Development of Serum-Free Media:Optimization of Nutrient Composition and Delivery Format

I. INTRODUCTION
In a previous edition, we introduced many of the basic formulation concerns to be considered when developing a serum-free basal medium (Jayme and Gruber, 1998). Since that publication, societal and safety concerns have prompted additional scientific requirements that ultimately impact the construction of serum-free medium. Safety concerns, to be discussed at some length later in this article, now recommend all medium components to be animal-origin free (Jayme, 1999). Consistent with these guidelines, every medium constituent must document a traceable history free of primary or secondary contact with animal-origin products. These regulatory requirements apply to all scales of production from developmental to production formats, and are also extended to bioreactor supplementation and to downstream purification processes. This article reviews the impetus for these new requirements and improvements in serum-free medium development consistent with these directives.

Although many practitioners of tissue culture consider their field to be an art, others characterize it as a science. Both positions may be appropriate, considering the circumstances surrounding the ontogeny of medium development. The advantages of cultivating cells and tissues in a defined nutrient medium were first recognized over eighty years ago. However, since its inception, maturation of the tissue culture field may be characterized as a series of fits and starts predicated mostly upon perceived societal, economic, or experimental needs of the time. Tissue culture has now inarguably gained legitimacy as a research tool and has enabled many scientific investigations to reach successful conclusions. Yet investigators have continued to press for more fully defined cell cultivation environments. Theoretically, a fully defined cell culture medium system would eliminate confounding variables and allow the most accurate assessment and description of dependent variables. Although relatively simple in both concept and statement, practical elimination of tissue culture variables has proved to be no trivial undertaking. As the field has matured and cell-based applications have become more complex, there has been a transition from simple salt solutions to simple basal media and to more complex, nutrient enhanced media, all of which required some level of supplementation by animal sera.

The ultimate goal - to eliminate serum and develop effective serum-free or even protein-free or chemically defined media - proved more challenging. Merely eliminating the serum additive proved ineffective due to the broad diversity of serum functions within a cell culture environment. The presumptive serum-free medium formulation must contain all requisite materials to support the synthesis of new cells (proliferation), as well as all the vitamins, minerals, energy substrates, lipids, and inorganic ions essential to maintain all normal and/or genetically engineered physiological functions.

As such, the field of tissue culture may be generally described as always being in the discovery mode. As specific cellular requirements are identified, appropriate changes are made in media formulae to support that particular cellular function. Certain cell types, important to a thorough understanding of normal and dysfunctional physiology, have been nutritionally fastidious and difficult to propagate in vitro. Often, these cell types were more effectively cultured and more persistently retained differentiated cellular function in the absence of serum. Finally, the target end point evolved from supporting cell proliferation or increasing total biomass to optimizing the bioreactor yield of a biomolecule (monoclonal antibody, recombinant protein, virus) of interest.

Innovations in proteomics, gene therapy, and bioand chemoinformatics have also added momentum to serum-free medium development. Societal issues, such as the global concerns regarding bioterrorism (e.g., anthrax, smallpox), immunodeficiency virus (HIV), and transmissible spongioform encephalopathy (prion) contamination, have prompted an amalgamation of scientific and regulatory positions. Public attention has focused on practices relevant to pharmaceutical and vaccine development. As the entire body of tissue culture knowledge focused on these biotechnology applications, analytical techniques facilitated acceleration in the design and development of numerous next-generation serum-free media. As biotechnology era products matured from discovery phases through production and advanced clinical trials and, eventually, into commercialization phases, more regulatory fervor has been engendered, encompassing all portions of cell culture from cell growth and expansion to gene amplification and expression, and biological product harvest and purification.

A. History
In its infancy (1950s), tissue culture was conducted primarily in glass vessels, the matrix of choice at that time. Commensurate with increased popularity in the 1960s, tissue culture systems scaled up to larger formats to include microcarrier and stir tank reactor formats and practices. Although representing a significant step in scale, the tissue culture field nonetheless languished under the daunting sterility practices and requirements associated with reuseable glass vessels. In the 1970s, sterile disposable plastic culture flasks were developed and, after 30 years, remain the mainstay of today's practitioners. In the 1980s, the tissue culture landscape was transfigured by the introduction of tissue culture-derived bioproducts into the commercial marketplace. In the 1990s, tissue culture expanded directly into therapeutic products, including skin replacement for burn patients, cartilage for focal articular surface repair, and experimental cell and gene therapy regimens.

During the last decade and in parallel to the maturation of tissue culture there has been explosive growth in the knowledge and practices pertaining to molecular and cellular biology. As pathological situations have been ascribed to the presence or absence of specific genes or proteins, the fields of genomics and proteomics have assumed both academic and therapeutic importance. The desire to produce specific proteins in culture evoked the ability to select, clone, and express the gene encoding that specific target protein of interest. Once gene expression for the specific protein was verified in a stable cell bank, it became the responsibility of the process development function to maximize product yield through optimized upstream and downstream activities. Optimization of the in vitro cultivation conditions typically involves a careful balance between maintaining the narrowly defined in vivo environment of normal cells and perturbing that environment (physicochemically and nutritionally) to stress the cells into differentially producing copious quantities of the desired bioproduct.

B. Serum as a Culture Additive
This audience is familiar with the evolution of tissue culture beginning with simple salt solutions to simple and more complex media formulations supplemented with serum, extracts, hydrolysates, or peptones. Technical considerations and disadvantages to the use of serum in cell culture applications have been reviewed previously (Jayme and Greenwold, 1991) and included physiological variability, quality control complexity, cellular specificity, downstream processing artifacts, adventitious agent contamination, cost and availability, growth inhibitors, and proteolytic enzyme activity (Freshney, 2000).

A curious paradox concerning the use of serum as a cell culture additive is that cells cannot proliferate in serum alone. The specialized microenvironment required by cultured cells is mimicked in vitro by replacing serum with a mixture of metabolic precursors, macromolecules, and biophysical elements (Ham, 1982; Bottenstein et al., 1979). Serum functions in cell culture (Freshney, 2000; Jayme and Blackman, 1985) and factors to consider in selecting serum for specific applications (Hamilton and Jayme, 1998) have been reviewed previously.

C. Regulatory Impacts
In 2001, the council of Europe adopted regulation (EC) 999/2001 known as the transmissible spongiform encephalopathies (TSE) regulation. Its purpose was to establish principles and guidelines to minimize the risk of transmission of a TSE via human or veterinary medicinal products. This regulation applied to all materials used in the preparation of active and excipient substances and included all source materials and reagents (including culture medium) used in production of an end product. The U.S. Food and Drug Administration (FDA), although not in regulation form as yet, issued letters (1991, 1993, and 1996) and a guidance document (1997), issuing strong recommendations toward the reduction or elimination of all animal-origin components used in the manufacture of FDA-regulated products. These strong regulatory positions have resulted in concerted efforts toward the elimination of serum and other animal-derived factors (e.g., serum albumin, transferrin) and performance qualification of substitute serum-free formulations.

II. CURRENT ISSUES IN NUTRIENT MEDIUM DEVELOPMENT
Failure in transitioning to a presumptive serum-free medium may be attributable to a variety of factors, including ineffective cellular adaptation and cultivation protocols or suboptimal nutrient composition. Three general approaches may be implemented to achieve a serum-free culture environment: (1) replacing serum with recombinant cytokines and with nonanimal transport or adhesion factors necessary for proliferation or production; (2) adapting or genetically modifying parental cells to reduce or eliminate their requirement for serum specific factors; or (3) developing or supplementing a basal formulation with low molecular weight constituents to yield an enriched, protein-free, biochemically defined nutrient medium optimized for the target application.

A useful exercise at the outset is an analysis of the primary motivation for eliminating serum (Jayme and Greenwold, 1991). Is it because serum is ill-defined and variable from lot to lot so that you are uncertain of all factors that impact your culture environment? Are you concerned with cost and availability of qualified serum additives that "work" in your system? Is your cell type unable to grow with serum supplementation because of overgrowth by contaminating fibroblasts or inhibitory or differentiating serum factors? Does serum contain other elements that mask or inhibit a normal biological function you desire to study? Are you concerned with potential adventitious contaminants or degradative enzymes? Is your project exclusively a laboratory research study or will its results ultimately be transferred to pilot or productionscale environments for diagnostic or biopharmaceutical applications? Results from this self-directed analysis may lead the investigator along different paths toward the development and optimization of serum-free culture medium (Waymouth, 1984; Jayme, 1991).

This section, focuses exclusively on three categories: (1) problematic constituents of basal nutrient media, (2) manufacturing process issues and, (3) concerns regarding cell maintenance under serum-free conditions. We will comment on several emerging trends in a subsequent section.

A. Nutrient Medium Constituents
1. Raw Material Definition and Standardization
Efforts to design effective serum-free formulations are often thwarted by the improper selection of basal components. In many instances, compendial specifications do not exist for certain medium ingredients. Unlike many constituents of classical medium formulations that may have compendial specifications for pharmaceutical use, novel growth and attachment additives frequently used as components of serum-free media remain unstandardized. Some additives are quantitated based upon protein content, whereas others are described in units of activity determined by bioassay performance results. Although percentage purity is frequently reported, there may be substantial variation among suppliers regarding the percentage and type of impurity. The presence or absence as raw material contaminants of certain micronutrients or cytotoxic elements may be amplified under serum-free cultivation conditions.

Some serum-free media formulations contain substantial levels of ill-defined hydrolysates, peptones, or extracts that exhibit lot-to-lot variability in biochemical composition and biological performance properties. For example, plant hydrolysates exhibit variable performance as a function of inherent seasonal growth variations, maturity at harvest, mechanism of harvest, or processing differences. These same additives may also create regulatory concerns regarding adventitious contaminants if primary or secondary processing involved contact with animal-origin materials. If possible without compromising biological activity, substitution of protein-free or chemically defined formulations devoid of macromolecular constituents may offer both technical and regulatory benefits. Protein additives to serum-free media should be synthetic in origin or be treated by processes validated to eliminate adventitious contaminants. Ensure that all medium components will be available from at least one reliable supplier at a cost and consistent quality commensurate with projected application needs.

Even where such compendial descriptions exist, specifications and analytical tests may vary significantly among global sources and suppliers. Despite efforts toward international harmonization of specifications, there remain United States (USP), European, and Japanese pharmacopoeias and an U.S. National Formulary (NP). The USP contains legally recognized standards of identity, strength, quality, purity, packaging, storage, and labeling. The NF includes standards for excipients, botanicals, and other "nondrug" ingredients that could feasibly be present in media formulations. Although there are similarities among the three international pharmacopoeias, there remain procedural or reporting differences in component identification, strength, stability, sterility, or endotoxin determination that may lend themselves to significant formulation or performance differences within serumfree culture environments.

2. Water
An often overlooked and undervalued component is water, the principal constituent of liquid cell culture medium. Given the wide variation in source materials, processing and storage methods, and quality parameters, water could readily qualify as a key variable component of the cell culture environment. Without the protective benefits of elevated serum proteins, variations in water quality could introduce variables that critically impact serum-free cultures, such as beneficial or cytotoxic trace metals, organic materials, bioburden, or bacterial endotoxin. The effects of these various contaminants will vary by cell type and by concentration, but each must be rigorously considered, controlled, and evaluated for effect. A useful guideline might be for investigators to use water that meets "water for injection" (WFI) standards to minimize these variables. We recommend that WFI should be produced fresh using validated procedures and should conform to published quality standards and specifications established through a routine testing program (Freshney, 2000).

3. Dissociating Enzymes
Trypsin is the most common enzyme used for tissue disaggregation or dissociation, as it is well tolerated by a wide range of cell culture applications. However, with the transition to serum-free culture environments without animal-origin constituents, the origin of trypsin materials is problematic. Trypsin is a naturally occurring protein in the pancreas of most mammals. The traditional sources of trypsin have been bovine or porcine pancreatic tissues, neither of which is acceptable under the current regulatory definitions as primarily and secondarily animal-origin free. Raw materials are routinely treated with gamma irradiation to destroy parvovirus and other likely adventitious contaminants.

Various protease alternatives to pancreatic trypsin have been commercialized from fungal, plant, and other nonanimal sources. These dissociating enzyme preparations appear to function equivalently to serine proteases derived from animal tissues and may exhibit other practical advantages for small-scale and production-scale cell culture applications. It is anticipated that these nonanimal trypsin substitutes will reduce the risk of introducing adventitious virus to adherent cultures and will serve as a superior alternative to porcine trypsin for regulated applications.

B. Manufacturing Process Issues
1. Storage and Stability
The functional shelf life of a particular nutrient medium may vary, depending on the professional perspective. A quality assurance professional with an analytical chemistry background might be challenged by the reality that medium formulations do not exhibit a "potency" analogous to a pharmaceutical agent. Nutrient media are formulated to deliver a reasonably homogeneous range of 30-70 different constituents. However, process variables, such as formulation water temperature, speed and duration of mixing, and filtration media, will differentially impact the postprocessing active concentration of the various ingredients.

The postfiltration stability of medium constituents is also highly variable. Some ingredients, such as ascorbic acid, break down quite rapidly in aqueous medium, whereas other components are quite stable with refrigerated storage. Most nutrient media formulated without glutamine and stored in the dark at 2-8°C should perform stably for 6-12 months. Once L-glutamine or other relatively unstable constituents have been added to the medium, performance should be monitored for individual applications. Spontaneous deamidation of glutamine to yield pyroglutamate and ammonia can be problematic due both to glutamine limitation and to sensitivity of serum-free cultures to ammonia.

Perhaps the most reliable determinations of medium shelf life are derived from cell-specific applications. Such analysis should not be limited to the observation of cellular growth kinetics, but should also include the investigation of biological function, as the production of biomolecules and other cellular functions may be impaired earlier than the proliferative rate.

2. Light Sensitivity
Deterioration of tissue culture medium components by exposure to high-intensity light has been documented since the late 1970s (Wang, 1976; Wang and Nixon, 1978; Spierenburg et al., 1984; Parshad and Sanford, 1977). Fluorescent lighting caused the deterioration of riboflavin, tryptophan (Wang, 1976; Lee and Rogers, 1988), and HEPES buffer (Zigler et al., 1985; Lepe-Zuniga et al., 1987) in medium. Riboflavin (vitamin B2) is broken down by visible and ultraviolet light exposures below 540nm. These photo effects on riboflavin and HEPES are mediated through the generation of superoxide radicals. These free radicals are relatively short lived, but are highly reactive and deleterious to most hydrocarbon moieties in their immediate vicinity, particularly ringed heterocycles. The resultant peroxides are also cytotoxic, particularly in serum-free environments. Of course, cell types exhibit varying sensitivities to light-induced medium cytotoxicity (Spierenburg et al., 1984), but it is generally recommended that medium, chemstocks, and cell cultures be protected from light to minimize the possibility of deleterious photo effects.

3. Lipid Delivery Mechanisms
Lipids play an integral cellular role in signal transduction, cellular communication, and intermediary metabolism. Lipids are an essential, albeit underinvestigated, ingredient of serum-free medium formulations. From a medium development perspective, lipids present unique challenges (Darfler, 1990; Gorfien et al., 2000).

Traditionally, lipid constituents were obtained from animal sources, similar to the lipid elements present in human intravenous feeding solutions. Sourcing biologically active sterols from nonanimal sources to meet customer and regulatory requirements was problematic due to biochemical differences that resulted in a significant loss of bioactivity. Traditionally, ovine cholesterol derived from lanolin has been the primary sterol additive source for cell culture. Synthetic cholesterol and plant-derived sterols have exhibited encouraging performance as substitutes for ovine cholesterol in serum-free culture.

The ability to solubilize lipids in aqueous medium and to maintain them stably following filter sterilization has raised additional challenges. Historically, lipid delivery was accomplished through attachment to albumin or other serum-derived proteins. With the demand to eliminate animal-derived proteins, ethanol dissolution was initially investigated, but it proved less desirable due to limited lipid dissolution capacity and stability. Pluronic-based microemulsions overcame some of these issues (Stanton, 1957; Schmolka, 1977), but the production scale was limited by the capacity of the microfluidization apparatus. Pluronicbased emulsions also exhibited variable stability, and sterile filtration of single-strength nutrient medium following the addition of concentrated lipid emulsion effectively removed all supplemental lipid. Advances in cyclodextrin-based technology appear to have overcome many of these practical limitations to lipid delivery (Walowitz et al., 2003).

4. Vendor Audits
As noted earlier, under serum-free or protein-free environments, cultured cells are exquisitely sensitive to fluctuations in medium quality associated with any component of the manufacturing conversion process. Absent the protective and detoxifying contribution of serum proteins, raw material impurities may exert a greater impact on culture performance. Manufacturing protocols should be consistent with current good manufacturing practices (cGMP). Prompted by quality compliance and regulatory issues, most media manufacturers have instituted a raw material qualification program, including routine vendor audits to assess and control the quality and consistency of individual medium components. Quality documentation for all raw materials and process components should be carefully scrutinized during routine audits of nutrient medium suppliers to ensure compliance with appropriate standards and specifications.

C. Cell Maintenance under Serum-Free Cultivation Condition
Unique constraints exist for cell cultivation in serum-free environments that have been reviewed exhaustively elsewhere (Bottenstein et al., 1979; Ham, 1982; Jayme and Blackman, 1985). Three critical elements are noted briefly here.

1. Adaptation
Experimental procedures for adapting cultures to serum-free medium have been described previously (Jayme and Gruber, 1998). Our global interactions with investigators attempting to transition to serum-free culture suggest that many failures may be attributed to inadequate or inappropriate efforts to adapt cells to a novel exogenous environment. Key concerns include the growth state of the cellular inoculum, cell seeding density, subcultivation techniques, and biophysical attributes of the cell culture system.

Typically, cultures may be transitioned from serumsupplemented medium to serum-free medium over a period of 3-6 weeks, following a weaning protocol that sequentially adapts cells in a proportionate mixture of conditioned and fresh media over a period of multiple subcultures. Following adaptation, cells should be recloned in serum-free medium to establish both master and working cell banks. Adapted cell banks should be verified for consistent biological performance properties and absence of adventitious agents for cGMP applications.

2. Cryopreservation and Recovery
Two principal criteria for successful cell cryopreservation and recovery are to initiate the freeze with a healthy cell population and to ensure that both cryopreservation and recovery procedures minimize cellular insult. Log-phase cultures should be maintained with normal proliferative characteristics for a minimum of three passages in the selected serum-free medium prior to cryopreservation.

Because historical cryopreservation protocols included serum or albumin in the freezing medium, we are frequently asked if the addition of animal origin proteins is required. Our experience indicates that it is not necessary to include these additives to achieve high viability recovery of cryopreserved cells. Cells previously adapted to serum-free medium conditions have been cryopreserved successfully in a formulation consisting of equal portions of conditioned and fresh medium, supplemented with 5-10% (v/v) dimethyl sulfoxide (DMSO) as a cryoprotectant. The cryoprotectant agent is necessary to minimize the disruption of cellular and organellar membranes by ice crystals that form during freezing. To accommodate cellspecific variations, titration of DMSO may be required to determine the optimal cryoprotectant concentration for viable cell recovery.

Recovery protocols remain controversial and may vary by cell type as alternative optimization schedules are developed. A generally recommended procedure is a rapid thawing of cryovials in a 37°C water bath and an immediate dilution of cryoprotectant by inoculating cells into prewarmed nutrient medium. During recovery from a cryopreserved state, cells are particularly sensitive to mechanical disruption. Consequently, vigorous trituration, centrifugation, and other physical stresses should be avoided or minimized. When cellular recovery is evidenced by adequate observable increases in cell density, spin down cells gently and remove as much as possible of the medium containing the residual cryoprotectant and replace on a volume-for-volume basis with fresh prewarmed medium.

3. Adherent Cultures
Attachment-dependent cultures pose additional challenges to the development of serum-free media. Many adherent cell types have the capacity to deposit complex extracellular matrices, utilizing exogenous attachment factors and synthesized glycosaminoglycans for cell-to-matrix attachment purposes. In serumsupplemented medium, such attachment factors (e.g., fibronectin, vitronectin) were contributed by the serum additive. Within a serum-free environment, there are three general approaches to improving cell attachment: (1) selection or genetic modification of the cell line to augment native adherence, (2) modification of substratum properties to facilitate attachment and spreading (Griffith, 2000; Han and Hubbell 1996), and (3) nutrient medium supplementation with attachmentpromoting factors or precursors. Although some progress has been made with medium modification, many potential factors are commercially unattractive from cost and animal origin perspectives and from their propensity to adsorb to container surfaces.

III. EMERGING TRENDS
We have chosen to focus on four trends that are exerting a significant impact on the development of serum-free culture environments: (1) format options for delivering nutrient medium, (2) outsourcing of biological production activities, (3) alimentation options to improve bioreactor productivity, and (4) expanded cell culture applications.

A. Format Evolution
To achieve production-scale economies, biotechnology manufacturers have transitioned to larger and more efficient bioreactors. Previous nutrient medium format options were limited to single-strength liquid medium in bulk containers or to ball-milled powder configurations. However, both of these historical options presented technical and logistical challenges for large-scale biological production applications.

Development of liquid media concentrates (50× subgroupings) facilitated stable solubilization of complex constituents of serum-free medium into a format that was readily reconstituted in either batch or continuous mode to yield production volumes of nutrient medium (Jayme et al., 1992). Liquid media concentrates have been utilized commercially for vaccine and recombinant protein. Combining this technology with an in-line mixing device facilitated commercial production of >30,000 liter batches of liquid medium dispensed directly into bulk containers from common ingredients (Jayme et al., 1996).

To minimize the shipment of water, many production-scale applications prefer a dry format, but encounter performance variations, incomplete solubilization, and hygienic concerns with the hydration of serum-free formulations produced as a ball-milled powder. Milling within a stainless steel hammer mill (FitzMill) overcame technical concerns regarding thermal inactivation of heat-labile components and regulatory concerns regarding sanitization associated with ceramic ball-milling processes.

A novel approach to producing a dry-form nutrient medium was introduced by the application of fluid bed granulation to serum-free formulations, yielding a granular medium format (Fike et al., 2001). This alternative approach, termed advanced granulation technology (AGT), resulted in the homogeneous distribution of trace elements and labile components onto granules that dispersed and dissolved rapidly within a medium formulation tank. Upon hydration, medium granules yield a single-strength nutrient medium with superior biological performance relative to the ball-milled powder format of the identical formulation and with specified pH and osmolality without requirement for manual titration (Radominski et al., 2001).

B. Outsourcing
Given the extended developmental lead time and associated expense from identifying a lead candidate, through process development, clinical investigation and regulatory submission, and eventually culminating in an approved biological product, many companies are choosing to defer capital investment in production capacity or hiring of manufacturing personnel pending regulatory approval and commercial success. Such companies may develop mutually beneficial partnerships with contract manufacturing organizations with the ability to produce gram-to-kilogram quantities of a biological product.

Similarly, many companies have identified core competencies that they uniquely possess and have chosen to outsource noncritical capabilities, finding it financially advantageous to hire, buy, or acquire source technology. Various instances of consolidation within the biotechnology industry have resulted in acquisition by the pharmaceutical industry of entrepreneurial firms with intellectual property assets. Other biotechnology companies have opted to outsource significant requirements to contract research organizations. Others have chosen to defer capital investment in media or buffer kitchens or cell banking capabilities by contracting these ancillary services or purchasing ready-to-use materials.

C. Bioreactors
Inevitably, the transition from bench-scale shake flasks or spinner cultures to pilot or production-scale stirred tank or airlift bioreactors encounters various scale-up challenges. In addition to the classical engineering issues of agitation, gas transfer, and control of temperature and pH, there may be qualitative or quantitative adjustments in nutrient formulation or delivery schedule. Nutrient medium qualified for the batch culture of specific cells may benefit from additional optimization if cultures are to be expanded in a bioreactor that has a linear nutrient flow, such as a hollow fiber or horizontal plate-style bioreactor. Variations in nutrient consumption kinetics dependent on the bioreactor environment, such as mechanical stress, dissolved oxygen content, and gas sparging, may profoundly impact cellular energetics and predispose cells to alternative metabolic pathways that alter cell functionality or product quality.

To extend bioreactor longevity and provide more consistent product quality, process development groups often augment traditional batch bioreactor regimens by supplementing exhausted nutrients in fed-batch or perfusion mode (Mahadevan et al., 1994). While a concentrated nutrient feed cocktail may be developed solely by eliminating inorganic salts and buffering components from the base medium, superior cell viability and biological productivity may be achievable through analysis of spent medium (Fike et al., 1993). Determination of component exhaustion kinetics by quantitative analysis of spent medium produced by high-density cultures can yield valuable information regarding metabolite reduction or enrichment to optimize culture productivity. Nutrient modifications derived from iterative analysis, resulting either in adjustment of the initial formulation or inperiodic or continuous addition of concentrated nutrient supplements, have resulted in enhanced bioreactor longevity and specific productivity (Jayme, 1991). Although initially more laborious, this method often yields a simple, customized nutrient cocktail that permits optimized adjustment of individual nutrients and avoids inhibitory effects resulting from excessive additives or metabolic by-products.

D. Applications to Cell Therapy and Tissue Engineering
Tissue engineering applies engineering and life sciences techniques to develop biological substitutes that restore, maintain, or improve tissue or organ function. Integration of biomaterials and biological scaffolding (Griffith, 2000; Han and Hubbell, 1996) with a suitable cell type(s) and a bathing nutrient fluid can expand progenitor cells along a desirable lineage or repopulate a deficient tissue with healthy cells possessing the desired biological function. Cells may be obtained from autologous, allogeneic, or xenogeneic sources or be derived from immortalized cell lines or stem cell progenitors. Given the intended therapeutic application, nutrient media for each specialized application will ultimately need to be acceptable by both technical and regulatory criteria.

IV. CONCLUSIONS
The justifications for eliminating serum from cell culture are numerous, and serum-free media have now been developed for a broad array of biotechnological applications. The past few years have addressed regulatory concerns regarding the potential contamination of biopharmaceuticals by adventitious agents introduced via medium constituents of animal origin or defects in the manufacturing process. Eliminating all animal-origin components has proved daunting, but numerous protein-free nutrient formulations have been specifically designed and developed to eliminate questionable components without sacrificing biological performance. Coincident to the removal of animalorigin components has been the concomitant increase in the biochemical definition of medium composition. Biochemical definition will ultimately translate to enhanced production consistency for prospective biopharmaceuticals synthesized in animal cell-based bioreactors and for biomaterials designed for tissue repair or regeneration, drug delivery, or genetic therapy. Defined serum-free nutrient media and scaffolding matrices compatible with technical and regulatory requirements are under active investigation for the three-dimensional regeneration of bone, ligaments, skin, blood vessels, nerves, and organ functions. These applications extend into many of today's problematic areas, such as arthritis, diabetes, cancer, cardiovascular disease, congenital defects, or sports injuries.

Over the past decade, a striking series of trends has transformed the landscape of tissue culture. The electronic availability of information and data has reduced response times and established new international standards. Biotechnology and tissue culture have become globalized in both needs and concerns. The synergy of molecular biology techniques into functional genomics and proteomics has revolutionized and condensed the developmental pipeline for cell-based products. Enhanced social and safety consciousness has redefined our fundamental responsibility for generating products and research capabilities that will meet global criteria for scientific, economic, regulatory, and ethical acceptability.

References
Bottenstein, J., Hayashi, I., Hutchings, S., Masui, H., Mather, J., McClure, D. B., Ohasa, S., Rizzino, A., Saeo, G., Serrero, G., Wolfe, R., and Wu, R. (1979). The growth of cells in serum-free hormonesupplemented media. Methods Enzymol. LVIII, 94-109.

Darfler, E J. (1990). Preparation and use of lipid microemulsions as nutritional supplements for culturing mammalian cells. In Vitro 26, 779-783.

Fike, R., Dadey, B., Hassett, R., Radominski, R., Jayme, D., and Cady, D. (2001). Advanced granulation technology (AGT): An alternate format for serum-free, chemically-defined and protein-free cell culture media. Cytotechnology 36, 33-39.

Fike, R., Kubiak, J., Price, E, and Jayme, D. (1993). Feeding strategies for enhanced hybridoma productivity: Automated concentrate supplementation. BioPharm. 6(8), 49-54.

Freshney, R. I. (2000). "Culture of Animal Cells: A Manual of Basic Technique," 4th Ed. Wiley-Liss, New York.

Gorfien, S. E, Paul, B., Walowitz, J., Keern, R., Biddle, W., and Jayme, D. (2000). Growth of NS0 cells in protein-free, chemically-defined medium. Biotechnol. Prog. 16(3), 682-687.

Griffith, L. G. (2000). Polymeric biornaterials. Acta Mater. 48, 263-277.

Ham, R. G. (1982). Importance of the basal nutrient medium in the design of horrnonally defined media. In "Growth of Cells in Hormonally Defined Media" (G. H. Sato, A. B. Pardee, and D. A. Sirbasku, eds.), pp. 39-60. Cold Spring Harbor Laboratory, New York.

Hamilton, A. O., and Jayme, D. W. (1998). Fetal bovine serum lot testing. In "Cell Biology: A Laboratory Handbook" (J. E. Celis, ed.), Vol. 1, pp. 27-34. Academic Press, New York.

Han, D. K., and Hubbell, J. A. (1996). Lactide-based poly(ethylene glycol) polymer networks for scaffolds in tissue engineering. Macromolecules 29, 5233-5235.

Jayme, D. W. (1991). Nutrient optimization for high density biological production applications. Cytotechnology 5, 15-30.

Jayme, D. W. (1999). An animal origin perspective of common constituents of serum-free medium formulations. Dev. Biol. Stand. 99, 181-187.

Jayme, D. W., and Blackman, K. E. (1985). Review of culture media for propagation of mammalian cells, viruses and other biologicals. In "Advances in Biotechnological Processes" (A. Mizrahi and A. L. van Wezel, eds.), Vol. 5, pp. 1-30. A1. R. Liss, New York.

Jayme, D. W., DiSorbo, D. M., Kubiak, J. M., and Fike, R. M. (1992). Use of nutrient medium concentrates to improve bioreactor productivity. In "Animal Cell Technology: Basic and Applied Aspects" (H. Murakami, S. Shirahata, and H. Tachibana, eds.), Vol. 4, pp. 143-148. Kluwer, New York.

Jayme, D. W., and Greenwold, D. J. (1991). Media selection and design: Wise choices and common mistakes. Bio/Technology 9, 716-721.

Jayme, D. W., and Gruber, D. E (1998). Development of serum-free media and methods for optimization of nutrient composition. In "Cell Biology: A Laboratory Handbook" (J. E. Celis, ed.), Vol. 1, pp. 19-26, Academic Press, New York.

Jayme, D. W., Kubiak, J. M., Battistoni, and Cady, D. J. (1996). Continuous, high capacity reconstitution of nutrient media from concentrated intermediates. Cytotechnology 22, 255-261.

Lee, M. G., and Rogers, C. M. (1988). Degradation of tryptophan in aqueous solution. J. Parenteral Sci. Tech. 42, 20-22.

Lepe-Zuniga, J. L., Zigler, J. S., Jr., and Gery, I. (1987). Toxicity of light exposed HEPES media. J. Immunol. Methods 103, 145.

Mahadevan, M. D., Klimkowsky, J. A., and Deo, Y. M. (1994). Media replenishment: A tool for the analysis of high-cell density perfusion systems. Cytotechnology 14, 89-96.

Parshad, R., and Sanford, K. K. (1977). Proliferative response of human diploid fibroblasts to intermittent light exposure. J. Cell Physiol. 92, 481-485.

Radominski, R., Hassett, R., Dadey, B., Fike, R., Cady, D., and Jayme, D. (2001). Production-scale qualification of a novel cell culture medium format. BioPharm 14(7), 34-39.

Schmolka, I. R. (1977). A review of block polymer surfactants. J. Am. Oil Chem. Soc. 54, 110-116.

Spierenburg, G. T., Oerlemans, E T., van Laarhoven, J. P., and de Bruyn, C. H. (1984). Phototoxicity of N-2-hydroxyethylpiperazine- N2-ethanesulfonic acid-buffered culture media for human leukemic cell lines. Cancer Res. 44, 2253-2254.

Stanton, W. B. (1957). Polymeric nonionic surfactants. Soap Chem. Spec. 33, 47-49.

Walowitz, J. L., Fike, R. M., and Jayme, D. W. (2003). Efficient lipid delivery to hybridoma culture by use of cyclodextrin in a novel granulated dry-form medium technology. Biotechnol. Prog.

Wang, R. J. (1976). Effect of room fluorescent light on the deterioration of tissue culture medium. In Vitro 12, 19-22.

Wang, R. J., and Nixon B. T. (1978). Identification of hydrogen peroxide as a photoproduct toxic to human cells in tissue culture medium irradiated with daylight fluorescent light. In Vitro 14, 715-722.

Waymouth, C. (1984). Preparation and use of serum-free culture media. In "Methods for Preparation of Media, Supplements, and Substrata for Serum-Free Animal Cell Culture," pp. 23-68. A. R. Liss, New York.

Zigler, J. S., Jr., Lepe-Zuniga, J. U, Vistica, B., and Gery, I. (1985). Analysis of the cytotoxic effects of light-exposed HEPEScontaining culture medium. In Vitro Cell Dev. Biol. 21, 282-287.