Continuous hybridoma suspension cultures with and without cell retention: kinetics of growth, metabolism and product formation

Journal 0

of Biotechnology,

1992 Elsevier

BIOTEC

Science

22 (1992) Publishers

3 I-40 B.V.

31 All

rights

reserved

016%1656/92/$05.00

00656

Continuous hybridoma suspension cultures with and without cell retention: kinetics of growth, metabolism and product formation G. Schrnid, C.R. Wilke and H.W. Blanch Department

of Chemical (Received

Engineering, 2 October

lJnit>ersity

of California,

1990; revision

accepted

Berkeley, 18 March

California,

U.S.A.

1991)

Summary

A laboratory scale bioreactor was constructed from glass and polycarbonate materials whereby a track-etch membrane (3 pm pore diameter) was integrated into its two-part bottom flange. The reactor performance was evaluated for continuous hybridoma suspension cultures under various conditions of cell retention. A total retention experiment demonstrated that this type of stirred tank reactor cannot be operated at near zero growth rate conditions. Instead, at steady viable cell concentrations of 2: 3 x 10” ceils per ml, specific growth and death rates were estimated at 0.60 f 0.06 d-‘. Specific substrate (glucose, glutamine, O,, amino acids) consumption, by-product (ammonia, alanine, amino acids) and product (antibody) production rates as well as various apparent molar yield coefficients were obtained and are compared to metabolic quotients and yield coefficients previously calculated from standard continuous culture experiments, i.e., without cell retention, at specific growth rates of 0.63 and 1.24 dd’. Furthermore, steadyCorrespondence

to and present address: G. Schmid, Pharma Research New Technologies, PRTM, Bld. F. Hoffmann-La Roche AG, CH-4002 Basle, Switzerland. Nomenclature: D, dilution rate [ = volumetric feed rate F/reactor volume V] (d-l); Jo. specific growth rate (d-t); R, retention rate (-); k,, specific death rate (d-l); tr, total cell count (cells per ml); rrdr

66/302,

dead cell count (cells per ml); n,, viable cell count (cells per ml); p, product concentration (mM); s, substrate concentration (mM); (mAb), antibody concentration (pg ml-‘); 9,, specific product formation rate (mmol per 10’ cells per d); rlS, specific substrate production rate (mmol per 10’ cells per d); qmAb, antibody metabolic quotient (mg per IO’ cells per d); r, time (d); Yr,‘S, apparent yield of product substrate [ = 9,, /rls] (mol mol - ’ ). glut, glucose; NH s, ammonia; represented by their 3 letter code.

from

Subscripts:

lac,

lactate;

mAb,

antibody;

F, feed

stream.

Amino

acids

are

32

state data on viable cell and antibody concentrations, spec. mAb productivities, and space-time yields determined before and after a step change (2.5fold increase) in dilution rate at identical specific growth rates k are presented. Mammalian cell metabolism; Hybridoma; Monoclonal tinuous culture; Cell retention; Microfiltration

antibody; Bioreactor;

Con-

Introduction A major goal in the process development of animal cell culture plants has been to substantially increase the reactor cell concentrations and achieve elevated space-time yields or volumetric productivities. Besides research and development efforts focused on packed bed and fluidized bed reactors or hollow fiber systems, a variety of means have been explored to effectively retain mammalian cells in the conventional stirred tank or airlift bioreactor. Most of these approaches either make use of internal, moving or rotating (“spin filter”) polymeric membranes and stainless steel or ceramic filters or of external cross-flow filtration devices (Himmelfarb et al., 1969; Thayer, 1973; van Wezel et al., 1985; Tolbert et al., 1981; Feder and Tolbert, 1985; Tolbert et al., 1985; Wagner and Lehmann, 1988; Avgerinos et al., 1990; Takazawa et al., 1988a; Scheirer, 1988; Reuveny et al., 1985). Internal (settling zones) and external sedimentation or centrifugation have likewise been evaluated as methods of cell retention or recycle (Shimazaki et al., 1986; Sato et al., 1983; Tokashiki et al., 1988; Takazawa et al., 1988b; Brown et al., 1991). In this communication we present experimental data on hybridoma growth, metabolism, and product formation obtained with a laboratory scale stirred bioreactor equipped with a microporous membrane for in-situ cell retention.

Materials

and Methods

Cell line, medium composition and growth conditions

Cell line AB2-143.2 is an SP2/0-derived murine hybridoma that produces an IgG,,, antibody to benzene-arsonate (Hornbeck and Lewis, 1985). Cells were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% fetal bovine serum (HyClone, Logan, UT, U.S.A.) and 1% each of 100 X MEM nonessential amino acids and 11 g I-’ Na-pyruvate (all except serum from Gibco Laboratories, Grand Island, NY). Buffering capacity was provided by replacing 44 mM sodium bicarbonate with 25 mM HEPES sodium salt (N-2-Hydroxyethylpiperazine-N”-Zethanesulfonic acid) (Shipman, Jr., 1969). An extra 20 mM NaCl was added to ensure a constant osmolality. For continuous suspension cultures

33

initial (and feed) concentrations of glucose and glutamine were 13.5 mM and 4.9 mM, respectively. All the experiments were performed without antibiotics and periodic mycoplasma samples were negative (Schmid et al., 1992a). Cell culture bioreactors

A one liter round-bottom glass reactor (Pegasus, Scarborough, Ontario, Canada) with 600 ml working volume was used for all continuous suspension culture experiments without cell retention (Schmid et al., 1992a). Continuous cultures with total or partial cell retention were performed in a 1.5 I custom-built flat-bottom bioreactor (700 ml working volume) made from glass and polycarbonate materials. A 142 mm diameter microporous track-etch membrane (Nuclepore, Pleasanton, CA, U.S.A.) with 3 pm pore diameter was integrated into the two-part bottom flange. Further details on this reactor are described by Schmid et al. (1992b). With both reactor configurations agitation was provided by a 5-cm diameter axial flow turbine with four 45 o pitched blades operated at 180 & 10 rpm. The oxygen transport was via surface aeration and the partial pressure in the liquid was controlled at 70% of air saturation by varying the oxygen concentration in the headspace. Experiments with partial cell retention (D = 0.97 d-l; R = 0.61) did need additional sparging of oxygen via a stainless steel filter to maintain the desired dissolved oxygen concentration. Temperature was maintained at 37.0 + 0.2 o C with a circulating water bath. The pH-value was controlled at 7.2 + 0.1 by the addition of 0.5 N sodium hydroxide. Any samples (5 ml including purge) were taken twice daily. A multi-channel peristaltic pump (Gilson, Middleton, WI, U.S.A.) was used for medium addition and product removal. Metabolite and product analyses

The cell samples were diluted 1 : 1 with 0.16% trypan blue in normal saline and counted on a hemacytometer; any nonviable cells stained blue. An average of four determinations were used to calculate the viable cell concentration and percent viability. The remainder of each sample was centrifuged to pellet the cells. The supernatant was preserved with sodium azide and frozen for later analyses. Glucose was measured using a clinical glucose analyzer; lactate was determined by an enzymatic assay (Sigma Chemical Co., St. Louis, MO, U.S.A.); ammonia was measured with an ion-selective electrode (Orion, Boston, MA, U.S.A.). Samples were reacted with o-phthaldialdehyde (OPA) to form fluorescent amino acid derivatives which were subsequently separated by high performance liquid chromatography (HPLC) using an RP-18 column with gradient elution from 10% methanol (0.1 M sodium acetate, pH 6.8 as balance) to 95% methanol over 30 min. Antibody concentrations were determined by high pressure Protein A affinity chromatography. Samples were loaded on a 30 X 4.6 mm AffiGel MAPS Protein A column (BioRad, Richmond, CA U.S.A.) with 0.1 M phosphate buffer, pH 8.0 and eluted using 0.1 M citrate buffer, pH 2.5 (Custer and Schmid, 1987). Additional details on these assay procedures are presented elsewhere (Schmid et al., 1992a).

F, so, rl=O

1 D = F/V

I 1 V, s. n -----l

(1

-

R) * F. s. n t

R * F. s. n = Fig. 1. Scheme

Determination

of the reactor

0

t

configuration.

of specific growth and death rates and of metabolic quotients

The described experiments were performed in constant-volume continuous-flow bioreactors without and with partial or total cell retention. Fig. 1 provides a scheme of the reactor configuration. Note that all metabolic quotients have been calculated using viable cell concentrations. Material balances around the reactor yield the following equations for cell growth: dn,/dt

= pn, - kdnv - (1 - R) Dn,

(1)

dn,/dt

= kdnv - (1 - R) Dn,

(2)

Addition dn/dt

of Eqs. (1) and (2) assuming n = n, + nd gives: =pn,-

(1 -R)D

Thusatsteadystate:p=(l-R)D(n/n.)=(l-R)D+k, Specific metabolic quotients for substrate consumption product (antibody) formation are defined by: qs= [D(s,-s)

-ds/dt]/n,

(3)

(4) and by-product as well as (5)

qp= [D(P-P,)

+dp/dt]/n,

(6)

4 mAh

+

(7)

=

[D(mAb)

d(mAb)/dtlp,

Results and Discussion Following up on an experimental set-up described by Schmialek et al. (1976), we had a 700 ml working volume laboratory scale bioreactor constructed whereby any 142 mm diameter flat membrane can be integrated into the two-part bottom flange. Track-etch polycarbonate screen-type filters that have defined cylindrical pores with 3 pm diameter were used for the in situ retention of cells. The stirrer was positioned 0.5-1.0 cm above the membrane surface and operated at 180 rpm so as to provide adequate tangential flow and to minimize membrane fouling.

35

Initially, we performed an experiment with total cell retention using a dilution rate of 0.4 d- ’ similar to the lowest dilution rates evaluated by us in standard continuous culture without cell retention (Schmid et al., 1987, 1992a). In these

(A) VIABLE CELLS TOTAL CELLS

.

0 . .

D=0.4 R=l

.d’y

-.a-2

,.//-.

o-

/o’ ,t/

, 4

a

6

TIME

10

(d)

(B)

a-

0

2

4

2’

6

TIME

10

(d)

4

Fig. 2. Continuous hybridoma suspension culture rate 0.4 d-‘; total retention (R = 1). (A) Viable glutamine, ammonia and alanine concentrations.

a

6 TIME

0

/

d-’

I I

/

a

10

(d)

with total cell retention. Culture conditions: and total cell concentrations. (B) Glucose, (C)-(E) Concentrations of selected amino

dilution lactate, acids.

36

(D)

0

2

4

6 TIME

a

10

(d)

(E)

TIME

(d)

Fig. 2 (continued).

studies we, like others (Ray et al., 1989; Miller et al., 1988; Boraston et al., 1984; Low et al., 1987; Tovey and Brouty-Boye, 1976) noticed a decline in viable cell concentrations at low dilution rates and a dramatic decrease in viability with stirred tank reactors. Observed elevated specific death and growth rates may be attributed to high toxin and low nutrient concentrations and/or increased hydrodynamic stress due to longer residence times of cells in the bioreactor. A cell retention reactor operated at steady-state conditions in total retention mode (R = 11, i.e., p = k,, is another means of investigating if near zero death rate maintenance of cells can be achieved in such a stirred system. After batch growth of hybridoma cells for 4 days (pmax = 1.32 d-‘) continuous feed into and cell-free product removal from the vessel were initiated with a dilution rate of 0.4 d-‘. At days 8 and 9 viable cell concentrations reached a steady value of = 3 X lo9 cells per ml, whereas total cell concentrations still increased monotonically (Fig. 2A). An estimation of specific growth and death rates from Eqs. (1) and (3) at unsteady-state conditions indicate Al.= k, = 0.60 f 0.06 d-’ at this stage. The specific growth rate falls well below F,,,~~ but later increases again as the value for k, increases from 0.15-0.20 d-’ (days 6 and 7) to 0.60 d-r. Interestingly, this coincides with the exhaustion of major nutrients glucose and

37 TABLE

1

Total cell retention compared to standard continuous culture data (A) Metabolic quotients for substrates, by-products and product (antibody) Substrate/ by-product/ product

Concentration (mM)

q (mmol per 10’ cells per d n

Feed h

p=0.63d-’

Glucose Lactate Glutamine Alanine NH, Antibody

13.50 1so 4.90 0.24 0.38 0

@=kd=0.60d-’ R=l 0.10 19.59 0.05 2.01 4.51 98.62

(B) Metabolic quotients for selected amino acids Amino acid Concentration (mM) Feed ’ @=kd=0.60d-’ R=l Aspartate 0.10 0.02 Glutamate 0.19 0.34 0.11 0.01 Asparagine Serine 0.45 0.18 Histidine 0.23 0.01

/L = 1.24 d-’ R=O

1.75 (2.37) 0.65 (0.23) (0.54) (12.90)”

0.77 (0.29) (0.53) (14.29)

7.91 (15.90) 1.64 (0.69) (1.64) (22.64)

q (mmol per 10’ cells per d a p=0.63d-’ p=l.24d-’ R=O 0.010 0.011 (0.012) (0.020) (0.043) (0.172) 0.013 0.018 (0.003) 0.035 0.040 0.087 0.029 N.A. ’ N.A.

Glycine Threonine Arginine Tyrosine Methionine

0.51 0.61 0.29 0.32 0.22

1.10 0.31 0.01 0.17 0.01

(0.077) 0.039 0.037 0.020 0.027

(0.091) N.A. 0.052 0.021 0.035

0 0.148 0.097 0.045 0.057

Valine Phenylalanine Isoleucine Leucine Lysine

0.69 0.37 0.70 0.68 0.63

0.25 0.18 0.23 0.15 0.34

0.057 0.025 0.062 0.069 0.038

0.071 0.021 0.068 0.083 0.043

0.114 0.047 0.146 0.175 0.070

n Values in parentheses indicate production rates; h mean values of 6 reactor feed samples; ’ not available; d specific mAb productivity in mg per 10’ cells per d.

glutamine (Fig. 2B) and a shift in hybridoma metabolism resulting in maxima for lactate and glutamate concentrations (Fig. 2B, C> (assessed by independent methods of analysis) and readjustment of apparent yield coefficients Y,&g,Uc and Y’ By day 8 methionine, histidine, and arginine concentrations also becoGm”‘%l”i;ing (Fig. 2D, E). Indeed, metabolic quotients calculated for the above experiment with total cell retention at days 8 and 9 (CL = 0.60 d-‘1 compare favourably with data obtained from a continuous culture experiment without retention performed at a dilution rate of 0.41 d-’ (p = 0.63 d-l). Table 1 lists the calculated spec. consumption and production rates and compares this data with results obtained in standard continuous suspension culture (R = 0) at p = 0.63 d-’ and 1.24 d-’ (Schmid et al.,

38 TABLE

2

Apparent by-product Total cell retention

yield coefficients compared to standard

continuous

Apparent by-product yield coefficients (mol mol- ‘)

Specific

Y' lUC.glUC

1.40 0.14

1.49 0.13

0.85 0.37 0.03 2.19

0.69 0.38 0.05 2.28

GA.gluc Y~HXGLN CLA.GLN Y~LU.GLN YdLY.SER

growth

culture

data

rate p

0.60*0.06 d-’

0.63 *00.04 d-’

1.24iO.02

d-’

R=O

R=l

2.01

0.09 I .oo 0.42 0.10

0

1992a). For example, aspartate, asparagine, and glutamate are consumed (or much less produced) under these conditions instead of being produced as they are at high specific growth rates. There is also good agreement in the spec. oxygen consumption rate of = 5 mmol per lo9 cells per d. Finally, the estimated apparent by-product yield coefficients Y’ are reflective of a specific growth rate of 0.60 d-’ when compared to standard continuous culture data (Table 2). In conclusion, it seems to be impossible to operate stirred tank bioreactors anywhere near zero growth rate conditions because of increased specific death rates k, inherently associated with such a system. In later experiments the effects of different dilution and retention rates on cell growth, metabolism and antibody formation were investigated (Schmid et al., 1992b). One such study evaluated the effects of a 2.5fold increase in dilution rate. Table 3 lists steady-state values for viable cell and antibody concentrations, specific mAb production rates, and space-time yields. Before and after the step change in dilution rate cells grew at nearly identical specific growth rates of 0.59 d-’ and 0.57 d-‘, respectively. With viable cell concentrations and space-time yields increased by a factor of 2.5 and antibody concentrations and specific productivities unchanged the behaviour of this hybridoma line under the conditions of the

TABLE

3

Continuous hybridoma trations, specific mAb rate

suspension production

culture without and with partial cell retention. Steady-state concenrates and space-time yields before and after a step change in dilution

Steady-state values for Viable cell count (cells ml-‘) Antibody concentration (pg ml-‘) Spec. productivity (mg per 10’ cells per d) Space-time yield (mg I-’ d-‘)

No cell retention (D=0.41d-’ p=0.59d-‘)

2.0x 106 104.1 21.3

42.7

Partial

R=O;

cell retention

(D = 0.97 d-‘; ~=0.57d-‘1 5.1 x 10” 117.6

22.4 114.1

R = 0.61;

39

described in situ cell retention bioreactor seemed to be predictable at least at the observed cell concentrations. Unfortunately, experiments at higher dilution rates, i.e., D = 2 d-’ or 3 d-l, were unsuccessful due to blocking of the membrane. The steady-state and transient responses to a step change in dilution rate with a complete analysis of metabolite (glucose, lactate, glutamine, ammonia, and selected amino acids) and antibody concentrations and metabolic quotients will be published elsewhere (Schmid et al., 1992~).

Acknowledgement G.S. was supported by a postdoctoral scher Austauschdienst (DAAD).

fellowship from the Deutscher Akademi-

References Avgerinos, G.C., Drapeau, D., Socolow, J.S., Mao, J., Hsiao, K. and Broeze, R.J. (1990) Spin filter perfusion system for high density cell culture: production of recombinant urinary type plasminogen activator in CHO cells. Bio/Technology 8, 54-58. Boraston, R., Thompson, P.W., Garland, S. and Birch, J.R. (1984) Growth and oxygen requirements of antibody producing mouse hybridoma cells in suspension culture. Dev. Biol. Stand. 55, 103-111. Brown, PC., Lane, J.C., Wininger, M.T. and Chow, R. (1991) On-line removal of cells from continuous suspension cultures. In: Production of Biologicals from Animal Cells in Culture: Research, Development, and Achievements. Proceedings of the 10th ESACT Meeting (Avignon, France, May 7-11, 1990), Buttetworths, Oxford, pp. 416-420. Custer, L. and Schmid, G. (1987) Unpublished results. Feder, J. and Tolbert, W.R. (1985) Mass culture of mammalian cells in perfusion systems. Am. Biotech. Lab. Jan/Feb, 24-36. Himmelfarb, P., Thayer, P.S. and Martin, H.E. (1969) Spin filter culture: the propagation of mammalian cells in suspension. Science 164, 555-557. Hornbeck, P.V. and Lewis, G.K. (1985) Idiotype connectance in the immune system II. A heavy chain variable region idiotope that dominates the antibody response to the p-azobenzenearsonate group is a minor idiotope in the response to trinitrophenyl group. J. Exp. Med. 161, 53-71. Low, K.S., Harbour, C. and Barford, J.P. (1987) A study of hybridoma cell growth and antibody production kinetics in continuous culture. Biotechnol. Tech. 1, 239-244. Miller, W.M., Blanch, H.W. and Wilke, CR. (1988) A kinetic analysis of hybridoma growth and metabolism in batch and continuous suspension culture: effect of nutrient concentration, dilution rate, and pH. Biotechnol. Bioeng. 32, 947-965. Ray, N.G., Karkare, S.B. and Runstadtler, P.W., Jr. (1989) Cultivation of hybridoma cells in continuous cultures: kinetics of growth and product formation. Biotechnol. Bioeng. 33, 724-730. Reuveny, S., Velez, D., Miller, L. and Macmillan, J.D. (1985) Comparison of cell propagation methods for their effect on monoclonal antibody yield of fermentors. J. Immunol. Methods 86, 61-69. Sato, S., Kawamura, K. and Fujiyoshi, N. (1983) Animal cell cultivation for production of biological substances with a novel perfusion culture apparatus. J. Tissue Cult. Methods 8, 167-171. Scheirer, W. (1988) High-density growth of animal cells within cell retention fermentors equipped with membranes. In: Spier, R.E. and Griffiths, J. (Eds.), Animal Cell Biotechnology, Vol. 3, Academic Press, New York, pp. 263-281. Schmialek, P., Geyer, A., Miosga, V., Nundel, M. and Zapf, B. (1976) An automatic apparatus for continuous suspension culture with a concentrating device. Biotechnol. Bioeng. 18, 1497-1506.

Schmid, G., Wilke, C.R. and Blanch, H.W. (1987) Continuous suspension culture of hybridomas with cell retention. Presented at the 194th ACS National Meeting, New Orleans, LA, U.S.A., paper no. 74. Schmid, G., Wilke, C.R. and Blanch, H.W. (1992a) Suspension culture of hybridoma ceils in Hepesbuffered medium: metabolism and product formation in batch and continuous culture. Biotechnol. Bioeng., in preparation. Schmid, G., Wilke, C.R. and Blanch, H.W. (1992b) Continuous suspension cultures of hybridoma cells using in-situ cell retention. Biotechnol. Lett., in press. Schmid, G., Blanch, H.W. and Wilke. C.R. (1992~) Continuous hybridoma cultures with cell retention: steady-state and transient responses to a step change in dilution rate. Enzyme Microb. Technol.. submitted for publication. Shimazaki, K., Nishimura, N. and Shibai, H. (1986) New separation system for animal cells with large-scale centrifugation. Ann. N.Y. Acad. Sci. 469, 63-72. Shipman, C., Jr. (1969) Evaluation of 4-(2-hydroxyethylj-I-piperazineethane-sulfonic Acid (HEPES) as a tissue culture buffer. PSEBM 130, 305-310. Takazawa, Y., Tokashiki, M., Murakami, H., Yamada, K. and Omura, H. (1988a) High-density culture of mouse-human hybridoma in serum-free defined medium. Biotechnol. Bioeng. 31, 168-172. Takazawa, Y., Tokashiki, M., Hamamoto, K. and Murakami, H. (1988b) High cell density perfusion culture of hybridoma cells recycling high molecular weight components. Cytotechnology 1, 171-178. Thayer, P.S. (1973) Spin filter device for suspension cultures. In: Kruse, P.F., Jr. and Patterson, M.K., Jr. (Eds.), Tissue Culture Methods and Applications, Academic Press, New York, pp. 345-351. Tokashiki, M., Hamamoto, K., Takazawa, Y. and Ichikawa, Y. (1988) High-density culture of mouse-human hybridoma cells using a new perfusion culture vessel. Kagaku Kogaku Ronbunshu 14.337-341. Tolbert, W.R., Feder, J. and Kimes, R.S. (1981) Large scale rotating filter perfusion system for high density growth of mammalian suspension cultures. In Vitro 17, 885-890. Tolbert, W.R., Lewis, C.J., White, P.J. and Feder, J. (1985) Perfusion culture systems for production of mammalian cell biomolecules. In: Feder, J. and Tolbert, W.R. (Eds.), Large Scale Mammalian Cell Culture, Academic Press, Orlando, pp. 97-123. Tovey, M. and Brouty-Boye, D. (1976) Characteristics of the chemostat culture of murine leukemia L1210 Cells. Exp. Cell Res. 101, 346-354. Van Wezel, A.L., van der Velden-deGroot, C.A.M., de Haan, J.J., van den Heuval, N. and Schasfoort, R. (1985) Large scale animal cell cultivation for production of cellular biologicals. Dev. Biol. Stand. 60, 229-236. Wagner, R. and Lehmann, J. (1988) The growth and productivity of recombinant animal cells in a bubble-free aeration system. Trends Biotechnol. 6, 101-104.