DIALYSES CULTURE 0F MAMkfiALIAN CELLS Thesis for the degree of M. S. MECHEQAN SYATE UfiéVERSfiY Chbfiyi C. Homg - 297}. ‘ “ ‘ “infill-n [J L I B R A R Y 1 Michigan State 3 University OCT 1 9 2001 DIALYSIS CULTURE OF l‘iiAlx IViALIAN CELLS By Chi-byi C. Horng A THESIS Submitted. to Michigan State University in partial fulfillment of the requirements for the degree of 5‘ MASTER OF SCIENCE Department of Microbiology and Public Health ’ ' 1971 _. u u= CL' re 1211 ini dia EllC 81m lact derf and Com l’oir ABSTRACT DIALYSIS CULTURE or IxiAl‘viIxsiA LIAN CELLS By Chi—byi C. Horng In order to determine the feasibility of applying the dialysis technique to mammalian cell culture systems, a growth study was made with L-Strain mouse-fibroblast cells in a coil dialyzer system. To min— imize the possible deteriorating effect of dialysis, a small reservoir—to— culture volume ratio of 2:1 was utilized, with intermittent renewal of the s reservoir medium. Observations relating to dialysis were limited to the late logarithmic or early stationary phase in the growth cycle. A lag period of about 12 hours, observed after dialysis was initiated, was interpreted as being the result of the diluting out of the dialyzable growth factors present in the conditioned medium. The pres- ence of this lag period would not nec‘e’s sarily' affect the ”efficiency of the dialysis application. Following the lag period, logarithmic growth resumed, the glucose‘concentration increased to a higher level, and accumulated lactic acid decreased in the culture. It was thus demonstrated that—stress derived from dialyzable nutrients and metabolic products can be released and cell growth promoted by dialysis. Furthermore, since the serum component of the medium was not necessary in the large medium reser- voir, a significant cost reduction appeared feasible. CHI—BYI C. HORNG However, subsequent dialysis with intermittent renewal of the reservoir medium did not significantly affect growth. The cells at this stage showed low viability, decreased glucose consumption and dinnin- ished lactic acid production. It was concluded that dialysis culture has limitations but these restrictions might be eliminated by inaproving the physical construction of the culture system. Modification of the physical design of the culture vessel and its relevant units apparently is necessary before a final assessment can be made of dialysis culture for the mam— malian cell. It was suggested that consideration be given to the applicability of dialysis systems to monolayer, as Opposed to suspension, culture of primary cells. Problems such as cell dissociation, nutritional environ- ment, culture surface and environmental. regulatory systems were dis- cussed. It was proposed that a model system be employed for further investigation: Mammary secretory cells could be employed as the cell type; the cell dissociation procedure‘éould be improved; and regulatory factors, such as hormones, could be incorporated into the medium in place of serum so that the specialized cell functions might be maintained or triggered in the in vitro culture environment. DIALYSIS CULTURE OF NIAIVIMALIAN CELLS BY ‘51 RI' Chi—byi CIHorng A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of 4.. MASTER OF SCIENCE Department of Nlicrobiology and Public Health 1971 TO MY WIFE HUEY—YU ii ACKNOWLEDG ME N TS I wish to express sincere appreciation to my major advisor, Dr. Philipp Gerhardt, and to the members of my guidance committee, Dr. William F. l\/IcLirnans, Dr. Marvis Richardson and Dr. D. E. Schoenhard, for their patience, cooperation and assistance during the various phases of my graduate training and research. Experimental work and thesis preparation were done in Dr. McLimans' laboratory at Roswell Park Memorial Institute, Buffalo, New York. Thanks are extended to Dr. McLimans for his helpful discussions and invaluable direction, to Mr. Bruce D. Styles for his excellent technical assistance with autoanalysis and to Miss Suzanne J. Zajac for assistance with grammar and typing of this manuscript. The financial support of a fellowship from the Taiwan Provincial Government, Republic of China, is gratefully acknowledged. l. 2. 3. TABLE OF CONTENTS PRELIMINARIES 1.1 List of Tables. . . . . . ....... 1.2 List of Figures . INTRODUCTION . . . ..... 2.1 Historical. . . . . . . . . . . ............ 2. 1.1 Early Development of Tissue Culture . 2.1. 2 Agitated Fluid Suspension Culture . 2.1. 3 Development of Dialysis Culture in Microbial Systems . . . . . . ....... 2. 1. 4 Dialysis Culture of Manamalian Cell Suspensions . ............... 2.2 Motives and Objectives, ............ . . . . MATERIALS AND METHODS, , 3.1 Cell Strain and Stock Culture ........ 3.1.1 Cell Straini, . ........ 3.11.2 Stock Culture. .......... 3.2 ’Culture and Dialysis system, ...... .. . 3.3 Measurements, . . . . . . . . . ...... iv 11 13 17 17 17 18 20 24 TABLE or CONTENTS (CON'I‘II‘JUED) Page 4. RESULTS 25 4.1 Preliminary Establishment of Experimental Conditions. . . . . . . . . .. . . . . . . . . . . . . 25 4.1.1 Cell Clumping and Stock Culture. . . . . 25 4. 1. 2 Bacterial Contamination . . . . . i. . . . 28 4.1.3 Culture Volume. . . . . . . . . . . . . . 30 4. 1. 4 Centrifugation of the Inoculum . . . . . . 31 4.1. 5 Improvement of Agitation .. . . . . . . . 34 4.2 "Growth Trials" of Dialysis Culture . . . . . . . . 39 4. 2.1 Cell Growth in Dialysis Culture . . . . . 39 4. 2. 2 Dialysis Effects on the Exhaustion of Nutrients and Accumulation of Metabolic Products. . . . . . . . . . . . 43 5. DISCUSSION.............'............. 52 5.1 Objectives Achieved in the Experiment. . . . . . . 52 5. 1.1 Applicability of Dialysis Culture to Mammalian Cell Suspensions . . . . . . 52 5.1.2 Discovery ofa Lag Period. . . . . . . . 53 5. 1. 3 Limitation of Dialysis Culture . . . . . . 55 5.1.4 Economics Aspect . . . . . . . . . . . . 56 5. 2 Comparison with a Bacterial Culture System . . . 59 5. 3 Further Extrapolation of Dialysis to Mammalian CellCulture..................... 63 TABLE OF CONTENTS (CONTINUED) CONCLUSIONS........................ ()7 RECOMIAENDATIONS.................... 69 BIBLIOCil-{APi-IY........................ '73 MATERIAL REFERENCES. . . . . . . . . . . . . . . . q. . 82 ‘1-..- vi 1.1 LIST OF TABLES Table Page 1 I Change of viable cell population, glucose and lactic acid concentration during dialysis culture 48 2 Calculated glucose consumption and lactic acid production during dialysis culture 49 3 Cost and yield analySiS—-a comparison of batch vs. dialysis culture 5? vii Figure 10 11 1.2 LIST OF FIGURES Page Schematic of culture vessel assembly 21 Growth curve of L-cell suspension culture in spinner flask , 27 Maintenance of logarithmic growth in the stock suspension culture of L-cells - 29 Effect of culture volume on cell growth 32 Effect of centrifugation of inoculum on cell growth 33 Cell growth in culture vessel with dialysis coil 35 Checking toxicity of medium from culture vessel with dialysis coil by spinner culture 36 Growth comparison between culture vessel with dialysis coil and culture vessel without the coil 38 Dialysis culture and cell growth ' 42 Correlation between cell growth and glucose concentration in the culture vessel and reservoir 44 Correlation between cell growth and lactic acid concentration in the culture vessel and reservoir 46 viii I. 2. IN TR OD UC TION 2.1 Historical _-_ 2.1.1 Early development 'of tissue culture In early methods of cultivation of mammalian cells, a frag- ment of fresh tissue, generally embryonic, was placed in a drop of plasma (11) or saline solution (61), and for a few days thereafter mi- gration and multiplication of cells might be observed. However, the phenomenon was irregular and of Short duration, and no increase in the mass of the tissue was observed. Subsequently, culture of animal tissue cells developed, as Earle has suggested (32), into three general groups: static matrix cultures, static surface substrate cultures, and agitated fluid suspen- sion culture. The first and earliest group, static matrix cultures, owes its development to Carrel's work during the 1910's and 1920's. The prob- lems of providing the cells with the necessary food and removing the catabolic substances from the medium were solved by the use of a bi- phaSic‘medium composed of plasma-clot and fluid medium, by the trans- fer of tissue sections every 48 hours (12) and by the construction of con- tainers and instruments permitting the aseptic handling of the cultures (13). However, disadvantages of the plasma—clot substrate, such as clouding and liquefying of the culture and size limitation, led to the .1.-- -. development of the second group, static surface substrate cultures. As the culmination of a long series of experiments, Earle‘s group grew cells successfully first under perforated cellophane (33) and then di— rectly on the glass floor of the flask (28). Instead of transferring the culture by excising a fragment of the cell sheet and implanting it in a new culture flask, the cell suspension could be prepared from the cell sheet and used as the inoculum (27). Through these efforts, the mass culture of cells as monolayers on glass could be done routinely and reproducibly (30). Since that time these procedures have been gener- ally adopted as conventional tissue culture techniques. While there was no inherent size limitation in the solid sub- strate culture method, the advantages to being able to entirely elimi- nate the surface substrate as a complicating factor in the growth of large cultures were so significant that a new group of experiments was initiated to explore the possibility of obtaining rapidly-proliferating cultures with the cells freely in suspension in the nutrient fluid. Owens and Gey made the first attempt in 1953 by providing the culture tubes with a rapid rotatory tumbling action so as to maintain the medium in constant. motion and the cells--a lymphoblastic strain of tumor origin. which never did grow on glass--in continuous suspension (78). Earle and his co-workers elaborated on this idea of "tumbling cell culture” to establish rapidly’proliferating fluid-suspension cultures of pure strain L cells from the mouse (11). Cultures were maintained in rota- ting roller tubes. Factors such as viscosity, circulation and initial "—H‘ “"r'" cell pOpulation were studied in these fluid-suspension cultures. 2.1.2 Agitated fluid. suspension culture Subsequent developments in the agitated fluid suspension cul- ture can be reviewed from three aspects: design of culture vessels, increase in culture volume and control of culture environment. Various designs in the culture vessel appeared during the early stage of development. Graham and Sirninovitch reported the prop— agation of monkey kidney cell strain in roller tubes rotated around their horizontal axis at 40 to 50 rpm (45); these roller tubes were similar to those used by Earle and his associates for cultivating mouse connective tissue cells (L strain) in suspension. Concurrently, the roller tube was modified by Powell into a hexagonal—Sided roller tube so that the fluid medium successively collected in and was discharged from the angled space between the longitudinal faces of the tubes (82). Good growth of ascites tumor cells was thus demonstrated. Making an imaginative par- allelism between mammalian cell suspension and fermentation in the antibiotics industry, Earle and his colleagues successfully grew L cells in an Erlenmeyer flask mounted on a New Brunswick type shaker (30). Similar techniques have been used by Kuchler and Merchant for stEdying growth cycle of L cells (57). The shaker system was further developed by Earle_e_t_ :11 who, instead of an Erlenmeyer flask, used a flat-bottorn boiling flask with a gas inlet filter and outlet filter provided for continual gassing (31). In addition to the shaker flask cultures, a technique em- ploying a suspended magnetic stirrer bar for agitation was introduced by Cherry and [lull in 1956 (14). A spinner culture apparatus was then developed by McLimans and his colleagues in 1957 (65). The apparatus consisted of a teflon-covered magnetic bar suspended in a Pyrex vessel. The submerged culture thus employed showed an actively proliferating mammalian cell growth, able to support viral growth (23), and the capacity to be extrapolated .to larger types of equipment (65). As a result of the increasing demand for kinetic studies of mammalian cells from the standpoint of nutritional, immunological, biochemical and growth characteristics 'as well as an interest in the feasibility of obtaining cell products, attempts were made to scale up culture systems. Scale—up prototypes were developed with all—glass culture vessels by making minor modifications to increase the size. Thus, modification of Cherry's magnetic stirring bar culture resulted in the development of the stirrer flask, Centrifuge stirrer flask and fil- ter bottom stirrer flask with a capacity to 2 liters (15). The original roller tube was modified into a 6-liter roller bottle rotated.on a ball mi11\‘(1\5). The suspended magnetic stirring bar was placed to rest on the floor of carboy-type bottles which had a volume of up to 9 liters (100). Conversely, the spinner apparatus was modified and described as "minispinner" of a volume from 5 to 50 milliliters; ”orthospinner" types of which are commercially available; and ”magnaspinner" the capacity of which is '4’ and 15 liters (73). McLimans et a1. pointed out that this process is difficult to scale up or to extrapolate to larger ves- sels. Through initial investigation of the relative toxicity of types of construction used in larger fermentor systems as well as determination of satisfactory antifoam. agents (41), they proposed that a 5-liter New Brunswick type fermentor be used for scale-up studies (64). A 20-1iter stainless fermentor used in antibiotic fermentations was soon adapted by them for the culture of man‘imalian cells (102). Many inves— tigators have based their studies on this proposal. Rightsel, McCalpin and McLean have studied cell growth in 5, 7. 5 and 30-liter feririentors (83). Among recent developments in use of these large—scale fernientors, two groups are of particular interest. In Pirbright, England, a culture vessel with automatic pH control was scaled up fr01n 5 liters (90) to a pipe-line connected 30-liter fermentor used for the culture of hamster kidney cells (91) and the production of Serniliki Forest virus from these cells (92). At Roswell Park Memorial Institute, in Buffalo, New York, a pilot plant, initiated with a 6-1iter vibromixer (97) and scaled up to capacities of 20, 50, 250 and 1250 liters, was established for the culture of human leukemia (74). M Initial success with scale-up cultures revealed new problems and questions which had not received serious study. The delicate nature of mammalian cells and their inferior growth rate, compared with that of the microbial system, called attention to the necessity of improving the culture environment so that a better understanding of cell functions might be achieved. Various factors, including medium composition, 4-, gas aeration or overlay, agitation, sterilization, pH, oxidation—reduc- tion potential, OSITIOl'dI‘lty, viscosity and temperature have been ass- sessed , the culn‘iination of which has been the development of con'iplete monitor and autonnatic control systems for both batch culture and con— tinuous cultures. The experimental studies have been well documented, and only selected examples will be cited for illustration. The majority of the culture systems used Eagle's medium (25) with some modification. The medium contains a variety of anaino acids, vitamins, glucose, and balanced salts, with the addition of l-to-lO per cent serum. Methylcellulose was used by Earle and his associates (29), and its protective effect was confirmed by Koza and Motejlova (56). Whole or dialyzed serum seems to be a necessary component of the medium for a consistent growth pattern; the growth factor activity ap— pears to be carried by small micelles of serum proteins (94). Attempts have been made to minimize or eliminate the serum requirement; suc— cess apparently depends on cell type (72). It has been suggested that the serum component may be replaced by pep'tone in autoclavable medi- um (77). Chemically defined medium has been formulated with 0 per \. \ ~ cent serum; in the presence of methylcellulose, the medium was able to provide apparent growth of L cells (48). The importance of controlling the overlaying gas phase of the culture was pointedlvout as early as 1956 by Earlegtgl. (31). Having realized that pH drift was largely due to C02 generated by the cells and C02 equilibration between liquid and gas phase, these investigators ‘1 circulated a constant, slow flow of gas containing COZ through each culture flask (31). Another important component. of the gas phase is 02‘ Its influence on cell growth rate was established by Cooper 33:11.. in 1958 and assessed to be related to (liquid—phase oxygen level or oxi— dation-reduction potential (ORP) (17). Various devices include manual adjustment of gas flow rates of independent supplies of C02 and air (9); automatic control over the surface with adjustable mixtures of C02 and air, with another gas mixture, used as a control, entering the culture liquid directly beneath the impeller (90); and sensitive feedback monitor systems for both pH and p02. Among the latter are a system which controls ambient conditions, described by Thompson_e_._t_al. (93); a gas monitor and control unit, the Meta-Stat, described by HarrisflaL. (46); and an aeration apparatus described by Daniels and Browning (19). Daniels Stall. further emphasized the importance of ORP (21). Results of recent experiments demonstrated that incubation of medium prior to inoculation induces desirable qualities reflected in better growth (22). Higher levels and increased rates of cell growth are associated with the initial QRP before inoculation (21). I Physical factors have a critical effect. Temperature is generally controlled at 37°C by use of an incubator (21); a direct ther- mistor control (20); or a circulating water‘jacket (46). Agitation has been achieved by use of a ball mill (15); a shaker (30); a suspended magnetic stirring bar (65); a magnetic stirring bar resting on the floor (100); a non-suspended magnetic stirring bar rotating at one point on bottom bulges (93); a suspended inipt-llcr (64); and a vibron'iixer uti- lizing the Bernoulli effect (97). Attention has been directed toiviscosity since suspension techniques came into use. The effect of osrnolarity' was noted when it was discovered that different cell types exhibit spe- cific optimum tonicity (79). Concomitant with these development were attempts to control the nutritional substrates via the continuous introduction of fresh nutrients as well as the removal of cellular waste products. One of these systems employs a constant population as maintained by the bal- anced addition and removal of medium and the continuous elimination . or harvesting of the cell crop. The other, however, maintains growth constancy without loss of cell mass. The former is the so-called "continuous cell culture" and the latter is "dialysis cell culture. " Both have been described in relation to their use for propagation of inicro- organisms. In 1950 Monod advanced his concept and theory concerning a continuous culture in microbial systems (71); the feasibility of contin- uous culture of mammalian cells was first reported by Cooper, Burt and .‘Wilson in 1958 (17). Sustained growth in a constant environment was achieved by the continuous introduction of fresh medium at a fixed rate and the simultaneous removal of spent medium and cells. For control of the "fixed rate, ” three methods were designed by Cooper et a1. (18): constant rate of medium flow iniithe chemostat; constant -— cell density in the turbidostat; and a combination in the chen'io—turbidostat. , 9 The cheniostat method v-gas studied further by Cohen and Eagle (16) and by Pirt and Callow (80). 2. 1.3 Development of dialysis culture in naicrobial systen’is Historically, in both microbial systems and mammalian cell culture, dialysis techniques were applied before continuous culture techniques. As early as 1896 Ivfetchnikoff, R03»: and Saliinbeni demon- strated the diffusibility of cholera toxin by implanting cellophane sacs containing Cholera vibrios in the peritoneal cavity of animals (70). Shortly thereafter, this__in fliiapplication was extrapolated to an in vitro situation by Carnot and Fournier (10); a collodion sac containing pneumococci cultures was suspended in a laboratory flask containing ordinary growth medium, and the presence of diffusible toxin was fur- ther demonstrated. In addition, they reported that properties were acquired which were different from those found in ordinary culture, such as greater capsulation, prolonged viability and more persistent virulence. Dialysis culture was subsequently employed in microbial sys- tems i133 wide variety of culture processes which included cell produc- tion, fornaation of nondiffusible and diffusible cell products and inter- biotic culture systems (87). Among these are important contributions by Gerhardt and his associates. Their efforts were directed primarily toward achieving concentrated cultures of bacteria. A biphasic system consisting of a layer of solid agar medium overlaid with a small volume 10 of broth was first emplOyed (96). The solid phase served as a reser- voir both for supplying nutrients and for rei’i'ioving Inetabolic products. The system was demonstrated to successfully increase cell concentra- tion of various bacterial species (96) and of gonococci (39). This con-- cept of a culture system employing a large reservoir phase and a small culture phase was succeeded by a dialysis flask technique in which a semipermeable naeiribrane served to separate a reservoir liquid Inedium and a culture liquid medium (40). In order to achieve independent con- trol of the component Operations and to adapt to the larger scale of size, dialysis fermentor systems were developed in which growth was achieved in a fermentor remote from a nutrient reservoir, but connected with it by conduits and pumps. Dialysis was accomplished with membrane tubing in either the fermentor or the reservoir or, most satisfactorily, with a membrane sheet in a plate—and—frarne dialyzer which was remote from, but connected with, both vessels (38). A thorough review of dialysis culture of microorganisms as well as a theoretical treatnaent of the subject has been reported (87). Some further deveIOpInents are: a differential dialysis culture employing a small intermediate product chamber separated from the culture chamber by a membrane filter and from the reservoir chamber by a semipermeable dialysis membrane for concentrating macromolecular products (47); a demonstration of the applicability of’dialySis culture to the production of diffusible cell products (1) and their kinetic analysis (2);)an’ attempt made to employ dialysis as a means to alleviate the product feedback control (3); and ll an extrapolation of liquid medium dialysis to gas phase dialysis (50). Dialysis culture seems to be particularly feasible for some, poorly growing organisms. The fact that: cultivation of fastidious ini- crobes can be facilitated by using dialysis techniques was recently il- lustrated by Rightsel and Wiygul. l\/iycobacterium lepraeniuriuni was ——._.. _-__._....._— ”-5— .—_-..._.—._—— demonstrated to grow in a cell-free environment in either an in vivo implanted or an in vitro maintained cell-irnperrneable diffusion chain- ber (84). 2.1. 4 Dialysis culture of mammalian cell suspensions Essentially, dialysis culture is a technique by which cell popu- lations can be cultivated by supplying nutrients from and removing met— abolic products to a reservoir medium through a dialysis membrane while preventing the cells from being diluted out by the medium. This principle was employed in mammalian cell culture by Graff and McCarty as early as 1957, the same year that the spinner apparatus was employed for suspension cell culture. In their cytogenerator, instead of a dialysis membrane, they employed fritted glass candles to introduce nutrients and to remove cellular products. Growth constancy was maintained without loss of cell mass (44). However, in contrast with the well documented culture techniques employed for mammalian cell suSpen- sions, the dialysis system seems to have received less attention than it merits. Among the few early reports was that of Mount and Moore who lZ claimed that incorporating a continuous flow dialyzer to conventional suspension culture vessels provided a satisfactory proliferation for extended periods of time with a conconaitant increase in cell yield and in cell viability over that of non—dialysis control cultures. They re- ported this achievement at a Tissue Culture Association meeting (75), but no detailed data were published thereafter and the dialysis system was not employed in their later work. Cori, in an effort to establish continuous culture of HeLa cells in the chemostat (42) and continuous culture of virus from l-leLa cells in the lysostat (43), reported prom- ising effects from a dialysis system, although this culture system was not detailed. By employing a dialysis culture chamber, Langloisgtal; were able to achieve a high level of virus concentration elaborated by leukemic myeloblasts (58). In these reports, no information was pro- vided about the dialysis culture of mammalian cell suspensions in terms of details of culture techniques and growth patterns. On the other hand, discouraging results were reported by Sommer regarding the possibility of concentrating a mammalian cell population by dialysis. A decreased population of L cells, as compared \ \ to the control, was observed in a dialysis flask with a'reservoir; the culture volume-ratio Of the reservoir to the flask was either 4:1 or 10:1 (89). Z- Z 31911133121913.1533}: Dialysis culture of mammalian cells appears not to have been thoroughly treated. Thus a careful evaluation of the potential of this system, employing precise cell culture techniques, is warranted. Cul- tivation of mammalian cells is more difficult than cultivation of the more rapidly proliferating microorganisms. Yet, the application of dialysis to mammalian cell culture systems would appear more urgent and mean- ingful than in the instance of microbial culture. Interesting aspects which support this view include: 1. One of the puzzling aspects in mammalian cell culture is the limited cell population that can be achieved. It would be of both practical and theoretical interest to determine if this limit can be extended by adapting a dialysis system. This rationale was derived from Gallup and Gerhardt's work which demonstrated that Serratia naarcescens can be cultivated with the dialysis system to a virtually unlimited population with viable cell counts in excess of 1012 cells/nil and partial cell volume of 50 per cent (38‘). 2. Theoretically, the efficiency of dialysis culture is de-_ scribed by the formula: X/Xnd : SrO/SfO (l + Pm Am/prm) (87) where X is the cellpdensity for dialysis culture, Xnd is that of non-dialysis culture, Sr0 is the initial substrate 13 14 concentration in the reserxoir, SfO is that in the fermen- tor, Pm is the pern'ieability coefficient, Am is the total‘ area of the membrane, Vf is the volume of the fermentor, and ’Llln‘l is the maximui'n growth rate constant. The relative benefit of dialysis, as revealed by the term Pm Ain/prnl, favors particularly the slower-growing mammalian cells. With a lowerjunl, the efficiency of the system should be higher for mammalian cells than for microorganisms. 3. Among the various components of the mammalian culture medium, serum presents the greatest obstacle in terms of cost and variability of growth response. It has been sug— gested that the growth—prornoting factors of serum may be non—dialyzable (‘34). Applicability of dialysis culture to man‘imalian cells would mean that the serum component of the medium may be limited to smaller volumes, with considerable decrease in cost. ‘Efforts to eliminate the serum component have not been routinely successful. How- ever, effort has not been directed toward the incorporation into the culture system of a membrane impermeable to serum factors. The present work is an initial attempt to apply a dialysis system to mammalian cell culture. The first and-na'ain objectivewas to determine the feasibility of applying the dialysis concept to cell culture systems. SOPlIlSthaiCd culture systems as employed in current culture techniques were not available; therefore, a rather primitive type of. culture vessel and dialysis design were adapted. Ivforeover, the experimental design was directed toward providing a minimum stress condition to negate the possible deteriorating effect of dialysis. The. study was therefore not primarily concerned with achieving maximum efficiency. The second objective was to clarify a previously-described failure (89), so that the mechanisms involved might be revealed. L— mouse-fibroblast cells were chosen for this study. The success achieved thus far with dialysis culture techniques can be attributed to deveIOpmental work with microbial systems. The present work is an attempt to extrapolate techniques for microbial cul— ture to mammalian cell culture. Consequently, comparisons will be made between the results obtained from this work and those pertaining to the use of Inicrobial systems. This was the third objective. Although suspension culture is an ideal system for culture of mammalian cell lines, it is of little value in the instance of primary cultures. To determine the applicability of dialysis techniques to mann- mal'ian‘ cell culture in general, a further extrapolation should beiniade from suspension culture to monolayer culture. This constituted the fourth objective. The final objective was to discuss the possibility of employing the dialysis cultureitechnique to assist in maintaining cell function and/or differentiation in vitro. Although the dialysis technique alone cannot 16 maintain cell differentiatimii, its intrinsic character of separating dia— lyzable molecules from non-dialyzable ones pern'iits an interesting approach to the in vitro study of cell differentiation. Only through further extrapolation can dialysis culture of rnarn- malian cells be applied to biochemical engineering and/or the utilization of manii‘na‘ian cells in the fermentation industry. This is the ultimate interest of the author. 3. iATlSRlALS A1 ID hf'lETliODS 3.1 Cell. Strain and Stock Culture —-_—— —-———~.——_...—- 3.1.1 Cell. Strain Earle's strain L mouse fibroblast was used in the study. This strain was originated fr01n normal subcutaneous connective tissue of an adult. C3H strain mouse. It had been treated with a carcinogen, ZO- methyl-cholanthrene, for 111 days and then subcultured without the carcinogen (26). The strain was shown capable of producing sarcomas on injection into C311 strain mice (26). L cells have been widely employed as a model cell type since the initiation of agitated fluid suspension culture. Two sources of L cells were used. First, a LB cell strain which had been adapted to suspension culture, was obtained from Dr. W. Munyon of Roswell Park Memorial Institute. ' The cells were subcultured in spinner flasks as stock culture and were employed for some preliminary studies such as effect of centrifugation on inoculation and toxicity of medium. Subsequent to an accidental contamination, the stock culture was terminated. Although the source of'the contamination was afterwards. identified as the serum in the medium, this strain .was replaced by a second strain. Both strains had comparable growth patterns. " The second strain, L-,9_Z9, was kindly provided by Dr. S. Weiss of the Monsanto Company, St. Louis. This strain is the first cloned 17 18 mammalian cell isolated from Earle‘s L strain mouse. fibroblast (86). The cells were adapted to suspension culture and proven to be PPLO- free. 3.1. 2 Stock culture For the stock culture of L-929, both monolayer and suspension subcultures were routinely established. RPMI 1X Eagle 1955 (7 M) was the medium used for monolayer culture. It is a modification of Eagle’s h'iixiiinuinlEssential Medium. The medium contained 10 per cent inactivated fetal calf serum and 100 U/rnl of penicillin and streptomycin. The dissociating agent employed for subculture was 0. 2% trypsin in PBS (8 M). For routine subculture, the T—3f) flask was used. The G-90 was used for mass culture. Aliquots of used medium were mixed with fresh medium to initiate a subculture. Neither HCl nor NaHCO3 was used for pH adjustment; the latter was achieved by raising with aeration or fresh medium and lowering with 5 per cent C02 in com-- pressed air. For routine subculture, a culture split of 1:4 was prac— ticed. A split of 1:2 was sometimes employed to ensure luxuriant growth. — Eagle's Minimum Essential lVledium (9 N1) was used for sus- pension culture. The medium differs from that used for monolayer culture in its high concentration (of phosphate buffer and the absence of l9 calcium salt. The medium was prepared by adding gluta mine to a final concentration of 29. 2 rug/l, penicillin and streptomycin to 100. U/inl,_ and inactivated fetal calf serum to 10 per cent. Routine culture was carried out in a spinner flask at a culture volume of 100 ml. Inoculation was done without centrifugation. With an initial cell population between 1. 5 x lOS/rnl to 2. O x 105 and with via- bility of the inoculum above 95 per cent, cell growth began after inocu- lation without an extensive lag phase. Adjustment of pH was achieved by aeration; exchanging a rubber stopper for a gauze stOpper resulted in a rise of pH. The culture was maintained continuously with sufficient medium replenishment to keep the cells at a logarithmic growth phase. Cell counting and pH measurement were performed daily. 3. 2 Culture and Dial;sis Systeiri _—-__-—__. _. “‘_-_.‘_.—.-_——"_.__- The dialysis systeia was essentially a coil dialyzer placed in- side its Special fermentor. Culture vessels were designed to meet the objectives of feasibility for dialysis operation and maintenance of sterile conditions. The culture vessel is shown diagrammatically in Figure 1. It has been described by Nchimans et al. (66) as monitor, dialyzer and chemostat. The flask is a 500-1111. water-jacketed, pyrex glass vessel. The top is constructed of 316 stainless steel, with appropriate holes provided for insertion of sampling, feeding, circulating and temperature ' sensoring elements mounted in No. 4, 6, or '7 silicone stoppers. The stopper units are held in place by simple "Y" clamps. Appropriate connectors are provided for gas flow by means of stainless steel tube- ports. The stainless steel top is securely fitted to the flange neck of the flask by a collapsible under-ring. A gas-tight seal is assured by means of a recessed Teflon gasket on the underside of the steel head. Dialysis tubing-~Visking regenerated cellulose (16 M) with an average pore size of 24 A), flat width of 1 cm, total length of 170 cm, surface area of 340 cm2--was inserted on a type 316 solid stainless steel coil. The coil dialyzer rested on or was suspended in the culture vessel. The two ends of the dialysis tubing were connected with rubber tubing sleeves to glass tubingsmounted on silicone stoppers. Tie straps (11 M) were used around-the rubber tubing sleeves so that a tight 20 21 . c -.\_ a . Q 3’ .gvux-flfi‘.‘ wast?" - 50‘“ -1101. , o 9‘?) ’ ‘3}‘~U ) '+ .0. c.‘ t ‘ --—_ —_- —---- ..- - .J I I I L .-- l ‘ ‘ ' u I. .' - - . . .‘ i' - ,iru‘AVM” “v i U U ‘ I --,----Ja_~_- A .l 'l - I I . - - m I I I i ‘1 —’ OI s O- - —-- \ q--- :( ’ —---.cn- - --~l ’ ' l ’-¢—----PP-- ’0 I - .m:--------‘-‘JC“."I‘ -- I ’ II || - -- C Q, ...... u-OQJaqb‘.-fl ‘5 a.-- -" ~-m-‘-‘----" ‘.r\ ’ \ ! k I \ 7 “ . )‘\""~~--‘--'-‘--P . “ I \ \ \ \- I I an. I 5 on---‘~‘-~h--~‘~"‘ - .0-- -.p-c::—II:J-- ”" q \ l I -' § \ (" ‘W—----¢-‘ ‘: i‘ ‘ . ' l‘ u‘ -:"‘ l . \‘ -‘s--~-~c_--"' 'r .‘ ~~~..------."‘-"‘\ \ ‘ . ' I I \ " ~-_~~_---__-o ' o " ' u- ‘0- ‘-------------— / Figure 1. Schematic of culture vessel assembly showing: 1. culture vessel; 2. reservoir medium outlet; 3. reservoir medium inlet; 4. sterile connector for sampling; 5. feeding and inoculation unit; 6. thermometer; 7. gassing unit; 8. gas vent. ZZ connection could be ensured. For suspension of the coil dialyzer, four of the silicone stoppers inside the culture vessel were cut into grooves, The coil was suspended under the silicone stoppers by use of 4 Silk OO Sutpacks (5 Ni) which were fastened around. the grooves. Another vessel, of the same size but without the coil dialyzer, served as the reservoir vessel. Feeding of medium and inoculation were performed through another glass feeding bottle. The feeding bottle was connected with the culture vessel by silicone tubing. Sampling was carried out with a 5 m1 syringe via a sterile connector (Figure 1). Circulation of the medium between the reservoir and the dial- ysis tubing was achieved by use of a peristaltic pump (1 M) with a flow rate of 55 i 5 ml per minute. The vessel was gassed with 5 per cent C02 in compressed air (10 M). The gas passed through a cotton filter, a gas washing bottle and a filtered manifold before entering the culture vessel. Another cotton filter acted as a gas vent. Flow of gas was adjusted manually to achieve an optimum pH level. \ By use of a circulating water bath (Radiometer 12 M), ternper- ature was maintained'at 37°C in the water-jacketed units, including the CUIture vessel, the reservoir and the gas washing bottle. A thermom- eter was inserted in both vessels for constant temperature check. Connection between thepculture vessal and the reservoir, sam- Pling unit, feeding bottle and gas unit was achieved by use of I. D. 1/8 23 inch silicone tubing (3 M) with the exception of the piece inserted into the pump. For the latter, rubber tubing of 1/8 inch 1. D. was used. Because of the abrasive action of the pump, daily replacen'ient of the rubber tubing was necessary. Agitation was achieved by use of a teflon-coated magnetic stir- ring bar, 1. 5 inches in length, resting onithe floor of the culture vessel. The culture vessel was mounted on a magnetic stirrer (4 M) and the stirring bar rotated'at 100 to 150 rpm. The entire unit, after being cleaned and assembled, was tested for air leaks. It was then autoclaved as an assembled unit at 1210C for 45 minutes. To ensure adequate steriliZation, glass—distilled water was added to each vessel. Thus, the vessels generated their own steam. After autoclaving, the condensed water was withdrawn with a syringe and evaporated to dryness with flowing gas. To initiate a culture, a portion of medium was introduced into the culture vessel and gassed with 45 per cent C02 overnight. After the proper pH' was achieved, the vessel was inoculated with cells and the volume adjusted with medium to give a desired cell population. \. 3. 3 Mir: 2': s ur e m e nt s Routine measurement was made of pH and cell concentration. The former was measured by a lVlicro Electrode Unit pH meter (13 .M), and the latter by trypan blue (6 M) vital staining and a Rosenthal counting Chamber (2 M). An autoanalyzer (15 M) was used for sin'iultaneous automatic analysis for both glucose and lactic acid. For glucose analysis, the method described by Frings, Ratliff and Dunn was eiriployed (35). Ali- quots were quantitatively sampled and separated from each other by air and were drawn through flow tubing. O-Toluidine reagent was added, and they were passed through a single mixing coil, incubated in a heating bath for 40 minutes and cooled through a water-jacketed single coil. Their adsorbance was measured at 630 mu and recorded. For lactic acid analysis, an enzymatic method without dialysis 0f the samples, developed by Hochella and Weinhouse (49), was carried Ont. Samples were drawn and mixed with pH 9. 6 glycine buffer. A dye Solution, 3—p-Nitropheny1-Z-p-iodopheny1-5-pheny1tetrazolium chloride “NIL-and the enzyme reagent—-containing diaphorase, NAT-)3", and lactic dC-L‘hydrogenase-«were added. During an incubation time of 6. 5 minutes, the dye was reduced and the adsorbance which developed was measured at 500 mu. The principle employed is that under the catalysis of lactic dehydrogenase, NAD+ is reduced by lactic acid to NADH. Instead of measuring the ultraviolet adsorbance of the 'NADH, the NADH is used to reduce the INT in another enzyme reaction catalyzed by diaphorase. Z4 4. R 17381.1 LTS ———-——-¢-——-- _ 4.1 Preliminary Establishment of Experimental Conditions 4.1.1 Cell clumi'iing and Stock culture The experiments performed during the course of this study are essentially considered as "growth trials, ” the term used by Humphrey (50) in his dialysis study. For reproducible "growth trials“ a stock culture maintained in an actively growing state was used as the inoculum so that the "growth trials" would have a standardized starting point-- which is of ultimate importance (52). The experimental Operating pro- cedure employed in this study for preparation of the inoculum was based upon the following criteria: 1. The L-929 had been adapted to suspension culture. 2.. It was PPLO—free. 3. It showed a generation time of less than 24 hours. 4. The inoculum had a viability higher than 95 per cent. 5. It initiated growth above 1. 5 x 105/m1. In meeting these criteria, the first difficulty encountered was that \of-cell clumping. Cell clumping is a well—established problem. Although the objective of agitated fluid suspension culture is to grow the cell as an individual unit, cell clumping frequently occurs and has been described as groups of 2. to 4 cells or as large matrices (65). At the beginning of the study, it was found that when a confluent monolayer culture was maintained in an incubator for an additional twow, an... 'w 25 26 days beyond the cmifluent stage, the cells associated themselves to fi— brous material which was visible after trypsinization. The cell line seemed to maintain its originality as a fibroblast and elaborated fibrous material. When these cells were inoculated as a suspension culture, cell clunaping occurred. Cell number in the clump varied from Z to 20. Cluiriped cells generally showed lower viability. Some clumps even ,.,. , showed entirely dead cells. As a result, the generation time was usually ‘1 longer than 30 hours, and cell counting could not be performed precisely. The situation was improved by diluting the cells with fresh me- dium and increasing the rate of stirring, but the clumps could not be "" completely removed. Finally, an attempt was made to transfer the cells back to the monolayer culture. After several subcultures, with a split ratio of 1:2 to encourage luxuriant growth and with subculture at the early stage of the confluent monolayer, the cells were transferred to suspension with satisfactory results. The cells appeared in suspension in the fluid medium as discrete units, with a generation time of about 20 hours. Shown in Figure 2 is a typical growth curve of L-cells in a spinner flask. The generation time indicated in this example was 20 hours; viability was maintained above 96 per cent up to the 75th hour, when the cells were already in the stationary phase. The pH of the culture was maintained generally between 7. 1 and 7. 6. Adjustment of the pH was achieved by simply exchanging the silicone stopper for a sterilized gauze stopper for a few hours. The accumulated C02, which caused the lowering of the pH, equilibrated with the atmosphere. As an Z7 33 ~73 :- ~74 375‘ 92‘“ .: ~23 > I. “'72 l6~— 14— {l—n l2“- ,. lO_ 1 .CELL POPULATION ( XI 05/ M L) l I l l l l l l l l 20 40 60 80 100 I20 I40 ELAPSED TIME (HR) Figure 2. Growth curve of L-cell suspension culture in spinner flask. Ina-i 4|. .‘. um |n___u'.v_.-_—. r-”-.-. IA illustration, adjustment of pll from the 38th hour to the 40th hour caused a rise in pH of from 7.1.3 to 7. 42. Because the conventional pH adjust- ing agents (HCl and NaI-lCO3) were not used, an interfering effect was not evident in the adjustment process. Release from the pH stress resulted in a better proliferation, as shown in the 92nd hour in Figure 2. a"? For the stock culture, the cells were generally maintained in the logarithmic phase of growth so that they could be kept in a highly uniform state. Figure 3 shows the maintenance of logarithmic growth, r" together with desirable viability and pH levels. The inoculum used for -' the "growth trials" was fr01n the late stages of the logarithmic phase, I with a cell population of about 1 x 106/ml. 4. l. 2 Bacterial contamination Another difficulty encountered in the beginning stage in the "growth trial" was bacterial contamination. . Procedures employed to locate the source of contamination and to maintain sterile conditions were as follows: “‘“‘ 1. Improvements in the culture system were made, including the use of a sterile connector for sampling and a sterile aspirator as the gas washing unit; autoclaving the connected system as a whole unit; and routinely testing its air-tight- DCSS. 29 VIABILITY (°/o) OO 6— O) .b 8 TOTAL CELL POPULATION (XI07ML) l l l l l l l I l l l 20 40 so 80 100 120 ELAPSEO TlME (H R) Figure 3. Maintenance of logarithmic growth in the stock suspension culture [of L-cells. _ J 2. The improved'culturc system was tested for sterility by running the system with sterile nutrient broth for a day or two. 3. The original LB strain was exchanged for a PPLO—free L-929 strain. 4. Fluid—thioglycolate medium and brain—heart-infusion broth were employed as a routine sterility test for each batch of medium, stock culture and "growth trial" culture. The source of contamination was thereafter traced by means of a sterility test and was found to be in the serum. The medium has since been sterilized by Millipore filtration and dispensed in 100-ml sterile vaccine bottles. 4.1. 3 Culture volume The culture volume employed for a particular vessel appears to influence botht he effectiveness of agitation and the efficiency of gassing. The former arises because different culture volumes give different char- acteristics of agitation. The latter, being determined by the-”ratio of surface to volume, influences, in turn, the oxidation-reduction potential and pH adjustment. The culture volume conventionally employed is be- tween one-half to one-fourth of the total capacity of the culture vessel. For example, Earle_e_t a_l; used 400 ml for a 1. 5-liter flat-bottom boiling flask in the shaker culture (31); McLi-mans_e_3_t_a_l_. used volumes 31 up to 3 liters in the 5-1ll(‘1'clInptllcl‘-Zié_fl.1£tl.€f(l fermentor (64). Because of restrictions iiriposed by the limited availability of culture vessels and dialyzer coils, atterripts were 111.1ch to employ a culture volume of 400 ml so that the entire coil dialyzer could be sub- merged in culture miediui‘n. The result is shown in Figure 4. The "growth trials" were done without the coil dialyzer. In contrast to a desirable growth pattern obtained with a culture volume of 150 ml, the 400-ml culture volume failed to achieve satisfactory growth. 4.1. 4 Centrifugation of the inoculum Conventional techniques of mammalian cell suspension culture use an inoculum of seed cells together with the used medium of the stock culture. In order to have an initial precise growth environment--e. g. , a high glucose concentration in fresh medium and a zero concentration of lactic acid-—efforts were made to initiate the "growth trial" without used inediuma Aliquots of 21 ml of LB cells from the spinner culture, with a cell population of 7. 17 x 105/ml and a viability of 88. 6 per cent, were centrifuged at 1, 000 rpm for 5 minutes and suspended-in 10 ml of fresh medium. The cell suspension was inoculated in a culture vessel without a dialyzer, at a total culture volume of 100 m1. As shown in Figure 5, both the cell population and the viability decreased sharply during the first two hours. Although both were recovered later, cen- trifugation of the inoculum brought about a lag period as long as 60 hours. Inoculation has since been done without centrifugation. TOTAL CELL POPULATION (XIOS/ML) (D A N cc to ISO ML 400 ML 1 l l l l l l l 20 4O 60 80 ‘ELAPSED TIME (HR) Figure 4.6 Effect of culture volume on cell growth. IOO CELL POPULATION ‘( x I 04/940 33 “A? W Ail LE .h I llll I II 20 40 so so ELAPSED TIME (HR) Figure 5. Effect of centrifugation of inoculum on cell growth. 4. l. 5 In'ipr0\.v'ernent of agitatiox'i The agitation system employed in the ”growth trials" was achieved by means of a teflon—coated stirring bar resting on the floor of the culture vessel and rotated via a magnetic stirrer located below the culture vessel. This system appeared satisfactory for growth in a culture vessel without a coil dialyzer. However, the presence of a coil dialyzer in the same vesselgreatly interfered with apparent cell prolif- eration. Repeated attempts to obtain growth in the absence of a dialyzing medium were made using a culture vessel containing a dialyzer coil. As shown in Figure 6, the coil appeared-to have an inhibiting effect on cell growth. It was suspected that this Inight be due either to cell at— tachment on the dialyzer membrane or to the toxicity caused by the dia- lysis tubing and/or by the inside steel coil. Conclusive evidence could not be found for the former possibility; microscopic examination revealed little cell attachment on the membrane. ,To check the latter possibility, a sample of used medium was extracted from, the culture vessel and separated from the cells by centrifugation. The medium was added to a spinner flask; growth was measured and compared to a control. As shown in Figure 7, no toxicity was evident. It was finally discovered that many cell clumps had precipitated on the outside bottom of the coil. The presence of the coil greatly min- imized the dispersing effect of the magnetic stirrer for the portion of the fluid outside the coil. As a result, cells'precipitated and formed VIABILITY (°/o) 5mm TOTAL CELL POPULATION (x10 I 2 —. l l J l J J l l I 20 4O 60 80 I00 ELAPSED TIME (HR) Figure 6. Cell growth in culture vessel with dialysis coil. 3t) I5 -—~ .12. 0 IO - 2.3 Lott: 8L“ 9 ?_< 2 Q l— < .J 2) (L O 0. _J ._J Lu 0 .J . E O l— 2 .. I l l I l l l l 20 4O 60 80 TIME (HR) Figure 7. Checking toxicity of medium from culture vessel with dialysis coil by spinner culture. A———A————A : Spinner culture with the medium from coil vessel. 30 ml inoculum + 30 ml medium from culture vessel + 40 ml fresh medium. . 1... -. . *"r‘vn—Ou ...,. f 0 G 2 Control spinner. 30 ml inoculum + 70 ml fresh medium. clumps. An attempt was then made to improve the stirring effect by suspending the coil with four silk strings so that it. would nOt have direct contact with the bottonn of the culture vessel. This seemed to improve the situation. The culture vessel, with a total capacity of 500 ml, is 5 cm in height. The coil dialyzer was also designed to be of the same height. It was decided to employ a culture volume of 150 ml and to raise the coil so that it was separated from the floor of the vessel by a distance of 0.5 cm. As a result, only about one—fifth of the dialyzer coil was sub-' merged in the medium. Figure 8 shows the growth in the culture vessel containing the coil, as compared to another without a coil. The presence of a dia— lyzer coil in the culture vessel no longer caused significant interference. IOO >_ h“ ”at, A ' t: 96 1'4”“ :1 \— (I) g 9‘" . Ae— 1"! > '—\‘I' 92 DIALYSIS 12 -— m IO _ m 8 '- '/ WITHOUT COIL V‘ ITH COIL 6 TOTAL CELL POPULATION (XIO /ML) 4: I5 I L l J I I I l ' 20 ' 4Q 60 so ELAPSED TIME (HR) Figure 8. Growth comparison between culture vessel with dialysis coil (without running dialyzing medium) and culture vessel without the coil. the arrow. Preliminary dialysis was attempted as indicated by 4. 2 "Growth Trials" of Dialysis Culture _,__~o 1 -—i. 2.1 Cell growth in oiolysis culture On the basis of the following, attempts were niiade to apply the dialysis system at the early stationary phase or late logarithmic phase of growth: Essential effects of dialysis, as revealed in the microbial system, are prolongation of logarithmic growth and in- crease in cell viability (87). It does not significantly alter the slope of the logarithmic growth curve. In other words, before entering the stationary phase, the essential nutrients might not have been exhausted and the metabolites might not have accumulated sufficiently to influence the growth pattern. As a consequence, applying dialysis at the early stationary phase or late logarithmic phase would be as ef— fective as applying it at the beginning. The possibility was considered that there might be some essential growth factors which were not provided by the medium and which would have to be generated by the grow- ing cells themselves. Concentration of these so-called ”conditioned medium factors” should be higher in the sta- tionary phase than in the lag phase. They might be diluted below the required concentration if dialysis were applied from the beginning. To avoid this possible deteriorating --~ 39 ,r 4:0 effect. of dialysis, it appeared preferable to start dialysis at either the early stationary phase or late logarithmic phase. 3. Applicability of; initiating dialysis at the stationary phase was actually suggested by the work of Abbott and Gerhardt (I). In their demonstration experiment of sali- cylic acid fermentation, dialysis was successfully initiated at an. early stationary phase and then repeated by inter- Inittent replenishment of the dialysis reservoir. 4. To reveal the true mechanisna of dialysis culture, initiating dialysis at the stationary phase and successively replen- ishing the reservoir is preferable to applying dialysis cul- ture only once at the beginning. In the former case, metab- olites can be analyzed so that their roles in limiting the rate of cell growth can be a ssessed. The experiment shown in Figure 8 was run simultaneously with two ”growth trials. " One, without a coil, served as a control culture; the other had a dialyzer coil but was carried out under the same condi- tionsx'.‘ .‘After indicating comparable growth patterns in the logarithmic. phase, the cell populations of both cultures were measured every 4 hours at the late logarithmic phase in order to assess the beginning of the stationary phase. After the stationary phase had been reached, at the 59th hour, the control vessel was washed with pyrogen-free sterile dis- tilled water and fed with 300 ml of fresh medium. It was then connectedwr m“ 1‘. with the dialyzer of the other culture vessel and thus served as a res— ervoir, with a reservoir~to—culture volume ratio of 2:1. After appli— cation of the dialysis systei‘n, growth in the dialyzed culture vessel showed a. lag period of about: 12 hours :1 nd then increased. A mechanical accidc it of pumping interfered with continuation of the ”growth trial. ” Another "growth trial" was made using a culture vessel with the dialysis coil but without medium running in the coil. After a normal growth period, dialysis was initiated at the late stage of the logarithmic phase. Again, there was a lag period (if approximately 12 hours followed by a revival of growth at a typical logarithmic rate. As shown in Figure 9, the rate of this renewed logarithmic growth was not less than that of the first one. It can be definitely concluded that dialysis did show a growth—proxnoting effect. This second logarithmic growth did not continue as long as the first one. It was soon followed by another stationary phase. After the stationary phase had been reached, dialysis was attempted by renewing the reservoir medium. There was still some effect of dialysis, al— though the high cell population and low viability made the effect less evi- dent." It should be noted that as the cell population increased, cell viability progressively decreased. This decrease in cell viability was not extensively related to the dialysis process. At the late stage of the ”growth trial, " cell clumps, with the cell number between 2 and 10, appeared. Most of the clumped cells were non—viable. .. IOO -~ (0 O I VIABILITY (°/o) 00 en I (I) O I . “7.3 .\J 0'1 RENEW RESA’IED. I N J) 03 I I O I DIALYSIS \L (I) CElL POPULATION (x IOS/Mo 0') .b. I I I I I I I I I 20 ~’ 40 60 80 too I20 ELAPSED TIME (HR) _ Figure 9. Dialysis culture and cell growth. Initiation of dialysis and renewing of reservoir medium are shown by arrows. The decreasein ccH-vkdnlMy was nota pllcffiafl- lasindi- cated in Figure 9, pH of the culture was i‘I‘iaintaim-rd at a fairly‘stable level. 4.2.2 QDialyshseffects<;10--5. I -i - l l __ I -_ 1 I 12 (43-55) 7.60>:105 44 3 5.83X10'5 37.7 4.96x10-5 1...... -l I I 12 I : (55—67) 9.58x105 40.6 4.24X10‘5 59,3 l 6.19X10‘5 . _L ____J_ L i T M I l I I I 28 ' I l :- i I (67-95) I 1.19X103 43.6 3.66x10-5I 59.1 4.97 x10-5: 1. , __ l 13 (95-108) 1.27x106 I 52.3 4.12x10-5 43.0 3.39X10-5‘ ._,_ 12 (108-12ml 1.32x106 1 19.3 11.46x10-5 17.7 1.34x10-5 ‘ i “J. I . 12 ,(120-132) 1.29x106 24.7 1.91X10'5 12.7 0.98X1O'5 I - l _-______. ._._.__ 50 where —rS is the rate of substrate utilization. due to growth and metab— olism, rg is the rate of growth of the organism, and Y}; and YE are eanirical rate constants. . The first term on the right of the equation represents the portion of substrate used in cell growth. The second term is the portion used for cell maintenance. It can be expected that as the cell population increases, a highi-r amount of substrate will be used for the second term. This is shown in column (1,) of Table 2.. The total glucose consumption per ml per hour increased from 20 ug/ml/hr to 52.. 3 ug/n'il/hr. Because of this maintenance term, the duration of the logarithmic phase achieved with dialysis cannot be as long as that of the first logarithmic phase. After the 108th hour the glucose consumption dropped to an abnormal level. Dividing column (I.) by viable cell con- centration provided glucose consumption given in glucose consumed per Cell per hour, which revealed the importance of the first term, the sub- strate used for cell growth. The highest value was in the beginning of growth, when the cells were in an actively dividing state. The level de— creased to a more or less stable range. Again, after the 108th hour, the cells showed an abnormal nutrient consumption. It seemed that cell\growth could be divided into two phases. During the first phase,, nutrient consumption could be interpreted by use of a mathematical model and cell growth could be prmnoted by dialysis culture. During the second phaSe, on the other hand, nutrient consumption was abnormal and growth was unresponsive to dialysis effect. Lactic acid production is dependent on viable cell population and- -,.._..... _. 51 physiological state of the cells. The latter is revealed in column (1V, ), the lactic acid produced per cell per hour. The values are scattered in a range between 7. 47 ug/cell/hr and 3. 39 ug/cell/hr, except for the rather low level revealed in the abnormal second phase. 5 . DIS C USS l_('.-L‘-i\i 5.1 Objectives it'lcl'iieved in the Experin'ient _— ~.____.__— ——-—-—~ —‘.._—.._—____.__. ._ --._-_——— 5. l. 1 Applicability of dialysis culture to n‘iai'nmalian cell susi'imisions l\»‘iai‘nrnalian cell culture, like most biological processes, is subject to various limiting factors. These factors can be classified as follows: 1. Physical environment. Factors such as agitation, viscos- ity, osmolarity, etc. , fall into this category. These factors, if poorly controlled, will cause cell sedimentation, cell clumping and mechanical injury. 2. Chemical environment. Included in this category are nu- .—-.——. trients, Inetabolic products, pH, oxidation-reduction po- tential, etc. These factors are determined by the nutrients supplied in the medium, the gaseous environment over the surface of the culture and the metabolic state of the cells. The nutrients and metabolic products can be further divided into dialyzable and non—dialyzable. Furthermore, nutri— ents include those which can be supplied in the medium and others which have to be generated by the cells themselves. It is unreasonable to have an overall statement as to whether or not dialysis can be [applied to mammalian cell suspension culture. How- ever, taking into consideration the above limiting factors of mammalian Cell culture, the applicability of dialysis culture can easily be seen. For-- 52 53 example, it cannot be expected that dialysis would have any promising effect on physical environnaent. In fact, the presence of a dialyzer in a culture system might be another limiting factor which would have to be critically considered in the physical construction. Likewise, for the chemical environrnent, dialysis will not be able to dilute out non—dialyz— able metabolic products, to supply non-dialyzable nutrients or to supply nutrients which are not present in the medium. Actually, one of the limitations of dialysis is that it might be possible to dilute out some di- alyzable growth factors which can only be generated by the cells. However, within a certain range, it cannot be denied that dia- lysis can be applied to mammalian cell culture to promote cell growth. As demonstrated in this study, after application of the dialysis system, a new phase of logarithmic growth resumed, exhausted glucose concen- tration increased and accumulated lactic acid was diluted out. In other words, stress derived from dialyzable nutrients and metabolic products can certainly be released by dialysis. 5.1. 2 Discovery of a_lag period \. \ In an attempt to reveal the mechanism of dialysis culture, this study utilized a small volume ratio (1:2) with intermittent renewal of the reservoir medium, and applied dialysis at the late log phase or early stationary phase rather than at the initiation of the culture. A lag period was observed after applying dialysis. This'lag period was not previously 54 reported. It is suspected that the significance of the lag period i'i'iight be related to the n'iea‘:lizitiiis;;n'i which would explain a. prior failure of a dialy- sis triz—zl in a siiriilar experiment (89). A rationale is that there might be some growth factors generated by the cells themselves. These factors might be the ones termed "conditioned medium factors. ” As shown in Figure 5, centrifugation of inoculum diminished these factors so that cell multiplication waned until a sufficient concentration was again gen- erated. For an inoculum size of 10 per cent without centrifugation, the concentration of these factors could be diluted 10 times and still be ef- fective, although there is usually a lag phase. However, if dialysis were run from the beginning, in the case of a volume ratio of 1:10 the factors would be diluted 100 times. As a result, good proliferation would not take place. This is the same rationale that caused the author to apply dialysis at a later stage so that the n'iore populated and highly active cells would generate a higher amount of these factors. Yet, there is still a good opportunity to observe the lag period after initiating the dialysis system. Theoretically, although a lag phase would still be present, the application of the dialysis system at the later stages of growth shouldnot affect the system's efficiency since it is generally at later stages that its growth-promoting effect is shown, namely, prolongation of the log phase and maintenance of higher viability at the stationary phase. 5.1.3 Lin'iitatit‘in of (ll;:‘:‘w:€l5-; culture ._--,_____— M..__—_-.. . } ,-._--._..._—.. _....- -_-. The concept of rate-lii'niting factors should be borne in mind for interpretation of the efficiency of any biologic'tal process. It has been mentioned in one of the previous sections that there are some limiting factors involved in mammalian cell suspension culture. Among these factors are some which can he released by dialysis and others upenwhich dialysis would not have any effect. These factors might excrt their in— fluence simultaneously, or some of them might become predominant at certain stages of growth as rate-limiting factors. This work appears to have demonstrated that the cell population of a mammalian cell culture can be increased by a dialysis system. How- ever, as the cell population increases, the physical construction of the culture vessel assumes a. critical role. In a poorly constructed vessel, it might be more difficult to Inaintain cells in homogeneous suspension without resultant precipitation and clumping. A magnetic stirrer, which grinds the cells underneath it, is another critical factor. As a result of its use, viability of the cells decreases rapidly despite the application of dialysis. ‘ Nevertheless, the difficulties encountered in the present study should not be considered an impediment to further investigations. What is emphasized here is that in the application of dialysis culture, those factors which seem to have no relationship to dialysis should neverthe- less be taken into consideration. If these factors can be made non—rate-limiting, the continuous process of dialysis i‘night‘ promote cell growth to an unlimited concentration. There is a high cell population of blood cells in the blood. vessels, 5. Z x 109 to 4. 5 x 109 for erythrocytes and 5.0 X 106 to 9. 0 x 106 for leucocytes, because of a smooth flow and because of the existence of a smooth layer of endotheliurn; in like manner, it is essential that an improvement in the physical design of the culture vessel be made before assessing the efficacy of improving the chemical environirient through a dialysis system. Among critical factors are the delicate cell membrane which is liable to be destroyed b y drastic stirring and, at the same time, to clump together without efficient stirring, and a larger cell size which, again, calls for a compromise between stirring and sedimentation. All of these problems do not exist in a microbial systeni. 5.1.4 Economics aspect Since’ serum growth-factors are usually described as being protein fractions (94) and are thus non—dialyzable, an attempt was made to use serum—free medium in the reservoir. Applicability of dialysis culture to mammalian cell suspensions, as demonstrated in the present study, would result in restriction of serum to the small Volume of a fermentor. Consequently, there is a noteworthy economical reduction. A cost and yield analysis, based on 80 per cent cell yield efficiency of dialysis culture, is shown in Table 3. The cell population of dialysis, U7 K] Table 3. Cost and yield analysis—-a comparison of batch vs. dialysis culture. . Dialyms Culture System Batch Culture Vessel Reservoir Total Volume (1) 1, 000 1, 000 10, 000 Max. Cell Pop. (106/1441) 1.4 12. 32a -— Total Cells Ha1wested(xlOlZ) 1. 4 12. 32 -- -._ —— —- - “m Weight of Cells Harvestedb (gm) 1, 400 12’ 320 __ u——-—.—— _ — Total. Serurn—free lVIediaC used (1) 900 900 10,000 Total Sera Used—(:71: ----- 100 ”flMT;:-—_“__m*“_m::-_ Total Cost Serum-freed media $450. $450. $5, 000. Total Cost Serae . $3, 750. $3, 750. --$ . Total‘C‘Z‘ost Media $4, 200. . ' $9, 200. Media C08:}:;r gm Cells - $3.00. $0. 75 — —-—— .__._——— Media Cost Ratio: 2l-fold decrease —--- ‘4— Cost Decrease for the Batch Fermcntor Cell'Yield (1. 4 x1012): $3,150. ——-— —— ‘——-— _— —m——.———-—n_ -. ——-—-———.—_——- -‘-_ a. 1.4 x 106x 11, 000/1, 000 x 0.8 = 232 x 107/m1 b. lOé/mg c. 10%serum d. $0.50/1 e. $35.5/1 Figures in b. d. 6. were provided by Dr. l\/IcLirnans. 5:6; calculated fron’i relexmrt data, is not 1.1urely imaginative. In continuous cell culture, hl‘chirne-‘ms (Tl. al., achieved. a cell population to 8-10 x 109/1111, t and Earle et al. , to 20 x 106/1111. There is no reason, through irnprove— ment of physical environn'icnt and thus abolition of rate—liiniting factors, that dialysis cannot achieve an equally high population, or an even higher value since no cell loss is involved. Further evidence was provided by Graham and Silninovitch who, using a strain of monkey kidney cells grown in a suspension system, achieved a concentration of 107 cells/ml through the expediency of Inaking complete medium changes every 24 hours (45). A four-fold decrease in cost is indicated. It should be sufficiently Optimistic to encourage further attempts to improve the dialysis culture system. By employing intermittent renewal of reservoir medium, further cost reduction becomes a possibility. 5. 3 Comparison with '0. Bacterial Culture System .__._~__-_._ .fl---_._ ‘—§———yr--..._ . - *-—.r4_.—.. In addition to being niulticcllulzir and eult-.;-iryotic. as compared with the unicellular and prolzaryotic microorganisms, mammalian cells exhibit seine unioue characteristics which must be tak'm into consider- ation if a bElClZGI‘lFtl culture system is to be applied. Among these char— acteristics are: 1. Naked cell membrane. Vithout a protective cell wall, ——-— the mammalian cell ITICIDl)I‘Eln€ is directly exposed to the culture environment. The delicate cell membrane is liable to be destroyed by drastic mechanical force, fluid dynamic effect and surface tension. Thus, agitation, which is necessary to maintain the cells in suspension, should not be too violent as this will cause mechanical and fluid shearing injury. If the dialyzer-in—reservoir dialysis or the independent-dialyzer dialysis is to be employed, circulation of the cell suspension is necessary, and pre- vention of damage to the cell suspension is dependent on the type of pump used and the speed of circulation. Gassing appears as an overlay on the culture surface since sparging becomes impractical as long as a media serum compbnent is recluired. It has been demonstrated that damage caused by fluid shearing is of the first order of reaction with respect to cell number and that L 929 cells are more 59 60 sensitive than llel-.a. S3 cells (4). lvlore precise measure— ments, including tolerance to. mechanical force and surface tension, appear to be necessary if large—scale cultivation of nianin'ialian cells is to be employed for industrial pur— poses. Another problein related to the cell membrane is that of cell clumping. Basically, mammalian cells are rnulticellular. Aggregation of cells is likely to occur if conditions are favorable. Consequently, cell clumping appears as the cell suspension becomes more concentrated. Larger cell size. In a static fluid suspension, a cell par- ticle is subject to two forces: an upward buoyancy equiv- 4. 3 , 1 ,. , . . . ‘ alent to 3- 7ft {9 and a eounward gravxty equivalent to é’flrDF, where r is the radius of the cell, f is the specific gravity of the cell and [0‘ is the specific gravity of the sus- pension fluid. The net force, equal to;— ‘)Tr3(IO-p‘), causes sedimentation of the cell. By comparing this sedimentation force of mammalian cells with that of bacteria, taking a radius ratio of 10, the difference can be well appreciated. In order to keep the cells in a suspension state, an agita- titm device must be employed, which creates a mobility characterized by Stokes' formula: , 1 Al 2 .___.__.____ 6 7T9 r where u is the mobility for spherical particles, moving in a continuous liquid under the influence of unit force, andg 61 is the \v'iscorz‘ity of the liquid. Again, the larger rnam- malian cells cannot be mobilized by a given agitation syst— tern as efficiei‘itly as the smaller bacteria. Nutritional requirement. The nutritional requirement of mammalian cells is n'iuch more complicated than that of microorganisms. The formulae for culture medium are comprised of the energy source, vitamins, amino acids, and inorganic salts. In addition, various amounts of serum must be employed, with consideration also given to "con- ditioned medium” factors. The growth-pronioting factors in serum are still chemically unknown. In one suspension culture, the serum was replaced by insulin for the growth of L-cells (76) and HeLa cells (7). Conditioned medium is most readily demonstrated via cell clones. The principle was first employed by Sanford, Earle and Likely (86). By means of procedures involving restriction of the volume of culture medium and pre—conditioning of the medium, the L—9Z9 cell strain was isolated as single, completely sep- arated tissue cells (86). The substance in conditioned medium can be produced from either unirradiated or x- irradiated feeder cells (81). Relationship between feeder cells and clone-forming cells does not appear to be cell- type specific (8). ,Conditioned medium from normal leucocytes permits the growth of leukemic marrow cells (51). 617’, A linear relationship was demonstrated bethCn the dose of the conditioned median: and the number of colo- . nies IUl‘lTlll'lg (8). There n‘iight be different entities in different cell types since the substance has been described as heat—stable in mouse embryo cells (81) but theririolabile in chick embryo cells (85). The substance seems to consist of macromolecules characterized as being removable from the medium by prolonged high speed centrifugation (85), precipitable by 50-100 per cent saturation with ammonium sulfate (8), and non-dialyzable (81). In the present study, a lag period was observed after initiating dialysis culture at either late logarithmic growth phase or early stationary phase. The lag period has been interpreted as resulting from conditioned medium. If this interpretation is correct, it appears that the re exists in conditioned medium a sub- stance which is dialyzable. Since the observation is rather indirect, its significance needs to be characterized further. 5. 3 li‘urt'lzer .l‘l‘atrapolation o1.- Di._tli.'s.'is to Tir’lai‘riziuzliiin Cell Culture .__..‘._.—---_-_..- ___—_—__—_.——-._.. --___-- _ .__. n“-..-...-- ._._.- ..____, P‘“ I The experimental work in this s y has clearly deirionstrated that the dialysis technique is grapplicable to mammalian cell suspension culture. llowever, the use of established cell lines for the inamzfacture 1 for use in man is generally forbidden and the approved 1" of products designi- primary cell has not yet been grown in Si.r‘:')(‘l'!.‘3l0n culture despite exten- sive efforts by some. workers. Consequently, it would be valuable to investigate the feasibility of extrapolation of the applicability of dialysis culture of mammalian cell suspensions to monolayer culture of primary cells. Currently, while the Evitro cultivation of primary cells can be done without a great deal of difficulty, what is of challenge is the production of cells which would not only deznonstrate luxuriant growth but which would also maintain their specific function. Achievement of this aim is a prerequisite of fermentative utilization of animal cells, which is a prOSpective field of biochemical engineering. Since this as- pect is of interest, a preliminary effort will be made to determine if a possible approach might be conceived. Prevalent techniques for primary cell culture are still far from being perfect. Questions have been raised about the physiological—sig- nificance of the cultivated cells (68). Thus,_i_n_vitro cultivation of cells leads to dedifferentiation, survivor cells might be those more resistant ones selected from the rather barbaric culture environment, and little 63 (5-1 is known about nutritional factors in serum-—which is an essential ine— diurn cornpoi'ient in current culture techniques. In order to produce cells cmriparable to the cells in the original tissue, it n'iight be essentia to mimic the 113:3; ystem as closely as possible. By comparison of the flvivo and}: :iLrp systems, the following major problems are revealed: 1. Cell dissociation- For the isolation of the individual cell from its original tissue, perfusion agents, mechanical devices and dissociating agents are used either alone or in various combinations. As a result, the cell membrane might be altered and key enzymes or regulating factors might be lost before the cell can be cultivated. Z. Nutritional environment. The importance of the constancy of the internal environment in regulating the activities of living tissue was pointed out by Claude Bernard in as early as 1857 (5). However, the culture environment, in terms of gaseous and liquid medium conditions, is still greatly inferior to theiflone. Of particular interest is the fine distribution and efficient transport of the _i_n_v_i‘\_1_o__circu- latory system which it is almost impossible to reproduce 1211;19- 3. Attaching surface. Tissue cells in vivo are associated with each other either directly or through connective tissue. In the in vitro system, glass or plastic is chosen rather arbitrarily —- —- 65 as the material o which the cells attach. lt many he pos- sible that these artificial surfaces might bring about phys- ical or chemical changes in the cell membrane. Among naturally—occurrin; connective tissue, collagen has been ‘ used in. the in vitro cell culture with promising effects (66). .- —— —-- g...“ 4. legulatory system“ Another in'iportant doctrine developed F) by Bernard, in 181 , is the concept of homeostasis (6). According to him, the environment is not only the product of tissue metabolism; it reacts in turn upon the tissues themselves and regulates their activity. Indeed, honneo- stasis is the unique characteristic of niulticellular organ- isms. It does not take a great deal of imagination to relate in vitro cell dedifferentiation and in vivo cancer to aber- —_ _— w...‘ rations in a connnunication system. If tissue cells can be induced to be autonomous by disturbing the_ig 1&3 regula- tory system (36), how can _i_n_ vitro cells be maintained in a differentiated state without their regulatory system? It is speculated that in'iprovernent of culture environment might be essential, but not sufficient, for maintaining cell functions 1232119. Cybernetics, the study of methods of control and communications which are common to both living organisms and niachin-es (101), naust be the key language to be learned for a full understanding of both' in vivo and in vitro systems. All of these problems are not encountered in suspension culture. ()(i For the extrapolation of dialysis techniques from suspension culture to Inonolayer culture, and for the cultivation of primary cells competent to Inaii‘itain cell functions, each of the prohlerns should be seriously studied. Some of the problems are obviously not pertinent to dialysis culture. However, failure to take them into consideration may create rate-liiniting factors which will make dialysis culture successful but impractical. , (i. CONCLUS .Ol'~~l.‘$ The present work utilized a small volume ratio with inter- mittent renewal of the reservoir medium and applied dialysis at the late log phase or early stationary phase rather than at— the initiation of the culture. The following conclusions may be reached as a result of care— ful interpretation of the experimental data: A lag period of about 12. hours was observed after applying dialysis. This might be the result of a dialyzable substance present in conditioned. medium. Presence of the lag period might not affect the efficiency of the dialysis application. Applicability of dialysis culture was demonstrated by the ob— servation of a resumed new phase of logarithmic growth, an increased higher level of glucose concentration, and a diluted lower level of lactic 7 acid concentration. Thus, the stress derived from dialyzable nutrients and metabolic products can be released by dialysis. Furthermore, economical cost reduction was achieved by the restriction of the expen- sive serum component of the medium to the small culture vessel. There has, however, been only limited success inflt—‘he use of dialysis culture techniques. Nevertheless, some of the limitations can be overcome by improvement in physical construction of the culture vessel. It is essential that physical limitation factors be removed before the chemical efficiency of dialysis culture may be assessed. The present study has only indicated the existence of the 'physical factors and the ()7 .. 68 necessity for inipro: enient, The results of further investigation would be of value to industry. Another interesting potential line for further investigation is the determination of whether dialysis culture of mammalian cell suspen— sions can be extrapolated to primary cell. monolayer culture. Problems such as cell dissociation, nutritional environment, attaching surface and regulatory systems must be considered in the investigation. 7. R FCOI‘viI‘xi .C l‘ZDA 'flO"\I3 As a possible approach, it is proposed that a i'riodel system, be developed which will be useful for further investigation of differentiation functions and respongns to regulatory substances, resulting in decrease in the stress of the culture environn’ient and. enabling consideration of scale-up possibilities. The prog‘osal would hopeful! y lead to experimental. wort: concerning the scale-up of_i_n_. vitro cultivation of differentiated cells so that this differentiated function might be utilized in the field of fer- mentation. Guidelines of the model system are as follows: 1. Cell type. Criteria for a suitable cell type for the model system are responsiveness to known regulatory factors and the existence of well-characterized specialized functions. The mammary secretory cell might be a good choice. At the _i__n_v_i_:p_level, functions of the mammary gland are subject to ovary, adrenal and pituitary regulationg(55). 'At the organ culture level, insulin alone in a chemically defined medium is capable of maintaining the cells in a healthy state. Insulin plus prolactin maintains initial mitotic activity and stimulates alveolar deveIOpment (24.). Among the other regulatory factors affecting its function are hormones such as hydrocortisone, progesterone, estrogen, etc. (62, 95). The uniqueness of the mammary cell in producing several products found nowhere else and 69 70 the vast literature concerning them, the differentiation of this cell in the adult anirii;‘i.l to proliferation and then to milk synthesis, the active rnté'tabolic rate, and the obvious depei‘idence upon external hormonal influence rnalie this cell an attractive location in which to study the control mechanisms of cellular differentiation (60). Culture technique. Among the current superior techniques used for primary culture are those which provide a thin film for gas diffusion with a large volume of medium so that a. constant. environment can be maintained (67). Another factor worthy of consideration is the separation of macro-- molecules from the other components of the medium so that the environment containing small molecular components can be constantly n’iaintained, while macromolecule compo- nents such as hormones can be added or withdrawn at will. It is conceivable that a dialysis system with these intrinsic characteristics might be the best candidate. Thus the plate- and—frame dialyzer might be used as a culture chamber. On one side ofthe chamber, nutrients can be provided from and metabolic products can be diluted into a large reservoir through a dialysis membrane. On the other side, gas can be diffused through a silicone membrane. Essential for this application might be a suitable substratum for the cells to attach. However, theoretical. estimation and 71 1“,)ll‘y'ti‘vlt‘2ll (l e sit", n must first he worked out. Rt‘fflllfil’f)l“)! factors. Although differentiated cells can be . ____ ,-.~__. ¢,-,.--..._,-_..- *v...c.n-_._. cultivated in vitro, these cells lose function in a certain ‘0 period of time (37, 59). The reason might be the loss of or diluting out of inducers, hormones, metabolic intermediates or other factors. Additionally, in; 9.1.3.9. culture tech niques might alter the cell membrane so that responsiveness to thc- regulatory factors might be decreased. The two aspects should be separately treated. A good culture technique should be able to maintain responsiveness. In the absence of regulatory factors, differentiated functions might not be demonstrated even though reSponse to the factors might still be present. friggingcultivated cells might be termed differentiated if they are potentially capable of maintaining or redeveloping their specific characteristics after exposure to the correct environment. It might be possible to use a carrier substratum in organ culture to replace the cell dissociation procedure. The carrier, such as ion exchange resin particles (98), might in turn serve as the inoculum. Regulatory factors can be employed at various culture stages in order to examine the response. A rationale of the necessity for this is that the mechanism of various hormonal activation is generally assumed to involve an allosteric interaction between the hormone and either the “H *v- , auhnile}mda5e or Huginernbrane wfiflixvhnfliitis associ— ated (53). £912H1£L2312 Serunuinxxlinthe convenuonalcmdture Inediunirnufln.conumn boflaessentnfl gromwh.fiuxors and regukfling agents. [Halong:19the SCIUUQ(NNTQNNM&HLC8U~ notlae replacmxl,<:ulh1re naedia are ncn chernically defhaed ammibiochenncalsfiamhes “in preeentdifilcuhfies. Since the serwurilevel hithe nnediunachoes notaflfectimnlk synthe- sisznearly as nquch as fi;affects cellsyurvival or reyflication, it has been speculated that, perhaps indirectly, a low serum level may enhance the maintenance of specific function (59). In fact, hormones have been added to chemically defined medium for the. maintenance of the histological pattern and Unszdveolar secretory appearance oftheimmuhrnary organ (54). BIBLIOGRAPHY 10. 11. 8 . BIBI .305; I? $313113: ABBOT'J‘, B. J. and I). GERFifiRDT. 1970. Dialysis fermentation. I. Enhanced pimduct‘icni of salicylic. acid from naphtl‘elene by PStL‘ltthT‘flOllfif-S flucnwesm-ns. Biotechnol. Bioeng. 3:577-589. ABBOTT, B. J. and P. ClilRllARD'I‘. 1970. Dialysis ferinentz-ition. II. Kinetic analysis of salicylic acid production VII—t cell growth and maintenance. Biotechnol. Bioeng. 12:591—601. ABBOTT, B. J. and P. GERHARDT. 1970. Dialysis fermentation. IlI. Anomalous inhibition of threonine biosynthesis in an auxo- trOphic n'iutant of I‘Ischcrichia coli. Biotechnol. Bioeng. 12: 603—613. AUGENSTEIN, D. C., A. J. SINSKEY and D. I. C. WANG. 1971. Effect of shear on the death of two strains of mammalian tissue cells. Biotechnol. Bioeng. 13:409-418. BERNARD, C. 1857. Lecons sur les proprietés physiologiques des alterations pathologiqucs des liquides de l'organisme. Lefon d'ouverture. 1. Paris. 'Dec. 9, 1957. / \ . BERNARD, C. 1878—9. Le’cons sur les phenomenes de la Vie commune aux aniniiaux et aux vegetaux. Paris. . BLAKER, G. J., J. R. BIRCH and S. J. Pirt. 1971. The glucose, insulin and glutamine requirements of suspension cultures of HeLa cells in a defined culture Inedium. J. Cell Sci. 25529-537. BRADLEY, T. R. and IVI. A. SUIVINER. 1968. Stimulation of mouse bone marrow colony growth_i_n vitro by conditioned medium. Aust. J. Exp. Biol. Med. Sci. 39:607-618. CAPSTICK, P. B., R. C. TELLING, W. G. CHAPIVIAN and L. 9' STEWART. 1962. Growth of a cloned strain-of hamster kidney cells in suspended cultures and their susceptibility to the Virus of foot-and-mouth disease. Nature 195:1163-1164. _ CARNOT, P. and L. FOURNIER. 1900. Recherches sur 1e pneu- mocoque et ses toxines. Arch. lyfed. Exptl. 13:357-378. CARREL, A. and M. T. BURROWS. 1911. Cultivation of tissues _i_n_vitro and its technique. J. Exp. 'lvied. 21:387—396. 73 14. 15. 16. 17. 18. 19. 20. 21. 22. 7 4. CARRICL, A. 1912. On the permanent life of tissue outside of the 01‘85-111'11‘3373‘. J. EXP. Ixied. 15:51.;~~‘:'.'.o. CARREL A. 1923'. A method for the. uh rsiolouical stud r of tissues ’ L 3 (5 in vitro. J. Exp. M‘ed. 38:40'7uél S. CHERRY, Vi". R. and R. N. HULL. 193e, Studies on the growth of Inarnrnalian cells in agitated fluid media. Anat. Rec. 1249183. CHERRY, V’i’. R. and R. N. HULL. 1960. Studies and observations on the growth of marmnalian cells in agitated fluid medium. J. Biochcrri. Niicrobiol. Technol. Eng. 2:267-285. COI’IEN, E. P. and H. EAGLE. 196:1. A sin'iplified cheniostat for the growth of nuammalian cells: characteristics of cell growth in continuous culture. J. Exp. Sled. 113:467-474. COOPER, P. D., A. 1V1. BURT and l. N. WILSON. 1958. Critical effect of oxygen tension on rate of growth of animal cells in continuous suspension culture. Nature 182:1508-1509. COOPER, P. D., J. N. WILSON and A. M. BURT. 1959. The bulk growth of animal cells in continuous suspension culture. J. Gen. Microbiol. 21:702-720. DANIELS, W. F. and E. W. BROWNING. 1962. An apparatus for aeration of tissue cells in suspended culture with controlled gas mixtures. Biotechnol. Bioeng. 4:79-86. DANIELS, W. F., D. A. PARKER, R. W. JOHNSON and L. E. SCHNEIDER. 1965. Controlled pH and oxidation-reduction potential with a new glass tissue-culture fermentor. Biotechnol. Bioeng. 7:529-553. DANIELS, W. F., L. H. GARCIA and J. F. -ROSENSTEEL. 1970. . The relationship of oxidation-reduction potential to the growth ' performance of tissue culture media poised prior to incubation. Biotechnol. Bioeng. 13:409—417. DANIELS, W. F., L. H. GARCIA and J. F. ROSENSTEEL. 1970. Oxidation-reduction potential and concomitant growth patterns of cultures of Earle‘s “L" cells in centrifuge bottle spinners. Biotechnol. Bioeng. 12:419-428. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 75 DAVIS, 13:. V., W. I". IviC LIi‘-.'I_AI*§S, F. L. CLOVER, C. J. KUCERA. D. \V. ZII‘lGLlilll and G. Vt". {Emil-f, 1957. Inwmtigation of the "L” Cell/llcri’ics Simpltx virus ii‘iteraction in sulin'iergcd culture (spinner). 1-‘1‘t:sent.:'d at the April 1957 meeting of the Federation of American Suctieties for Experimental Biology, Chicago, Illinois. DILLEY, W. G. 1971. lvtorphogzenic and mitogei‘iic effects of pro- lactin on rat mammary glands in vitro. Endocrinology 88: 514-517. EAGLE, II. 1959. Amino acid metabolism in mammalian cell cultures. Science 1302437-1137. EARLE, \V. R. 1943. Pm duction of malignancy in vitro. IV. The mouse fibroblast cultures and changes seen in the living cells. J. Nat. Cancer Inst. 4:165-212. EARLE, W. R., V. J. EVANS and E. L. SCHILLING. 1950. Extension of cellophane Substrate procedure to growth of in 1112 cultures of large areas. J. Nat. Cancer Inst. 10:94—3:‘)67. EARLE, W. R., V. J. EVANS, K. K. SANFORD, J. E. SHANNON, JR. and H. K. WALTZ. 19:31. Influence of glass and cellophane substrates on proliferation of strain L cells in tissue culture. J. Nat. Cancer Inst. 12:563-567. EARLE, W. R., E. L. SCHILLING, J. C. BRYANT and V. J. EVANS. 1954. The growth of pure strain L cells in fluid-sus- pension culture. J. Nat. Cancer Inst. 13:1159—1171. EARLE,’ W. R., J. C. BRYANT and E. L. SCHILLING. 1954. Certain factors limiting the size of the tissue culture and the development of massive cultures. Ann. New York Acad. Sci. _5_§3_:1000-10ll. “EARLE, w. R., J. c. BRYANT, E. L. SCHILLING and v. J. EVANS. 1956. Growth of cell suspension in tissue culture. Ann. New York Acad. Sci. onéb-682. — EARLE, W. R. 1958. Long—terni cultivation of animal tissues in large cultures. Fed. Proc. 11:967—974. EVANS, V. J. and W. R. EARLE, 1947. The use of perforated cellophane for the growth of cells in tissue culture. J. Nat. Cancer Inst. §_le3-119. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 76 li‘Rliljl‘vi/‘iN, R. R. 1961. Productiztin of concentrated cell sus- pensions; the sponge fermentor. 'l‘echnol. Eng. 3:333,?-35tl. FRINGS, C. 5., C. R. RA'I‘LIFF an: J. diocheni. I‘viicrobiol. 1R. T. DUNN. 1970. Autonualcd determination of glucose in serum or plasma. by a direct. o—toluidine procedure. Clin. Chem. 16:282—284. FURTll, J. 1969. Pituitary Cybernetics and neoplasia. 1969. PP 417-71. In The thirvey Lectures, Academic Press, New York,l969f- ‘CmuLANL s.rx,zx NUSSBAUhL w. J. IVIC DOUGALL and W. F. MC- LIIWANS. 1970. Studies on hormone production by human fetal pituitary cell cultures. lVIed. 134:27—32. —-".-— GALLUP, D. M. and P. GERIIARDT. Proc. Soc. Exp. Biol. 196 3. Dialysis fermentor systems for concentrated culture of microorganisms. Appl. Microbiol. 11:506-512. GERI-IARDT, P. and C. G. HEDEN. gonococci in clear liquid medium. 103:49-51. GERI—IARDT, P. and D. M. GALLUP. 1960. Concentrated culture of Proc. Soc. Exp. Biol. Med. 1963. Dialysis flask for concentrated culture of microorganisms. J. Bacteriol. 86: 919-929. GIARDINELLO, F. E., W. F. MC LIMANS and G. W. RAKE. 1958. The apparent toxicity of metallic materials of construction and antifoam agents for mammalian cell lines. Appl. Microbiol. _(_)_:30—35. GORI, G. B. 1965. Chemostatic concentrated cultures of hetero- ploid mammalian cell suspensions in dialyzing ferrnentors. 1. Experimental evidence and theoretical consideration. Appl. Microbiol. 13:93-«98. .— GORI, G. B. 1965. Continuous cultivation of virus in cell sus- Appl. Microbiol. _1__3_:909-9l7. pensions by use of the lysostat. GRAFF,S.am1K.S.iM:CARTY. Exp. Cell Res. 12:348-357. 1957. Sustained cell culture. 45. 4:6. 47. 48. 49. 50. 51. 52. 53. 54. 77 GRAUAI-ai, A. 1“. and L. L. Slk'llNO‘v'lTCll. 1955. Proliferation (a 1W‘>' L)OC. b.‘.l_). I. of Inonkey kidney cells in rotating (cultures. Proc. Biol. "fed. 89:326—537. HARRIS, Cz., D. T. MOUT‘W' \‘v', F. MC LlhiANS, K. TUNNAl-“l, S. SCHEELE and G. E. lviOOR' . .I’fi o\‘.’ I a _ I _ it. limit). Czas monitor and SIL"‘.".S . Biotechnol. Bioeng, control unit for cell culture 83 81489-509. HEROLD, J. D., J. S. SCHULTZ and P. GERHARDT. 1967. Differential dialysis culture for separation and concentration of arnacrornolecular product. Appl. Microbiol. 153192-1197. HIGUCIII, K. 1970. An improved chemically defined culture medium for strain L mouse cells based on growth responses to graded levels of nutrients including iron and zinc ions. J. Cell Physiol. 75:65-72. HOCHELLA, N. J. and S. WEINHOL’SE. 1965. Automated lactic acid determination in serum and tissue extracts. Anal. Biochem. 10:304—317. HUMPHREY, H. E. B. 1970. Dialysis culture of bacteria: dialyzer design, nutrient transfer, growth, and dialysis. Ph.D. Thesis, Department of Microbiology and Public Health, Michigan State University. ISCOVL‘, N. N., J. S. SENN, J. E. TILL and E. A. MC CULLOC-H. 1971. Colony formation by normal and leukemic human marrow cells in culture: effect of conditioned medium from human leukocytes. J. Hernatol. 21:1-5. JONES, W. 1., JR., R. B. HOOPER, R. E. LINCOLN and J. P. DOBBS. 1970. Experimental operating procedure for inoculum build—up for tissue-cell pilot plane. Technical Manual, Process Research Branch, Process Development Division, Agent Development and Engineering Laboratories, Fort Detrick, Frederick, Niaryland. JOST, J. P. and H. V. RICKENBERG. 1971. Cyclic AMP. Ann. Rev. Biochein. 49:741-774. JUERGENS, W. (3., F. E. STOCKDALE, Y. J. TOPPER and J. J. ELIAS. 1965. Hormone-dependent differentiation of mammary gland _i_n \Litro. Proc. Nat. Acad. Sci. USA 54:629- 634. ' —‘ 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 78 lx'li\l, U. 1965. Pituitary function and lim‘mo:'i.2il thcr.:ipy of experinmntai breast cancer. Cancer Res. 25:11-46—1161. KOZA, J. and A. RTOTEJ LOVA. 1961. Some conditions of the growth of cells in suspension cultures. Folia Biologica (Praha) 10:143—151. KUCl-lLF. i, R. J. and D. J. l\-'ll:3RCl'IAD-3T. l9 6. Propagation of strain L (Earle) cells in agitated fluid sus Proc. Soc. Exp, Biol. l\r’l(_:d. (1)2:803-806. 5 pens ion c ultures . LANGLOIS, A. J., It. A. BONAR, P. R. mo, D. P. BOLOGNESE, D. BEARD and J. W. BEARD. 1966. BAI strain A (inyeloblastosis) leukosis virus from niyeloblast tissue culture. Proc. Soc. Exp. Eel. Med. 123286—290. LARSON, B. .L. 1966. Biochemical differentiation in the in vitro mammary secretory cell. PP 607-618. In Carcinogenesis-T" a Broad Critique, A collection of papers presented at the 20th Annual Symposium on Fundamental Cancer Research at the Ivl. D. Anderson Hospital and Tumor Institute, University of Texas, Houston, 1966. William and Wilkins, Baltimore, Maryland. LARSON, B. L. 1969. Biosynthesis of milk. J. Dairy Sci. 2%: 737-37417. LEWIS, 1V1. R. and VJ. H. LEWIS. 1912. The cultivation of chick tissues in media of known constitution. Anat. Rec. 6:207—211. LOCKWOOD, D. H... F. E. STOCKDALE and Y. J. TOPPER. 1967. Hormone—dependent differentiation of mammary gland: sequence of action of hormones in relation to cell cycle. Nature 156: 945-946. "'— .‘MARR, A. (3., E. H. NILSON and D. J. CLARK. 1963. Maintenance requirement of E)- (131}; Ann. New York Acad. Sci. 193636-548. MC LIMANS, W. F., F. E. GIARDINELLO, E. V. DAVIS, C: J. KUCERA and G. \V. RAKE. 1957. Submerged culture of mammalian cells: the five liter fermentor. J. Bacteriol. 14:768-774. IVIC LIIVIANS, W. F. , E. V. DAVIS, F. .L. GLOVER and G. W. RAKE. 1957. The submergedculture of mammalian cells: the spinner culture. J. Iirmnunol. 19:428—433. ()6. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. «I \Q MC LII\.-'i.:'XT‘IS, W. F., D. '1‘. l\’f£)l_if\i'l‘, S. llOGITCl-l, E. J. CROLISE, G. llAlltRlS and (j. E. l‘yiOC-RE. 1966. A controlled environment system for study of mammalian c 011 physiology. Ann; New York Acad. Sci. 139:190—213. 1\'lC LIMEHVS, W. F., E. J. CROUSE, K. V. TUNNAl-l and G. E. 1\leORE. 1968. Kinetics of gas diffusion in inarnnqalian cell culture systems. I. Experimental. Biotechnol. Bioeng. 1.9.: 725-740. IVIC LIMANS, W. F. 1969. Physiology of the cultured inaniinaliau cell. PP 307—367. In G. L. Tritsch (ed. ), Axenic l\/laini'nalian Cell Reactions, Marcel Dekker, New York, 1969. MERCI—Ia‘-\NT, D. J., R. J. KUCHLER and W. H. kiUNYON. 1960. Population dynamics in suspension cultures of an animal cell strain. J. Biochein. Ivticrobiol. Technol. Eng. 21:253—365. METCHNIKOFF, E., E. ROUX and T. SALIMBENI. 1896. Toxine et antitoxine cliolerique. Ann. Inst. Pasteur _1_O:257-282. MONOD, J. 1950. La technique de culture continue, theorie et application. Ann. Inst. Pasteur 19:390-410. MOORE, G. E., D. MOUNT, G. TARA and N. SCHWARTZ. 1963. Growth of human tumor cells in suspension cultures. Cancer Res. 38517354741. MOORE, G. E. and K. ULRICH. 1965. Suspension cultures of mammalian cells. J. Surg. Res. 5:270-282. MOORE, G. E., P. l—IASENPUSCH, R. E. GERNER and A. A. BURNS. 1968. A pilot plant for mammalian cell culture. Biotechnol. Bioeng. 12:625—640. .. MOUNT, D. and G. E. NIOORE. 1962. Continuous flow dialysis for environmental control of suspension cell culture. Exerpta Med. I. 16:799. NAGLE, S. C., H. R. TRIBBLE, JR., R. E. ANDERSON and N. D. GARY. 1963. A chemically defined medium for growth of animal cells in suspension. Proc. Soc. Exp. Biol. 1\«Ied. 112: 340—344. " __.._ ORLANDO, M. D., O. C..BOWERSOX, JR. and J. M. RILEY. 1968. Growth of mouse fibi'oblastélike cells in an autoclavable medium. Biotechol. Bioeng. _4_:61-67. 79. 80. 81. 82. 83. 84-. 85. 86. 8'7. 88 89. 8t) 013717.13, O. VON 1-1. , G. O. CE Y and 1.1. If. CEY. 1953. A nee." iiietlwd for the culti\:xt'ion of rnazri‘i'nalian cells suspended in agitated iiiziil rnediui'n. Proc. Arn. Assoc. C.-i;:~.:er Res. 1:411. PIR '1‘, S. J. and E. J. 'l‘l’lACI‘iifRAY. 1964:. Environi‘nental influences on L and ERR rnzii‘ni'n..:_liai‘i cells in shake~11ask cultures. Exp. Cell is. 31:406-112. PIR'I‘, S. J. and D. S. CALLOW. 1964.. Continuous-flow culture of the ERR and L types of mammalian cells. Exp. Cell Res. 33:413-421. ‘PLUZNIK, D. II. and L. SACHS. 1966. The induction of clones of norn'ial mast cells by a substance fl‘OlT) conditioned medium. Exp. Cell Res. 43:553—563. POWELL, A. K. 1954. Culture of ascites tumour cells in \fitro. Brit. Emp. Cancer Campaign, 32nd Annual Report, pp. 123- 127. RIGHTSEL, W. A., H. MC CALPIN and I. W. 1\-1C LEAN. 1960. Studies on large-scale methods for propagation of animal cells. J. Biocheni. l\/Iicrobiol. Technol. Eng._2_:313-325. RIGHTSEL, W. A. and \V. C. WIYGUL. 1971. Growth of Dbfflil‘lfjffljlilll 1£PEEET§111I.iJL@.m cell-impern‘ieable diffusion chairibers. lnIcc. In'nnun. 3:127-—132. RUBIN, H. 1966. A substance in conditioned medium which enhances the growth of small numbers of chick embryo cells. Exp. Cell Res. 31:138-148. SANFORD, K. K., W. R. EARLE and G. D. LIKELY. 1948. The growth_i_n_ ijffio‘f single isolated tissue cells. J. Nat. Cancer Inst. 92229-246. SCHULTZ, J. s. and P. GERHARDT. 1969. Dialysis culture of microorganisms: design, theory, and results. Bacteriol. Rev. _3__3_:1-47. ‘ SIMINOVITCH, L., A. F. GRAHAM, S. 1%. LESLEY and A. NEVILL. 1956. Prepagation of L strain mouse cells in suspension. Exp. Cell Res. 13:299-308. SOMERS, K. D. 1963. Concentrated culture of mannnalian cell suspensions in dialysis flask and biphasic systems. Unpublished report. 81 90. 'I‘EI..I..Il‘{G, R. C. and C. J. STONE. 1964.. A method of automatic pH (30111101 of a bicarbonate—CO;§ buffer system for the submerged culture of l‘iamster kidney cells. Bioteclniol. Bios-11g. 6:14’1-7—158. 91. 'I‘ELLMIG, R. C. and R. ELSW‘ORTH. 1965. Submerged culture of hamster kidney cells in a stainless steel vessel. Biotechnol.. Bioeng. 713117—434. 92. TELLING, R. c., P. J. RADLETT and o. N. MOWAT. 1967. -<-~ in deep culture vessels. Biotechnol. Bioeng. 9:257-265. 93. 'THOMPSON, K. W., R. T. PRICE and S. Z. LEWIN. 1962. Control of ambient conditions in cell suspension cultures. Cancer Chemotherapy Rep. 20:45-55. 94. TOZER, B. T. and S. J. PIRT. 1964. Suspension culture of inani— malian cells and macromolecular growth—pronioting fractions of calf serui‘n. Nature 201:375-378. 95. TUCKER, H. A. 1969. Factors affecting mammary gland cell numbers. J. Dairy Sci. 52:720-729. 96. TYRRELL, E. A., R. E. MAC DONALD and P. GERHARDT. 1958. Biphasic system for growing bacteria in concentrated culture. J. Bacteriol. 15:1-4. 97. ULRICH, K. and G. E. 1\IOORE. 1965. A vibrating mixer for agitation of suspension cultures of mammalian cells. Biotechnol. Bioeng. 12507-515. 98. VAN WEZEL, A. L. 1967. Growth of cell—strains and primary cells on micro-carriers in homogeneous Culture. Nature 216:64-65. 99. VESTERGAARD—BOGIND, B. 1962. Automatic pH regulation of cell suspensions by a gastrometric pH—stat. Scand. J. Clin. Lab. Invest. 1:12461-465. ' \. 100. WALLACE, R.,A. D. ARIS and H. R. cox. 1959. Long-term ’ culture of tissue cells in magnet stirred fluid suspensions. Proc. Soc. Exp. Biol. Med. 101:553-557. 101. WIENER, N. 1948.- Cybernetics, or control and communication in the animal and the machine. The Technology Press, New York, 1948. ' 102. ZIEGLER, D. “7., E. V. DAVIS, W. J. TIIOMLAS and W. F. l\/IC LIME-f? 1958. The propagation of mammalian cells in a 20-1iter stainless fermentor. Appl. IVIicrObiol. 9:305. MATERIAL REFERENCES 10 11 12 13 Ni 1V1 1V1 IVI 1V1 9. 1x1}; TI;‘.1'\“.I.\.L 1115.1?le E l\‘ C ES .'-’_\'mcric;m Instrurnc-nt Cumpai'iy, Inc. S i. 1 v e r S p i' i n 1.1, his. r y] :1. :11. (1 Ha us scr 82 Son C. A. 1\"l£1x Lev y Cole—Paririer Instrument Company 7425 North Oak Pa rk Avenue Chicago, Illinois 606-18 Cole-Parrner Instrument Company 7425 North 0511:. Park Avenue Chicago, Illinois 606—18 Ethicon, Inc. Somerville, New Jersey Grand Island Biological Company Grand Island, New York 14072 Microbiological Associates, Inc. Bethesda, Maryland hiicrobiological Associates, Inc. Bethesda, Maryland Microbiological Associates, Inc. Bethesda, Maryland National Cylinder Gas Division of Chemetron Corporation Chicago, Illinois Panduit Corporation 17301 Ridgeland Avenue Tinley Park, Illinois Radiometer , Copenhagen, Denmark Radiometer Copenhagen, Denmark 82 Aininco Peristaltic Pump Fuchs Rosenthal Ultra Plane Heinocytonieter Silicone Tubing l\/Iicro—V hiagnetic Stirrer Silk 00 Sutupak terile Surgical Silk Type B U.S. P. Trypan Blue Stain (0. 4%) in Hank‘s Balanced Salt Solution RPIVII 1955 1X Eagle's lVIediurn O. 2% Trypsin Solution for RP Minimum Essential klediurn Eagle for Suspension Cultures, without L- glutainine 5% CO2 in Compressed Air Tie Straps Water Thermos tat '1\/Iicro Electrode Unit Type E ' pH 1\/ieter l 4 M 15 NI 16 M I- {eichhold (.‘lwrnicals, Inc, Cuyal‘mga I“?! l I :’~‘ , Ohio 'I‘tix'thnicon Crporation Ardsley, I‘Qew York 10502 Union ‘arbide Corporation Food Products Division 6733 W. 65th Street Chicago, Illinois Latex S or gical Tubing; Auto Anal; 7.0 r Dialysis Tubing (”Visking") l-legeneratcd Cellulose flllll lllllllllllllllll! Mill] Hill 31293 00989 3706