CA1312030C

CA1312030C - Method to increase antibody titer

Info

Publication number: CA1312030C
Application number: CA000581361A
Authority: CA
Inventors: Brian Maiorella; Duane Inlow; William Howarth
Original assignee: Cetus Corp
Current assignee: Novartis Vaccines and Diagnostics Inc
Priority date: 1987-11-18
Filing date: 1988-10-26
Publication date: 1992-12-29
Anticipated expiration: 2009-12-29
Also published as: WO1989004867A1; AU2805989A

Abstract

METHOD OF INCREASING PRODUCT EXPRESSION
THROUGH SOLUTE STRESS

Abstract of the Disclosure A method of determining the optimal level of product expression and cell growth of animal cell cul-ture is described. The method generally comprises culturing cells under conditions of solute stress, that is, under conditions whereby optimal cell growth is decreased yet levels of product expression are in-creased. In a preferred embodiment of the invention is described a method of increasing the yield of mono-clonal antibodies comprising culturing hybridoma cells in an environment of solute stress. One approach to the creation of such an environment is the addition of inorganic salts, organic polyols, or metabolic prod-ucts to the culture medium. One to three fold in-creases in antibody yield have been obtained by these methods.

Description

-1- 13l~`s~

METHOD OF INCREASING PRODUCT EXPRESSION
THROUGH SOLUTE STRESS

Field of the Invention The present invention is in the qeneral field of biochemical engineering. More specifically, this invention is in the field of c011 and tissue cul-ture dealing primarily with somatic hybrid cell cul-ture.

Backqround of the Invention With the advent of hybridoma technology and the accompanying availability of monoclonal antibod-ies, the application o-f such antibodies has escalated into a variety of areas of the biological sciences.
For example, monoclonal antibodies have been used for the study of cell surface antigens, for affini-ty puri-fication of proteins, for histocompatibility testing, for studying various viruses and for radioimmunoassay.
More recently, it has been recognized that monoclonal an~ibodies may have medical application for drug tar-geting and immunotherapy (C.H. Poynton and C.L.Reading, (1984) Exp Biol 44:13-33). With the in-creased application of the antibodies in the biologi-cal and medicinal sciences, there has come a con-comitant demand for high levels of an-tibody produc-tion.
To date, efforts have been undertaken todevelop culture condi-tions -to maximize cell culture growth and thereby increase resultant product yield.

~ .

Early work in the development of chemically defined animal cell culture media focused on the formulation of such media to achieve rapid cell proliferation ~P.R. White, (1946) Growth 10-231-289, and C. Waymouth (1974) J Natl Cancer Inst 53:1443-1448). Such media incorporats specific nutrients, especially amino acids, vitamins, purines, and pyrimidines. Today some of the more widely used basal media for mammalian cell cultures include Hams F-12, Dulbecco's modified Eagle's medium (DM~), RPMI 1640, and Iscove's modified DME. All of these above-referenced basal media are also supplemented with several trace metals and salts, including the major cations (potassium, sodium, cal-cium, magnesium and the like) with concentration val-ues near isotonic levels. The role of inorganic nu-trition in cell culture is discussed in a number of references including R.A. Shooter and G.O. Gey, (1952) Br J Exp Pathol 33:98-103; C. Waymouth, (1974) supra;
J.R. Burch and S.J. Pert, (1971) J Cell Sci 8:693-700;
~.G. Ham, Growth of Cells in Hormonally Defined Media, Cold Spring Harbor Conferences on Cell Proliferation, Vol. 9, Sato, Pardee and Sirbashin, eds., 1982.
Culture media have been developed spe-cifically for low serum and serum-free mammalian cell cultures for production of monoclonal antibodies. One such serum-free medium is disclosed in European Patent Publication 076,647, published 13 April 1983. Other media have been developed by changing levels of sup-plements such as trace elements, vitamin and hormone additives wherein variations in the traditional basal media are slight. References to such media include, for example, D. Barnes and G. Sato, (1980) Cell _:649-655; W.L. Cleveland et al (1983) J Immunol Meth 56:221-234; N. Iscove and F. Melchers, (1978) J
Exp Med 147:923-933; T. Kawamo-to et al (1983) Anal Biochem 130:445-453; J. Kovar and F. Franek, (1984) _3_ l 31 20~0 Immunol Lett 7:339-345; H. Murakami et al (1983) Agric Biol Chem 47(8):1835-1840; H. Murakami et al (1982) Proc Natl Acad Sci USA 79:1158-1162; H. Muzik et al (1982) In Vitro 18:515-524; and S.D. Wolpe, ~'In Vitro Immunization and Growth of Hybridomas in Serum-Free Medium", in J.P. Mather, ed., "Mammalian Cell Culture," Plenum Press, New York, 1984.
In addition to providing the right kinds and amounts of nutrients, the culture medium must also provide suitable physiochemical conditions. Param-eters that are important for clonal growth of hybrid-oma cell culture include osmolarity, pH buffering, carbon dioxide tension, and partial pressure of oxygen. These all must be adjusted to optimal values for multiplication of each type of cell with, prefer-ably, minimal or no amounts of serum and minimal amounts of protein. Other physical factors such as temperature and illumination must also be controlled carefully.
Effor-~s to increase antibody yield have focused primarily on means to optimize cell growth and cell density. The optimal conditions for cell growth of ma-mmalian cell culture are generally within narrow ranges for each of the parameters discussed above.
For example, typical culture conditions for mammalian hybridoma cell culture use a basal culture medium sup-plemented with nutritional additives, pH in the range of 6.8 to 7.4 at 35-37C.
As a general point of reference, antibody titers from murine hybridoma cell lines are highly variable from cell line to cell line and range typically from 10 to 350 ug/ml (K.J. Lambert et al (1987) Dev Indust Microbiol 27:101-106). Human mono-clonal antibody expression from human/human or human/
mouse fusions are also highly variable from cell line to cell line and range typically from 0.1 to 25 ug/ml 1 3 1 203~
4- ;
(R. Hubbard, "Topics in Enzyme and Fermentation Biotechnology," chap. 7, pp. 196-263, A. Wiseman, ed., John Wiley & Sons, New York, 1983). These values are indicative of culture conditions that are op-timized for cell growth and cell viability.
Another example from the literature documents ~hat, at least for some cell lines, mono-clonal antibody production proceeds even after a cul-ture stops growin~ (D. Velez et al, (1986) J Imm Methods 86:45-52; S. Reuveny et al, (1986) ibid at p.
53-59). Thus, one strategy for increasing monoclonal antibody yield has been to develop culture conditions that allow growth of hybridomas to higher cell densi-ties and to recover the antibodies late in the sta-tionary phase of cell culture. W. Arathoon and J.Birch, (1986) Science 232:1390-1395 reported that a 1,000 liter hybridoma fermentation produced about 80 grams of monoclonal antibody during the growth phase and another 170 grams of antibody during an extended stationary/death phase. It is not known the means, if any, by ~hich -the stationary phase of growth was ex-tended.
Another approach from the literature to in-creasing antibody production is to achieve high cell densities by cell recycle o-r entrapment methods.
Examples of these me-thods include ho]low fiber re-actors (G.L. Altshuler et al (1986) Biotechnol Bioenq XXVIII, 646-658); static maintenance reactors (J.
Feder et al, Canadian Paten-t 1,210,352 issued 26 Au~. 1986;
ceramic matrix reactors (A. Marcipar et al (1983) _ nals N.Y. Acad Sci 413:416-420); bead immobili~ed reactors (K. Nilsson et al (1983) Nature 302:629-630);
perfusion reactors (J. Feder and W.R. Tolbert, (1985) Amer Biotechnol Lab III:24-36) an~ others. In some cases, a res-ting" cell culture s-tate is reported to be achieved by reducing levels of nu-trien-ts in the ~. " .

.

medium (as by reducing serum or protein supplement levels) with antibody production continuing while growth is slowed.
While a variety of methods to increase anti-body yield from hybridoma cell culture are beingexplored, the primary focus is still on the optimiza-tion of cell growth. We have discovered that culture conditions for growth optimization and for optimal product expression may differ and that product expres-sion can be increased under condi-tions of solu-te stress, created by the addition of certain solutes, notwithstanding the resulting growth inhibitory effects.
The concep-t of subjecting animal cells, especially mammalian cell cultures, to an environment of solute stress to produce higher product expression yields, such as increased antibody titers, has not been reported. One means for introducing such an en-vironment to the culture is through salt addition which is easily monitored by measuring the osmolarity of the culture medium.
Media osmolarity for mammalian cell culture is usually held in the range of 280-300 mOsM/kg (W.B.
Jakoby and I.~. Pas-tan, Meth Fnzymol, vol. LVIII, 25 "Cell Culture", Academic Press (1979), pp. 136-137).
Of course, the optimal value may depend upon the spe-cific cell type. For example, as reported in "Tissue Culture, Methods and Applications", edited by P.F.
Kruse, Jr., and M.K. Patterson, Jr., ~cademic Press 30 (1973) p. 704, human lymphocytes survive best a-t low (about 230 mOsM), and granulocytes at higher osmolar-ities (about 330 mOsM). Mouse and rabbit eggs develop optimally in vivo at around 270 mOsM, 250-280 mOsM
being satlsfactory, while above 280 mOsM development is retarded. Iscove reports 280 mOsM to be optimum for grow-th of murine lymphocytes and hema-topoietic 1 -~ 1 20 30 cells, and Iscoves modified DME is adjusted for this growth promo~ing osmolarity (N.N. Iscove, ~1984) "Method for Serum-Free Cul-ture of Neuronal and Lymphoid Cells," pp. 169-185, Alan R. Liss, ed., New York.
The spread of quality control osmolarity values on a number of commercially available tissue culture media is provided in a table beginnin~ at page 706 in the "Tissue Culture, Methods and Applica-tionsl' reference, supra. The osmolarity values given therein reflect the 280-300 mOsM/kg range used for mammalian cell culture.
Ano-ther means to introduce an environment of solute stress in the cell culture is through the addi-tion of cellular metabolic products, such as lacticacid and ammonia. These products are generally known to be growth inhibitory agents and strategies to reduce the level of these products in the culture medium in or~er to enhance cell growth have been reported. T. Imamura et al (1982) Anal Biochem 124:353-358; A. Leibovitz, (1963) Am J Hyq 78:173-180; S. Reuveny et al (1986) J Immunol Meth 86:53-59;
J.S. Thorpe et al (1987) "The Effect of Waste Products of Cellular Metabolism on Growth and Protein Synthesis in a Mouse Hybridoma Cell Line", Paper #147 presented at American Chemical Society National Meeting, Aug.
30-Sept. 5, 1987, New Orleans, La.--Symposium on Nutrition and Metabolic Regulation in Animal Cell Culture Scale-Up; and M.W. Glacken et al (1986) _iotech & Bioeng XXVIII:1376-1389.
Contrary to the -teaching in the art which cautions against major adjustments to culture media osmolarity and other physiochemical parameters, we have found that introducing an environment of solute stress during fermentation can favor an increase in specific (per cell) antibody expression and/or in-~ 7 t ~ 1 2030 creased culture longevity which can result in an increase inantibody titer. It is to such a concept that this invention is directed. Briefly, in a preferred embodiment of the invention, an approach to mammalian cell culture which further optimizes yield of antibody production has been developed in which hybridoma cells are cultured uncler conditions of controlled solute stress. Optionally, the method incorporates prior art advances including the culture of hybrid mammalian cell lines in serum-free media or in high density culture to reduce costs and facilitate purification.

Summary of the Invention Therefore, this invention is directed to a method for increasing expression of a protein in a mammalian cell culture, which already provides for all cell growth requirements, and recovering the protein from the culture, the expression i5 increased above the expression level at optimal growth, comprising adding a solute to the cell medium at a level above that for optimal cell growth to create stress on the cell, as expressed by an inhibitory effect on cell growth or cell densiky; and recovering and purifying the protein from the cell culture.
In another aspect of this invention is provided a method to determine the solute level in a cell culture medium ko produce the highest protein product expression from a cell, the culture medium has all the requirements necessary for optimal growth comprising; determining the concentrations of solutes necessary for optimal cell growth;
increasing the concentrations of one or more solutes to place the cell under stress; and determining the concentrations which produce more protein product.

.

- 8 - l 3 1 203 0 A preferred method of this invention comprises culturing human IgM-producing hybridoma cells and another preferred method comprises culturiny hybridoma cells which produce IgG.
These and other objects of the invention will be apparent from the following description and claims. Other embodiments of the invention embodying the same or equivalent principles may be used and substitutions may be made as desired by those skilled in the art without departing from the present invention and the purview o~ the appended claims.

Brief Description of the Drawinqs Figure l shows the effect of 400 mOsM media on antibody yields of human/human/murine trioma D-234 cells in serum-free HL-1 media. The closed circles represent cell growth in 300 mOsM media and the open circles represent the resulting IgM antibody yield. The closed squares represent cell growth in 400 mOsM media and the open squares represent resulting IgM antibody yield.
Figure 2 shows the effect of ammonium chloride on production of antibodies of D-234 cells. The closed circles represent cell growth in the absence of ammonium chloride and the open circles represent the resulting IgM antibody yield. The open triangles represent cell growth in the presence of 10 mM ammonium chloride and the closed triangles represent resulting antibody yield.
DescriPtion of the Preferred Embodiments As used herein the term "hybridoma" refers to a hybrid cell line produced by the fusion of two or more cell lines to produce an immortal cell line producing a desired product - 8a - 1 31 203D

(such as an antibody). The term includes hybrids produced by the fusion of a myeloma cell line and an antibody producing cell (such~as a splenocyte or plasma cell). The term also includes prog-:, f.

1 3 I 203~) g eny of heterohybrid myeloma fusions (the result of a fusion with human B cells and a murine myeloma cell line) subsequently fused with a plasma cell, referred to in the art as trioma cell lines.
5As used herein the term "animal" refers -to any mammalian, insect or invertebrate species.
"Mammalian" indicates any mammalian species, and includes rabbits, mice, dogs, cats, primates and humans, preferably humans.
10As used herein the term "solute" indicates a water soluble agent, including but not limited to in-organic salts and the correspon~ing ions thereof;
~` o~ganic polyols, including~g~c~ro~ ~ d sugars such '' as, for example, glucose, mannose, fructose and mannitol; and metabolic products such as, for ex-ample, lactate or ammonia; which is effective in pro-ducing increase~ product expression.
As used herein -the term "solute stress~
refers to the addition of solutes in such concentra-tions, a-t least above that concentration determined for optimal cell growth, that produce a growth in-hibitory effect or reduced final cell density, that is, a growth rate or maximum cell density less than that determined for optimal growth. ~owever, the level of product expressed at this reduced growth level is comparatively greater than that level of ex-pression achieved at the optimal growth rate owing to an increase in specific (per cell) product expression rate or an increase in longevity of the culture.
30As used herein the term "osmolality'l refers to the total osmotic activity contributed by ions and nonionized molecules to a media solution. Osmolality, like molality, relates to weight of solvent (mOsM/kg H2O) while osmolarity, like molarity, relates to vol-ume (mOsM/liter solution). Osmolality is one method used to moni-tor solute stress. St~ndard osmolality -10- 1 Jl 2~0 refers to the optimum range of clonal growth of mam-malian cells which occurs at 290-~30 mOsM/kg.
In a preferred embodiment of the invention, - methods have been developed for the high~level produc-tion of mammalian, preferably human~ monoclonal anti-bodies for use as diagnostic reagents or for use in human therapy. In particular, a method of determining the op-timal level of product expression in mammalian cell culture has been developed wherein the concentra-tion of a solute of interest in a culture mediwn com-position for optimal product expression is different than the culture medium cornposition determined for optimal cell growth, which method comprises:
a) growing the mammalian cell culture in medium to determine optimal cell growth;
b) varying the concentration of -the solute in the culture medium to a concentration above that optimal for cell growth which concentration is effec-tive to create an environment of solute stress on the cell culture;
c) monitoring the product expression under the varying solute concentrations to determine op-timal product expression; and d) selecting the solute concentration that provides the optimal combination of cell growth and product expression which allows for optimal productiv-ity.
Following the methodology set forth herein, one is able to determine -the solute concentration that provides the optimal combination of cell growth and product expression for any particular cell line of interest. Once -the solute concentration has been determined, one is able to create an environmen-t of controlled solute stress for culturi.ng the mammalian cell lines and thereby s-timulate specific (per cell) product expression and/or increase culture longevity, I 3 1 ~030 notwithstanding the inhibitory growth effect on the cul-tured cells.
The mammalian cell culture used in the present invention includes, but is not limited to, any of a number of cell lines of bo-th B-cell and T-cell origin including murine thymic lymphoma cells, human myeloma cell lines, and human lymphoblastoid cells and hybridomas. Accordingly, the product -to be op-timized includes growth factors, lymphokines, and monoclonal antibodies. The cell cultures may include cell lines which are found to naturally produce such desired pro-ducts, or have been manipula-ted by genetic engineering techniques to produce recombinant products.
Solute stress is introduced into the cell culture fermentation by the addition of one or more solutes which effectively inhibit optimal cell growth.
The solute can be added at various time periods during the fermentation including prior to, during or aEter the addition of cells. While such changes to the cul-ture media negatively affect the growth of culturedcells (given the narrow growth parameters known for optimal cell growth) the present invention lies in the discovery that culturing cells in such an environment of solute stress can positively impact specific cell productivi.ty and culture longevity, thereby increasing product yield.
Solute stress which is effective in increas-ing the product yield can be achieved by increasing the concentrati.on of a solute already presen-t in a culture medium or introducing a new solute to the medium.
In the method of the invention, a sub-lethal solute concentration range is firs-t determined in order to study the solute inhibi-tory growth effect.
This determination is necessary as each cell line may have unique tolerance levels to the selected solute.

- 12 - 1 3 ~ 2 0 ~ O
As a second step, various sub-lethal concentrations are studied in more detail to establish the conditions for optimal cell productivity which is responsible for increased product expression. From the data thus generated, one may determine the solute concentration that provides for the optimal combination of cell growth and product expression.
The following discussion, concerning the various types of solutes that may be used in the methods of the present invention, also provides a number of preferred concentration ranges that have been determined for specific hybridoma cell lines. Other cell lines may have somewhat different tolerance levels. These ranges are provided as a guide for determining the optimal combination of growth and product expression levels for a variety of cultured cells and are not to be construed as a limitation of the invention.
The concentration ranges provided herein are a good indicator of a possible concentration range for the specific cell line of interest.
The solutes of the invention comprise a number of inorganic salts and ions thereof, including, for example, sodium chloride, potassium chloride, calcium chloride, magnesium chloride and ~he like, and combinations thereof.
Preferred salts include sodium chloride and combinations of sodium chloride and potassium chloride. An effective concentration range for the increased production of monoclonal antibodies by the cell lines D-234 and T-88, using salts such as sodium chloride is 340 to 460 mOsM/kg, with 350 to 400 mOsM/kg being more preferred for the cell line D-234 and 400 to 450 mOsM/kg being more preferred for the cell line T-88. An effective concentration for the increased per cell productivity of monoclonal antibodies by the cell line 454A12, using sodium chloride, is about 400 mOsmol/kg.
The concentration values given above, as well as all concentration ranges provided herein regardless of the method of solute concentration meas-urement used, have been established prior to the addition of cells. However, the solute may be added before, duxing or after cell addition. The timing of the solute addition is generally not critical, as it has been found that increasing solute stress by, for example, salt addition, may be performed at various time points during the exponential phase of the growth cycle to achieve an increase in antibody yield. Of course, one skilled in the art will appreciate that the concentration of the metabolic solutes will increase during the course of -the fermentation.
In addition to the aforementioned salts, it has been found that solutes which are generally be-lieved to have inhibitory growth effects may also be used in the present invention. For example, lactic acid, a major metabolic end product of glycolysis in hybridoma cell culture, participa-tes in the lowering of the pH during growth, producing sub-optimal growth conditions. The lactate ion itself, may also be growth inhibitory. Efforts have been made to reduce lactic acid production by replacing glucose with alternative sugars (i.e., fructose and galactose) that are less easily metabolized to lactate. It has been assumed that reduction of the level of lactate in the culture medium would enhance both cell growth and antibody production.
However, the present invention demonstrates that the presence of lactate during fermentation can effectively increase antibody yield notwithstanding its inhibitory growth effec-ts. Using the methodology of the present invention, a sub-lethal concentration range (0 to 100 mM sodium lactate) was first deter-mined in order to study the lactate inhibition effect.
Various sub-lethal concentrations of sodium lactate are subsequently tested for the effect on produc-t - 14 13~2030 expression. For the cell line D-234, an effective concentration range for sodium lactate is 40 to 60mM.
Ammonia is another substance that has concerned cell culturists due to its negative effects on cell growth.
It is produced by cellular metabolism of amino acids as well as by spontaneous decomposition of glutamine. It has been assumed that reduction of ammonia in hybridoma cultures would banefit both cell growth and antibody production. However, as demonstrated herein, an increase in antibody titer was observed despite the inhibition of cell growth in the presence of ammonium chloride. For the cell line D-234, a preferred concentration range for ammonia chloride addition is 3 to 20mM, with 10-15 mM being more preferred.
The organic polyols useful in the invention include glycerol, polypropylene glycol and a variety of low molecular weight sugars including, for example, glucose mannose, fructose and mannitol. Of these organic polyols, glucose is preferred, and for the cell line D-234, an effective concentration range for glucose is 6 to 20 g/l, with 7 to 15 g/l being preferred. Another preferred organic polyol is polypropylene glycol. For the cell line 454A12, an effective concentration of polypropylene glycol is about 8~1/L.
Ths method of the invention is operable with any of a variety of well-known and/or commercially available mammalian cell culture media. Such suitable culture media includes serum-free media such as HL-l (Ventrex Labs, Portland, ME), HB104 (Hana Biologicals, Berkeley, CA), Iscove's DME medium (Gibco, Grand Island, NY) and RPMI-1640 medium (Gibco) or media supplemented with serum. The hybridomas used in the present method are preferably adapted for growth and maintenance in serum-free medium for large-scale, reproducible spinner culture production of monoclonal antibodies using, for example, a step-wise method.
The method of the invention has been shown to increase antibody titer regardless of the presence ~, ;, ~ 15 - 1 3 1 2030 or absence of serum in the medium. The cell lines used in the present invention may be cell lines of diverse mammalian origin. Rat, mouse and human embodiments are contemplated, with human embodiments illllstrated in the examples which follow. The antibodies may be of any class, including IgG
and IgM, with IgM and IgG types being specifically exemplified herein. The human embodiments are the products of triomas synthesized by somatic cell hybridization using a mouse x human parent hybrid cell line and Epstein-Barr virus (EBV)-transformed human peripheral blood lymphocytes (PBLs) or splenocytes from non-immunized volunteers or volunteers immunized with available Gram-negative bacterial vaccines or inactivated Gram-negative bacteria. Fresh PBLs or splenocytes (not transformed) may be used, if desired~ A
detailed description of the synthesis of the hybridomas, including the fusion protocol, ELISAs and hybrid screening procedure exemplified in the following examples is disclosed in Canadian Application Serial No. 578,020.
8riefly, the mouse-human heterohybrid fusion partner designated F3B6 was constructed by fusing human PBL B
cells obtained from a blood bank with the murine plasmacytoma cell line NSl obtained from the American ~ype Culture Collection (ATCC) under ATCC No. TIB18 (P3/NSl/1-AG4-1). The resulting hybrid cells were adapted for growth in 99%
serum-free medium and deposited with the ATCC under ATCC No.
HB-8785.
The heterohybrid F3B6 cells and positive EBV-transformed PBL B cells were then used to construct hybridoma cells lines which secrete antibodies illustrative for use in the method of the present invention. A preferred strategy for preparing and identifying such hybrids follows~
Cells (PBLs, splenocytes, etc.) are panned on cell-wall lipopolysaccharide (LPS) (an endotoxin of a gram-negative bacteria which produces bacteremia) coated tissue culture plates, the EBV transformed and fused to the tumor fusion partner (mouse myeloma x human s cell or rate myeloma).
Panning involves incubation of the popula-tion of immunocompetent cells on a plastic surface coated with the relevant antigen. Antigen-specific cells adhere.
Following removal of non-adherent cells, a population of cells specifically enriched for the antigen used is obtained. These cells are transformed by EBV and cultured at 103 cells per microtiter well using an irradiated lymphoblastoid feeder cell layer.
Superna-tants from the resulting lymphoblastoid cells are screened by ELISA against an E. coli Rc LPS and a Salmonella Re LPS. Cells that are positive for either Rc or Re lipid A LPS are expanded and fused to a 6-thioguanine-resistant mouse x human B cell fusion partner. If the mouse x human B cell fusion partner is used, hybrids are selected in ouabain and aza-serine. Supernatants from the Rc or Re positive hybrids are assayed by ELISA against a spectrum of Gram-negative bacteria and purified Gram-negative bac-terial LPSs. Cultures exhibiting a wide range of activity are chosen for in vivo LPS neutralizing activity. Many but not all antibodies so produced are of the IgM class and most demonstrate binding to a wide range of purified lipid A's or rough LPS's. The antibodies demonstrate binding to various smooth LPS's ~5 and to a range of clinical bacterial isolates by ELISA.
Two of the hybridoma cell lines which produce the Gram-negative bacteri.al endotoxin blocking antibodies described above were used to illustrate the methods of the present invention. D-234 and T-88 are representative of hybridomas used in the methods of the present invention to produce increased yields of their respective monoclonal antibodies. D-234 was adapted to growth and maintenance in serum-free medium for large-scale production of monoclonal antibodies.
The D-234 hybridoma was crea-ted from a fusion of the 1 3 1 L 0 3 (~

heterohybrid fusion partner F3B6 and human B lympho-cytes; a hybridoma sample adapted for growth in serum-free media was deposited with the ATCC under ac-cession number HB-8598. The T-88 hybridoma is a fusion product of the same heterohybrid F3B6 and human splenocytes from a lymphoma patient. ~ sample of this hybridoma (that was not adapted for growth in serum-free media) was deposited with the ATCC under acces-sion number Hs-9431. In addition, a subsequent hybridoma passage of D-234 was deposited with the ATCC under accession number HB-9543. These latter two hybridoma cell lines are specifically exemplified in the following examples.
The murine-murine hybridoma cell line, 454A12, used as an example here was formed from the fusion of a mouse splenocy~e and a mouse myeloma fusion cell partner.
This hybridoma produces IgG monoclonal antibodies specific for human transferrin receptor. The 454A12 hybridoma, its production, and the antibody it produced were described in U.S. patent application, Serial No. 069,867, "Anti-human Ovarian Cancer Immunotoxins and Methods of Use Thereof", filed July 6, 1987, applicants Bjorn, M.J. et al.

Examples The following examples are illustrative of this invention. They are not intended to be limiting upon the scope thereof.

Example 1 Culture of D-234 A one ml ampoule of frozen D-234 stock (ATCC
HB-9543) was thawed quickly in a 37 C water bath. The contents were aseptically added to 100 ml prewarmed, pregassed, serum-free HL-l medium (Ventrex Labs, Portland, Me) supplemented with 0.1% Pluronic-~ polyol S
i .

-17a- 1 31 2330 F-68 and 8 mM L-glutamine in a 250 ml Erlenmyer flas~
with a loosely fitted plastic screw cap. The flask was placed in a humidified incubator (36.5C, 90%
relative humidity and 5% CO2) and cultured with shaking at 100-120 rpm.
This parent culture was subcultured during mid-exponential phase, about 2-4 days after inocula-tion, when the cell density was approximately 5 x 105 to 1 x 106 viable cells per ml. The subcultures were grown in the daughter flasks under the same culture ~.
;

1312~30 conditions as above, starting with the initial inoculum of 1 x 105 and 5 x 104 viable cells/ml. The cells were counted using a Coulter Counter, and viability was determined by trypan blue exclusion using an hemocytometer. Maximum total cell densities were around 1.7 million with viable cell densities around 1 million.
For standard batch production, the cultures were allowed to grow to completion which occurs about 7 to 10 days irom planting by which time cell viability had declined to 30~ or less. The cells were harvested by centrifugation (3,000 rpm for 5 min) to separate the cells and purify the antibodies.
The resulting antibody yield was determined by enzyme-linked immunoadsorbent assay (ELISA) using a standard IgM ELISA but modified by using a high salt (i.e., at least 0.5 M NaCl) assay buffer. IgM titers were around 40 ug/ml.

Example 2 Effect of Salt Addition on IqM Production In D-234 The following treatments were set up in 100 ml working volume shake flasks at standard planting C~ densities in HL-1 with 0.1% Pluronic~ ~-68 and 8 mM
glutamine. ~ 3.75 M salt solution (27:1 molar ratio NaCl:KCl) was used to increase salt concentration beyond that of the standard HL-l medium.
Approximately 1 x 105 viable cells/ml were used to inoculate the aforementioned culture medium, which was used as the control sample. In addition, 1 x 105 viable cells/ml were inoculated into a ~00 mOsM
initial osmolari.ty medium. A third sample was formed by inoculating the standard osmolarity medium and, after 88 hours of culture, -the 3.75 M salt solution was added to a final concentration of ~00 mOsM. At this time point, the cell density was determined indi-~ ~rc~c~ rk cating that the culture contained ~ 1.2 x 106 vc/ml.The cells in each of the three cultures were cultured for 9 days, during which time the cell viability and cell density levels were monitored. The IgM titers were determined for each of the three experimental runs. The results of these experimental runs ~re pro-vided in Figure 1 and in Table 1 below. As indicated therein, a twofold increase in final IgM titers over the control (~90 mg/L) was correlated with prolonged viability and increased syecific IgM production rates in 400 mOsM cultures where growth rate and cell den-sity are reduced.

1 31 ~0~
--~o--TABLE I
D-234 Summary Table 300 mOsM 400 mOsM "Add Salt~
Control Inikial (at 88 Hours) Maximum Total 23 12 22 Cell Density (10 /ml) Maximum Viable 15 7.5 14 Cell Density (105/ml) Ave. Expo- 0.033 0.028 0.032 nential Growth (0-66 hr) (0-89 hr) (0-89 hr) Rate mu (1/hr) Final IgM 41 88 58 Concentration (mg/L) Ave. Exponen- 0.24 0.56 0.40 tial IgM Produc-tion Rate (mg/109/hr) For the 400 mOsM/kg initial culture, expo-nential growth rate "mu" and maximum cell density were reduced, which was indicative of solute s-tress. The duration of the culture was increased in the h~gh osmolarity culture, and the specific IgM productivity rate was twofold to thxeefold higher than the control.
The extra IgM over and above the control was produced after the peak in viable cell density.

13120~) A 1.5-fold increase in final IgM -titer to ~58 mg/L was observed in the culture where salt was added at 88 hours. Specific IgM production rates increased from one day after salt addition into the viable cell decline (versus the control, where produc-tion rate declined aftex the viable cell peak)l even though there appeared to be little, if any, difference in the growth curve compared to the control.
For the D-234 cell line, salt addition near the peak viable cell density has an IgM production enhancing effect in the decline phase without any extension of the viable cell curve. This suggests that specific IgM produc-tion rates can be increased without slowing growth (and limiting ultimate cell densities) early in culture. However, for D-234, final titers are not as high as those achieved in slow growing (limited cell density) cultures planted in high osmolarity medium.

Example 3 Effect of Inoculation Density and Timinq of Salt Addition Using the methods described in the foregoing examples, the effects of initial inoculation density of D-234 on the specific cell productivity and timing of the salt addition were explored.
A control was run at the standard osmolarity of 300 mOsM medium using 5 x 104 planted cultures.
These cells exhibited good growth, bu-t viable cell densities were lower than that produced for the 1 x cultures (and total cell density of 1.6 versus 1.9 million) with an extension of the viable phase from six to seven days. However, final IgM titers were similar. At 370 mOsM, 5 x 10 cells/ml inoculated cultures resulted in significant growth slowing and -22- 1312~
lowerlng of viable cell density and titers, about half compared with 1 x 105 planted cultures.
Various solute stress conditions were tested using the 5 x 104 inoculation density cul-ture. Titers for 300, 340 and 370 cultures were 40, 75, and 35 mg/
L, respectively. It was found that adding salt at day one instead of at day zero to the 370 mOsM allowed the 5 x 104 culture to reach viable cell densities (5 x 105 cells/ml) and a titer (65 mg/L IgM) approaching the 1 x 105, 370 culture values (6 x 105 cells/ml and 75 mg/L IgM).
From the results of the previous experimen-t, 3~0 and 370 mOsM were chosen as osmolarities to test with salt added on day 0, 1, 2, or 3. The results indicated that adding salt at different times to the 370 mOsM culture resulted in a slight increase (60 to 65 mg/L final IgM) in final titer concentration.
For the 340 mOsM culture, the addition of salt at day 1 and day 2 led to higher titers (~110 mg/
20 L) than did day 3 addition (~90 mg/L) or day 0 (~70 mg/L).

Ex~ple 4 Effect of Salt Addition on T-88 Growth and IqM Production T-88 cells were grown in replicate 100 ml working volume shake flasks of HL-1 media with 0.1~
w/v Pluronic~ polyol E'-68, 8 mM glutamine and 5~ added fetal calf serum at 300 mOsM (control); 340 mOsM;
400 mOsM; and 450 mOsM. Like the above examples, osmolality was increased by the addi-tion of a 3.75 M
sal-t solution with a 27:1 molar ratio NaCl:KCl. The cultures were grown for 7 days, during which -time the cell density and cel:L viability were periodically monitored.

-23- l 3 1 ~0~() Comple-te growth curves were generated for the control and for the 400 mOsM flasks. The 400 mOsM
growth curve showed slow growth and reduced cell den-sity, therefore indicating solute stress had occurred.
The duration of the culture was extended, during which IgM production over and above the control was ob-tained. The specific IgM production rate was higher at 400 mOsM over most of the culture period. Table 2 shown below, illustrates that a 30~ reduction in total cell density and a 20 to 25% increase in final IgM
titer for the 400 and 450 mOsM shake flasks was achieved. IgM produced per million cells from day three to day four was about -two times higher at 400 and 450 mOsM compared with the control and 340 mOsM
treatment. Exponential phase doubling time (Td) for the 400 mOsM treated flasks was higher than for the control (27 versus 20 hours).

Table 2 T-88 ~ 5~ FCS Summary Table Control mOsM mOsM mOsM mOsM
Maximum Total 23 24 17 17 Cell Density (105/ml) Final IgM 37 35 43 46 Concentra-tion (mg/L) IgM Produced 6 7 15 11 per Million Cells From Day 3 to Day 4 (ug/105 cells/
day) Ave. Exponen- 0.037 0.034 tial Growth (Td 20) (Td 27 Rate mu (1/hr) -24- l 31~030 Example 5 Effect of Lactate on D-234 Grow~h and IgM Production This example describes the effect of sodium lactate on growth, viability, and IgM production of D-234.
Approximately 1 x 10 cells/ml of D-234 were grown in 250 ml shake flasks (agitated at 100 rpm) in HL-1 medium containin~ 0.1~ Pluronic~ polyol F-68 and B mM glutamine. A l M stock solution of sodium lac-tate (pH 7.4) in HL-1 was added to the medium. A pre-liminary screen of -the effect of a broad range of sodium lactate concentrations (0-100 mM) on D-234 growth and IgM production was run. It was determined that growth was greatly inhibited by levels of added lactate above 40 mM. Cell densities at day four were reduced at all levels of lactate tested with critical drop between 40 and 60 mM.
The results of this experiment are given in Table 3 below.

Table 3 Effect of Na Lactate on D-234 Growth and IqM Production Initial Lactate Total Cell Density IgM
mM 1 x 105/ml (% Viability) uq/ml Day 2 Day 4 Day 4 ~ 7 0 4.7 (95)21.0 (90) 10 24 5.6 (96)15.0 (92) 20 35 5.1 (92)12.9 (89) 22 54 ~0 2.2 (93)4.1 (87) 19 61 2.1 (81)2~6 (65) 15 28 100 2.2 (72)2.0 (50) 11 1~

-25 13~2030 The results indicate that the production of IgM by D-234 was increased with increasing concentr~-tions of sodium lactate up to 60 mM where growth was extremely inhibited, and IgM production peaked at 61 ug/ml compared to the control at 24 ug/ml. Even at 80 mM added lactate, the level of IgM produced was similar to that seen for the control, even though the cell density was only 12% of the control. Specific (per cell) productivity was increased up to 14-fold (at 60 mM added lactate).

Example 6 Effect of NH4Cl on D-234 Growth and IqM Production The hybridoma D-234 was grown in HL-1 serum-free medium supplemented with 0.1% Pluronic:~ polyol F-6~, 10 mM glutamine and 10 mM NH4Cl. A control wasalso run without NH4Cl. One hundred ml cultures in 250 ml shake flasks were inoculated at an initial den-sity of 1 x 105 viable cells/ml (91% viability).
As illustrated in Figure 2, the addition of mM NH4Cl inhibited the growth, reduced both viability and the maximum total cell density of the culture (2.3 x 106/ml for the control vs 1.1 x 106/ml when 10 mM NH4Cl was added). However, this stress condition prolonged the stationary/decline phase and resulted in a 2-fold increase in the production of IgM.

Example 7 Effect of Hiqh Glucose Concentration on Antibody Production - 30 ~ The hybridoma D-234 was grown in HL-l medium (Ventrex) which already contains 5.5 g/l. A 500 g/l ` stock solution of glucose was used -to increase the glucose level of the HL-l medium. The -total glucose ~ `~rCyG/l~ rnci~k 26 ~ 1 31 2 030 levels tested in this example were 5.5 (control), 10.5, 1505, and 25.5 g/l.
The 10.5 g/l glucose culture grew more slowly than the control and began to die sooner. While the control reached a maximum of 8.7 x 105 viable cells/ml, the 10.5 g/l stressed culture reached 7.1 x 105 viable cells/ml.
However, the death phase of this culture was longer than the control resulting in higher antibody production: 85 versus 67 mg/l.
The 15.5 g/l glucose culture proved to be very stressful for D-234 resultiny in a low maximum viable cell density (4.3 x 105 viable cells/ml) and producing IgM at 50 mg/l. The 25.5 g/l glucose condition proved to be lethal.

Deposition~of Cultures The hybridomas used in the above examples to illustrate the method of the present invention were deposited in and accepted by the American Type Culture Collection (ATCC), 12301 Parklawn Drive, Rockville, Maryland, USA, under the terms of the Budapest Treaty. In addition, the mouse x human fusion partner F3B6 adapted to 99% serum-free medium which partner was the source of these hybridomas was similarly deposited with the ATCC. The deposit dates and the accession numhers are given below:

Culture Deposit Date Accession No.

D-234 10 August 1984 HB-8598 D-234 17 September 1987 HB-9543 T-88 19 May 1987 HB-9431 F3B6 18 April 1985 HB-8785 The deposits above were made pursuant to a contract between the ATCC and the assignee of this . ` ',;t `: 'i ~...... ..

~ 31 2030 patent application, Cetus Corporation. The contract with ATCC provides for permanent availability of the progeny of these cell lines to the public on the issu-ance of the U.S. patent describing and identifying the deposit or the publications or upon the laying open to the public of any U.S. or for~ign patent application, whichever comes first, and for availability of the progeny of these cell lines to one determined by th~
U.S. Commissioner of Paten-ts and Trademarks to be entitled thereto according to 35 USC 122 and the Commissioner's rules pursuant thereto (including 37 CFR 1.14 with particular reference to 886 OG 638).
The assignee of the present application has agreed that if the cell lines on deposit should die or be lost or destroyed when cultivated under suitable con-ditions, they will be promptly replaced on notifica-tion with a viable culture of the same cell line.

Samples o~ the 454A12 hybridomas had been deposited with In Vitro International Inc., (formerly at 7885 JacXson Road, Suite 4, Ann Arbor, Michigan 48103, U.S.A., currently at 611 P. Hammonds ~erry Road, Linthicum, Maryland 21090, U.S.A., telephone number ~301)789-3636) on June 18, 1985, under the Accession No.
IVI10075. This deposit was made under the Budapest Treaty and will be maintained and made accessible according to the provisions thereof.

28 - l 3l 2 03 n SUPP~EMENTARY DISCLOSURE

To the Principal Disclosure is added Fig. 3 of the drawings and Examples 8 and 9.

In the drawings:
Figure 3 shows the effect of sodium chloride on specific production rate of IgG antibody hybridoma ~54A12.

Exarnple 8 ~ffect of PolyPropylene Glycol on IqG Production The following experiment showed that when polypropylene glycol (PPG) was added to hybridoma 454Al2 cell culture, it increased the IyG production of khe hybridoma by 40%. Though PPG limited the maximum cell density achievable by the culture, it slowed the decline in culture viability a~ter the peak density had been attained.
The hybridoma 454A12 was grown in 125 ml shake flasks filled to 50 ml with HL~l and 4mM glutamine. The test sample contained 8~1/L of polypropylene glycol whereas the control was without the PPG.
It was observed that the test sample exhibited an exponential phase growth rate similar to the control at 0.054 hour l. However, the test sample experienced a lag in growth of one day, and a higher exponential phase death rate of 0.0059 hr~l. Additionally, the test sample had a lower maximum cell density than the control. The test sample reached a maximum cell density of only 1.3 million cells/ml, whereas the control reached a maximum cell density of 2 million cells/ml.
Beyond the maximum cell density peak, the decline in cell viability in the test sample was slower than in the control. The test sample y:ielded a final IgG concentration - 29 - l 3 1 203Q

of 63 ~g/ml, which was about 40~ higher than the control, which ~ielded 46 ~g/ml of IgG.

Example 9 Effect of Sodium Chloride on I~G Production In another experiment, the 454A12 hybridomas were grown in shake flasks with commercially available HL~1 medium (Ventrex) supplemented with 8mM glutamine.
Osmolality of the standard (control) HL-l medium was 300 mOsmol/kg. In the test sample, the osmolality was increased to 400 mOsmol/kg using sodium chloride. The result of solute stress in the test sample was evidenced by a 50 decrease in maximum cell density. In the stressed condition, the specific IgG production rate per cell increased throughout the culture period. ~he increase in specific productivity was greatest during the post exponential phase period of the culture when specific productivity was more than 60~ higher under the stressed condition (Figure 3).

Claims

1. A method for increasing expression of a protein in a mammalian cell culture, which already provides for all cell growth requirements, and recovering the protein from the culture, where the expression is increased above the expression level at optimal growth, comprising adding a solute to the cell medium at a level above that for optimal cell growth to create stress on the cell, as expressed by an inhibitory effect on cell growth or cell density; and recovering and purifying the protein from the cell culture.

2. A method in accordance with claim 1, wherein the solute is selected from the group consisting of sodium chloride, potassium chloride, calcium chloride, magnesium chloride, sodium lactate, ammonia, glycerol, polypropylene glycol, glucose, mannose, fructose, mannitol, or mixtures thereof.

3. A method in accordance with claim 2, wherein the protein is produced by a hybridoma from the group consisting of D-234(ATCCHB-8598), D-234(ATCCHB-9543), T-88(ATCCHB-9431),

4. A method in accordance with claim 3, wherein the mammalian cell culture is composed of D-234 cells and the osmolality of the medium with the addition of sodium chloride is in the range of 350 to 400 mOsmol/kg.

5. A method in accordance with claim 3, wherein the mammalian cell culture is composed of T-88 cells and the osmolality of the medium with the addition of sodium chloride is in the range of 400 to 450 mOsmol/kg.

6. A method in accordance with claim 2, wherein lactate is added as sodium lactate and the sodium lactate concentration is in the range of 10 to 100 mM.

7. A method in accordance with claim 2, wherein the sodium lactate concentration is in the range of 40 to 60 mM.

8. A method in accordance with claim 2, wherein ammonia is added as ammonium chloride and the ammonium chloride concentration is in the range of 3 to 20 mM.

9. A method in accordance with claim 8, wherein the ammonium chloride concentration is in the range of 10-15 mM.

10. A method in accordance with claim 2, wherein the polypropylene glycol concentration is about 8 µl/L.

11. A method in accordance with claim 2, wherein the glucose concentration is in the range of 7-15 g/L.

12. A method to determine the solute level in a cell culture medium to produce the highest protein product expression from a cell, where the culture medium has all the requirements necessary for optimal growth comprising:
determining the concentrations of solutes necessary for optimal cell growth; increasing the concentrations of one or more solutes to place the cell under stress; and determining the concentrations which produce more protein product.

CLAIMS SUPPORTED BY THE SUPPLEMENTARY DISCLOSURE

13. The method in accordance with claim 2 wherein the protein is produced by a hybridoma consisting of 454Al2 (IVI 10075).

14. The method in accordance with claim 1, wherein the mammalian cell culture is comprised of 454A12 hybridomas, and the osmolality of the medium with the addition of sodium chloride is about 400 mOsmol/kg.