Improvement of Shikonin Productivity in Lithospermum erythrorhizon Cell Culture by Alternating Carbon and Nitrogen Feeding Strategy Venkatesh Srinivasan Department of Chemical Engineering,State University of New York, Buffalo,New York 13260 Dewey D.Y.Ryu* Department of Chemical Engineering,University of California,Davis, California 95616 Revised April 22,1993/Accepted April 27,1993 Stationary phase cell suspension cultures of Agrobac- was concluded in our earlier work16 that a low nitrogen terium tumefaciens transformed Lithospermum ery- level in the culture medium was essential for shikonin throrhizon respond to additions of sucrose-rich(C-rich) production by the cell line.In fact,the addition of excess medium with a 2-3-fold increase in the accumulation of shikonin derivatives and a 3-3.5-fold increase in nitrate to a production stage batch culture upon nitrate the accumulation of soluble phenolics while showing exhaustion reduced the product yield by 70-80%.How- a modest (10-30%)increase in cell concentration. ever,this nitrate effect was found to depend on the level Conversely,the addition of nitrate-rich(N-rich)medium of sucrose (carbon source)present at the time of nitrate resulted in 25-35%increase in biomass concentration but only 2-9%increase in shikonin production and addition,thereby suggesting that the inhibitory effect of ~3%increase in the yield of soluble phenolics.Repeated high nitrogen levels on shikonin yield was in fact mediated additions of C-rich medium resulted in only a modest by availability of the carbon source.A similar conclusion (less than 10%)improvement in shikonin production was also reached by Davies24 regarding the influence of over the levels obtained after the first application. nitrogen on polyphenol production by cell cultures of Paul's No obvious correlation could be discerned between scarlet rose.Our studies have also indicated a general intracellular ATP levels or protein synthesis patterns and the pattern of shikonin accumulation following the trend of increasing volumetric productivity of shikonin addition of C-rich medium,suggesting that the precursor derivatives in batch culture with increasing initial sucrose diversion mechanism is not generally applicable in concentrations in the range of 2-5 wt %Therefore a our cell line.It was found that alternating feeding logical strategy was to test the response of the culture of N-rich and C-rich media could be used as an effective strategy for enhancing the productivity of plant to additions of sucrose.Accordingly,when the sucrose secondary metabolite.1993 John Wiley Sons,Inc. level in the medium had dropped below 5 g/L,a known Key words:L.erythrorhizon.shikonin.carbon and quantity of sterile sucrose solution was added to the culture. nitrogen feeding However,it was found that the addition of sucrose alone (data not shown)resulted in negligible sucrose uptake and no enhancement of shikonin yield.Cell viability was INTRODUCTION also found to decrease.The inclusion of low levels of The infuence of various environmental parameters such nitrate and other salts alleviated this problem,suggesting as light intensity,medium composition,dissolved oxy- that the uptake of sucrose by stationary phase cultures gen,and temperature on the production of shikonin was limited by the depletion of other nutrients,possibly derivatives (naphthoquinone pigments)by cell cultures of phosphorus and nitrogen.Consequently,C-rich and N-rich Lithospermum erythrorhizon is well documented.6.7.9.17 Op- media(see Materials and Methods)were used in the ensuing timized media for growth and production of cell line experiments.Several reports in the literature2-5.12.14 have M-18 have been formulated.18 In an earlier article,16 indicated a stimulatory effect by elevated levels of carbon we reported the influence of the nitrogen source on source on natural product yields. shikonin production and the activities of the biosynthetic In this article we report on the continued stimulation enzymes phenylalanine ammonia lyase and 3-hydroxy- of shikonin production by stationary phase cultures of 3-methylglutaryl CoA reductase in an Agrobacterium L.erythrorhizon by additions of sucrose-rich media and tumefaciens transformed cell line of L.erythrorhizon.It we suggest a control strategy for improving the yield of secondary metabolites in this cell line.We have also To whom all correspondence should be addressed. examined the possibility of a general correlation between Biotechnology and Bioengineering,Vol.42,Pp.793-799(1993) 1993 John Wiley Sons,Inc. CCC0006-3592/93/070793-07
Improvement of Shikonin Productivity in Lithospermum erythrorhizon Cell Culture by Alternating Carbon and Nitrogen Feeding Strategy Venkatesh Srinivasan Department of Chemical Engineering, State University of New York, Buffalo, New York 13260 Dewey D.Y. Ryu* Department of Chemical Engineering, University of California, Davis, California 956 16 Revised April 22, 1993/Accepted April 27, 1993 Stationary phase cell suspension cultures of Agrobacteriurn turnefaciens transformed Lithospermum erythrorhizon respond to additions of sucrose-rich (C-rich) medium with a 2-3-fold increase in the accumulation of shikonin derivatives and a 3-3.5-fold increase in the accumulation of soluble phenolics while showing a modest (10-30%) increase in cell concentration. Conversely, the addition of nitrate-rich (N-rich) medium resulted in 25-35% increase in biomass concentration but only 2-9% increase in shikonin production and - 3% increase in the yield of soluble phenolics. Repeated additions of C-rich medium resulted in only a modest (less than 10%) improvement in shikonin production over the levels obtained after the first application. No obvious correlation could be discerned between intracellular ATP levels or protein synthesis patterns and the pattern of shikonin accumulation following the addition of C-rich medium, suggesting that the precursor diversion mechanism is not generally applicable in our cell line. It was found that alternating feeding of N-rich and C-rich media could be used as an effective strategy for enhancing the productivity of plant secondary metabolite. 0 1993 John Wiley & Sons, Inc. Key words: L. erythrorhizon shikonin carbon and nitrogen feeding INTRODUCTION The influence of various environmental parameters such as light intensity, medium composition, dissolved oxygen, and temperature on the production of shikonin derivatives (naphthoquinone pigments) by cell cultures of timized media for growth and production of cell line M-18 have been formulated.18 In an earlier article,16 we reported the influence of the nitrogen source on shikonin production and the activities of the biosynthetic enzymes phenylalanine ammonia lyase and 3-hydroxy- 3-methylglutaryl CoA reductase in an Agrobacterium tumefaciens transformed cell line of L. erythrorhizon. It Lithospermum erythrorhizon is well d~cumented.~,’~~,’~ O P- * To whom all correspondence should be addressed. was concluded in our earlier work16 that a low nitrogen level in the culture medium was essential for shikonin production by the cell line. In fact, the addition of excess nitrate to a production stage batch culture upon nitrate exhaustion reduced the product yield by 70430%. However, this nitrate effect was found to depend on the level of sucrose (carbon source) present at the time of nitrate addition, thereby suggesting that the inhibitory effect of high nitrogen levels on shikonin yield was in fact mediated by availability of the carbon source. A similar conclusion was also reached by Da~ies’,~ regarding the influence of nitrogen on polyphenol production by cell cultures of Paul’s scarlet rose. Our studies have also indicated a general trend of increasing volumetric productivity of shikonin derivatives in batch culture with increasing initial sucrose concentrations in the range of 2-5 wt %. Therefore a logical strategy was to test the response of the culture to additions of sucrose. Accordingly, when the sucrose level in the medium had dropped below 5 g/L, a known quantity of sterile sucrose solution was added to the culture. However, it was found that the addition of sucrose alone (data not shown) resulted in negligible sucrose uptake and no enhancement of shikonin yield. Cell viability was also found to decrease. The inclusion of low levels of nitrate and other salts alleviated this problem, suggesting that the uptake of sucrose by stationary phase cultures was limited by the depletion of other nutrients, possibly phosphorus and nitrogen. Consequently, C-rich and N-rich media (see Materials and Methods) were used in the ensuing experiments. Several reports in the literat~re~-~,’~?~~ have indicated a stimulatory effect by elevated levels of carbon source on natural product yields. In this article we report on the continued stimulation of shikonin production by stationary phase cultures of L. erythrorhizon by additions of sucrose-rich media and we suggest a control strategy for improving the yield of secondary metabolites in this cell line. We have also examined the possibility of a general correlation between Biotechnology and Bioengineering, Vol. 42, Pp. 793-799 (1993) 0 1993 John Wiley & Sons, Inc. CCC QQ06-3592/93/070793-Q7
secondary metabolic activity and intracellular protein and Cellular Adenosine Triphosphate Content ATP levels. Cells of known fresh weight were extracted in 3 mL of ice-cold 90%dimethylsulfoxide (DMSO)by vortexing for MATERIALS AND METHODS 30 s.Extracts were stored at-70C until further analysis Parallel controls revealed that the loss of adenosine triphos- Cell Line phate (ATP)in cell extracts was in the range of 1-7% Lithospermum erythrorhizon cell line transformed with during this period.This difference has been ignored in the Agrobacterium tumefaciens Ti plasmid was obtained from measurements.Prior to measurement,extracts were thawed Dr.Chang Ryul Liu at the Genetic Research Institute, at 4C and centrifuged at 500g for 10 min.The supernatant Taeduck,Korea. was diluted 100-fold in N-(2-hydroxyethyl)piperazine-N'- 2-ethanesulfonic acid (HEPES)buffer (0.5 mM,pH 7.8) Cell Cultivation and 0.1-mL aliquots were used for ATP determination by luminometry.A Turner TD-20e luminometer (Turner De- Cell suspensions were cultivated as described in our earlier paper.16 Thirteen-day-old cells were transferred (20%in- signs,Mountain View,CA)was used to monitor the emitted luminescence of the ATP coupled luciferin-luciferase enzy- oculum v/v)into production medium (M-918 with 5 wt% matic reaction.Lumit-QM luciferin-luciferase and diluent sucrose and without hormones).The culture volume was were obtained from Lumac BV (Lumac BV,Landgraaf, approximately 150 mL after inoculation.The medium was Netherlands).The ATP was quantified by comparing the adjusted to pH 5.9 and overlaid with liquid paraffin 1:6 relative luminescence units (RLUs)measured from cell (v/v)before autoclaving.Liquid paraffin was used to fa- extracts with those produced by known ATP standards cilitate in situ extraction and measurement of product.8 (prepared in HEPES buffer)measured by the same assay. Cultivation was carried out in 500-mL shake flasks with foam-plug closures at 25C in the dark at 125 rpm on a rotary shaker.In this article,control conditions refers to cul- Soluble Phenolics tivation in hormone-free M-9 medium as described above. Total soluble phenolics in the cell-free culture medium were Sucrose-rich (C-rich)and nitrate-rich (N-rich)media measured spectrophotometrically by the method of Swain were prepared as follows.To prepare C-rich medium, and Hillis as reported by Seitz et al.15 Caffeic acid was hormone-free M-9 medium!8(20%concentration without used as the standard. sucrose or nitrate)was prepared as the basal medium and the concentrations of sucrose and nitrate were adjusted to give a final concentration of 30-40 g/L sucrose and RESULTS AND DISCUSSION 15-25 mg/L nitrogen (as nitrate)after addition to the Figure 1 illustrates the growth and production of shikonin culture.The volume of addition was adjusted to compen- derivatives under control conditions (hormone-free M-9 sate for the cumulative sample volume withdrawn from medium,5 wt sucrose).Typically nitrate was consumed the flask.Approximately 5-6-mL samples were taken. by days 6-8 and sucrose by days 12-15.A two-to Hormone-free 2.5-fold concentrated M-9 medium18 was threefold increase in fresh weight was typically observed. designated as N-rich.Upon addition of N-rich medium,the Production of shikonin derivatives started between days 6 resultant concentration of nitrogen was typically between and 8 and the maximum concentration varied between 140 50 and 65 mg/L and sucrose was between 20 and 30 g/L. and 200 mg/L in various experiments.The data shown here It must be mentioned that the C-rich and N-rich definitions represent the mean of duplicate experiments. are based upon the nitrogen content of the medium. Figures 2A-D illustrate the effects of adding C-rich and N-rich media.Soluble phenolics were measured as Measurement of Cell Growth,Product,and an indication of general secondary metabolic activity.All Nutrient Levels experiments were conducted in duplicate.Days 1-13 in Cell growth,sucrose,and nitrate in the medium were Figure 2A represent cultivation under conditions of the con- measured by previously published methods.16 Shikonin trol experiment.A maximum cell yield of about 140 g/L concentration was measured by the method of Heide was obtained.Shikonin production was observed to start on et al. day 5 and a concentration of ~125 mg/L was measured on day 11.The soluble phenolic content of the medium Cellular Protein Content increased from ~20 mg/L on day 5 to ~160 mg/L on day 12.Shikonin and phenolic concentrations did not The cell mass of known fresh weight was homogenized in increase further until day 13.On day 13,C-rich medium 5 mL Tris-HCI (0.2 M,pH 7.1)using a tissue dismember- was added to the culture aseptically.Sucrose and nitrate ator(Fisher Scientific).The homogenate was centrifuged at concentrations following the addition are shown in the 500g for 10 min and the protein content of the supernatant figure.Sharp increases in soluble phenolics and shikonin was measured by the method of Bradford!using bovine derivative yields were observed after a lag of 24-36 h. serum albumin as the standard. Shikonin concentration increased almost twofold over a 794 BIOTECHNOLOGY AND BIOENGINEERING,VOL.42,NO.7,SEPTEMBER 20,1993
secondary metabolic activity and intracellular protein and ATP levels. MATERIALS AND METHODS Cell Line Lithosperrnurn erythrorhizon cell line transformed with Agrobacteriurn tumefaciens Ti plasmid was obtained from Dr. Chang Ryul Liu at the Genetic Research Institute, Taeduck, Korea. Cell Cultivation Cell suspensions were cultivated as described in our earlier paper.16 Thirteen-day-old cells were transferred (20% inoculum v/v) into production medium (M-9l' with 5 wt % sucrose and without hormones). The culture volume was approximately 150 mL after inoculation. The medium was adjusted to pH 5.9 and overlaid with liquid paraffin 1:6 (v/v) before autoclaving. Liquid paraffin was used to facilitate in situ extraction and measurement of product.' Cultivation was carried out in 500-mL shake flasks with foam-plug closures at 25°C in the dark at 125 rpm on a rotary shaker. In this article, control conditions refers to cultivation in hormone-free M-9 medium as described above. Sudrose-rich (C-rich) and nitrate-rich (N-rich) media were prepared as follows. To prepare C-rich medium, hormone-free M-9 medium'' (20% concentration without sucrose or nitrate) was prepared as the basal medium and the concentrations of sucrose and nitrate were adjusted to give a final concentration of 30-40 g/L sucrose and 15-25 mg/L nitrogen (as nitrate) after addition to the culture. The volume of addition was adjusted to compensate for the cumulative sample volume withdrawn from the flask. Approximately 5 -6-mL samples were taken. Hormone-free 2.5-fold concentrated M-9 medium" was designated as N-rich. Upon addition of N-rich medium, the resultant concentration of nitrogen was typically between SO and 65 mg/L and sucrose was between 20 and 30 g/L. It must be mentioned that the C-rich and N-rich definitions are based upon the nitrogen content of the medium. Measurement of Cell Growth, Product, and Nutrient Levels Cell growth, sucrose, and nitrate in the medium were measured by previously published methods.16 Shikonin concentration was measured by the method of Heide et a1.8 Cellular Protein Content The cell mass of known fresh weight was homogenized in 5 mL Tris-HC1 (0.2 M, pH 7.1) using a tissue dismemberator (Fisher Scientific). The homogenate was centrifuged at 500g for 10 min and the protein content of the supernatant was measured by the method of Bradford' using bovine serum albumin as the standard. Cellular Adenosine Triphosphate Content Cells of known fresh weight were extracted in 3 mL of ice-cold 90% dimethylsulfoxide (DMSO) by vortexing for 30 s. Extracts were stored at -70°C until further analysis. Parallel controls revealed that the loss of adenosine triphosphate (ATP) in cell extracts was in the range of 1-7% during this period. This difference has been ignored in the measurements. Prior to measurement, extracts were thawed at 4°C and centrifuged at 500g for 10 min. The supernatant was diluted 100-fold in N-(2-hydroxyethyl)piperazine-N'- 2-ethanesulfonic acid (HEPES) buffer (0.5 mM, pH 7.8) and 0.1-mL aliquots were used for ATP determination by luminometry. A Turner TD-20e luminometer (Turner Designs, Mountain View, CA) was used to monitor the emitted luminescence of the ATP coupled luciferin-luciferase enzymatic reaction. Lumit-QM luciferin-luciferase and diluent were obtained from Lumac BV (Lumac BV, Landgraaf, Netherlands). The ATP was quantified by comparing the relative luminescence units (RLUs) measured from cell extracts with those produced by known ATP standards (prepared in HEPES buffer) measured by the same assay. Soluble Phenolics Total soluble phenolics in the cell-free culture medium were measured spectrophotometrically by the method of Swain and Hillis as reported by Seitz et al.15 Caffeic acid was used as the standard. RESULTS AND DISCUSSION Figure 1 illustrates the growth and production of shikonin derivatives under control conditions (hormone-free M-9 medium, 5 wt % sucrose). Typically nitrate was consumed by days 6-8 and sucrose by days 12-15. A two- to threefold increase in fresh weight was typically observed. Production of shikonin derivatives started between days 6 and 8 and the maximum concentration varied between 140 and 200 mg/L in various experiments. The data shown here represent the mean of duplicate experiments. Figures 2A-D illustrate the effects of adding C-rich and N-rich media. Soluble phenolics were measured as an indication of general secondary metabolic activity. All experiments were conducted in duplicate. Days 1-13 in Figure 2A represent cultivation under conditions of the control experiment. A maximum cell yield of about 140 g/L was obtained. Shikonin production was observed to start on day 5 and a concentration of -125 mg/L was measured on day 11. The soluble phenolic content of the medium increased from -20 mg/L on day 5 to -160 mg/L on day 12. Shikonin and phenolic concentrations did not increase further until day 13. On day 13, C-rich medium was added to the culture aseptically. Sucrose and nitrate concentrations following the addition are shown in the figure. Sharp increases in soluble phenolics and shikonin derivative yields were observed after a lag of 24-36 h. Shikonin concentration increased almost twofold over a 794 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 42, NO. 7, SEPTEMBER 20, 1993
00 200 180 (100) 150 120 邑 B 100 14BIeM 80 40 8101214161820 Time (days) Figure 1.Cultivation under control conditions in regular production medium(M-9 medium with 5 wt sucrose and without phytohormones):(biomass,(O)nitrate,(sucrose,and ( shikonin.Data represent means of duplicate experiments. period of 8 days following the addition of fresh C- study this issue,the reverse sequence of media additions rich medium.A maximum concentration of ~220 mg/L was tried(Figs.2C,D).Following cultivation under control was measured.Correspondingly,the concentration of conditions for about 13 days,N-rich medium was added to phenolics increased almost 3.5-fold over a 6-day period the culture on day 13(Fig.2C).While cell growth was following the addition of C-rich medium.A maximum of strongly stimulated (~43%increase),the yield of shikonin -550 mg/L was measured.Following dilution of the derivatives remained the same (~140 mg/L)during the culture with fresh C-rich medium,the cell concentration 8-9-day period following the addition of N-rich medium. dropped from 140 g/L to ~100/L,and subsequently a Soluble phenolics were not measured in this experiment. modest increase in cell fresh weight(~15%)over a period On day 23,C-rich medium was added to the culture.After of 8 days could be observed.The N-rich medium was added a lag of 24-48 h,fresh accumulation of shikonin was to the culture on day 24(Fig.2A)to further test the validity noticed.This increase continued for a period of 10 days.A of earlier observations regarding the inhibitory effect of maximum concentration of 200-220 mg/L was measured, high nitrate levels.Almost all the added nitrate and ~83% reflecting 35%increase over levels on day 23. of the added sucrose was consumed over a period of 9 days Given the evidence of the stimulatory effect of C-rich following the addition of N-rich medium(Fig.2B).There medium on shikonin productivity,a natural question arises. was a noticeable increase in cell fresh weight (~25%) Will successive additions of C-rich medium result in further over the fresh weight on day 24.Conspicuously,there product accumulation or is there a limit to the stimulatory was almost no further increase in the concentration of effect?Might there be some limitation imposed by the cell's soluble phenolics.A meager 3-4%increase was recorded. biosynthetic apparatus at some stage? Likewise,the shikonin concentration showed only a very To address these questions,C-rich medium was added to slight increase (~9%)over the level on day 24.These cultures after 12 days of cultivation under control condi- results further confirmed previous observations regarding tions(Fig.2E).The anticipated increase in shikonin yield the influence of excess nitrogen on shikonin production. was observed.Shikonin levels increased ~2.2-fold from They also reveal that shikonin production and general 150 to 325 mg/L over 9-10 days following addition of secondary metabolic activity are strongly stimulated by the C-rich medium.Cell growth increased by approximately addition of C-rich medium.Also,N-rich medium supported 25%in this experiment.However,further addition of growth of the culture while preventing further product C-rich medium on day 24 resulted in very little con- accumulation. sumption of the added sucrose (Fig.2F)and showed only Due to continuously changing conditions during batch 12%increase in shikonin yield.A 20%increase in cell culture,there is continuous variation in cellular physiology. fresh weight was observed.A medium preparation without It may be argued,therefore,that the diverse effects of C-rich sucrose (C-rich salts)was also added on day 24 to some and N-rich media on cell growth and shikonin production, flasks in a parallel experiment (data not shown).While when administered in alternating succession,are really cell concentration increased by about 11%,product levels effects caused by altered cellular physiology during this remained unchanged from levels on day 24.These results period rather than direct influences of the added media.To suggest the existence of some intrinsic limitation to the SRINIVASAN AND RYU:ALTERNATING CARBON AND NITROGEN FEEDING 795
80 150 100 50 0 0 2 4 6 8 10 12 14 16 18 20 Time (days) Figure 1. Cultivation under control conditions in regular production medium (M-9 medium with 5 wt % sucrose and without phytohormones): (m) biomass, (0) nitrate, (0) sucrose, and (0) shikonin. Data represent means of duplicate experiments. period of 8 days following the addition of fresh Crich medium. A maximum concentration of -220 mg/L was measured. Correspondingly, the concentration of phenolics increased almost 3.5-fold over a 6-day period following the addition of C-rich medium. A maximum of -550 mg/L was measured. Following dilution of the culture with fresh C-rich medium, the cell concentration dropped from 140 g/L to -1OO/L, and subsequently a modest increase in cell fresh weight (-15%) over a period of 8 days could be observed. The N-rich medium was added to the culture on day 24 (Fig. 2A) to further test the validity of earlier observations regarding the inhibitory effect of high nitrate levels. Almost all the added nitrate and -83% of the added sucrose was consumed over a period of 9 days following the addition of N-rich medium (Fig. 2B). There was a noticeable increase in cell fresh weight (-25%) over the fresh weight on day 24. Conspicuously, there was almost no further increase in the concentration of soluble phenolics. A meager 3-4% increase was recorded. Likewise, the shikonin concentration showed only a very slight increase (-9%) over the level on day 24. These results further confirmed previous observations regarding the influence of excess nitrogen on shikonin production. They also reveal that shikonin production and general secondary metabolic activity are strongly stimulated by the addition of C-rich medium. Also, N-rich medium supported growth of the culture while preventing further product accumulation. Due to continuously changing conditions during batch culture, there is continuous variation in cellular physiology. It may be argued, therefore, that the diverse effects of C-rich and N-rich media on cell growth and shikonin production, when administered in alternating succession, are really effects caused by altered cellular physiology during this period rather than direct influences of the added media. To study this issue, the reverse sequence of media additions was tried (Figs. 2C, D). Following cultivation under control conditions for about 13 days, N-rich medium was added to the culture on day 13 (Fig. 2C). While cell growth was strongly stimulated (-43% increase), the yield of shikonin derivatives remained the same (-140 mg/L) during the 8-9-day period following the addition of N-rich medium. Soluble phenolics were not measured in this experiment. On day 23, C-rich medium was added to the culture. After a lag of 24-48 h, fresh accumulation of shikonin was noticed. This increase continued for a period of 10 days. A maximum concentration of 200-220 mg/L was measured, reflecting 35% increase over levels on day 23. Given the evidence of the stimulatory effect of C-rich medium on shikonin productivity, a natural question arises. Will successive additions of C-rich medium result in further product accumulation or is there a limit to the stimulatory effect? Might there be some limitation imposed by the cell's biosynthetic apparatus at some stage? To address these questions, C-rich medium was added to cultures after 12 days of cultivation under control conditions (Fig. 2E). The anticipated increase in shikonin yield was observed. Shikonin levels increased -2.2-fold from 150 to 325 mg/L over 9-10 days following addition of C-rich medium. Cell growth increased by approximately 25% in this experiment. However, further addition of C-rich medium on day 24 resulted in very little consumption of the added sucrose (Fig. 2F) and showed only 12% increase in shikonin yield. A 20% increase in cell fresh weight was observed. A medium preparation without sucrose (C-rich salts) was also added on day 24 to some flasks in a parallel experiment (data not shown). While cell concentration increased by about 11%, product levels remained unchanged from levels on day 24. These results suggest the existence of some intrinsic limitation to the SRlNlVASAN AND RYU: ALTERNATING CARBON AND NITROGEN FEEDING 795
420 2A C-Rlch N-RIch 360 720 300 600 240 490 (7/6)ySe 180 360 120 240 0 10 15 20 30 Time (days) 420 2B 360 C-Rich N-Rich 300 2a0 120 90 60 15 20 25 30 20 25 Time (days) Timo (days) 100 2D 420 2E N-Rich C-Rich 380 300 60 240 120 1a0 40 60 20 20 25 30 Time (days) Time (days) Figure 2.Additions of C-rich or N-rich media at various stages during cultivation.Additions are indicated by arrows and appropriate labels.Typically, days 1-13 represent cultivation under control conditions and subsequent data are for cultivation in specific environments as indicated:(O)nitrate,() sucrose,()biomass fresh weight,and (shikonin,and (A)soluble phenolics.Data represent means of duplicate experiments. 796 BIOTECHNOLOGY AND BIOENGINEERING,VOL.42,NO.7,SEPTEMBER 20,1993
' 420 2A C-Rlch N-Alch 11 1 360 - 240 160' + - 0 - '0 120' 180 5 g 80 120 'D - 0 40 60 0 0 0 0 5 10 15 20 25 30 35 Time (days) I ZB C-Rch 1 Time (days) 0 5 10 15 20 25 30 35 40 Time (days) t 2c Time (days) Time (days) Figure 2. Additions of C-rich or N-rich media at various stages during cultivation. Additions are indicated by arrows and appropriate labels. Typically, days 1-13 represent cultivation under control conditions and subsequent data are for cultivation in specific environments as indicated: (0) nitrate, (0) sucrose, (U) biomass fresh weight, and (0) shikonin, and (A) soluble phenolics. Data represent means of duplicate experiments. 796 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 42, NO. 7, SEPTEMBER 20, 1993
were cultivated in either the growth medium (hormone-free Schenck and Hildebrandt medium with 5 wt sucrose)or the normal production medium.No shikonin is produced in the growth stage cultures.A classic pattern of variation in cellular ATP content was observed.The ATP content increased from low levels during the lag phase and attained a maximum value in the mid-to late-exponential growth phase (Fig.3A).Following this,a steady decrease was observed in the stationary phase and lag phase levels were reached again.Cellular protein,however,varied in different ways in the growth and production stages.In the growth medium,three distinct regions were observed (Fig.3A). Protein content per fresh weight of cell mass increased during the lag and early exponential growth phases.It dropped to ~33%of peak value and remained constant 25 30 at this value during the midexponential phase,probably Time (days) reflecting a brief period of balanced growth,and dropped an additional~60%during the late-exponential phase and Figure 2.(continued) remained essentially constant through the stationary phase. In the production medium,however,cellular protein content continued stimulation of shikonin production by C-rich appeared to follow the growth curve (Fig.3B).A pattern medium.It may be noted (Fig.2C)that it was possible to similar to the control culture(production stage,days 1-12, stimulate production by C-rich medium by first cultivating Fig.3B)was also observed in the two stages of cultiva- in N-rich medium.These observations indicate that there tion following additions of C-rich medium (days 13-35, may in fact be a biosynthetic potential for secondary Fig.3B).Cellular protein content appeared to be a function metabolite synthesis that is exhausted by repeated cultiva- of culture dilution,and increases in protein content occurred tion in a C-rich environment but which can be successfully over the same durations as increases in biomass.Rapid replenished by brief periods of cultivation in N-rich or increases were also observed in cellular ATP content.We growth-supporting conditions.This biosynthetic potential had hoped to see some characteristic changes in the patterns could potentially involve energy reserves,levels of key of cellular protein and ATP levels that might correlate to biosynthetic enzymes,and other metabolic precursors.As secondary metabolic activity,but this was not observed. a result,a control strategy,based on cycling the culture The pattern of variation in ATP content was qualitatively between C-rich and N-rich conditions,may make it possible similar in C-rich or N-rich medium.Clearly,a simple to obtain higher culture productivity for extended periods criterion such as cellular ATP or protein content is not a of time. sufficient indicator of the proposed biosynthetic potential. It is well established that the start of secondary metabolic Keeping in mind the complex regulatory mechanisms and activity in many microorganisms corresponds closely to the myriad of environmental stimuli (light,nutrient,and the depletion of cellular ATP.10.11 While the cessation of physical stresses,temperature,and infection by pathogens) cellular protein synthesis is usually a good indication of that are known to affect secondary metabolite synthesis the onset of secondary metabolism,it fails to be so in those in culture plant cells,these processes are most likely cases where limited cell growth is observed.Besides,the mediated by complex intracellular signals in addition to early biosynthetic pathway of shikonin derivatives consists those considered here.The detailed intracellular responses of two branches.These are the phenylpropanoid branch of our cell line to nutrient cycling need to be studied leading to the synthesis of p-hydroxybenzoic acid from further. L-phenylalanine and the terpenoid pathway leading to the The results described above therefore suggest that the synthesis of the high-energy intermediate geranyl pyrophos- cellular response,namely,cell growth in N-rich medium phate (GPP)from mevalonic acid(MVA)via isopentynyl and enhanced production in C-rich medium,is not sig- pyrophosphate and dimethylallyl pyrophosphate.Many of nificantly affected by the alternating C-rich and N-rich the enzymes in the latter pathway such as GPP synthase environments during the period of time in which these ex- and MVA pyrophosphate decarboxylase are ATP-requiring periments were conducted.The yield coefficients of cell enzymes suggesting that the energy status of the cell may and product were estimated to understand the nutrient re- influence the production of shikonin derivatives.In an quirement under different cultivation conditions (Table I). attempt to gain some understanding of the intracellular The product yield,Yprx,is approximately 2-fold better in parameters that the proposed biosynthetic potential might C-rich condition but is only one 16th in N-rich condition. correlate to,we measured the cellular protein and ATP This is probably so since a substantial portion of the prod- concentrations under nonproducing and producing condi- uct was produced in the late exponential and the stationary tions.The results are shown in Figures 3A and B.Cells phases,a period when cell growth was not significant,a SRINIVASAN AND RYU:ALTERNATING CARBON AND NITROGEN FEEDING 797
100 . , . , . , . , . , . , . , . fi 2F 0 5 10 15 20 25 30 35 40 Time (days) Figure 2. (continued) continued stimulation of shikonin production by C-rich medium. It may be noted (Fig. 2C) that it was possible to stimulate production by C-rich medium by first cultivating in N-rich medium. These observations indicate that there may in fact be a biosynthetic potential for secondary metabolite synthesis that is exhausted by repeated cultivation in a C-rich environment but which can be successfully replenished by brief periods of cultivation in N-rich or growth-supporting conditions. This biosynthetic potential could potentially involve energy reserves, levels of key biosynthetic enzymes, and other metabolic precursors. As a result, a control strategy, based on cycling the culture between C-rich and N-rich conditions, may make it possible to obtain higher culture productivity for extended periods of time. It is well established that the start of secondary metabolic activity in many microorganisms corresponds closely to the depletion of cellular ATP.'o," While the cessation of cellular protein synthesis is usually a good indication of the onset of secondary metabolism, it fails to be so in those cases where limited cell growth is observed. Besides, the early biosynthetic pathway of shikonin derivatives consists of two branches. These are the phenylpropanoid branch leading to the synthesis of p-hydroxybenzoic acid from L-phenylalanine and the terpenoid pathway leading to the synthesis of the high-energy intermediate geranyl pyrophosphate (GPP) from mevalonic acid (MVA) via isopentynyl pyrophosphate and dimethylallyl pyrophosphate. Many of the enzymes in the latter pathway such as GPP synthase and MVA pyrophosphate decarboxylase are ATP-requiring enzymes suggesting that the energy status of the cell may influence the production of shikonin derivatives. In an attempt to gain some understanding of the intracellular parameters that the proposed biosynthetic potential might correlate to, we measured the cellular protein and ATP concentrations under nonproducing and producing conditions. The results are shown in Figures 3A and B. Cells were cultivated in either the growth medium (hormone-free Schenck and Hildebrandt medium with 5 wt % sucrose) or the normal production medium. No shikonin is produced in the growth stage cultures. A classic pattern of variation in cellular ATP content was observed. The ATP content increased from low levels during the lag phase and attained a maximum value in the mid- to late-exponential growth phase (Fig. 3A). Following this, a steady decrease was observed in the stationary phase and lag phase levels were reached again. Cellular protein, however, varied in different ways in the growth and production stages. In the growth medium, three distinct regions were observed (Fig. 3A). Protein content per fresh weight of cell mass increased during the lag and early exponential growth phases. It dropped to -33% of peak value and remained constant at this value during the midexponential phase, probably reflecting a brief period of balanced growth, and dropped an additional -60% during the late-exponential phase and remained essentially constant through the stationary phase. In the production medium, however, cellular protein content appeared to follow the growth curve (Fig. 3B). A pattern similar to the control culture (production stage, days 1 - 12, Fig. 3B) was also observed in the two stages of cultivation following additions of C-rich medium (days 13-35, Fig. 3B). Cellular protein content appeared to be a function of culture dilution, and increases in protein content occurred over the same durations as increases in biomass. Rapid increases were also observed in cellular ATP content. We had hoped to see some characteristic changes in the patterns of cellular protein and ATP levels that might correlate to secondary metabolic activity, but this was not observed. The pattern of variation in ATP content was qualitatively similar in C-rich or N-rich medium. Clearly, a simple criterion such as cellular ATP or protein content is not a sufficient indicator of the proposed biosynthetic potential. Keeping in mind the complex regulatory mechanisms and the myriad of environmental stimuli (light, nutrient, and physical stresses, temperature, and infection by pathogens) that are known to affect secondary metabolite synthesis in culture plant cells, these processes are most likely mediated by complex intracellular signals in addition to those considered here. The detailed intracellular responses of our cell line to nutrient cycling need to be studied further. The results described above therefore suggest that the cellular response, namely, cell growth in N-rich medium and enhanced production in C-rich medium, is not significantly affected by the alternating C-rich and N-rich environments during the period of time in which these experiments were conducted. The yield coefficients of cell and product were estimated to understand the nutrient requirement under different cultivation conditions (Table I). The product yield, Yplx, is approximately 2-fold better in C-rich condition but is only one 16th in N-rich condition. This is probably so since a substantial portion of the product was produced in the late exponential and the stationary phases, a period when cell growth was not significant, a SRlNlVASAN AND RYU: ALTERNATING CARBON AND NITROGEN FEEDING 797
3A 9.9 7120 8.25 100 歪 6.6 80 4.95 60 3.3 40 1.65 20 10 15 20 TIme (days) ,3B Product yield C-Rich 640 480 320 180 30 40 Time (days) Figure 3.Variation in cellular ATP and protein content in (A)growth and (B)production media.In(B),Days 1-13 represent cultivation under control production conditions and subsequent cultivation is under C-rich conditions.Biomass concentration(dashed line)and product yield (dotted line)are indicated schematically.Numerical values may be found in Fig.2E.Additions of C-rich media are indicated by arrows.(Biomass in (A),both (ATP and (A)protein content in (A)and (B).All data are mean values from duplicate experiments. situation similar to cultivation in C-rich medium.However, metabolite production can occur simultaneously in this cell further studies revealed that cell growth and production line. could occur simultaneously.It can be seen (Fig.3B)that We have shown that addition of sucrose-rich medium shikonin production during cultivation under control con- to stationary phase cultures of L.erythrorhizon strongly ditions (days 1-13)started before protein synthesis had stimulates general secondary metabolic activity indicated by ceased.The same was also true for cultivation under C. the rapid accumulation of soluble phenolics and specifically rich conditions.In fact,in C-rich conditions,substantial the accumulation of shikonin derivatives.A relatively small accumulation of product was observed while cellular pro- increase in biomass was also observed.On the other hand, tein levels continued to rise.Thus,while protein synthesis N-rich medium stimulates biomass growth and represses the follows cell growth fairly closely and while elevated ni- accumulation of shikonin derivatives.Although most of the trogen levels were found to be inhibitory to shikonin product accumulated in the late-exponential and stationary production,it is evident that cell growth and secondary phases in control cultures,we did not observe a general 798 BIOTECHNOLOGY AND BIOENGINEERING,VOL.42,NO.7,SEPTEMBER 20,1993
3A .3B 0 5 10 15 20 25 Tlme (days) J - F E" 5 . - 640 9.9 1 120 t Product yield / C-Rich 't C-Rich / I 0 10 20 30 40 " Time (days) Figure 3. Variation in cellular ATP and protein content in (A) growth and (B) production media. In (B), Days 1-13 represent cultivation under control production conditions and subsequent cultivation is under C-rich conditions. Biomass concentration (dashed line) and product yield (dotted line) are indicated schematically. Numerical values may be found in Fig. 2E. Additions of C-rich media are indicated by arrows. (W) Biomass in (A), both (0) ATP and (A) protein content in (A) and (B). All data are mean values from duplicate experiments. situation similar to cultivation in C-rich medium. However, further studies revealed that cell growth and production could occur simultaneously. It can be seen (Fig. 3B) that shikonin production during cultivation under control conditions (days 1 - 13) started before protein synthesis had ceased. The same was also true for cultivation under Crich conditions. In fact, in C-rich conditions, substantial accumulation of product was observed while cellular protein levels continued to rise. Thus, while protein synthesis follows cell growth fairly closely and while elevated nitrogen levels were found to be inhibitory to shikonin production, it is evident that cell growth and secondary metabolite production can occur simultaneously in this cell line. We have shown that addition of sucrose-rich medium to stationary phase cultures of L. erythrorhizon strongly stimulates general secondary metabolic activity indicated by the rapid accumulation of soluble phenolics and specifically the accumulation of shikonin derivatives. A relatively small increase in biomass was also observed. On the other hand, N-rich medium stimulates biomass growth and represses the accumulation of shikonin derivatives. Although most of the product accumulated in the late-exponential and stationary phases in control cultures, we did not observe a general 798 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 42, NO. 7, SEPTEMBER 20, 1993
Table I.Nutrient partitioning in different culture conditions. 3.Davies,M.E.1972.Polyphenol synthesis in cell suspension cultures of Paul's scarlet rose.Planta 104:50-65. Yield coefficient Control N rich C rich 4.Davies,M.E.1972.Effects of auxin on polyphenol accumulation and the development of phenylalanine ammonia-lyase activity in YXIN 0.671±0.12 0.374±0.06 0.53±0.26 darkgrown suspension cultures of Paul's scarlet rose.Planta 104 YxIC 0.16±0.05 0.06±0.006 0.017±0.003 66-77. YPix 0.067±0.02 0.004±0.001 0.12±0.02 5.Do,C.B.,Cormier,F.1991.Effects of low nitrate and high Individual yield coefficients represent the averages from five sets of data sugar concentrations on anthocyanin content and composition of for control conditions and four sets of N-rich or C-rich data.YxIN g grape (Vitis vinefera L.)cell suspension.Plant Cell Rep.9: dry weight cells/mmol NO3.Yxic=g dry weight cells/mmol sucrose. 500-504. Yp/x mg shikonin derivatives/g dry weight cells. 6.Fujita,Y.,Hara,Y.,Suga,C..Morimoto,T.1981.Production of shikonin derivatives by cell suspension cultures of Lithospermum erythrorhizon.II.A new medium for the production of shikonin exclusiveness of growth and shikonin production.Thus, derivatives.Plant Cell Rep.1:61-63. the stimulation of shikonin production by C-rich medium 7.Fujita,Y.1988.Shikonin production by plant (Lithospermum ery- cannot be explained entirely by diversion of precursors throrhizon)cell cultures.pp.225-236 In:Y.P.S.Bajaj (ed.),Biotech- nology in agriculture and forestry,vol.4,Medicinal and aromatic for growth into the shikonin biosynthetic pathways as plants I.Springer-Verlag,Berlin,Heidelberg. suggested by Mizukami et al.3.The lack of repeated stimu- 8.Heide,L.,Tabata,M.1987.Enzyme activities in cell-free extracts lation by C-rich medium indicated a saturation of metabolic of shikonin producing Lithospermum erythrorhizon cell suspension activity.We also observed an 8-11%drop in cell viability cultures.Phytochemistry 26:1645-1650. between days 24 and 36 from the value on day 24 (71% 9.Heide,L,Nishioka,N.,Fukui,H.,Tabata,M.1989.Enzymatic regulation of shikonin biosynthesis in Lithospermum erythrorhizon measured by phenosafranin dye exclusion).It is also pos- cell cultures.Phytochemistry 28:1873-1877. sible that successive additions of C-rich medium resulted 10.Janglova,Z.,Suchy,J.,Vanck,Z.1969.Regulation of biosynthesis in altered cellular physiology.Interim cultivation in N-rich of secondary metabolites.7.Intracellular adenosine-5'-triphosphate medium alleviated this problem. concentration in Streptomyces aureofaciens.Folia Microbiol.13 Although the literature is rich with instances of hor- 208-210. 11.Madry,N.,Sprinkmeyer,R.,Pape,H.1979.Regulation of tylosin monal regulation and regulation by specific precursors of synthesis in Streptomyces:Effects of glucose analogs and inorganic secondary metabolites,the effect of carbon and nitrogen phosphate.Eur.J.Appl.Microbiol.Biotechnol.7:365-370. source is still significant.As the relatively small changes 12.Merillon,J.M.,Rideau,M.,Chenieux,J.C.1984.Influence of sucrose in cell and product yields following switching between on levels of ajmalicine,serpentine and tryptamine in Catharanthus roseus cells in vitro.Planta Med.50:497-501. media demonstrates,it is possible to significantly improve 13.Mizukami,H.,Konoshima,M.,Tabata,M.1977.Effect of nutritional overall shikonin production by cycling the environment of factors on shikonin derivative formation in Lithospermum erythrorhi L.erythrorhizon cell cultures between N-rich and C-rich zon callus cultures.Phytochemistry 16:1183-1186. states.It is expected that by cultivation under semicontin- 14.Sarkissian,G.S.,Fowler,M.W.1986.The metabolism and uous or continuous conditions,better control over product utilization of carbohydrates by suspension cultures of plant cells. and cell yields can be achieved.Such studies are being pp.151-181 In:M.J.Morgan (ed.).Carbohydrate metabolism in cultured cells.Plenum,New York and London. actively considered. 15.Seitz,H.U.,Eilert,U.,De Luca,V.,Kurz,W.G.W.1989.Elicitor We thank Dr.Liu Chang Ryul at the Genetic Research Institute in mediated induction of phenylalanine ammonia lyase and tryptophan Daeduck,Korea for providing the Lithospermum erythrorhizon cell decarboxylase:Accumulation of phenols and indole alkaloids in cell line and the National Science Foundation for its partial support of suspension cultures of Catharanthus roseus.Plant Cell Tissue and this research. Organ Culture 18:71-78. 16.Srinivasan,V.,Ryu,D.D.Y.1992.Enzyme activity and shikonin production in Lithospermum erythrorhizon cell cultures.Biotechnol. References Bioeng.40:69-74. 17.Tabata,M.,Mizukami,H.,Hiraoka,N.,Konoshima,M.1973.Pig- 1.Bradford,M.M.1976.A rapid and sensitive method for the quan- ment formation in callus cultures of L.erythrorhizon.Phytochemistry titation of microgram quantities of protein using the principles of 13:927-932. protein-dye binding.Anal.Biochem.72:248-254. 18.Tabata,M..Fujita,Y.1985.Production of shikonin derivatives 2.Cormier,F.,Crevier,H.A.,Do,C.B.1990.Effects of sucrose by plant cell cultures.pp.207-218 In:M.Zaitlin,P.Day,and concentration on the accumulation of anthocyanins in grape (Vitis A.Hollaender (eds.)Biotechnology in plant science.Academic, vinifera)cell suspension.Can.J.Bot.68:1822-1825. Orlando,FL. SRINIVASAN AND RYU:ALTERNATING CARBON AND NITROGEN FEEDING 799
Table I. Nutrient partitioning in different culture conditions. Yield coefficient Control N rich C rich YXIN 0.671 0.12 0.374 f 0.06 0.53 z 0.26 YXlC 0.16 * 0.05 0.06 * 0.006 0.017 * 0.003 YPlX 0.067 2 0.02 0.004 f 0.001 0.12 -c 0.02 Individual yield coefficients represent the averages from five sets of data for control conditions and four sets of N-rich or C-rich data. YXIN = g dry weight cells/mmol NO3-. YX/C = g dry weight cells/mmol sucrose. Yplx = mg shikonin derivatives/g dry weight cells. exclusiveness of growth and shikonin production. Thus, the stimulation of shikonin production by C-rich medium cannot be explained entirely by diversion of precursors for growth into the shikonin biosynthetic pathways as suggested by Mizukami et al.13. The lack of repeated stimulation by C-rich medium indicated a saturation of metabolic activity. We also observed an 8-11% drop in cell viability between days 24 and 36 from the value on day 24 (71% measured by phenosafranin dye exclusion). It is also possible that successive additions of C-rich medium resulted in altered cellular physiology. Interim cultivation in N-rich medium alleviated this problem. Although the literature is rich with instances of hormonal regulation and regulation by specific precursors of secondary metabolites, the effect of carbon and nitrogen source is still significant. As the relatively small changes in cell and product yields following switching between media demonstrates, it is possible to significantly improve overall shikonin production by cycling the environment of L. erythrorhizon cell cultures between N-rich and C-rich states. It is expected that by cultivation under semicontinuous or continuous conditions, better control over product and cell yields can be achieved. Such studies are being actively considered. We thank Dr. Liu Chang Ryul at the Genetic Research Institute in Daeduck, Korea for providing the Lithospermum erythrorhizon cell line and the National Science Foundation for its partial support of this research. References 1. 2. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein using the principles of protein-dye binding. Anal. Biochem. 72: 248-254. Cormier, F., Crevier, H.A., Do, C.B. 1990. Effects of sucrose concentration on the accumulation of anthocyanins in grape (Vitis vinifera) cell suspension. Can. J. Bot. 68: 1822-1825. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. Davies, M. E. 1972. Polyphenol synthesis in cell suspension cultures of Paul’s scarlet rose. Planta 104: 50-65. Davies, M.E. 1972. Effects of auxin on polyphenol accumulation and the development of phenylalanine ammonia-lyase activity in darkgrown suspension cultures of Paul’s scarlet rose. Planta 104: 66-77. Do, C.B., Cormier, F. 1991. Effects of low nitrate and high sugar concentrations on anthocyanin content and composition of grape (Vitis vinefera L.) cell suspension. Plant Cell Rep. 9: 500-504. Fujita, Y., Hara, Y., Suga, C., Morimoto, T. 1981. Production of shikonin derivatives by cell suspension cultures of Lithospermurn erythrorhizon. 11. A new medium for the production of shikonin derivatives. Plant Cell Rep. 1: 61-63. Fujita, Y. 1988. Shikonin production by plant (Lithospermum erythrorhizon) cell cultures. pp. 225-236 In: Y. P. S. Bajaj (ed.), Biotechnology in agriculture and forestry, vol. 4, Medicinal and aromatic plants 1. Springer-Verlag, Berlin, Heidelberg. Heide, L., Tabata, M. 1987. Enzyme activities in cell-free extracts of shikonin producing Lithospermum erythrorhizon cell suspension cultures. Phytochemistry 26: 1645-1650. Heide, L., Nishioka, N., Fukui, H., Tabata, M. 1989. Enzymatic regulation of shikonin biosynthesis in Lithospermum erythrorhizon cell cultures. Phytochemistry 28: 1873- 1877. Janglova, Z., Suchy, J., Vanck, Z. 1969. Regulation of biosynthesis of secondary metabolites. 7. Intracellular adenosine-5’-triphosphate concentration in Streptomyces aureofaciens. Folia Microbiol. 13: 208 - 21 0. Madry, N., Sprinkmeyer, R., Pape, H. 1979. Regulation of tylosin synthesis in Streptomyces: Effects of glucose analogs and inorganic phosphate. Eur. J. Appl. Microbiol. Biotechnol. 7: 365-370. Merillon, J. M., Rideau, M., Chenieux, J. C. 1984. Influence of sucrose on levels of ajmalicine, serpentine and tryptamine in Catharanthus roseus cells in vitro. Planta Med. 50: 497-501. Mizukami, H., Konoshima, M., Tabata, M. 1977. Effect of nutritional factors on shikonin derivative formation in Lithospermum erythrorhizon callus cultures. Phytochemistry 16: 1183- 1186. Sarkissian, G. S., Fowler, M. W. 1986. The metabolism and utilization of carbohydrates by suspension cultures of plant cells. pp. 151-181 In: M.J. Morgan (ed.), Carbohydrate metabolism in cultured cells. Plenum, New York and London. Seitz, H. U., Eilert, U., De Luca, V., Kurz, W.G. W. 1989. Elicitor mediated induction of phenylalanine ammonia lyase and tryptophan decarboxylase: Accumulation of phenols and indole alkaloids in cell suspension cultures of Catharanthus roseus. Plant Cell Tissue and Organ Culture 18: 71-78. Srinivasan, V., Ryu, D. D. Y. 1992. Enzyme activity and shikonin production in Lithospermum erythrorhizon cell cultures. Biotechnol. Bioeng. 40: 69-74. Tabata, M., Mizukami, H., Hiraoka, N., Konoshima, M. 1973. Pigment formation in callus cultures of L. erythrorhizon. Phytochemistry 13: 927-932. Tabata, M., Fujita, Y. 1985. Production of shikonin derivatives by plant cell cultures. pp. 207-218 In: M. Zaitlin, P. Day, and A. Hollaender (eds.), Biotechnology in plant science. Academic, Orlando, FL. SRlNlVASAN AND RYU: ALTERNATING CARBON AND NITROGEN FEEDING 799