GB2147579A - Production of L-amino acids from alpha-keto acids by fermentation - Google Patents

Abstract

L-amino acids are prepared by contacting a logarithmically growing culture of suitable microorganisms e.g. a brevibacterium, corynebacterium or E. Coli with the corresponding alpha-keto acids. The conversion of alpha-keto acid to L-amino acid takes place via a transamination reaction driven by a coupled enzyme system which is provided by the growing culture.

Description

SPECIFICATION Production of L-amino acids from alpha-keto acids This invention relates generally to the biological production of L-amino acids. More specifically, it has been found that an L-amino acid may be prepared from its corresponding alpha-keto acid precursor by feeding the aipha-keto acid to an actively, logarithmically growing culture of a suitable microorganism in a nutrient medium. There are numerous microorganisms capable of the biotransformation of alpha-keto acids to L-amino acids.

It is known that alpha-keto acids may be converted enzymatically into their corresponding Lamino acids. For example, U.S. 2,749,279 discloses a process wherein alpha-ketoglutaric acid and ammonia are treated with a biological enzyme-coenzyme catalyst. Suitable biological catalysts may be obtained from animal tissue, from rapidly growing plants or from bacteria in the form of autolysates or extracts of the cell culture. Similarly, U.S. 4,304,858 discloses the conversion of water soluble alpha-ketocarboxylic acids to the corresponding amino acids in the presence of a substrate specific dehydrogenase, ammonium ions, and nicotinamide-adeninedinucleotide (NAD + /NADH). Transformation of aipha-keto acids to amino acids using restingcell suspensions has been reported by Yamada et al., The Production of Amino Acids, pp.

440-46 (1972).

SUMMARY OF THE INVENTION It now has been found that biotransformation of an alpha-keto acid to the corresponding Lamino acid can be mediated more efficiently by an actively growing culture of microorganisms.

Microorganisms capable of transforming alpha-keto acids into L-amino acids are grown in a nutrient medium. The alpha-keto acid precursor of the desired L-amino acid is fed to the actively, logarithmically growing culture. The transamination reaction is driven substantially to completion by a coupled enzymatic system provided by the microorganisms of the growing culture, allowing for continuous recycling of the amino donor (L-glutamate) and the coenzyme (NADH or NADPH). The L-amino acid may be recovered from the culture broth.

It is a primary object of this invention to provide a highly efficient process for the biological transamination of alpha-keto acids to their corresponding L-amino acids.

Moreover, it is intended that the process of this invention be useful with a broad range of microorganisms and for production of a broad range of L-amino acids.

It is an additional object to provide a process in which a single strain of microorganisms can produce sufficient quantities of all required enzymes to convert a given alpha-keto acid to Lamino acid.

A further object is to provide for a high level of alpha-keto acid conversion to L-amino acid over a short reaction period.

A still further object is to provide a method which produces a high yield of L-amino acids in the presence of a low concentration of the amino donor.

DESCRIPTION OF THE INVENTION Bacterial growth generally may be divided into three sequential phases. The lag growth phase is a period in which there is littie or no increase in the number of microorganisms in the culture.

In the logarithmic or exponential growth phase, the number of cells increases exponentially over time. In the stationary phase, there is iittle or no cellular growth or division.

It has been found that the microbial transamination of an alpha-keto acid to L-amino acid can be conducted with increased efficiency by adding the alpha-keto acid to a logarithmically growing culture of suitable microorganisms. Fed-batch fermentation according to the method of this invention results in high yields at high conversion rates. This process takes advantage of a coupled enzyme system present in the growing microorganisms to drive the reaction to completion. The L-amino acid product may be recovered from the culture broth.

The Reaction The alpha-keto acid to L-amino acid biotransformation is an enzymatic transamination reaction The biotransformation is mediated by a coupled enzyme-coenzyme system comprising a transaminase. a glutamate dehydrogenase, the coenzyme NADP + /NADPH (or NAD + /NADH), and a dehydrogenase. Alternatively, the latter component of the enzyme system may be an hydrogenase for the recycling of NADP + (or NAD +) back to NADPI-1 (or NADH).

The coupled reaction system may be depicted graphically as follows:

(NADPH) or L-amino (NADH) a-ketoglutarate acid (hydro-) i J i J < genasej or --- -- gluta- --trans- Iluta- ase hy dro- dehydro genase genase 8 t 8 f (NADP ) L-glutamate a-keto or acid (NAD+) (3) (2) (1) The desired transamination reaction (step (1)), alpha-keto acid to L-amino acid, normally is reversible and would go merely to equilibrium, thus limiting the efficiency of the conversion to the amino acid.By coupling the transamination reaction to other enzymatic reactions. the transamination theoretically may be driven to completion. although chemical degradation reactions may be present which will decrease the actual yield.

The coupled enzymatic reaction designated as step (2) above, recycles one of the transamination products, alpha-ketoglutarate, back to L-glutamate. This step, a reductive amination, is mediated by glutamate dehydrogenase and requires the presence of ammonium ions. The Lglutamate available for the transamination of the alpha-keto acid to the L-amino acid is continuously replaced by this recycle step.

Step (2) utilizes one of the coenzymes NADPH or NADH, which is converted to NADP + or NAD +. A third enzymatic reaction, step (3), converts this NADP + (NAD +) product back into NADPH (NADH) via either a dehydrogenase reaction or an hydrogenase reaction. Thus, the NADPH (NADH) required for step (2) also is continuously replenished. The NH4 + required for step (2) is supplied in the nutrient medium.

The enzymatic reaction system itself is known. However, prior uses of this system have required the addition of three different microorganisms to supply the reaction components of the three steps depicted above, U.S. 3,183,170 (Kitai et al.) or the addition of microorganisms, amino donors and cofactor (Yamada et al. The Microbial Production of Amino Acids, pp.

440-46 (1976)). Moreover, these prior methods use resting cell suspensions.

The invention disclosed herein utilizes a single microorganism strain to provide the entire coupled enzymatic reaction system. A logarithmically growing culture of suitable microorganisms, with the provision of ammonium ions in the culture medium, will mediate all three steps of the biotransformation of alpha-keto acids to their respective L-amino acids. In the method of this invention, step (3) in the coupled enzyme system utilizes a dehydrogenase reaction for converting NADP + (NAD +) to NADPH (NADH).

The Reactants The coupled enzymatic reaction system is present in actively growing microorganisms. This naturally occurring system now can be utilized for the efficient conversion of alpha-keto acids to L-amino acids. By providing suitable microorganisms with the proper substrate, the conversion is carried out in high yields by the microorganisms themselves.

The types of microorganisms suitable for use in this bioconversion process are considerably varied. For example, bacterial strains known to be glutamic acid producers have been utilized with success. These include Brevibacterium thiogenitalis (ATCC No. 31723), Brevibacterium lactofermentum (ATCC No. 1 3655), Brevibacterium ammoniagenes (ATCC No. 13746), Brevibacterium glutamigenes (ATCC No. 13747), Brevibacterium flavum (ATCC No. 13826) and Corynebacterium hercules (ATCC No. 13868). The ATCC No. indicated is the designation number given each culture by the American Type Culture Collection, 1 2301 Parklawn Drive, Rockville, Maryland 30852, where each of the above cultures has been deposited. It is expected that other glutamic acid producers also will be suitable for use with this process.

In addition, it has been found that E. coli HB101, not normally considered a glutamic acid producer, may be used in this process with yields equivalent to the above-mentioned microorganisms. E. coli is not a glutamic acid producer in the sense that it does not excrete glutamic acid into the medium. However, all microorganisms produce small intracellular amounts of the substance. The enzyme system of interest here requires only catalytic amounts of glutamic acid to start the reaction and E. Coliappears to generate sufficient amounts of glutamic acid to complete step (2) of the coupled system.

It is apparent that there is a wide variety of microorganisms useful in this bioconversion process. The primary requirement is that the microorganism be capable of absorbing and converting an alpha-keto acid to the corresponding L-amino acid. In order to most efficiently carry out the bioconversion, the microorganism should be capable of producing sufficient quantities of the enzymes involved in steps (1) through (3) of the coupled enzyme system to drive the transamination reaction beyond the equilibrium point substantially towards completion.

Finally, the microorganism preferably should be capable of excreting the L-amino acid into the culture medium for convenient and economical recovery.

The alpha-keto acid selected for use with this process will depend upon which L-amino acid is the desired product. For example, phenylpyruvic acid is converted to L-phenylalanine, alphaketoisocaproic acid to L-leucine, alpha-ketoisovaleric acid to to L-valine, pyruvic acid to L-alanine, oxaloacetic acid to L-aspartic acid, ss-hydroxy-aípha-ketobutyric acid to L-threonine, p-o.yphenyl pyruvic acid to L-tyrosine, indole pyruvic acid to L-tryptophan, alpha-keto-P-methyl valeric acid to L-isoleucine, etc. As can be seen, conventional L-amino acid precursors are used in the inventive process.

It is preferred that the alpha-keto acid be added to the culture in aqueous solution as this form facilitates pumping the precursor into the culture broth. However, it desired, the alpha-keto acid may be in other physical forms, such as a solid or slurry.

The purity of the alpha-keto acid precursor has been shown to affect the L-amino acid yield rather dramatically, at least in the case of the conversion of phenylpyruvate to phenylalanine (see Example 5). In adition to the increased amounts of alpha-keto acid available in purified precursor, it is believed that the microorganisms may display less sensitivity to the purified alpha-keto acid.

The concentration of the aqueous solution of alpha-keto acid, if that is the form in which it is fed to the culture, may vary considerably. For example, if this invention is used in a continuous fermentation, the solution would be more dilute than for batch fermentation. The upper limit on concentration may depend on the solubility of the alpha-keto acid in the desired solvent. It is expected that levels of about 1.0 to 200 gm/l may be suitable, with the concentration depending on the particular alpha-keto acid, as well as on the factors discussed above. The alpha-keto acid most conveniently is dissolved in water but other solvents compatible with the fermentation, additional growth medium, for example, may be used if desired.

The nutrient medium should be chosen to provide optimal growth conditions for the microorganism used. Media variation and adjustment is well within the capability of persons skilled in this art and need not be discussed in detail here.

In addition to the basic nutritional requirements, however, the medium should contain a source of NH + . The ammonium ions are used by the microorganisms both to increase cell mass and for step (2) of the coupled enzymatic reaction system. Thus, the levels of ammonium ion in the culture medium should be geared to the desired production of amino acid and cell mass. This adjustment of NH4 + level also is well known to persons skilled in this art. As an example, the ions conveniently may be supplied by adding to the nutrient medium ammonium sulfate in a concentration of about 10 to about 50 gm/l. Alternatively, urea, ammonium chloride, ammonium nitrate, ammonium acetate, etc. may be used as the NH4 + donor.

The Reaction Process The bioconversion process of this invention requires a culture of actively growing microorganisms. This culture may be prepared by inoculating a quantity of sterile nutrient or growth medium with a seed culture of the desired strain. The seed culture used to inoculate the nutrient medium preferrably is allowed to grow for about 6 to 40 hours prior to use as inoculum. The length of time for growing the seed culture prior to inoculation will vary depending on the bacterial strain used. It is desired that the seed culture comprise a healthy, growing cell mass sufficient to form a good basis for establishment and growth of the fermentation culture.

The size of the fermentation reactor will depend on the specific application of this process. It is well within the knowledge of those skilled in the art to adjust the quantities of medium and seed culture to the requirements of a specific reactor or desired yield.

The culture is allowed to grow in the fermentation reactor until the culture has become acclimated to the reactor environment and has grown to a cell mass sufficient to withstand addition of the alpha-keto acid. This is particularly important when impure precursor is used.

Optical density may be a convenient measure of cell growth. An optical density ("O.D.") (640 nm) of approximately 10 generally indicates that the culture is ready for the addition of the alpha-keto acid. This corresponds to a growth period of approximately 6-40 hours, depending on the organism and growth conditions. However, this may be somewhat variable and should not be viewed as a strict requirement, but rather left to the experience and discretion of the worker.

The culture is grown at a temperature compatible with maximal growth of the bacterial strain used. The general range will be from ambient temperatures up to about 37 C. For Corynebac teria and Brevibacterium. the preferred temperature is about 30"C; for E. Coli. the preferred temperature is about 37 C. The pH of the culture preferably should be about 7-8. Aeration may be provided if necessary for the bacterial strain selected.

At this stage, the reactor is fed with the desired alpha-keto acid. The alpha-keto acid preferably is in an aqueous solution as described above. The alpha-keto acid concentration in this aqueous solution may be from about 1.0% to about 20% (wt/wt). The solution pH may be from about 7 to about 14. The pH may be adjusted to lower pHs with glacial acetic acid or other biologically compatible acids (such as sulfuric acid or citric acid) or with gaseous carbon dioxide. The solution may be sterilized by steam (e.g., at 121 C for 15 minutes). Milli-pore filtration (e.g., 0.22 iz pore size) or by other convenient means.

If necessary for continued culture growth. a sterile glucose solution may be fed into the fermentor with or at about the same time as the alpha-keto acid. However, excess glucose remaining in the culture broth at the end of the fermentation may interfere with the recovery of the amino acid product. If that occurs, the quantity of glucose provided to the culture should be iimited to the amount calculated to be utilized by the cell mass during fermentation.

The addition of alpha-keto acid (and glucose, if desired) preferably is made aseptically to avoid c:nt,,minating the reactor. The reactor culture broth concentration of the alpha-keto acid probably will be in the range of about 20 gm/l to about 50 gm/l for a slug fed batch reactor.

For continuous fermentation, the concentration may be lower. However. it is desired to have the alpha-keto acid concentration as high as possible, without disrupting metabolism, in order to maximize its utilization by the microorganisms during the logarithmic growth phase.

The culture then is allowed to grow for an additional period, preferably about 20 to about 72 hours. This period, during which the alpha-keto acid Is biotransformed into the L-amino acid.

preferably is as short as possible, while still allowing for maximum yields. By the process of this invention, maximum expected yields may be as high as 100% in as short a period as 23 hours, as shown for the production of L-phenylalanine in Example 6. After the growth period, the Lamino acid product may be recovered from the culture broth by conventional means.

The following Examples are given for illustrative purposes only and are not meant to limit the invention described herein.

Example 1 The following seed culture media were prepared: Seed 1 Ingredients Growth1 30.00 gm Glucose2 100.00 gm 2.50 gm Corn steep liquor 2.50 gm 2.50 gm NZ Amine B 2.50 gm 1.00 gm SzHPO4 1.00 gm 0.25 gm MgSO4.7H2O 0.25 gm 10.00 gm (NH4)2So4 3 50.00 gm 1.00 cc Biotin stock solution 1.00 cc 1.00 cc Thiamine-HCl stock solution4 1.00 cc 20.90 gm MOPS buffer5 20.90 gm 1 - Per liter deionized water.

2 - A 70t (wt/vol) glucose stock solution.

3 - 100 mg d-biotin per liter deionized water.

4 - 1.0 gm thiainine-HCl per liter deionized water.

5 - MOPS = morpholinopropane-sulfonic acid.

All ingredients except for the glucose were combined and steam sterilized at 121 do for 15 minutes. The glucose solution was sterilized separately by the same method and then added aseptically to the rest of the media. Flasks (250 cc) of seed culture medium were used to grow two seed cultures for each of six glutamic acid-producing bacterial strains.The strains and the optical densities for each strain after the seed cultures were allowed to grow for 8 hours are as follows: Strain Optical Density (640nm) Brevibacterium 3.9 + 0.1 thiogenitalis (ATCC 31723) Brevibacterium 11.2 + 0.8 lactofermentum (ATCC 13655) Brevibacterium 6.0 + 0.4 ammoniagenes (ATCC 13746) Brevibacterium 9.2 + 0.2 glutamigenes (ATCC 13747) Brevibacterium 13.3 + 0.3 flavum (ATCC 13826) Corynebacterium 12.3 + 0.7 hercules (ATCC 13868) The sterilized growth medium was added in 50 cc aliquots to 250 cc indented shake flasks.

Each flask was inoculated with a single seed culture on a 1 % vol/vol basis. The shake flask cultures were allowed to grow at 300 RPM shaking speed for 1 7 hours at 30"C for both the Brevibacterium and the Corynebacterium cultures.

An aqueous sodium phenylpyruvic acid (Na-PPA) solution was used. The solution was an unpurified sodium hydroxide hydrolyzate of 5-benzylidenehydantoin. Preparation of the hydrolyzate of the 5-benzylidenehydantoin (Hamsphire Chemical Co.) was according to the following procedure: 2233 gm deionized water and 223.3 gm NaOH were combined in a 3 liter stainless steel beaker and heated to boiling. While continuing to boil, 350 gm 5-benzylidenehydantoin was added slowly. Boiling was continued at 101-103"C for 2 hours. The solution was mixed during the reaction with an electric mixer. After boiling, the mixture was cooled down quickly in an ice water bath. The final volume of this hydrolyzate was 1,810 ml, with a Na-PPA-N20 concentration of 162 gm/l, a pH of 13.51 and a color of clear amber.A one molar solution of the sodium phenylpyruvate (hydrolyzate) 1.2M NaOH, 0.5M sodium carbonate and 0.5M urea.

The Ph of the solution was adjusted to 8 using 2N acetic acid and was then sterilized using a Milli-pore filter (0.22 it pore size).

A separate glucose solution (70% wt/vol in deionized water) was prepared and was steam sterilized at 121 C for 1 5 minutes. The glucose solution was added aseptically to the sterile Na PPA solution at a volumetric ratio of 3:7. A volume of the composite solution was added to each shake flask so that a 10 gm/l Na phenylpyruvate concentration in the culture broth was achieved.

Each shake flask culture was allowed to grow for an additional 23 hours, for a total growth period of 40 hours. At that time, broth samples were removed and assayed for L-phenylalanine using a high pressure liquid chromatograph (HPLC) system. Milli-pore filtered (0.22 IL pore size) broth samples were derivatized using a Dansylchloride technique prior to injection onto the HPLC.

Dry cell weights also were determined for each broth sample. A 10 cc sample of broth from each flask was centrifuged at 10,000 RPM for 30 minutes. The wet cell paste was dried in an oven at 80"C for 24 hours, followed by 1 5 minutes in a vacuum dessicator. The dry cell sample then was weighed.

For each shake flask, the average rate of L-phenylalanine production over the 23 hour period was calculated per gram dry weight of cell mass. The test results for duplicate shake flasks for each strain of microorganism are shown in Table Table 1 Dry Cell % Conversion Microorganism Wt. (gm/l) mM (gm/l) L-phenylalanine (mole/mole) Average Rate ATCC &num;31723 39 # 0.8 40.9 # 1.0 (6.8 # 0.2) 67 # 1.5 45.6 # 1.9 ATCC &num;13655 22.4 # 3.2 25.7 # 4.5 (4.3 # 0.8) 42.3 # 7.5 49.6 # 1.7 ATCC &num;13746 34.7 # 0.6 28.2 # 0.3 (4.65 # 0.05) 46.2 # 0.5 35.3 # 0.2 ATCC &num;13747 46.5 # 12.6 29.4 # 1.5 (4.9 # 0.3) 48.2 # 2.5 39 # 9.6 ATCC &num;13826 51.6 # 5.1 41.2 # 7.9 (6.8 # 1.3) 67.6 # 12.9 34.4 # 3.2 ATCC &num;13868 42.5 # 1.5 38.4 # 2.1 (6.4 # 0.4) 63.1 # 3.5 39.3 # 0.8 1-M L-phenylalanine produced/hour/gm dry cell weight.

Example 2 The following culture medium (luria broth) was prepared: Ingredient Quantity Bacto-Tryptone 10.0 gm Bacto-Yeast Extract2 5.0 gm NaCl 3 5.0 gm MOPS buffer 20.9 gm 1 - Per liter deionized water.

2 - Available from Difco.

3 - MOPS = inorpholinopropane-sulfonic acid.

The ingredients were combined and were steam sterilized at 121 C for 15 minutes. The sterilized medium was added in 50 cc aliquots to 250 cc indented shake flasks. Each shake flask was inoculated with a culture of Escherichia coli HB 101 on a 1% (vol/vol) basis.

Th cultures in the flasks were allowed to grow for 8 hours at 37 C and 300 RPM shaking speed for use as seed cultures. At the end of this period the optical density (640nm) was 6.27 i 0.03 for the duplicte shake flask experiments. This seed culture then was used to inoculate fresh 250 cc shake flasks, which were allowed to grow 17 hours before being fed with the alpha-keto acid.

A sterile aqueous Na phenylpyruvic acid-glucose solution was prepared as in Example 1. This solution was fed to the test shake flask cultures at the end of the 17 hour growth period (optical density = 10.3 # 1.3), as in Example 1, to achieve a final broth Na phenylpyruvic acid of 10 gm/l. After an additional 23 hours of growth, samples were removed and dry cell weights and L-phenylalanine concentrates determined as in Example 1. The results are shown in Table 2. Table 2 Dry Cell % Conversion Wt. (gm/l) mM (gm/l) L-phenylalanine (mole/mole) Average Rate 19.4 # 0.3 31.8 # 5.8 (5.3 # 1.0) 52.2 # 9.5 71.3 # 13.8 1- M L-phenylalanine produced/hour/gm dry cell weight.

Example 3 The following seed culture medium was prepared: Ingredients Quantity Hysoy I 3.0 gm L-Isoleucine 1.0 gm MgSO4'7H2O 0.4 gm (NH4)2S04 10.0 gm KH2PO4 3.0 gm Trace elements2 1.0 cc CaCO3 15.0 gm Biotin stock solution3 1.0 cc Thiainine-HCl stock solution4 1.0 cc Deionized water 700.0 cc 1 - available from Sheffield Co.

2 - Trace element stock solution consisted of: Ingredient Quantity ZnSO4 . 7H2O 8.8 gm FeSO4 . 7H2O 10.0 gm CuSO4 . 5H2O 0.06 gm Na2B4O 10H20 0.088 gin Na Mo 4 .2H20 0.053 gm MnSO4 . H2O 7.5 gm CoCl2 . 6H2O 0.12 gm CaC12 0.055 gm Deionized water 900.00 cc The trace element stock solution was adjusted to pH 2 with concentrated H2SO4 and deionized water added to bring the final volume to 1000 cc.

3 - 100 mg d-biotin per liter deionized water.

4 - 1.0 gm thiainine-HCl per liter deionized water.

The above medium was adjusted to pH 7.5 with 5N NaOH, and deionized water added to bring the volume to 900 cc. The medium was steam sterilized at 121 for 15 minutes. A stock solution glucose (50 gm/100 cc deionized water) was steam sterilized separately at 121 C for 15 minutes. After cooling, the sterile glucose was added aseptically to the sterile medium to bring the final medium volume to 1000 cc. The pH was adjusted to 7 with 5 N glacial acetic acid.

Fifty cc aliquots of the seed culture medium were added to each of two 250 cc indented shake flasks. The flasks were inoculated with Brevibacterium thiogenitalis ATCC No. 31723 and grown for 8 hours at 30 C with shaking (300 RPM shaking speed) to form seed cultures. At the end of the 8 hours growth, the optical density (640nm) was about 10.0.

Growth medium was prepared in the same manner as the seed medium, except that 50 gm of (NH4)2SO4 was used. Fifty cc aliquots of the growth medium were distributed to two 250 cc indented shake flasks. The flasks then each were inoculated with a 2.5 cc sample from one of the seed cultures. After inoculation, the shake flask cultures were grown for 33 hours at 30 C with shaking (300 RPM).

A sterile glucose solution (70% wt/vol in deionized water) was prepared as in Example 1. An aqueous unpurified 17.5% (wt/wt) solution of Na alpha-ketoisocaproic acid (density = 1.125 gm/cc; pH = 13.5) was used. The solution was a sodium hydroxide hydrolyzate of isobutylidenehydantoin prepared according to the method described in Example 1 using a sufficient quantity of the isobutylidenehydantoin to form a 15% solution in water and NaOH prior to boiling. A one molar solution of the Na alpha-ketoisocaproic acid (hydrolyzate) contained about 1.2M NaOH, 0.5M sodium carbonate and 0.5M urea. The Na alpha-ketoisocaproic acid solution was adjusted to pH 7 with concentrated glacial acetic acid, and steam sterilized at 121 C for 15 minutes. Equal volumes of the sterile glucose solution and sterile Na alpha-ketoisocaproic acid solution were mixed aseptically and the pH readjusted to 7 with glacial acetic acid.

Volumes of the combined solution were added to the shake flask cultures so that the final broth concentrations of alpha-ketoisocaproic acid were as shown in Table 3. The cultures were allowed to grow for an additional 62 hours. After this growth period, samples were taken, Milli pore filtered (0.22 IL pore size) to remove the cells, and analyzed for L-leucine using a high pressure liquid chromatograph (HPLC) assay as described in Example 1. The results are shown in Table 3.

Table 3 Na c-Ketoisocaproic Molar Acidl L-Leucine Produced2 Yield3 0.03l2M (4.10 gum/1) 0.025M (3.28 gm/l) 80.0% 0.0354M (4.61 gm/l) 0.037M (4.85 gm/l)4 105.084 0.0624M (8.12 gm/l) 0.058M (7.61 gm/l) 93.78 1 - Added at 33 hours growth.

2 - After an additional 62 hours growth.

3 - Molar Yield = gm L-Leucine x 100 grfl Na a-Ketoisocaproic Acid 4 - Analytical error on HPLC assays is approximately +10E, suggesting that this number is an analytical error on the high side.

Example 4 The seed culture medium and growth medium of Example 3 were used to grow Brevibacterium thiogenitalis ATCC No. 31 723 for the purpose of biotransforming alpha-ketoisovaleric acid to L-valine. Seed cultures and test shake flask cultures were prepared according to the method described in Example 3, using this strain.

An aqueous unpurified 4.53% (wt/wt) solution of Na alpha-ketoisovaleric acid solution was used. The solution was a sodium hydrolyzate of isopropylidenehydantoin prepared according to the method described in Example 1 using a sufficient quantity of the isopropylidenehydantoin to form a 6% solution in water and NaOH prior to boiling A one molar solution of the Na alphaketoisovaleric acid contained about 1.2M NaOH, 0.5M sodium carbonate and 0.5M urea. The solution pH of 13.8 was adjusted to 7.0 with glacial acetic acid. The solution was steam sterilized at 121 C for 15 minutes.

A separate glucose solution (70 /0 wt/vol in deionized water) was prepared and was steam sterilized at 121"C for 15 minutes. Seventy parts of the sterile alpha-ketoisovaleric acid solution were added aseptically to 30 parts of the sterile glucose solution. Volumes of this mixture were added to the test shake flask cultures at various stages of growth such that the final broth concentrations of Na alpha-ketoisovaleric acid was either 10 or 15 gm/l, as shown in Table 4.

The total growth period was 40 hours for each test shake flask culture. At that point, samples were withdrawn, Milli-pore filtered (0.22 p pore size) and analyzed for L-valine using an HPLC assay as in Example 1.

In a control study using the same microorganisms and procedures, Na alpha-ketoisovaleric acid was omitted from the fermentation. The controls produced an average of .0205M (2.4 gm/l) L-valine from glucose. This has not been accounted for in the test results shown in Table 4.

Table 4 Na a-Keto- Time of L-Valine Molar isovaleric Acid Addition Produced Yield2 0.0725M 16 hrs .0846 + .008M (10 gm/l) (9.90 + 0.90 gm/l) 998 0.0725M 24 hrs .0910 + .003M (10 gm/l) (10.65 + 0.35 gm/l) 3 10663 0.1087M 16 hrs .1004 + .005M (15 gm/l) (11.75 + 0.55 gm/l) 78% 1 - After 40 hours total growth period.

2 - Molar Yield = gm L-Valine x 100 gm Na a-ketoisovaleric Acid 3 - Analytical error on HPLC assays is approximately +10t, suggesting that this number is an analytical error on the high side.

Example 5 This Example compares the biotransformation efficiency of Brevibacterium thiogenitalis ATCC No. 31723 when fed with a KOH 5-benzylidenehydantoin hydrolyzate (K phenylpyruvic acid) and when fed with a 98% pure Na Phenylpyruvic acid. The alpha-keto acid solutions were prepared as follows: K Phenylpyruvic Acid--An unpurified KOH hydrolyzate of 5-benzylidenehydantoin (obtained from Hampshire Chemical) was prepared as in Example 1 from 3M KOH per mole 5benzylidenehydantoin The pH of the hydrolyzate was adjusted to 8.0 with CO2 gas prior to feeding the shake flasks.

Na Phenylpyruvic Acids solid Na phenylpyruvate monohydrate (98%) (obtained from Aldrich Chemical Co.) was dissolved in 0.6M aqueous KOH to a concentration of 98 gm/l (as Na phenylpyruvate monohydrate (Na PPa H20). The pH was adjusted to 8.0 with CO2 gas prior to feeding the shake flasks.

The seed and growth media used were prepared as described in Example 1, except that the MOPS buffer was replaced with 20 gm/l CaCO3 and the glucose plus other media components were autoclaved together for 10 minutes at 110"C (instead of separate steam sterilization). Seed cultures were prepared as in Example 1, using Brevibacterium thiogenitalis ATCC No. 31 723.

The seed cultures were used to inoculate shake flasks as described in Example 1. After an initial 1 7 hour growth period, separate shake flasks were fed with the phenylpyruvic acid solutions in amounts such that the final culture broth concentrations of phenylpyruvate were 76.22 mM (12.51 gm/i as phenylpyruvic acid). The shake flasks were allowed to grow, and molar yields were determined at 23 hours (40 hours total growth) and at 28 hours (45 hours total growth). HPLC assays were done as in Example 1. The results. shown in Table 5, indicate that the ability of Brevibacterium thiogenitalis ATCC No. 31 723 to biotransform phenylpyruvate to L-phenylalanine is not fully realized when impure precursor is used in this process.

Table 5 Phenylpyruvate L-Phenylalanine Production (mM) Molar Molar 23 Hrs Yield 28 Hrs Yield K phenyl- 35.96 # .85 47.2 # 1.1 49.52 # 12.7 65 # 16.7 pyruvate Na phenyl- 67.80 + 0 89.0 + 0 69.19 + 1.2 90.8 + 1.6 pyruvate 1 - Duplicate shake flask tests.

2 - Time from addition of phenylpyruvate.

3 - Molar yield = mM L-Phenylalanine x 100 mM Phenylpyruvate Example 6 This Example was conducted using sodium phenylpyruvic acid (Na-PPA) purified by precipitation from a sodium hydroxide hydrolyzate of 5-benzylidenehydantoin. The hydrolyzate was prepared according to the method of Example 1.

An acetic acid precipitate was then prepared from the hydrolyzate. A 1 500 ml volume of hydrolyzate was placed in a 200 cc glass beaker, in an ice water bath. Glacial acetic acid (conc.

99.7%, 17.4M) was added gradually while stirring to lower the pH to 9. The temperature was not allowed to rise above 30 C. Precipitate began to form at about pH 12.95; after 15 minutes, the pH had reached 9.

The mixture was allowed to stand, covered and under a hood, for 1 hour at 23 C. The slurry was filtered through &num;4 Whatman paper, without washing. The wet precipitate was placed in a vacuum oven at 50 C for 18 hours. The dry precipitate was put through a &num;40 mesh sieve. The final product was 79% pure Na-PPa-H2O, with other components including acetate, Na+ and small amounts of CD3-- and organic nitrogen.

Seed and growth culture media were prepared as described in Example 5. Brevibacterium thiogenitalis ATCC No. 31723 was grown on the seed medium as in Example 5. The seed culture was used to inoculate 250 cc shake flasks.

After an initial 18 hour growth period, the precipitated Na-PPA-H20) was redissolved in 0.3M KOH (100 gm/l as Na-PPA-H20) was added to the shake flasks to achieve the final broth concentrations indicated in Table 6. The cultures were allowed to grow for an additional 23 hour period. HPLC assays for L-phenylalanine were done as described in Example 1. The results are shown in Table 6. The results are approximate as the evaporative loss of liquid from the shake flasks was not measured.

Table 6 Growth at 23 hrs. Molar4 Test No. Na-PPA-H2O L-Phenylalanine (O.D.) 640 nm) Yield 1 0 0 34 # 4 2 1.73 # .12 1,8 # .09 41 # 2 100 3 3.06 # 0 3.69 # .02 40 # 2 100 4 3.3 # .08 3.57 # .19 24 # 1 100 5 5.71 # 0 6.32 # .19 27 # 1 100 6 5.75 # .36 6.18 # .11 23 # 0 100 7 6.80 # .05 6.05 # .05 43 # 2 88 8 8.16 # .44 7.4 # 0 43 # 0 90 9 9.13 # .36 6.99 # .14 27 # 0 76 10 9.45 # 0 8.35 # .15 36 # 7 88 11 9.49 # 0 7.82 # .12 24 # 1 82 12 12.91 # .04 10.05 # .95 37 # 5 78 Duplicate shake flask experiments.

Colorimetrically assayed PPA concentration in the medium (gm/l).

Concentration of L-Phenylalanine (gm/l) produced in 23 hours.

M L-phenylalanine 4 Molar yield = x 100.

M Phenylpyruvate

Claims (12)

1. A process for the preparation of L-amino acid comprising: (a) growing microorganisms which are capable of transforming an alpha-keto acid into the corresponding L-amino acid, in a nutrient medium, (b) feeding the desired alpha-keto acid to the logarithmically growing microorganisms, and (c) allowing the growing microorganisms to convert said alpha-keto acid to the corresponding L-amino acid

2.A process for the biological transamination of an aklpha-keto acid to the corresponding Lamino acid using the coupled enzyme system of a logarithmically growing bacterial culture which comprises: (a) contacting a logarithmicaily growing culture of microorganisms capable of transforming alpha-keto acids into the corresponding L-amino acids with the desired alpha-keto acid in the presence of a nutrient medium, and (b) allowing the coupled enzyme system of said growing culture to drive the alpha-keto acid to L-arnino acid transamination reaction.

3. The process of Claim 1 or 2 in which said microorganisms substantially comprise a single bacterial strain.

4. The process of Claim 1 or 2 in which said microorganisms are selected from the group comprising glutamic acid excreting bacteria and Eschenchia coli

5. The process of Claim 4 in which said glutamic acid producing excreting bacteria are elected from the group comprising Rrevibacteriurn species and Corynebacterium species.

.). rl.e process of any of Claims 1 to 5 in which the L-amino acid is recovered from the culture medium.

7. The process of any of Claims 1 to 6 in which the microorganisms are grown for about 6 to about 40 hours before addition of the alpha-keto acid and for from about 20 to 72 hours while the alpha-keto acid is fed thereto.

8. The process of any of Claims 1 to 7 in which said alpha-keto acid is present in the amount of about 0.5% (wt/vol) to about 5.0 M0 (wt/vol) in the culture broth.

9. The process of any of Claims 1 to 8 in which said alpha-keto acid is present in an aqueous solution.

10. The process of Claim 9 in which the concentration of said aqueous solution is from about 1.0% (wt/vol) to about 20% (wt/vol).

11. The process of any of Claims 1 to 10 in which the said alpha-keto acid is phenylpyruvic acid, alpha-ketoisocaproic acid, alpha-ketoisovaleric acid, pyruvic acid, oxaloacetic acid, p- hydroxy-alpha-ketobutyric acid, p-oxyphenyl pyruvic acid, indole pyruvic acid or alpha-keto-ss- methylvaleric acid.

1 2. The process of Claim 1 or 2 substantially as hereinbefore described.

1 3. L-amino acid prepared by (a) culturing in nutrient medium microorganisms capable of transforming a precursor alpha keto acid to the desired L-amino acid, (b) causing the culture to be in an actively, logarithmically growing phase, (c) contacting the logarithmically growing culture with some precursor alpha-keto acid, and (d) allowing said growing culture to convert the alpha-keto acid to the corresponding L-amino acid.

1 4. The L-amino acid of Claim 1 3 which has been recovered from the culture medium.

1 5. The L-amino acid of Claim 1 3 or 1 4 which is prepared from a precursor alpha-keto acid selected from phenylpyruvic acid, alpha-ketoisocaproic acid, alpha-ketoisovaleric acid, pyruvic acid, oxaloacetic acid, p-hydrnxyl-alpha-ketobutyric acid, p-oxyphenyl pyruvic acid, indole pyruvic acid and alpha-keto-ss-methyEvaleric acid.

1 6. L-amino acid produced according to the process of any Claims 1 to

1 2.