Production of anthraquinones by immobilized - Springer Link

0 downloads 0 Views 830KB Size Report
Mar 18, 1994 - Abstract The production of anthraquinones by Fran- gula alnus Mill. plant cells was used as a model system to evaluate the performance of a ...

Appl Microbiol Biotechnol (1995) 43:416-423

© Springer-Verlag 1995

L. Sajc • G. Vunjak-Novakovic • D. Grubisic N. Kova6evi6 • D. Vukovi6 • B. Bugarski

Production of anthraquinones by immobilized Frangulaalnus Mill. plant cells in a four-phase air-lift bioreactor

Received: 18 March 1994/Received revision: 20 September 1994/Accepted: 28 September 1994

Abstract The production of anthraquinones by Frangula alnus Mill. plant cells was used as a model system to evaluate the performance of a liquid-liquid extractive product-recovery process. The shake flask experiments have shown higher production of anthraquinones in cell suspension and flask cultures of calcium-alginate-immobilized cells when silicone oil was incorporated into the medium, compared to a control without silicone oil. An external-loop air-lift bioreactor, developed and designed for the production and simultaneous extraction of extracellular plant cell products, was regarded as a four-phase system, with dispersed gas, non-aqueous solvent and calcium-alginate-immobilized plant cells in Murashige and Skoog medium. Continuous extraction of anthraquinones by silicone oil and n-hexadecane inside the bioreactor resulted in 10-30 times higher cell productivity, compared to that of immobilized cells in a flask. Based on the mixing pattern, immobilized biocatalyst extraparticle and intraparticle diffusional constraints and the kinetics of growth, substrate consumption and product formation, a mathematical model was developed to describe the time course of a batch plant cell culture.

L. Sajc (5::~) • D. Vukovi6 " B. Bugarski Department of Chemical Engineering, Faculty of Technology, Belgrade University, Karnegijeva 4, Belgrade, Yugoslavia. Fax: 381 11 3220847 G. Vunjak-Novakovic Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA D. Grubisic Department of Plant Physiology, Faculty of Biology, Belgrade University, 29. novembra 142, Belgrade, Yugoslavia N. Kova~evib Faculty of Pharmacy, Belgrade University, Vojvode Stepe 450, Belgrade, Yugoslavia

The model showed satisfactory agreement with four sets of shake flask experiments and three bioreactor production cycles.

Introduction The recovery of extracellular plant cell products by coupling in situ extraction directly to the bioreactor has been used successfully as a technique for kinetic enhancement as well as for product concentration. Integration of biosynthesis with product recovery may therefore be considered to be a systems approach to bioprocessing with a view to improved process design. Although plant cells offer a potential for production of valuable phytochemicals, this potential is currently incompletely realised because of problems of slow growth, low product yield, genetic instability of highly producing cell lines, shear sensitivity and lack of understanding of the basic physiology and biochemistry of plant cell systems (Zenk 1978; Shuler 1981; Yeoman et al. 1982; Fowler 1985, 1988; Buitelaar and Tramper 1992). Only a few successful examples of the productio n of fine chemicals from plant cell culture, e.g. shikonin, ginseng compounds and berberine, have been achieved when selection or media manipulation to generate high-yielding cell lines and efficient bioprocessing techniques are combined. Immobilized plant cells can lead to increased production levels of some valuable secondary metabolites, when compared to suspension cultures (Brodelius and Mosbach 1982; Lindsey and Yeoman 1984; Bringi and Shuler 1990). Improved yields with cells immobilized in calcium alginate have been reported by Wichers et al. (1983), Asada and Shuler (1989), Vanek et al. (1989), Subramani et al. (1989) and Kim and Chang (1990). Immobilization also offers advantages from an engineering point of view. The retention of plant cells enables easy separation of cells from medium (which can be regarded as a Newtonian fluid)

417

and, if the product is extracellular, it allows continuous product removal. The immobilization of cells also permits high cell loading and continuous reutilization of biocatalysts as a production system (Scragg 1991a, b). A new approach that could improve productivity of plant cells is the integration of bioconversion and downstream processing by end-product removal and accumulation in the extractive phase (Holden and Yeoman 1987; Mavituna et al. 1987a, b; Brodelius and Pendersen 1993; Sim and Chang 1993). As aqueous media have an inherently low capacity for low-polarity plant cell extracellular products, the strategy is based on the addition of an extraction phase to provide an artificial site for product accumulation and to adjust global transport equilibria to favour secretion. This approach leads to enhanced mass-transfer rates, decreased levels of product inhibition and facilitated product recovery. Unfortunately, specificity of the extraction phases is difficult to achieve and, for a group of related end-products, product distributions could be different in the accumulation phase compared with the extracellular medium. As the solvent must perform optimally for both the reaction and product recovery steps, solvent selection strategies for extractive biocatalysis should be contemplated (Daugulis 1988). An external-loop air-lift bioreactor was developed and designed for the continuous production and extraction of plant cells extraceUular metabolites. This system was regarded as a four-phase system, with dispersed gas, liquid solvent and calcium-alginate-immobilized cells in Murashige and Skoog (1962) plant cell medium. This bioreactor design retains the advantages of using immobilized cells as biocatalysts, but utilizes continuous liquid-liquid extraction of plant cell products in a four-phase flow. The production of anthraquinones by Frangulaalnus ceils, which excrete them into the medium, was used as a model system to evaluate the bioreactor performance.

Materials and methods Chemicals Murashige and Skoog (MS) basal medium, kinetin, indole-3-butyric acid, 2,4-dichlorophenoxyacetic acid, 6-benzylaminopurine, sodium alginate (medium viscosity), silicone oil (density, p = 960 kg/m 3, viscosity, # = 0.05 Pas), n-hexadecane, Tris(hydroxymethyl) aminomethane (TRIS), and 2,3,5-triphenyltetrazolium chloride (TTC) were purchased from Sigma (St. Lous, Mo.). All other chemicals were reagent grade.

medium, supplemented with 30 g/1 sucrose, i mg/1 2,4-dihydroxy phenylacetic acid and 1 rag/1 6-benzylaminopurinein a 125-ml conical flask. All media were sterilized at 115 °C (103 kPa, 15 lb/in2) for 30 min. The cells were cultivated on a gyratory shaker at 100 rpm. The suspension cultures were subcultured every 4 weeks by splitting the flask contents into three new flasks with 30 ml MS medium

Immobilization of cells To improve mechanical stability of the calcium alginate gel and the productivity of the immobilized cells, the following immobilization procedure was used. Sodium alginate, sterilized by ~ rays, was dissolved in sterilized water at 25 °C during mixing on a magnetic stirrer for 2 h. A preparation of 100 ml 2% (w/v) sodium alginate solution was mixed with 50 ml thick cell suspension 4 weeks old. Since a higher productivity of immobilized large cell aggregates (in a reduced time after immobilization) was expected, the cell suspension was not filtered before mixing with sodium alginate. The cell/alginate suspension was subsequently extruded dropwise through a 22-gauge needle into 0.2 M CaC12 solution, and beads 4 mm in diameter (nearly uniform size) were formed. The beads were left to harden for 30 min in CaC12 solution, rinsed three times with sterile water and transferred into MS medium. To increase bioreactor cell loading, each bead was cut into eight pieces, a size equivalent to a sphere of 2.5 mm diameter.

Assays

Dry cell weight Samples of cell suspension (5 ml) and immobilized cells (after dissolving of calcium alginate in 12 2.5-mm beads with 1% sodium citric acid solution at 50 °C for 30 min) were filtered through filterpaper (Whatman 41, previously dried in an oven at 70 °C for 24 h and weighed), placed on Buchner funnel and rinsed three times with distilled water. After drying at 70 °C for 72 h, samples were weighed and the total dry cell weight was calculated from the known aliquot.

Sugar The total sugar concentration, expressed as the sum of sucrose, glucose and fructose, was determined by assaying hydrolysates for reducing sugar. The reducing sugar concentration was determined by the o-toluidine method: 0.1-ml samples of medium were mixed with 0.1 ml of 5 M HC1 and hydrolysis was performed for 30 min in a boiling water bath: after cooling, the solution was made up to 3 ml total with distilled water and a 0.5-ml sample of this solution was mixed with 2 ml o-toluidine reagent and heated for 8 rain in a boiling water bath. Absorbance was detected at 630 nm (Shimatzu UV-VIS 1202 spectrophotometer). A calibration curve in the range 0-50 g/1 was obtained by using standard sucrose solutions, o-Toluidine reagent was prepared by dissolving 6 ml distilled o-toluidine and 0.15 g thiourea in 94 ml glacial acetic acid. The reagent was kept in the dark.

Anthraquinones Cultivation of cells Callus cultures of Frangul alnus, initially derived from seedlings, were maintained on MS medium containing 30 g/1 sucrose, 7 g/1 agar, 10 mg/1 indole-3-butyric acid and 1 rag/1 kinetin. Suspension cultures were initiated by adding 5 g fresh cell weight to 30 ml MS

Various anthraquinone derivatives, predominantly present as glucosides, were expressed as equivalents of dantron (1,8-dihydroxyanthraquinone), according to the European and Yugoslavian Pharmacopeia by the modified method described by van den Berg and Labedie (1984). Anthraquinones in 0.5-ml samples of silicone oil

418 were extracted with MeOH (5 x 3-ml). Extraction was by vortexing (1 min) and centrifugation (3 min, 3000 rpm). The top methanol fractions were collected, evaporated, and the residue taken up in 20 ml 5 M HC1 (containing 20 g FeC1J1 5 M HC1). Hydrolysis of glycosides and oxidation of anthrones were performed simultaneously in this solution, by heating for 30 min in a boiling water bath. After cooling, the aglycones were extracted with ether (3 x 10 ml). The ether solution was washed with distilled water and evaporated. The residue was taken up in 2 ml 1 M NaOH and the absorbance was determined at 500 nm. A calibration line in the range 0-20 mg dantron/l 1 M NaOH was obtained by using standard dantron solutions.

,--[

16

: .-~.

o

V?

Cell viability was determined by staining cells with TTC. A sample of cell suspension (1 ml), or 12 2.5-ram alginate beads were mixed with 2 ml 1% (w/v) TTC in TRIS buffer (pH = 7.5) and left in the dark for 24 h. Cell material was rinsed once with distilled water and subsequently extracted with 3 ml 95% EtOH at 60 °C for 30 min. Absorbance of the EtOH solution was determined at 485 nm, and viability calculated per gram dry cell weight.

t

Four sets of 12 125-ml conical flasks, containing MS growth medium (as described in Materials and methods) were prepared. The first and second set contained Cell suspension and immobilized cells, respectively, without silicone oil; 3-ml portions of silicone oil (1% medium volume) were added to each flask in the third set containing cell suspension and in the fourth set containing immobilized cells. At 4-day intervals, starting 2 days after innoculation, three flasks in each set were harvested and dry cell weight, total sugar and product concentration, and cell viability were determined in three samples taken from each flask.

Bioreactor experiments Batch cultivation of immobilized cells was performed in a fourphase, 250-ml external-loop air-lift bioreactor, shown in Fig. 1. Air and solvent were introduced through a combined distributor at the bottom of the riser section. A sintered glass plate (160 Inn) was used for air distribution, and five 1-mm-diameter openings for solvent distribution• Flow rates of air and solvent were 0.7 vvm and 0.05 vvm respectively. These values were kept constant in all bioreactor experiments. Solvent flow rate was adjusted to provide complete drop coalescence in the top of the horizontal section, necessary for the continuous extraction procedure. Under these conditions, an oxygen transfer rate of 2.35x10-6mmolO2 1 MS- ~s- ~ was determined. An oxygen uptake rate of 5.6 x 10- 8 mmol 02 1 MS- ~s- t of 50 g dry cell weight/l MS, determined separately in a batch stirred-tank reactor, implies that the oxygen requirements of immobilized cells are two orders of magnitude lower than the oxygen supply, even at 35% bioreactor loading. However, for plant cell cultivation, mixing is more important than aeration. Mixing studies have shown that the flow in the riser section approaches good mixing and, according to the mixing performance and the tank in the series model, the riser section could be described by a single stirred-tank reactor, and the bioreactor as a whole by four stirred-tank reactors in series. At these gas and solvent flow rates the mixing time was 27 s, and solvent and gas retentions were 1.3% and 1.5% respectively. The dantron mass-transfer coefficient

o

Q pc

',~. .



~A

~.

f~

5

;>~rnedium~>

< solvent Aair [~'--I I - - - ' ~ - - - I _ L

Shake flask experiments



I

Cell viability

Experimental

]

1 2 3 4 5 6 7 8

.

bioreactor

air humidifier membrane air pump solvent/medium s e p a r a t o r peristaltic pump DO electrode DO m e t e r pH e l e c t r o d e

.

.

.

.

.

.

.

.

.

solvent

,

9 pH m e t e r 10 level i n d i c a t o r 11 t e m p e r a t u r e i n d i c a t o r 12 13 14 15 16

air condenser i n o c u l a t i o n / s a m p l i n g port medium reservoir s o l v e n t reservoir data a c q u i s i t i o n system

Fig. 1 The experimental set-up kla= 1.25x10-2s -1, and the dantron distribution coefficient K = 3 (at a dantron concentration of 6 mg/l MS, which corresponds to the concentration measured in the cell suspension medium) imply that a high mass-transfer rate of plant cell product across the MS medium/silicone oil interface is possible. The extraction process seems to be limited mainly by the low solvent capacity of 500 mg/l at 25 °C. The effect of the continuous removal of cell products from MS medium on the kinetics of product formation by immobilized plant cells in the four-phase system was determined in three production cycles. The first two were carried out with the same immobilizedcells at a 25% bioreactor loading (40 g dry cell weight/l). In the first run, silicone oil was used as solvent, and in the second, n-hexadecane. With regard to the low capacities of both silicone oil and n-hexadecane (500 mg dantron/l at 25 °C), every 48 h the entire solvent volume of 50 ml was replaced with fresh to prevent solvent saturation by cell products. The third run was performed with 35% bioreactor loading (50 g dry cell weight/l) and silicone oil as solvent. The entire solvent volume was changed every day with fresh oil. In all three bioreactor experiments, (hormone-free) MS production medium was used. During each production cycle (10 days), every 2 days three samples were taken from the bioreactor and the dry cell weight, total sugar and product concentration, and cell viability were assayed.

Results

S h a k e flasks K i n e t i c s of g r o w t h , s u b s t r a t e c o n s u m p t i o n a n d p r o d u c t f o r m a t i o n of b o t h s u s p e n d e d a n d i m m o b i l i z e d cells,

419 as well as the effect of silicone oil present in the culture on cell viability and productivity, were determined as a prerequisite for experiments in a four-phase bioreactor system. These experiments were carried out simultaneously, with approximately the same amount of inoculated cell material. Figure 2a shows changes in dry cell weight, substrate and product concentration in cell suspension culture without silicone oil. Each point is an average of nine measurements on days 2, 6, 10 and 14. Figures 3a, 2b and 3b show changes in dry cell weight, and substrate and product Concentration in cell suspension culture with silicone oil, in immobilized culture without silicone oil and in immobilized culture with silicone oil respectively. After the short lag phase, during the first 2-4 days, the dry cell weight in cell suspension culture without 35,

I

.

.

.

.

.

.

I



silicone oil sharply increased and the stationary phase was reached on the 14th day, when nearly all the sugar was consumed (Fig. 2a). During the same time the immobilized cells in culture without silicone oil grew much more slowly (Fig. 2b). Product concentration in cultures without silicone oil was determined by silicone oil assay, after separation of cells and extraction of product from medium with silicone oil. It was observed that immobilized cells produced approximately five times more anthraquinones (0.45 mg/g dry cell weight) after 14 days, than suspended cells (0.10 mg/g dry cell weight). In cell cultures with silicone oil, slow growth and slow sugar consumption of both suspended (Fig. 3a) and immobilized cells (Fig. 3b) were observed. The lag phase of the suspended cells was prolonged and, after 14 days in culture, cells produced approximately three times more (0.30 mg/g dry cell weight) than suspended

1.0

i

.

.

.

.

.

.

i

1.0



3030. 0.8

0,8

25-

25. 03

~'

o.e r© r-.,

20.

¢D t~ 15.

0.4

E 13_

X

'~" 2009

o.e

~" a

15E3

0.4

E 0..

)< 10-

10. 0.2

0.2 5~

0

.

.

.

.

.

.

I

.

.

.

.

.

''~

i

35

'

0

a

1.0

0! 35.

30-

.

. . . . .

,

. . . . . .

,

.

30.

1.0

0.8

0.8

25-

25-

¢9 20-

0.6 O

CO 20-

0

E 0.4 ~ '

.co...r~ IS-

15-

O

X

0.6

//~

1o.4~,

/

X

10-

lO-

~'

0.2

0.2

5-

0

0 0

b

t

0

14

7 [day]

Fig. 2a, b Model predictions (solid line) and experimentally observed profiles (averages + SD) for cell growth, X (e), substrate consumption, S (0), and product formation, P (@), in a cell suspension culture (shake flask experiments), b immobilized cell culture (shake flask experiments). DCW dry cell weight

b

......

, ...... 7

, • 14

0

t [day]

Fig. 3a, b Model predictions (solid line) and experimentally observed profiles (averages + SD) for cell growth, X (e), substrate consumption, S (©), and product formation, P (Q), in a immobilized cell culture with silicone oil (shake flasks experiments), b immobilized cell culture (shake flasks experiments). DCW dry cell weight

420

cells in culture without silicone oil. During this period, immobilized cells in the culture with silicone oil produced more anthraquinones (0.65 mg/g dry cell weight) than both immobilized cells in the culture without silicone oil and suspended cells in the culture with silicone oil. Considering product formation in all four sets of shake flask experiments, immobilized cells produce more than cells in suspension. Product recovery by silicone oil improved the productivity of both suspended and immobilized cells. However, the anthraquinone content of the cells, both immobilized and in suspension, was 90% higher than in silicone oil and excretion was less than 10%. It was found that silicone oil present in the culture has no effect on cell viability. In all four sets of experiments, the viability of cells/g dry cell weight was constant during the cultivation period. The values of cell viability were within __ 15 % of the initial cell viability, determined after subculturing. Bioreactor

In the first production cycle, when silicone oil was used as solvent, 40 g dry cell weight/1 immobilized cells (25 % bioreactor loading) produced 6.5 mg/g dry cell weight of anthraquinones after 10 days (Fig. 4). This value is approximately ten times higher than that observed for immobilized cells in flask culture with silicone oil. When all the sugar was consumed, the medium in the bioreactor was replaced with fresh. The process was carried on with the same immobilized cells, but the product was separated with n-hexadecane. The same anthraquinone yield was observed in this batch (Fig. 5). ,

,

,

,

50-

The third production cycle was performed with 35% bioreactor loading (50 g dry cell weight/l), and silicone oil as solvent, which was completely changed every 24 h. As a result, anthraquinone yields up to 17.5 mg/g dry cell weight have been achieved (Fig. 6). When cultivated in a four-phase external-loop bioreactor on MS production medium (hormone-free), immobilized cells excreted significantly higher amount of anthraquinones (30%-50%) than immobilized cells in shake flask culture with silicone oil (10%), although the cell content was approximately the same.

40 50. i

40.

u?

~

*

m

30

o

30.

D

O

E

•~. 2 0 . X

a. 10

10.

0

, 2

.. 4

, 6

.

, 8

0 10

t [day]

Fig. 5 Model predictions (solid line) and experimentally observed profiles (averages _+ SD) for cell growth, X (e), substrate consumption, S (O), and product formation, P (Q), in the bioreactor (40 g dry cell weight, DCW/1) and n-hexadecane as solvent

........, 40 ]

50

,

,

,

,

~

I

f

~

,, 4 0

I

il

/ 40

~

~

~

~

30

O

30-

¢'t

l

E

20

-,

o

0..

X

'

10

10-

0-

0 0

2

4

6 t

8

10

[day]

Fig. 4 Model predictions (solid line) and experimentally observed profiles (averages _+ SD) for cell growth, X (e), substrate consumption, S (O), and product formation, P (@), in the bioreactor (40 g dry cell weight, DCW/1) and silicone oil as solvent

01 0

2

4

6

8

10

0

t [day]

Fig. 6 Model predictions (solid line) and experimentally observed profiles (averages _+ SD) for cell growth, X (e), substrate consumption, S (O), and product formation, P (G), in the bioreactor (50 g dry cell weight, DCW/1) and silicone oil as solvent

421

described by a Monod model:

Discussion The model was developed to describe the time course of a batch culture of plant cells with sucrose as the limiting substrate in the presence of silicone oil. For wellmixed batch studies in shake flask, it was assumed that extraparticle mass-transfer constraints were negligible, and that the sugar was uniformly distributed throughout the medium. The batch, four-phase external-loop air-lift bioreactor was also assumed to perform as a single well-mixed batch reactor. The mixing time in this system was 27 s, while substrate utilization by immobilized cells required approximately 10 days. Balance equations for product, cells and substrate in constant-volume aerobic batch system are: dP d--~-=rp

(1)

dX dt = rx

(2)

dS dt - rs

(3)

where X is the cell concentration (g dry cell weight 1-1), t is the time (days), P is the product concentration (g 1-1), rp is the product formation rate (g 1-1 d a y - 1), rx is the cell growth rate (g 1-1 day-1), S is the substrate concentration (g 1-1) and r~ the substrate consumption rate (g 1-1 d a y - 1). Product formation in culture of immobilized or suspended F. alnus plant cells was described by a Luedeking-Piret type rate equation: (4)

rp =

7+S

where the factor S/(7 + S) is introduced to ensure that rp approaches zero as S approaches zero. fi (day" 1) is the maximum specific production rate of cells and 7 (g/l) the cell product saturation coefficient. The rate of cell growth for both suspended and immobilized cells, for sugar as a limiting substrate, was Table 1 Kinetic p a r a m e t e r s for b a t c h cultures of p l a n t cells with sucrose as the limiting s u b s t r a t e in the presence of silicone oil

#max SX rx - Ks + S

(5)

where gmax(day- 1) is the maximum specific growth rate of cells, and K~ (g/l) the Monod cell growth saturation coefficient. The rate of sugar consumption was described by: (6)

r~ = rx + rp + msX

where ms (day- 1) is the maintenance coefficient for cells on carbon substrate. Equations 4-6 have been solved as a system of coupled first-order differential equations, using a numerical routine. The results obtained for the kinetic parameters #max, Ks, fl, 7 and ms are shown in Table 1. Calculated profiles showed satisfactory agreement with four sets of shake flask experiments and three bioreactor production cycles, as shown in Figs. 2-8. The maximum specific growth rate #max= 0.33 d a y - 1 was obtained in flask suspension culture without silicone oil. When silicone oil was added #maxdecreased both in the flask suspension culture (from 0.33 d a y - 1 to 0.07 day-1) and in the immobilized cell culture (from 0.06 d a y - 1 to 0.04 d a y - 1). K~ values for the cell suspension (1.6 g/l) and immobilized cells (2.0 g/l) in the presence of silicone oil were significantly lower than in oil-free cultures (20.0 g/1 and 16.0 g/l). The maximum specific production rate, fl, of immobilized plant cells in flask culture with and without silicone oil was approximately the same ( ~ i x 10 -4 day -1) and higher than that for cell suspension culture. In bioreactor experiments, growth of immobilized cells was minimal, and for each production cycle it could be approximated by d X / d t = 0. The calculated fl values for the first two runs (the same immobilized cells, but different solvents) were the same, 6.5 x 10 -4 day -1, and approximately six times higher than the values in immobilized cell flask cultures with silicone oil. The highest value, fl = 2.5 x 10-3 day-1, calculated for the third run (when silicone oil was changed every 24 h) was two orders of magnitude higher than that for the cell suspension without silicone oil. Low 7 values, calculated for all three bioreactor

Culture

#max ( d a y - 1)

Ks (g/l)

fl (day - x)

7 (g/l)

ms (day - 1)

S h a k e flasks Cell s u s p e n s i o n Cell s u s p e n i o n s + silicone oil I m m o b i l i z e d cells I m m o b i l i z e d cells + silicone oil

0.33 0.07 0.06 0.04

20.0 1.60 16.0 2.00

3.0 x 6.0 x 1.0 x 1.3 x

13.0 22.0 26.0 29.0

4,0 x 3.0 x 3.5 x 3.0 x

Bioreactor Run 1 Run 2 Run 3

----

----

6.5 × 10 . 4 6.5 x 10 - 4 2.5 x 1 0 - 3

1.00 1.00 1.00

4.0 × 10 . 6 4.0 x 10 . 6 4.0 x 10-6

l0 s 105 104 104

10-6 10-6 10 . 6 10-6

422

production cycles (1.0 g/l), imply that product formation kinetics could be approximated by a zero-order reaction. It seems that as the mass-transfer rate of product from medium to silicone oil increases, the product formation rate increases. The possible explanation for this is that immobilized cells produce faster to compensate for the detected lower cell product level in the extracellular space. The observed production rate results from simultaneous reaction within the cell and diffusion of substrate within the calcium alginate particles with immobilized cells, for negligible extraparticle mass constraints in a well-mixed system. The effectiveness factor, t/, which represents the ratio of the observed reaction rate to the reaction rate in the absence of diffusion, is defined as: r~ bs

I / = r~i,

(7)

For immobilized cells in the bioreactor d X / d t = 0 and ? ~ S. Equations 4 and 6, with these simplifying assumptions, give: -

obs = r . + m s x

=

+ m3X.

(8)

The Thiele modulus, ~b, which represents the ratio of the diffusion time of a system to the reaction time, for zero-order kinetics is defined (Levenspiel 1972; van't Riet and Tramper 1991) by:

d, ((fl + ms)X~'/2 6- \

-

tl = l ~5

Suggest Documents