WO1994018310A1 - Growth enhancing media supplement for the culture of mammalian cells - Google Patents

Abstract

Methods are provided for the production of a growth enhancing media supplement (GEMS) from Cohn fraction IV4 derived from human serum. Methods are also provided for the culture of a diverse array of mammalian cells in GEMS, which effectively replaces fetal bovine serum in these cultures.

Description

GROWTH ENHANCING MEDIA SUPPLEMENT FOR THE CULTURE OF MAMMALIAN CELLS

Technical Field of the Invention This invention relates to a method for preparing from human plasma Fraction IV₄ a growth-enhancing media supplement for the culture of mammalian cells. The invention also relates to the product of this method, that is a growth-enhancing media supplement, methods for culturing cells in this supplement, and the cell cultures resulting from these methods.

Background

Cell growth in culture requires that (1) the cells stay healthy, and (2) the cells are able to go through mitosis and proliferate.

Mammalian cells are generally considered challenging to grow under the essentially artificial conditions of Ln vitro culture. The absolute requirements for growth have not been defined for most mammalian cell types, but it is known that cells require a basal medium containing salts, glucose, amino acids, minerals, and vitamins, plus a supplement containing biological factors such as transferrin, insulin, and other growth factors. Different cell types have different requirements for survival and optimal growth.

Currently, the supplementing biological factors are most often derived from an animal source such as bovine serum. Certain cell types will grow in media supplemented only with the albumin fraction derived from serum. However, many cell types require other factors in serum that have not yet been identified, but are known to be necessary for the cells to thrive in an artificial medium. Certain cell types require factors that are not present in adult serum but only in fetal serum. This situation has led to the extensive use of fetal bovine serum (FBS) to supplement the media used to support the growth of these cells. However, the use of fetal serum has several serious drawbacks, including high cost and pending government regulations which may further decrease or eliminate the availability of fetal serum (Hodgson, J. 1993 Bio\Technology 11:49-53) . Due to the conditions surrounding collection of fetal calf blood, and the fact that only a relatively small number of fetal calves are available, there is considerable batch to batch variation in FBS. Certain batches of FBS may even be toxic to cells due to contamination and the development of endotoxins during handling. Local feeding of the cows can also influence the content of FBS collected at a given site, and it is considered impractical to attempt to pool fetal calf blood from diverse geographical locations. Moreover, the products obtained from cells grown in FBS might be contaminated by bovine proteins and/or bovine viruses which could preclude their use for human therapy.

To address the problems of high cost, unavailability, batch to batch variation, and foreign proteins associated with fetal animal serum, researchers have proposed its replacement in whole or in part by various fractions obtained from human serum. Human serum is routinely collected, using sterile techniques, from volunteers at multiple plasma centers in diverse geographical locations. The collected human serum is then pooled and processed to obtain valuable therapeutic fractions containing, for instance, the clotting factor VIII or human immunoglobulins. The technique for fractionating human serum by succeeding rounds of ethanol precipitation combined with pH changes was first described by Cohn, E.J. , et al., (1946) J. Am. Che . Soc. 68:465, and fractions of human serum are still referred to by the "Cohn fraction" number corresponding to this original description. In the course of fractionating human serum to obtain a specified product, there are certain fractions which have traditionally been discarded as "waste fractions" simply because they did not contain a therapeutic amount of the specific product sought.

Nominative "waste fractions" of human serum have been proposed as sources of media supplements to replace FBS. MacLeod, A.J. , EP 0143648, disclosed the use of a fraction obtainable from Cohn fraction II + in combined, or Cohn fraction V, under stringent preparation conditions, which could support the growth of mouse-mouse and rat-rat hybridoma cell lines, virus-transformed human peripheral lymphocytes, and Namalva and HEP-2 human tumor lines.

Stemerman, M.B., et al., U.S. Patent No 4,443,546, disclosed a method to obtain somatomedins by heating Cohn fraction IV to 100°C in the presence of acetic acid. The resulting somatomedin-containing fraction was combined with various growth factors to provide a supplement for the culture of endothelial cells and smooth muscle cells.

MacLeod, A.J. , EP 440 509, disclosed a supplement produced from a Cohn fraction IV or Cohn fraction II and III combined, in which immunoglobulins were removed by polyethylene glycol precipitation.

Uthne, K.O., et al, US Patent No 3,953,290, disclosed various cell growth factors, having molecular weights of 4,000 to 6,000 Daltons, isolated from various serum fractions, including Cohn fraction IV.

Ng, P.K., et al., US Patent No 4,452,893, disclosed a supplement obtained from Fraction IV which is substantially free of components having a molecular weight higher than 250 kDa. These removed components were considered to be inhibitory to cell growth.

MacLeod, A.J. , (1988) Biotechnology 37:41-56 is a review article which described a method to produce from fraction IV₄ a supplement to support the growth of certain types of hybridomas.

MacLeod, A.J. , et al., (1985) Develop. Biol. Standard 60:55-61, is a review article which disclosed the use of fraction IV₁ as a supplement to support the growth of certain types of hybridomas as well as Namalva and HEP-2 cells.

Polet, H., et al., (1975) J. Ex . Med. 142:949-959, disclosed the use of fraction V to grow activated human lymphocytes.

Glasser, L. , et al. , (1985) Blood 66:267, addressing the issue of storage variables on human neutrophils, disclosed that Cohn fractions IV₁ and IV_^- were not effective as preservatives.

Houlgatte, R. , et al. , (1988) BBRC 150:723-730, disclosed that Cohn fraction IV promoted the synthesis of nerve growth factor in fibroblast-like L cells.

Pickart, L. (1980) J. Supramolecular Structure 13:385-394 disclosed a peptide extracted from Cohn fraction IV₁ having a molecular weight range of 300-1000 Da, which stimulated chick chondrocyte metabolism.

Gowan, L.K., et al., (1987) Endocrinology 121:449-458, disclosed a form of insulin-like growth factor II with a molecular weight of 15,000 isolated from Cohn fraction IV.,_ . MacLeod, A.J. et al . , ( 1987 ) Deve lop . Biol. Standard.

66:561, disclosed Cohn fraction IV., as a source for a medium supplement.

Antoniades, H.N., et al., EP 264 748, disclosed supplement derived from Cohn fraction V.

Congote, L.F. (1987) In Vitro Cellular & Dev. Biol. 23:361, described the extraction of an erythropoietin-like factor from Cohn fraction V.

Jo, E-C, et al., (1991) Biotechnol. and Bioengineer. 38:247-253, disclosed the culture of melanoma cells supplemented by fractions IV_A plus V.

Polet, H. et al., (1975) J. Immunol. 117:1276-1281, disclosed that plasma proteins derived from fractions III and IV bind and remove free Con A, thus reducing the proliferation of lymphocytes.

Gorin, E. , et al, (1982) Biochem, J. 204:221-227, disclosed that Cohn fraction IV contains a 50,000 kDa lipase that is heat-labile and that can elevate lipid accumulation in cultured fibroblasts.

Fenje, P, US Patent No 3,769,415, disclosed a method for the production of a killed rabies virus vaccine containing protein from fraction V.

Posner, B.I., (1978) J. Clin. Endocrinol, Metab. 47:

1240-1250, disclosed the extraction of insulin-like activity from Cohn fraction IV .

Thomou, H. , et al., (1980) J. Cell. Sci. 44:285-297, disclosed the extraction from Cohn fraction III of mitogenic factors for human lymphocytes and mouse fibroblasts.

Teschner, R. , et al., (1980) Z. Naturforsch. 35:117-123, disclosed the supplementation of culture medium with Cohn fraction V.

Two abstracts by MacLeod, A.J. , et al, described the use of Cohn fraction IV₁ and proteins derived therefrom to culture cells: MacLeod, A.J. (1987) Develop. biol. Standard 66:562; and MacLeod, A.J. , et al., (1987) Develop, biol. Standard 66:561.

Chen, D-M. et al., (1977) Immuno1. Commun. 6:395-410, reported that Cohn fraction III and IV increase concanavalin-induced proliferation of murine spleen lymphocytes.

EP 415 666 (Sasai, S. , et al) disclose the use of a media supplement derived from human serum by a salting-out process to culture lymphokine-activated killer (LAK) cells.

In addition to the issue of support of cell growth, there is a general concern for the possible presence in human serum of active viruses pathological to humans. The use of pasteurization to inactivate any viruses remaining after Cohn fractionation IV₁ was suggested by MacLeod, A.J. in EP 440 509. EP 415 666 disclosed viral inactivation by treatment of the supplement with chemicals such as ethylene oxide or glutaraldehyde.

Taken together, the above described publications have created the impression among those of skill in the art that the derivation of growth-enhancing media supplements from human serum fractions would entail rather complex methods applied in a trial and error fashion, with no guarantee of success for any given mammalian cell type.

What is needed is a simple, economically practical method to produce from human serum a media supplement suitable for the culture of diverse types of mammalian cells, including hybridomas, mammalian cell lines, hematopoietic cells, and primary cells derived from normal tissue as well as tumors.

Brief Description of the Figures

Figure 1 depicts schematically the preparation of the growth-enhancing media supplement (GEMS) .

Figure 2 shows the molecular weights of proteins in pasteurized GEMS detected by HPLC.

Figure 3 shows the molecular weights of proteins, as detected by HPLC, in GEMS filtered through a virus-removing membrane.

Figure 4 depicts the growth of hybridoma cell line 9199 in pasteurized GEMS (PAST-G and PAST-V) and GEMS filtered for virus removal (ASAHI-G and ASAHI-V) compared with FBS (FBS- G and FBS-V) .

Figure 5 depicts the growth of hybridoma cell line 9105 in GEMS compared with FBS.

Figure 6 compares the growth of MB231 cells in 10% FBS with their growth in 0.5% FBS supplemented with GEMS.

Figure 7 compares the growth of peripheral blood lymphocytes (PBL) in various media supplements. Figure 8 depicts the support of lytic activity in lymphokine-activated killer cells (LAK's) via culture in 2% GEMS compared with AIM V and 2% AB serum.

Figure 9 shows the proliferation of tumor-infiltrating lymphocytes in GEMS.

Figure 10 depicts the growth of primary rabbit kidney cells (PRK's) in media supplemented with GEMS.

Figure 11 is a tabular depiction of the growth of hematopoietic progenitor CD34+ cells in media variously supplemented with lipids, GEMS (pasteurized or filtered) , or 25% FBS/HS.

Summary and Description of the Invention

The invention is directed to a method for producing a human-derived growth-enhancing media supplement (GEMS) from human serum Cohn fraction IV comprising resuspending the

Cohn fraction IV₄ paste in a resuspension buffer, pH 7.4 to 8.4, stirring and readjusting the pH of the suspension to achieve a stable pH between 6.4-8.0, clarifying the suspension and filtering the supernatant through a sterilizing filter.

The invention is further directed to methods for inactivating or removing viruses from the GEMS by either pasteurization or filtration through a virus-removing membrane, or both.

The invention is also directed to the GEMS resulting from the above processes. The invention is further directed to methods for culturing mammalian cells in GEMS and the cell cultures resulting from these methods.

The invention method for producing GEMS begins with a starting material designated "Cohn fraction IV₄ paste" derived from human serum. While it is possible to make the Cohn fraction IV₄ paste from serum obtained from only one or a few donors, it is much preferred to make the paste from a pool of many donors, most preferably more than 3000 donors. The large pool provides the advantage of averaging out any batch to batch variations which might otherwise occur among batches obtained from only a few donors. Thus the user of GEMS is spared the time and cost of trying several batches of media supplement in order to find the one that works for a given cell. Typically, a scientist who is forced to use FBS for a long-term project must purchase an entire lot of FBS in order to be assured of a supply, and even then the supply will inevitably run out in the long-term. Another advantage of GEMS over FBS is the fact that human blood is obtained, handled, and stored under optimal conditions that preclude contamination, degradation, and the formation of endotoxins.

Herein, the term "Cohn fraction IV₄" refers to a paste obtained essentially as follows: Plasma from one or more human donors is collected, frozen, and then partially thawed at a controlled temperature not exceeding 6°C forming a cryo-suspension, pooled, and centrifuged at about 5000 g until all the cryo-precipitate has been separated from the cryo-suspension. The resulting supernate, the cryo-poor plasma, is collected in a temperature-controlled tank and cooled to about -5°C. The pH is adjusted to about 6.9 and the ethanol concentration adjusted to about 20% v/v by the addition of a pH 4.0 buffer/ethanol solution prechilled at about -15°C. All subsequent precipitation steps with ethanol and centrifugation are done at about -5°C. The resulting suspension is mixed to achieve complete precipitation and then centrifuged at about 5000 g. The centrifugate (supernate, Fraction I+II+III) is collected. The Fraction I+II+III centrifugate (supernate) is adjusted to a pH of about 5.2 while maintaining the same alcohol cencentration with a prechilled pH 4.0 buffer/ethanol solution. The resulting suspension is mixed to achieve complete precipitation and centrifuged at about 5000 g. The centrifugate (supernate, Fraction IV-l) is adjusted to about 140 meq/L by the simultaneous addition of a pH 5.95 buffer and IN sodium bicarbonate solution. Ethanol is then added to a final concentration of about 40% v/v. The suspension is mixed to achieve complete precipitation and centrifuged at about 5000g. The resulting precipitate, Fraction IV₄ paste, is then collected.

The Cohn fraction IV₄ paste is then suspended in a buffer having a pH of about 7.4 to about 8.4, most preferably 7.8, stirred to form a homogeneous suspension, and the pH is then adjusted to achieve a stable pH between 6.4 to 8.0, most preferably 7.3. Herein, the term "stable pH" refers to a pH value that does not change by more than about 0.1 pH points when the suspension is stirred. The suspension is then clarified, and the supernatant is collected. Herein, the term "clarification" refers to any process which physically separates a precipitating solid phase from a liquid in which the solid is suspended. Clarification may be achieved via centrifugation, filtration, or decantation after a period of time sufficient to allow the solids to settle. After clarification, the supernatant is filtered through a sterilizing filter to remove microorganisms. Herein, the term "sterilizing filter" refers to a membrane device with pore sizes averaging 0.2 μm in diameter which will allow the passage of large proteins but which will detain bacteria and any larger microorganisms. The GEMS is now suitable to be added to basal medium to promote the in vitro growth of mammalian cells.

For certain therapeutic and diagnostic uses of GEMS, it is desirable to remove or inactivate any contaminating viruses which might be present in the Cohn fraction IV₄ starting material.

For virus removal, the GEMS may be filtered through various types of virus-removing membrane filters which have pore sizes small enough to retain viruses but large enough to allow proteins to flow through. For instance, a typical virus-removing system has, in sequence, a 75nm pore membrane and one or two 35nm pore membranes.

For virus inactivation, it is preferred to pasteurize the GEMS since pasteurization, i.e. heating to 60°C for 10 hours, has been shown to kill all known viruses. Prior to pasteurization, however, it is preferred to stabilize the biological growth components via the addition of various combinations of sucrose, sorbitol, glycine, and ascorbate. This stabilization procedure ensures that unidentified biological factors present in GEMS will survive the heat treatment (pasteurization) without forming complexes with other factors, becoming degraded, or otherwise being inactivated.

The stabilized, pasteurized, GEMS is then subjected to a procedure to remove the stabilizers in order not to compromise the growth of the cells. The various methods used for removal of stabilizers are dialysis, diafiltration, gel filtration, and size-exclusion chromotography. Herein, the term "dialysis" refers to a process whereby the GEMS containing stabilizer is suspenαed in a bag having a semipermeable membrane with pore sizes of approximately 1000 Daltons which allow the stabilizer molecules to be drawn by osmosis out of the GEMS, through the pores, and into a surrounding buffer solution having a lower osmolarity. Higher molecular weight molecules are retained within the bag. The surrounding buffer solution is replaced at intervals until the stabilizers have been sufficiently removed from the GEMS in the suspended dialysis bag. However, a preferred method for removing stabilizers involves "diafiltration" , also known as "molecular wash at constant volume" . Prior to diafiltration, the volume of the GEMS may be reduced via ultrafiltration in order to conserve the amount of buffer subsequently required for diafiltration. A third alternative for the removal of stabilizers is "gel filtration" using a desalting gel. These procedures are described in Example 1 below.

It may be advantageous to remove trace amounts of human IgG from the GEMS when cultured cells secrete a desired human IgG product into the medium supplemented with GEMS. In this case, the purification of the desired human IgG product is facilitated when there is no extraneous human IgG originating from the GEMS in the medium. Prior to adding GEMS to the basal medium, the GEMS is subjected to a treatment with Protein A or Protein G, both of which selectively bind IgG. This treatment may be conducted by column adsorption or batch adsorption. As the GEMS is contacted with Protein A or Protein G, the trace amounts of IgG are bound and separated from the remaining GEMS components, and the IgG-free GEMS is collected. Other affinity matrices known to adsorb IgG or other classes of Ig molecules may also be used, depending on which class of Ig molecules one desires to remove.

The GEMS products of the processes described above are useful in the culture of diverse mammalian cell types. For certain cell types, GEMS alone is sufficient to supplement a basal medium, completely replacing the fetal bovine serum which would otherwise be required. As described above, a human-derived supplement such as GEMS may be considered highly advantageous over fetal bovine serum for applications involving human therapy because non-human proteins are not present in GEMS. For other cell types, GEMS can replace up to 95% of the otherwise required fetal bovine serum (FBS) , leading to a great reduction in cost and an increase in availability and convenience.

Certain hybridomas are particularly challenging or impossible to grow in medium without FBS. Several of these hybridomas have been shown to thrive and produce their desired antibody product in medium supplemented only with GEMS (Examples 3 and 4 below) . The growth of hybridomas in the GEMS of the instant invention was reported in J.Tissue Culture Methods 14:39-44, February 13, 1992. This report did not disclose the method to produce the GEMS of the instant invention.

Cells of certain mammalian cell lines have been thought to absolutely require FBS for growth in vitro. Some of these problematical cell lines require attachment to a surface in order to proliferate, and they were previously found not to attach without FBS supplement to their medium. Surprisingly, it has been found that adherent CHOK₁ cells will grow in medium supplemented only with the GEMS of the instant invention when the surfaces of their flasks are pretreated with collagen, fetuin, or 10% FBS followed by a wash. Another unexpected result is that adherent CHO (Chinese hamster ovary) cells and C-127 (murine mammary epithelial) cells transfected with vectors encoding recombinant human Factor VIII will grow and express active Factor VIII when grown in media supplemented with GEMS alone, without pretreatment of their flask surfaces. In fact, the transfected C-127 cells were found to perform even better in GEMS than in 10% FBS. Likewise, GEMS may be used in place of FBS in the culture of adherent cell lines such as MB157, MB231, and Vero as well as chondrocytes, epithelial cells, human diploid strains, monkey kidney cells, canine kidney cells, fibroblasts, and certain hematopoietic cells that usually require a stro al layer for growth.

GEMS may also be used in place of FBS to culture cell lines which do not require attachment, but rather grow in suspension. Examples of non-adherent cell lines which may be cultured in GEMS include K562 cells, H69 cells, certain types of CHO cells, BHK-21 cells, HELA cells, baby hamster kidney cells (BHK) , human umbilical vein endothelial cells (HUVEC) , HEP-G2 cells, U0937 cells, human osteosarcoma cells (OS2) , and NIH 3T3 cells.

Peripheral blood lymphocytes (PBL's) are particularly challenging to grow in culture and have traditionally been thought to require FBS or whole human serum for proliferation. It has been found that 2% GEMS of the instant invention will support the proliferation of PBL's in culture to a level comparable to that afforded by 10% FBS.

PBL's may be activated by contact with IL-2 and/or a combination of anti-CD3 and anti-CD28 antibodies to become lymphokine-activated killer cells (LAK's) which can lyse certain target cells. The support of lytic activity of LAK's has likewise been thought to require FBS or whole human serum. Surprisingly, it was found that GEMS can replace FBS or whole human serum to support the proliferation and activation of LAK's. Tumor infiltrating lymphocytes (TIL's) have also been considered challenging to grow in culture, usually requiring fetal serum for propagation and lytic activity. GEMS was found to support the growth and lytic activity of TIL's derived from breast tumors or melanoma's. GEMS may also support the growth and lytic activity of TIL's derived from many different types of solid tumors, allowing the genetic manipulation of TIL's in vitro for therapeutic and/or diagnostic purposes.

Hematopoietic progenitor cells are selected from bone marrow or peripheral blood samples on the basis of antibody binding to cell-surface antigen CD34+. Previously, it was shown that optimal growth of CD34+ cells in culture was obtained by supplementation of the culture medium with 12.5% FBS plus 12.5% horse serum (25% FBS/HS) . Surprisingly, it was found that filtered (i.e. non- pasteurized) GEMS at 5 mg/ml or 0.2 mg/ml supports the growth of CD34+ cells to a level comparable to that of 25% FBS/HS. Pasteurized GEMS also supports the growth of bone marrow-derived CD34+ cells to an extent comparable to that of 25% FBS/HS. Pasteurized GEMS supports the growth of peripheral blood derived CD34+ cells at approximately one- third the rate of growth stimulated by 25% FBS/HS. A lower growth rate may be acceptable when all the advantages of GEMS are considered.

Cells derived from solid primary tissue, including both normal and tumor tissues, are considered particularly difficult to grow in culture because they generally require attachment to substrate as well as a diverse and undefined array of biological growth factors. Fortuitously, it was discovered that the GEMS of the instant invention supports the growth of cells derived from primary tissue, notably primary rabbit kidney tissue. This finding suggests the use of GEMS to expand ex vivo cells from an individual patient's tumor for diagnostic and/or therapeutic purposes.

In summary, the GEMS of the instant invention serves as a replacement for animal serum in general, and fetal serum in particular, when cells are grown in culture for a variety of medical purposes.

GEMS replaces FBS for the culture of certain hybridomas that are difficult or impossible to grow in serum-free media formulations. GEMS may thus make monoclonal antibody production from hybridomas more economical as well as more acceptable to regulatory agencies because of the absence of non-human proteins in the monoclonal antibody product.

Likewise, GEMS replaces FBS for the culture of established cell lines that were formerly thought to require FBS. GEMS is particularly valuable for the culture of transformed cell lines expressing recombinant protein products. It is expected that recombinant protein products will be more easily purified from GEMS supplemented media as compared with FBS-supplemented media due to the absence of non-human proteins in GEMS.

GEMS supports the in vitro growth and activity of cells of the immune system, including PBL'S, LAK'S, TIL'S, hematopoietic stem cells and hematopoietic progenitors. Thus GEMS can be used to generate large numbers of immune- system cells for the purposes of adoptive immunotherapy whereby autologous or allogeneic cells are generated in culture for replacement to the patient when needed.

The generation of large numbers of immune-system cells in culture also provides the opportunity to genetically alter the cells so that they produce therapeutic proteins when replaced to the patient. Lymphocytes generated in culture may also be genetically altered to express on their surfaces antigen-recognition proteins such that LAK's or TIL's will be targeted to pathological cells in the patient for therapy or diagnosis. For the culture of cells requiring attachment, GEMS can be used in conjunction with individual extracellular matrix components such as collagen and fetuin or with complete extracellular matrices such as those derived from bovine cornea or human embryonic cells.

GEMS can potentially be used in primary culture to amplify the number of cells isolated from a human solid tumor, thereby allowing the development of therapy and diagnostic tests based on an individual's unique tumor cell type. Given a sufficient quantity of the patient's tumor cells, antibodies may be raised specifically against those tumor cells. The specific antibodies may be conjugated with a diagnostic label or a toxin to form a "silver bullet" directed against the individual patient's tumor tissues and metastatic cells.

The following experimental examples are intended for enablement of the practice of the invention and for illustration of the principles of the invention. It is assumed that, given these examples, one of ordinary skill in the art of the invention will be able to devise alternative ways to practice the invention, and no limitation of the invention is implied by the presentation of these experimental examples.

EXAMPLE 1 Preparation of Pasteurized Growth-Enhancing Media Supplement (GEMS)

Cohn Fraction IV₄ paste was obtained essentially as described in Cohn, E.J., et al., (1946) J. Am. Chem. Soc. 68:465. Briefly, frozen plasma from several donors was collected from plasma centers across the U.S. The frozen plasma was partially thawed at a controlled temperature not exceeding 6°C forming a cryo-suspension, pooled, and centrifuged at about 5000g until all the cryo-precipitate had been separated from the cryo-suspension. The resulting supernate, the cryo-poor plasma, was collected in a temperature-controlled tank and cooled to about -5°C. The pH was adjusted to about 6.9 and the ethanol concentration adjusted to about 20% v/v by the addition of a pH 4.0 buffer/ethanol solution prechilled at about -15°C. All precipitation steps with ethanol and subsequent centrifugation were done at about -5°C. The resulting suspension was mixed to achieve complete precipitation and then centrifuged at about 5000 g. The centrifugate (supernate, Fraction I+II+III) was collected. The Fraction I+II+III centrifugate (supernate) was adjusted to a pH of about 5.2 while maintaining the same alcohol cencentration with a prechilled pH 4.0 buffer/ethanol solution. The resulting suspension was mixed to achieve complete precipitation and centrifuged at about 5000 g. The centrifugate (supernate, Fraction IV-l) was adjusted to about 140 meq/L by the simultaneous addition of a pH 5.95 buffer and IN sodium bicarbonate solution. Ethanol was then added to a final concentration of about 40% v/v. The suspension was mixed to achieve complete precipitation and centrifuged at about 5000 g. The resulting precipitate, Fraction IV₄ paste, was collected.

Materials

IV₄ paste - stored at -25°C HEPES - Sigma Cell Cultured tested H9136

0.9%NaCl - USP Irrigation or intravenous saline

NaCl (Sigma grade)

D-Sorbitol - Sigma cell culture tested (S3899) Lot 26F-

00915 Glycine - Sxgma cell culture tested (S1888) Lot 97F-

0895 Sodium (L) -ascorbate - Sigma (A7631) Corning 0.2μm disposable filters Falcon Fast flow disposable filters (o.22μM) Sartorious - "Sartoban" 0.8μm/0.2μm Capsule filter

Depyrogenated glassware, pressure chamber, stir bars, centrifuge bottles and caps.

Media Supplement preparation was carried out at 5°C, all buffers were at 5°C.

Resuspension Buffer

I. 20mM HEPES, 0.9% NaCl pH 7.86 ± 0.5 (5°C) a. 4.76g HEPES to 1 liter of normal saline

+12 ml of IM NaOH. Procedure:

1) The amount of IV₄ paste needed was determined by dividing the desired suspension volume by 8.33.

2) Cold (+5°) resuspension buffer (7.33 x weight of IV₄ paste) was added to the IV₄ paste and stirred for 1. 5 hours at 5°C. After 1 hour, the suspension was light brown in appearance with no large paste clumps.

3) The pH was adjusted to 7.3 ± 0.2 (5°C) by adding IM NaOH or IN HCI.

4) Stirring was continued for another 0.5 hours at 5°C.

5) The suspension was centrifuged at 10,000 x g for 20 min (5°C) .

6) Supernatant was decanted into Millipore Stainless steel pressure tank and filtered through a 0.8/0.2μm

Sartoban (Sartorius) capsule into a pyrogen free container, using a positive pressure of 10 psi. This supernatant comprised the growth-enhancing media supplement (GEMS) . For viral inactivation by pasteurization, the following steps were carried out.

7) Stabilizer addition:

Stabilizers as described in either 7A or 7B below were added slowly with rapid, but not foaming, stirring (5°C).

7A) 18% Sucrose/ 0.1% L-ascorbate [0.2g sucrose per ml of supernatant, O.lg sodium (L) ascorbate per 100ml of supernatant] .

7B) 15% Sorbitol/10% glycine/0.1% sodium ascorbate were added in order:

1) Sorbitol; 15g per 100ml original supernatant. Dissolved.

2) Glycine; lOg per 100ml original supernatant

3) sodium (L) ascorbate; O.lg per 100ml original supernatant, covered and mixed for 1 hour at 5°C. The stabilized supplement was a clear amber color.

8) The solution was filtered through a 0.2μm Falcon fastflow Corning (0.2 μm) disposable filter unit and dispensed into 50ml polypropylene centrifuge tubes.

9) Pasteurization (60°C for 10 hours) .

The stabilized supplement was placed in a water bath containing an ice/water mix. A timer was set to begin heating at 10PM and to stop at 9AM, (11 hours) . The tubes were kept submerged to the liquid fill level. The incubator was tested to assure that it would reach 60°C during the first hour of heating.

10) The pasteurized GEMS was stored at 5°C when diafiltration or desalting was planned within 24 hours. For longer periods prior to desalting, the supplement was stored at -25°C.

11A) The pasteurized GEMS was rendered essentially free of the stablizers by concentrating the GEMS to approximately half of its volume using an ultrafiltration system with PTGC membranes (polysulfone membranes with nominal molecular weight cut-off of 10,000) , followed by molecular washing and diafiltration using the same UF system with ImM HEPES, 0.9% NaCl pH 7.2. The buffer was used in an amount approximately ten times the volume of the media concentrate. After diafiltration, the GEMS was adjusted to a protein concentration of approximately 50 mg/ml protein with lOmM HEPES, 0.9% NaCl, pH 7.2 buffer.

11B) An alternate method to remove the stabilizers is desalting by gel filtration. Briefly, a Sephadex G-25 column was equilibrated with HEPES buffer (0.02M HEPES, 0.9% NaCl, pH 7.4), the stabilized suspension was loaded onto the column, and the void volume peak was collected, which was the desalted GEMS.

Proof of viral inactivation: Encephalomyocarditis viruses and sindbis viruses were added to test samples of GEMS prior to pasteurization step 10. After completion of pasteurization, no viral activity was detectable by an assay based on changes in morphology of target cells (Reed and Muench (1938) "A simple method of estimating fifty percent endpoints." Am. J. Hyg. 27:493-497) .

Polyacrylamide gel electrophoresis (PAGE) combined with Biolmage analysis demonstrated that the pasteurized GEMS contained proteins with apparent molecular weights ranging from 75 kDa to 670 kDa.

HPLC analysis of the pasteurized GEMS demonstrated prominent protein peaks corresponding to molecular weights of about 171 kDA, 362 kDa, and 1,044 kDa (Fig. 2) .

EXAMPLE 2 Preparation of PLANOVA (Asahi) -filtered Growth-Enhancing Media Supplement.

The supplement was prepared as in Example 1 through step 6. The resulting supernatant from step 6 was passed through virus-removing membrane filters, PLANOVA , Asahi Chemical Industry Co. , Tokyo, Japan. The supernatant was first passed through a PLANOVA filter having a mean pore size of 72 +/-4 nm (PLANOVA 75) and then through 1 or 2 PLANOVA's having a mean pore size of 35 +/-2 nm (PLANOVA 35) .

PAGE combined with Biolmage analysis demonstrated that the PLANOVA-filtered GEMS contained proteins with apparent molecular weights ranging from 75 kDa to 670 kDa.

HPLC analysis of the PLANOVA-filtered GEMS demonstrated prominent protein peaks corresponding to molecular weights of about 151 kDa, 382 kDa, and 538 kDa (Fig. 3) .

EXAMPLE 3 Culture of hybridomas and cell lines using GEMS.

The effect of pasteurized GEMS on cellular growth and expression levels was tested on Human and Murine Hybridoma cell lines that produce Human or Murine IgG. These cells were non-adherent cell lines. Two different adherent cell lines producing recombinant human Factor VIII were also tested in GEMS. The cell lines used in the experimentation were as listed:

Human Monoclonal Core Lipopolysaccharide...

78-1-2 producing Human IgM antibody.

Murine cell line 12A8 producing an IgG to Protein C. ιι ii it PCV310 " " " " " " " " " F8/1-5-6 " " " " FVIII. ιι it ii 88—60 " " " " " CHO cell line D-28 producing recombinant human factor VIII C-127 cell line D-28 " "

These cell-lines were seeded at 1-2E5 cells/ml and split from 50-96 hours depending on the doubling time of the cells.

Basal medium (BM) contained RPMI/Ham's F12 and sodium bicarbonate Penicillin/Streptomycin and glutamine.

Frozen cell lines were thawed, assessed for viability by trypan blue exclusion, and resuspended at 10 cells/ml in BM. The cells were plated at 10 c/ml in 20 ml of BM + 20% FCS in 75 cm tissue culture flasks.

The concentration of FCS in BM was stepped down from 20% to 4% slowly. At each stage 20%, 10%, 5%, 4%, three splits were completed i.e. viability ) 85% and cell density of 5 x 10 c/ml per split. When the culture was stable at 4% FCS + BM and 2% FCS + 2% GEM + BM, experimentation was begun. For each experiment or monthly test of the stability of GEMS a 4% FCS (2 mg/ml total protein) control was run. Two T75 flasks containing a volume of 20 ml at a cell density of 5 x 10 c/ml for each cell line were used as starting cultures. Each stationary T75 flask contained 18 ml of the required media condition, to which 2 ml cell suspension was added to yield a final volume of 20 ml. The flasks were incubated at 37°C for 56-96 hours in a 6% C0₂-in-air atmosphere. Cell densities were calculated using a hemacytometer and magnification 10 on the microscope. After the second split, 18 ml of each cell suspension was transferred to 50 ml centrifuge tubes and centrifuged at 1000 revolutions per minute for 15 minutes. Thirteen ml of cell supernatant was aspirated off, and cells were resuspended in 5 ml. The next flask was seeded at 2 x 10 c/ml. At split #3, 4 ml of supernatant were removed for antibody testing and cells were resuspended in 5 ml of media and again seeded at 2 x 10 c/ml. The fourth split was like split #3. For the final split, the procedure for split #3 was repeated except 4 flasks were seeded in 4% FCS and 2% FCS + 2% GEM to increase the cell density in preparation for the next experiment.

The concentration of Human Immunoglobulin M (IgM) was determined using a double antibody sandwich type ELISA. The human IgM ELISA was used to quantitate the amount of human IgM in samples from the anti-CLPS monoclonal antibody supernatants, produced from cell line 78-1-2.

The concentration of the mouse immunoglobulin G (IgG) was also determined by ELISA. This test was used to quantitate the amount of murine monoclonal antibody that is produced from Cell lines 12A8, PCV310, F8/1-5-6 and 88-60.

The concentration of the Factor VIII product from the cell lines CHO D-28 and C-127 was determined using a one stage activated partial thromboplastin time assay (APTT) .

Micronized silica was used as an activator and human Factor VIII deficient plasma was used as a substrate. The APTT assay is thus a screening test for deficiencies in the intrinsic system of coagulation of which Factor VIII is a part. The test sample was added to Factor VIII deficient plasma, and the degree of correction of the clotting time was measured. The test clotting time was then compared to a standard curve constructed using various dilutions of a lyophilized reference of known potency. The potency of the sample was estimated from the standard curve.

The stability of GEMS produced from different Fraction IV₄ discard pastes and stored at -25°C, 5°c, 21°- 25°C, and 35°C was determined by tracking cell growth and antibody production over a period of six months. The stability of GEMS stored in Basal Media [RPMI+HAMSF12 (1: 1)ratio] at 5°C was also tested for 5 months. Experimental conditions and results were as follows:

Cell Line 78- 1- 2 GEMS #T0965 Cells/ml Hu IgM

E5 μg/ml

FCS (4%) 8.7 ± 1.5 0.52 ± 0.26

FCS (2%) + GEM (2%) 7.9 ± 2.7 0.58 ± 0.19 GEM -70°C 6.0 ± 2.5 0.51 ± 0.27

GEM 5°C 6.7 1 2.0 0.56 ± 0.31

GEM 21°C 6.5 ± 1.6 0.44 ± 0.27

GEM 35°C 7.5 ± 1.6 0.42 ± 0.23

The best conditions for both cell density and antibody level for this cell line were obtained using basal media with 2% Fetal Calf Serum (FCS) and 2% GEMS. The optimum storage conditions for GEM was 5°C. Under these conditions the IgM antibody level was higher than that obtained with 4% FCS.

Cell Line 12A8

FCS (10%) FCS (2%) GEMS (2 mg/ml) GEMS (I mg/ml)

Using the GEMS at 2mg/ml final concentration yielded cell densities and antibody levels greater than those obtained using 2% FCS. In fact the GEMS at this concentration for cell line 12A8 produced as much IgG to Protein C as would be obtained with 4% FCS, although the cell density was lower.

The doubling times for this cell line were 18.5hr with 10% FCS; 29.5hr with 2% FCS; 25.6hr with GEMS at 2 mg/ml and 40.9hr with GEMS at 1 mg/ml.

Although GEMS supported cell growth and antibody expression this cell line performed better in FCS as low as 2%.

Cell Line D28-CHO GEMS #TQ956

Cells/ml FVIII

E5 U/ml FCS (10%) 7.9 ± 2.7 0.28 ± 0.10

FCS ( 2%) 3.1 ± 1.0 0.02 ± 0.01

GEMS (2 mg/ml) 2.5 ± 1.1 0.03 ± 0.01

The CHO cell line when grown in GEMS at 2 mg/ml did support the cells and the FVIII product expression was found to be greater than that with FCS at 2%.

GEMS also supported the expression of biologically active recombinant human Factor VIII from transfected C-127 cells, a murine mammary epithelial cell line which adhered to the flasks and grew well in GEMS without pretreatment of the flask's surface with any exogenous extracellular matrix or FBS. Factor VIII was expressed from these C-127 cells at levels equal to or greater than that seen with 10% FBS supp1ement. The use of a higher concentration of GEMS may increase the amount of Factor VIII produced. Addition of Von

Willebrand's Factor is expected to increase the stability of the FVIII product.

Cell Line F8/1- 5- 6 GEMS #T0956

Cells/ml aFVIII

E5 Mu IgG μg/ml

FCS (10%) 11.8 ± 3.1 5.2 ± 0.21

FCS ( 2%) 6.8 ± 1.3 4.2 ± 0.19

GEMS (2 mg/ml) 3.6 ± 0.5 4.2 ± 0.17 While the cell density was lower than that achieved with 10% and 2% FCS, the Murine IgG to FVIII antibody level was comparable to that achieved with 2% FCS.

Cell Line 88-60 GEMS #2830M1761179 αlgG .FVIII. μg/ml

FCS (10%) 1.2 1 0.1 FCS ( 2%) 0.7 ± 0.1

GEMS (2 mg/ml) 0.9 ± 0.1

GEMS (1 mg/ml)

0.8 ± 0.1

GEMS at 2 mg/ml with this cell line yielded a better cell density and antibody production than that achieved with 2% FCS.

Summary: GEMS supported growth and protein expression in all six cell lines tested. It worked as well if not better than 2% FCS depending on the cell line and the stability of the protein produced by the cell.

GEMS was found to be stable for at least 5 months when stored as a 50 mg/ml concentrated solution at 5°C or as a mixture with basal media at a final concentration of 2 mg/ml. It could be kept at -25°C, 21- 25°C, or 35°C with only a small loss in potency. EXAMPLE 4 Growth of hybridoma cell lines 9199 (Mouse/human) and 9105 (Rat/rat) using GEMS.

Cell line 9199 (Mouse/Human) hybridoma secreting human immunoglobulin and cell line 9105 (Rat/Rat) hybridoma secreting rat immunoglobulin were evaluated for growth support and secretion in medium supplemented with GEMS. These cell lines were chosen because both were found to be particularly challenging to grow in serum-free medium.

Both GEMS products (Pasteurized and Asahi filtered) were added to RPMI:DMEM base medium at a final total protein concentration of 2 mg/ml each with no other supplements.

Cultures were successively passaged and compared to Fetal Bovine Serum 5 percent (approximately 2.5 mg/ml total protein) Cell counts and cell viability were measured at each subculture and used to generate growth curves.

Samples of conditioned supernatant were collected at regular intervals and assayed for secreted immunoglobulin concentrations.

Both GEMS products were assayed by a human IgG ELISA assay and were found to contain substantial quantities of human immunoglobulin. The secretion data for cell line 9199 was adjusted for background IgG levels.

RESULTS:

1. GROWTH SUPPORT (Figures 4 and 5)

Cell growth and viability for both cell lines was equivalent to 5 percent FBS in both the pasteurized and the Asahi 25 filtered GEMS products at 2 mg/ml. ANTIBODY SECRETION (Table 1)

Cell line 9199:

The concentration of secreted antibody was slightly greater in medium supplemented with both GEMS products compared to five percent FBS and serum free medium (MF28) .

Cell line 9105:

The concentration of secreted antibody was approximately equivalent to five percent FBS in medium supplemented with both GEMS products.

TABLE I

GEMS ASAHI FILTERED = 176.6 UG/ML HUMAN IgG (T.P.= 22.0 MG/ML)

16.0 UG/ML @ T.P. of 2 MG/ML

GEMS . PASTEURIZED = 420.9 UG ML HUMAN IgG (T.P.= 50 MG/ML)

16.8 UG/ML @ T.P. of 2 MG/ML

EXAMPLE 5 Removal of IgG from GEMS.

When GEMS is to be used to support the growth of hybridomas secreting human IgG, it may be desirable to remove the IgG component from the GEMS in order to simplify purification of the product from the culture media. Both pasteurized and Asahi filtered GEMS products are suitable for IgG removal.

Materials: Pharmacia Protein A Sepharose CL6B BioRad column, 1 X 9.7 cm

In this example, four alternative forms of GEMS were used as starting material for removal of IgG:

(1) Pasteurized Sorb/Gly/ascorbate suspension from Example 1, steps 7B, 8-9.

(2) Pasteurized Sucrose/ascorbate suspension from Example 1, steps 7A, 8-9.

(3) Pasteurized, dialyzed Sorb/Gly/ascorbate suspension from Example 1, steps 7B, 8-11A. (4) Pasteurized, dialyzed Sucrose/ascorbate suspension from Example 1, Steps 7B, 8-11A.

The lyophilized gel was swollen in HEPES buffered saline

(HBS: 0.02 M HEPES, 0.9% NaCl, pH 7.4) . After 15 minutes, the gel was washed with approximately 200 ml of HBS per gram of lyophilized Protein A gel at 23° C. The BioRad column was poured and packed at approximately 1.5 ml/min.

The column was loaded with either of the above listed forms of GEMS at 1 ml/min.

The flow-through material was recovered as GEMS having human IgG essentially removed.

To assess the amount of immunoglobulins (Ig) removed from the G__IMS, the Ig bound to the column was eluted with 50 ml of IM acetic acid, 0.1M glycine-HCl, pH 3.0. Assays for IgG, IgA, and IgM content were performed for the flow-through materials as well as the eluted Ig. Assays for human serum albumin (HSA) were performed in order to use HSA content as an internal control based on the assumption that HSA does not bind to Protein A.

IgM content was tested by the well-known ELISA assay. IgG, IgA, and HSA were tested by radial immunodiffusion assay (RID) . Briefly, for RID, a test sample antigen was applied to a well cut into a gel matrix incorporated with the antiserum corresponding to the antigen to be tested. As the antigen diffused through the gel, it reacted with the antiarum to form a precipitin ring. The resulting diameter of the precipitin ring was measured and compared to the diameter of the precipitin rings of the antigen standards of known concentrations to quantitate the test sample antigen.

Results: No detectable IgG (assay sensitivity = >20 μg/ml) was found in the flow-through samples. In contrast, IgG was found in the eluted material. It was calculated that at least 93% of the IgG was removed from the GEMS.

EXAMPLE 6 Growth of cells in GEMS with attachment factors.

Materials:

Ham's F12, Gibco 211L 4185/320-1765 RPMI 1640, gibco 18K6431320-1875 A Fetal bovine serum, Hyclone A1110-41110701 Fetuin, Sigma 124F8890/F2379

CollagenVI (Type IV) , Sigma 88F3844.C7521

GEMS, Lots 1 and 2

CHOK, cells (ATCC # CCL 61) at passage #8

CHOK₁ cells (ATCC CCL 61) were harvested from serum containing medium and washed once with RPMI:Ham's F12 serum free medium. Aliquo_s of cells were dispensed into duplicate tissue culture wells (lX10E6/well) containing RPMI:Ham's F12 supplemented with GEMS at a 2 mg/ml concentration. The sets of wells were pretreated (O/N at 37 deg. C, 7% C02) with 10% FBS, collagen type VI (0.025, 0.05, 0.1, 0.5 mg/ml) or fetuin (0.07, 0.13, 0.25, 0.50 mg/ml) . The cultures were incubated at 37°C, 7% C0₂ and observed daily for plating efficiency, cell growth and morphology. Cultures were harvested and compared to 10% FBS control wells at 72 hours.

The growth of the CHOK., cells was examined in RPMI 1640: Ham's F12 supplemented with GEMS (lot 2) (with 10% FBS pretreatment) . The test was compared to 10% FBS supplemented RPMI 1640: Ham's F12 medium (positive control) . Cells were seeded at 1X10 into duplicate sets of T25 flasks. Duplicate flasks were harvested on days 1, 2, 3, 6 and 7, cell yields were recorded, growth curves were prepared, and population doubling times calculated.

Results: GEMS resulted in higher cell density than basal medium alone when the substrate was pretreated with attachment factors. Addition of Fetuin at 0.5 mg/ml to GEMS resulted in slightly improved morphology up to 48 hours. Addition of Collagen to GEMS resulted in a slight tendency towards flattening out of cells by 72 hours. In the absence of FBS or FBS pretreatment, cells attach but do not flatten out. Cell morphology was approximately the same in base medium alone as base medium with GEMS.

Cell yields for cultures pretreated with 10% FCS are shown in Table 2. TABLE 2

10% FBS PRETREATMENT

GEMS at 1:25 dilution supported CHOKI₁ cell growth equivalent to 10% FBS containing medium when the substrate was pretreated with FBS. The log phase(0-48 hours) population doubling times were: GEMS - ιι.7 hours 10 % FBS -11.5 hours

When the substrate was pretreated with collagen at 0.5 mg/ml for the GEMS (1:25) medium, the population doubling time was also equivalent to that for 10% FBS (1₂ hours and 11.6 hours, respectively) and the cell yields were equivalent (1.6 X 10⁶ and 1.7 X 10⁶ respectively)

EXAMPLE 7 Use of GEMS for growth of cell lines MB231, H69, and K562.

MB231 is a human breast cell line which requires attachment, H69 is a human small cell lung line, and K56₂ is a human promyelocyte cell line. The latter two cell lines can grow in suspension.

Long-term culture in 24-well cluster plates: i.o x lo^A K562 cells or 1.5 X 10 H69 cells were plated in each well. After the first 3 days of culture cell density was to ₄.₄ X 10 cells/well.

Initial test conditions were: 2.5% and 1% fetal bovine serum (FBS) with and without 5% GEMS, 2.5% GEMS, 1.25% GEMS, 0.625% GEMS, and 0.312% GEMS. After the first 3 days of culture, cells were transferred to 0% and 1% FBS with and without GEMS at the above concentrations and then cultured for an additional 12 days.

4.4 X 10 K562 cells per well were cultured in RPMI medium plus 1% and 0% FBS with and without the above concentrations of GEMS. Cell number was determined every three days for 18 days.

MB231 cells were cultured in 96-well microtiter plates using varying concentrations of FBS (10%, 2%, 0.5%) , GEMS (0%, 2%, 4%), with or without laminin and fibronectin at lOOμg/ml.

The MTT assay was employed to determine the number of viable cells. MTT is a soluble tetrazonium salt which is reduced to insoluble, colored formazans by the metabolic activity of living cells. The crystal violet assay was used to determine cell number based on its staining of protein.

Results: GEMS alone at concentrations between 1% and 4% supported the growth of cell lines MB231, H69, and K562 as well as or better than did FBS at concentrations between 1% and 10%. A typical result is shown in Figure 6.

EXAMPLE 8 GEMS used to support the growth and lytic activity of human lymphocytes.

Human lymphocytes were obtained from peripheral blood mononuclear cells and tumor specimens. For collection of peripheral blood lymphocytes, 50 ml of whole heparinized blood was separated on Ficoll-Hypaque and the mononuclear cells were harvested from the interface. The cells were washed 3X and resuspended for counting.

Cells were cultured in 10 different conditions as follows:

AIM V (serum-free medium, GIBCO)

4% GEMS + supplement 2% GEMS + supplement 1% GEMS + supplement

supplement 61.5

The supplement consisted of hydrocortisone (0.5 μg/ml) , insulin (10 μg/ml) , EGF (5 ng/ml) , T₃ (10^" M) , selenium (2 ng/ml) , linoleic acid BSA (5 μg/ml) . The supplemented RPMI without GEMS also received transferrin (20 μg/ml) .

All cultures received lOOOU/ l of recombinant IL2. The cells were placed in 25cm 2 flaskes at a densi.ty of 5 X 105 cells/ml and were incubated for 4 days at 37°C in a 5% C0₂ atmosphere.

Materials:

GEMS, Interleukin-2 , Anti-CD3 and Anti-CD28 monoclonal antibodies.

Lymphocyte Activation:

Lymphocytes at 2-5 X 10 cells/ml were either:

1. Incubated, in tissue culture ware coated with Anti- CD3 Ab, in indicated medium containing .5 μg/ml of Anti- CD28 AB and 25 units/ml of IL-2. 2. Incubated in medium containing 1000 units/ml of IL- 2. Cultures were split at around 1.5 X 10 cells/ml and reseeded 2-5 X/10 cells/ml.

Lytic assay: A standard Cr release assay was used to determine the lytic activity of activated lymphocytes on

K562 cells. K562 cells were first loaded with Cr 51 and subsequently added to the cells to be tested for lytic activity at an effector to target ratio of 2.5:1, 10:1, and 25:1. The plates were incubated at 37°C for 4 hours after which 100 μl of supernatant was removed from each well and counted in a scintillation counter. % cytotoxicity was calculated using the formula

experimental CPM minus spontaneous release CPM

100 X maximum release CPM minus spontaneous release CPM

Tumor infiltrating lymphocytes (TIL's) were isolated from solid specimens obtained from surgery or malignant fluid specimens. Solid specimens were minced, and, if necessary, digested with enzymes. Malignant fluid specimens were washed and resuspended in medium containing either IL-2 or medium containing a combination of anti-CD3, anti-CD28, and IL-2. Cells were incubated at 37°C in a fully humidified incubator.

Growth assays: Effects on growth of the above listed media conditions were compared with the effects of 2% AB serum, HDL, and HB104. Cells were collected and counted between 4 and 25 days in culture. Ongoing DNA synthesis was confirmed by a standard H-thymidine assay.

Results: Figure 7 exemplifies the results of a typical PBL proliferation experiment conducted over 25 days. Until day 15, 2% GEMS supported growth to an extent essentially equivalent to that afforded by HDL, HB104, and 2% AB serum.

Table 3 further demonstrates that 2% GEMS was comparable to 10% FCS in supporting the proliferation of PBL'S. TABLE 3

GEMS Q LAK PROLIFERATION

CELL RECOVERY X 10 ^*6 (% FCS)

The incorporation of H-thymidine at days 9 and 11 of PBL culture confirmed that 2% GEMS supported DNA synthesis as well as did 10% FCS or AIM V.

Figure 8 shows that 2% GEMS supported the lytic activity of lymphokine-actvated killer (LAK) cells. Results using 2% GEMS were typically comparable to results obtained using AIM V or 2% AB serum.

Figure 9 demonstrates that tumor-infiltrating lymophocytes (TIL's) obtained from a breast tumor sample and a melanoma sample proliferated in medium supplemented with GEMS.

EXAMPLE 9 Growth of primary rabbit kidney (PRK) cells in GEMS.

Primary cells were obtained from rabbit kidney and grown in glass tubes containing RPMI plus either 10% FBS as a control, or lower concentrations of FBS supplemented with 2% GEMS. After several days in culture, cell counts were compared as shown in Figure 10. GEMS supplemented 1% or 2% FBS to support growth of PRK cells to a level comparable to that obtained with 10% FBS.

EXAMPLE 10 The use of GEMS to sustain the proliferation of CD34+ hematopoietic progenitor cells in serum-free medium.

Methods: CD34+ cells were purified using Dynal immunomagnetic beads, from bone marrow or leukapheresis products from patients who had been mobilized with G-CSF during the recovery from chemotherapy induced leukopenia. The selected CD34+ cells were then seeded into suspension cultures at 10 /ml. The control media called HLTM (for human long term culture medium) , was a McCoy 5A base medium supplemented with amino acids, vitamins, monothioglycerol and hydrocortisone and contained 12.5% fetal bovine serum (FBS) and 12.5% horse serum (HS) from pretested batches. This media has been previously shown to give optimal performance in the proliferation and differentiation of CD34+ cells. The growth factors IL-3 (30OU/ml) , G-CSF(30OU/ml) , GM-CSF(300U/ml) and SCF(20ng/ml) were added to all of the cultures. The base medias to which GEMS was added also included Iscove's modified Dulbecco's medium (IMDM) without serum. The cultures were then harvested and the proliferation index was determined by dividing the final number of cells by the initial number. In addition, a sample of the cultured cells was stained with antibodies to CD15 and CDllb and analyzed by flow cytometry to determine the differentiation state of the cells. This phenotyping protocol defines a population of CD15+CDllb- cells that represent myelocytes and promyelocytes. Propidium iodide was also added to these samples to identify the non-viable cells present. Cell samples from the cultures were also plated in methyl cellulose cultures containing recombinant growth factors and incubated for an additional 14 days to estimate the number of cells that form colonies containing >50 cells and the proportion of granulocyte-macrophage colony forming cells, macrophage colony forming cells, erythroid colony forming cells (BFU-E) and mixed colony forming cells. These data are presented as indices of the final numbers of colonies divided by the initial numbers of colonies. In addition the numbers of cluster forming cells which produce less than 50 cells was estimated and presented in a similar manner.

Alternatively, samples of 2500 CD34+ cells/well were incubated in 96 well microtiter plates containing control media or different concentrations of GEMS for 7 days and pulsed overnight with H-Thymidine. The cells were harvested onto filters and counted in a scintillation counter to determine cpm.

Results (Figure 11) :

Cultures of peripheral blood CD34+ cells in a McCoy's or

IMDM base medium supplemented with GEMS.

Four preparations of selected CD34+ cells were evaluated: two preparations from bone marrow and two from leukapheresis products. Filtered GEMS at 5 mg/ml or 0.2 mg/ml in a McCoy's base medium had equivalent proliferation of cell numbers (PI day 11) when compared to the control HLTM medium containing 25% FBS/HS. Viabilities and the percent of GM colony forming cells were slightly lower in the GEMS supplemented media but CD15+ CDllb- cells were in similar proportions. In contrast, the cultures containing pasteurized GEMS produced only one third the number of cells as the control when McCoys 5A base was used with peripheral blood CD34+ cells but performed in a similar manner to filtered GEMS or the control when IMDM base was used with bone marrow derived CD34+ cells.

Expansion of colony forming cells in these cultures showed a predominant expansion of GM colonies and clusters. In all four of these experiments the addition of GEMS resulted in expansion of colony and cluster forming cells to a similar degree to that observed with the control media.

Thvmidine incorporation from CD34+ cells cultured in a McCoy's or IMDM base medium supplemented with GEMS.

Five preparations of selected CD34+ cells were evaluated in a thymidine incorporation assay. These included one peripheral blood and three bone marrow preparations. Two of the preparations cultured in McCoy's base medium, RT2739 (peripheral blood) and RP9 (bone marrow) show incorporation levels at about two thirds that of the control HLTM media containing 25% FBS/HS. The activity of GEMS was observed over a range from 5 to 0.312 mg/ml.

Two bone marrow CD34⁺ preparations that were studied, RP12 and RP13 were cultured in varying concentrations of GEMS in either McCoys or IMDM (295) medium. Similar results were observed with a bimodal pattern of activity. Concentrations of GEMS from 0.625 to 2.5 mg/ml gave incorporation levels equivalent or greater than the HLTM control.

Claims

What is claimed is:

1. A method for producing a human-derived growth- enhancing media supplement for the in vitro culture of cells, comprising; (a) providing an amount of a paste comprising human serum Cohn fraction IV₄,

(b) contacting said paste with a resuspension buffer, said buffer having a pH of about 7.4 to about 8.4 and said buffer being in an amount sufficient to form a first suspension,

(c) adjusting the pH of said first suspension while stirring to form a homogeneous second suspension having a stable pH of about 6.4 to about 8.0,

(d) clarifying said second suspension to form a supernatant,

(e) filtering said supernatant through a sterilizing filter to form said growth-enhancing media supplement.

2. The method of claim 1 wherein said Cohn fraction IV is derived from a pool comprising sera obtained from at least about 3000 human donors.

3. The method of claim 1 further comprising filtering said growth-enhancing media supplement through a virus-removing membrane filter.

4. The method of claim 1 further comprising the steps of

(f) adding to said growth-enhancing media supplement at least one stabilizer selected from the group consisting of sucrose, sorbitol, glycine, and ascorbate to form a stabilized third suspension,

(g) heating said third suspension at 60°C to about 62°C for at least 10 hours, and

(h) separating said stabilizer from said third suspension to form a pasteurized growth-enhancing media supplement.

5. The method of claim 3 further comprising, after step (h) , filtering said pasteurized growth-enhancing media supplement through a sterilizing filter.

6. The method of claim 4 wherein said separating step (h) is accomplished by a method selected from the group consisting of dialysis, diafiltration, and gel filtration.

7. The method of claim 1 further comprising contacting said supernatant of step (d) with an Ig-binding component to form an Ig molecule/Ig-binding component complex, and removing said complex from said supernatant.

8. The method of claim 3 further comprising contacting said growth-enhancing media supplement with an Ig-binding component to form an Ig molecule/Ig-binding component complex, and removing said complex from said growth- enhancing media supplement.

9. The method of claim 7 or claim 8 wherein said Ig molecule is IgG and said Ig-binding component is Protein A or Protein G.

10. The method of claim 7 or claim 8 wherein said removing step is accomplished by batch adsorption or column adsorption.

11. The growth-enhancing media supplement produced by claims 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10.

12. A method for culturing cells in vitro, comprising;

(a) providing an amount of a paste comprising human serum Cohn fraction IV₄,

(b) contacting said paste with a resuspension buffer, said buffer having a pH of about 7.4 to about 8.4 and said buffer being in an amount sufficient to form a first suspension,

(c) adjusting the pH of said first suspension while stirring to form a homogeneous second suspension having a stable pH of about 6.4 to about 8.00,

(d) clarifying said second suspension to form a supernatant,

(e) filtering said supernatant through a sterilizing filter to form a growth-enhancing media supplement, (f) adding said growth-enhancing media supplement to a basal medium to form a complete growth medium, and

(g) culturing said cells in said complete growth medium.

13. The method of claim 12 wherein said Cohn fraction IV₄ is derived from a pool comprising sera obtained from at least about 3000 human donors.

14. The method of claim 12 further comprising filtering said growth-enhancing • media supplement through a virus- removing membrane filter.

15. The method of claim 12 further comprising the steps of adding to said growth-enhancing media supplement at least one stabilizer selected from the group consisting of sucrose, sorbitol, glycine, and ascorbate to form a stabilized third suspension, heating said third suspension at 60°C to about 62°C for at least 10 hours, and separating said stabilizer from said fourth suspension to form a pasteurized growth-enhancing media supplement.

16. The method of claim 15 further comprising, after said separating step, filtering said pasteurized growth- enhancing media supplement through a sterilizing filter.

17. The method of claim 15 wherein said separating step is accomplished by a method selected from the group consisting of dialysis, diafiltration, and gel filtration.

18. The method of claim 12 further comprising contacting said supernatant of step (d) with an Ig-binding component to form an Ig molecule/lg-binding component complex, and removing said complex from said supernatant.

19. The method of claim 15 further comprising contacting said growth-enhancing media supplement with an Ig-binding component to form an Ig molecule/Ig-binding component complex, and removing said complex from said growth- enhancing media supplement.

20. The method of claim 18 or claim 19 wherein said Ig molecule is IgG and said Ig-binding component is Protein A or Protein G.

21. The method of claim 18 or claim 19 wherein said removing step is accomplished by batch adsorption or column adsorption.

22. The method of claim 12, 13, 14, 15, 16, 17, 18, 19 or 20, wherein said cells are selected from the group consisting of hybridomas, immune-system cells, primary tumor cells, and members of mammalian cell lines.

23. The method of claim 22 wherein said cells are members of a cell line selected from the group consisting of CHO cells, MB157 cells, MB231 cells, H69 cells, K562 cells, C- 127 cells, and Vero cells.

24. The method of claim 22 wherein said immune-system cells are selected from the group consisting of peripheral blood lymphocytes, lymphokine-activated killer cells, tumor-infiltrating lymphocytes, and CD34+ cells.

25. The method of claim 23 wherein said cells express a recombinant protein.

26. The method of claim 25 wherein said recombinant protein comprises human Factor VIII.

27. A cell culture produced by the method of claim 21.

28. A cell culture produced by the method of claim 22.

29. A cell culture produced by the method of claim 23.

30. A cell culture produced by the method of claim 24.

31. A cell culture produced by the method of claim 25.

32. A cell culture produced by the method of claim 26.