% Sucrose10% Mannitol 1% HSA 90% 95%5% Sucrose______________________________________

G. Transduction of Cells with Recovered Vector Supernatant

NIH3T3 TK.sup.- cells were seeded at 1.times.10.sup.5 cells/well in a six-well plate as an adherent target cell line and incubated at 37.degree. C. overnight to yield approximately a 60% confluent monolayer. A 10-fold serial dilution of vector supernatants (10.sup.4, 10.sup.-5, and 10.sup.-6) containing 8 .mu.g/ml Polybrene were added in duplicate to appropriate wells on day 2. The six-well plate was centrifuged at 2,500 rpm for 90 min. at 32.degree. C. by using a plate holder in a Beckman GS-6KR centrifuge (Beckman, Fullerton, Calif.). After centrifugation, the plate was incubated at 32.degree. C. overnight. On day 3, the wells were refed with 5 ml of medium containing 800 .mu.g/ml G418. The plate was incubated at 37.degree. C. for 3 days. On day 5, medium was replaced to remove dead cells. On day 8, G418-resistant colonies were stained and counted.

In additional experiments, HUT 78 cells, a human T-cell leukemia line, as a suspension target cell line, were transduced with an amphotrophic vector PA317/G1MD.1-15 which carried the MDR1 gene encoding p-glycoprotein (Endicott, et al., Ann. Rev. Biochem., Vol. 58, pgs 137-171 (1989)).

Cells were seeded at 1.times.10.sup.6 cells/well of a 24-well plate (Costar, Cambridge, Mass.). A higher number of HUT 78 cells were needed to cover the bottom of the well. Two milliliters of vector supernatant containing 8 .mu.g/ml of Polybrene was added to each well. The plate was centrifuged at 2,500 rpm for 90 min. at 32.degree. C. and incubated at 32.degree. C. overnight. The next day the wells were refed with fresh medium and incubated at 37.degree. C. for 3 days. Cells were collected and washed two times with Hank's balanced salt solution (HBSS) and assayed for rhodamine afflux pump by FACS analysis and for the MDR1 gene by polymerase chain reaction (PCR) assay. As a control, untransduced HUT 78 cells were assayed.

FACS analysis was performed to determine transduction of HUT 78 cells with vector (PA317/G1MD.1-15) by a modification of the method described previously (Kessel et al., Cancer Res., Vol. 51, pgs. 4665-4670 (1991)). Cells were incubated at 37.degree. C. for 20 min. then centrifuged at 1,500 rpm for 10 min. Cells were resuspended in basal medium and affluxed at 37.degree. C. for 45 min. Cells were centrifuged and resuspended in HBSS (1.times.10.sup.4 cells/ml). Flow cytometry was performed using an EPICS C Cytometer (Coulter, Hialeah, Fla.).

The PCR was employed to screen for the presence of the G1MD1 retroviral sequence in HUT 78 cells after transduction. Genomic DNAs from cells subjected to G1MD1 transduction as well as untransduced cells were isolated using the Elu-Quik system (Schleicher & Schuell, Keene, N.H.). Each 100 .mu.l reaction included 1 .mu.g of genomic DNA using standard reaction conditions: 1.5 mM MgCl.sub.2, 0.2 mM dATP, 0.2 mM dGTP, 0.2 mM dCTP, 0.2 mM TTP, 50 mM KCl, 10 mM Tris-HCl, pH 7.3, 1 .mu.M synthetic primers (MDR1), 0.01 units/.mu.l Perfect Match Polymerase Enhancer (Strategene, La Jolla, Calif.), and 0.025 units/.mu.l AmpliTaq DNA polymerase (Perkin-Elmer Cetus, Emeryville, Calif.). DNA was amplified using 30 cycles of 3-min. incubations at 94.degree. C. followed by 61.degree. C. and then by 72.degree. C. To avoid detection of endogenous MDR1 genes, MDR1, primers were derived from different exons (exons 4 and 6), which are assayed by greater than 23 kb of intervening sequence. The amplified fragments were separated by electrophoresis in a 1.4% agarose gel and visualized by ethidium bromide staining, and the size was determined using DNA molecular weight markers (GIBCO BRL, Grand Island, N.Y.).

The effect of centrifugation of vector on target cell transduction was investigated. Nine different retroviral vectors were assayed for their ability to transduce NIH-3T3 TK cells with or without centrifugation. An increase in titer as determined by the number of G418-resistant colonies was observed in all experiments. Representative results are shown in Table VII. This method has now been used over 50 times with NIH-3T3 TK cells as the target cell. The titer increases in samples varied from 1 to 18-fold.

In addition to NIH-3T3 TK cells, a human leukemia line that grows in suspension, HUT 78, was also transduced with and without centrifugation. In this experiment, HUT 78 cells were transduced with amphotropic vector PA317/G1MD.1-15, which can express the broad specificity efflux pump responsible for multidrug resistance (MDR1). FACS analysis shown in FIG. 5B demonstrates that centrifugation (19.5%) yielded approximately three timed more cells capable of rhodamine effluex than were obtained without centrifugation (5.6%). Consistent with this result, amplification of MDR1 sequences in these cells by PCR revealed that cells transduced using centrifugation contained more G1MD proviral sequences than cells transduced without centrifugation (FIG. 5A), directly demonstrating that centrifugation leads to increased transduction efficiency.

TABLE 7______________________________________COMPARISON OF VECTOR TRANSDUCTION OF NIH-3T3 TK.sup.-CELLS WITH AND WITHOUT CENTRIFUGATION(2,500 RPM. 90 MIN) Vector Titer (cfulml)Culture No centri- Centrifu- Titermethods Producer cells fugation gation increase______________________________________Cell Cube PA317/G1NaSvAd.24 2.0 .times. 10.sup.6 2.2 .times. 10.sup.7 11x 3.4 .times. 10.sup.6 2.7 .times. 10.sup.7 8x 2.6 .times. 10.sup.6 2.3 .times. 10.sup.7 9xRoller bottle PA317/G1Na.40 2.6 .times. 10.sup.7 2 .times. 10.sup.8 8x PA317/G1NaSvAd.24 1.7 .times. 10.sup.6 1.1 .times. 10.sup.7 7x 1.1 .times. 10.sup.6 1.5 .times. 10.sup.7 14x PA317/L1XSN 5.7 .times. 10.sup.5 5.5 .times. 10.sup.6 10x PA317/G1F1SvNa.19c 6.0 .times. 10.sup.5 7.8 .times. 106 13x PA317/G1II2R.alpha.SvNa3 1.1 .times. 10.sup.6 1 .times. 10.sup.7 9x PA317/G1IL2RBSvNa4 2.0 .times. 10.sup.5 3.3 .times. 106 17x PA317/G1NaBcI2G6 6.0 .times. 10.sup.5 9.5 .times. 10.sup.6 16x PA317/G1T2SvNa24 1.2 .times. 10.sup.5 7 .times. 10.sup.6 14xT75 PE501/G1NaBcLa1 5 .times. 10.sup.5 1.8 .times. 10.sup.6 4x______________________________________ Titers determined by G418 selection method.

An efficient transduction method is a key component for clinical utilization in clinical trials. Most protocols incorporate a 37.degree. C. incubation temperature for various transduction times. A novel transduction procedure has been developed using the combination of centrifugation and 32.degree. C. overnight incubation. Centrifugation has been used for improved detection of other viruses, and chlamydia in clinical specimens (Ripa and Mardh Nongonococcal Urethritis and Related Infections, Holmes, et al., eds., American Society for Microbiology, Washington, D.C., pgs. 323-327, 1977; Heggie and Huang, J. Virol. Methods, Vol 41, pgs. 1-7, 1993) The combination of centrifugation and overnight transduction of NIH-3T3 TK cells at 32.degree. C. resulted in a 4- to 18-fold increase compared to the overnight transduction at 32.degree. C. The actual mechanism for the success of this centrifugation method is not well understood; however, aggregation of vector particles and/or vector particles with debris may play a role. The supernatants from roller bottles and cCellCube appeared to be more viscous than supernatant from the 175 flask, although all supernatants were filtered through a 1.2 .mu.m filter. Transduction by centrifugation on human leukemic HUT 78 cells also demonstrated a significant increase in transduction efficiency as determined by FACS and PCR analysis. This technique may result in a dramatic impact for successful gene therapy, especially where vector is in limites concentration. Studies of transduction on clinically relevant target cells such as peripheral blood lymphocytes (PBL), tumor infiltrating lymphocytes (TIL), bone marrow cells, and tumor cells are in progress.

H. Results and Discussion

High titer vectors were produced from retroviral vector producer cells in various monolayer culture systems. Vector supernatant is clarified, and the volume of supernatant containing retroviral vectors was reduced 16 to 25 times with vector recovery ranging from 91% to 96% in a Millipore Pellicon tangential flow filtration system. Subsequently, vectors were purified by a two-step procedure including polyethylene glycol (PEG) precipitation and sucrose gradient centrifugation. The vector recoveries were from 100% to 110% by a standard vector titer assay. This suggests that the vector preparations do not contain free gp70 or empty vital particles which compete for virus binding sites with infectious retroviral vectors. The retroviral vectors, upon recovery, may be lyophilized successfully.

In addition, the above example demonstrates a large-scale technology for retroviral vector production, as well as procedures for concentration, purification, and lyophilization of retroviral vectors, whereby one may obtain a purified composition having a high titer of retroviral vectors. This example demonstrates that retroviral vectors for gene therapy can be produced efficiently, concentrated, purified, and stored in scale-up quantities, and subsequently allows one to transduce a variety of target cells with high efficiency.

EXAMPLE 2

Retroviral vector producer cells of the cell line PA317/G1TK1SvNa.7 (described in PCT Application No. W095/06486, published Mar. 9, 1995) were inoculated into a packed-bed air-lift bioreactor. Continuous medium perfusion was initiated three days after cell inoculation. The perfusion medium was DMEM+2% FBS, plus 10 ppm of an antifoaming agent, and 0.1% Pluronic F-68. Vector supernatant was accumulated in a sterile container (20 liters) at 4.degree. C. The vector titer was on the order of 10.sup.7 cfu/ml.

The supernatant was concentrated 2-fold through tangential flow filtration (Millipore, Pellicon regenerated cellulose membrane, NMWC, 300K). This was followed by diafiltration of the 2.times. concentrated supernatant with 4.times. volume of a buffer of 0.02M Bis-Tris+0.1M NaCl, pH 5.5. After completion of buffer exchange, the supernatant was concentrated further by 5-fold. The concentration/diafiltration was carried out as one step. The total concentration factor was 10-fold. Table VIII below shows the concentration factor, vector titer, and vector recovery during the concentration/diafiltration process.

TABLE VIII______________________________________Concentration and diafiltration of vector supernatant Initial 2X Concentrate Post Diafiltration______________________________________Volume (1) 20 10 2Concentration Factor -- 2 10Titer (cfu/ml) 4.5 .times. 10.sup.6 8.6 .times. 10.sup.6 4.2 .times. 10.sup.7Recovery (%) -- 95 .+-. 10 93 .+-. 10______________________________________

The concentrated/diafiltered vector supernatant was filtered through a 0.22 .mu.m filter before being purified by ion-exchange chromatography. A variety of ion-exchange resins were tested under different buffer conditions. The optimal buffer condition was 0.02M Bis-Tris at pH 5.5. The resin which gave the highest vector recovery and purification was Toyopearl Super Q 650M (Tosohaas). This resin is formed from a matrix of a copolymer of ethylene glycol and methacrylate, and has attached to said resin quaternary amine groups having the formula --N.sup.+ --(CH.sub.3).sub.3. The ion-exchange resin was packed into a column (diameter 16 mm, height 150 mm). The column was washed and equilibrated with 10 column volumes of buffer (0.02M Bis-Tris+0.1M NaCl, pH 5.50). The concentrated and diafiltered vector supernatant (30 ml) was loaded onto the column at a flow rate of 1.5 ml/min. After loading the sample, the column was washed with buffer A until the O.D. reading at 280 nm, which was monitored continuously via a UV monitor, reached baseline level. Then the column was eluted with a linear NaCl gradient made of 80 ml of buffer A and 80 ml of buffer B (0.02M Bis-Tris+1.2M NaCl, pH 5.50). The linear NaCl gradient was generated in a gradient mixer. Vector fractions were collected during elution. A purification factor of 40 was achieved by a single ion-exchange chromatography with vector recovery of 35% as indicated in Table IX.

TABLE IX______________________________________ Protein Specific Purifi- Titer concentration activity cation RecoverySteps (cfu/ml) (.mu.g/ml) (cfu/.mu.g) factor (%)______________________________________Super- 5 .times. 10.sup.6 1,100 5 .times. 10.sup.3 -- --natant (2%FBS)Diafiltra- 5 .times. 10.sup.7 1,600 3 .times. 10.sup.4 6 95tion andconcentra-tionIon- 3 .times. 10.sup.7 20 1.2 .times. 10.sup.6 240 35exchangechroma-tography(Toyo-pearlSuper Q)______________________________________

Table X below shows the comparison of vector recovery from different resins under different buffer pH values. Fractional TMAE (EM Separations Technology, Gibbstown, N.J.) is formed from a resin which is a copolymer of oligoethylene glycol, glycidyl-methacrylate, and pentaerythrol-dimethacrylate. The resin is charged with trimethylaminoethyl, C.sub.2 H.sub.4 N(CH.sub.3).sub.3 groups.

TABLE X______________________________________ Resins Toyopearl Fractogel Super Q TMAEElution Linear Linear______________________________________pH 5.5 7.0 5.5 7.0Recovery diafiltration 35% 1% 0.2% 0.5%(%)______________________________________

The purified retroviral vector from the ion-exchange chromatography had a relatively high salt concentration (0.9M NaCl) and a pH of 5.50. The purified vector supernatant was concentrated 2.times. through tangential flow filtration (Millpore, Minitan, NMWE 100K regenerated cellulose membrane). The 2.times. concentrated supernatant was buffer exchanged with 4.times. Volume of the formulation for lyophilization. (5% sucrose, 10% mannitol, in water). After completion of the buffer exchange, the supernatant was concentrated further 5-fold. A total concentration factor of 10-fold was achieved during this second concentration/diafiltration step. At this stage, the vector was ready for lyophilization.

Vector supernatant which was diafiltered into the lyophilization buffer (5% sucrose, 10% mannitol, in water) was spiked with human serum albumin (HSA) to a final concentration of 1%. The complete formulation was filtered through a 0.22 .mu.m filter. The filtered vector solution was placed aseptically into lyophilization vials (1 ml per vial into 3 ml vials). The vector was lyophilized in a lyophilizer following a 21/2 day lyophilization cycle to achieve a final moisture content of 1 to 2%. Vector recovery in the order of 90% is achieved. The lyophilized vector was stored at 4.degree. C. until use.

All publications cited herein are incorporated by reference.

It is to be understood, however, that the scope of the present invention is not to be limited to the specific embodiments described above. The invention may practiced other than as particularly described and still be within the scope of the accompanying claims.

Claims

1. A method of purifying infectious retroviral vector particles to obtain a purified retroviral supernatant containing infectious retroviral particles, comprising:

(a) generating retroviral vector particles by culturing retroviral vector producer cells;

(b) obtaining a supernatant containing said retroviral vector particles from the culture of retroviral vector producer cells;

(c) concentrating said supernatant;

(d) diafiltering said supernatant;

(e) subjecting said supernatant to ion-exchange chromatography;

(f) concentrating said supernatant; and

(g) diafiltering said supernatant to obtain a purified retroviral supernatant containing infectious retroviral particles.

2. The process of claim 1 and further comprising:

(h) lyophilizing said infectious retroviral vector particles.

3. The method of claim 1 wherein said supernatant is concentrated in steps (c) and (f) through a tangential flow filtration system.

4. The method of claim 1 wherein said subjecting of said supernatant to ion-exchange chromatography in step (e) comprises subjecting said supernatant to ion-exchange chromatography on an ion-exchange chromatography resin including an anion exchange group having the formula: --O--L--N.sup.+ --(R).sub.3.sup.-, wherein L is a linker group and R is an alkyl group having one or more carbon atoms.

5. The method of claim 2 and further comprising:

(i) reconstituting said retroviral vector particles.

6. The method of claim 4 wherein R is methyl.

7. The method of claim 4 wherein each of L and R is ethyl.