Journal Elsevier

of Biotechnology,

BIOTEC

315

13 (1990) 315-323

00496

Continuous itaconic acid production by immobilized biocatalysts H. Kautola, Laboratory

N. Vassilev

* and Y.-Y. Linko

of Biotechnology and Food Engineering, Helsinki University of Technology, Espoo, Finland

(Received

14 September

1989; accepted

7 November

1989)

Summary The continuous itaconic acid production from sucrose with Aspergihs terreus TKK 200-5-3 mycelium immobilized on polyurethane foam cubes was optimized in column bioreactors using statistical experimental design and empirical modelling. The highest itaconic acid product concentration calculated on the basis of the obtained model was 15.8 g 1-t in the investigated experimental area, when sucrose concentration was 13.5%, aeration rate 150 ml mm’ and residence time 178 h. From sucrose with immobilized A. terreus TKK 200-5-3 mycelium itaconic acid production was stable for at least 4.5 months in continuous column bioreactors. In comparison, using glucose as substrate and immobilized A. terreus TKK 200-5-l mycelium as biocatalyst similar stability was obtained with higher product concentration. The omission of copper sulphate from the production medium gave the highest itaconic acid product concentration (26 g 1-r) from 9% glucose with 0.25% ammonium nitrate and 0.095% magnesium sulphate. Itaconic

acid; Continuous

production;

Immobilized

biocatalyst;

Sucrose;

Glucose

Introduction Research on the biotechnical production of organic acids using immobilized mycelium as biocatalyst has recently increased (Vaija and Linko, 1986; Ju and

Correspondence to: H. Kautola, Laboratory of Biotechnology of Technology, SF-02150 Espoo, Finland. * Present address: Bulgarian Academy of Sciences, Institute

0168-1656/90/$03.50

0 1990 Elsevier Science Publishers

and Food Engineering, of Microbiology,

B.V. (Biomedical

Helsinki

University

1113 Sofia, Bulgaria.

Division)

316

Wang, 1986; Horitsu et al.. 1988; Kautola and Linko, 1989). Itaconic acid is of great interest as an intermediate for manufacturing of polymeric materials and ~-substituted pyrrolidones (Milsom and Meers, 198.5). Itaconic acid has been produced successfully with immobilized mycelium in the laboratory scale from glucose with polyacrylamide (Horitsu et al., 1983), polyurethane (Kautola et al.. 1987) gels. (‘elite (Kautola et al., 3985) or porous disks (Ju and Wang, 1986) as carriers. Further. Kautola et al. (1985) have reported itaconic acid production with agar or alginatr immobilized A. terrez4.s from xylose. In most cases repeated batch cultivations were employed. In continuous bioreactors the carriers used in itaconic acid production have been Celite (Kautola et al.. 1985), porous disks (Ju and Wang, 1986) and polyacrylamide gel (Horitsu et al.. 1983: Kautola, 1988). Rychtera and Waae (1981) have also reported the continuous itaconic acid production from glucose by a .1 liter conventional chemostat. We have earlier reported itaconic acid production from sucrose in repeated batch shake flask cultivation with Aspergihs ferreus TKK 200-5-J immobilized on polyurethane foam cubes (Kautola et al.. 1989). The production medium (sucrose and ammonium nitrate concentration and initial pH) for itaconic acid production was then optimized. In this paper we report further optimization of the aeration rate and residence time in the itaconic acid production using a continuous glass column bioreactor.

Materials and Methods Microorgunisms Aspergilfus terreus TKK 200-5-3 used in itaconic acid production from sucrose was maintained on Czapek-agar slants containing sucrose and A. terreus TKK 200-5-l for the production from glucose on potato-dextrose-agar slants. lmmohilization The 130 ml water jacketed glass columns with 0.95 g (about 70 pieces) of one centimeter polyurethane foam (Espe, Espoo, Finland) cubes with an average pore size of 1.5-1.7 mm were sterilized by autoclaving. Each sterile column with 95 ml of growth medium at 36Oc‘ was inoculated with spores from one fresh (1 l-day-old A. terreus TKK 200-5-3 or 4-day-old A. terreus TKK 200-5-l) slant cultivated at 28°C. The growth medium contained 6% sucrose or glucose. 0.4% ammonium nitrate. 0.095% magnesium sulphate, 0.004% copper sulphate and 0.0088% potassium dihydrogenphosphate at pH 3.1. The packed bed bioreactora were aerated with 100 ml min ’ sterile air for one week, during which the mycelium grew on the carrier. After the immobilization continuous production was carried out in the same bioreactors with production medium. Production medium The production medium contained 0.25% ammonium nitrate, 0.004% copper sulphate, 0.095% magnesium sulphate and 5, 6.5. 9, 10, 13.5 or 15%: sucrose. respectively or 9% glucose at pH 3.0 (modified from Kautola et al., 1989).

317

Continuous production

Substrate and air were led into the 130 ml water jacketed (36’ C) glass column from the bottom of the reactor and product outlet was at the top of the bioreactor. The aeration rate used was zero, 22 ml min-‘, 75 ml mm-‘, 100 ml mm’, 128 ml mm’, or 150 ml min-’ and the residence time was 24, 52.2, 120, 187.8, 190, or 216 h, respectively. Itaconic acid determination

Itaconic acid was determined using HPLC Varian 5000 (Varian Associates, Inc., Walnut Creek, CA, U.S.A.) with a 30 cm Aminex HPX-87 ion exchange column (BioRad Laboratories, Richmond, CA, U.S.A.) and 0.4 mM sulphuric acid as the mobile phase at 75°C. UV-detector (Knauer Variable Wavelength Monitor, Wissenschaftlicher GerPtebau Dr. Ing. H. Knauer GmbH, Berlin, FRG) was used at 210 nm. Samples taken from the final product solution were boiled for 20 min and filtered before analyzing. Experimental

design

The itaconic acid production in continuous packed bed bioreactors with immobilized A. terreus TKK 200-5-3 mycelium from sucrose was optimized using experimental design and empirical modelling based on the data obtained from the previous repeated batch experiments (Kautola et al., 1989). Three chosen independent variables, sucrose concentration (X,, 5%) aeration rate (X,, ml mm-‘) and residence time (X,, h) were used in 23-factorial experimental design with 6 star points and 4 replicates at the center point (Tables 1 and 2). Total number of experiments was 18 (Box et al., 1978). To estimate responses of the product itaconic acid concentration (Y, g I-‘) as the dependent variable, SPSS/PC + data programme (Statistical Package for the Social Sciences, SPSS, Inc., Chicago, IL, U.S.A.), was used to calculate the second degree polynomials (Thompson, 1982). The result of each experiment was calculated as the average of itaconic acid product concentration from 5 samples during 10 d after two weeks of continuous cultivation.

TABLE 1 NOMENCLATURE Name

unit

Description

%

Significance level of probability Multiple correlation coefficient Coefficient of determination Sucrose concentration Aeration rate Residence time Itaconic acid concentration

P R R2 x, x2

X3 Y

min-’ h g1-’ ml

31x

Results and Discussion The itaconic acid production from sucrose was optimized continuous column bioreactors. The effect of medium composition in continuous column cultivations using either glucose or sucrose gated. Systematic optimizution

of ituconk

systematically in and the stability were also investi-

ucid production

Based on the results of the experiments the following mathematical model, in which all terms regardless of their significance were included, was obtained (Ey. 1): Y = -8.98292

+ 2.3574X,

+0.00803838X, -0.173036X,”

- 0.14312X,*

XZ + 0.00609567X, +0.000512318X;

+ 0.0688927X, X, + 0.00035844

X, X,

- 0.000576627X$

R = 0.83, R’ = 0.69

(1)

(P: * < 0.05, significance level of probability; R = multiple correlation coefficient; R’ = coefficient of determination.) Eq. (1) shows that the aeration rate had the most significant effect on the itaconic acid concentration in the investigated area. Fig. la shows as response surfaces the itaconic acid product concentration in the continuous column bioreactor with immobilized mycelium as a function of sucrose concentration and aeration rate at

TABLE

2

INDEPENDENT VARIABLES SUCROSE CONCENTRATION (X,, ‘%), AERATION RA-Ik ( XL. ml min-‘) AND RESIDENCE TIME (X,, h) AND RESULTS (ITACONIC ACID CONCENTRATION. Y, g I-‘) IN THE EXPERIMENTAL PLAN WHERE CK= 1.415 Experiment

x, (S)

1

6.5

2 3

13.5 6.5 6.5 13-i 6.5 13.5 17.5 5.0 15 .(J 10.0 10.0 10.0 10.0 10.0 10.0 10.0 10.0

4 6

7

x 9 10 II

12 13 14 I5 16 17 18

X2 (mlmln-') 22 22 12X

72 12X 12x 27

12x 15 7s 0 150 75 75 75

x?(h) 52.2 52.2 52.2 I x7.7 i7 2 1x7.7 1x7 7 I x7.7 1’0.0 110.0 I20.0 120.0 14 0 716.0 I70.0

15

170.0

75

120.0

75

120.0

319

15 SUCROSE

CONCENTRATION(%)

1

3 4

4.5

7

5

9

11

13(

. )

4

\

3 c

3

(b)

il

1

o

75 AERATION

RATE

(ml

min

150 -1

)

Fig. 1. Itaconic acid concentration (g 1-l) in a continuous column bioreactor with immobilized A. rerreus mycelium as a function of: (a) sucrose concentration and aeration rate at the residence time of 120 h, (b) aeration rate and residence time at sucrose concentration of lo%, and (c) sucrose concentration and residence time at the aeration rate of 150 ml n-tin-‘, shown as response surfaces.

320

SUCROSE

CONCENTRATION(S)

Fig. 1 (continued).

the residence time of 120 h. In this case the highest investigated aeration rate gave the best calculated result of 14.1 g I-’ when initial sucrose concentration was 12.2%. aeration rate 150 ml mini ’ and residence time 120 h at the center point. Fig. 1 b. showing itaconic acid concentration in response surfaces as a function of aeration rate and residence time at sucrose concentration of 10% in the center point of the investigated area, gives the calculated result of 14.0 g I- ’ with a productivity of 2.04 g 1-l d-’ and the yield of 14% based on initial 10% sucrose concentration at the center point, 150 ml mini’ aeration and 165 h residence time. Fig. lc suggests, however, that a sucrose concentration of 13.5% would give as high as 15.85 g 1.l’ itaconic acid product concentration at the aeration rate 150 ml min.- ‘, and residence time 178 h. We found earlier in repeated batch cultivations that the best itaconic acid concentration can be reached with about 10% sucrose concentration (Kautola et al., 1989). The highest product concentration 13.3 g I- ’ is lower than the 14.0 g 1.. ’ calculated from 10% sucrose concentration and the 15.85 g 1-l obtained with 13.5% sucrose concentration from Eq. (1). In this case the yield was 11.7% based on initial sucrose concentration and the productivity 2.14 g I ’ d- ’ was 2.25 times higher than that obtained in repeated batch cultivations. The highest itaconic acid concentration obtained in actual experiments, 17.6 g 1-l with a productivity of 3.5 g l- ’ d-’ was reached at 10% sucrose concentration, 150 ml min-’ aeration rate and 120 h residence time.

321 I

I

I

I

I

I

I

I

: 20 u a c-I'5 e-l z 0 10 0 a +5 H I

I

I

I

I

I

I

4

6

8

10

12

14

16

I

-

18

T I M E (weeks) Fig. 2. Continuous itaconic acid production with immobilized A. terreus mycelium in column bioreactors from 9% sucrose (0) and 9% glucose (A) with aeration rate of 100 ml mm’, residence time of 190 h and initial pH 3.0 at 36°C with medium containing 0.25% NH,NO,, 0.004% CuS0,.5H20 and 0.095% or 0.25% NH,NO, and 0.095% MgSO,.7H,O (- - -) or 0.095% MgSO,. MgSO,.7H,O ( -) 7H *O (. 1. -). Arrows indicate the change of substrate composition.

Effect of medium composition on ituconic acid production Itaconic acid production in continuous column cultivation reached its highest value when the production medium contained only carbon source, ammonium nitrate and magnesium sulphate. The omission of copper sulphate from the production medium based on the observation by Batti and Schweiger (1963) and Rivibe et al. (1977) increased itaconic acid product concentration in this case about 1.4-fold. The lack of ammonium nitrate in the production medium drastically decreased itaconic acid production. This is in a good agreement with the earlier results on xylose where Kautola et al. (1985) reported that in the continuous itaconic acid production by immobilized mycelium the inclusion of magnesium sulphate and the omission of potassium dihydrogenphosphate was necessary for stable production. In Fig. 2 the continuous itaconic acid production with immobilized mycelium in column bioreactors was shown both from 9% sucrose or 9% glucose with the aeration rate of 100 ml mm-‘, residence time of 190 h and initial pH 3.0 at 36°C with different medium composition. The arrows in Fig. 2 indicate the change of substrate composition. In comparison of using sucrose or glucose as substrate, Fig. 2 also shows that itaconic acid production (26 g 1-r) from 9% glucose with immobilized A. terrew TKK 200-5-l mycelium was about 1.7 times higher than the 15.5 g I-’ with immobilized A. terreus TKK 200-5-3 from 9% sucrose. The concentration obtained with glucose in this work was markedly higher than that of 18.2 g 1-l reached by Ju and Wang (1986) using porous disk 1.5 1 bioreactor, but their yield

322

was 36% based on initial 5% glucose concentration in contrast to 29% in this work based on initial 9% glucose concentration. The productivity of 3.5 g 1 ’ d ’ obtained at the itaconic acid concentration of 26 g 1 ’ was low in comparison with the 28.8 g I-’ d ’ at 11 g I- ’ from 6% glucose obtained using Celite as carrier (Kautola et al., 1985). This may be partly due to for example the difference in the shape (height : diameter ratio) of the bioreactors and the glucose concentration. The variations in the relative quantity of itaconic acid released by the cells cannot he explained at present and must further be studied. In any case, continuous production of itaconic acid with immobilized biocatalysts is much more efficient than conventional chemostat employing free cells which produced 7.8 g 1 ’ from 5% glucose (Rychtera and Wase, 1981). Stability of the immobilized biocatalyst The stability of the itaconic acid production in continuous column cultivation with immobilized A. terreus TKK 200-5-l or TKK 200-5-3 mycelium was excellent for at least 4.5 months both from sucrose and glucose as can be seen from Fig. 2. The itaconic acid concentration of 26 g I-’ from 9% glucose with a volumetric productivity of 3.5 g 1 ’ d-.’ was maintained for a long period. Inasmuch as Kautola et al. (1987) also found itaconic acid production was stable for at least 14 months on polyurethane gel immobilized mycelium in repeated batch cultivation from 6% glucose, an immobilized biocatalyst system appears to be suitable in the continuous production of itaconic acid.

Conclusion The model obtained with the complete medium with sucrose gave as the highest calculated itaconic acid product concentration 15.85 g ll’ in continuous column cultivation with immobilized A. terreus TKK 200-5-3 mycelium on polyurethane foam cubes when sucrose concentration was 13.5%. aeration rate 150 ml min ’ and residence time 178 h. The simplified production medium containing only glucose or sucrose as the carbon source, and ammonium nitrate and magnesium sulphate was suitable for production of itaconic acid with the increased product concentration of 1.4-fold. Glucose gave about 1.7 times higher itaconic acid level than sucrose. Immobilized Aspergilfus terreus mycelium is an excellent biocatalyst, and stable continuous itaconic acid production for at least 4.5 months was possible.

References Batti, M. and Schweiger. L. (1963) Process for the production of itaconic acid. US Patent 3.078.217. Box, G., Hunter, W. and Hunter, J. (1978) Statistics for Experiments. John Wiley and Sons. New York. p. 374.

323 Horitsu, H., Takahashi, Y., Tsuda, Aspergillus terreus immobilized

J., Kawai, K. and Kawano, Y. (1983) Production of itaconic acid by in polyacrylamide gels. Eur. J. Appl. Microbial. Biotechnol. 18,

358360. Horitsu, H., Takahashi, Y., Ada&i, S., Xioa, R., Hayashi, T., Kawai, K. and Kautola, H. (1988) Production of organic acids by immobilized cells of fungi. In: Moo-Young, M. (Ed.), Bioreactor Immobilized Enzymes and Cells: Fundamentals and Applications, Elsevier Applied Science Publishers Ltd., Essex, pp. 287-300. Ju, N. and Wang, S. (1986) Continuous production of itaconic acid by Aspergillus terreus immobilized in a porous disk bioreactor. Appl. Microbial. Biotechnol. 23, 311-314. Kautola, H. (1988) Continuous itaconic acid production by immobilized cells in airlift reactor. Abstracts of the 8th International Biotechnology Symposium, Paris, July 17-22, 1988, Imprimerie Chirat, St-Just-la-Pendue, D75, p. 289. Kautola, H., Gronlund, B., Horitsu, H., Linko, Y.-Y. and Linko, P. (1987) Optimization of itaconic acid production from glucose by free and immobilized Aspergilh terreus. In: Neijssel, O.M., Van der Meer, R.R. and Luyben, K.Ch.A.M. (Eds.), Proc. 4th Eur. Congr. Biotechnol., vol. 1, Elsevier Science Publishers, Amsterdam, pp. 106-109. Kautola, H. and Linko, Y.-Y. (1989) Fumaric acid production from xylose by immobilized Rhizopus arrhizus cells. Appl. Microbial. Biotechnol. 31, 448-452. Kautola, H., Vahvaselkl, M., Linko, Y.-Y. and Linko, P. (1985) Itaconic acid production by immobilized Aspergih terms from xylose and glucose. Biotechnol. Lett. 7, 167-172. Kautola, H., Vassilev, N. and Linko, Y.-Y. (1989) Itaconic acid production by immobilized Aspergilh terreus on sucrose medium. Biotechnol. Lett. 11, 313-318. Milsom, P.E. and Meers, J.L. (1985) Gluconic and itaconic acids. In: Moo-Young, M. (Ed.), Comprehensive Biotechnology, vol. 3, Pergamon Press, Oxford, pp. 681-700. Riviere, J., MOSS,M.O. and Smith, J.E. (1977) Industrial Applications of Microbiology, Surrey University Press, London, pp. 159-161. Rychtera, M. and Wase, D.A.J. (1981) The growth of Aspergillus terreus and the production of itaconic acid in batch and continuous cultures. The influence of pH. J. Chem. Technol. Biotechnol. 31, 509-521. Thompson, D. (1982) Response surface experimentation. J. Food Proc. Preserv. 6, 155-188. Vaija, J. and Linko, P. (1986) Continuous citric acid production by immobilized Aspergillw niger: reactor performance and fermentation kinetics. J. Mol. Catal. 38, 237-253.