320 PARK ET AL. J.FERMENT.BIOENG., TABLE 1.Culture data from single and two-stage cultures of L.erythrorhizon immobilized in polyurethane foam Single-stage Two-stage SHCK SHIK SHCKSHCK SHIK /SHIK /M-9 /M-9 Final cell mass 17.45 12.48 12.63 17.25 7.5 (g DCW/ Final shikonin conc. 25 112 56 12 78 100 (mg/0 Culture time 18 20 7/11 7/9 7/9 (d) Volumetric productivity 1.4 5.6 3.1 0.8 4.9 (mg/I/d) Culture times for two-stage culture indicate the durations in the first stage/second stage;the cultivation period in the first stage was fixed at 7 d. duction medium,M-9.They reported that,with a selected cell line of high yield,shikonin production was maximized with a two-stage culture system;cells were propagated in a large quantity in the first stage containing MG-5 medium, which lasted for 9 d,and then 1.5 g//of shikonin titre was induced during the subsequent 14-d second stage with M-9 Time (d) medium. In the present study,single and two-stage culture sys- FIG.5.Typical time course of shikonin production in a packed- bed bioreactor with immobilized L.erythrorhizon cells. tems were compared by using the immobilized plant cells; the media used were SHCK and SHIK for the single-stage and SHCK/SHIK,SHCK/M-9 and SHIK/M-9 combina- Fig.4). tions for the two-stage cultures.Culture data of final cell Direct comparison of the productivity in the present growth and shikonin titre are summarized in Table 1.It bioreactor system with that in other studies,such as the was noted that the single-stage culture with SHIK medium one by Tabata et al.(1),is very difficult because the activ- was much more effective for shikonin production than the ities of the cell lines used are quite different;the shikonin two-stage culture with SHCK/M-9.The two-stage culture synthetic activity of the cell line used in the present study is with SHIK/M-9 gave higher shikonin production than lower by an order of magnitude than those used by Tabata SHCK/M-9 but lower cell growth.In terms of volumetric er al.However,it is still considered that the present im- productivity,there was about a 15%increase with the mobilization technique can offer similar improvements to single-stage culture (SHIK)compared to the two-stage, systems with cell lines of higher activity,since it can con- SHIK/M-9 system.This is probably due to the fact that siderably reduce the culture steps and process time.We effective induction of shikonin is supported by enhanced believe that room still exists for further optimization of the productivity in SHIK medium.It has been reported that bioreactor system,and that it should be able to give higher natural auxin IAA (indole-3-acetic acid)is more favorable productivity than the present result. for shikonin production than the synthetic auxins such as 2,4-D(2,4-dichlorophenoxyacetic acid)(1).This agreed well with the result obtained from the present work REFERENCES employing immobilized cells that shikonin production was 1.Tabata,M.and Fujita,Y.:Production of shikonin by plant cell higher in SHIK than in SHCK medium (Fig.4).Noting cultures,p.207-218.Biotechnology in plant science.Academic that SH medium does contain ammonium (ca.2.6 mM), Press,New York,N.Y.(1985). which was reported to be inhibitory for shikonin biosyn- 2.Hulst,A.C.,Tramper,J.,Brodelius,P.,Leon,J.C.,and thesis (1),the present results indicate that the synthetic Luyben,K.A.M.:Immobilized plant cells;respiration and oxy activity can be controlled by using appropriate growth gen transfer.J.Chem.Tech.Biotech.,35B,198-204 (1985). regulators like IAA.Immobilization in the polymer 3.Nakajima,H..Sonomoto,K.,Morikawa,H.,Sato,F. matrices should also provide the plant cells with more Ichimura,K.,Yamada,Y.,and Tanaka,A.: Entrapment of favorable conditions for shikonin production as described Lavandula vera cells with synthetic resin prepolymers and its ap- plication to pigment production.Appl.Microbiol.Biotechnol. above. 24,266-270(1986). Operation of packed-bed bioreactor A typical time 4.Mavituna,F.and Park,J.M.:Growth of immobilized plant cells course of shikonin production in packed-bed bioreactor is in reticulate polyurethane foam matrices.Biotech.Lett.,7 (9), shown in Fig.5.The packed-bed reactor system was a 637-640(1985). single-stage batch culture using SHIK medium for both 5.Shenk,R.U.and Hildebrandt,A.C.:Medium and techniques cell growth and shikonin production.As in the flask cul- for induction and growth of monocotyledonous and ture,the cells grew to yield fully-packed immobilized cell dicotyledonous plant cell cultures.Can J.Bot.,50.199-204 aggregates and the shikonin produced was directly ad- (1972). sorbed to the polyurethane matrices,resulting in a practi- 6.Fujita,Y.,Hara,Y.,Suga,C.,and Morimoto,T.:Production of shikonin derivatives by cell suspension cultures of Lithospermum cally clear culture broth.It was noted that the product titre erythrorshizon.Plant Cell Reports,1,61-63 (1981). was higher than in flask culture,although slow cell growth 7.Chaplin,M.F.and Kennedy,J.F.(eds.):Carbohydrate analysis; and the initial lag in shikonin production were similar(see a practical approach,p.3.IRL Press,Washington,D.C.(1986)320 PARK ET AL. TABLE 1. Culture data from single and two-stage cultures of L. erythrorhizon immobilized in polyurethane foam Single-stage Two-stage SHCK SHIK SHCK SHCK SHIK /SHIK /M-9 /M-9 Final cell mass 17.45 12.48 12.63 17.25 7.5 (g DCW//) Final shikonin conc. 25 112 56 12 78 (mg//) Culture time ~ 18 20 7/11 7/9 7/9 (d) Volumetric productivity 1.4 5.6 3.1 0.8 4.9 (mg/l/d) Culture times for two-stage culture indicate the durations in the first stage/second stage; the cultivation period in the first stage was fixed at 7 d. duction medium, M-9. They reported that, with a selected cell line of high yield, shikonin production was maximized with a two-stage culture system; cells were propagated in a large quantity in the first stage containing MG-5 medium, which lasted for 9 d, and then 1.5 g/l of shikonin titre was induced during the subsequent 14-d second stage with M-9 medium. In the present study, single and two-stage culture sys￾tems were compared by using the immobilized plant cells; the media used were SHCK and SHIK for the single-stage, and SHCK/SHIK, SHCK/M-9 and SHIK/M-9 combina￾tions for the two-stage cultures. Culture data of final cell growth and shikonin titre are summarized in Table 1. It was noted that the single-stage culture with SHIK medium was much more effective for shikonin production than the two-stage culture with SHCK/M-9. The two-stage culture with SHIK/M-9 gave higher shikonin production than SHCK/M-9 but lower cell growth. In terms of volumetric productivity, there was about a 15%0 increase with the single-stage culture (SHIK) compared to the two-stage, SHIK/M-9 system. This is probably due to the fact that effective induction of shikonin is supported by enhanced productivity in SHIK medium. It has been reported that natural auxin IAA (indole-3-acetic acid) is more favorable for shikonin production than the synthetic auxins such as 2, 4-D (2, 4-dichlorophenoxyacetic acid) (1). This agreed well with the result obtained from the present work employing immobilized cells that shikonin production was higher in SHIK than in SHCK medium (Fig. 4). Noting that SH medium does contain ammonium (ca. 2.6 raM), which was reported to be inhibitory for shikonin biosyn￾thesis (I), the present results indicate that the synthetic activity can be controlled by using appropriate growth regulators like IAA. Immobilization in the polymer matrices should also provide the plant cells with more favorable conditions for shikonin production as described above. Operation of packed-bed bioreactor A typical time course of shikonin production in packed-bed bioreactor is shown in Fig. 5. The packed-bed reactor system was a single-stage batch culture using SHIK medium for both cell growth and shikonin production. As in the flask cul￾ture, the cells grew to yield fully-packed immobilized cell aggregates and the shikonin produced was directly ad￾sorbed to the polyurethane matrices, resulting in a practi￾cally clear culture broth. It was noted that the product titre was higher than in flask culture, although slow cell growth and the initial lag in shikonin production were similar (see J. FERMENT. BIOENG., 3( x. = ~"-o. Q A \ ,9 --+ r5 x \ Time ( d ) FIG. 5. Typical time course of shikonin production in a packed￾bed bioreactor with immobilized L. erythrorhizon cells. Fig. 4). Direct comparison of the productivity in the present bioreactor system with that in other studies, such as the one by Tabata et al. (1), is very difficult because the activ￾ities of the cell lines used are quite different; the shikonin synthetic activity of the cell line used in the present study is lower by an order of magnitude than those used by Tabata et al. However, it is still considered that the present im￾mobilization technique can offer similar improvements to systems with cell lines of higher activity, since it can con￾siderably reduce the culture steps and process time. We believe that room still exists for further optimization of the bioreactor system, and that it should be able to give higher productivity than the present result. REFERENCES 1. Tabata, M. and Fujita, Y.: Production of shikonin by plant cell cultures, p. 207-218. Biotechnology in plant science. Academic Press, New York, N.Y. (1985). 2. Hulst, A.C., Tramper, J., Brodelius, P., Leon, J.C., and Luyhen, K. A. M.: Immobilized plant cells; respiration and oxy￾gen transfer. J. Chem. Tech. Biotech., 35B, 198-204 (1985). 3. Nakajima, H., Sonomoto, K., Morikawa, H., Sato, F., Iehimura, K., ¥amada, Y., and Tanaka, A.: Entrapment of Lavandula vera cells with synthetic resin prepolymers and its ap￾plication to pigment production. Appl. Microbiol. Biotechnol., 24, 266-270 (1986). 4. Mavituna, F. and Park, J. M.: Growth of immobilized plant cells in reticulate polyurethane foam matrices. Biotech. Lett., 7 (9), 637-640 (1985). 5. Shenk, R. U. and Hiidebrandt, A.C.: Medium and techniques for induction and growth of monocotyledonous and dicotyledonous plant cell cultures. Can J. Bot., 50, 199-204 (1972). 6. Fujita, Y., Hara, Y., Suga, C., and Morimoto, T.: Production of shikonin derivatives by cell suspension cultures of Lithospermum erythrorshizon. Plant Cell Reports, 1, 61-63 (1981). 7. Chaplin, M. F. and Kennedy, J. F. (eds.): Carbohydrate analysis; a practical approach, p. 3. IRL Press, Washington, D.C. (1986)