WO1999064560A1 - Methods for monitoring the stability of protein or cellular product secretion by cells in a fermenter culture - Google Patents

Abstract

The invention features a method for monitoring the stability of a protein or other cellular product secreted by cells in a fermenter culture. The method involves, harvesting from a fermenter culture a first sample of cells that secrete a protein or cellular product and incorporating the cells from the first sample into gel microdrops (GMDs), and measuring secretion of the protein or cellular product by cells in one or more GMD samples over time. The protein secreted by the cells may be, for example, an antibody, cytokine, or growth factor. The cellular product secreted by the cells may be, for example, an antibiotic, amino acid, vitamin, or carbohydrate.

Description

METHODS FOR MONITORING THE STABILITY OF PROTEIN OR CELLULAR PRODUCT SECRETION BY CELLS IN A FERMENTER

CULTURE

Background of the Invention

An important aspect of the production of proteins or other cellular products by cells is the stability of gene expression under large scale culture conditions. This is particularly important for recombinant protein production, where expression of a recombinant vector can vary from cell to cell and over time. Moreover, this is also important in the stable expression of other cellular products that are the result of a biosynthetic pathway that has been recombinantly introduced into a given host cell. The ideal production process has a robust level of expression which can be maintained through many cell generations without the selection pressure of antibiotics or other drugs. In practice, cell lines producing recombinant proteins can often lose their high levels of expression even when maintained under selection pressure. Therefore, the evaluation of a fermentation process typically includes monitoring for signs of expression instability.

A number of parameters have been used to monitor the condition of cells in fermenters. Examples include monitoring of glucose uptake, lactate accumulation, and oxygen consumption, each of which monitor the metabolism of the culture as a whole. Periodic viable cell counts can be performed to determine the growth rate of the population. Currently, periodic measurements of the specific productivity of a cell line are performed by assaying the amount of protein produced and the number of cells in the culture. Specific productivity, expressed as picograms of recombinant protein or an amount of a given cellular product produced per cell per day, thus can detect only the average loss of expression of the total population. Furthermore, this method is unable to distinguish between a drop in productivity due to fermenter conditions which are physiologically unfavorable to the total cell population versus a drop due to the development of a subpopulation of cells of lower productivity. Moreover, since all of these methods measure the condition of the total cell population, only changes involving a significant proportion of the culture are detectable. Summary of the Invention

This invention provides a method for monitoring the stability of protein or cellular product secretion that allows for the testing of individual cell productivity within a large population of cells, such as cells within a fermenter culture. The method of the invention allows for a more discriminating evaluation of the productivity of cells from fermenters cultures, since the method measures productivity of individual cells in the population. Moreover, the method of the invention provides rapid results and is relatively simple to perform. The method of the invention generally involves harvesting a sample of the cells from a fermenter culture, incorporating cells from the sample into gel microdrops (GMDs), performing a GMD secretion assay to assess protein or cellular product secretion by individual cells in the sample and then repeating this procedure with a second sample of cells, harvested at a later time than the first sample of cells, to thereby monitor the stability of protein or cellular product secretion by the culture. The procedure can be repeated with additional cell samples, harvested over time, to continue monitoring the stability of protein or small molecule secretion by the culture.

Accordingly, in one aspect, the invention provides a method for monitoring stability of protein secretion by cells in a fermenter culture, comprising: • (a) harvesting from a fermenter culture a first sample of cells that secrete a protein;

(b) incorporating cells from the first sample into gel microdrops (GMDs) to form a GMD sample;

(c) measuring secretion of the protein by cells of the GMD sample using a GMD secretion assay; and

(d) repeating steps (a) through (c) on a second sample of cells, wherein the second sample of cells is harvested from the fermenter culture at a time later than when the first sample of cells was harvested to thereby monitor stability of protein secretion by cells in the fermenter culture. The method can further comprise step (e): repeating steps (a) through (c) on a third sample of cell/s, wherein the third sample of cells is harvested from the fermenter culture at a time later than when the second sample of cells was harvested. The method can still further comprise step (f): repeating steps (a) through (c) on a fourth sample of cells, wherein the fourth sample of cells is harvested from the fermenter culture at a time later than when the third sample of cells was harvested. The method can still further comprise step (g): repeating steps (a) through (c) on a fifth sample of cells, wherein the fifth sample of cells is harvested from the fermenter culture at a time later than when the fourth sample of cells was harvested. The method of the invention can be used to monitor the stability of protein secretion by any type of cells that is suitable for recombinant protein production, including both eukaryotic and prokaryotic cells. In a preferred embodiment, the method is used to monitor a culture of mammalian cells, such as CHO cells, COS cells, NS/O cells and the like. In another embodiment, the method is used to monitor a culture of yeast cells.

In a preferred embodiment, the protein whose production is monitored is an antibody secreted by the cells of the culture. In another preferred embodiment, the protein whose production is monitored is a protein encoded by an expression vector incorporated into the cells (i.e., a recombinant protein). In a particularly preferred embodiment, the protein whose production is monitored is an antibody encoded by at least one expression vector incorporated into the cells (i.e., a recombinant antibody).

The method of the invention allows for the monitoring of large scale fermenter cultures. In one embodiment, the fermenter culture comprises at least 10 liters of culture medium. In another embodiment, the fermenter culture comprises at least 50 liters of culture medium. In yet another embodiment, the fermenter culture comprises at least 100 liters of culture medium.

Monitoring of the culture is achieved by assaying at least two samples of cells, wherein the second sample is harvested at a later time than the first sample. In one embodiment, the second sample of cells is harvested at least one day later than the first sample of cells. In another embodiment, the second sample of cells is harvested at least two days later than the first sample of cells. In yet another embodiment, the second sample of cells is harvested at least three days later than the first sample of cells. The method can further comprise recovering a cell population with stable expression of the protein.

In a related second aspect, the invention provides a method for monitoring the stability of cellular product secretion by cells in a fermenter culture, comprising:

(a) harvesting from a fermenter culture a first sample of cells that secrete a cellular product;

(b) incorporating cells from the first sample into gel microdrops (GMDs) to form a GMD sample; (c) measuring secretion of the cellular product by cells of the GMD sample using a GMD secretion assay; and

(d) repeating steps (a) through (c) on a second sample of cells, wherein the second sample of cells is harvested from the fermenter culture at a time later than when the first sample of cells was harvested to thereby monitor the stability of cellular product secretion by cells in the fermenter culture.

The method can further comprise step (e): repeating steps (a) through (c) on a third sample of cells, wherein the third sample of cells is harvested from the fermenter culture at a time later than when the second sample of cells was harvested. The method can still further comprise step (f): repeating steps (a) through (c) on a fourth sample of cells, wherein the fourth sample of cells is harvested from the fermenter culture at a time later than when the third sample of cells was harvested. The method can still further comprise step (g): repeating steps (a) through (c) on a fifth sample of cells, wherein the fifth sample of cells is harvested from the fermenter culture at a time later than when the fourth sample of cells was harvested. The method of the invention can be used to monitor the stability of cellular product secretion by any type of cell/s that is suitable for recombinant cellular product production, including both eukaryotic and prokaryotic cells. In a preferred embodiment, the method is used to monitor a culture of mammalian cells, such as CHO cells, COS cells, NS/O cells and the like. In another embodiment, the method is used to monitor a culture of fungal cells, preferably, yeast cells. In yet another embodiment, the method is used to monitor a culture of bacterial cells In a preferred embodiment, the cellular product whose production is monitored is an antibiotic, preferably, a polyketide antibiotic, secreted by the cells of the culture. In another preferred embodiment, the cellular product is an amino acid, preferably, an essential amino acid, such as, for example, histidine, isoleucine, leucine, lysine, methionine, cysteine, phenylalanine, tyrosine, threonine, tryptophan, valine, or arginine. In even another related embodiment, the cellular product is a vitamin, preferably, a B vitamin (e.g., B2, B6, or B12), vitamin C, vitamin K, or riboflavin. In still another preferred embodiment, the cellular product is a carbohydrate, for example, a saccharide, or a polymer thereof. In another preferred embodiment, the cellular product whose production is monitored is a cellular product produced by an enzyme encoded by an expression vector incorporated into the cells. In a particularly preferred embodiment, the cellular product whose production is monitored is a antibiotic produced by an enzyme encoded by at least one expression vector incorporated into the cells. The method of the invention allows for the monitoring of large scale fermenter cultures. In one embodiment, the fermenter culture comprises at least 10 liters of culture medium. In another embodiment, the fermenter culture comprises at least 50 liters of culture medium. In yet another embodiment, the fermenter culture comprises at least 100 liters of culture medium. Monitoring of the culture is achieved by assaying at least two samples of cells, wherein the second sample is harvested at a later time than the first sample. In one embodiment, the second sample of cells is harvested at least one day later than the first sample of cells. In another embodiment, the second sample of cells is harvested at least two days later than the first sample of cells. In yet another embodiment, the second sample of cells is harvested at least three days later than the first sample of cells. The method can further comprise recovering a cell population with stable expression of the protein.

Other features and advantages of the invention will be apparent from the following detailed description and claims. Brief Description of the Drawings

Figures 1Λ-1D are FACS histograms of the gel microdrop secretion assays of samples of a fermentation of the D8/E cell line, secreting the recombinant antibody D2E7. Figure 1 A represents antibody secretion by the fermenter inoculum (AFI 615.75), Figure IB represents antibody secretion by the first harvest (AFF 610A), Figure IC represents antibody secretion by the fourth harvest (AFF 610D), and Figure ID represents antibody secretion by the eighth harvest (AFF 61 OH).

Figures 2A-2B are bar graphs depicting the gene copy numbers for the D2E7 antibody heavy chain (Figure 2A) or the D2E7 antibody light chain (Figure 2B) for cells harvested from the seed train and the fermenter of the D8/E cell culture. Samples 615 0.6, 615 5.0 and 615 75 were taken from the fermenter seed inoculum culture as it was expanded from 0.6 liters to 5 liters and then to 75 liters. Samples 610A, 610D and 61 OH were taken from the fermenter at the first, fourth and eight harvest.

Figures 3A-3B depict the antibody secretion profiles, as determined using the gel microdrop secretion assay, of cells from the end of the fermenter run 607 (Figure 3 A) or the fermenter run 702 (Figure 3B).

Figure 4 is a bar graph depicting the antibody productivity (expressed in picograms of antibody per cell in 24 hours) of high and low expressor subpopulations of the 702 fermenter run over a period of 4 weeks (for the low expressor subpopulation) or ten weeks (for the high expressor subpopulation). The low expressor subpopulation (702hLo) is represented by the bar graphs on the left, whereas the high expressor subpopulation (702hHi) is represented by the bar graphs on the right.

Figures 5Λ-5B are bar graphs depicting the gene copy numbers for the D2E7 antibody heavy chain (Figure 5A) or the D2E7 antibody light chain (Figure 5B) for cells from the final harvest of the 607 fermenter run (607H) or the 702 fermenter run (702H) or for their sorted subpopulations that produce low antibody levels (607H low and 702H low, respectively) or high antibody levels (607H high and 702H high, respectively). Detailed Description of the Invention

This invention pertains to methods for monitoring the stability of protein secretion (e.g., antibody secretion) from cells in fermenter cultures. Alternatively, the invention provides a method for monitoring the stability of cellular product secretion, i.e., a non-protein based product (e.g., an antibiotic, an amino acid, a vitamin, or a carbohydrate) from a cell in a fermenter. The methods of the invention allow for the testing of individual cell productivity within a large population of cells and thus allow for a more discriminating evaluation of the productivity of cells from fermenters cultures. The invention is based, at least in part, on the use of gel microdrop encapsulation and gel microdrop secretion assays to monitor protein or cellular product secretion. Although gel microdrop technology is known in the art, it has not heretofore been used to monitor the stability of protein or cellular product secretion over time by cells within fermenter cultures.

As demonstrated in the Examples, the appearance within a fermenter culture of a subpopulation of cells with lower productivity of, for example, a protein, was detected using the gel microdrop secretion assay, by assaying sequential harvests of the culture over time. Subsequent analysis of the culture by Southern blot analysis confirmed the presence of lower productivity cells that had lower copy numbers of the expression vector encoding the protein. The gel microdrop method of the invention enables one to detect the nature of this instability in a very short period of time (e.g. , within six hours), rather than relying on the Southern blot approach which takes several days and, furthermore, allows one to distinguish between factors which inhibit the metabolism of all cells in the culture (e.g., accumulation of waste products) from factors which may effect only a subset of cells (genetic instability). Moreover, the method of the invention can be used to recover a population with stable expression of the protein or cellular product of interest.

In order for the full scope of the invention to be clearly understood, the following definitions are provided. I. Definitions

The term "protein" is intended to include any selected polypeptide of interest that is capable of being expressed in a cell. Preferably, the protein is secreted and may be, for example, an antibody, a cytokine, a growth factor, or a hormone. The term "cellular product" is intended to include any non-protein based cellular product that a cell is capable of producing naturally or upon genetic manipulation. Preferably, the cellular product is an antibiotic, an amino acid (e.g., an essential amino acid), a vitamin, a carbohydrate, or any non-protein based molecule suitable for use as a pharmaceutical, nutrient, dietary supplement, or food additive (e.g., a flavoring, colorant, or preservative) that can be expressed from a cell. In a preferred embodiment, the antibiotic is a polyketide antibiotic.

The term "antibiotic" is intended to include any substance produced by a cell that can inhibit the growth of or destroy a microorganism.

The term "amino acid" is intended to include any member of a group of organic compounds marked by the presence of an amino group (NH₂) and a carboxyl group (COOH). The term is intended to include any amino acid found in nature (of which there are over 80) and preferably, those 20 amino acids necessary for protein synthesis, and thus, the growth of any living organism. In one preferred embodiment, the amino acid may be an "essential amino acid" and this term is intended to include any amino acid required by an animal, e.g., a human, that must be obtained from the diet.

The term "vitamin" is intended to include any group of organic substances other than proteins, carbohydrates, fats, minerals, and organic salts which are essential for normal metabolism, growth, and development of an organism, e.g., a. human. In a preferred embodiment, the vitamin is a B vitamin (e.g., Bl, B2, B6, B 12, or a combination thereof) vitamin C, vitamin K, or riboflavin.

The term "carbohydrate" is intended to include a group of chemical substances containing only carbon, oxygen, and hydrogen, e.g., sugars, and also polymers thereof, e.g., glycogen, starches, dextrins, celluloses, alginate, curdlan, levan, phosphomannan, poly-beta-hydroxybutyrate, scleroglucan, and xanthans (see also Table 5). The term "fermenter culture" is intended to refer to a large scale cell culture (e.g., greater than 10 liters, more preferably greater than 50 liters and even more preferably greater than 100 liters) that is used to generate proteins (e.g., secreted proteins) or cellular products (e.g., an antibiotic, an amino acid, a carbohydrate, or a vitamin), which are then harvested for further use.

The term "harvesting" (as in "harvesting from a fermenter culture") refers to removing or isolating a sample of cells from the fermenter culture.

The term "gel microdrops" (abbreviated "GMDs") refers to particles having a very small volume (e.g., 10"¹⁰ to 10"⁵ ml) and comprising at least one gel region, which provide a mechanical matrix capable of entrapping or surrounding (without necessarily contacting) a biological entity, such as an individual cell. Preferably, the GMD consists entirely of gel, in which case containment of the biological entity (e.g., cell) occurs by entrapment of the biological entity by the gel matrix. Examples of suitable polymer gels for encapsulation include agarose, alginate, agar, carrageenan, polyacrylamide, collagen, gelatine and fibrinogen, with agarose being most preferred. The diameter of the GMD typically is 0.2-1000 μm, more preferably 5-100 μm.

The term "incorporating cells into gel microdrops" refers to encapsulating cells within a gel microdrop, for example using procedures described herein.

The term "gel microdrop sample" refers to a sample of cells that have been incorporated (i.e., encapsulated) into gel microdrops.

The term "gel microdrop secretion assay" (or "GMD secretion assay") refers to an assay in which proteins or cellular products are secreted into the matrix of a GMD by a cell entrapped within the GMD, are labeled with a detectable label (e.g., a fluorescent label) and the GMD-associated label is detected (e.g., by flow cytometry). Preferably, the GMD secretion assay is performed quantitatively (i. e., the amount of GMD- associated label is quantitated).

The term "cell" is intended to include any cell that expresses a desirable protein or cellular product, i.e., non-protein based molecule (or a combination thereof), and can be cultivated in a fermenter. In one embodiment, the cell is a eukaryotic cell such as a mammalian cell. In another embodiment, the cell is a eukaryotic cell such as a fungal cell or an insect cell (that is amenable, for example, for baculovirus expression). In yet another embodiment, the eukaryotic cell is a yeast cell. In still another embodiment, the cell is a plant cell. In yet a further embodiment, the cell is bacterial cell. In preferred embodiments, the cell may be a particular strain of any of the foregoing cells selected for the expression of a particular product, such as, for example, a particular antibiotic. In another embodiment, the cell may be altered by genetic manipulation, e.g. , to incorporate a heterologous polynucleotide sequence/s, e.g., which can be transfected. In a preferred embodiment, such a polynucleotide sequence/s may encode a protein/s that is itself the desired product or, alternatively, may be an enzyme that is involved in a biosynthetic or catalytic pathway that liberates a desired cellular product, for example, an antibiotic, amino acid, vitamin, carbohydrate, etc. Accordingly, the cell may be transformed with one or more vectors comprising such sequences and the polynucleotide sequences may encode one or more genes. In addition, the term is intended to include progeny of the cell originally transfected. Techniques for culturing the above-mentioned cells are known in the art (see, e.g., Large-Scale Mammalian Cell Culture Technology, Lubiniecki, A., Ed., Marcel Dekker, Pub., (1990), Bacterial Cell Culture: Essential Data, Ball, A., John Wiley & Sons, (1997), Molecular and Cell Biology of Yeasts, Yarranton et al., Ed., Van Nostrand Reinhold, Pub., (1989); Yeast Physiology and Biotechnology, Walker, G., John Wiley & Sons, Pub., (1998); Baculovirus Expression Protocols, Richardson, C, Ed., Humana Press, Pub., (1998); Methods in Plant Molecular Biology: A Laboratory Course Manual, Maliga, P., Ed., C.S.H.L. Press, Pub., (1995); Current Protocols in Molecular Biology, eds. Ausubel et al, John Wiley & Sons (1992), Sambrook, J. et al, Molecular Cloning: A Laboratory Manual. 2nd, ed., Cold Spring Harbor Laboratory, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1989), and Bergey's Manual of Determinative Bacteriology, Kreig et al, Williams and Wilkins (1984)).

II. Methods of Use

The invention provides a method for monitoring stability of protein or cellular product secretion by cells in a fermenter culture, comprising: (a) harvesting from a fermenter culture a first sample of cells that secrete a protein or cellular product; (b) incorporating cells from the first sample into gel microdrops (GMDs) to form a GMD sample;

(c) measuring secretion of the protein or cellular product by cells of the GMD sample using a GMD secretion assay; and (d) repeating steps (a) through (c) on a second sample of cells, wherein the second sample of cells is harvested from the fermenter culture at a time later than when the first sample of cells was harvested to thereby monitor stability of protein or cellular product secretion by cells in the fermenter culture.

The Gel Microdrop (GMD) Composition

Various methods for incorporating cells into GMDs to form a GMD sample are known in the art (see e.g., U.S. Patent No. 4,399,219; U.S. Patent No. 4,401,755; U.S. Patent No. 4,643,968; U.S. Patent No. 4,647,536; U.S. Patent No. 4,916,060; U.S. Patent No. 4,959,301; U.S. Patent No. 5,055,390; U.S. Patent No. 5,225,332; Weaver, J.C. et al. (1984) Ann. N. Y. Acad. Sci. 434:363-372; Weaver, J.C. (1986) Biotech, and Bioeng. Symp. 17:185-195; Williams, G.B. et al. (1987) Ann. N. Y. Acad. Sci. 501 :350-353; Weaver, J.C. et al. (1988) Bio/Technology 6:1084-1089; Powell, K.T. and Weaver, J.C. (1990) Bio/Technology 8:333-337; Williams, G.B et al. (1990) J. Clin. Microbiol. 28:1002-1008; Weaver, J.C. et al. (1991) Bio/Technology 9:873-877; Kenney, J.S. et al. (1995) Bio/Technology 13:787790; Weaver, J.C. et al. (1997) Nature Medicine 3:583-585). Typically, GMDs are formed by adding a conventional cell suspension containing a large number (e.g., millions) of cells to liquefied gel (e.g., 37 °C molten agarose), dispersing in a non-aqueous medium (e.g., mineral oil) and transiently cooling to cause gelation. The resultant gel microdrops then can be removed from the non- aqueous medium and suspended in an aqueous medium. Using Poisson statistics as a guide, the size of GMDs which have a high probability of containing zero or one initial cell within the GMD can be obtained. Preferably, cells are incorporated into GMDs using a CellSys 100™ Microdrop Maker (One Cell Systems, Inc., Cambridge, MA), using the CelMix™ Emulsion Matrix and CelBioGel™ Encapsulation Matrix reagents supplied by the manufacturer (One Cell Systems, Inc., Cambridge, MA), according to the manufacturer's instructions. Detection Methods

Methods for measuring the secretion of a protein or cellular product of interest by cells of the GMD sample using a GMD secretion assay are also known in the art (see e.g., Gray, F. et al. (1995) J Immunol. Methods 182:155-163; Kenney, J.S. et al. (1995) Bio/Technology 13:787-790; Weaver, J.C. et al. (1990) Methods Enzymol. 2:234-247; Powell, K.T. and Weaver, J.C. (1990) Bio/Technology 8:333-337; Weaver, J.C. et al. (1991) Bio/Technology 9:873-877; Sullivan, S. and McGrath, P. (1995) Mol. Cell. Biol. 6:445A). Typically, cells secreting a protein are encapsulated in biotin-conjugated agarose and a biotinylated antibody directed against the protein secreted by the cell is linked in the gel matrix through streptavidin. This biotinylated antibody is used to retain the secreted protein within the GMD. The amount of bound protein is then quantified using a fluorescently-labeled secondary antibody and flow cytometry. Specific cells can be selected by fluorescence activated cell sorting (FACS) on the basis of the intensity of the GMD fluorescence. Examples of suitable fluorescent materials with which the secondary antibody can be labeled include fluorescein isothiocyanate, rhodamine, umbelliferone, fluorescein, dichlorotriazinyl-amine fluorescein, dansyl chloride and phycoerythrin. Similarly, antibodies against non-protein based cellular products, for example, antibiotics, are also known in the art (see, for example, Pauillac, et al. , Immunol. Methods, 164:165-73 (1993); Karkhanis, et al, J.Clin. Microbiol, 36:1414- 1418 (1998); Tsuchiya, et al, J. Antibiot, 32:488-495 (1979); Fujiwara, et al, J. Immunol. Methods, 45:195-203 (1981); Kitagawa, et al, J. Biochem., 92:585-590 (1982); Fujiwara, et al, Cancer Res., 44:4172-4176 (1984); Little, et al, Antimicrob. Agents Chemother., 26:824-828 (1984); Kachab, et al, J Immunol Methods, 147:33-41 (1992); U.S.P.N. 4,596768; U.S.P.N. 4,150,949; and Mizugaki, et al, J. Immunoassay, 17:133-144 (1996)). Moreover, techniques for modifying non-protein based cellular products to be effective immunogens have been previously described (Nolli, et al, Ann. 1st. Super Sanita, 27:149-154 (1991); Chevrie, et al, Biochimie, 76:171-179 (1994); and Fujiwara, et al, J. Immunol. Methods, 134:227-235 (1990)). Typically, for example, the cellular product of interest (e.g., an antibiotic, amino acid, vitamin, carbohydrate, etc.,) is a hapten that can be conjugated to a protein resulting in a conjugate that can be used to immunize a mammal, for example, a mouse or rabbit. Techniques for identifying, isolating, and culturing antibodies from an immunized animal are well established in the art (see, e.g., Using Antibodies: A Laboratory Manual, Harlow et al, C.S.H.L. Press, Pub. (1999)).

Although a preferred embodiment of the GMD secretion assay described above utilizes a biotin-streptavidin system to trap the secreted protein or cellular product within the GMD and a fluorescently-labeled secondary antibody to allow for detection and quantitation of the entrapped protein, it will be readily apparent to the ordinarily skilled artisan that modifications of this approach can be used. The key feature of the GMD secretion assay is that proteins or cellular products are secreted into the matrix of the GMD by cells encapsulated by the GMD remain entrapped within the GMD and are labeled with a detectable label to allow for detection and quantitation of the secreted protein or cellular product. Other prosthetic group complexes can be used to link the protein or cellular product within the matrix, such as a biotin-avidin system. Moreover, other detectable substances can be used to label the protein to allow for its detection, including, for example, various enzymes, luminescent materials, and radioactive materials. Examples of suitable enzymes include horseradish peroxidase, alkaline phosphatase, β-galactosidase, or acetylcholinesterase; an example of a luminescent material includes luminol; and examples of suitable radioactive material include 195 I,

¹³¹I, ³⁵S or ³H.

Sample Collection

To monitor the stability of protein or cellular product secretion by cells in a fermenter culture in accordance with the method of the invention, steps (a) through (c) of the method as described above are repeated on a second sample of cells, wherein the second sample of cells is harvested from the fermenter culture at a time later than when the first sample of cells was harvested. The precise timing of harvesting (i.e., the spacing of the first and second harvest, and any other subsequent harvests) will depend upon the particular type of cells cultured, the type of protein or cellular product being monitored, and the nature of the instability or instabilities that may be anticipated to occur. In one embodiment, the second sample of cells is harvested at least one day later than the first sample of cells. In another embodiment, the second sample of cells is harvested at least two days later than the first sample of cells. In yet another embodiment, the second sample of cells is harvested at least three days later than the first sample of cells. In other embodiments, the second sample of cells is harvested at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22 or 23 hours later than the first sample of cells. In still other embodiments, the second sample of cell is harvested at least 4, 5, 6, 7, 8, 9 or 10 days later than the first sample of cells.

To further monitor the stability of protein or cellular product secretion by cells in a fermenter culture in accordance with the method of the invention, steps (a) through (c) of the method as described above can be repeated on a third sample of cells, wherein the third sample of cells is harvested from the fermenter culture at a time later than when the second sample of cells was harvested. In another embodiment, the method further comprises repeating steps (a) through (c) on a fourth sample of cells, wherein the fourth sample of cells is harvested from the fermenter culture at a time later than when the third sample of cells was harvested. In yet another embodiment, the method further comprises repeating steps (a) through (c) on a fifth sample of cells, wherein the fifth sample of cells is harvested from the fermenter culture at a time later than when the fourth sample of cells was harvested. Steps (a) through (c) of the method can further be repeated on yet additional samples of cells harvested from the fermenter culture over time to allow for continuous monitoring of the culture over time.

Suitable Cell Types

The method of the invention can be performed using essentially any type of cell that secretes a protein or cellular product of interest and that can be cultured in fermenter cultures. In a preferred embodiment, the cells are eukaryotic cells, such as mammalian cells (such as CHO cells, COS cells, NS/O cells) or fungal cells, e.g., yeast cells. In another embodiment, the cells are prokaryotic (e.g., E. coli cells). In yet another embodiment, the cells are insect cells (e.g., a baculoviral expression system for protein production). Pref erred Proteins

The method of the invention can be applied in the monitoring of the stability of essentially any protein of interest secreted by the cultured cells, as long as the protein can be suitably detected by the GMD secretion assay (e.g., using an antibody specific for the protein). A preferred protein whose secretion is assayed is an antibody. Other preferred proteins whose secretion can be assayed include proteins of clinical utility, such as interleukins (e.g., IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8 etc.), interferons (e.g., a-, β- or γ-IFN), growth factors (e.g., GM-CSF), other cytokines (e.g., TNFα) erythropoietin, tissue plasminogen activator, streptokinase, clotting factors (e.g., Factor VIII, Factor IX), insulin, anti-angiogenesis factors, human growth hormone (HGH) and human chorionic gonadotropin (hCG). Other preferred proteins whose secretion can be assayed include cell surface molecules that have been engineered to be secreted from the cell (i.e., by deletion of transmembrane and cytoplasmic domains), such as soluble CD4, and fusions of polypeptides with portions of immunoglobulin molecules (e.g., CTLA4- Ig).

The protein may be a protein that is naturally secreted by the cell (e.g., an antibody that is secreted by a hybridoma) or the cell may be genetically engineered to secrete the protein, in which case the protein typically is encoded by an expression vector incorporated into the cells. When the protein is comprised of more than one subunit, the individual subunits may be encoded by different vectors incorporated into the cell (i.e., the cell may carry more than one expression vector) or the subunits may be encoded by a single expression vector within the cell. Recombinant DNA techniques are well known in the art for engineering a cell to express a secreted protein.

Cellular Products (Non-Protein Based Molecules)

In addition to the aspect of protein detection, the invention is also designed to provide a method for detecting a wide range of other cellular products (i.e., non-protein based molecules) expressed by a cell. These non-protein based molecules are the cellular products of animal cells, plant cells, or microorganisms and include molecules within the chemical classes comprising alkaloids, amino acids, carbohydrates, esters, lipids, nucleic acids, organic acids or alcohols, polyketides, and non-ribosomal peptides. Functionally, the cellular product can be an antibiotic, an aroma, a pharmaceutical other than an antibiotic, an enzyme, an enzyme inhibitor, a flavor, a flavor enhancer, a hormone, a pesticide, a pigment, a surfactant, a vitamin, or any agent stated in Tables 1- 9. Moreover, hundreds of different animal cells, plant cells, and microorganisms produce cellular products such as these and any of these molecules may be used to manufacture a corresponding antibody for use in applying the methods of the invention described herein. Accordingly, the invention is intended to include a method for identifying any cell that produces a cellular product of interest, preferably, e.g., antibiotics, polyketide antibiotics, amino acids, antifungals, vitamins, non-antibiotic pharmaceuticals, food additives (flavors, colorants, etc.), chemicals, and polymers that can be detected using an antibody-based assay.

It will be appreciated by the skilled artisan that using the teachings herein, and what is known in the art, would allow for the manufacture of an antibody against a cellular product using only routine experimentation. In addition, it will be appreciated that the use of animal cells, microorganisms, or plants in generating any number of the cellular products described herein is well known in the art (see, for example, Tables 1-2 and 4-9 adapted from, A Revolution in Biotechnology, Marx, J., Ed. Cambridge University Press, (1989); Cragg et al, J. Nat. Prod., 60:52-60 (1997); and, for example, Table 3 adapted from KOSAN Inc., http://www.kosan.com). Accordingly, any of the cellular products produced by any of the cells recited herein is intended to be encompassed by the methods of the invention.

Amino Acids

In one embodiment, such a cellular product may be, a product essential for life, i.e., a primary metabolite, such as, for example, an amino acid. It is known in the art that, of the twenty amino acids essential for life, humans can only synthesize twelve. Accordingly, the other eight amino acids must be supplied by the diet. Conveniently, a number of microorganisms, such as bacteria, can synthesize these essential amino acids and this production can be monopolized for the production of amino acids for human use (see, Table 1). Accordingly, the invention is intended to include methods for detecting any cell capable of producing any of the twenty amino acids, for example, lysine or tryptophan, useful as nutritional supplements for either humans or animals, for example, livestock. In a preferred embodiment, the invention is intended to include any of the essential amino acids, e.g. , histidine, isoleucine, leucine, lysine, methionine, cysteine, phenylalanine, tyrosine, threonine, tryptophan, valine, or arginine. In addition, the invention is also intended to encompass any amino acid (modified or unmodified) that are linked to produce, for example, an artificial sweetener, such as, e.g., aspartame, which is comprised of two amino acids, aspartic acid and phenylalanine.

Table 1. Production of Various Amino Acids by Microbial Fermentation

Vitamins

In another embodiment, the invention may be used to monitor the production of other cellular products such as vitamins. In particular, the B vitamins, for example, vitamin B2-aldehyde and vitamin B2-acid are produced by Schizophyllum commune, a Basidiomycete, and, for example, riboflavin, a reduction product of B2-aldehyde which can be catalyzed by Lactobacillus casei (Tachibanna et al. , J. Nutr. Sci. Vitaminol. 25:361-366 (1979); Tachibana et al, J. Nutr. Sci. Vitaminol, 26:419-426 (1980); Tachibana et al, J. Nutr. Sci. Vitaminol., 28:335-342 (1982). Other vitamins routinely produced by microorganism that may be subjected to the detection methods of the invention include vitamin C, vitamin B 12, vitamin A, and vitamin K.

Antibiotics

In a preferred embodiment, the cellular product produced and capable of being detected by the methods of the invention is a secondary metabolite, i.e., a nonessential product produced by the cell, such as, e.g., an antibiotic. Antibiotics are some of the most desirable secondary metabolites produced by microorganisms and the ability to optimally culture such organisms is an advantage of the invention. In one embodiment, the method of the invention involves the detection of a cell/s producing an antibiotic, such as penicillin produced from, for example, a strain of Penicillium crysogenum (Aldrige, S., New Scientist (1997)). Other antibiotics include gentamicins produced by Micromonosporα purpureα (Abou-Zeid e αl, Nαturwiss. 133:261-275 (1978)) and Daptomycin (Cubist Pharmaceuticals, Inc.) produced from Streptomyces roseosporus (Mchenny et αl, J. BαcterioL, 180:143-151 (1998)). Other preferred cellular products include penicillin G or V produced by Penicillium chrysogenum, cephalosporin C produced by Cephalosproium acremonium, and any antibiotic produced by the species Streptomyces, for example, Streptomyces rimosus (High Tech Separation News, Business Communications Co., (1998)). Table 2. Microbiological Production of Various Antibiotics by Fermentation

Polyketide Antibiotics and Other Polyketide Products

In a preferred embodiment, the method of the invention is capable of detecting a cellular product that is member of the chemical class of polyketides, such as, for example, polyketide antibiotics. Polyketides are complex natural products that are built from simple carboxylic acid monomers and are produced by a number of microorganisms. A number of naturally-occurring polyketides are highly desirable, and commercially successful, bioactive molecules useful in a wide range of applications in human and animal health (see Table 3, adapted from Kosan Sciences, Inc, http://www.kosan.com). It is known in the art that cellular products of this chemical class can be made immunogenic such that antibodies that bind the molecule can be produced. Accordingly, the skilled artisan will appreciate that the invention may be used to detect the presence and production of any of the polyketide agents presented herein and, for example, as set forth in Table 3, below.

Table 3. Polyketide Antibiotics and Other Polyketide Products

Moreover, the invention also provides the ability to detect novel polyketides that are produced from cells that have been manipulated using genetic engineering. It is known in the art that the enzymes in the biosynthetic pathway involved in polyketide synthesis can be manipulated to produce various "unnatural" polyketide products (see, e.g., Hutchinson, C, Bio/Technology 12:375-380 (1994); Kennedy et al, Science 284:1368-1372 (1999) and Cane et al, Science 282:63-68 (1998)). Accordingly, the novel polyketide product, using techniques described herein, can be made into a immunogen suitable for raising antibodies that can bind to the novel or "unnatural" polyketide product. The method of the invention is then modified to incorporate this antibody in order to detect cell/s expressing the variant polyketide as described. Anti-Fungal Agents

In another embodiment, the invention is suitable for detecting cells producing anti-fungal agents, for example, the sphingofungins, produced by Aspergillus fumigatus (ATCC 20857) (VanMiddlesworth et al, J. Antibiot. 45:861-867 (1992)). In addition, the anti-proliferative agent fumagillin also produced by Aspergillus fumigatus may be monitored using the methods of the invention (Casey et al, Am. J. ofOpth. 124:521-531 (1997)). And, in yet another embodiment, antifungal molecules for agricultural application are encompassed by the invention. For example, the bacteria Pseudomonas fluorescens and other Pseudomonas strains that produce the antibiotics 2,4- diacetylphloroglucinol and pheazine-1-carboxylic acid that inhibit the fungus, G. graminis, which destroys wheat crops, may be monitored using the methods of the invention (Stelljes, K., Agricultural Research, 47:10 (1999); Pesticide & Toxic Chemical News, Information Access Co., Pub. (1997)).

Other Non-Antibiotic Pharmaceuticals

In another embodiment, the invention is intended to include methods of detecting cells producing other cellular products such as non-antibiotic pharmaceuticals (see, e.g., Table 4).

Table 4. Microbiological Production of Pharmacologically Active Compounds

Carbohydrates and Polymers

In another embodiment, the invention also encompasses the method of detecting cells expressing a carbohydrate, preferably, a desirable polymer comprising a carbohydrate (see, e.g., Table 5).

Table 5. Microbiological Production of Polymers by Fermentation

Qther Products

In yet another embodiment, the methods of the invention may used for the detection of any other antibiotics, food pigments, food flavorings, alkaloids, herbicides, pesticides, or any other cellular product known in the art that can be produced by an animal cell, plant cell, or microorganism under culture conditions involving a fermenter and subject to being detected by an antibody (see, e.g., Tables 1-9 and also Barrios- Gonzalez et al, Biotechnol Ann. Rev. 2:85-121 (1996); From Ethnomycology to Fungal Biotechnology: Exploiting from Natural Resources for Novel Products, Singh, J., Ed., Plenum Press, Pub. (1999); Manual of Industrial Microbiology and Biotechnology, Demain, A. Ed., Am. Soc. of Microbiology, Pub., (1999); Biomining: Theory, Microbes, and Industrial Processes, Rawlings, Ed., R.G. Landes Co., Pub. (1997); Biotechnology of Industrial Antibiotics, Vandamme, E., Ed., Marcel Dekker, Pub. (1984); Industrial Biotechnology, Malik, V., Ed., Science, Pub. (1992); Biotechnology and Food Ingredients, Goldberg et al, Ed., Aspen Publishers (1991); Biotechnology and Food Process Engineering, Schwartzberg et al., Ed., Marcel Dekker, Pub.(1990); and Food Biotechnology: Techniques and Applications, Mittal, G., Technomic Pub. Co. (1992).

Table 6. Microbial Production of Aroma and Flavor Chemicals

Table 7. Other Cellular Products Produced by Microorganisms by Fermentation

Table 8. Industrial Chemicals Produced by Fermentation

Plant Products

In a embodiment, the cellular product to be detected by the methods of the invention is produced by a plant cell. It is know in the art that plant cells may be used as a source of various useful pharmaceuticals, nutrients, flavorings, colorants, and antimicrobials (see, e.g., Table 9; Engineering Plants for Commercial Products and Applications, Collins, G., Ed., Ann. N. Y. Acad. of Sciences, Pub., (1997); and Plant Cell and Tissue Culture for the Production of Food Ingredients, Fu, T-J., Ed., Plenum Pub., (1999)). Table 9. Cellular Products from Plants and Their Application

Moreover, technologies in the art now make it possible to isolate and grow totipotent plant cells which can be grown in liquid culture, e.g., in a fermenter. Typically, such cells are isolated from the root tip or meristem and cultured into a callus which can then be divided into single cells and cultured in the presence of appropriate nutrients and hormones. Such cells may be selected for various properties (e.g. , using growth conditions), mutagenized to acquire certain enhanced properties (e.g., using radiation), or transformed using genetic engineering to express a desirable molecule, for example, a cellular product. Techniques for conducting the above-mention manipulations are known in the art (see, e.g., Plant Cell Culture Secondary Metabolism: Toward Industrial Application, Discosmo, F., Ed., CRC Press, (1996); Transgenic Plants; A Production System for Industrial and Pharmaceutical Proteins, Owen, M., Ed., John Wiley & Sons, Pub., (1996); and Applied Plant Biotechnology, Chopra, V., Ed., Science Publishing, (1998)). Accordingly, any plant cell capable of being grown in a fermenter and capable of secreting a cellular product that can be bound by an antibody, is within the scope of the invention. Culture Size

The method of the invention allows for the monitoring of large scale fermenter cultures. Preferably, the fermenter culture comprises at least 10 liters of culture medium. More preferably, the fermenter culture comprises at least 50 liters of culture medium. Even more preferably, the fermenter culture comprises at least 100 liters of culture medium. Moreover, monitoring of fermenter cultures of 1000 or more liters can be accomplished using the method of the invention.

Recovery of Preferred Cell Culture/Clone

In another embodiment of the invention, following monitoring of the stability of protein secretion by cells in the fermenter culture, a cell population with stable expression of the protein can be recovered. This stable, recovered population can then be further cultured, if desired. To recover a population of cells with high levels of stable protein secretion, following the GMD secretion assay in which protein secreted by the cells (and entrapped within the GMD) is fluorescently labeled, fluorescence activated cell sorting (FACS) can be performed to sort for GMDs having large amounts of associated fluorescent label.

The present invention is further illustrated by the following examples which should not be construed as limiting in any way. The contents of all cited references, including literature references, issued patents, and published patent applications, as cited throughout this application are hereby expressly incorporated by reference.

The following reagents and methodologies were used in the Examples:

Materials and Methods

Cells and Expression Vectors

All cell lines described in the Examples are derived from the Chinese hamster ovary (CHO) cell line DUX Bl 1, a mutant of CHOK1 cells deficient in dihydrofolate reductase (DHFR). D2E7*peaBJ, a vector which directs the expression of the heavy and light chains of D2E7, was used to transfect CHO cells. D2E7 is a recombinant human antibody against human tumor necrosis factor alpha, described further in PCT Publication WO 97/29131. The vector carries the murine dhfr gene driven by the SV40 early promoter which functions as a selection and amplification marker (Kaufman, R.J. et al. (1985) Mol Cell. Biol. 5:1750). D8/E, a CHO line transfected with the D2E7 expression vector, was amplified via methotrexate selection to increase the number of copies of the vector and therefore the level of antibody secretion.

Preparation of Gel Microdrops and Antibody Secretion Assay

CHO cells were incorporated into gel microdrops (GMDs) using a CellSys 101® microdrop maker (One Cell Systems, Inc.) The cells' level of antibody secretion was determined using the hybridoma secretion assay protocol supplied by One Cell Systems Inc. and modified as follows. Cells washed once with cold phosphate buffered saline were encapsulated using 350μL of CelBioGel biotin-conjugated agarose and lOOμL of pluronic solution (Gibco/Life Technologies). The resulting GMDs were washed twice and incubated with streptavidin (Gibco) at 90μg/ml, washed again and incubated with biotin-conjugated goat anti-human IgG (Organon Technika) or biotin-conjugated murine monoclonal against human IgG (Zymed). After washing, the GMD-encapsulated cells were returned to growth medium in a 37°C incubator with a 5% CO₂ atmosphere for a period of time sufficient to allow an accumulation of secreted antibody within the drop. At the end of the incubation, GMDs were incubated with goat anti-human IgG conjugated with fluorescein isothiocyanate (FITC) or FITC-conjugated monoclonal against human IgG (Zymed). After washing the GMDs were analyzed by flow cytometry. Gates were chosen such that only GMDs containing cells were analyzed.

Fermentation Conditions and Sampling Strategies Cells harvested from the GMP fermenter runs as part of the product processing were washed twice with phosphate buffered saline and used for DNA extractions or secretion assays as described above. The production of D2E7 was via a batch/refeed strategy. Approximately eighty percent of the fermentation volume was harvested from the fermenters once every three days. The fermenter was replenished with fresh medium and incubated another three days before another harvest was made. This batch and refeed process was repeated until eight harvests, designated A through H were recovered. The production process was run under two different protocols. For the AFF607 run, the D8/E cells were cultured without methotrexate from the cell bank vial through culture in the 1000 L. fermenter. For the AFF610 and AFF702 runs, methotrexate at a final concentration of 500 nM was included in the cell culture medium until the inoculation of the 100 L. fermenter. No methotrexate was added to the cell culture medium used in the 100 L fermenter or in the 1000 L fermenter used for production of D2E7.

Analysis of Cells Recovered from the Encapsulation Assay

Subpopulations of cells identified using the antibody secretion assay were separated using a FACStar cell sorter. Cell lines established via the outgrowth of cells from individual GMDs were cultured in spinner flasks in the same medium used in the fermenters. Samples were taken daily and assayed for the number of viable cells and for the concentration of human IgG.

Vector Copy Number Determination

Genomic DNA was extracted from the recovered cells according to the protocol described in Current Protocols in Molecular Biology; Ausubel, F.M., Brent, R., Moore, D.M., Kingston, R.E., Seidman, J.G., Smith, J.A., and K. Struhl eds; Wiley Interscience, N.Y., N.Y. (1990). Ten micrograms of DNA isolated from each culture were digested with restriction endonucleases. Digested DNAs were subjected to agarose gel electrophoresis. Each gel was loaded with 5 μg of digested genomic DNA from each of the cell banks as well as a series of standards containing known amounts of D2E7 heavy chain and light chain DNA.

The standards were prepared as follows. The D2E7 expression vector was digested with EcoRI and Notl to release fragments carrying the heavy chain or light chain coding regions along with their upstream adenovirus promoters. The digested plasmid was diluted to concentrations which allowed the loading of D2Ε7 DΝA at levels equivalent to standard copy numbers. These standards were DΝAs in amounts equivalent to 100, 20, 8, 4, or 2 copies of these genes per genome. The resulting gels were blotted to nylon membranes and hybridized to radioactively-labeled probes as described in Current Protocols. Duplicate blots were hybridized with probes which consisted of either the D2E7 heavy or light chain DNA encoding the full length coding region. These DNA fragments were isolated from D2E7 expression vectors by restriction endonuclease digestion with Srfl and Not I. DNA fragments were isolated and purified by gel electrophoresis and subsequent extraction. Isolated DNA fragments were radioactively labeled using the Rediprime labeling system (Amersham Corp.). This system employs a random primer technology and nick translation reaction to produce probes with high specific activity. Blots were exposed to radioactive probes in hybridization solution, described in Current Protocols in

Molecular Biology at 65°C. Blots were washed under conditions of high stringency (0.1 X SSC, 0.1% SDS) and exposed to Kodak X-OMAT photographic film, or to BioRad Molecular Imager screens. In order to correct for variations in the loading of sample wells, blots were stripped of residual probe and probed with radiolabeled human S14 gene probes (Rhoads, D.D., Dixit, A., and Roufa, D. J. Mol. Cell. Biol. 6, 2774-2783 (1986). S 14 is a ribosomal protein expressed in all human cell types. The sequence has a high degree of homology with its homologue in other mammalian species. Each sample's signal with the S14 probe was used to normalize the signal seen with the D2E7 process.

EXAMPLE 1: The Secretion Profile of Cells Grown in

Fermenter Cultures Differs over Time

The D8/E cell line, carrying an expression vector for the recombinant antibody D2E7, was inoculated into a fermenter culture and cultured for a total of six weeks, as described above. Eight harvests of aliquots of the fermenter culture were taken over time and the secretion of D2E7 by cells in the culture was assessed using the gel microdrop secretion assay described above.

The secretion profiles of the initial fermenter inoculum (AFI 615.75) and the first harvest (AFF 610A), fourth harvest (AFF 610D) and eighth harvest (AFF 610H), illustrated by FACS histogram, are shown in Figures 1A-1D. The smaller flanking peaks in the figures represent the negative and positive control samples, which were prepared in parallel with the secretion assay and show background fluorescence and the maximal signal attainable from the microdrops, respectively. The results from the secretion assays depicted in Figures 1 A- ID show that the secretion profile of cells grown directly from the stored cell banks (i.e., the fermenter inoculum) differed from that of cells recovered from the final fermenter harvest. The latter showed a distinctly bimodal distribution of expression levels while the former showed a relatively uniform level of expression.

EXAMPLE 2: Vector Copy Number Determinations

To determine whether the number of copies of the D2E7 expression vector in the D8/E cell line changed over time during culture, Southern blot analysis was conducted as described above. The numbers of copies of the heavy chain and light chain genes, respectively, present in cells of different samples of the culture are depicted in the bar graphs of Figures 2A-2B. Samples 615 0.6, 615 5.0 and 615 75 were taken from the fermenter seed inoculum culture as it was expanded from 0.6 liters to 5 liters and then to 75 liters. Samples 610A, 610D and 61 OH were taken from the fermenter at the first, fourth and eight harvest. No significant difference in antibody heavy or light chain copy number could be detected by Southern blot analysis. The Southern blots probed for the recombinant human antibody heavy and light chains showed that while the level of antibody light chain appeared to rise with increasing culture time in the fermenter, heavy chain copy number appeared to be lower in the last fermenter sample (i.e, the eighth harvest). However, as the copy number determinations have a variation of about 30%, all of the samples are within the error of the test; and therefore no significant difference in copy number was detected from the total fermenter cell population.

EXAMPLE 3; Analysis of Subpopulations of the Fermenter Culture While the copy number experiments described in Example 2 were not able to determine if a loss of vector copies had occurred, the data from the GMD secretion assay described in Example 1 suggested that a genetic instability existed in the D8 E cell line. To confirm this hypothesis, we set out to isolate the subpopulations which produce different levels of fluorescence in the assay. These subpopulations were tested for their rate of antibody production and for the number of vector copies in the genome.

To analyze subpopulations of cells identified via the secretion assay, cells recovered from the last harvest of fermentation runs 607 and 702 were cultured in spinner flasks. Three million cells of each culture were resuspended in 100 μl PBS and encapsulated into gel microdrops and tested in the secretion assay as described above. The FACS histogram for the last harvest of fermentation run 607 (referred to as sample 607H) is shown in Figure 3A, whereas the FACS histogram for the last harvest of fermentation run 702 (referred to as sample 702H) is shown in Figure 3B. For each culture, the FACS histogram showed two subpopulations, one which appeared to represent a population of low expression cell lines (labeled as region R3 in Figures 3A- 3B) and one representing a population of high level expression cell lines (labeled as region R2 in Figures 3A-3B).

The FACStar Plus (Becton-Dickenson) was used to isolate representatives of each of the R2 and R3 populations to allow for the determination of each populations' productivity. The 702H sort contained 1.3 X 10⁵ cells for the low expression population and 1.5 X 10⁵ cells for the high expression population. For the 607H culture, 0.5 X 10⁵ cells and 1.5 X 10⁵ cells were collected by sorting for the low and high level expression populations, respectively. Each pool of sorted cells was divided among 3 wells of a 6 well cell culture dish and grown in fermenter culture medium containing penicillin- streptomycin and gentamycin. Upon growth to confluence, cells were transferred into spinner flask culture in fermenter culture medium containing penicillin-streptomycin and gentamycin. After adaptation to growth in suspension, the cultures were tested for their level of antibody productivity. The low and high producer lines were cultured in spinner flasks (50 ml volume) for 4 weeks with daily cell counts and samples taken for antibody productivity determinations by the gel microdrop secretion assay. All cultures were subcultured twice a week. The high expression subpopulation of the 702H sample was grown in spinner culture for an additional 6 weeks (for a total often weeks) to test for expression stability.

The antibody productivity of the low expressor (702hLo) and high expressor (702hHi) subpopulations of the 702 culture are illustrated in the bar graph of Figure 4, in which the antibody productivity is expressed as picograms of antibody expressed per cell in 24 hours. Each productivity value is the average of the daily values over a four day period of growth. The results shown in Figure 4 demonstrate that during the extended period of culture of the high producer cell line over a period often weeks, the high level of antibody production was maintained. Productivity of the high expression subpopulation was greater than three-fold higher than that of the low expression cells. The vector copy number in the 607H and 702H samples, and in the 607H low, 607H high, 702H low and 702H high subpopulations were determined by Southern blot analysis. The results are summarized in the bar graphs of Figure 5 A (heavy chain) and Figure 5B (light chain). The graphs represent the average of two experiments. Bars reflect the range of values observed. Copy number assessments of each subpopulation showed that the high and low cultures have gene copy numbers that correspond to their level of D2E7 production. The heavy and light chain copy numbers of the 607H subpopulations differ by 29% and 46%, respectively. Likewise, the 702H subpopulations differ by 54% and 65%, respectively. In our hands, the copy number assay has a variation of 30%. Therefore, three out of the four determinations represent significant differences in copy number.

Based on these data, it has been demonstrated that the GMD secretion assay method detects the appearance of a low producer subpopulation derived from the original producer line but which has lost copies of the expression vector from its genome. Thus, the gel microdrop secretion assay has been used successfully to monitor the stability of recombinant antibody secretion from transfected CHO cells.

Equivalents

Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. For example, the Examples and the specification sufficiently teach that the monitoring of any secreted cellular product capable of being identified using an antibody, can be accomplished using the methods of the invention. Moreover, the cellular product may be expressed by any living cell, from a cell in a homogeneous or heterogeneous culture, and by a cell drawn from any environment or reaction vessel in which cells can be cultured and monitored. Such equivalents are intended to be encompassed by the invention and the following claims.

Claims

We claim:

1. A method for monitoring stability of protein secretion by cells in a fermenter culture, comprising: (a) harvesting from a fermenter culture a first sample of cells that secrete a protein;

(b) incorporating cells from the first sample into gel microdrops (GMDs) to form a GMD sample;

(c) measuring secretion of the protein by cells of the GMD sample using a GMD secretion assay; and

(d) repeating steps (a) through (c) on a second sample of cells, wherein the second sample of cells is harvested from a fermenter culture at a time later than when the first sample of cells was harvested thereby monitoring stability of protein secretion by cells in a fermenter culture.

2. The method of claim 1, which further comprises (e) repeating steps (a) through (c) on a third sample of cells, wherein the third sample of cells is harvested from the fermenter culture at a time later than when the second sample of cells was harvested.

3. The method of claim 2, which further comprises (f) repeating steps (a) through (c) on a fourth sample of cells, wherein the fourth sample of cells is harvested from the fermenter culture at a time later than when the third sample of cells was harvested.

4. The method of claim 3, which further comprises (g) repeating steps (a) through (c) on a fifth sample of cells, wherein the fifth sample of cells is harvested from the fermenter culture at a time later than when the fourth sample of cells was harvested.

5. The method of claim 1, wherein the cells are eukaryotic cells.

6. The method of claim 1 , wherein the cells are mammalian cells.

7. The method of claim 1, wherein the cells are yeast cells.

8. The method of claim 1, wherein the cells are prokaryotic.

9. The method of claim 1, wherein the protein secreted by the cells is an antibody.

10. The method of claim 1, wherein the cells secrete a protein encoded by an expression vector incorporated into the cells.

11. The method of claim 1 , wherein the cells secrete an antibody encoded by at least one expression vector incorporated into the cells.

12. The method of claim 1, wherein the fermenter culture comprises at least 10 liters of culture medium.

13. The method of claim 1, wherein the fermenter culture comprises at least 50 liters of culture medium.

14. The method of claim 1, wherein the fermenter culture comprises at least 100 liters of culture medium.

15. The method of claim 1 , wherein the second sample of cells is harvested at least one day later than the first sample of cells.

16. The method of claim 1 , wherein the second sample of cells is harvested at least two days later than the first sample of cells.

17. The method of claim 1, wherein the second sample of cells is harvested at least three days later than the first sample of cells.

18. The method of claim 1 , further comprising recovering a cell population with stable expression of the protein.

19. A method for monitoring stability of cellular product secretion by cells in a fermenter culture, comprising:

(a) harvesting from a fermenter culture a first sample of cells that secrete a cellular product;

(b) incorporating cells from the first sample into gel microdrops (GMDs) to form a GMD sample; (c) measuring secretion of the cellular product by cells of the GMD sample using a GMD secretion assay; and

(d) repeating steps (a) through (c) on a second sample of cells, wherein the second sample of cells is harvested from a fermenter culture at a time later than when the first sample of cells was harvested to thereby monitoring stability of cellular product secretion by cells in a fermenter culture.

20. The method of claim 19, which further comprises (e) repeating steps (a) through (c) on a third sample of cells, wherein the third sample of cells is harvested from the fermenter culture at a time later than when the second sample of cells was harvested.

21. The method of claim 20, which further comprises (f) repeating steps (a) through (c) on a fourth sample of cells, wherein the fourth sample of cells is harvested from the fermenter culture at a time later than when the third sample of cells was harvested.

22. The method of claim 21, which further comprises (g) repeating steps (a) through (c) on a fifth sample of cells, wherein the fifth sample of cells is harvested from the fermenter culture at a time later than when the fourth sample of cells was harvested.

23. The method of claim 19, wherein the cells are eukaryotic cells.

24. The method of claim 19, wherein the cells are mammalian cells.

25. The method of claim 19, wherein the cells are yeast cells.

26. The method of claim 19, wherein the cells are prokaryotic.

27. The method of claim 19, wherein the cellular product secreted by the cells is an antibiotic.

28. The method of claim 27, wherein said antibiotic is a polyketide.

29. The method of claim 19, wherein the cellular product secreted by the cells is an amino acid.

30. The method of claim 29, wherein said amino acid is an essential amino acid.

31. The method of claim 19, wherein the cellular product secreted by the cell is a carbohydrate or a polymer thereof.

32. The method of claim 19, wherein the cellular product secreted by the cell is a vitamin.

33. The method of claim 32, wherein said vitamin is selected from the group consisting of a B vitamin, vitamin C, vitamin K, and riboflavin.

34. The method of claim 19, wherein the cells secrete a cellular product produced by an enzyme encoded by an expression vector incorporated into the cells.

35. The method of claim 19, wherein the cells secrete an antibiotic produced by an enzyme encoded by at least one expression vector incorporated into the cells.

36. The method of claim 19, wherein the fermenter culture comprises at least 10 liters of culture medium.

37. The method of claim 19, wherein the fermenter culture comprises at least 50 liters of culture medium.

38. The method of claim 19, wherein the fermenter culture comprises at least 100 liters of culture medium.

39. The method of claim 19, wherein the second sample of cells is harvested at least one day later than the first sample of cells.

40. The method of claim 19, wherein the second sample of cells is harvested at least two days later than the first sample of cells.

41. The method of claim 19, wherein the second sample of cells is harvested at least three days later than the first sample of cells.

42. The method of claim 19, further comprising recovering a cell population with stable expression of the cellular product.