Jotrm~ oF Bxoscmrcc~.ANDBmE~rOmEErU~G Vol. 91, No. 4, 382-389. 2001 Effect of Consumed Carbon to Nitrogen Ratio on Mycelial Morphology and Arachidonic Acid Production in Cultures of Mortierella alpina YASUHISA KOIKE, l H O N G JIE CAI, 1 K E N I C H I HIGASHIYAMA, 2 SHIGEAKI FUJIKAWA, 2 AND E N O C H Y. PARK 1. Laboratory of Biotechnology, Department of Applied Biological Chemistry, Faculty of Agriculture, Shizuoka University, 836 Ohya, Shizuoka 422-85291 and Institute for Fundamental Research, Suntory Ltd., 5-2-5 Yamazaki, Shimamot o-cho, Mishima-gun, Osaka 618-0001, 2 Japan Received 30 October 2000/Accepted 12 January 2001 The influence of the consumed carbon to nitrogen (C/N) ratio on arachidonic acid (AA) production and mycelial morphology was investigated in cultures of Mortierella alpina using shake flasks and a fermentor. The consumed C / N ratio was varied from 5 to 32 under the condition that the total initial amount of carbon and nitrogen sources was 50 g/l. Cellular yield increased markedly at C / N ratios below 7; carbon utilization was switched from cellular growth to lipid biosynthesis in the C / N ratio range of 7-15; lipid biosynthesis was most active when the C / N ratio was in the range of 15-32. However, for C / N ratios higher than 15, the mycelial concentration decreased due to nitrogen limitation but the lipid yield still increased. In the presence of excess nitrogen, the biomass concentration depended on the amount of the nitrogen source, but the AA yield was inversely related to this. On the other hand, in the presence of excess carbon, the fatty acid concentration increased with carbon source concentration but the A A concentration remained constant. From the viewpoint of A A production, the optimum C / N ratio was in the range of 15 to 20 with a balance between the amounts of carbon and nitrogen sources. When an enriched medium was used at a fixed C / N ratio of 20, the cellular and AA concentrations were shown to be proportional to the total concentrations of carbon and nitrogen sources in both flasks and the fermentor. The whole pellet size and width of pellet annular regions did not change with increasing C / N ratio for C / N ratios below 20 in the flask cultures. However, when the C / N ratio was higher than 20, these sizes increased in proportion to the C / N ratio. [Key words: Mortierellaalpina, arachidonic acid, mycelial morphology, C/N ratio] natural nitrogen source for cell growth during a long period of cultivation. Therefore, the kinetic models of the consumed C / N ratio described previously (4, 13-15) are not very appropriate to industrial cultivation. The other consideration is the morphology of filamentous microorganisms, which usually varies between "pellet" and "filamentous", depending on the culture conditions and the genotype of the strains. In the case of Mortierella ramanniana, producing ~'-linolenic acid, a higher amount of lipid was accumulated when fungus exhibited the pellet morphology than when it exhibited the filamentous morphology (16). To evaluate mycelial morphology quantitatively, Paul and Thomas (17) reviewed some of the image analysis techniques used to quantify the morphology of microorganisms from fermentation samples, and describe some of the parameters used for characterizing morphology, such as size, shape, roughness, and cell volume. Using image analysis morphology can be characterized automatically (18). We have also investigated the effects of shear stress on morphological change in Streptomyces fradiae culture and relationship between mycelial morphology and tyrosin productivity (19). In the case of Mortierella alpina cultures the morphology was affected by the dissolved oxygen concentration (20), mineral addition (21), and the natural nitrogen source (22). When the dissolved oxygen concentration was maintained at 20-50 ppm the morphology changed from filamentous to pellet (20). With the addition of potassium dihydrogen phosphate the morphology was Polyunsaturated fatty acids (PUFAs) are abundant in humans and have an important role in the structure and function of biological membranes (1). One of the PUFAs, arachidonic acid (5, 8, 11, 14-cis-eicosatetraenoic acid, AA), a biogenic precursor of prostaglandins and leukotrienes, has been the subject of intensive medical research (2, 3). Polyunsaturated fatty acids such as AA have been reported to accumulate in ceils, algae, yeast and fungi (4, 5). Mortierella fungi are known to accumulate AA to approximately 70% of the total fatty acids (6). In general, it is assumed that PUFAs are accumulated due to a metabolic shift in the Krebs cycle caused by activation of ATP-dependent citrate lyase. Therefore, a very important factor affecting lipid over-production is the metabolic shift caused by nitrogen limitation (7, 8). Several researchers have reported on the influence of various culture conditions on lipid production, i.e. composition of the culture medium (9), effect of nutrient limitation (8), oxygen availability (10), pH and temperature (11). Among these parameters, nitrogen limitation had a significant effect on lipid content and fatty acid composition under oxygen-non-limited culture conditions (12). However, these results were obtained from experiments in while a soluble nitrogen source was used. The nitrogen sources used in industrial production are natural, and are usually insoluble in liquid media. Microorganisms use the nitrogen leached out from the insoluble * Corresponding author, e-mail: yspark@agr.shizuoka.ac.jp phone/fax: +81-(0)54-238-4887 382 VoL 91, 2001 EFFECT OF C/N RATIO ON ARACHIDONIC ACID PRODUCTION AND MYCELIAL MORPHOLOGY filamentous, but in the case of the addition of sodium, calcium, and magnesium ions the predominant morphology was of large pellets with diameters of 2-3 m m (21). We have found that natural nitrogen sources such as yeast extract, gluten meal, or corn steep liquor resulted in the formation of circular pellets, but Pharmamedia, fishmeal, or soybean meal resulted in the formation of radial filamentous mycelia from a central pellet core (22). However, despite the strong influence of the consumed C / N ratio on productivity in mycelial culture, little is known about the effect of the consumed C / N ratio on the morphology of fungi that accumulate lipids within their mycelia. Here, we describe the results of an investigation on the morphology and AA productivity of the fungus, M. alpina. The effects of consumed C / N ratio are discussed with regard to morphological changes and A A productivity. The microscopic observations of the morphology are described based on the whole pellet size, the pellet core size, the ratio of the pellet annular region to the whole pellet size, and the width of the pellet annular region using image analysis for the characterization. MATERIALS AND METHODS M. alpina CBS 754.68 Microorganism and media was used throughout this study. For the seed culture, the medium used contained (g/0: glucose, 20; and yeast extract (Oriental Yeast Co. Ltd., Tokyo), 10. For the production culture, the medium used contained (g/0: glucose, 40; soybean oil, 2; soybean meal (SM) (Ajinomoto Co., Tokyo), 10; KH2PO4, 3; Na2SO4, 1; C a C I 2 . 2 H 2 0 , 0.5; and MgCIz.6H20, 0.5. To investigate the effects of consumed carbon to nitrogen (C/N) ratio on the mycelial morphology and production of lipids and AA, the consumed C / N ratio was varied from 5 to 32 with the constraint that the total initial concentration of glucose and the SM in batch culture was kept at 50 g/! as shown in Table I. When the glucose concentration was fixed at 40 g/l, the consumed C / N ratio was varied from 10 to 24. To investigate the effect of glucose, the C / N ratio was varied from 13.4 to 38.6 while also changing the initial SM concentration. These media conditions are listed in Table 1. The pH of the media used was adjusted to 6.0 before sterilization. Culture conditions A spore suspension of M. alpina CBS 754.68 maintained on agar slants at 103spores/ ml was inoculated into a test tube containing 10ml of seed medium. The spores were grown at 28°C on a reciprocal shaker at 120 spm (stroke per minute) for 30 h. For A A production, 5 ml of the seed culture was inoculated into a 500-ml Erlenmeyer flask containing 45 ml of the production medium. The culture was carried out at 28°C on the reciprocal shaker until the glucose was exhausted. All cultures were carried out in triplicate and the results were represented as an average value. In the case of jar fermentor culture, 50 ml of the seed culture was inoculated into a 3-1 jar fermentor (MDL300, Marubishi Co. Ltd., Tokyo) containing 1.5 1 of the production medium. The culture was carried out at 28°C until the glucose was exhausted. To avoid the effect of shear stress on the mycelial morphology the agitation and aeration rates were kept at 350rpm and 1 w m , respectively, throughout the experiments. To maintain the dissolved oxygen concentration at a level higher than 30%, oxygen-enriched air (NewLife, AlrSep Co., Bur- TABLE 1. Experimental conditions for varying the consumed C/N ratio Initial concentration (g//) Glucose Soybean meal 5.0 10.0 15.0 20.0 30.0 35.0 37.5 40.0 42.5 47.5 383 45.0 40.0 35.0 30.0 20.0 15.0 12.5 10.0 7.5 2.5 Residual concentration (g//) Glucose Nitrogen 0.3 0.0 0.1 0.1 0.5 0.9 0.3 5.7 4.2 32.6 0.8 0.5 0.5 0.6 0.3 0.1 0.1 0.0 0.0 0.0 Consumed C/N ratio (-) 4.9 5.6 6.7 8.6 12.0 14.9 18.9 20.4 23.7 31.7 Initial glucose concentration was 40 g/l, but SM concentration was varied from 5 to 40 g/l. 40.0 40.0 40.0 40.0 5.0 20.0 30.0 40.0 18.3 0.0 0.1 0.2 0.0 0.2 0.4 0.6 24.1 14.9 11.7 10.0 Initial SM concentration was 10 g/l, but glucose concentration was varied from 20 to 120 g/I. 20.0 40.0 60.0 80.0 100.0 120.0 10.0 10.0 10.0 10.0 10.0 10.0 0.0 5.7 19.5 31.9 36.9 48.2 0.0 0.0 0.0 0.0 0.0 0.0 13.4 20.4 23.4 27.1 34.4 38.6 falo, NY, USA) was used. Image analysis A binocular microscope (BX-60, Olympus Co., Tokyo) or a stereoscopic microscope (SZH-10, Olympus Co.) equipped with a monochrome CCD camera (XC-77CE, Sony, Tokyo) was used for the image analysis of mycelia. Captured images were fed into a computer (Macintosh 8500/150) and analyzed using image analysis software (IPLab Spectrum, Signal Analytics Co., Virginia, USA). The captured images were thresholded to obtain binary images. Opening was then applied to the binary images for improvement of their quality. Repeated opening cycles were performed on each pellet until only the pellet core remained. The subtraction of the pellet core from the whole pellet gave the annular region. Throughout these experiments, the annular regions were regarded as filamentous mycelia. For each sample, images of at least 50 elements (defined as either pellets or clumps of hyphae) were used for image processing and determination of morphological parameters. Average values were used to analyze the morphological parameters. Equivalent diameters of the projected area of the whole pellet (Din) and pellet core (Dpc) were calculated by assuming a circular morphology for both (22). The width of the pellet annular region (Lf) was also regarded to be the thickness of the filamentous mycelia around the pellet core. The L f was calculated using the whole pellet and pellet core diameters as follows: Lf---- D m - - D p c 2 Analytical methods The fungal mycelia were harvested by filtration and washed with 50 ml of distilled water. The filtered mycelia were dried at 105°C for 2 h 384 KOIKE ET AL. J. BloSci. BIOENO., and weighed to obtain the dry cell weight. Intracellular fatty acids (FA) and A A were extracted from 20 mg of dried mycelia, trans-methylated in methanolic HCI and quantified by gas chromatography using a flame-ionized detector (Shimadzu GC-14B, Shimadzu, Tokyo). A glass column (3 m m × 2 m) was packed with 5% Advance DS on 80/100 mesh Chromosorb W (Shimadzu, Kyoto). The temperatures of the column, injection port, and detection chamber were 190, 240, and 250°C, respectively. The gas pressures of nitrogen, hydrogen, and air were 4, 0.6, and 0.5 kg/cm 2, respectively. The residual glucose concentration was measured using the DNS method (23). The consumed C / N ratio was determined as follows, = a ( G l c o - Glc) + 1~Smo N ~'SMo- N where a and/9 indicate the ratios of carbon contained in the glucose and SM, respectively. The value for a was 0.4 and that for the/~ was 0.303. The ? value indicated the ratio of nitrogen in SM and was 0.0847 as detrermined from analytical data. Glc and N denote the residual concentrations of glucose and nitrogen, respectively. The subscript 0 indicates an initial concentration. The residual nitrogen concentration in the culture medium was measured by estimation of the ammonia formed using Nessler's reagent after micro-Kjeldhal digestion of cell-free medium samples. I II III 25 --O-x i " 15 i,° 0 4 B !, 3 1 o i ~~ 1.0 ., I ~ 0.8 3o 0.6 ~. 0.4 ~ AA/FA RESULTS 0.2 Effect of C / N ratio on arachidonic acid production in flask culture under a fixed initial a m o u n t of glucose and 0.0 SM Under the condition of a fixed initial amount of carbon and nitrogen sources of 50 g/1 over all, the C / N ratio was varied from 5 to 32. Mycelial growth and fatty acid production from M. alpina cells are divided into three phases as shown in Fig. I. In phase I (a consumed C / N ratio of less than 7) the carbon source is limited; in phase II (a consumed C / N ratio of 7 to 15) utilization of the carbon source switches from cell growth to lipid biosynthesis; in phase III (a consumed C / N ratio of 15 to 32) lipid biosynthesis is most active and the nitrogen source is limited. During phase I, the mycelia concentrations (X) and its concentration without fatty acids (X-FA) remained constant (Fig. 1A) and the ( X - F A ) / X ratio was higher than 0.9. The concentrations of F A and A A produced were very low (Fig. 1B), and the A A / F A ratio was only 0.25. Both the F A yield from consumed glucose (YFA/S) and the A A yield from consumed glucose (YAA/S) were also low (Fig. 1D). This indicates that the glucose was utilized only for mycelial growth and not for lipid biosynthesis below a consumed C / N ratio of 7. However, during phase II, the mycelial concentration increased; and the rate of lipid biosynthesis increased rapidly (Fig. 1B). Therefore, YFA/S and YAms increased markedly (Fig. 1D). In this phase, the carbon source was utilized not for cell growth, but for lipid biosynthesis. During phase III the mycelial concentration decreased linearly with the increasing consumed C / N ratio (Fig. IA). The YeA/s and YAA/S were high (Fig. 1D), indicating that mycelial growth was complete and lipid biosynthesis remained at the highest level. The ( X - F A ) / X ratio decreased from 0.8 to 0.6 (Fig. 1A), because of the limitation of the nitrogen source. C : , t 0.12 iI ~ (X-FA)/X I I o.o9 D "~ 0.06 e~ "'~ 0.03 0 ~V', :. , + YaWs t 0 10 20 30 C/N ratio FIG. 1. Effect of consumed C/N ratio on mycelial growth and lipid production in flask culture of M. alpina. The sum of the initial concentrations of glucose and SM was 50 g/1. A shows the mycelia concentrations (X) and mycelial weight except for fatty acids (X-FA); B, concentrations of arachidonic acid (AA) and total fatty acid (FA); C, ratios of (X-FA)/Xand AA/FA; D, AA yield (YAms)and FA yield (YFA/S)from consumed glucose. Cellular growth and Hpidbiosynthesis are classified into phases I, II, and III. Effect o f C / N ratio on A A production in the presence of an excess concentration o f either glucose or SM To investigate the effect of the consumed C / N ratio on AA production a fixed concentration of either glucose or SM, two separate experiments were carried out. In one, the SM concentration was varied from 5 to 40 g/l while the initial glucose concentration was fixed at 40 g/1 in batch culture (a consumed C / N ratio of 10 to 24). In the other, the glucose concentration was varied from 20 to 120g/l, but the initial SM concentration was fixed at 1 0 g / / ( a consumed C / N ratio of 13 to 39). The results VOL 91, 2001 I EFFECT OF C/N RATIO ON ARACHIDONIC ACID PRODUCTION AND MYCELIAL MORPHOLOGY -o- x - - I ~ X.FA II ,-, DlO • 1 5 I • X 30 ~ o ~ ~AA 0 .--.I~FA --o-AA --o- ~-A ~#'-"'~B "~4 ~ l0 ~ , r-.., r ~ o 3 II 1 ' 0 • ' " ' n ~ YaW~ i°o:i 5 10 • 20 [ ] 27 I ,[]ln [ ] 39 60 50 I 0.09 •23 5O - I ~ X.FA ,o ~5 385 - o - YaWs C ~" 0.10 40 o.0, , n , | 15 20 C/N ratio . i 25 0,00 . . . . . . . . . "~ ~" 10 15 20 25 30 35 40 C/N ratio FIG. 2. Effect of consumed C/N ratio on myceliai growth and lipid production in flask culture of M. alpina. The initial glucose concentration was 40 g/I and the SM concentration was varied (I); the initial SM concentration was 10 g/I and the glucose concentration was varied (II). A shows concentrations ofXand X-FA; B, concentrations of AA and FA; C, YA~S. are shown in Fig. 2. W i t h an increase in the SM concentration at a fixed initial glucose concentration (I in Fig. 2), when the consumed C / N ratio was close to 10, the concentration o f mycelia was highest and then decreased with increasing consumed C / N ratio. On the other hand, the FA concentration was independent o f the C / N ratio, but the A A concentration decreased slightly, and YAA/S also decreased. However, with an increase in the glucose concentration at a fixed initial SM concentration o f 10 g/! (II in Fig. 2), mycelial concentrations remained constant despite increases in the consumed C / N ratio. With an increase in the consumed C / N ratio, the A A concentration decreased, but the FA concentration increased linearly. The yield o f A A f r o m consumed glucose decreased linearly with increasing consumed C / N ratio. The fatty acid compositions in these cultures are shown in Fig. 3. In the case o f a fixed initial glucose concentration o f 40 g/l (Fig. 3I), when the consumed C / N ratio increased, i.e. when the nitrogen source was limiting, the p r o p o r t i o n o f saturated fatty acids increased and the p r o p o r t i o n o f the p o l y u n s a t u r a t e d fatty acid, A A ( 2 0 : 4 ) increased. In the case o f a fixed initial SM concentration o f 1 0 g / / (Fig. 3II), when the consumed C / N ratio increased a n d the carbon source was in excess, the p r o p o r t i o n o f the saturated fatty acids increased, and the A A concentration decreased to half the a m o u n t observed for consumed C / N ratios o f 20 and 39. These results indicate that for a fixed a m o u n t o f the c a r b o n source, the A A concentration increased with an increase in the a m o u n t o f the nitrogen source; for a fixed a m o u n t o f the nitrogen source the A A concentration increased with a decrease in the a m o u n t o f the carbon source. Effect o f high-concentration carbon and nitrogen ~" 20 10 0 FIG. 3. Effect of the consumed C/N ratio on fatty acid composition in flask culture ofM. a l p i n a . Data in I and II were obtained from the cultures the results for which are shown in Fig. 2I and 2II, respectively. DGLA denotes dihomo-?-linolenic acid. sources with a constant C / N ratio on A A production in flasks and in a fermentor Under a constant C / N ratio o f 20, cultures were carried out in flasks a n d in a fermentor, using m e d i a enriched 1- to 4-fold in b o t h SM and glucose. The results for the fermentor culture are shown in Fig. 4. Glucose was exhausted at 4 d, 7 d, 9 d, a n d 10d when the 1- to 4-fold enriched m e d i a were used, respectively. In all cultures, the nitrogen source was consumed completely by 6 d or before. Since glucose and SM were consumed completely there was no change in the consumed C / N ratio o f 20. Final mycelial concentrations increased with increases in the degree o f m e d i u m enrichment. The A A concentration depended on the degree o f m e d i u m enrichment up to a 3-fold increase, but decreased in the 4-fold enriched m e d i u m (Fig. 4C). In the case o f 3- and 4-fold enriched media, m a n y mycelial clumps were f o r m e d and attached to the glass walls and impeller o f the fermentor. Moreover, it was impossible to mix the culture b r o t h due to high viscosity in the case o f the 4-fold enriched medium. The yield coefficients, b o t h for the flasks a n d fermentor cultures are shown in Table 2. Cellular yields for glucose (Yx/s) decreased with an increase in the degree o f m e d i u m enrichment. The A A yield for glucose (YAA/S) in the 4-fold enriched m e d i u m also decreased to half to that in the control, while the FA yield for glucose (YFA/S) increased gradually. The ratio o f A A to FA decreased from 0.44 to 0.30 in the f e r m e n t o r culture and was similar to that o f the flask cultures. This result indicates that even though the consumed C / N ratio was in an o p t i m a l range, when the nutrient concentration was high, the mycelia metabolized the c a r b o n source 386 KOIKE ET AL. J. 100 a TABLE =.I • ~.o = 60 7!,020 So ( g / / ) a SM0 (g//)b .... a A 50 AA/FA 30 .2[ 40 10 80 20 120 30 160 40 17.96 (14.21) 27.40(27.63) 37.34(41.20) 43.40(49.85) 3.04 (2.78) 7.20 (6.79) 11.45 (12.97) 9.15 (14.63) 1.35 (1.27) 2.49 (3.17) 3.42 (3.75) 2.75 (3.62) Fatty acid ratio ( - ) 40 ~ Yield coefficients for the same C/N ratio in the flask and fermentor cultures Concentration (gl/) X FA AA •~ 2. BIOSCI. BIOENG., 0.44 (0.45) 0.35 (0.49) 0.30 (0.29) 0.30 (0.25) Yield from glucose ( - ) 21) lO O o i B i i i i t i i i i 3 1 I Yx/s YAA:S YFA/S I 2 I I I 4 6 Time 8 I I 10 1 FIG. 4. Concentrations of residual glucose (A), mycelia (B), and arachidonic acid (C) in fermentor culture of M. alpina. Initial concentrations (g//) of glucose and SM were 40 and 10 (O); 80 and 20 ( • ); 120 and 30 (n); and 160 and 40 (m), respectively. In the cases of concentrations of 120 and 30, and 160 and 40, sampling commenced after 5 d of cultivation, because insoluble SM remained in the culture. The final mycelial weight and AA concentration increased because mycelia growing on the glass wall inside the ferrnentor were included in the calculation. and tended to produce F A , but not A A . In the 4-fold enriched medium, concentrations o f mycelia and lipids were lower than those o f the flask cultures. The difference in yield coefficients between flask and fermentor cultures might be due to shear stresses affecting the mycelial m o r p h o l o g y due to mechanical agitation in the fermentor. Effect of the consumed C / N ratio on mycelial mor. phology in flask culture P h o t o g r a p h s depicting the m o r p h o l o g y o f mycelia cultured with different C / N ratios are shown in Fig. 5. The m o r p h o l o g i c a l changes occurring in relation to the consumed C / N ratio were similar despite the use o f three different culture media. With increasing consumed C / N ratio the whole pellet size increased; smooth or hairy pellets were formed in all cases (Fig. 5A). In the presence o f excess nitrogen, mycelia formed long and narrow pellets (Fig. 5B). On the contrary, in the presence o f excess carbon, mycelia f o r m e d hairy pellets (Fig. 5C). W h e n the carbon source concentration was high but the nitrogen source concentration was low, with the consumed C / N ratio higher than 30, the pellets grew larger than those observed at low C / N ratios, and were hairy. When the consumed C / N ratio was varied, the morphology was characterized and significant features are shown in Fig. 6. The whole pellet size (A in Fig. 6) did 0.45 (0.39) 0.35 (0.39) 0.31 (0.38) 0.27(0.33) 3.37(3.50)×10-2 3.55(3.45)×10 2 2.85(3.46)×10-2 1.72(2.37)×10-2 0.76(0.77)×10-1 1.03(0.71)×10-l 0-95(1.20)X10 I 0.57(0,96)X10-I Data are from the fermentor, but data in parentheses are from the batch culture. The consumed C/N ratio in both cultures was 20. a Initial glucose concentration. b Initial SM concentration. not change until a consumed C / N ratio o f 20 was reached, but then increased gradually with an increase in the consumed C / N ratio. However, the ratio o f the size o f the pellet annular region to the whole pellet size was 0.8 (data not shown), which was independent o f the consumed C / N ratio or the culture media. The Lf remained at 0.3-0.4 on average until a consumed C / N ratio o f 20 was reached, and then increased linearly with increasing C / N ratio (Fig. 6B). Effect of the consumed C / N ratio on mycefial morpholMorphological changes occurring ogy in the fermentor at a consumed C / N ratio o f 20 using the enriched media in the fermentor are shown in Fig. 7. W h e n the initial concentrations o f glucose and SM were 40 and 10 g/l, respectively, the whole pellet sizes were larger than 1 m m 2 (Fig. 7A), which was similar to the size o f those shown in Figs. 5 and 6. However, as the glucose concentration was increased to 120 a n d 160 g/l, the whole pellet size and Lf decreased. This suggests that when the medium was enriched the pellet size was smaller than in the control, but the mycelial concentration was higher (see Fig. 4). The m o r p h o l o g y o f mycelia in the fermentor was similar to that o f mycelia in the flask culture but the size was smaller (data not shown). A l t h o u g h the width o f pellet annular regions was 0.45 m m on average at the beginning o f the culture, the width was dependent on the degree o f m e d i u m enrichment, decreasing linearly. This means that the filaments a r o u n d the pellet core might be shaved-off by shear stress and might form new smaller entities. DISCUSSION The key to lipid accumulation is to allow the a m o u n t o f nitrogen supplied to the culture to become exhausted during cultivation. To investigate the effect o f limiting the carbon or nitrogen source, we varied the consumed C / N ratio under the condition that the sum o f the initial concentrations o f the carbon and nitrogen sources was fixed at 5 0 g / l . As indicated by m a n y researchers, the fatty acid yield increases with increasing consumed C / N ratio when nitrogen is limiting (Fig. 1). To confirm this, the consumed C / N ratio was varied VOL. 91, 2001 EFFECT OF C/N RATIO ON ARACHIDONIC ACID PRODUCTION AND MYCELIAL MORPHOLOGY A B 387 C • Z ,,, ............~........~!ii~¸ FIG. 5. Photographs of mycelia cultured at various C/N ratios in flasks. Numbers in photographs denote the consumed C/N ratios. Photograph A was of mycelia grown in experiments the results of which are shown in Fig. 1; and photographs B and C, of mycelia grown in experiments the results of which are shown in I and II in Fig. 2, respectively. at a fixed initial concentration of the carbon source. In this case the mycelial concentration depended on the nitrogen source concentration. When nitrogen was in excess, the proportion of polyunsaturated fatty acids increased, but that of saturated fatty acids decreased. On the other hand, when the consumed C / N ratio was increased at a fixed initial concentration of the nitrogen source, the fatty acid concentration depended on the initial concentration of the carbon source, but the concentration of A A was independent of this. When carbon was in excess, the proportion of unsaturated fatty acids decreased, but that of saturated fatty acids increased. Increasing the already high C / N ratio led to a decrease in the A A concentration to 50% (Fig. 3). When the C / N ratio is high the rate of the desaturation reaction becomes low, and the accumulated saturated fatty acids are not converted to polyunsaturated fatty acids. From these results, it was clear that the concentrations of carbon and nitrogen sources should be balanced and that a suitable C / N ratio for A A production was around 20. These results are similar to those obtained with other Mortierella sp. (15, 24). The optimum C / N ratio for lipid biosynthesis by Mortierella was 20 to 23. However, in the case of the oleaginous yeasts, Rhodotorula glutinis (13, 25) and Apiotrichum curvature (14, 26), the optimum C / N ratio was higher than 40, a ratio at which the nitrogen is probably suitably limited. Granger et al. (13) reported that the cell growth and lipid production of R. glutinis were independent of the C / N ratio below a consumed C / N ratio of 10. When the consumed C / N ratio was higher than 25 the lipid concentration was dependent on the ratio. However, for C / N ratios higher than 40, lipid synthesis declined. This indicates that the fungus required more nitrogen for lipid biosynthesis than does the yeast. The difference in the optimum C / N ratio between the yeast and fungus indicates the degree of nitrogen limitation. Using the optimum C / N ratio, when the initial concentrations of both carbon and nitrogen sources increased, cellular growth and lipid production were proportional to the increasing concentration until the carbon and nitrogen sources were 120 and 40g/l, respectively. However, at concentrations higher than this the A A concentration decreased despite an increase in the biomass and FA concentration. In the culture containing initial carbon and nitrogen concentrations of 160 and 40g/l, respectively, it was very difficult to mix completely the culture broth because of its high viscosity. Moreover, many mycelia were attached to the fermentor wall and to the impeller and its shaft. This probably resulted in local deficiencies of dissolved oxygen in the culture due to insufficient mixing in the reactor. Unsaturated fatty acids are synthesized by an aerobic mechanism which involves composed NAD(P)H-cytochrome b5 reductase, cytochrome bs, and a terminal cyanide-sensitive desaturase in the cytoplasm (27, 28). Higashiyama et al. (20) reported that the oxygen requirement for desaturation was higher than for cell growth and FA production. They showed that the optimum dissolved oxygen concentration for A A production was 10-15 ppm, but at 3 ppm the FA and X-FA yields did not decrease, but the A A yield was lower than that obtained at a high dissolved oxygen concentration. Therefore, if the viscosity of the culture broth is sufficiently low or the filament are converted to 388 KOIKE ET AL. J. BIOSCI.BIOENG., l0 2 • 0 1.o o.4 N~ 0.2 0.0 0 20 C/N ratio i0 30 40 FIG. 6. Effect of the consumed C/N ratio on morphological parameters of M. alpina. A and B show the whole pellet size (Am)and width of the pellet annular region (Lf), respectively. The morphological parameters pertain to mycelia grown in experiments the results of which are shown in Fig. 1 ( • ); and in Fig. 2I (/x); experiments, in Fig. 2II ( [] ). small pellets, dissolved oxygen might not be limited in the culture and then the A A production would be correlated with the nutrient concentration. The most interesting results are that the whole pellet size and the width o f the pellet annular region did not change until the consumed C / N ratio was less than 20, 2.0 A but then increased gradually with increasing consumed C / N ratio. W h e n the m e d i u m was enriched at a fixed C / N ratio o f 20, the whole pellet size and the width o f the pellet annular region decreased with increasing o f glucose and SM concentration. In the case o f the fermentor, the whole pellet sizes were smaller than those in the flask cultures and decreased over the course o f cultivation. The width o f the pellet annular region remained constant during cultivation. These results indicate that the filaments were b r o k e n by shear stress due to the mechanical agitation. The resulting segmented mycelia m a y have grown to form small mycelia in the presence o f nutrients. H i g a s h i y a m a et al. (29) reported that hairy mycelia o f M. alpina on the pellet surface were shaved off during cultivations, a n d those filaments aggregated or became entangled and increased in size. This result indicates that the m o r p h o l o g y observed in flask culture cannot be observed in the fermentor culture due to differences in the shear conditions and it was thought that the M. alpina mycelium used in this study is very sensitive to shear stress. In this study however, since the agitation rate was fixed at 350 r p m throughout, there m a y not have been a difference in shear stress conditions between the two types o f cultures. The difference in m o r p h o l o g i c a l size between cultures was due to the C / N ratios. In this study, to produce A A f r o m a culture o f M. alpina it was important to adjust the optimal C / N ratio in the m e d i u m to a r o u n d 15-20, by balancing the carbon and nitrogen concentrations. W h e n the consumed C / N ratio was kept constant at 20, the a m o u n t o f A A p r o d u c e d was p r o p o r t i o n a l to the carbon and nitrogen concentrations. This result will be useful for A A production in a fed-batch operation. The mycelial m o r p h o l o g y and the consumed C / N ratio were correlated in this study. The whole pellet size and the width o f the pellet annular region (layer o f filamentous mycelia a r o u n d the pellet core) were independent o f the consumed C / N ratio below 20 in both flask and fermentor cultures, but increased gradually with increasing C / N ratio. o4 ~O . . ~ 1.0 REFERENCES 0.5 1. Yamada, H., Shimizu, S., Shinmen, Y., Kawashima, H., and Akimoto, K.: Biotechnological processes for the production of 0.0 | | | | polyunsaturated fatty acids. J. Dispersion Sci. Technol., 10, 561-579 (1989). , 2. Carlson, S.E., Werkman, S.H., Peeples, J. M., Cooke, R.J., and Tolley, E.A.: Arachidonic acid status correlates with the 0.6 B A ~ ~.o ~'~ e-~ 0.4 first year growth in preterm infants. Proc. Natl. Acad. Sci. USA, 90, 1073-1077 (1993). 3. Gill, I. and Vallvety, R.: Polyunsaturated fatty acid. I. Occurrence, biological activities and applications. Trends Biotechnol., 15, 401-409 (1997). 4. Granger, L.M., Perlot, P., Goma, G., and Pareilleux, A.: o.2 0.0 t 40110 ! | 120/30 ! | 160/40 Glucose to SM concentration ratio (-) FIG. 7. Effect of medium enrichment on morphological changes during fermentor culture of M. alpina. Data for whole pellet sizes (A) and widths of the pellet annular region (B) were obtained after each cultivation. Eificiency of fatty acid synthesis by oleaginous yeasts: prediction of yield and fatty acid cell content from consumed C/N ratio by a simple method. Biotechnol. Bioeng., 42, 1151-1156 (1993). 5. Yongmanitchai, W. and Ward, O.P.: Omega-3 fatty acids: alternative sources of production. Proc. Biochem., 24, 117-125 (1989). 6. Shinmen, Y., Shimizu, S., Akimoto, K., Kawashima, H., and Yamada. H.: Production of AA by Mortierella fungi: selection of a potent producer and optimization of culture conditions for large-scale production. Appl. Microbiol. Biotechnol., 31, 11-16 (1989). 7. Hall, M. J. and Ratledge, C.: Lipid accumulation in oleaginous VOL 91, 2001 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. EFFECT OF C/N RATIO ON ARACHIDONIC ACID PRODUCTION AND MYCELIAL MORPHOLOGY yeast (Candida 107) growing on glucose under various conditions in a one- and two-stage continuous culture. Appl. Environ. Microbiol., 33, 577-584 (1977). Yamancki, H., Mori, H., Kobayashi, T., and Shimizu, S.: Mass production of lipids by Lipomyces starkeyi in microcomputer-aided fed-batch culture. J. Ferment. Technol., 61, 275280 (1983). Yoon, S.H. and Rhee, J . S . : Quantitative physiology of Rhodotoruala glutinis for microbial lipid production. Proc. Biochem., 18, 2-4 (1983). Choi, S.Y., Ryu, D.Y., and Rhee, J.S.: Effects of growth rate and oxygen on lipid synthesis and fatty acid composition of Rhodotorula gracilis. Biotechnol. Bioeng., 14, 1165-1172 (1982). Rolph, C. E., Moreton, R.S., and Harwood, J.L.: Acyl lipid metabolism in the oleaginous yeast Rhodotorula gracilis (CBS 3043). Lipids, 24, 715-720 (1989). Ykema, A., Verbree, E.C., Kater, M.M., and Smit, H.: Optimization of lipid production in the oleaginous yeast Apiotrichum curvatum in whey permeate. Appl. Microbiol. Biotechnol., 29, 211-218 (1988). Granger, L.M., Perlot, P., Goma, G., and Parellleux, A.: Kinetics of growth and fatty acid production of Rhodotorula glutinis. Appl. Microbiol. Biotechnol., 37, 13-17 (1992). Hassan, M., Blanc, P.J., Granger, L.M., Parellleux, A., and Goma, G.: Lipid production by an unsaturated fatty acid auxotroph of the oleaginous yeast Apiotrichum curvature grown in single-stage continuous culture. Appl. Microbiol. Biotechnol., 40, 483-488 (1993). Yokochi, T., Kamisaka, Y., Nakahara, T., Enoshita, L., and Suzuki, O.: Studies on production of lipids in fungi. XXI. Effect of cultural conditions on lipid productivity of Mortierella isabellina with a culture at high cell mass. J. Jpn. Oil Chem. Soc., 38, 241-248 (1989). (in Japanese) Hiruta, O., Futamura, T., Takebe, H., Satoh, A., Kamisaka, Y., Yokochi, T., Nakahara, T., and Suzuki, O.: Optimization and scale-up of y-linolenic acid production by Mortierella ramanniana MM 15-1, a high T-linolenic acid producing mutant. J. Ferment. Bioeng., 82, 366-379 (1996). Paul, G.C. and Thomas, C.R.: Characterisation of mycelial morphology using image analysis. Advances in Biochem. Eng./Biotechnol., 60, 1-59 (1998). Tucker, K.G., Kelley, T., Delgrazia, P., and Thomas, C.R.: Fully-automated measurement of mycelial morphology by image 389 analysis. Biotechnol. Prog., 8, 343-359 (1992). 19. Tamura, S., Park, Y.S., Toriyama, M., and Okabe, M.: Change of mycelial morphology in tyrosin production by batch culture of Streptomyces fradiae under various shear conditions. J. Ferment. Bioeng., 83, 523-528 (1997). 20. Higashiyama, K., Murakami, K., Tsujimura, H., Matsumoto, N., and Fujikawa, S.: Effects of dissolved oxygen on the morphology of an arachidonic acid production by Mortierella alpina IS-4. Biotechnol. Bioeng., 63, 442-448 (1999). 21. Higashiyama, K., Yaguchi, T., Akimoto, K., Fujikawa, S., and Shimizu, S.: Effects of mineral addition on the growth morphology and arachidonic acid production by Mortierella alpina 1S-4. J Am. Oil Chem. Soc., 75, 1815-1819 (1998). 22. Park, E.Y., Koike, Y., Higashiyama, K., Fujikawa, S., and Okahe, M.: Effect of nitrogen source on mycelial morphology and arachidonic acid production in culture of Mortierella alpina. J. Biosci. Bioeng., 88, 61-67 (1999). 23. Mandcls, M., Andreotti, R., and Roche, C.: Measurement of saccharifying cellulase. Biotechnol. Bioeng. Syrup., 6, 21-33 (1976). 24. Yokochi, T., Kamisaka, Y., Nakahara, T., and Suzuki, O.: Production of lipid containing 7-1inolenic acid by continuous culture of Mortierella ramanniana (Studies on production of lipids in fungi. XXIII). J. Jpn. Oil Chem. Soc., 42, 893-898 (1993). 25. Pan, J . G . and Rhee, J.S.: Kinetic and energetic analysis of lipid accumulation in batch culture of Rhodotorula glutinis. J. Ferment. Teclinol., 6, 557-560 (1986). 26. Ykema, A., Bakels, R. H. A., Vcrwoert, II., G.S., and Smit, H.: Growth yield, maintenance requirements and lipid formation in the oleaginous yeast Apiotrichum curvatum. Biotechnol. Bioeng., 34, 1268-1276 (1989). 27. Certik, M. and Shimizu, S.: Biosynthesis and regulation of microbial polyunsaturated fatty acid production. J. Biosci. Bioeng., 87, 1-14 (1999). 28. Ratledge, C.: Microbial lipid: commercial realities or academic curiosities. In Kyle, D. J. and Ratledge, C. (ed.), p. 1-15. Industrial applications of single cell oils. AOCS Press, Champaign, IL (1992). 29. Higashiyama, K., Fujikawa, S., Park, E.Y., and Okabe, M.: Image analysis of morphological change during arachidonic acid production by Mortierella alpina 1S-4. J. Biosci. Bioeng., 87, 489-494 (1999).