Title: Serum-free, low-protein media for rotavirus vaccine production
United States Patent: 6,656,719
Issued: December 2, 2003
Inventors: Gould; Sandra L. (Tinton Falls, NJ); Robinson; David K. (New York, NY); Distefano; Daniel J. (Cranford, NJ); Seamans; T. Craig (Westfield, NJ)
Assignee: Merck & Co., Inc. (Rahway, NJ)
Appl. No.: 176492
Filed: October 21, 1998
Defined serum-free, low protein media (LPKM), that supports 1) Vero cell growth for up to 20 passages, 2) Vero cell growth on microcarriers and 3) rotavirus production is provided. Maximum cell densities attained are 60-100% of that in serum-containing medium; the doubling time is equal to that for cells in serum containing medium. Rotavirus titers achieved in LPKM-1 are 50-100% of the serum-containing process. Finally, since LPKM-1 contains no animal-sourced proteins, the problems associated with the serum-containing rotavirus production process (i.e. lengthy wash steps before infection, potential introduction of adventitious agents and lot-to-lot variability) can be avoided; while maintaining nearly equivalent product titers.
SUMMARY OF THE INVENTION
Defined serum-free, low protein media provided herein, supports 1) Vero cell growth for up to 20 passages, 2) Vero cell growth on microcarriers and 3) rotavirus production. We refer to media as LPKM. Maximum cell densities attained are 60-100% of that in serum-containing medium and the mean doubling time is equal to serum containing medium. Rotavirus titers achieved in LPKM-1 are 80-100% of the serum-containing microcarrier process. Since LPKM media contain no animal-sourced proteins, the problems associated with the serum-containing rotavirus production process (i.e. lengthy wash steps before infection, potential introduction of adventitious agents and lot-to-lot variability of serum) are avoided while maintaining nearly equivalent product titers.
DETAILED DESCRIPTION OF THE INVENTION
Serum-free, low protein media are provided by the present invention. The media support Vero cell growth in static and agitated environments on standard cell culture surfaces and support rotavirus production. These media are referred to herein as LPKM-1, -2 and -3.
LPKM media contain many of the supplements found in K-1 medium (Taub and Livingston 1981). However, K-1 alone was not sufficient to sustain Vero cell growth for multiple passages. By adding recombinant human Epidermal Growth Fetor (rhEGF) and dexamethasone (dex), it is possible to directly transfer Vero cells to this medium (called K-1+rhEGF/dex) and sequentially subculture them at least 20 times (53 PDL). Dex and rhEGF have been shown to act synergistically to stimulate proliferation of primary human diploid fibroblasts (Baker et al. 1978) and have been used in serum-free media designed to support clonal growth and repeated passages of human diploid fibroblast cell lines (Bettger et al. 1981). Further improvements were made by changing the mixture of basal media from F-12: DMEM to F-12: Medium 199, which resulted in increased growth rates and final cell densities.
Animal sourced proteins were removed or replaced in LPKM-1. When possible, recombinant proteins expressed in Escherichia coli replaced animal-sourced proteins. Transferrin has long been recognized as a necessary component of serum-free medium for the growth of many different cell lines (Barnes and Sato 1980; Bjare 1992). However, some cells can use inorganic iron and chelated iron, e.g. ferric citrate (Franek and Dolnikova 1991), sodium nitroprusside (Darfler 1990) and pyridoxal isonicotinoyl hydrazone (Brock and Stevenson 1987). We have found that Vero cells grow in the absence of transferrin in 1:1 mixtures of DMEM:F-12 and F-12:199. No additional iron was required to replace transferrin. The combination of ferrous and ferric salts found in F-12: DMEM and F-12: Medium 199 presumably provide sufficient iron to support Vero cell growth.
Vero cells grown in static culture in LPKM-1 grew more slowly and attained a lower saturation density than did Vero cells grown in serum-containing medium. It is not known whether the increased doubling time is due to a general lengthening of the cell cycle, or if a smaller but constant proportion of cells proliferate in LPKM-1 at each passage. The modification of glucose, sodium bicarbonate, and sodium pyruvate concentrations in LPKM-2 resulted in Vero cell growth rate and maximum cell density in microcarrier culture equal to those reported in serum-containing medium. The addition of several trace elements, 2-mercaptoethanol, ethanolamine and the modification of some vitamin concentrations in LPKM-3 have yielded improvements in Vero cell growth in static culture.
Merten et al. (1994) have reported that a serum-free medium containing 40 mg/l protein (MDSS2) was able to support Vero cell growth in static and agitated environments. Vero cells were subcultured a total of 19 times and achieved the same growth rate as cells in their serum-containing control. However, the reported growth rate of cells in. MDSS2, 0.0098+/-0.0014 h-1 (71 h doubling time), is 40% less than that for Vero-cells in LPKM-1 (0.017 h-1). The maximum cell density attained in LPKM without a medium exchange is only about 85% of that in serum-containing medium. Clark et al. (1982) have suggested that Vero cells grown in serum-free media reach a lower saturation density because each cell spreads over a larger surface area. However, Vero cells grown in protein-free medium on polyvinyl formal (PVF)-treated surfaces reach saturation densities comparable to cells in serum-containing medium (Cinatl et al. 1993). These findings indicate that saturation density is a function of both the cell/substrate interaction and the medium composition.
We believe that this is the first report of rotavirus production using completely defined, serum-free, low protein (5 mg/l) media for both Vero cell growth and rotavirus production. The yield per cell of rotavirus in LPKM without a medium exchange was nearly equivalent to that for cells grown in serum-containing medium. The reduced virus titer was a reflection of the lower cell densities reached in the absence of a medium exchange. The virus stock was generated as follows: Vero cells were grown in serum-containing medium, washed extensively with phosphate buffered saline, then infected in serum-free basal medium containing trypsin. Although other viruses have been adapted to growth in serum-free media by repeated passaging, the possibility of adapting the rotavirus to propagate better in LPKM-1 was not explored.
Recently, other viruses have been produced using Vero cells grown in serum-free media. Maximal rabies virus titers in MDSS2 were 1.5-fold greater and were acheived 1-2 days earlier than in serum-containing medium (Merten et al. 1994). The titers of Coxsackievirus, poliovirus (3 strains), measles virus and Herpes simplex virus (2 strains) produced from Vero cells grown in PFEK-1, a protein-free medium, were also shown to be similar to titers produced in a serum-containing process (Cinatl et al. 1993). However, viral titers obtained from cells grown in PFEK-1 were equal to those from cells grown in serum-containing medium only when cells were grown on PVF-treated surfaces.
Because vaccines are generally administered to healthy individuals, the safety of the product is a primary concern. By eliminating animal-sourced proteins from LPKM media, the risk of introducing adventitious agents is reduced. We have also demonstrated that a tumorigenic phenotype is not induced in Vero cells, as measured by colony formation in soft agar, when sequentially passaged in serum-free medium. Finally, LPKM media contains only defined chemical components and recombinant proteins, thereby providing for improved lot-to-lot and process consistency.
Media and Chemicals
Media (Dulbecco's Modified Eagle's Medium, Medium 199 and Nutrient Mixture F-12), dexamethasone (dex), hydrocortisone (HC), and triiodothyronine (T3) were from Sigma Chemical Company (St. Louis, Mo.). Recombinant human epidermal growth factor (rhEGF) was from R&D Systems (Minneapolis, Minn.), and prostaglandin E1 (PGE1) was from BioMol (Plymouth Meeting, Pa.). Fetal bovine serum (FBS) and GlutaMax I were from Gibco (Grand Island, N.Y.). L-glutamine was from Mediatech (Washington, D.C.) or JRH (Lenexa, Kans.).
The compositions of LPKM media are provided in Table 1. Media of this invention can be formulated by standard methods known in the art. For convenience, a formulation for LPKM-1 is provided in Table 2. LPKM-2 and LPKM-3 can be similarly formulated by modifying the concentrated solutions shown for formulating LPKM-1. Alternatively, additional components can be added separately or individually, as will be understood by one of skill in the art.
Cell Propagation in Static Culture
Vero cells (ATCC, CCL-#81, passage 121) were grown either in Dulbecco's Modified Eagle's Medium (DMEM) with 10% FBS and 2 mM L-glutamine or in serum-free media with 2 mM L-glutamine. Cells were used between passages 123 and 146. K-1 serum-free medium was prepared as previously described (Taub and Livingstson, 1981). Other additives to the serum-free media were prepared as follows: rhEGF was reconstituted according to manufacturer's protocol, except that albumin was omitted from the reconsititution medium; dex was dissolved at 1 mg/ml in absolute ethanol. Final media concentrations of all components are as stated in the LPKM recipe table (Table 1). rhEGF and all hormones were stored at -20oC. and were frozen and thawed a maximum of two times.
For cell growth and virus production in static culture, Vero cells were grown in 75 cm2 cell culture flasks (Corning/Costar, Cambridge, Mass.). Cells were subcultured using 0.25% trypsin/1 mM EDTA solution, and plated at 1.3-4.0x104 cells/cm2. For cells in serum-containing medium, trypsin/EDTA was removed prior to cell detachment and cells were resuspended in an appropriate volume of serum-containing medium after detachment. For cells in serum-free medium, the trypsin/EDTA was not removed prior to cell detachment. Trypsin activity was quenched with an equal volume of a 1 mg/ml soybean trypsin inhibitor (STI) (Gibco) solution. The resulting cell suspension was centrifuged at 250x g for 10 minutes, the supernatant aspirated, and the cell pellet resuspended in an appropriate volume of serum-free medium. Cell viability was determined by exclusion of trypan blue dye (0.4%, diluted 1:1 with the cell suspension) and cell numbers were determined using a hemacytometer. Verification of improved cell density and/or growth rate was obtained by growing the cells for at least three passages in the media or additives identified in the screens.
Growth on Microcarriers
Microcarrier experiments were conducted in siliconized 250 ml spinner flasks (Bellco Microcarrier Flask) using Cytodex I microcarriers (Pharmacia) at a concentration of 3 g/l. Microcarriers were rehydrated and sterilized according to the manjufacturer's recommendations. After sterilization, beads were rinsed once with growth medium prior to inoculation. Cultures were inoculated with 1-2x105 cells/ml using cells grown in 225 cm2 cell culture flasks (Corning/Costar). After inoculation, the cultures were agitated at 35 rpm throughout both the cell growth and virus infection phases. Cultures were sampled daily for cell counts. Briefly, microcarriers were allowed to settle and the medium gently removed. An equal volume of crystal violet solution (Hu and Wang 1987) was added and the solution incubated at 37oC. for 30 minutes. The solution was then vigorously vortexed and the nuclei counted using a hemacytometer.
Screening of Media and Additives
The effect of media or additives on final cell density was determined using the following method. Cells were plated in duplicate wells for each test condition at a density of 2x104 cells/cm2 in 24-well plates. After incubation at 37oC. for 5 days, cells were removed using 0.5 ml trypsin/EDTA per well. Trypsin activity was quenched by adding 0.5 ml of 1 mg/ml STI and the cells thoroughly resuspended. Cell viability and concentration were determined as described above.
In some cases, cell growth was monitored using a Cellstat.TM. system (Cellstat Technologies, Palo Alto, Calif.). Cells were plated in triplicate wells for each test condition in 24-well plates at 2x104 cells/cm2. A Bioprobe.TM. lid was sterilized according to manufacturer's directions and placed on the 24-well plate. The lid and plate were then inserted into a measurement board located in an incubator. Cells were incubated for 5-7 days in a humidified atmosphere of 5% CO2. Medium conductivity in each well was automatically measured and recorded every hour and a growth curve generated. The slopes of the growth curves were used as an indicator of relative growth rate.
Tumorigenicity testing was done using colony growth in soft agar using standard techniques known in the art. (MacPherson and Montaigrier 1964).
A bovine-human reassortant rotavirus known in the art was used in experiments (F. Clark, Children's Hospital of Philadelphia) (U.S. Pat. No. 5,626,851, May 6, 1997). Bovine rotavirus strain WC3 and human rotavirus strain W179, which contains the G serotype 1 outer capsid protein, were used to coinfect MA104 (fetal green monkey kidney) cells. Progeny virus were plaque purified and a strain, W179-9, identified in which only the VP7 (or G) gene segment is derived from the human rotavirus strain. Vero cells used for infection were maintained between passages 125 and 138. Cells growing in serum-containing medium in cell culture flasks were rinsed twice with Dulbecco's Phosphate Buffered Saline (D-PBS) (approximately 2 minutes per wash) before infection. DMEM containing 2 mM L-glutamine but no FBS was added prior to infection. For serum-containing microcarrier cultures, cells were washed three times with D-PBS. Each wash consisted of a 30-45 minute incubation with agitation at 37oC. After washing, one volume of DMEM with 2 mM L-glutamine was added and the cells were infected. Cells propagated in serum-free media were sometimes washed with D-PBS and the medium changed to DMEM with 2 mM L-glutamine. In other cases, either 75% of the medium was removed and replaced with an equal volume of fresh medium immediately prior to infection or the medium was not exchanged. Virus was added from a stock solution at a multiplicity of infection (MOI) of 0.5. Trypsin/NaCl (Gibco), at 12.5 or 25 mg/ml, was added immediately after the virus inoculum. Infected cultures were either incubated at 37oC. or switched to 34oC. immediately after infection.
Cultures were sampled daily for 3-5 days following infection. One ml samples were removed and stored at -70oC. until assay.
Claim 1 of 2 Claims
What is claimed:
1. A Serum-free media for the production of rotavirus selected from the group consisting of LPKM-1, LPKM-2 and LPKM-3.