WO1987005929A1

WO1987005929A1 - Immortalized cells which produce tissue-specific products

Info

Publication number: WO1987005929A1
Application number: PCT/US1987/000710
Authority: WO
Inventors: Steven K. H. Foung; Linda B. Rabin
Original assignee: Genelabs Incorporated
Priority date: 1986-04-01
Filing date: 1987-04-01
Publication date: 1987-10-08
Also published as: EP0262211A1; AU7281787A

Abstract

An immortalized human tissue cell line formed by fusing a selected human, or non-human primate non-lymphocytic tissue cell with an immortalizing human/mouse hybridoma cell. The cell line is selected for its ability to produce a product normally produced by the tissue cell type, as exemplified by immortalized human liver cells capable of secreting a variety of liver proteins, such as fibronectin, factor VIII, and serum albumin, and immortalized pancreatic islet cells capable of secreting insulin.

Description

IMMORTALIZED CELLS WHICH PRODUCE TISSUE-SPECIFIC PRODUCTS

1. Field of the Invention

The present invention relates to immortalized human or non-human primate tissue cells, and their use in producing tissue-specific products, and to methods for producing the immortalized cells.

2. References

Ben-Jonathon. N. , et al Methods Enzymology 103:249

(1983). Bone, A.J., et al. In Vitro 18(2): 141 (1982). Boyum, A., Scand J Lab Clin Invest, 21:(Supp. 97)77. Carrell. R.W. , et al. Nature. 298:329-334 (1982). Clark. E.A., et al, Immunogenetics, 18:599 (1983). Cruetz, C.E., et al. In "Calcium Binding Proteins and Calcium Functions in Health and Disease", Siegel, F.L. et al, eds, North Holland Publishing company, NY, 1980. Guguen-Guillouzo, C. et al, from "Isolation,

Characterization, and Use of Hepatocytes (Harris, R.A., et al, eds.), p. 105, Elsevier Press (1983). Hynes, R.O., et al, J Cell Biol, 95:369 (1982). Knowles, B.B., et al, in Adv. in Hepatitis Res, (Chisari. F.V., et al, eds.), p. 196, Mason Publishing, NY (1984). Kohler, G., et al, Nature, 256:495 (1975). Krack, G., et al, in "Isolation, Characterization, and Use of Hepatocytes", ibid, p. 391. Lane, H.C., et al, J Exp Med, 155:333 (1982). Liang. T., et al. Endocrin, 115(6):2311 (1984). Liebman. H.A. , et al, PNAS, 82:3879-3883 (1985). Loughlin, J.S., et al , Am J Physiol, 246(Pt1):E145 (1984).

Oleivera, E.B., et al, J Biol Chem, 254:489-502 (1979). Rotblatt, F., et al, Biochem, 24: 4294-42300 (1985). Schmid. I., et al, Biochem, 12: 2711-2724 (1973). Sekiguchi. K. , et al, PNAS, 77:2661-2665 (1980). Takahashi. N., et al, PNAS, 84:2021-2025 (1984). Udenwald, W.F., et al, BBRC, 112:147-155 (1983). Wolf, G., et al, Endocrinol, 108:805-811 (1981).

3. Background of the Invention The ability to culture normal human tissue-specific cells in culture over extended periods allow a variety of tissue-specific cell products to be produced relatively easily and cheaply. In particular, stable cell cultures would provide a convenient source of many secretory products, such as protein and peptides. which are presently unavailable in commercially practical quantities, or which can be made only by recombinant DNA techniques. The latter approach generally involves laborious cloning and vector selection procedures, and for many mammalian proteins, often present difficulties in terms of expression, post-translational modification, such as glycosylation. stability, and secretion from a suitable host. By contrast, stable, tissue-specific cells in culture would provide a high-output source of naturally produced, processed, and/or secreted protein which could be isolated from culture medium or cells by known methods.

Another advantage of the cultured tissue-specific cells would be the ability to produce. in a single culture medium, a battery of different cell products. For example, cultured liver cells could produce, in a single culture, several valuable liver-secretory proteins, including factor VIII, α-1 antitrypsin, α-1 acid glycoprotein, hemopexin, fibronectin, and C-reactive protein. Alternatively, the cultured cells could be selected for high production of one or a few secretary products.

A basic the limitation in mammalian cell culture to date has been the inability to propagate specific organ or tissue cells in a differentiated form over extended culture periods. Although tissue-specific cells, such as liver parenchymal cells, pancreatic islet cells, and pituitary cells, can be maintained in culture for short periods, the cells either die within several weeks or shift toward a less differentiated state. For example, isolated human hepatocytes. when placed in culture, lose the ability to carry out normal liver-cell functions, such as serum albumin secretion and gluconeogenesis, over a period of 1-2 weeks . Similarly , pancreatic is let cells in culture show a gradual loss in insulin secretion over a 1-2 week period. The period of cell viability can be extended somewhat by certain strategems, such as by coculturing the tissue-specific cells with undifferentiated epithelial cells: however, the period of cell viability is still limited. By way of illustration, cultured hepatocytes, when cocultured with liver epithelial cells, can be maintained in a viable state (as evidenced by continued liver-cell functions) for at most 6-8 weeks (Guguen-Guillouzo).

Given the potential advantages of being able to grow tissue-specific cells, particularly human cells, stably in culture, it is not surprising that considerable effort has been invested in this problem. To date, however, these efforts have been largely unsuccessful. As mentioned above, efforts to extend the culture period of cultured hepatocytes, such as by coculturing the cells with undifferentiated epithelial cells, have produced only limited increase in culture times. Some malignant cells associated with specific human tissues, such as hepatoma cells derived from malignant liver tissue, are able to grow stably in culture, but the ability to produce many cell products normally produced in vivo may be lost. Further, many of the tumor-derived cell lines, such as hepatoma cell lines which have been characterized appear to contain integrated viral genomes (Knowles), and thus products isolated from the cells would be suspected of containing viral contaminants.

Attempts to produce stable, tissue-specific cells in culture by fusing the cells with related myeloma cells have also been attempted. This approach is analogous to the one used for immortalizing antibody-secreting human lymphocytes, by fusion with myeloma cells under defined selection conditions. Heretofore no successful application of this technique have been reported.

4. Summary of the Invention

It is therefore an important object of the invention to provide a method for producing human and non-human primate cells which can be maintained stably in culture over a period of several months or more, and which retain the ability to produce and, in many cases, secrete, tissue-specific products in culture.

Another object of the invention is to provide such cells which are a convenient source tissue-specific products. The invention includes an immortalized non-lymphocytic, non-malignant human or non-human primate cell derived from a selected tissue source whose cells are normally not capable of long-term growth in culture. The cells are formed by fusion of a mouse/human hybridoma with a human or non-human primate cell derived from the selected non-lymphocytic tissue. The fusion partner is produced by fusing mouse myeloma cells and human B-lymphocytes, and selecting fusion products which show stable human chromosome retention, as evidenced by continued HLA surface antigen production in culture. The fusion partner is fused with the tissue-selected cell under conditions which allow growth of successful trioma fusion products only. In a preferred embodiment of the invention, the fusion partner is one formed by fusing mouse myeloma cells and human B-lymphocγtes, selecting fusion products which show immunoglobulin secretion and HLA surface antigen production in culture, and treating the selected fusion products with a mutagen. The mutagenized fusion products are selected for those which retain the ability to produce HLA surface antigen, show no immunoglobulin secretion, and are unable to survive in a growth medium which allows growth of a successful product formed by fusing the fusion partner with such a human cell. An exemplary fusion partner has the characteristics of ATCC No.HB8464.

The hybrid cells produced by the fusion of the fusion partner with a selected non-lymphocytic cell is selected selected for production of one or desired cell products, such as factor VIII, α-1 antitrypsin, α-1 acid glycoprotein, hemopexin. fibronectin, C-reactive protein, transcortin, synexin I, synexin-II, and serum albumin, from cultured hepatocytes, insulin from cultured islet cells, and growth hormone from cultured pituitary cells. The fusion products may be selected for high production of specific products and/or for loss of production of certain normal specific cell products.

These and other objects and features of the invention will become more fully apparent from the following detailed description of the invention.

Detailed Description of the Invention

I. Definitions

As used herein, "trioma" refers to a cell line which contains genetic components originating in three originally separate cell lineages. As used in the context of this application, these triomas are stable, immortalized cells which result from the fusion of a murine myeloma/human hybridoma with a non-lymphocytic tissue cell from a human or non-human primate source. The murine myeloma/human hybridoma (the "immortalizing hybridoma" or "fusion partner") is an immortal cell line which results from the fusion of a murine myeloma or other murine tumor cell with human lymphoid cells derived from a normal (preferably non-immunized) subject. As described below, by careful selection and mutation, an immortalizing hybridoma which provides improved chromosomal stability, has human characteristics, and which does not secrete immunoglobulin is obtained.

"Non-secreting" Jiybridoma refers to a hybridoma which is capable of continuous reproduction and. therefore, is immortal, which lacks the capacity to secrete immunoglobulin.

A hybridoma "having human characteristics" refers to a hybridoma which retains detectable human-derived chromosomes, such as those producing human HLA antigen which will be expressed on the cell surface.

"Tissue-specific cells" refer to non-lymphocytic human or non-human primate cells derived from a selected tissue or organ, such as liver, pancreas, pituitary, nervous tissue, and adrenal gland, and which maintain, in a cultured state, the ability to produce and preferably secrete, tissue-specific products, such as secretory proteins. "Cell line" refers to various embodiments including but not limited to individual cells, harvested cells, and cultures containing cells so long as these are derived from cells of the cell line referred to. By "derived" is meant progeny or issue. It is further known in the art that spontaneous or induced changes can take place in karyotype during storage or transfer. Therefore, cells derived from the cell line referred to may not be precisely identical to the ancestral cells or cultures, and the cell line referred to includes such variants.

II . Preparing Stable Human and Non-Human Primate Cell Lines

A. The Immortalizing Fusion Partner Cells

The cells which make up the immortalizing hybridoma are murine myeloma cells and human lymphoid B cells. Murine myeloma cell lines are commonly available and may be obtained through the American Type Culture Collection (ATCC), located at the National Institutes of Health (NIH) in Bethesda, Maryland. Human lymphoid B cells are isolated from the blood of normal individuals using conventional techniques. Such procedures include density gradient purification and separation of B cells from T cells using standard sheep erythrocyte rosetting. B. Tissue-Specific Cells

The tissue-specific cells which are to be immortalized by fusion with the mouse/human fusion partner are isolated cells derived from a selected, non-lymphocytic tissue from a human or non-human primate. The latter source includes primates, such as chimpanzees, whose phylogenetic similarity to humans (Clark) permits successful fusion with the mouse/human fusion partner. The ability of the mouse/human fusion partner to fuse with and immortalize chimpanzee cells forms the basis of an immortalized chimpanzee cell line described in co-owned patent application for "Non-Human Primate Monoclonal Antibodies and Methods", Serial No. 767, 213, filed May 1, 1985. The cells are isolated from a tissue or organ which produces the selected tissue-specific product normally in vivo. The cells must also be capable of limited viability in culture, without appreciable loss of the desired tissue-specific property, for a short period until the cells have been successfully immortalized by fusion with the immortalizing partner.

A variety of methods for preparing tissue cells for culture have been developed. As indicated above, these techniques are aimed primarily at producing isolated cells which show good viability and retention of tissue-specific properties in short-term cell culture. These procedures are generally suitable for isolating tissue-specific cells for fusion with an immortalizing mouse/human hybridoma according to the present invention. Methods for isolating selected tissue cells will now be considered.

Several methods for producing isolated hepatocytes for cell culture have been reported (Krack, Guguen-Guillouzo). In general, these methods involve perfusing liver tissue with a solution of collagenase, minsing the enzyme-digested tissue to release free cells, and filtering the isolated cells through a cloth mesh. A simpler method of cell preparation which does not involve collagenase treatment has been implemented by the inventors, and has been shown to give good fusion efficiency with the immortalizing partner. In this method, the isolated tissue, such as fetal abortus or adult liver tissue, is homogenized in a ground glass homogenizer having a single cell clearance, and the homogenized cells washed several times in a suitable culture medium, as described in Example II.

Collagenase digestion is also widely used in methods for isolating pancreatic islet cells (Bone). The cell-isolation procedures generally follow those used in hepatocyte isolation. Alternatively, the islet cells can be isolated by mechanical dispersion, using the simple homogenization technique developed by the inventor, and described in Example VIII. Mammalian pituitary and hypothalamus cells have also widely studied culture, and a variety of methods are known for isolating anterior (Liang, Loughlin) pituitary cells, and primary pituitary cells (Ben-Jonathon) for cell culture. These methods generally follow enzyme-dispersion procedures like those outlined above for hepatocyte isolation. Similarly, the mechanical dispersion technique outlined above may be used.

Other cell types, such as adrenal medulla cells, myocytes, and nerve cells have also been isolated for use in cell culture, and these cells would also be suitable for immortalizing according to the method of the invention. C. Fusion Procedures

Fusions to form the murine-human non-secreting hybridomas and the immortalized tissue cell lines of the invention are performed by a modification of the method of Kohler and Milstein. Briefly, a mouse myeloma and human lymphocyte (to make the immortalizing hybridoma) or the hybridoma and the isolated tissue-specific cell (to make the immortalized tissue cell line) are combined in the presence of a fusogen such as polyethylene glycol under suitable conditions, e.g., 40%-50% polyethylene glycol (1000 to 4000 molecular weight) at between room temperature and 40°C, preferably about 37°C. Fusion requires about 5-10 minutes, and the cells are then centrifuged and screened. Alternatively, cell fusion, such as the fusion between the tissue cells and the immortalizing hybridoma, may be carried out by standard electrofusion methods. Example VII illustrates the method, involving fusion of adult human liver cells with an immortalizing mouse/human hybridoma cell line. The method used an electrofusion apparatus obtained by Amsco (Erie, PA), following the general fusion conditions suggested by the manufacturer.

D. Screening Procedures

Following the fusion procedure, screening for hybridized products is made by culturing cells centrifuged from the fusion medium in growth medium which is selective for the desired hybrids. Ordinarily, non-immortalized cells cannot survive repeated transfers on any medium, and hence will not survive repeated culturing of the centrifuged cells. Commonly used lines of immortalized murine myeloma cells, however, are incapable of growth on certain selective media which have been chosen to deprive them of their ability to synthesize DNA. Two very commonly used media of this description are "hypoxanthine-aminopterin-thymidine" or "HAT" medium and azaserine-hypoxanthine medium or "AH" medium.

Both of these selection media take advantage of the capacity of normal cells to utilize a "salvage" pathway for DNA synthesis under circumstances where the de novo process is inhibited. Aminopterin inhibits both de novo purine and pyrimidine nucleotide synthesis in normal cells and both thymidine and hypoxanthine are required for the salvage pathway. Azaserine inhibits only purine synthesis, so only hypoxanthine is required for the salvage pathway. The salvage process, which requires hypoxanthine phosphoribosyl transferase (HPRT) is generally inoperable in the commonly used murine myeloma cells (although they retain the de novo pathway). Since aminopterin (in the HAT medium) or azaserine (in the AH medium) are both inhibitors of the de novo DNA synthesis pathway, the murine myeloma cells are incapable of growth in either "HAT" or "AH" medium. Thus, only hybridized cells can both survive repeated transfers and grow in HAT or AH medium. Normal lymphocytes cannot survive because they are not immortalized, and do not survive repeated transfers: unhybridized myeloma cells cannot survive because they lack the salvage pathway which permits the use of hypoxanthine to overcome aminopterin or azaserine inhibition.

E. The Immortalizing Hybridoma

The selection procedures employed in producing the immortalizing hybridoma are aimed at selecting cells which show stable human characteristics, non-secretion of immunoglobulin, and sensitivity to a defined selection medium which can be used for selecting successful immortalized tissue cell hybrids. Briefly, the cells centrifuged from the fusion mixture of mouse myeloma and human lymphoid cells are diluted and plated in microtiter plates. Screening is done using AH or HAT medium growth, with selection of successful colonies being made on the basis of assay procedures related to stability and human character. From among the many colonies assayed, several are chosen which continue to produce immunoglobulin in the supernatant fluid for a suitable period of time, preferably in excess of six months (one criterion for stability). Here it is noted that the culture period of two-three months required to obtain a sufficient number of cells for mutagenesis, as described below, may also serve as a period for gauging stability of immunoglobulin secretion. The continued production of such immunoglobulin indicates that the characteristics conferred by the human lymphocyte partner have not been lost (lymphocytes which were unfused will, of course, not survive). Retention of human characteristics is assessed by assaying the cell surfaces for the presence of HLA antigen. The selected colonies continue to exhibit HLA antigen expression at their cell surfaces, and in fact, continued production of immunoglobulin and HLA surface antigen appear to be linked, as the two traits are invariably found in the same cells. The selected cells are then treated with a mutagen. such as 6-thioguanine, to destroy their ability to secrete immunoglobulin and confer HAT or AH sensitivity. This will make possible later fusion to give an immortalized tissue-specific cell, and subsequent use of the fusion product for protein secretion, without a contribution of immunoglobulin secretion from the fusion partner. The mutagenized cells are also selected for retention of HLA antigen expression on the cell surface.

Example I below describes the preparation of an immortalizing hybridoma which is generally useful in producing primate triomas according to the invention. The cell line, designated SBC-H20, has the selected characteristics noted in the example. The cell line was deposited with the ATCC on or about 13 December 1983 and given the designation ATCC HB 8464.

F. Immortalized Human or Non-Human Primate Cell Line Human or non-human tissue cell from above are fused under conditions like those outlined in Section IIC. Typically, the tissue cells are mixed with the immortalizing hybridoma cells at a ratio of between about 1:1 to 1:5, and preferably about 1:3. Here it is noted that cell mixtures containing greater numbers of the tissue cells, such as in a ratio of 1:1 and 5:1 are feasible. However, since the tissue cells are usually more difficult to obtain, it is generally preferred to use an excess of the immortalizing cells in the cell fusion mixture. The cell mixture is freed of serum by washing, and resuspended in polyethylene glycol to promote cell fusion. After a suitable incubation period, the cells are washed, resuspended in culture medium, and plated on miαrotiter wells. Since the unfused tissue-specific cell is unable to grow in culture, the selection medium can simply be the HAT or AH medium which will discriminate against unfused immortalizing hybridoma cells. The resulting fusion products are thus easily selected on the basis of the presence of colonies of growing cells, as seen by light microscopy.

G. Tissue-Specific Products Where the immortalized cells are selected for the ability to produce a selected tissue-specific product, the microtiter plate cells or medium is assayed for the presence of that product. This can be done, for example in the case of secretory proteins, by means of a solid phase reporter-linked immunoassay. In the usual solid-phase assay method, a solid surface coated with antibodies specific against the selected protein is reacted with the cell culture supernatant, binding supernatant proteins immunospecifically to the support surface. Because of the close structural relationship between human and primate (particularly chimpanzee) secretory proteins, antibodies which are specific against selected human secretory proteins, such as human serum albumin (HSA) may also be used to screen for corresponding non-human secretory proteins.

Purified goat or rabbit serum antibodies against a variety of human secretory proteins and peptide hormone are commercially available. Also mouse monoclonal antibodies against many human secretory proteins are available.

Where a protein-specific antibody is not commercially available, the antibody can be produced by conventional serum antibody techniques for raising antibodies in rabbits or goats. These procedures generally require a source of purified or nearly purified protein for animal inoculation. Procedures for purifying many human secretory products have been reported, as referenced below. Alternatively, conventional monoclonal antibody techniques may be used for generating antibodies against selected human products.

After reacting the cell medium with the support-bound antibodies, to bind antibody-specfic secretory protein, the support is washed to remove non-specifically bound protein and then reacted with a soluble, reporter-labeled antibody which is also specific against the secretory protein of interest. The reporter label on the soluble antibody is a fluorophore. chromophore, enzyme, or radioisotope label. Techniques for labeling antibodies with reporter molecules are well known, and typically involve the use of a bifunctional coupling reagent, such as a diN-hydroxysuccinimide, or a suitable carboxyl or amine activating agent, such as a soluble carbodiimide. to couple one or more reporter molecules to the antibody. Such methods are well known to those in the field. The labeled antibody may be specific against the same or a different antigenic determinant as the support-bound antibody, but in any case, must be able to bind to the selected protein, with such bound immunospecifically to the first antibody carried on the solid support. The presence of the selected protein in the cell medium is confirmed by the presence of label on the washed support. Example III below illustrates an enzyme-linked immunoassay (ELISA) for detecting human serum albumin (HSA), complement C₃ (C₃), and human fibronectin (HFN) secreted by immortalized liver. Example VI below illustrates the use of an immunoassay employing an anti-factor VIII antibody for determination of factor VIII secretion from immortalized liver cells.

Alternatively, where antibodies specific against a selected protein are not readily available, standard biochemical techniques for detecting fractionated proteins may be used to identify the presence of a given protein in cell medium. By way of example, the cell medium can be fractionated by gel electrophoresis and examined for the presence or absence of a protein-staining band known to be associated with the selected protein. The presence of certain functional proteins or peptides may also be identified by protein activity assays, such as the ability of the protein to produce a specific biological effect. As an example, the presence of one or more complement components in cell medium can be assayed by the ability of the medium to effect cell mediated lysis in the presence of a complement fraction which is deficient in the complement factor of interest, as illustrated in Example VI.

After identifying one or more viable hybrid cell colonies which secrete the protein of interest, the cells can be subcloned to ensure monoclonality. Typically, subcloning is done by a limiting dilution technique in which parent hybrid cells are diluted to produce no more than about 1 cell per well, and the cells replated in multi-well microtiter plates. Clonal populations which show the desired protein secretory property are then isolated. Subcloning techniques can also be used to select for increased production of a given protein, or for production of certain tissue-specific proteins only. To select for high producing cell strains, the subcloned cells are examined for high levels of secreted protein in the culture medium. The identified high-producer cells can be taken through one or more additional rounds of subcloning to further select cells which are high in secretory activity. As part of the screening process, the cells can be additional examined for the presence of other secreted proteins, and cell selection can be made in part on the basis of decreased levels of these unwanted proteins. The latter selection procedure would be useful particularly to eliminate proteins which would be difficult to purify away from the secreted protein of

III. Human or Non-Human Primate Cell Products

Experiments conducted in support of the present invention, and reported particularly in Example IV, demonstrate that human non-lymphocytic tissue cells, such as human liver cells, can be immortalized so as to be capable of continued replication in culture over a period of at least several months, and at the same time, retain their normal cell-specific secretory functions. The results are consistent with earlier studies by the inventors and their colleagues showing that the mouse/human fusion partner described herein is capable of immortalizing human and non-human primate lymphocytes, to produce cells which are stable in culture for periods of up to several years, with little diminishment in secretory activity. For both lymphocytic and non-lymphocytic cell types, the fusion process is characterized by a high efficiency of successful hybrids, and of these, a high percentage of hybrid cells which are active and stable in secreting tissue-specific proteins.

The high efficiency of cell fusion, for both lymphocytic and non-lymphocytic cells, appears to be related to the initial selection, in forming the fusion partner, of mouse/human hybrids which are stable for human chromosome retention, as evidenced by continued immunoglobulin secretion and HLA production. That is. the fusion partner is preselected for a stable arrangement of human and mouse chromosomes, which apparently favors chromosome stability in the trioma fusion product. The ability of the fusion partner to form stable fusion products with such diverse cells as human lymphocytes, chimp lymphocytes, and human liver cells indicates the general applicability of the method to human cell immortalization.

Another significant feature of the fusion process, as it applies to the present invention, is the high percentage of cells which secrete one or more selected tissue-specific products. From the data presented in Example III, it is seen that successful liver cell hybrids contain a high percentage which secrete HSA (between about (18-28 %), complement C₃ (between about 28-40%), HFN (between about 20-40%), and all three liver secretory proteins (between about 10-15 %). These proteins were assayed as representative liver secretory proteins only, and the ability of many cells to secrete all three proteins indicates that a complete range of tissue-specific secretory products are produced in at least some significant portion of the successful fusion hybrids.

Immortalized liver cells, formed in accordance with the methods of Example IV. secrete several proteins and peptides which would be valuable to produce in quantity by simple cell culture methods. These secretory products include, in addition to HSA and complement C₃:

1. Factor VIIIC. a blood clotting factor which, together with von Willebrand factor (vWF), forms the factor VIII complex. Factor VIIIC is an important regulatory protein in the blood cell coagulation cascade, and is the protein deficiency in hemophilia A (Ginsburg). As described in Example VI, successful fusion products of adult liver cells with immortalizing hybridoma cells yielded a high percentage of cells which secrete factor VIII. In the case of PEG fusions, about 12.8% of the fusion products were positive for factor VIII, and in the case of fusion by electrofusion, about 10.5%. Protein isolation from culture can be performed by known methods for purifying the factor from human serum (Rotblatt).

2. Factor IX, the blood clotting factor whose deficiency is responsible for hemophilia B. The factor can be isolated by published methods for obtaining the factor in isolated form from human serum (Liebman).

3. α-1 antitrypsin, a protein which is implicated in hereditary emphysema . The protein can be isolated from cell culture supernatant according to published methods (Carrell).

4. Hemopexin, a protein factor involved in hemolytic anemia. Known methods for isolating the protein from human serum (Takahashi), are applicable to protein isolation from culture fluid. 5. Fibronectin, a widely studied glycoprotein that is involved in a number of binding reactions, including the binding of collagenous and glycosaminoglycan constituents of connective tissue to actin and DNA (Hynes). The protein has been proposed for use in replacement therapy in septic patients.

Plasma fibronectin, the form secreted by the liver, can be isolated by described procedures (Sekiguchi).

6. C-reactive protein, an anti-inflammatory agent and complement action initiator. The protein would be valuable in the treatment of .............. , and can be isolated from a culture supernatant by known methods (Oleivera).

7. Transcortin, a corticosteroid binding agent whose isolation from cell medium would follow published procedures (Wolf). 8. Synexin I and 2, both cell fusogen and calcium binding agents. The proteins are purified from a culture serum by known methods (Cruetz, Udenwald). Immortalized pancreatic islet cells, whose isolation and fusion for immortalization are described in Example VIII, can be used for production of human insulin in culture. The method has a number of advantages over current recombinant methods for making insulin in that the insulin is secreted in processed, mature form without need for extracellular enzymatic, digestion of the hormone, the immortalized cells can be selected for high insulin production, and no recombinant cloning and selection techniques are required. Examples VIII and IX describe methods for immortalizing pancreatic cells and obtaining secreted insulin.

Hypothalamic and anterior pituitary cells are also sources of a number of important human proteins. Hypothalamic cells secrete antidiuretic hormone (ADH, vasopressin. and oxytocin, all of which can be isolated by known techniques. Among the important secretory hormones produced by anterior pituitary cells are leutinizing hormone (LH), growth hormone (GH) corticotropin (ACTH) and follicle-stimulating hormone (FSH). Examples X and XI below describe procedures for isolating and immortalizing pituitary cells, selecting stable culture cells which produce and secrete GH in cell culture. The method enjoys the same advantages over recombinant methods for making pituitary hormones. In addition, the method allows for a several hormones to be made and secreted in a single culture system.

The following examples illustrate various aspects of the invention, but are in no way intended to limit the scope thereof. Example I Preparation of Immortalizing Hybridoma SBC-H20 Mouse myeloma cell line SP20/08A2 was obtained for use as the immortalizing partner from Frank Fitch, University of Chicago. This cell line is freely available and can be used without restriction. Other mouse myeloma lines are also readily available. Human peripheral B lymphocytes were isolated from the heparinized plasma of a normal human donor by Ficoll-Hypaque gradient as described in reference .11. The peripheral B lymphocytes B and myeloma cells were mixed at a 1:1 ratio, washed once in RPMI 1640 medium (Gibco). and pelleted at 250 x g for 10 min. The pellet was gently resuspended in 1 ml of RPMI with 40-45% (V/V) polyethylene glycol solution, MW 1430-1570 (BDH Chemicals, Poole, England) which was pre-warmed to 37°C. After two min at room temperature, the cell suspension was diluted to 6 ml with RPMI, centrifuged at 500 x g for 3 min, and, beginning 8 min from the onset of fusion, the cell pellet was washed with RPMI containing 10% fetal calf serum (FCS). The pelleted cells were plated in multi-well trays using suitable dilutions to obtain individual clones. The colonies were grown on AH selection medium containing 2 μg/ml azaserine and 100 μM hypoxanthine, and successful clones were assayed for immunoglobulin production and for HLA surface proteins using the assay methods described in reference 6.

A hybrid clone which showed stable immunoglobulin production for 6 months, and which was consistently producing HLA surface protein, was selected. This clone was placed in Iscove's medium (IMDM) (Gibco) containing 10% FCS, 2 mM glutamine, 100 unites penicillin, 100 mg streptomycin per ml, as well as the mutagen 6-thioguanine (Sigma, St. Louis, MO). The concentration of 6-thioguanine was progressively increased to 2 x 10 ^-5 M ouabam over a period of approximately 30 days. The resultant mutant hybrids were subcloned, and the colonies tested for immunoglobulin secretion. A non-secreting subclone which was HAT/AH sensitive, resistant to 10 ouabain, and which retained the ability to produce HLA surface antigen was selected. A sample of this cell line which is designated SBC-H20 was deposited with the ATCC and has the deposit identifying no. ATCC HB 8464. The characteristics of this murine-human hybridoma include: sensitivity to HAT and AH media, resistance to ouabain (Sigma) to a concentration of 10 ^-6 M, non-secretion of immunoglobulins, human chromosomal stability over time, and production of HLA surface protein.

Example II Isolation of Human Hepatocytes Human fetal liver tissue was obtained from a therapeutic abortus and placed in Iscove's Medium (IMDM) with 20% fetal calf serum (FCS). The tissue was dispersed mechanically by several gentle strokes in aground glass homogenizer with a single cell clearance. The cells were washed three times, with centrifugation, in IMDM medium without FCS.

Example III

Preparing Immortalized Liver Cells Isolated hepatocytes from Example II were suspended to a final cell concentration of about 3 x 10⁷ cells/ml in IMDM. The isolated cells were mixed with the hybridoma cell line SBC-H20 (Example I), at a cell ratio of about 1:3. The cells were washed in IMDM without serum and pelleted. The pellet were pelleted at 200 x g for 10 minutes and resuspended gently in 1 ml of 55% IMDM: 45% polyethylene glycol (v/v) MW 1430-1570 (BDH Chemicals, Poole, England) which was prewarmed to 37°C. The fused cells were resuspended in IMDM containing 10% FCS and 100 urn hypoxanthine. 19 urn thymidine (HT medium) and plated in rαicrotiter wells at

1 x 10⁵ cells/well and in 24 well trays, at 1 x 10⁶ cells/well. Cultures were grown in a humidified incubator at 37 C in 6% CO₂. After 24 hours, the medium was changed to the selection medium consisting of HT medium with 800 nm aminopterin (HAT medium). The HAT selection medium was used for 14 days prior to switching to HT medium, at which point unfused SBC-H20 cells and unfused hepatocytes were 100% non-viable, as evidenced by trypsin blue inclusion.

Hybrid cells formed by fusion. of the SBC-H20 fusion partner with the hepatocytes were observed in 49 of the 120 microtiter wells (Group A) and 45 of the 48 wells in the 24-well trays (Group B), as evidenced by the presence of colonies which were visible by light microscopy. Some of the wells or trays contained two or more distinct colonies.

The wells or trays containing visible colonies were tested for ability to secrete human serum albumin (HSA), complement C₃ (C₃), and fibronectin (HBN), as evidenced by the presence of one or more of these liver secretory proteins in the culture medium. The enzyme-linked immunoassay used to detect the presence of the proteins is described in Example IV below. The results obtained for the 45 fused cell colonies in Group A. and the 49 fused cells in Group B are shown in Table I. As seen, several of the fused cell lines were active in secretion of one or more of the liver proteins. and 5 colonies in Group A and 8 colonies in Group B secreted all three proteins.

From these two groups, selected hybridomas were cloned by limiting dilution to ensure monoclonality. Multiple monoclonal cell lines were isolated which continued to secrete HSA. complement C . and HBN.

Table 1

Parent CompleFibroAll three hybrids* Number Albumin ment nectin factors

Group A 45/48 8 17 20 5

Group B 49/120 14 20 9 8

*Supernatants from hybrids cultured in 24-well trays (Group A) and in microtiter trays (Group B).

Example IV Identification of Liver Cell Products

Selected hybridomas from the Group A and Group B colonies in Example III were subcloned by limiting dilution to insure monoclonality, and reassayed for parenchymal secretory proteins. Four parent hybrid cell lines, designated B3, D4, D5, and D8, were subcloned, and each parent gave multiple monoclonal sublines, as indicated in Table II below. Each of the subclonal lines was assayed for the cell secretion of HSA, C₃ and HFN by the following enzyme-linked immunoassay.

Goat anti-human antibodies specific against human albumin, complement C₃, and fibronectin were obtained from Cappell Labs (Malvern, PA) Microtiter trays were coated with one of the three different anti-sera overnight with 50 nl/well of a 1:1000 dilution (lug/ml)at 4 C in a moist chamber. The following day, 150 nl of a 0.2 % gelatin solution were added to each well for one hour at room temperature, to saturate non-specific binding sites in the wells. The trays were then washed four times with cold PBS /0.05% Tween-20.

Analyte and control samples (50 ul) were added to the trays and incubated for 1 hour at room temperature. The analyte samples were undiluted medium obtained from the growing monoclonal cultures. The control samples contained human plasma diluted 1:20,000 (complement control), purified human fibronectin at 5 ug/ml, and human serum albumin, at 25 ug/ml. After incubation, the trays were washed three times with the above Tween solution. Peroxidase-conjugated goat antibodies against human C₃, albumin, or fibronectin were purchased from Cappell Labs. The peroxidase-conjugated antibody (50ul), at a final dilution of about lug/ml antibody protein, was added to each of the wells, and incubated for one hour at room temperature, to bind the conjugated antibody to analyte secretory protein in sandwich fashion. After washing three time with the above Tween solution, substrate mixture was added and the peroxidase enzyme reactions were allowed to develop for about thirty minutes at room temperature before arresting the reaction by the addition of 100 microliters of 10% SDS. The substrate mixture was prepared by mixing 7.2 ml of a0.1M citrate/HCL buffer, pH 4.2, 4.8 ml of 2.5 mM ABTS. and 80 ul of a 1 M H₂O₂ solution. The mixture was prepared about 30 minutes prior to addition to the washed trays. The assay mixture/SDS solutions were read 415 or 405 nm. As seen from the results in Table II, each of the parent hybrids B3, D4, and D8, (all HFN secretors) yielded multiple subclones which were also competent in HFN secretion only. Similarly parent hybrid cell line D4, which was active in producing all three liver proteins, yielded multiple subclones, each of which was competent in secreting all three proteins assayed.

Table 2

Parent hybrid Number Albumin Complement Fibronectin

B3 3 0 0 3

D4 9 2 3 9

D5 8 0 0 8

D8 3 0 0 3

Example V

Product:ion of Fibronectin in Culture

Parent hybrids from Example III which showed HFN secretion, but neither ΗSA or C₃ secretion, are further subcloned and selected for high HFN producers. Each HFN parent hybrid is subcloned by limiting dilution on 48 well-microtiter plates, and wells containing colonies are assayed for concentration of HFN in the cultured supernatants. as described in Example IV. Each of ten high producing subclones identified in this manner is further subcloned by limiting dilution on a 48 well microtiter plate, and wells showing cell colonies are again assayed for HFN levels. Five subclonal cell lines which show the highest levels of HFN in the culture supernatant are combined and grown to confluence in 50 ml flasks containing IMDM medium with 10% FCS. The cultures are then suspended in 250 ml IMDM medium containing 10% FCS in 850 cc roller bottles. After 7 days of incubation, the medium is harvested by low-speed centrifugation. to remove the cells, and the HFN is purified from the medium, according to published methods (Sekiguchi). The cells removed by centrifugation are resuspended in 250 ml of the above IMDM/FCS/HEPES growth medium, and incubated as above in roller bottles for two days, then treated as above to harvest cell medium. The cell resuspension. growth, and medium isolation steps may be carried out repeatedly over a several month period.

Example VI Factor VIII Production in Culture

A. Human Fetal Cells

Human fetal liver cells were obtained as in Example II, and were fused with SBC-H20 cells (Example I) at a cell ratio of about 1:4 lever cells:fusion cells. Fusion was carried out with polyethylene glycol, and successful fusion products selected first in HAT medium, then for the ability to produce liver secretory products substantially as described in Example III. Fusion products which produced one or more liver secretory products were also tested for secretion of factor VIII, using a modification of the Coatest™ enzyme for factor VIII obtained commercially from Helena Labs. Several of the fused cells showed factor VIII secretion by this assay. Factor VIII secretion was also examined by an an ELISA immunoassay, using an anti-factor VIII antibody obtained from Symbiotics, Inc. In this procedure, microtiter plates were coated with the anti-factor VIII antibody. Sample cell supernatants and coagulation ARP obtained from Helena Labs was added to the wells, incubated for 1/2 hour, then washed away. To the wells was next added biotinylated anti-factor VIII prepared by biotinylating the above antibody according to conventional methods. After several washings to remove non-specifically bound material, avidin conjugated alkaline phosphate was added to the wells. The enzyme was assayed in the presence of the substrate p-nitrophenaol phosphate according to standard assay procedures. The assay also gave positive factor VIII identification in the fused cells which were positive for factor VIII by above enzyme test.

B. Adult Liver Cells Adult liver fusion cells were obtained as in

Example VII below, and tested for factor VIII secretion by the ELISA immunoassay test described in part A above. Successful fusion products, as selected on HAT medium, produced factor VIII in about 12.8% of of cells fused by PEG and in about 10.5% of the cells produced by electrofusion.

Example VII Immortalized Adult Liver Cells Human adult liver tissue was obtained by biopsy, and the tissue was mechanically dispersed as in Example II. The cells were mixed, after washing, with SBC-H20 cells, at a cell ratio of about 1:3, and fused in the presence of polyethylene glycol as in Example III. Cell selection in the presence of HAT, and further cell selection for liver secretory proteins was carried out as in Example VIII. Immortalized cells capable of secreting albumin, complement C₃, fibronectin, and or factor VIII were observed. Successful adult liver cell fusion products which secreted one or more of the above secretory proteins were also obtained by cell fusion by electrofusion, carried out substantially according to the manufacturer's instructions (Amsco, Erie, PA).

Example VIII

Preparation and Fusion of Pancreatic Islet Cells

Human fetal pancreatic tissue is obtained from a therapeutic abortus and placed in Iscove's Medium

(IMDM) with 20% fetal calf serum (FCS). The tissue is dispersed mechanically by several gentle strokes in a ground glass homogenizer with a single cell clearance.

The cells are washed three times, with centrifugation, in IMDM medium without FCS.

Alternatively, a suspension of free islet cells cells are prepared from fetal human tissue by a modified enzyme dispersion technique (Gordon). Briefly, the tissue is perfused with a collagenase solution, minced with scissors after a selected period of exposure to collagenase, suspended in enzyme-free perfusate to stop the collagenase reaction, and filtered through sterile nylon mesh cloth.

The isolated cells are mixed with the fusion partner of Example I, at a cell ratio of about 3:1, and the cells are fused as described in Example III.

The fused cells are resuspended in IMDM containing 10%

FCS and 100 um hypoxanthine, 19 um thymidine (HT medium) and plated in microtiter wells at 1 x 10⁵ cells/well and in 24 well trays, at 1 x 10⁶ cells/well. Cultures are grown in a humidified incubator at 37 C in 6%

CO₂. After 24 hours, the medium is changed to the selection medium consisting of HT medium with 800 nm aminopterin (HAT medium). The HAT selection medium is used for 14 days prior to switching to HT medium, at which unfused SBC-H20 cells and unfused islet cells are 100% non-viable, as evidenced by trypsin blue inclusion. Hybrid cells formed by fusion of the fusion partner with the islet cells are detected by the presence of colonies which are visible by light microscopy.

The wells containing visible colonies are tested for ability to secrete human insulin using an enzyme-linked immunoassay to the type described in Example IV. to detect the presence of insulin in the cell medium. Briefly, goat anti-human insulin antibodies are obtained from a commercial source. The antibodies are absorbed to microtiter plate wells, the wells incubated with gelatin to saturate non-specific binding sites, and the cell medium incubated in the wells, to bind insulin immunospecifically to the surface-attached antibodies. After washing the wells, goat anti-human insulin antibodies which have been derivatized with peroxidase are added, and the amount of peroxidase-linked enzyme which has bound to the wells determined colorimetrically, as above. Parent hybrids which give insulin secretion are further subcloned to ensure monoclonality of the selected cell lines.

Example IX Insulin Production in Culture The insulin-secreting cell lines from Example VIII may be further subcloned and selected for high producers of insulin, substantially as described in Example V above. Subclonal cell lines which show the highest levels of insulin in the culture supernatant are combined and grown to confluence in 50 ml flasks containing IMDM medium with 20% FCS. The cultures are then suspended in 250 ml IMDM medium containing 10% FCS in 850 cσ roller bottles. After 7 days of incubation, the medium is harvested by low-speed centrifugation. to remove the fused cells, and the insulin is purified from the medium, in accordance with known methods.

The cells removed by centrifugation are resuspended in 250 ml of the above IMDM/FCS/HEPES growth medium, and incubated as above in roller bottles for two days, then treated as above to harvest cell medium.

Example X

Preparation and Fusion of Pituitary Cells

Human fetal pituitary tissue is obtained from a therapeutic abortus and placed in Iscove's Medium (IMDM) with 20% fetal calf serum (FCS). The tissue is dispersed mechanically by several gentle strokes in a ground glass homogenizer with a single cell clearance. The cells are washed three times, with centrifugation, in IMDM medium without FCS.

Alternatively, a suspension of pituitary cells are prepared from fetal human tissue by a modified enzyme dispersion technique (Ben-Jonathon). Briefly, the tissue is perfused with a collagenase solution, minced with scissors after a selected period of exposure to collagenase, suspended in enzyme-free perfusate to stop the collagenase reaction, and filtered through sterile nylon mesh cloth.

The isolated pituitary cells are fused with the mouse/human fusion partner of Example II, substantially by the procedure described in Example III.

The fused cells are resuspended in IMDM containing 10%

FCS and 100 um hypoxanthine, 19 um thymidine (HT medium) and plated in microtiter wells at 1 x 10⁵ cells/well and in 24 well trays, at 1 x 10⁶ cells/well. Cultures are grown in a humidified incubator at 37 C in 6% CO₂. After 24 hours, the medium is changed to the selection medium consisting of HT medium with 800 nm aminopterin (HAT medium). The HAT selection medium is used for 14 days prior to switching to HT medium, at which unfused SBC-H20 cells and unfused islet cells are 100% non-viable, as evidenced by trypsin blue inclusion. Hybrid cells formed by fusion of the fusion partner with the pituitary cells are detected by the presence of colonies which are visible by light microscopy.

The wells containing visible colonies are tested for ability to secrete human growth hormone

(hGH), using an enzyme-linked immunoassay to the type described in Example IV, to detect the presence of hGH in the cell medium. Briefly, goat anti-hGH antibodies are substituted for goat anti-HSA, C₃, or HFN antibodies in Example IV and the goat anti-human insulin antibodies in Example VII, in a sandwich-type immunoassay to detect the presence of hGH in the cell Parent hybrids which give hGH secretion are further subcloned to ensure monoclonality of the selected cell lines.

Example XI Production of hGH in Culture The hGH-secreting cell lines from Example X may be further subcloned and selected for high producers of insulin, substantially as described in Example V above. Subclonal cell lines which show the highest levels of hGH in the culture supernatant are combined and grown to confluence in 50 ml flasks containing IMDM medium with 20% FCS. The cultures are then suspended in 250 ml IMDM medium containing 10 % FCS in 850 cc roller bottles. After 7 days of incubation, the medium is harvested by low-speed centrifugation, to remove the fused cells, and the hGH is purified from the medium, according to published methods. The cells may be be repeatedly supplied with fresh growth medium, and the growth medium harvested after a selected growth period, as above, to provide a renewed supply of medium containing hGH over a seven-month period.

415 or 405 nm with a Titertech Multiscan Reader.

While exemplary embodiments and uses have been described herein, it will be appreciated that the invention encompasses a broad range of human and non-human immortalized tissue cells which can be used in stable cell culture for production of tissue-specific human or non-human primate products.

Claims

IT IS CLAIMED :

1. An immortalized normal human or non-human primate tissue cell line which is capable of stably producing in culture a selected product of a non-lymphocytic tissue cell from a human or non-human primate source, said cell line comprising a fusion partner produced by fusing mouse myeloma cells and human B-lymphocytes, and selecting fusiqn products which show stable human chromosome retention, as evidenced by continued HLA surface antigen production in culture, and fused with the fusion partner, an isolated cell obtained from a human or non-human primate source which produces the selected product in vivo.

2. The cell line of claim 1, wherein the fusion partner is produced by fusing mouse myeloma cells and human B-lymphocγtes, selecting fusion products which show immunoglobulin secretion and HLA surface antigen production in culture, treating the selected fusion products with a mutagen, and selecting mutagenized products which retain the ability to produce HLA surface antigen, show no immunoglobulin secretion, and are unable to survive in a growth medium which allows growth of a successful product formed by fusing the fusion partner with such a human cell.

3. The cell line of claim 2, wherein the fusion partner has the characteristics of ATCC NO.HB

8464.

4. The cell line of claim 1, which is selected for increased cell production of a selected tissue-specific secretory product normally produced in vivo by the human cell.

5. The cell line of claim 1, wherein the human cell is selected from the group consisting of hepatocytes, pancreatic islet cells, and pituitary cells.

6. The cell line of claim 5, wherein the human cell. is a hepatocyte, and which is capable of secreting one or more proteins selected from the group consisting of factor VIII, α-1 antitrypsin, α-1 acid glycoprotein, hemopexin, fibronectin, C-reactive protein, transcortin, synexin I, synexin-II, and serum albumin.

7. The cell line of claim 6, wherein the secreted proteins are selected from the group consisting of serum albumin. C-reactive protein, fibronectin, and factor VIII.

8. A method of producing an immortalized cell line which is capable of stably producing in culture a selected product of a non-lymphocytic tissue cell from a human or non-human primate source, comprising providing a fusion partner produced by fusing mouse myeloma cells and human B-lymphocytes, selecting fusion products which show stable HLA surface antigen production in culture, treating the selected fusion products with a mutagen. and selecting mutagenized fusion products which retain the ability to produce HLA surface antigen, and are unable to survive in a growth medium which allows growth of a successful product formed by fusing the fusion partner with such a tissue cell. obtaining non-lymphocytic cells isolated from a selected human or non-human primate tissue which produces the selected product in vivo, fusing the fusion partner with the non-lymphocytic cells obtained, and selecting fusion products which are capable of producing the selected cell product.

9. The method of claim 8, wherein the non-lymphocytic cells are human hepatocytes, pancreatic islet cells, and anterior and posterior pituitary cells.

10. The method of claim 8, for use in producing tissue-specific human secretory proteins in culture, wherein the immortalized cell line is grown in culture , and the secretory protein is isolated from the medium used in culturing the cell line.

11. The method of claim 10, for use in producing liver secretory proteins selected from the group consisting of factor VIII, α-1 antitrypsin, α-1 acid glycoprotein. hemopexin. fibronectin, C-reactive protein, transcortin, synexin I, synexin-II, and serum albumin, wherein the human non-lymphocytic cells are hepatocytes.

12. The method of claim 10, for use in producing insulin, wherein the human, non-lymphocytic cells are pancreatic islei cells.

13. The method of claim 10, for use in producing prolactin, growth hormone, or ACTH, wherein the non-lymphocytic cells are anterior pituitary cells.

14. The method of claim 10. wherein the said selecting further includes selecting fusion products which show increased production of such secretory proteins.