- Email: [email protected]
Enhanced production of l -lactic acid by ammonia-tolerant mutant strain Rhizopus sp. MK-96-1196
JOURNAL OF BIOSCIENCE AND BIOENGINEERING Vol. 97, No. 1, 19–23. 2004
Enhanced Production of L-Lactic Acid by Ammonia-Tolerant Mutant Strain Rhizopus sp. MK-96-1196 SHIGENOBU MIURA,1 LIES DWIARTI,2* TOMOHIRO ARIMURA,1 MINAKO HOSHINO,1 LIU TIEJUN,2 AND MITSUYASU OKABE2 Tokyo Laboratories, Musashino Chemical Laboratory Ltd., 16-1 Miyamae 1-Chome, Suginami-ku, Tokyo 168-0081, Japan1and Department of Applied Biological Chemistry, Faculty of Agriculture, Shizuoka University, 836 Ohya, Shizuoka 422-8529, Japan2 Received 1 August 2003/Accepted 6 October 2003
By a monospore isolation technique, Rhizopus sp. MK-96-1 was selected from colonies of Rhizopus sp. MK-96, which was isolated from the soil sample collected in Fujieda, Japan, and used as a parent strain. By the ammonia-concentration-gradient agar plate technique after mutation using N-methylN¢-nitro-N-nitrosoguanidine (NTG) method, a mutant strain designated Rhizopus sp. MK-96-1196 producing more than 90 g/l L-lactic acid under pH control using liquid ammonia in an airlift bioreactor was successfully isolated. Compared with the parent strain, this mutant strain produced about twofold the amount of L-lactic acid in half fermentation time under the same culture conditions. Ammonium L-lactate was recovered and purified as free L-lactic acid via n-butyl L-lactate. The ammonia used for pH control in the fermentation broth was recovered as liquid ammonia during the recovery and purification process and subsequently reused for the next fermentation. Thus, we have developed a new highly purified L-lactic acid production process without producing recalcitrant wastes, e.g., CaSO4 (gypsum). [Key words: Rhizopus sp. MK-96-1196, ammonia concentration-gradient agar plate, esterification, n-butyl lactate]
agar plate technique after N-methyl-N¢-nitro-N-nitrosoguanidine (NTG) mutation (6). Furthermore, we investigated a new process of recovering and purifying L-lactic acid from the culture broth of this mutant strain.
Recently, many researchers have reported on the production of L-lactic acid by fermentation using Rhizopus oryzae (1–3). For example, Yin et al. achieved a production level of more than 80 g/l from corn starch (1). However, they used calcium carbonate as a pH-controlling agent during the fermentation. Because a considerable amount of calcium sulfate (gypsum) is produced during the conversion of calcium lactate to free L-lactic acid, the resulting gypsum might cause considerable environmental problems. Even if lactic acid bacteria such as Lactobacillus brevis are used for lactic acid production, the situation is similar to that using R. oryzae, because calcium carbonate is exclusively used as the pH-controlling agent in the fermentation by lactic acid bacteria (4, 5). In any case, a pH-controlling agent other than calcium carbonate is necessary for large scale L-lactic acid fermentation. In this connection, ammonia seems to be one of the best pH-controlling agent based on the following reasons: (i) ammonia can be recovered from the fermentation broth after harvesting in the recovery and purification of L-lactic acid and subsequently recycled for successive fermentations; (ii) ammonia can also be used as a nitrogen source during fermentation. However, since ammonia is generally toxic to microbes, we attempted to isolate a new ammonia-tolerant mutant strain producing L-lactic acid with a higher yield than conventional strains using a novel screening method combining colorimetric detection and ammonia-concentration-gradient
MATERIALS AND METHODS Microorganism Rhizopus sp. MK-96-1 was selected from colonies of Rhizopus sp. MK-96 isolated from a soil sample collected in Fujieda, Japan, and used as a parent strain. Culture medium The monospore medium contained the following (g/l): soluble starch, 10; Polypepton (Japan Pharmaceutical, Tokyo), 1.0; KH2PO4, 1.0; MgSO4 × 7H2O, 0.25; agar, 2.0; Triton X-100, 0.15; and bromocresol green, 0.2. The seed medium contained the following (g/l): glucose, 120; (NH4)2SO4, 1.35; KH2PO4, 0.3; MgSO4 ×7H2O, 0.25; and ZnSO4 ×7H2O, 0.04. The fermentation medium contained (g/l): corn starch, 120; (NH4)2SO4, 3.0; KH2PO4, 0.25; MgSO4 ×7H2O, 0.15; and ZnSO4 ×7H2O, 0.04. Corn starch was liquefied by adding 0.2% a-amylase (Hs; Nagase Biochem. Ind., Kyoto) at 95°C for 40 min, followed by autoclaving at 121°C for 20 min. Culture method Spores were grown on PDA slants at 30°C for 8 d. They were collected using a platinum loop and suspended in distilled water. A 500-ml flask containing 100 ml of the seed medium was inoculated to give a spore concentration of 106 spores/ml and incubated in a rotary shaker (170 rpm) at 30°C for 18 h. Sterilized calcium carbonate (1 g) was added to each flask. A 500-ml flask containing 100 ml of the fermentation medium was inoculated with 5% of the above inoculum and cultured in the same manner as the seed culture. Sterilized calcium carbonate (1 g) was added to each flask. In the case of fermentation using the 3-l airlift bioreactor, 5% of the inoculum grown for 18 h in a 500-ml
* Corresponding author. e-mail: [email protected] phone/fax: +81-(0)54-238-4883 19
20
MIURA ET AL.
Erlenmeyer shaking flask was transferred into 2 l of the production medium and cultured at 30°C. The pH was automatically controlled at an indicated value using 10% liquid ammonia throughout the fermentation period. In the case of fermentation using the 100-l airlift bioreactor (working volume, 70 l), the production medium was sterilized at 121°C for 20 min by sparging raw steam into the bioreactor. Monospore isolation of high-yield L-lactic acid-producing strain by colorimetric method The colonies showing large halos (yellow) on the monospore agar medium containing bromocresol green as a detection agent were selected, followed by 300ml-flask fermentation. NTG mutation NTG solution was added to the spore solution to give a concentration of 1.8 mg/ml followed by shaking at 30°C for 120 min. Spore viability was nearly 1% under these conditions. The spores were washed three times with 0.1 M citrate buffer (pH 7.0) containing 0.001% Tween 20. The resulting spores were spread on a potato dextrose agar (PDA) slant and incubated at 30°C for 5–6 d. Isolation of ammonia-tolerant and high-yield L-lactic acidproducing strain A high-yield L-lactic acid-producing strain was detected among more than 1000 colonies of Rhizopus sp. MK96 by a colorimetric detection method (described earlier), followed by shaking flask culture (1st screening). After treatment with NTG, the resulting spores were spread on an ammonium-concentrationgradient agar plate. The plate (80 ´230 cm square-type) with a lid was covered with two layers of the agar medium as follows. First, 200 ml of agar medium containing 10 g/l ammonium chloride was poured into the plate such that there was a gradual increase in the medium depth from one side of the plate to the other. After the bottom layer hardened sufficiently in the horizontal position, another 200 ml of the agar medium was added and allowed to harden. This technique provided an agar plate with a gradually increasing ammonium chloride concentration from 0% to 10%. The colonies growing in the area with a comparatively higher ammonium chloride concentration were isolated, followed by the 2nd screening using the shaking flask culture. Finally, the high-yield L-lactic acidproducing strains selected in the 2nd screening were subjected to a final screening using a 3-l airlift bioreactor under pH control using 10% liquid ammonia (3rd screening). Synthesis of n-butyl lactate from ammonium lactate produced by Rhizopus sp. MK-96-1196 The synthesis of n-butyl lactate and its subsequent hydrolysis to L-lactic acid followed Filachione and Catztello’s esterification method using a reaction flask, a Vigreux column and a condenser (Sibata Scientific Technology, Tokyo) (7). Analytical methods Lactic acid concentration was measured by HPLC using a CLC-ODS column (Shimadzu, Kyoto). The eluant, 0.1% H3PO4, was used at a flow rate of 1.0 ml/min, and the detection wavelength was UV 210 nm. Residual starch concentration was measured as residual total sugar using the phenol-sulfuric acid method (8). The optical purity of the lactic acid was measured by HPLC. The analytical conditions were as follows: column, CHIRALPAK MA(+) (Daicel Chem. Ind., Tokyo); eluant, 2 mM CuSO4; flow rate, 0.5 ml/min; temperature, room temperature; detection, UV 254 nm. Bioreactor The schematics of the 3- and 100-l airlift bioreactors and its accessories are shown in a separate paper (9).
RESULTS AND DISCUSSION First screening (selection using agar plate containing bromocresol green) Fifty-five colonies in the monospore medium showing comparatively larger yellow halos as shown in Fig. 1 were selected from among more than 1000 colo-
J. BIOSCI. BIOENG.,
FIG. 1. Screening for a strain producing high amounts of L-lactic acid. The spore solution (0.1 ml) was spread on an agar-plate containing bromocresol green and cultivated at 30°C for 3 d.
FIG. 2. Ammonium-tolerant colonies growing on the ammoniumconcentration-gradient agar plate. The plate was incubated at 30°C for 3 d. Colonies growing in the area with a comparatively higher ammonium chloride concentration (denoted by arrow) were selected, followed by the 3rd screening using a 3-l airlift bioreactor.
nies of Rhizopus sp. MK-96. The selected strains were subjected to 300-ml-flask fermentations and one strain tentatively named Rhizopus sp. MK-96-1 was confirmed to produce more than 60 g/l L-lactic acid under pH control using calcium carbonate. Second screening after mutation followed by ammonium-concentration-gradient agar plate technique The spore solution of the strain (Rhizopus sp. MK-96-1) showing the highest productivity in the above-mentioned flask fermentations was treated with NTG. One milliliter of the resulting spore solution was spread on the ammonium concentration-gradient agar plate. As shown in Fig. 2, there were a few colonies in the high-ammonium-concentration region. One hundred fifty-one colonies appearing in the comparatively high ammonium concentration range were selected and then subjected to the second screening using 300ml Erlenmeyer flasks. Figure 3 shows a histogram of the distribution of L-lactic acid productivity of each ammonium-tolerant mutant strain. Although, as seen from the figure, the productivity distribution shows a relatively normal probability curve in which the center exists between 30–40 g/l, a few strains however showed a high L-lactic acid productivity of more than 60 g/l. Third screening (selection using 3-l airlift bioreactor) Twenty-one strains producing more than 50 g/l L-lactic acid
VOL. 97, 2004
PRODUCTION OF L-LACTIC ACID BY A MUTANT STRAIN
FIG. 3. Distribution of L-lactic acid concentration produced by 151 colonies selected in the second screening. Each cultivation was carried out in a 300-ml Erlenmeyer flask on a rotary shaker at 30°C for 3 d.
in the 2nd screening were used for the third selection using a 3-l airlift bioreactor under pH control using 10% liquid ammonia. As shown in Table 1, six strains have higher specific L-lactic acid productivities under pH control using 10% liquid ammonia than the parent strain under pH control using calcium carbonate. Among these six strains, strain 9-46-2 showing the highest L-lactic acid specific productivity of 0.81 g/l/h/g-cells under pH control using 10% liquid ammonia was selected as the most promising L-lactic acid producer and designated Rhizopus sp. MK-96-1196. Comparison of parent (Rhizopus sp. MK-96) and mutant (Rhizopus sp. MK-96-1196) strains under pH control using 10% liquid ammonia Figure 4 shows the concentrations of L-lactic acid and residual glucose in the culture broth of the parent and mutant strains during fermentation. This figure clearly shows that the glucose con-
FIG. 4. Comparison of the fermentation time courses of the parent strain (Rhizopus sp. MK-96-1) with those of the mutant strain (Rhizopus sp. MK-96-1196). Symbols: open, mutant strain; closed, parent strain. Circles, L-Lactic acid; triangles, glucose. Culture conditions: temperature, 30°C; aeration rate, 1 vvm; pH 6.5 with 10% liquid ammonia.
sumption rate of the parent strain was lower than that of the mutant strain, resulting in decreased L-lactic acid production rate, which could be due to ammonia intolerance. Optimization of culture conditions for enhanced production of L-lactic acid by Rhizopus sp. MK-96-1196 The culture conditions such as inoculum size, initial starch concentration, pH during fermentation and aeration rate under pH control using 10% liquid ammonia was optimized using the 3-l airlift bioreactor (9). Based on this optimization study, a 100-l airlift bioreactor was used to recover and purify lactic acid from the fermentation broth on a pilot scale. Recovery of ammonia and free L-lactic acid from fermentation broth on a pilot scale Selection of suitable alcohol for esterification of ammonium lactate As shown in Table 2, various types of alcohol are usable, including n-butanol, iso-butanol, sec-butanol, n-hexanol, iso-hexanol and octanol. Primary alcohols in gen-
TABLE 1. Screening results in the 3rd screening using the 3-l airlift bioreactor Strain
Timea (h) 37 66 97 66 62 40 49 73 55 30 56 56 69 80 50 46 42 60 53 50 61 49
Yieldb (g/g) 0.68 0.76 0.68 0.73 0.80 0.73 0.66 0.81 0.72 0.67 0.80 0.73 0.70 0.66 0.76 0.77 0.76 0.72 0.71 0.71 0.71 0.76
Parent 5-2 7-6 7-11 7-6 7-11 8-7 8-27 5-2-8 8-7-3 5-20-1 5-21-1 6-21-1 7-6-7 7-6-8 7-11-7 8-4-1 8-7-5 8-27-1 9-46-2 9-46-4 9-46-5 a Time of cultivation. b L-Lactic acid yield based on the consumed glucose.
21
Lactic acid (g/l) 81.6 91.2 81.6 87.6 96.0 87.6 79.2 97.2 86.4 80.4 96.0 87.6 84.0 79.2 91.2 92.4 91.2 86.4 85.2 85.2 85.2 91.2
Biomass (g/l) 3.4 2.5 2.0 3.1 2.5 3.0 3.2 2.1 2.3 5.5 2.8 2.3 2.3 2.5 3.6 2.8 2.7 2.4 3.5 2.1 2.2 3.8
Specific productivity (g/l/h/g-cells) 0.65 0.55 0.43 0.43 0.62 0.73 0.51 0.63 0.68 0.64 0.68 0.68 0.53 0.40 0.51 0.71 0.80 0.60 0.46 0.81 0.63 0.49
22
J. BIOSCI. BIOENG.,
MIURA ET AL.
TABLE 2. Comparison of various alcohols in esterification of lactic acid Alcohol n-Butyl alcohol iso-Butyl alcohol sec-Butyl alcohol n-Hexyl alcohol n-Octyl alcohol iso-Octyl alcohol
Time (h) 2.5 5.0 7.0 5.5 2.4 1.8
Conversion (%) 62 43 12 73 78 65
eral showed a higher conversion to ester than did secondary alcohols. n-Hexanol and octanol having high boiling points showed the highest conversion rate of 73–78%. Among the tested alcohols, although the conversion rate of n-butanol to ester was slightly lower than those of the other alcohols, it easily formed an azeotrope with water. Because this azeotropic composition was easily separated into two phases, namely n-butanol and water, the reflux of n-butanol to the esterification reaction system can be easily accomplished. Thus, n-butanol was selected as the best alcohol for the esterification of ammonium lactate. Effect of molar ratio of n-butanol to ammonium lactate Although increasing the molar ratio of n-butanol to ammonium lactate increased the conversion rate of n-butanol to n-butyl lactate and decreased the reaction time, as seen from Fig. 5, a ratio of 2.5 moles of n-butanol to 1 mole of ammonium lactate appeared to be practical and economical for industrial scale L-lactic acid production. Effect of reaction pressure on esterification of ammonium lactate The reaction pressure for the esterification was investigated. Table 3 shows a comparison of the yield of the by-product, dibutyl ether, under different reaction pressures: normal pressure (760 mmHg) and reduced pressure (250 mmHg). As seen in the table, when the esterification was carried out under normal pressure (760 mmHg), the temperature of the reactor reached 130.1°C, thereby producing 0.56% dibutyl ether (by-product) based on the amount of n-butanol charged at first. In contrast, when the same operation was carried out under reduced pressure, the reaction lasted for 10 h and the final temperature in the reactor was 101.1°C. The amount of dibutyl ether produced was only 0.02%.
FIG. 5. Effect of molar ratio of n-butanol to ammonium lactate on conversion to n-butyl lactate. Reaction mixture: 0.95 mole of ammonium lactate produced by fermentation and 2.5 moles of the corresponding alcohol. The initial temperature was 120°C.
TABLE 3. Comparison of pressure in esterification Pressure (mmHg)
Temperaturea (°C)
Reaction time (h) 3 10
Conversion (%)
Dibutyl ether (%) 0.56 0.02
760 131.1 97.4 250 101.1 98.5 a Final. Reaction mixture and initial temparature are the same as in Table 2.
Based on the data reported by Filachione and Costello (7) and our supplemental experimental results described above, a new process for the recovery and purification of free L-lactic acid is proposed in Fig. 6. Recovery and purification of L-lactic acid from fermentation broth on a pilot scale experiment To confirm the recovery yield and purity of the product (L-lactic acid) on a pilot scale, a 100-l airlift bioreactor study was carried out under the optimum conditions described in a separate paper (9). Figure 7 shows the time course of the fermentation. The resulting ammonium lactate produced during the course of the fermentation was 91 g/l in the fermentation broth (65 l) followed by recovery and purification using the proposed process, as shown in Fig. 6. Hydrolysis The resulting n-butyl lactate was hydrolyzed to free L-lactic acid and n-butanol in a hydrolyzing
FIG. 6. A proposed recovery and purification process from ammonium lactate to n-butyl L-lactate. 1, Airlift bioreactor; 2, filter; 3, evaporator; 4, ammonia-removing can; 5, esterification-promoting can; 6, distillation column; 7, distillation column; 8, two-liquid-phase separation can; 9, two-liquid-phase separation can; 10, condenser; 11, condenser; 12, flash vaporizing component; 13, distillation can; 14, condenser; 15, receiver; 16, hydrolysis can; 17, distillation column; 18, two-liquid-phase separation can; 19, condenser.
VOL. 97, 2004
PRODUCTION OF L-LACTIC ACID BY A MUTANT STRAIN
23
FIG. 7. Fermentation time course in the 100-l airlift bioreactor under optimum conditions. Symbols: circles, L-lactic acid; triangles, biomass; rhombuses, glucose.
FIG. 9. HPLC chromatogram of final products in L-lactic acid fermentation by Rhizopus sp. MK-96-1196.
and esterification of L-lactic acid with n-butanol, followed by the hydrolysis of the resulting n-butyl lactate to free L-lactic acid and n-butanol. Compared with the conventional calcium lactate process, this method enables the production of highly purified free L-lactic acid without producing recalcitrant wastes, e.g., CaSO4 (gypsum). REFERENCES
FIG. 8. Mass balance of raw materials and product in the L-lactic acid production process.
can (Fig. 6-16) using a strongly acidic cation-exchange resin as an acid catalyst. Figure 8 shows the mass balance of the fermentation, recovery and purification processes. As seen from the figure, 65 moles of ammonium L-lactic acid was produced from 42 moles of glucose and 110 moles of ammonia during the fermentation. With the addition of 192 moles of n-butanol and a small amount of sulfuric acid, 63 moles of purified n-butyl lactate was produced from 65 moles of ammonium L-lactic acid. The resulting ammonium L-lactic acid was hydrolyzed to 63 moles of free L-lactic acid, the optical purity of which was 99.7%, as shown in Fig. 9. The L-lactic acid finally obtained showed no color and emitted no odor and manifested highly satisfactory thermal stability. Almost all of the ammonia used for pH control in the fermentation was recovered mainly in the evaporator (Fig. 6-3) and the ammonium-removing can (Fig. 6-4) and reused as a pH control agent in subsequent fermentations. n-Butanol was also recovered in the two liquid separators (Fig. 6-8, 6-9, 6-18) and reused in subsequent esterifications. In conclusion, we have developed a new highly purified L-lactic acid production process consisting of fermentation by the ammonia-tolerant mutant, Rhizopus sp. MK-96-1196
1. Yin, P. M., Nishina, N., Kosakai, Y., Yahiro, K., Park, Y. S., and Okabe, M.: Enhanced production of L(+)-lactic acid from corn starch in a culture of Rhizopus oryzae using air-lift bioreactor. J. Ferment. Bioeng., 84, 249–253 (1997). 2. Kosakai, Y., Park, Y. S., and Okabe, M.: Enhancement of L(+)-lactic acid production using mycelia flocks of Rhizopus oryzae. Biotechnol. Bioeng., 55, 461–470 (1997). 3. Yin, P. M., Yahiro, K., Ishigaki, T., Park, Y. S., and Okabe, M.: L(+)-Lactic acid production by repeated batch culture of Rhizopus oryzae in airlift bioreactor. J. Ferment. Bioeng., 85, 96–100 (1998). 4. Yen, Y. H. and Cherie, M.: Separation of lactic acid from whey permeates fermentation broth by electro dialysis. Trans. I. Chem. E., 69, 200–205 (1991). 5. Vick Roy, T., Blanch, H. W., and Wilke, C. R.: Lactic acid production by Lactobacillus delbrueckii in a hollow fiber fermentor. Biotechnol. Lett., 4, 483–488 (1982). 6. Yahiro, K., Takahama, T., Park, Y. S., and Okabe, M.: Breeding of Aspergillus terreus mutant TN-484 for itaconic acid production with high yield. J. Ferment. Bioeng., 79, 506– 508 (1995). 7. Filachione, E. M. and Costello, E. J.: Lactic esters by reaction of ammonium lactate with alcohols. Ind. Eng. Chem., 44, 2189–2191 (1952). 8. Dubois, M., Gilles, K. A., Hamilton, J. K., Rober, P. A., and Smith, F.: Calorimetric method for determination of sugars and related substances. Anal. Chem., 28, 350–356 (1956). 9. Miura, S., Arimura, T., Hoshino, M., Kojima, M., Lies, D., and Okabe, M.: Optimization and scale-up of L-lactic acid fermentation by mutant strain Rhizopus sp. MK-96-1196 in airlift bioreactors. J. Biosci. Bioeng., 96, 65–69 (2003).
Enhanced Production of L-Lactic Acid by Ammonia-Tolerant Mutant Strain Rhizopus sp. MK-96-1196 SHIGENOBU MIURA,1 LIES DWIARTI,2* TOMOHIRO ARIMURA,1 MINAKO HOSHINO,1 LIU TIEJUN,2 AND MITSUYASU OKABE2 Tokyo Laboratories, Musashino Chemical Laboratory Ltd., 16-1 Miyamae 1-Chome, Suginami-ku, Tokyo 168-0081, Japan1and Department of Applied Biological Chemistry, Faculty of Agriculture, Shizuoka University, 836 Ohya, Shizuoka 422-8529, Japan2 Received 1 August 2003/Accepted 6 October 2003
By a monospore isolation technique, Rhizopus sp. MK-96-1 was selected from colonies of Rhizopus sp. MK-96, which was isolated from the soil sample collected in Fujieda, Japan, and used as a parent strain. By the ammonia-concentration-gradient agar plate technique after mutation using N-methylN¢-nitro-N-nitrosoguanidine (NTG) method, a mutant strain designated Rhizopus sp. MK-96-1196 producing more than 90 g/l L-lactic acid under pH control using liquid ammonia in an airlift bioreactor was successfully isolated. Compared with the parent strain, this mutant strain produced about twofold the amount of L-lactic acid in half fermentation time under the same culture conditions. Ammonium L-lactate was recovered and purified as free L-lactic acid via n-butyl L-lactate. The ammonia used for pH control in the fermentation broth was recovered as liquid ammonia during the recovery and purification process and subsequently reused for the next fermentation. Thus, we have developed a new highly purified L-lactic acid production process without producing recalcitrant wastes, e.g., CaSO4 (gypsum). [Key words: Rhizopus sp. MK-96-1196, ammonia concentration-gradient agar plate, esterification, n-butyl lactate]
agar plate technique after N-methyl-N¢-nitro-N-nitrosoguanidine (NTG) mutation (6). Furthermore, we investigated a new process of recovering and purifying L-lactic acid from the culture broth of this mutant strain.
Recently, many researchers have reported on the production of L-lactic acid by fermentation using Rhizopus oryzae (1–3). For example, Yin et al. achieved a production level of more than 80 g/l from corn starch (1). However, they used calcium carbonate as a pH-controlling agent during the fermentation. Because a considerable amount of calcium sulfate (gypsum) is produced during the conversion of calcium lactate to free L-lactic acid, the resulting gypsum might cause considerable environmental problems. Even if lactic acid bacteria such as Lactobacillus brevis are used for lactic acid production, the situation is similar to that using R. oryzae, because calcium carbonate is exclusively used as the pH-controlling agent in the fermentation by lactic acid bacteria (4, 5). In any case, a pH-controlling agent other than calcium carbonate is necessary for large scale L-lactic acid fermentation. In this connection, ammonia seems to be one of the best pH-controlling agent based on the following reasons: (i) ammonia can be recovered from the fermentation broth after harvesting in the recovery and purification of L-lactic acid and subsequently recycled for successive fermentations; (ii) ammonia can also be used as a nitrogen source during fermentation. However, since ammonia is generally toxic to microbes, we attempted to isolate a new ammonia-tolerant mutant strain producing L-lactic acid with a higher yield than conventional strains using a novel screening method combining colorimetric detection and ammonia-concentration-gradient
MATERIALS AND METHODS Microorganism Rhizopus sp. MK-96-1 was selected from colonies of Rhizopus sp. MK-96 isolated from a soil sample collected in Fujieda, Japan, and used as a parent strain. Culture medium The monospore medium contained the following (g/l): soluble starch, 10; Polypepton (Japan Pharmaceutical, Tokyo), 1.0; KH2PO4, 1.0; MgSO4 × 7H2O, 0.25; agar, 2.0; Triton X-100, 0.15; and bromocresol green, 0.2. The seed medium contained the following (g/l): glucose, 120; (NH4)2SO4, 1.35; KH2PO4, 0.3; MgSO4 ×7H2O, 0.25; and ZnSO4 ×7H2O, 0.04. The fermentation medium contained (g/l): corn starch, 120; (NH4)2SO4, 3.0; KH2PO4, 0.25; MgSO4 ×7H2O, 0.15; and ZnSO4 ×7H2O, 0.04. Corn starch was liquefied by adding 0.2% a-amylase (Hs; Nagase Biochem. Ind., Kyoto) at 95°C for 40 min, followed by autoclaving at 121°C for 20 min. Culture method Spores were grown on PDA slants at 30°C for 8 d. They were collected using a platinum loop and suspended in distilled water. A 500-ml flask containing 100 ml of the seed medium was inoculated to give a spore concentration of 106 spores/ml and incubated in a rotary shaker (170 rpm) at 30°C for 18 h. Sterilized calcium carbonate (1 g) was added to each flask. A 500-ml flask containing 100 ml of the fermentation medium was inoculated with 5% of the above inoculum and cultured in the same manner as the seed culture. Sterilized calcium carbonate (1 g) was added to each flask. In the case of fermentation using the 3-l airlift bioreactor, 5% of the inoculum grown for 18 h in a 500-ml
* Corresponding author. e-mail: [email protected] phone/fax: +81-(0)54-238-4883 19
20
MIURA ET AL.
Erlenmeyer shaking flask was transferred into 2 l of the production medium and cultured at 30°C. The pH was automatically controlled at an indicated value using 10% liquid ammonia throughout the fermentation period. In the case of fermentation using the 100-l airlift bioreactor (working volume, 70 l), the production medium was sterilized at 121°C for 20 min by sparging raw steam into the bioreactor. Monospore isolation of high-yield L-lactic acid-producing strain by colorimetric method The colonies showing large halos (yellow) on the monospore agar medium containing bromocresol green as a detection agent were selected, followed by 300ml-flask fermentation. NTG mutation NTG solution was added to the spore solution to give a concentration of 1.8 mg/ml followed by shaking at 30°C for 120 min. Spore viability was nearly 1% under these conditions. The spores were washed three times with 0.1 M citrate buffer (pH 7.0) containing 0.001% Tween 20. The resulting spores were spread on a potato dextrose agar (PDA) slant and incubated at 30°C for 5–6 d. Isolation of ammonia-tolerant and high-yield L-lactic acidproducing strain A high-yield L-lactic acid-producing strain was detected among more than 1000 colonies of Rhizopus sp. MK96 by a colorimetric detection method (described earlier), followed by shaking flask culture (1st screening). After treatment with NTG, the resulting spores were spread on an ammonium-concentrationgradient agar plate. The plate (80 ´230 cm square-type) with a lid was covered with two layers of the agar medium as follows. First, 200 ml of agar medium containing 10 g/l ammonium chloride was poured into the plate such that there was a gradual increase in the medium depth from one side of the plate to the other. After the bottom layer hardened sufficiently in the horizontal position, another 200 ml of the agar medium was added and allowed to harden. This technique provided an agar plate with a gradually increasing ammonium chloride concentration from 0% to 10%. The colonies growing in the area with a comparatively higher ammonium chloride concentration were isolated, followed by the 2nd screening using the shaking flask culture. Finally, the high-yield L-lactic acidproducing strains selected in the 2nd screening were subjected to a final screening using a 3-l airlift bioreactor under pH control using 10% liquid ammonia (3rd screening). Synthesis of n-butyl lactate from ammonium lactate produced by Rhizopus sp. MK-96-1196 The synthesis of n-butyl lactate and its subsequent hydrolysis to L-lactic acid followed Filachione and Catztello’s esterification method using a reaction flask, a Vigreux column and a condenser (Sibata Scientific Technology, Tokyo) (7). Analytical methods Lactic acid concentration was measured by HPLC using a CLC-ODS column (Shimadzu, Kyoto). The eluant, 0.1% H3PO4, was used at a flow rate of 1.0 ml/min, and the detection wavelength was UV 210 nm. Residual starch concentration was measured as residual total sugar using the phenol-sulfuric acid method (8). The optical purity of the lactic acid was measured by HPLC. The analytical conditions were as follows: column, CHIRALPAK MA(+) (Daicel Chem. Ind., Tokyo); eluant, 2 mM CuSO4; flow rate, 0.5 ml/min; temperature, room temperature; detection, UV 254 nm. Bioreactor The schematics of the 3- and 100-l airlift bioreactors and its accessories are shown in a separate paper (9).
RESULTS AND DISCUSSION First screening (selection using agar plate containing bromocresol green) Fifty-five colonies in the monospore medium showing comparatively larger yellow halos as shown in Fig. 1 were selected from among more than 1000 colo-
J. BIOSCI. BIOENG.,
FIG. 1. Screening for a strain producing high amounts of L-lactic acid. The spore solution (0.1 ml) was spread on an agar-plate containing bromocresol green and cultivated at 30°C for 3 d.
FIG. 2. Ammonium-tolerant colonies growing on the ammoniumconcentration-gradient agar plate. The plate was incubated at 30°C for 3 d. Colonies growing in the area with a comparatively higher ammonium chloride concentration (denoted by arrow) were selected, followed by the 3rd screening using a 3-l airlift bioreactor.
nies of Rhizopus sp. MK-96. The selected strains were subjected to 300-ml-flask fermentations and one strain tentatively named Rhizopus sp. MK-96-1 was confirmed to produce more than 60 g/l L-lactic acid under pH control using calcium carbonate. Second screening after mutation followed by ammonium-concentration-gradient agar plate technique The spore solution of the strain (Rhizopus sp. MK-96-1) showing the highest productivity in the above-mentioned flask fermentations was treated with NTG. One milliliter of the resulting spore solution was spread on the ammonium concentration-gradient agar plate. As shown in Fig. 2, there were a few colonies in the high-ammonium-concentration region. One hundred fifty-one colonies appearing in the comparatively high ammonium concentration range were selected and then subjected to the second screening using 300ml Erlenmeyer flasks. Figure 3 shows a histogram of the distribution of L-lactic acid productivity of each ammonium-tolerant mutant strain. Although, as seen from the figure, the productivity distribution shows a relatively normal probability curve in which the center exists between 30–40 g/l, a few strains however showed a high L-lactic acid productivity of more than 60 g/l. Third screening (selection using 3-l airlift bioreactor) Twenty-one strains producing more than 50 g/l L-lactic acid
VOL. 97, 2004
PRODUCTION OF L-LACTIC ACID BY A MUTANT STRAIN
FIG. 3. Distribution of L-lactic acid concentration produced by 151 colonies selected in the second screening. Each cultivation was carried out in a 300-ml Erlenmeyer flask on a rotary shaker at 30°C for 3 d.
in the 2nd screening were used for the third selection using a 3-l airlift bioreactor under pH control using 10% liquid ammonia. As shown in Table 1, six strains have higher specific L-lactic acid productivities under pH control using 10% liquid ammonia than the parent strain under pH control using calcium carbonate. Among these six strains, strain 9-46-2 showing the highest L-lactic acid specific productivity of 0.81 g/l/h/g-cells under pH control using 10% liquid ammonia was selected as the most promising L-lactic acid producer and designated Rhizopus sp. MK-96-1196. Comparison of parent (Rhizopus sp. MK-96) and mutant (Rhizopus sp. MK-96-1196) strains under pH control using 10% liquid ammonia Figure 4 shows the concentrations of L-lactic acid and residual glucose in the culture broth of the parent and mutant strains during fermentation. This figure clearly shows that the glucose con-
FIG. 4. Comparison of the fermentation time courses of the parent strain (Rhizopus sp. MK-96-1) with those of the mutant strain (Rhizopus sp. MK-96-1196). Symbols: open, mutant strain; closed, parent strain. Circles, L-Lactic acid; triangles, glucose. Culture conditions: temperature, 30°C; aeration rate, 1 vvm; pH 6.5 with 10% liquid ammonia.
sumption rate of the parent strain was lower than that of the mutant strain, resulting in decreased L-lactic acid production rate, which could be due to ammonia intolerance. Optimization of culture conditions for enhanced production of L-lactic acid by Rhizopus sp. MK-96-1196 The culture conditions such as inoculum size, initial starch concentration, pH during fermentation and aeration rate under pH control using 10% liquid ammonia was optimized using the 3-l airlift bioreactor (9). Based on this optimization study, a 100-l airlift bioreactor was used to recover and purify lactic acid from the fermentation broth on a pilot scale. Recovery of ammonia and free L-lactic acid from fermentation broth on a pilot scale Selection of suitable alcohol for esterification of ammonium lactate As shown in Table 2, various types of alcohol are usable, including n-butanol, iso-butanol, sec-butanol, n-hexanol, iso-hexanol and octanol. Primary alcohols in gen-
TABLE 1. Screening results in the 3rd screening using the 3-l airlift bioreactor Strain
Timea (h) 37 66 97 66 62 40 49 73 55 30 56 56 69 80 50 46 42 60 53 50 61 49
Yieldb (g/g) 0.68 0.76 0.68 0.73 0.80 0.73 0.66 0.81 0.72 0.67 0.80 0.73 0.70 0.66 0.76 0.77 0.76 0.72 0.71 0.71 0.71 0.76
Parent 5-2 7-6 7-11 7-6 7-11 8-7 8-27 5-2-8 8-7-3 5-20-1 5-21-1 6-21-1 7-6-7 7-6-8 7-11-7 8-4-1 8-7-5 8-27-1 9-46-2 9-46-4 9-46-5 a Time of cultivation. b L-Lactic acid yield based on the consumed glucose.
21
Lactic acid (g/l) 81.6 91.2 81.6 87.6 96.0 87.6 79.2 97.2 86.4 80.4 96.0 87.6 84.0 79.2 91.2 92.4 91.2 86.4 85.2 85.2 85.2 91.2
Biomass (g/l) 3.4 2.5 2.0 3.1 2.5 3.0 3.2 2.1 2.3 5.5 2.8 2.3 2.3 2.5 3.6 2.8 2.7 2.4 3.5 2.1 2.2 3.8
Specific productivity (g/l/h/g-cells) 0.65 0.55 0.43 0.43 0.62 0.73 0.51 0.63 0.68 0.64 0.68 0.68 0.53 0.40 0.51 0.71 0.80 0.60 0.46 0.81 0.63 0.49
22
J. BIOSCI. BIOENG.,
MIURA ET AL.
TABLE 2. Comparison of various alcohols in esterification of lactic acid Alcohol n-Butyl alcohol iso-Butyl alcohol sec-Butyl alcohol n-Hexyl alcohol n-Octyl alcohol iso-Octyl alcohol
Time (h) 2.5 5.0 7.0 5.5 2.4 1.8
Conversion (%) 62 43 12 73 78 65
eral showed a higher conversion to ester than did secondary alcohols. n-Hexanol and octanol having high boiling points showed the highest conversion rate of 73–78%. Among the tested alcohols, although the conversion rate of n-butanol to ester was slightly lower than those of the other alcohols, it easily formed an azeotrope with water. Because this azeotropic composition was easily separated into two phases, namely n-butanol and water, the reflux of n-butanol to the esterification reaction system can be easily accomplished. Thus, n-butanol was selected as the best alcohol for the esterification of ammonium lactate. Effect of molar ratio of n-butanol to ammonium lactate Although increasing the molar ratio of n-butanol to ammonium lactate increased the conversion rate of n-butanol to n-butyl lactate and decreased the reaction time, as seen from Fig. 5, a ratio of 2.5 moles of n-butanol to 1 mole of ammonium lactate appeared to be practical and economical for industrial scale L-lactic acid production. Effect of reaction pressure on esterification of ammonium lactate The reaction pressure for the esterification was investigated. Table 3 shows a comparison of the yield of the by-product, dibutyl ether, under different reaction pressures: normal pressure (760 mmHg) and reduced pressure (250 mmHg). As seen in the table, when the esterification was carried out under normal pressure (760 mmHg), the temperature of the reactor reached 130.1°C, thereby producing 0.56% dibutyl ether (by-product) based on the amount of n-butanol charged at first. In contrast, when the same operation was carried out under reduced pressure, the reaction lasted for 10 h and the final temperature in the reactor was 101.1°C. The amount of dibutyl ether produced was only 0.02%.
FIG. 5. Effect of molar ratio of n-butanol to ammonium lactate on conversion to n-butyl lactate. Reaction mixture: 0.95 mole of ammonium lactate produced by fermentation and 2.5 moles of the corresponding alcohol. The initial temperature was 120°C.
TABLE 3. Comparison of pressure in esterification Pressure (mmHg)
Temperaturea (°C)
Reaction time (h) 3 10
Conversion (%)
Dibutyl ether (%) 0.56 0.02
760 131.1 97.4 250 101.1 98.5 a Final. Reaction mixture and initial temparature are the same as in Table 2.
Based on the data reported by Filachione and Costello (7) and our supplemental experimental results described above, a new process for the recovery and purification of free L-lactic acid is proposed in Fig. 6. Recovery and purification of L-lactic acid from fermentation broth on a pilot scale experiment To confirm the recovery yield and purity of the product (L-lactic acid) on a pilot scale, a 100-l airlift bioreactor study was carried out under the optimum conditions described in a separate paper (9). Figure 7 shows the time course of the fermentation. The resulting ammonium lactate produced during the course of the fermentation was 91 g/l in the fermentation broth (65 l) followed by recovery and purification using the proposed process, as shown in Fig. 6. Hydrolysis The resulting n-butyl lactate was hydrolyzed to free L-lactic acid and n-butanol in a hydrolyzing
FIG. 6. A proposed recovery and purification process from ammonium lactate to n-butyl L-lactate. 1, Airlift bioreactor; 2, filter; 3, evaporator; 4, ammonia-removing can; 5, esterification-promoting can; 6, distillation column; 7, distillation column; 8, two-liquid-phase separation can; 9, two-liquid-phase separation can; 10, condenser; 11, condenser; 12, flash vaporizing component; 13, distillation can; 14, condenser; 15, receiver; 16, hydrolysis can; 17, distillation column; 18, two-liquid-phase separation can; 19, condenser.
VOL. 97, 2004
PRODUCTION OF L-LACTIC ACID BY A MUTANT STRAIN
23
FIG. 7. Fermentation time course in the 100-l airlift bioreactor under optimum conditions. Symbols: circles, L-lactic acid; triangles, biomass; rhombuses, glucose.
FIG. 9. HPLC chromatogram of final products in L-lactic acid fermentation by Rhizopus sp. MK-96-1196.
and esterification of L-lactic acid with n-butanol, followed by the hydrolysis of the resulting n-butyl lactate to free L-lactic acid and n-butanol. Compared with the conventional calcium lactate process, this method enables the production of highly purified free L-lactic acid without producing recalcitrant wastes, e.g., CaSO4 (gypsum). REFERENCES
FIG. 8. Mass balance of raw materials and product in the L-lactic acid production process.
can (Fig. 6-16) using a strongly acidic cation-exchange resin as an acid catalyst. Figure 8 shows the mass balance of the fermentation, recovery and purification processes. As seen from the figure, 65 moles of ammonium L-lactic acid was produced from 42 moles of glucose and 110 moles of ammonia during the fermentation. With the addition of 192 moles of n-butanol and a small amount of sulfuric acid, 63 moles of purified n-butyl lactate was produced from 65 moles of ammonium L-lactic acid. The resulting ammonium L-lactic acid was hydrolyzed to 63 moles of free L-lactic acid, the optical purity of which was 99.7%, as shown in Fig. 9. The L-lactic acid finally obtained showed no color and emitted no odor and manifested highly satisfactory thermal stability. Almost all of the ammonia used for pH control in the fermentation was recovered mainly in the evaporator (Fig. 6-3) and the ammonium-removing can (Fig. 6-4) and reused as a pH control agent in subsequent fermentations. n-Butanol was also recovered in the two liquid separators (Fig. 6-8, 6-9, 6-18) and reused in subsequent esterifications. In conclusion, we have developed a new highly purified L-lactic acid production process consisting of fermentation by the ammonia-tolerant mutant, Rhizopus sp. MK-96-1196
1. Yin, P. M., Nishina, N., Kosakai, Y., Yahiro, K., Park, Y. S., and Okabe, M.: Enhanced production of L(+)-lactic acid from corn starch in a culture of Rhizopus oryzae using air-lift bioreactor. J. Ferment. Bioeng., 84, 249–253 (1997). 2. Kosakai, Y., Park, Y. S., and Okabe, M.: Enhancement of L(+)-lactic acid production using mycelia flocks of Rhizopus oryzae. Biotechnol. Bioeng., 55, 461–470 (1997). 3. Yin, P. M., Yahiro, K., Ishigaki, T., Park, Y. S., and Okabe, M.: L(+)-Lactic acid production by repeated batch culture of Rhizopus oryzae in airlift bioreactor. J. Ferment. Bioeng., 85, 96–100 (1998). 4. Yen, Y. H. and Cherie, M.: Separation of lactic acid from whey permeates fermentation broth by electro dialysis. Trans. I. Chem. E., 69, 200–205 (1991). 5. Vick Roy, T., Blanch, H. W., and Wilke, C. R.: Lactic acid production by Lactobacillus delbrueckii in a hollow fiber fermentor. Biotechnol. Lett., 4, 483–488 (1982). 6. Yahiro, K., Takahama, T., Park, Y. S., and Okabe, M.: Breeding of Aspergillus terreus mutant TN-484 for itaconic acid production with high yield. J. Ferment. Bioeng., 79, 506– 508 (1995). 7. Filachione, E. M. and Costello, E. J.: Lactic esters by reaction of ammonium lactate with alcohols. Ind. Eng. Chem., 44, 2189–2191 (1952). 8. Dubois, M., Gilles, K. A., Hamilton, J. K., Rober, P. A., and Smith, F.: Calorimetric method for determination of sugars and related substances. Anal. Chem., 28, 350–356 (1956). 9. Miura, S., Arimura, T., Hoshino, M., Kojima, M., Lies, D., and Okabe, M.: Optimization and scale-up of L-lactic acid fermentation by mutant strain Rhizopus sp. MK-96-1196 in airlift bioreactors. J. Biosci. Bioeng., 96, 65–69 (2003).