Biotechnology Letters Vol No i0 Received August 12 715-718 (1986) HIGH DENSITY CELL CULTURE OF RHODOTORULA GLUTINIS USING OXYGEN-ENRICHED AIR Jae Gu Pan~, Moo Young Kwak, and Joon Shick Rhee Department of Biological Science and Engineering Korea Advanced Institute of Science and Technology POB 150, Chongyang, Seoul, KOREA SUMMARY Fed-batch culture of Rhodotorula glutinis, an oleaginous yeast, was carried out to obtain high biomass concentration. With air,final concentration of biomass was I00-II0 g/L, whereas with oxygen-enriched air, 185 g/L was obtained. Obligate aerobic metabolism of }J. glutinis under oxygen limitation and low oxygen requirement under lipidaccumulating condition were found to be advantageous for the high density culture of this microbe. INTRODUCTION Recently, interests for the microbial lipid production have been renewed because of the advent of biotechnology and of the unstable prices for most of the agricultural products (Ratledge, 1982). Rhodotorula glutinis is an extensively studied SCO (Single Cell Oil) microbe (Yoon & Rhee, 1983). Oleagenicity of this yeast is relatively well known under various limiting conditions (Rattray, 1984), and continuous culture as well as batch culture was known to be suitable for accumulation of the intracellular lipid (Gill, et ai.,1977; Yoon & Rhee, 1983). However, SCO process is far from industrial realization because of economics, although SCO processes using industrial wastes such as whey (Davies & Gordon, 1984) and peat oxidates (Andreevskaya & Zalashko, 1979) were reported recently. In the study described in this communication, fed-batch culture of R. glutinis carried out to obtain high biomass concentration using oxygen-enriched air. Improvement of volumetric productivity by using high density culture along with several advantages of the high density culture (Matsumura, 1983) would make a positive step to realizing the SCO process in foreseeable future. MATERIALS AND METHODS Microorganism: Rhodotorula glutinis NRRL Y-I091 was obtained from Northern Regional Research Center, USDA, Peoria, IL. U.S.A. Method of culture: Composition of medium used for the high density culture was as follows: KH2P04, 12.5; Na2HPO4, 1.0; (NH~)2SO4, 5.0; MgSO4.7H20, . ; CaCI2.2H20, 0.25; Yeast Extract(Difco), 1.9; Trace minerals, 0.25 mL. A solution of trace minerals contained following 715 minerals in gram perL 5N-HCI: FeSO4"7H20, 40; CaCI2"2H20 , 40; MnSO4.nH20, i0; AICIB-6H20, I0; COC12, 4; ZnSO4.7H2 O, 2; Na2MoO4.2H20 , 2; CuC~-2H20, I; H3B04, 0.5. Feeding solution contained 9.0 g MgS04 .7H20 and 20.0 g Yeast Extract in L of 60% glucose solution. The solutions of trace minerals, glucose, and MgSO 7H20 were sterilized separately and mixed aseptically. Culture was carried out in a custom-made L fermenter with L initial working volume. Culture pH was adjusted to 5.5 with N NaOH or 33% ammonia water using custome-made automatic pH controller. Dissolved oxygen was monitored by steam sterilizable D.O. probe (New Brunswick Scientific) and was maintained at 30-60% of air saturation by controlling manually agitation speed (max. 1,500 rpm) and by controlling aeration rate (0.2 - v/v/m). For the aeration with oxygen-enriched air, premixed oxygen enriched air (40% 02 + 60% air) replaced the compressed air. Foam was suppressed by adding Antifoam A (Dew Corning, Food grade). Glucose feeding was controlled manually according to residual concentration measured intermittently during the culture (1-5%). For the calculation of biomass yield, culture volume was measured after stopping aeration and agitation at each sampling time. Analytical Method: Biomass concentration was measured by dry cell weight (105 ~ overnight). Volume of cells was measured by centrifuging the cells at 7,000 g for I0 min. Residual glucose was analyzed by glucose oxidase/catalase assay method (Sigma). Residual nitrogen was analyzed by micro-Kjeldahl method (A.O.A.C., 1980). Lipid content was measured by a modified percolation method (Choi, et al., 1982) followed by Folch's washing (Folch et al., 1957). RESULTS When R. glutinis was cultured in fed-batch culture without any limitation, growth was continued until limited by availability of dissolved oxygen (Fig. la). Dissolved oxygen fell down to zero when biomass concentration reached about 25 g/L, and growth was continued even under oxygen-limiting condition to reach ii0 g/L concentration. Lipid content was about 20%. Instantaneous biomass yield was not changed until the specific growth rate was slowed down to zero. R_. $1utinis did not produce any fermentative by-products even under oxygen-limiting condition. Another fed-batch culture was carried out to produce intracellular lipid by limiting the nitrogen source for the cells previously grown under oxygen limiting condition (Fig. Ib). Number of cells remained constant after the exhaustion of nitrogen source and content of the intracellular lipid increased up to 40%. Dissolved oxygen increased hr after the start of lipid accumulating condition. In order to ensure adequate supply of oxygen, oxygen-enriched air (40% oxygen + 60% air) was used in the fed-batch culture (Fig. 2). There was no oxygen limitation throughout the culture. Final concentration of biomass was 185 g/L, the volume of which corresponded to 75% of total volume of culture broth when measured by centrifugation. Classical two phases were observed in the high density culture that yielded 40% lipid content. 716 .5 _D.O. 60 ul c" >- 100 1010 I o o O~ ul tO E c~ El 20 ~L , I 50 I a , o I00 ~ o I ~ 20" 40 Time, hr Fig. ]0 ~ 60 80 100 120 TIRE, hr Fed batch culture of R. glutinis with air i. a) O_-limited growth: Dissolved oxygen fell to zero at the Z indicated point. b) 02-limited growth was shifted to nitrogen limitation (N.L.) 200 t ! O. ~ O O o "'., "~ e %. 06- . ~ 9. . . . . O 150 o ,_1 - - r cq ,-n loo 50 -40 I2: IJu -30 c).o,.-o .,,.~.) 1.-- 50 P Z U (3 20 E Z | % "l"q-n i 40 ! , I 80 oj o TIME.hr Fig. 2. Fed-batch culture of R. glutinis with oxygen-enriched air (40% 02 + 60% air). Dissolved oxygen was maintained at 40% (• 20%) of air saturation. 717 DISCUSSION Final concentration of biomass (185 g/L) obtained in this work seems to be close to maximum cell concentration in the continuously stirred tank reactor, because volume of cells at this concentration corresponded to 75% of broth volume after centrifugation. The viscosity of yeast culture broth was found to increase dramatically above the concentration 200 g/L (Mori, et al., 1979). Lipid content in the high density culture was somewhat low as compared to those of batch and continuous culture (60%). The reason for the low lipid content in high density culture is not clear at present. It may be, however, caused by high concentration of salts added during the course of culture, because in high density culture it was difficult to maintain the concentration of salts in the range physiologically adequate for the synthesis of lipid. Moreover, unused minerals were to be concentrated because of increasing concentration of biomass. Oxygen requirement in lipid-accumulating phase was low as shown in Fig. lb. It was consistent with the result analyzed theoretically using mass and energy balance method (Pan & Rhee, 1985a). Lower requirement for oxygen is related with lower generation of metabolic heat (Minkevich & Eroshin, 1973; Pan & Rhee, 1985b) Obviously, these facts should be the positive aspects of SCO process over the other processes because usual limitation of bioreactor productivity comes from mass or heat transfer capacity. Obligatory aerobic metabolism of R. glutinis is advantageous for the high density culture of this organism, since oxygen-limited growth could continue without causing accumulation of the fermentative byproducts that was found to be the main constraints in a high density culture of E. coli B (Pan et al., 1986). REFERENCES Andreevskaya, V. D., and Zalashko, M. V. (1979). Prik. Biokhim. Mikrobiol., 15, 522. AOAC (1980) in "Official Methods of Analysis of the AOAC", 13th ed., pp. 858. Choi, S. Y., Ryu, D. D. Y., and Rhee, J. S. (1982) Biotechnol. Bioeng. 24, 1165. Davies, J., and Gordon, T. (1984). in "Proceedings of 6th Australian Biotechnology Conference," pp. 127, Univ. of Queensland, St. Lucia, Brisbane. Folch, J., Lees, M., and Sloane-Stanley, G. H. (1957). J. Biol. Chem. 226, 497. Gill, C. 0., Hall, M. J., and Ratledge, C. (1977). Appl. Env. Microbiol., 33, 231. Matsumura, M. (1983). Ferment. Ind. (in Japanese), 41, 102. Mori, H., Yano, T., Kobayashi, T., and Shimizu, S. (1979). J. Chem. Eng. Japan., 12, 313. Pan, J. G., and Rhee, J. S. (1985a). Biotechnol. Bioeng., 28, 112. Pan, J. G., and Rhee, J. S. (1985b). Korean J. Chem. Eng., 2, 81. Pan, J. G., Rhee, J. S. and Lebeault, J. M. (1986). Biotechnol. Lett., Submitted. Ratledge, C. (1982). Prog. Ind. Microbiol., 16, 119. Rattray, J. B. M. (1984). JAOCS, 61, 1701. Yoon, S. H., and Rhee, J. S. (1983). JAOCS, 60, 1281. 718 . cation, fed-batch culture of R. glutinis carried out to obtain high biomass concentration using oxygen-enriched air. Improvement of volu- metric productivity by using high density culture along. Biotechnology Letters Vol 8 No i0 715-718 (1986) Received August 12 HIGH DENSITY CELL CULTURE OF RHODOTORULA GLUTINIS USING OXYGEN-ENRICHED AIR Jae Gu Pan ~, Moo Young Kwak, and Joon Shick Rhee Department. 20 E Z | oj o Fig. 2. Fed-batch culture of R. glutinis with oxygen-enriched air (40% 02 + 60% air) . Dissolved oxygen was maintained at 40% (• 20%) of air saturation. 717 DISCUSSION Final