WO1991014774A1 - A method of immobilizing cells onto a support material - Google Patents

Abstract

A process is described for the efficient immobilization of cells to surfaces, for example glass fibre. The process of immobilization is predicted, modified and controlled by an understanding of the interaction of the surface tensions of the cells, the suspending liquid or growth medium and the support substrate. The process of cell adhesion described herein involves culturing the cells and determining their surface tension by standard techniques known to those skilled in the art. The cells of known surface tension are then suspended in a liquid or culture broth of known surface tension, also determined by standard techniques, in which the support material, of known surface tension, is maintained. Under these suitable conditions the cells will rapidly and spontaneously adhere or become immobilized to the support material. The adhesion is sufficiently strong so as to become irreversible after a time under normal, operating conditions.

Description

A METHOD OF IMMOBILIZING CELLS ONTO A SUPPORT MATERIAL

FIELD OF THE INVENTION

The present invention relates to a method for the efficient immobilization of cells on support materials for the production of cell derived biochemicals. More specifically this invention relates to a method for the efficient immobilization of cultured plant cells on glass fibre support material for the production of plant cell derived biochemicals.

DESCRIPTION OF THE PRIOR ART Plants synthesize a diverse range of natural products, many of which are of crucial importance to the medical and pharmaceutical industry. These biochemicals include drugs (digitalis, codeine, morphine, vinblaεtine, vincristine, reserpine, among others) flavours, essences and pigments. Cell culture technology has made it possible to use relatively undifferentiated plant cells to produce some of the more valuable phytochemicals under controlled conditions. In recent years, there has been increasing difficulties in securing a stable supply of the source plants, for these compounds. There has been a decrease in plant resources due to greatly increased demand, increased labour costs, and political or agricultural difficulties in cultivating the source plants. The development of alternative methods for the production of medicinal compounds is necessary and of considerable economic importance. Cell culture and fermentation technology offer a solution to this problem.

By removing cells from the organized structure of the whole plant, cell culture places strict demands on the environmental conditions within a bioreactor. The development of specialized bioprocesses, in which mild agitation is used to produce efficient mixing and a relatively homogeneous cell suspension has allowed for the production of high biomass levels in precisely controlled bioreactors and defined culture media.

However, the production of commodity biochemicals is frequently associated with the late growth phase of plant cells when cell growth declines and cell to cell contact increases with resulting aggregation. ''For this reason interest has focused on promoting cell-cell interaction from suspension culture conditions in order to promote increased biochemical production. This may be accomplished by immobilization of plant cells (the biocatalyst) by physically restraining the cells on or within a support. The immobilization of biocatalysts (e.g. cells, organelles and enzymes) onto, or within a support has become a powerful and well-accepted technology with biotechnological applications in bioroedical and industrial research and development. For example, various biosensors using immobilized enzymes are important diagnostic tools in medicine, and are useful probes in research and industry (Mosbach 1976, 1987a,b, 1988, Methods in Enzy ology, Vols. XLIV, 135, 136 and 137). Likewise, immobilized cells have broad applications in the production of useful biochemicals such as ethanol, amino acids, organic acids, antibiotics, enzymes and hormones, among others used by the commodity chemical, food and pharmaceutical industries. Immobilized biocatalysts include enzymes, enzyme-systems comprising two or more enzymes coupled for sequential catalysis, or cells of plant, animal, fungal or bacterial origin that are preferably, irreversibly anchored in some fashion to a support-matrix. The support materials are well known in the art and may include either soluble or insoluble substances, such as hydrous transition metal oxides, alginate, agarose, polyacrylamide, celite, cellulose, porous ceramics, sand, stainless steel supports, and insoluble polymeric materials such as polyurethane foam, nonwoven fabrics, diethyla inoethyl (DEAE) SEPHADExS wood-chips, nylon fibres and polyester pads among others. Effective application of these materials in the process of immobilization requires that specific protocols be adopted that may include chemical activation and cross-linking of polymers and enzymes or cells, covalent attachment of enzymes or cells, or both to the'support, entrapment, encapsulation, adsorption, or a combination of these. The literature provides examples where such methods have been used to immobilize biocatalysts efficiently while maintaining viability and activity. However, some methods may result in adverse effects that include decreased activity because the biocatalyst is damaged during the immobilization procedure. For example, covalent attachment of -chymotrypsin with EupergitSyc for dipeptide synthesis was found to lead to enzyme deactivation because of exposure to n-propanol. As a further example European Patent Application

0197784, published October 15, 1986 discloses a method of immobilizing biological material onto glass fibres wherein the biological material is mixed with a polymer, coated onto glass fibres and insolubilized by adding a cross- linking and/or condensing agent. In contrast to the present invention, this European Patent application teaches that the cross linking of the polymer holds the cells within the matrix.

United States Patent 4,546,083 provides a method and device for culturing cells in vitro. This device contains a spool of fibre helically wound about a spindle. The spindle is slotted along its length and it is through these slots that the nutrient fluids flow. It is disclosed that generally any fibre-forming material can be utilized in this device. In this patent the cells are deposited on the fibre by forced filtration.

United States Patent 4,407,954 discloses an apparatus comprising a multiplicity of fibers suitable for supporting selected organisms. Although glass fibres are described in this patent, because the surface of the fibres are smooth it is difficult for the organisms to stick. Canadian Patent 1209067, issued Aug. 5, 1986, provides a process for the immobilization of one or more hydrocarbon utilizing microorganisms on a plastic carrier in which the immobilization is carried out in an aqueous nutrient medium to which a minor amount of water immiscible hydrocarbon substrate has been added.

As discussed the immobilization of cells and enzymes is a well established technology that will continue to provide viable solutions to complex medical and industrial applications. Excellent and detailed discussions, and principles of the various procedures of immobilization have been presented in the four volumes of Methods in Enzvmoloσy (Vol. XLIV, 135, 136 and 137 c£. c t).

However, the techniques of immobilization by absorption of cells and enzymes to support materials is perhaps the simplest method to apply, although the theory perhaps is the least well understood. Although this method of immobilization appears to be "simple" (e.g. the particles in suspension are mixed with the support and adsorption ensues) , the forces involved in mediating adsorption are complex, and conflicting opinions exist concerning the mechanism(s) of adhesion to various support materials and how to predict, control and modify the process.

Archambault __ ___. (1989, Biotechnol. Bioeng. 33; 293-299) observed a secretion from cells that appeared to act as a "gluing agent". Bornman & Zaσhrisson (1987 in Methods in Enzymology, Vol. 135, pp. 421-433) noted that protoplasts failed to adhere to non-lectin-treated Cytode: 1 or even to positively charged Cytodex 2 beads, and media with high levels of Ca²⁺ were found to promote attachment. Dunn (1987 in Methods in Enzymology, Vol. 135, pp. 300-307) found that the nature of the solid-support was not influential in determining the adhesion of mixed cells to surfaces, including glass, metal and plastics; the formation of a polysaccharide slime could mediate attachment to such surfaces. Betinec & Mendelson (1988, Enz. Microb. Technol. ___: 418-425) noted the immobilization of Bacillus subtilia maerofibrββ onto several support materials including cellulose and steel and attributed immobilization to the morphogenesis of the macrofibres. Several reports indicate that hydrophobic interactions and electrostatic phenomena that occur when cells and solid supports are introduced into a liquid are of paramount importance in determining efficient and rapid adsorption. The adhesion of cells in suspension to a solid surface immersed in a liquid is seen to include two independent processes. In order for cells to firmly attach to any solid support-material the cells must first be transported to, and make contact with the surface; this involves the physical-chemical interactions prevalent in the system. The subsequent binding of cells to the surface and to adjacent cells will occur because of the time- dependent production of extracelluar mucilaginous, adhesive secretions which act to agglutinate cells to the surface and other cells. This inherent complexity has no doubt restricted the use of adsorption as a mode of immobilization because there appears to be no generally applicable techniques that serve to guide researchers toward achieving efficient biocatalyst loading on well- defined supports.

It is generally believed that high biocatalyst densities adherent to the support are not achievable, and the strength of adhesion is insufficient for many practical applications. However, the process of firm and irreversible adhesion of microorganisms to surfaces is a well-documented phenomenon crucial to environmental, biomedical and technological areas. Adsorption and adhesion of particles, cells, proteins, and microorganisms are known to result in thrombus formation, dental plaque and caries formation, and occurs in waste-water and sewage treatment, liquid-flow conduit-fouling, and metal leaching by microorganisms. This demonstrates convincingly that particles and surfaoss oan be hsld together tor extended periods. In all of these examples large populations of cells, usually greater than can be accommodated in free suspension, may occur in mixed consortia under natural and complex conditions. The strength of adhesion is often sufficient to maintain firm attachment in altered environments and high fluid-flow velocities. The use of adsorption techniques for immobilization of cells for biotechnological applications is now restricted because of our limited understanding of the forces involved in spontaneous adsorption and adhesion of cells to surfaces and how to optimize these forces for adhesion under operational conditions. This invention provides for immobilization strategies using adsorption methodologies. The advantages may include chemical modification of the support-carrier to yield the most desirable surface properties (e.g. hydrophobicity, charge) while retaining the mechanical characteristics of the support material, coupled with the simplicity of application under physiological or operational conditions. Microorganisms and cells in culture, proteins and other particles suspended in solution will become firmly and irreversibly attached to or immobilized on solid surfaces upon contact under conditions managed to promote such immobilization. When the cells are viable, continued growth results in so-called "biofilms". This spontaneous adhesion of particles to surfaces can be used to advantage in many technological and biotechnological applications. However, conflicting and contradictory opinions regarding the utility of the process and the methods that are useful in optimizing adhesion have led us to conclude that more clarity is required. Herein we will discuss the adhesion of cells in suspension to support-surfaces as predicted by a thermodynamic model and modified by electrostatic interactions prevalent in the system. We will show that the process of cell adsorption (adhesion) can be predicted, controlled and modified through an understanding and modification of the forces mediating the response. Canadian Patent 1,156,167, issued Nov.l, 1983 discloses a method of attaching cells to a substrate by first coating the cells with colloidal particles.

Immobilization strategies used previously for plant cells have included encapsulation in calcium alginate or agarose gels and entrapment by a polyurethane foam matrix. However, these methods suffer from drastically reduced mass transfer capacities, and perhaps most importantly limited scale-up capabilities, and the gels are . prone to degradation. Summary of Invention

According to the present invention there is provided a method of immobilizing cells onto a hydrophilic or hydrophobic support material comprising the steps of: determining the cellular surface tension of a suspension of cells; selecting the optimum liquid medium wherein the difference between the surface tension of the cells and the surface tension of the liquid medium is maximized whereby to maximize the adhesion of the cells to the support material, i) in the case where the support is hydrophilic the surface tension of the cells is less than the surface tension of the support, and is greater than the surface tension of the medium; and ii) in the case where the support is hydrophobic the surface tension of the cells is less than the surface tension of the medium and is greater than the surface tension of the support; exposing the support material to a suspension of the cells in the liquid culture medium whereby the cells will immobilize to the support material.

In another embodiment of the present invention there is provided a method for the production of biochemicals from plant cells immobilized onto a support material said method comprising the steps of: determining the cellular surface tension of a suspension of plant cells; selecting a liquid culture medium, wherein the medium has a surface tension which is less than the surface tension of the cells; selecting as a support a hydrophilic solid material with a surface tension higher than that of the cells exposing said support material to a suspension of plant cells in the liquid culture medium, whereby the cells will immobilize to the support material; incubating the immobilized cells; and recovering the biochemicals derived from the plant cells. Brief Description of the Figures:

Figure 1 depicts the extent of adhesion of various suspension - cultured plant cells species as a function of substrate surfape tension. Species shown in the Figure are Digitalis pupurea (1), Catharanthus roseus (2), Ipo oea batatas (3) , Datura innoxia (4) and Papaver somniferum (5) . The representative error bar (95 % confidence limit) is typical for all values. Figure 2 illustrates the percentage of C.roseus cells immobilized as a function of the glass fibre substrate surface tension. A line was drawn as the best fit curve.

Figure 3 demonstrates the effect of glass fibre surface coating on the total culture growth (i.e. including immobilized and non-immobilized cells) and the actual immobilized biomass.

Figure 4 shows, the percentage of C.roseus cells immobilized using the untreated fibre glass substrate after 4 days in culture as a function of initial inoculum biomass measured as fresh weight. Errors are + 5% of mean value. Figures 5a, 5b and 5c illustrates the comparison of alkaloid accumulation; catharanthine, ajmalicine and tryptamine accumulation on C.roseus cells grown in suspension and as immobilized cultures. Detailed Description of the Preferred Embodiments.

Exploiting the spontaneous adhesive behaviour of cultured cells to inexpensive fibre surfaces o^*f varying surface tensions may provide a eoβt-βffβctivβ immobilization process for the production of commodity biochemicals from cell cultures. Success of this strategy required full characterization of the forces involved in cell adhesion and the parameters which modify the adhesion process. The present invention has elucidated the fundamental mechanisms governing the cell adhesion process and has provided insight into the selection of optimum support material and suspending liquid conditions for attaining high levels of cell adhesion. As an example, high levels of plant cell adhesion to needled glass fibre supports has been demonstrated.

Immobilization makes the cells amenable to manipulation for fermentation purposes. The initial adhesion of the cells to various inorganic substrate surfaces can be manipulated according to well defined surface thermodynamic principles that consider the hydrophobicity of the surface. A thermodynamic model of particle adhesion has been used to predict the extent of adhesion of suspension-cultured cells to various substrates. The extent of adhesion depends on the relative values of the substrate (Υ_sv), liquid (Yχ_v), and cellular (γ_cv) surface tensions. Briefly, when:

Ylv Ycv cells adhesion increases with decreasing surface tension of the substrate. Alternatively, when: Ylv Ycv the opposite pattern occurs. For the case of equality:

Ylv ^β Ycv cell adhesion is negligible for conditions described herein and is independent of the value of γ_sv. The model provides a method of determining the cellular surface tension (γ_cv) for suspension-cultured cells in order to select the optimum liquid medium (γι_v) ^and substrate (γ_sv) surface tensions that will maximize the spontaneous adhesion of the oβl s to the substrate.

The thermodynamic approach to the adhesion phenomenon assumes electrical charge interactions to be constant and considers the process to be governed by an interfacial free energy balance involving van der Waals forces. The changes in the free energy of adhesion (Δp dh^*. can be determined: ΔF^adh - _γsc - _γcl - _γsl <0 where γ_sc is the solid-cellular interfacial tension; Y_cl is the cellular-liquid interfacial tension; and γ_sι is the solid-liquid interfacial tension. The distinguishing cases implied by this model in terms of the relative values of the cellular (Y_ev) , liquid (Yι_v), and substrate (Y_sv> surface tensions are represented in Facchini, Neumann and DiCosmo 1988b, Appl. Microbiol. Biotechnol. ___: 346-355.

A kinetic analysis of cell adhesion has determined that the process occurs instantaneously, i.e. immediately after cell-substrate contact. The saturation level of adhesion is dependent on thermodynamic considerations and is not influenced by the bulk concentration of cells in suspension. Importantly, however the rate of cell-substrate attachment increases with increasing cell densities (from 0.1 to 10 % packed cell volume) . These results suggest that the attractive van der Waals forces are considerably greater than the repulsive electrostatic forces between the cells and the substrate (both are negatively charged) . Electrostatic forces were investigated further by considering more fully the effect of electrostatic interactions in terms of the suspending liquid pH and ionic strength. A significant decrease in the net negative cell charge (as mediated by low pH of high multivalent cation strength) resulted in increased levels of adhesion. Since cell-substrate electrostatic repulsion is of minor consequence (in this example) , it is likely that the saturation level of adhesion to any surface is a function of thermodynamic considerations and the extent of electrostatic repulsion between the approaching cells and the cells which have adhered to the substrate surface. When the cell charge is large the increasing density of adhering cells gradually increases the repulsion between the approaching cells and the "effective" substrate surface. The effect of electrostatic repulsion is small in comparison with the magnitude of the attractive van der

Waals forces. This is beneficial since the effective range of pH and ionic strength in a culture medium is limited.

A variety of different plant cells have been use in the Examples of the present invention. The thermodynamic model of the adhesion of cells to a support material is based on the surface tension of the various components. Such parameters are not specific to plant cells and therefore it would be reasonable to predict that the adhesion of other cellular materials could be controlled and predicted according to the present thermodynamic model. Examples of such cells include: bacteria, yeasts, animal cells, and artificial cells such as hydridomas.

According to the thermodynamic model when γι_v < _Ycv cell adhesion increases as the surface tension of the substrate increases. It was found in the present invention that when glass fibres, which have a surface tension of at least 60 mJ/m² were used, a high degree of cell adhesion was reached. Any other hydrophilic material that has a high surface tension would also be operable according to the present invention. Other inorganic fibre material suc as carbon fibres could be used.

Further, according to the thermodynamic model whe Ylv ^> Ycv cell adhesion increases with decreasing surface tension of the substrate. Hydrophibic material which has low surface tension would be preferred under these conditions. Such hydrophobic support material could include for example plastics.

These results have been used for the development of an immobilized plant cell bioreactor exploiting the spontaneous adhesion of the cells. The surface tension of a typical plant cell culture medium (e.g. Murashige and Skoog; Murashige and Skoog, 1962, Plant Physiol. 15, 473- 497) is approximately 47 mJ/m². Since C. roseus cells in culture have a surface tension of approximately 55 mJ/m², a substrate with a very high surface tension will produce the greatest levels of adhesion from thermodynamic considerations. Glass fibres, with relatively high surface tension (at least 60 mJ/m²), were chosen since these would provide the necessary large surface area. This fibre glass material used had various coatings to provide a range of substrate surface tensions to test the predictions of the thermodynamic model in a practical situation. The percentage of cells retained by the fibre glass materials increased with increasing relative surface tension of the fibre coating. Uncoated glass fibres had the highest surface tension and resulted in optimum cell retention. This material was then incorporated into a 5 L immobilized plant cell bioreactor design, with great success. The following examples set forth various embodiments of the invention but are not to be construed as limiting. Examples

Example 1: Effect of Inoculum Concentration on the Kinetics and Extent of Cultured Plant Cell Adhesion to Polymer Surfaces Under Static Conditions.

An understanding of the spontaneous adhesion of suspension-cultured plant cells to various solid substrates is of critical importance in improving the classical air¬ lift or stirred-batch bioreactor process for the large- scale production of fine biochemicals and important pharmaceuticals from plant cell cultures. Elucidation of the fundamental mechanisms of plant cell adhesion is necessary in order to predict, control, and modify the phenomenon so that the full efficiency of the-system and biosynthetic potential of plant cell culture reactors is realized.

Our previous work on the initial interaction of suspension-cultured C. roseus cells with polymer substrates has demonstrated that in the absence of varying electrostatic forces or specific biochemical interactions . the extent of adhesion can be predicted using a thermodynamic model of particle adhesion to solid surfaces that considers only the role of van der Waals interactions (Facchini et al. 1988a, Biotechnol. Bioeng. , 32, 935-938; Facchini ≤£ &1, 1988b, Appl. Microbiol. Biotechnol., 29, 346-355).

The previous experiments were performed under static conditions using a relatively short cell-substrate exposure time of 20 min. However, since the efficiency of product accumulation of the plant cell culture bioreactor process improves with longer fermentation times typically on the order of many weeks and because the system must be optimized for immobilization purposes, it is necessary to consider the kinetics of plant cell adhesion. Understanding cultured plant cell adhesion kinetics will aid in developing a bioreactor system with minimal biofouling problems. Also, this information is essential to optimize an immobilization technique that exploits the spontaneous adhesion of the cells to various substrates. The plant cell adhesion process has two distinct stages; (i) the initial interaction of the plant cell with the surface mediated only by physicochemical forces (Facchini et al. 1988a and 1988b, op. pit.) and (ii) the firm retention of the plant cells on the substrate resulting from the secretion by the adherent plant cells of high- molecular-weight polysaccharides and glycoproteins (Robins St al. 1986, FEMS Microbiol. Lett 155, 143-149). Due to the complexity of the process, this investigation examines only the initial phase, i.e., considering the cell only as a deformable polymer particle. Specifically, 'we will focus on the events occurring prior to the 20-min exposure times used previously. This will allow investigation of the initial interactions.of adhesion before extracellular material can.be secreted thus modifying liquid, substrate, and cellular surface tensions.

Suspensions of C. roseus cells (line LBE-1) were maintained as stock cultures in a Murashige and Skoog medium (Murashige and Skoog, 1982, pp. cit.) containing: 0.5 mg liter "-^• α-naphthalenacetic acid, 0.1 g liter "¹ kinetin, and 3% (w/v) sucrose. Cells were propagated as 75-ml cultures in 250-ml Delong flasks in the dark at 27*c on a gyratory shaker (120 rpm) . Subculturing was performed every 14 days using a one to four dilution of cells to medium.

For all experiments cell suspensions were prepared by diluting a 75-ml cell culture harvest 6 days after subculturing with 6 volumes of distilled water and filtering the dilute suspension through a 500-, 350-, and 210 μ nylon mesh filter series under gentle vacuum. The resulting suspension was centrifuged three times at 600g for 3 min and the pellet resuspended each time in fresh 0.1 M sodium phosphate buffer (pH 5.5). The final suspension consisted of greater than 97% small aggregates (two to five cells) with low levels of extracellular polysaccharides and proteins in the suspending medium. The sodium phosphate buffer chelates Ca²⁺ that may be present, thus reducing the effect of divalent cationic bridging as described previously (Facchini ei al. 1988b, op. cit. : Absolo ___ al. 1985, J. Colloid Interface Sci. 104, 51-59). The cell suspensions were adjusted to 0.1, 1, and 10% packed cell volume (PCV) concentrations using 0.1 M sodium phosphate buffer adjusted to pH 5.5. A small quantity of cells was adjusted to a final concentration of 40% PCV at pH 5.5. The substrate materials used to perform the cell adhesion experiments (Table 1) were prepared as described previously (Facchini e_£ al. 1988a and 1988b op; cit. ) . Fluorinated ethylene propylene (FEP) was prepared as a smooth film by heat pressing the polymer between chromic acid-cleaned glass slides. Polystyrene (PS), polyethylene terephthalate (PET) , and sulfonated polystyrene (SPS) were commercially available as smooth films. With the exception of SPS the polymers were cleaned in a sonicating bath with 95% ethanol; the SPS samples were immersed for 10 sec in n-hexane. The substrates were subsequently air-dried and mounted on chromic acid-cleaned microscope slides. The surface tensions of the substrates were determined from contact angle data using the equation of state approach (Neumann e_fc ftl. , 1974, J. Colloid Interface Sci. 49, 291- 304) and are summarized in Table 1. Table l: Polymer substrates used in adhesion experiments

Itolymer Method of Source Contact Angle Surface Tension Preparation GH2O (degrees) mJ/m²

Sulphonated used as Dσw Chemical 24 + 3 66.6 Polystyrene received Co.

(SPS)

Polyethylene used as Oelanese Ltd., 60 + 3 47.0 Terephthalate received Toronto (PET)

Polystyrene used as Dσw Chemical 95 + 2 25.6 (PS) received Co.

Fluorinated heat press Ccranercial 110 + 3 16.4 Ethylene-Prop- Plastics Toronto ylene (FEP) Static adhesion tests to the various substrates were performed by pipetting 1 ml of the different cell suspensions (0.1, 1, and 10% PCV) into wells formed in Teflon®blocks separated from the polymers by Silastic® gaskets. An alternative method involved filling the wells with 975 μl of 0.1 M sodium phosphate buffer (pH 5.5) and carefully layering 25 μl of a concentrated cell suspension (40% PCV) on the surface of the liquid. This resulted in an effective concentration within each well of 1% PCV. The cells were incubated at 25*C for various times ranging from 15 to 20 min. At the end of each exposure time the Teflon wells with surfaces still attached were submerged and subsequently inverted into a distilled water bath at 25*C for 15 min to stop the experiment and remove nonadherent cells. The blocks were then removed from the water bath, and the surfaces separated and subsequently air-dried.

The extent of adhesion was assessed using an automatic image analysis system (Omnicon(53000, Bausch ^' Lomb, Rochester, NY) , as described previously (Facchini et al. 1988a, op. cit.) . For each experiment six wells per substrate were examined with 10 area fields per well at 32X magnification covering the entire exposed surface. The experiments were performed in triplicate and the data from each of the individual field measurements averaged and expressed as the percentage area of each surface covered by cells.

The contact angle measurements and the calculated surface tensions are presented in Table 1. For all cell concentrations the extent of adhesion of cells to the various polymeric substrates increased as a function of exposure time and reached an upper limit characteristic for each surface. When 1 and 10% PCV concentrations were used the extent of adhesion on all of the various substrates reached a stable plateau. Adhesion of the plant cells to all surfaces at 0.1% PCV concentration appeared to continue to increase after the 20-min test period. It is expected that extrapolation would lead to a plateau value characteristic of the polymer substrate under investigation. With each of the cell concentrations, used the rates of adhesion and the final saturation levels increased following the sequence SPS < PET < PS < FEP. However, the.rate of adhesion to all surfaces was greatest for the 10% PCV concentration suspension and decreased with decreasing cell concentration. However, at any given plant cell concentration the rate of adhesion to the -various substrates are not equivalent as predicted by a theory suggesting kinetic control of the adhesion process (Ruckenstein et al. 1976, Thromb. Haemostatis 36, 334-342). The level of plant cell adhesion decreases linearly with increasing substrate surface tension in all cases in accordance with thermodynamic predictions for a system that includes a suspending liquid surface tension of approximately 72.5 mJ m~² and a cellular surface tension of approximately 55 mJ m"² (Facchini ___ al. 1988b op. cit.) . However, by considering these initial events the first steps can be taken to maximize or minimize the adhesion process as dictated by the desired application. Example 2: yhe Effect of pH_f ionic strength and cation valency on the adhesion of suspension - cultured C. roseus cells to various polymer substrates.

A typical cultured plant cell growth medium is a complex composition of various salts, sucrose and organic constituents; the pH is initially set between 5.5 and 6.0 and varies during the growth cycle of the cell cultures. The effect of solutions containing various concentrations of sodium chloride (NaCl) , calcium chloride (CaCl₂) and aluminium chloride (AICI3), and the influence of the suspending-liquid pH on the extent of adhesion of suspension-cultured C. roseus cells to various substrates will be investigated. The effect of these factors will be related to the electrostatic properties of the cells.

Cells of c. roseus were prepared as i-ft Example 1. For all experiments, the cell suspension were adjusted to a final 1% packed cell volume (PCV) concentration in the appropriate test liquid. Cells were found to be greater than 98% viable in all test solutions for a 1 h period, as determined by dye exclusion of a 0.5% (w/v) Evan's Blue stain (Taylor e+. al. 1980, J. Exp. Bot. 31, 571-576). The substrate materials used for static testing of plant cell adhesion were as described in Example 1. For pH experiments a 0.1 M phosphate buffer was used with the pH adjusted in the range of 2-8. Ionic strength experiments were performed using concentrations up to 1.0 M of the salts NaCl, CaCl2 and AICI3. For all ionic strength experiments, the pH was adjusted to near 5 using 1.0 M NaOH.

The velocity of a particle in a given electric field strength can be used to characterize its surface charge. The electrophoretic mobility of the cultured C. roseus cells in the various test solutions was determined using a Pen Kem System 3000 automated electrokinetic analyzer (Pen Kem Inc., Croton-On-Hudson, NY, USA) . Dilute suspensions of the very fine, washed cells (approximately 0.1% PCV) were prepared in each test solution just before loading into the electrophoresis chamber and measurements were made immediately.

The effect of suspending-liquid pH strength on the extent of adhesion of cultured C. roseus cells to various substrates under static conditions was determined as described previously in Example 1, using 1 ml of the 1% PCV cell suspension in the appropriate test solution.

C. roseus cells have a constant net negative charge in the pH range of 5.5-8.0. Below pH 5.5 the absolute value of the electrophoretic mobility decreases and reaches a second plateau below pH 2.5. Although an isoelectric point was not observed, the net negative cell charge is relatively small in the lower plateau region. No significant differences in this pattern were Observed with cell age during the 14 day growth cycle of the cell cultures.

Levels of adhesion increased with decreasing polymer surface tension following the same sequence as shown in Example 1 [(SPS) < (PET) < (PS) < (FEP) ] at each pH value of the suspending liquid. The linear relationship between the extent of adhesion and the substrate surface tension is consistent with predictions of a thermodynamic model, used previously to describe the cultured-piant cell- adhesion process (Facchini ___ al. 1988a; 1988b, op. cit? Facchini et al. 1989, Biotechnol. Appl. Bioche . 11, 74- 82). Altered pH of the suspending liquid does not affect the value of i_v which for distilled water is 72.5 mJ/m².

Ionic concentrations up to 1.0 M alter the surface tension of the liquid by < 0.5 mJ/m². The surface tension of the cultured cells was previously determined to be approximately 55 mJ/m2. Since Y^y > Y_cv, the observed pattern of increasing adhesion to substrates of decreasing surface tension agrees with the thermodynamic model predictions.

The suspending-liquid pH has a moderate effect on the extent of adhesion of the cells to each surface, compared with the greater influence of substrate surface tension (hydrophobicity) . The lowest levels of adhesion were observed at the higher pH values tested (6 and 8) . Adhesion increased by more than 50% at pH values of 3 and 4, but decreased slightly at pH 2.

A decreased net negative charge on the cultured plant cell leads to increased adhesion, since the degree of diffuse double-layer repulsion is reduced. The decrease in pH might also result in increased cell aggregation, due to the decreased repulsion between cells in suspension. However, the increase in cell adhesion is not due to adhesion of cell aggregates; the distribution of the adherent cells remains uniform, indicating that the cells consistently approach the surface as small aggregates of two to five cells. A decrease in the electrophoretic mobility indicates a reduction in thickness of the diffuse ionic double layer surrounding both the cell and the surface. This may allow for closer approach of the cell to the surface before repulsive interactions, which result from the overlap of the double layers, come into effect. However as we will discuss, decreased repulsion between approaching cells and cells already adsorbed may be more important than decreased cell-substrate electrostatic repulsion, in allowing for greater surface coverage by the adherent plant cells. A decrease in the net negative cell charge was observed with increasing ionic strength for all the cations tested (Na⁺,Ca²⁺ and Al³⁺) . However, the slope of the decrease increased with increasing cation valency. The electrophoretic mobility was near zero (i.e. the cells were uncharged) when l.OM CaCl2 or 0.1M AICI3 was.used. At concentrations of A1³⁺>0.1M a reversal in cell charge was observed with a maximum positive electrophoretic mobility value of approximately +1.5mV~¹s~¹ reached at l.OM AICI3. For all buffers tested, the effect of substrate surface tension was again observed to be in agreement with thermodynamic considerations. The adhesion of cells was not detectably affected by increased concentrations of NaCl up to l.OM in the suspending liquid. However, the extent of adhesion increased with increasing concentrations of CaCl₂ to l.OM. Using AICI3, the maximum level of adhesion occurred at 0.1M, with lower levels at both higher and lower concentrations, of Al³⁺ ions. A statistical analysis of variance (ANOVA) demonstrated a dependence of the extent of adhesion on ionic strength for CaCl2 and A1C1₃ at a confidence limit consistently > 90%.

The relationship between the electrophoretic mobility and the extent of cell adhesion for increasing cationic concentrations in the suspending liquid of NaCl, cacι₂ and AICI3 is consistent with the effects of pH. Increasing concentrations of the monovalent cation Na⁺ did not affect cell adhesion within the sensitivity range of the assay technique used. The relatively small decrease in cell charge with increasing Na⁺ concentration did not alter the degree of cell-substrate or cell-cell repulsion by an amount great enough to produce a detectable effect on adhesion. However, the near neutral electrostatic condition of the cells at l.OM Ca²⁺ increased the extent of adhesion, due to essentially complete removal of repulsive cell-substrate and cell-cell interactions. This condition was also achieved using 0.1M Al³⁺. At higher Al³⁺ concentrations, ionic adsorption of the cation to the cell surface caused charge reversal from a negative to a moderately positive value. Since the usually negatively charged surface of the substrate would also be expected to undergo charge reversal under these conditions (i.e. l.OM Al³⁺) and since all cells are now positively charged, repulsion is expected to increase between the positively- charged surfaces of the plant cells and the substrates, as well as between the cells themselves. Indeed, adhesion decreased at l.OM Al³⁺ from the maximal level attained at 0.1M, supporting this view.

The effects of pH, cation valency and ionic strength, which influence electrostatic properties of the cell and substrate surfaces, act to modify the extent of adhesion, as controlled by interfacial tensions. However, the maximum extent of adhesion of the cells to any substrate may still be related to the balance of the attractive potential between the cells and the substrate and the repulsive interactions between the increasing density of adherent cells and the approaching cells. Thus, electrostatic repulsion appears to increase dynamically throughout the adhesion process, as more cells adhere to the surface. The plateau levels of adhesion may represent the extent of surface coverage, when the electrostatic repulsion between adsorbed and approaohing cells is sufficient to prevent any further cell-substrate adhesion.

This example supports the suggestion that the electrostatic repulsion between adherent cells and cells approaching the surface out of suspension, along with interfacial tensions, defines the maximum extent of adhesion of cultured plant cells to surfaces. Also, the maximum extent of adhesion has been shown to change with a consistent pattern during growth cycles of the cell cultures. These results support the suggestion of the primary role, in controlling the extent of adhesion of cultured plant cells is interfacial tensions, with cell- cell electrostatic repulsion playing a modifying role. Example 3:

Adhesion of Various Species of Suspension - Cultured Plant Cells to Inert Substrates. in this Example the adhesion of seven different species of suspension-cultured plant cells to a range of polymeric substrates was examined.

The species of suspension-cultured plant cells examined were C. roseus (line LBE-1) , Anchusa officinalis. D. innoxia. p. purpurea, j_t fratatas, p. βorønjferurø, and Thalictru rugosum. The growth medium and culture conditions for all species were identical to those described in the previous Examples for C. roseus cultures except for the T. rugosum cultures (obtained from H. Pedersen, Rutgers University), which were grown using the phytohor one, 2,4-dichlorophenoxyacetic acid, rather than the combination of alpha-naphthalenacetic acid and kinetin used for the other cultures. All cultures had been maintained as stock suspensions and subcultured into fresh medium every 10 days for at least 6 months.

Adhesion testing was performed using cells from early linear growth phase cultures, 6 or 8 days after subculturing into fresh medium. Cells were prepared as described in Example 1 and were resuspended in distilled water at a density of 1% packed cell volume (PCV) . The polymer substrates used for adhesion experiments, their method of preparation, measured contact angles and surface tension values calculated using the equation of state approach are summarized in Table 1. Adhesion experiments were performed as described previously in Example 1.

The effective cellular surface tensions were determined by measuring the contact angle formed by drops of distilled water on layers of the cells. Smooth layers were produced by collecting cells from a 250 ml sample of a 1% PCV filtered and washed suspension on a 0.45 μ HAWP

Millipore filter under very gentle vacuum. The resulting 1 mm thick cellular layers were placed immediately in a desiccator for 30 minutes. A 20 μl volume of double distilled water at 25*C and pH 7.0 was used to determine the advancing contact angle. Measured contact angles were stable for up to 60 minutes after removal from the desiccator. Contact angles on 10 different drops were determined each on a previously unused area of the cellular layer. All experiments were repeated in triplicate. The extent of adhesion of 5 of the suspension- cultured plant cell species to the various polymer substrates as a function of substrate surface tension is shown in Fig. 1. For graphical clarity the data for T. rugosum and A. officinalis are not shown. The pattern of adhesion exhibited by T. rugosum was nearly identical to those patterns shown for C. roseus and I. batatas, while the extent of adhesion of A. officinalis to each of the polymers was generally lower than T. rugosu . C. roseus. and I. batatas. The increasing trend in adhesion with decreasing substrate surface tension reported previously (Facchini et al. 1988b op. pit.) was observed for each of these 4 species as well as for D. purpurea. However, the extent of adhesion of D. purpurea cells to each of the polymer substrates was greater than levels observed for T. ruαosu . c. roseus. and I. fefltfltfli as shown in Fig. 1. Alternatively, adhesion of D. innoxia and

P. somniferum to all polymers was low and generally independent of the surface tension of the substrate (Fig. 1) . Since adhesion, of all species demonstrating a dependence on substrate surface tension, was greatest of FEP, the extent of adhesion to this polymer was used to index the relative adhesiveness of the various species suspended in distilled water. By considering the extent of adhesion to FEP the seven species examined can be categorized into 4 groups: 1) very adhesive (D. purpurea. 2) moderately adhesive (T. rugosum. c. roseus and

I. batatas) ; 3)' mildly adhesive (A. officinalisl and 4) poorly adhesive (D. innoxia and P. somniferum) .

The approximate surface tension values for the various species of suspension-cultured plant cells used herein were estimated using the water contact angle method. This method has been applied to numerous micro-organisms and cells including bacteria (Busscher ___ a}. 1984 Appl. Environ. Microbiol., 48, 980-983; Absolom et al.. 1983, Appl. Environ. Microbiol. 46, 90-97), yeast (Mozes and Rouxhet, 1987, J. Microbiol. Methods, 6, 99-112), erythrocytes (Absolom §__ al. 1985 pp. cit.) , and suspension-cultured plant cells (Facchini __t §J_. 1988b op. cit.) . The results of contact angle measurements of each of the cultured plant cell species tested, as well as the calculated surface tensions determined using an equation of state approach (Neumann ___ al. , 1974, op. cj.t.) are listed in Table 2. Measurement of contact angles on layers of cultured plant cells is difficult because the drop remains on the surface of the cellular layer for only a short time before penetrating. However, using repeated measurements to improve accuracy, and due to the widely different contact angles observed for the various species, the technique was used successfully in this study to elucidate general differences in^*cellular surface tensions and index the relative hydrophobicity of the cells. From Table 2 it is apparent that the surface tensions of the various cultured plant cell species can be categorized into three groups: 1) high - (P. somniferum, D. innoxia. and A. officinalis); 2) moderate - (C. roseus. T. rugosum. and I. batatas); and 3) low - (D. purpurea) . The observed levels of adhesion of the various cultured plant cell species correlate with these measured estimates of cellular surface tensions as predicted by the thermodynamic model. For example, the calculated cellular surface tensions for P. somniferum and D. innoxia were 70 and 69 mJ/m², respectively. Since these values are close to the surface tension of water (72.5 mJ/m²) the thermodynamic model predicts that adhesion will be low. Alternatively, the increasing difference between the surface tension of water and the cellular surface tension of the cultured species in the order. A. officinalis <

C. roseus < T. rugosum < I. batatas < D. purpurea results in increasingly negative values of^F^adn and, subsequently, increased adhesion.

In conclusion, the surface properties of different species of suspension-cultured plant cells are quite variable. A thermodynamic approach accurately describes the adhesion of the relatively hydrophobic cells since the magnitude of the attractive van der Waals force is large and repulsive electrostatic interactions can be ignored. However, the attractive interaction between hydrophilic cultured plant cells, suspended in distilled water, and an inert substrate are small and are not likely to be greater than the magnitude of the repulsive electrostatic potential. Adhesion of these cells will occur only when the electrostatic repulsive force is reduced by the addition of cationic species to the suspending liquid. Repulsive electrostatic interactions result from overlapping diffuse ionic double layers of the plant cell and the substrate. The negative surface charges of these solids are oeunterbalanoed with oppositely oharged ions, some of which are bound to the surface, with the rest distributed in the diffuse layer. Increasing the concentration of cationic charge in the suspending liquid will reduce the extension of the ionic double layer. This will allow a for closer approach of the cell to the substrate and, subsequently, increased levels of adhesion. Table 2: Measured water contact angles and calculated effective surface tensions for the various suspension- cultured plant cell species during linear growth phase.

Species Contact Angle Surface Tension

(mJ/m²)

65

53

69 42

48 70

47

Example 4:

Immobilization of Cultured C. roseus Cells using a Fibre glass Substrate.

Exploiting the spontaneous adhesive behavior of cultured plant cells to a wide range of surfaces offers the potential for the development of a simple and effective immobilized plant cell bioreactor. In the previous Examples the fundamental mechanisms that control the extent of plant cell adhesion to polymer surfaces were examined. Using an adhesion model that considers only thermodynamic and electrostatic interactions the adhesion of suspension- cultured C. roseus cells could be predicted and controlled. These studies were performed using polymer substrates and suspending liquids with well defined physico-chemical properties. In this Example the adhesion model is used to describe the immobilization efficiency of a fibre glass substrate treated with various coatings to produce geometrically identical substrates with a range of surface properties. The substrate configuration selected was a woven fibre glass mat since this provides a large surface area for plant cell adhesion and, therefore, will optimize the effectiveness of a surface-immobilization strategy. In addition, the inert nature of the fibre glass material and surface coatings provide the potential to examine the effect immobilization on secondary metabolism in cultured plant cells while minimizing the concerns discussed above. Furthermore, an untreated fibre glass substrate is used to immobilize suspension-cultured C. roseus cells and examine the effects on growth and indole alkaloid accumulation.

Cell suspensions of patharanthus roseus were prepared as described in previous examples.

The fibre glass substrates used for immobilizing suspension-cultured C. roseus cells were approximately 10- 15 mm thick needled fibre glass mats with a uniform fibre diameter of 10 μm (PPG Industries, Inc., Pittsburgh, PA). The surface of the glass fibre is modified to give a functional group coating. These selected surface glass fibre materials, obtained from PPG Industries Inc. provide a range of substrate surface tensions (γ_sv) which are listed in Table 3. The surface tensions of the various fibre glass substrates were calculated from an eguation-of- state approach (Neumann e__ al. 1974 c-p.. cit.) using the measured contact angle of double-distilled water on an individual fibre. Briefly, the water contact angle on the fibre was determined using a modified Wilhelmy balance method (Li et al. 1984, J. Adhesion 17: 105-122).

Immobilization-efficiency is defined herein as the capacity of the substrate to retain a growing biomass in the immobilized state measured as a percentage of immobilized cells versus the total biomass (i.e. immobilized and freely-suspended cells) in the culture system. In order to test the immobilization- efficiency of the coated fibreglass samples, each sample mat was cut into a 3 cm x 3 cm square, accurately weighed, rinsed in distilled water and placed into a 250 ml Erlenmeyer flask containing 50 ml of fresh culture medium. The dry weight of each fibre glass sample was approximately 1.4 grams. The medium and fibre glass substrate were autoclaved together as described above. Each flask was inoculated with 1.2 grams (fresh weight) of 14 day old suspension-cultured C. roseus cells. Control flasks did not include a fibre glass sample and were cultivated as suspension-cultures. After 7 days some of the cells in each flask were immobilized on the fibre glass substrate while others remained in suspension. At this time the fibre glass substrates, with cells attached, were removed from the flask, washed once with distilled water, and partially dried using a vacuum filtration apparatus. Cells in suspension were collected by vacuum filtration and kept separate. All cells were dried for 48 hours at 60*C prior to dry weight determination. The dry weight of the immobilized cells was calculated by subtracting the weight of the fibre glass substrate. All experiments were performed in triplicate.

The loading capacity and time course experiments were all performed using the untreated fibre glass substrates. For all experiments a 3 cm x 3 cm square piece of the fibre glass mat was weighed and sterilized together with the culture medium as described above. Loading capacity experiments were performed by inoculating triplicate flasks with a range of inoculum densities (measured as fresh weight) and collected after 96 hours in culture. The percentage of cells immobilized relative to total biomass was determined as described above. Time course experiments were performed using free-cell suspensions and immobilized cultures collected at a two day interval during the course of a 16 day growth cycle.

The adhesion of suspension-cultured Cl roseus cells to the various glass fibre substrates plays a crucial role in the immobilization process as shown in Fig. 2 where the relationship between the surface tension of the fibre glass substrate and its inherent immobilization-efficiency measured as a percentage of cells immobilized is illustrated. The immobilization-efficiency increases with increasing fibre glass substrate surface tension.

The trends shown in Fig. 2 are enhanced by the combination of spontaneous cell adhesion and entrapment processes. The high surface tension substrates mediate greater initial cell loading than the lower surface tension substrates. The greater adherent-cell density is predicted from thermodynamic considerations (i.e.) the extent of cell adhesion per unit surface area will increase with increasing substrate surface tension provided that the experimental conditions disclosed herein are satisfied. For example, the untreated fibre glass substrate (high γ_sv) achieves a greater adherent-cell density on the glass fibres than the phenyl-coated substrate (low γ_sv) . Consequently, fewer suspended cells remain in flasks containing the (high γ_sv) substrate because more cells were initially immobilized by adhesion. Selection of the high surface tension substrate makes the most efficient use of the available inoculum under conditions described in this Example. Although C. roseus were used in this Example other species can be used resulting in similar adhesion efficiency. It should be noted that adhesion efficiency decreases as γ_cv approaches Y_sv.

Cell viability tests revealed that cultures immobilized using any of the various glass fibre substrates, and cultures cultivated as suspensions, contained approximately 80 to 85% viable cells. Fig. 3 illustrates the effect of immobilizing C. roseus on the various glass fibre substrates relative to total culture growth (i.e. including immobilized and non-immobilized cells) .

The fibre glass-substrate surface coatings have no effect on cell viability or cell growth as compared to the untreated fibre glass material. Various surface coatings on the glass fibre substrate were tested in order to elucidate the effect of a wide range of surface tensions on the immobilization-efficiency of the substrate. All the coatings induced a lowering of the value of Y_sv relative to the bulk fibre glass material, and these alterations reduced the immobilization-efficiency of the substrate.

The untreated fibre glass material has the greatest potential as an immobilization substrate because the majority of free-cells are efficiently immobilized. The various surface coatings did not improve the immobilization-efficiency under the conditions employed. Fig. 4 illustrates the amount of free-cells immobilized as a function of the inoculum biomass (free-cells measured as fresh weight) after 4 days incubation including a 1.3 gram sample of the fibre glass substrate in a 250 ml Erlenmeyer flask containing 65 ml of medium with orbital shaking at 120 rpm) . Greater than 92% of the cells were immobilized when 2.5 grams (fresh weight) or less of cells were inoculated into the flasks. When the free-cell weight was 3 grams (fresh weight) approximately 55% were immobilized; the percentage immobilized decreased to less than 40% when the free-cell biomass was increased to 5 grams. The fibre glass substrate demonstrates an initial cell loading threshold to achieve greater than 90% loading-efficiency, of approximately 1.9 grams (fresh weight) of cells per gram (dry weight) of the substrate. All experiments described herein used a free-cell biomass approximately one half of the threshold value. Table 3. Surface coatings, water contact angles, and surface tensions of the fibre glass substrates used for immobilization experiments.'^*

Coating Chemical nature Contact angle Surface tension θ H₂0 (degrees) (mJ m~²)

Example 5: production of Secondary Metabolites from c. roseus

Immobilized on a Glass Fibre Substrate.

In the present Example, immobilizing C. roseus cells using glass fibres resulted in a decreased specific accumulation of tryptamine, catharanthine, and ajmalicine, relative to suspension cultured cells (Figs. 5a, 5b, and

5c) . No alkaloids were detected in the culture media indicating that cell leakage did not occur.

Alkaloids were extracted from samples that had immediately been frozen in liquid nitrogen after fresh weight determination and stored at -20^*C for not longer than 14 days. After thawing, the cell sample was extracted in 50 ml of methanol in a sonicating bath for 15 minutes.

The methanol extract was reduced to dryness under vacuum, and the residue taken up in 20 ml of bicarbonate buffer (sodium carbonate/sodium bicarbonate 6:4 (w/w), pH 10), and partitioned three times into ethyl acetate. The ethyl acetate phase was evaporated under vacuum, and the final extract taken up in 1.0 ml of methanol.

Quantitative determinations by the method of external standards were performed using a Hewlett-Packard

1090®high performance liquid chromatography (HPLC) system, equipped with a U.V. detector. Separation was accomplished using a Pierce RP-18 Spheri-sδ) (220 x 4.6 mm) column. Initial conditions of 60:40 methanol/water (v/v) containing 0.1% triethylamine were anintained for 4 minutes, followed by a convex solvent gradient to 90:10 methanol/water over 9 minutes. These final conditions were maintained for a further 2 minutes. Alkaloid elution was routinely monitored at 230 nm, and alkaloids tentatively identified on the basis of retention times and U.V. spectra. Medium alkaloid levels were determined by adjusting the spent medium to pH 10 with 1.0 M NaOH, and partitioning three times into equal volumes of ethyl acetate. The ethyl acetate phase was reduced to dryness under vacuum and the final extract taken up in 1.0 ml of methanol. Quantitative determinations were performed as above.

Despite the difficulties in comparing data, there are similarities in the membrane entrapment and cotton fibre immobilization techniques, and the fibre glass immobilization method reported here, especially in terms of the inert nature of the substrate materials used. Moreover, the response of the cultures to immobilization by these methods, ie., suppression of secondary metabolic activity relative to suspension cultures, suggests that the direct cell to cell contact achieved by these methods inhibits secondary metabolic processes (Payne et al., 1988 Biotechnol. Bioengineer. 31 905-912). The immobilization methods used in this study and by Payne et al. (1988 op. cit.) ensure a direct cell to cell contact which is considerably different from calcium alginate immobilization procedures (Brodelius and Mosbach, 1982 In: Advances in Applied Microbiology 28,1-26,). These results also support the suggestion that the observed increases in secondary metabolite accumulation by calcium alginate immobilized cells are caused by properties of the polysaccharide matrix or ionic environment of the alginate gel. Apparently, these conditions are not reproduced by immobilization techniques employing inert substrates such as fibre glass or cotton fibres. We also suggest that the variable response of cultured plant oβllβ to widely different immobilization techniques in terms of secondary metabolite accumulation implicates a species-specific phenomenon. This may be most important when considering the role of direct cell to cell contact in immobilized plant cell cultures. A better understanding of this phenomenon may elucidate some of the regulatory control mechanisms of cultured plant cell secondary metabolism. This knowledge is essential if the biosynthetic potential of cultured plant cells is to be commercially exploited. Although the surface-immobilization of cultured C. roseus cells using a fibre glass support resulted in decreased secondary metabolic activity relative to suspension cultures, we believe that the strategy deserves further consideration. The material possesses excellent scale-up potential because it is inexpensive, inert, re- usable, and can be produced in any geometric configuration desired. The immobilization technique relies on the passive processes of cell adhesion and entrapment, thus labour costs are reduced. The inert nature of the substrate eliminates the concern of the indeterminate variables associated with other immobilization techniques. This will provide the opportunity to investigate the true nature of cell-cell contact phenomena. Additional benefits of, and aspects concerning, the surface immobilization of plant cell cultures are given by Archambault et al. (1989 pp. pit.) .

Claims

CL IMS

1. A method of immobilizing cells onto a.hydrophilic or hydrophobia support material, comprising the steps eft

(a) determining the cellular surface tension of a suspension of cells;

(b) selecting the optimum liquid medium wherein the difference between the surface tension of the cells and the surface tension of the liquid medium is maximized whereby to maximize the. adhesion of the cells to the support material,

(i) in the case where the support is hydrophilic the surface tension of the cells is less than the surface tension of the support, and is greater than the surface tension of the medium; and (ii) in the case where the support is hydrophobic the surface tension of the cells is less than the surface tension of the medium and is greater than the surface tension of the support;

(c) exposing the support material to a suspension of the cells in the liquid culture medium whereby the cells will immobilize to the support material.

2. A method of immobilizing cells on a support for biotransformation purposes comprising the steps of;

(a) determining the cellular surface tension of a suspension of cells;

(b) selecting a liquid culture medium wherein the medium has a surface tension which is less than the surface tension of the cells;

(c) selecting as a support a hydrophilic solid material with a surface tension higher than that of the cells; and

(d) exposing said support material to a suspension of the cells in the liquid culture medium whereby the cells will immobilize to the support material. 3. The method of claim 2, wherein the support is a hydrophilic inorganic solid material.

4. The method of claim 3, wherein the support is a glass fibre solid material. 5. The method of claim 2, wherein the cells are living.

6. The method of claim 2 wherein the difference between the surface tension of the cells and the surface tension of the liquid medium is maximized, which results in increased immobilization.

7. The method of claim 6 wherein the difference between the surface tension of the cells and the surface tension of the liquid is in the range of about 6 to about 20 mJ/m². 8. The method of claim 7 wherein the immobilization of the cells to the support material increases as the surface tension of the support material increases. 9. The method of claim 8 wherein the surface tension of the support material is at least 30 mJ/m². ιo. The method of claim 9 wherein the surface tension of the support material is at least 60 mJ/m². 11. A method of immobilizing living plant cells on a support for culturing purposes comprising the steps of:

(a) determining the cellular surface tension of a suspension of living plant cells;

(b) selecting a liquid medium wherein the medium has a surface tension which is about 6 to about 20 mJ/m² less than the surface tension of the cells;

(c) selecting as a support glass fibre material with a surface tension higher than 60 mJ/m²; and

(d) exposing said support to a suspension of the cells in the liquid culture medium; whereby the cells will immobilize to the support material. 12. The method of claim 11 wherein the suspension of cultured cells ranges from 0.1 to 10% packed cell volume and the biomass inoculum is not greater than 1.9 gram, fresh weight, of cells per gram, dry weight of substrate. 13. The method of claim 1 wherein adhesion increases as the net negative charge on the cultured cells decreases.

14. The method claim 13 wherein the net negative charge on the cells can be reduced by one of the group consisting of: decreasing the pH of the liquid medium, increasing the ionic strength of the liquid medium and increasing the cationic valency of the salts of the liquid medium.

15. The method of claim 1 wherein the cells to be immobilized are selected from the group consisting of: bacteria, fungi^', animal cells and plant cells.

16. The method of claim 11 wherein the plant cells are selected from the group consisting of Catharanthus roseus. Anchusa officinalis. Datura innoxia. Digitalis purpurea. iPQrcea £a£≤±£s, papaver somniferum and Thalictrum rugosum. 17. The method of claim 1 wherein the support material is glass fibre.

18. A method for the production of biochemicals derived from plant cells immobilized onto a hydrophilic support material said method comprising the steps of: (a) determining the cellular surface tension of a suspension of plant cells;

(b) selecting a liquid culture medium wherein the medium has a surface tension which is less than the surface tension of the cells; (c) selecting as a support a hydrophilic solid material with a surface tension higher than that of the cells;

(d) exposing said support material to a suspension of and the plant cells in the liquid culture medium; whereby the cells will immobilize to the support material;

(e) incubating the immobilised cells/ and

(f) recovering the biochemicals derived from the plant cells.

19. The method of claim 18 wherein the cell derived biochemical is selected from the group consisting of: tryptamine, catharanthine and ajmalicine.

20. The method of claim 18 wherein the support material is glass fibre.

21. A method of immobilizing cells on a support in biotransformation purposes comprising the steps of:

(a) determining the cellular surface tension of a suspension of cells; (b) selecting a liquid culture medium wherein the medium has a surface tension which is greater than the surface tension of the cells;

(c) selecting as a support a hydrophobic solid material with a surface tension lower than that of the cells; and

(d) exposing said support material to a suspension of the cells in the liquid culture medium; whereby the cells will immobilize to the support material.

22. The method of claim 21, wherein the differences between the surface tension of the cells and the surface tension of the liquid medium is maximized, which results in increased immobilization.

23. The method of claim 21, wherein the difference between the surface tension of the cells and the surface tension of the liquid is in the range of about 6 to about 20 mJ/m².

24. The method of claim 21, wherein the immobilization of the cells to the support material increases as the surface tension of the support material decreases. 25. The method of claim 24 wherein the surface tension of the support material is not greater than 4J0- mJ/m².

26. The method of claim 25 wherein the surface tension of the support material is not greater than 20 mJ/m². 27. A method of immobilizing living plant cells on a support for culturing purposes comprising the steps of:

(a) determining the cellular surface tension of a suspension of living plant cells;

(b) selecting a liquid medium wherein the medium has a surface tension which is about 6 to about 20 mJ/m² greater than the surface tension of the cells;

(c) selecting as a support a hydrophobic material with a surface tension not greater than 20 mJ/m²;and

(d) exposing said support to a suspension of the cells in the liquid culture medium; whereby the cells will immobilize to the support material.