CA2274896A1 - Method for production of d-proline derivatives - Google Patents

Abstract

The invention relates to new microorganisms which are capable of utilizing a proline derivative in the form of the racemate or its optically active isomers, of the general formula (I), in which R1 means alkyl, acyl or hydrogen, R2 means hydrogen or hydroxy and R3 -NH2, aryloxy or alkoxy, as the only source of nitrogen, the only source of carbon, or as the only carbon and nitrogen source. These microorganisms are used for new methods of production of D-proline derivatives.

Description

Process for preparing D-proline derivatives The present invention relates to novel microorganisms which are able to utilize, as sole nitrogen source, as sole carbon source or as sole carbon and nitrogen source a proline derivative in the form of the racemate or its optically active isomers of the general formula R3 COR' N
~R' where R1 is alkyl, acyl or hydrogen, R2 is hydrogen or hydroxy and R3 is -NH2, aryloxy or alkoxy. These microorganisms or their cell-free enzymes are used for a novel process for preparing a D-proline derivative of the general formula II and a D-proline derivative of the general formula VI
RI r,, CORD RI ,,,, COOH
,N
li ~ VI
R, N'R' where Rl, RZ and R3 have the meaning given above .
D-proline and its derivatives are important intermediates for preparing pharmaceuticals (J. Org.
Chem., 1994, 59, 7496-7498).
A plurality of processes are known for preparing D-proline.
JP-A-92183399 describes a process for preparing D-proline starting from (DL)-proline using microorganisms of the genus Candida or Trichospora.
This process has the disadvantage that the reaction REPLACEMENT SHEET (RULE 26) time is too long and D-proline is obtained in low yield.
JP-A-07127354 describes a process for preparing D-proline starting from ornithine using microorganisms of the species Proteus mitajiri. Disadvantages of this process are that, firstly, the starting material ornithine is too expensive, and secondly that D-proline is obtained in poor yield.
In addition, JP-A-07289275 describes a process for preparing D-proline starting from L-proline. In this process, L-proline is racemized using microorganisms of the genus Escherichia which have racemate activity to form (DL)-proline which is then cultured with L-proline-degrading microorganisms, in order to obtain D-proline. It is a disadvantage of this process that the resultant D-proline must be further purified and this is associated with high losses of yield.
DE-A-43 30 678 comprises a process for preparing N-carbamoyl-D-proline starting from a bicyclic hydantoin. In this process, the bicyclic hydantoin is converted to N-carbamoyl-D-proline by Agrobacterium radiobacter microorganisms. This process has the disadvantage that, firstly, the conversion time is too long, and secondly that the corresponding D-proline derivative is only obtained in moderate yield.
It is an object of the present invention to isolate microorganisms which can be used for a simple and industrially practicable process for preparing D-proline derivatives of the general formulae II and VI. The corresponding products are to be isolated here in good yield with good enantiomeric purity.
This object is achieved by the microorganisms, and enzyme extracts therefrom, according to Claim 1 together with the enzymes according to Claim 5 and together with the process according to Claim 6 and 10.
The microorganisms according to the invention can be isolated from soil samples, sludge or wastewater using conventional microbiological techniques.
According to the invention, these microorganisms are isolated in such a manner that these microorganisms are cultured in a conventional manner in a medium comprising a proline derivative of the general formula I in the form of the racemate or one of its optically active isomers - as sole carbon and nitrogen source or - as sole nitrogen source together with a suitable carbon source or - as sole carbon source together with a suitable nitrogen source.
From the resultant culture, expediently, those are then selected which utilize an L-prolinamide of the general formula I as sole nitrogen source.
The radical R1 in the proline derivative of the general formula I is alkyl, acyl or hydrogen. The radical RZ is hydrogen or hydroxy. The radical R3 is -NH2, alkoxy or aryloxy.
Alkyl is defined below as a C1_5 alkyl group, substituted or unsubstituted. Suitable substituents are, for example, halogen, in particular fluorine, chlorine and bromine, and hydroxy. Examples of an unsubstituted C1_5 alkyl group are methyl, ethyl, propyl, isopropyl, butyl, isobutyl, tertiary butyl, pentyl, isopentyl (isoamyl). Examples of a substituted C1_5 alkyl group are chloromethyl, trifluoromethyl, 2-bromopropyl and hydroxymethyl.
Acyl is defined below as acetyl, formyl, propionyl, butyryl, pentanoyl, benzoyl or phenylacetyl.
Alkoxy is defined below as a C1_5 alkoxy group, substituted or unsubstituted. Suitable substituents are, for example, the substituents mentioned above for alkyl. Examples of an unsubstituted C1_5 alkoxy group are methoxy, ethoxy, propoxy, isopropoxy, butoxy, pentoxy, isopentoxy or isobutoxy. Examples of a substituted C1_5 alkoxy group are trifluoromethoxy, chloromethoxy, 2-chloropropoxy and hydroxymethoxy.

Aryloxy is defined below as benzyloxy and phenyloxy.
As suitable carbon source, the microorganisms can utilize, for example, sugars, polyols, pentones, carboxylic acids or amino acids as growth substrate. As sugars, use can be made of hexoses, such as glucose or fructose, pentoses, such as xylose, and disaccharides, such as maltose or sucrose. As polyol, use can be made of, for example, glycols, glycerol, sugar alcohols, such as mannitol, sorbitol or inositol. As carboxylic acids, use can be made of dicarboxylic or tricarboxylic acids or their salts, such as oxalic acid, malonic acid, succinic acid, glutaric acid, adipic acid, pimelic acid, azelaic acid, sebacic acid, malefic acid, fumaric acid, sorbic acid, agaricic acid, citric acid, 1,2,3-propanetricarboxylic acid, citrate, fumarate, oxalate or malate. As amino acids or their salts, use can be made of all proteinogenic amino acids or their salts, such as aspartic acid, glutamic acid, aspartate, glutamate, glutamine, asparagine, alanine, valine, leucine, isoleucine, proline, tryptophan, phenyl-alanine, methionine, glycine, serine, tyrosine, threonine, cysteine, histidine, arginine.
Preferably, as carbon source, use is made of a sugar or sugar alcohol.
As suitable nitrogen source, the microorganisms can utilize, for example, ammonium compounds or amino acids or their amides. As ammonium compound, use can be made of, for example, ammonium chloride, ammonium sulphate, ammonium carbonate or ammonium acetate. As amino acid, use can be made of alanine, valine, leucine, isoleucine, proline, tryptophan, phenyl-alanine, methionine, glycine, serine, tyrosine, threonine, cysteine, asparagine, glutamine, lysine, arginine or histidine. Expedient representatives of amino acid amides are prolinamide, alaninamide, glycinamide, serinamide and valinamide. Other suitable nitrogen sources are peptones, piperidincarboxamides, pyridincarboxamides or hydroxycarboxamides.

As piperidincarboxamide, use can be made of pipecolinamide, as pyridincarboxamide, use can be made of nicotinamide and as hydroxycarboxamide, use can be made of lactamide.
As selection and culture medium, use can be made of those customary in the specialized area, such as that described in Table l, for example.
During culture and selection, the effective enzymes of the microorganisms are expediently induced.
As enzyme inducer, use can be made of L- and DL
lactamide or their N-methylated derivatives such as N-methyl-L-lactamide and N,N-dimethyl-L-lacatamide.
Preferably, use is made of the L-enantiomer.
Expediently, the inducer concentration is between 0.05 and 10 g/1, preferably between 0.5 and 3 g/l.
Usually, culture and selection are performed at a temperature of from 10 to 40°C, preferably from 25 to 35°C, and at a pH between 4 and 10, preferably between 6 and 8.
Preferred microorganisms are L-prolinamide-utilizing members of the genus Aureobacterium, Klebsiella, Aeromonas, Serratia or Pseudomonas. In particular, microorganisms of the species Klebsiella pneumoniae, Aeromonas sobria, Serratia plymuthica, Pseudomonas fluorescens or Aureobacterium sp. LS10 (DSM 10203) and their functional equivalents, variants and mutants, are isolated.
The microorganisms Aureobacterium sp. LS10 were deposited in accordance with the Budapest Treaty on the 29.08.1995 with the DSM-Deutsche Sammlung von Mikroorganismen and Zellkulturen GmbH, Mascheroderweg 1 b, D-38124 Braunschweig.
"Functions of the equivalent variants and mutants" are taken to mean microorganisms or enzymes which essentially have the same properties and functions as the original microorganisms or enzymes obtained therefrom. Such variants and mutants can be formed by chance, eg by UV-irradiation.

Taxonomic description of Aureobacterium sp.
LS10 (DSM 10203).
Gram-positive coryneform rods, strictly aerobic, no acid formation or gas formation from glucose Mobility +
Spores -Catalase +
meso-diaminopimelic acid in cell wall: none Peptidoglycan type: B2(3, [L-Hsr]D-Glu(Hyg)-Gly-D-Orn 16S r DNA sequence similarity:
Sequencing of the region having the greatest variability gave 98.90 agreement with Aureobacterium luteolum. However, Aureobacterium luteolum is non mobile, ie the Aureobacterium sp. according to the invention is a novel species.
The enzymes according to the invention, the L-proline-ester hydrolases or L-prolinamide-hydrolases, which are able to hydrolyse proline derivatives of the formula I can be obtained, for example by digestion of the microorganisms which is customary to those skilled in the art. For this purpose, use can be made of, for example, the ultrasonic method, French press method or lysozyme method.
In the first stage of the process according to the invention, expediently, an L-proline derivative of the general formula C
I

where RZ has the meaning already given, is racemized to the corresponding proline derivative of the general formula R.~ ~ sC00H
~ ~~_ V
N
H
where RZ has the meaning given. The racemization can be carried out either in a known manner, for example according to EP-A-0 057 092, or the racemization takes place in a strongly alkaline environment.
The second stage, the esterification, the preparation of amides or the alkylation or acylation at the nitrogen atom of the racemic proline derivative to form the proline derivative of the general formula I
likewise takes place in a manner known per se. The esterification can be performed by reaction with an alcohol in accordance with Organikum [Organic chemistry], 1976, pp. 498ff., the amides can be prepared, for example, by ammonolysis with ammonia according to Organikum [Organic chemistry], 1976, pp. 509ff., the alkylation at the nitrogen atom can be performed by reaction, for example, with an alkyl halide according to Organikum [Organic chemistry], 1976, p. 257 and the acylation at the nitrogen atom by reaction, for example, with a carboxylic acid halide.
The conversion of the proline derivative of the general formula I to the L-proline derivative of the general formula III
Ri C~DI~
rn N
y is carried out according to the invention either using microorganisms and/or enzyme extracts therefrom or using enzymes having L-proline ester hydrolase activity _ g _ or L-prolinamide hydrolysis activity. In this bio transformation, in addition to the L-proline derivative of the general formula III, the D-proline derivative of the formula II is produced. The two compounds are isolated, if appropriate.
In principle, the biotransformation is possible using all microorganisms which utilize a proline derivative of the general formula I in the form of the racemate or its optically active isomers as sole nitrogen source, as sole carbon source or as sole carbon and nitrogen source.
Microorganisms which are particularly suitable for the process are the microorganisms described above of the genera Aureobacterium, Klebsiella, Aeromonas, Serratia or Pseudomonas. In particular, the micro-organisms of the genus Aureobacterium, such as Aureobacterium sp. LS10 (DSM 10203) and its functionally equivalent variants and mutants are suitable for the biotransformation.
The biotransformation can be carried out, after conventional culture of the microorganisms, using resting cells (non-growing cells which no longer require carbon and energy sources) or with growing cells. Preferably, the biotransformation is carried out using resting cells.
For the biotransformation, use can be of media customary to those skilled in the art, such as low-molecular-weight phosphate buffer, tris buffer or the medium described in Table 1. Preferably, the biotransformation is carried out in the medium according to Table 1.
Expediently, the biotransformation is carried out with single or continuous addition of the proline derivative of the formula I in such a manner that the concentration does not exceed 40o by weight, preferably 20o by weight.
The pH of the medium can be in a range from 4 to 10, preferably in a range from 5 to 8. Expediently, _ g _ the biotransformation is carried out at a temperature of from 10 to 50°C, preferably from 20 to 40°C.
Preferably, the biotransformation is carried out at high cell density, in particular at a cell density of from OD6so = 20 to OD6so = 60.
The biotransformation can be carried out in the absence of the alcohols formed in the proline ester hydrolysis, preferably in the absence of ethanol, propanol, butanol, isopropanol, isobutanol, isopentyl alcohol (isoamyl alcohol) and benzyl alcohol. The corresponding alcohols can be removed, for example, by adsorption.
The D-proline derivatives of the formulae II or III produced in this manner can be isolated by conventional work-up methods.
Preferably, the process is carried out without isolating the racemic proline derivative of the formula V.
The proline derivatives according to the formula VI are prepared by hydrolysing the proline derivatives according to formula II. The hydrolysis is carried out in a manner customary to those skilled in the art, for example according to Organikum [Organic chemistry] p. 476 and p. 477.

Examples:
Example 1:
Selection of prolinamide-utilizing microorganisms Firstly, a minimal medium (see Table 1) which meets the growth requirements of many microorganisms was prepared:
Table l: Minimal medium.
Na2S04 0.1 g/1 Na2HP04 ~ 2H20 2 g/1 .

KHZPOQ 1.0 g/1 NaCl 3.0 g/1 MgCl2 ~ 6H20 0 g/1 .

CaCl2 ~ 2H20 14 mg/1 .

FeCl3 ~ 6H20 0 mg/1 .

Trace element solution 1.0 ml/1 Vitamin solution 1.0 ml/1 pH 7.0 As carbon source (C source) fructose (5 g/1) was added. To enrich microorganisms which are able to hydrolyse prolinamide, DL-prolinamide (1 g/1) was added as sole nitrogen source (N source) to this basal medium. Various batches were then inoculated with soil samples from different locations and incubated (30°C, 120 rpm) until clearly visible growth could be observed. One aliquot of this culture was then reinoculated into a volume of fresh medium of equal volume and incubated again up to marked turbidity. This process was repeated three times. The enriched micro-organisms were then isolated on a solid medium (same composition as liquid medium, only with the addition of 20 g/1 of agar) and purified. In this manner, more than 20 different bacterial isolates were obtained which were able to utilize prolinamide as sole N source.

Example 2:
Testing the selected microorganisms for hydrolytic activity and selectivity to prolinamide, proline esters or N-methylproline esters The isolates obtained by the methods described in Example 1 were propagated in the medium described there, in each case using L-prolinamide (1 g/1) as N
source and a small amount of yeast extract (0.2 g/1).
The cells thus produced were harvested by centrifugation and then resuspended and washed in buffer solution (50 mM phosphate buffer, pH 7.0). After resuspension in buffer again, the ability to hydrolyse L-prolinamide was tested with resting cells. For this purpose, a suitable amount of cells was incubated (30°C) with the respective substrate (4 g/1) in buffer solution (50 mM phosphate buffer, pH 7.0). At various time points, aliquots were withdrawn and examined via thin-layer chromatography for the release of L-proline or N-methyl-L-proline. A plurality of isolates showed the hydrolytic activity which was desired in each case of L-prolinamide and various esters of L-proline and N-methyl-L-proline. In the case of prolinamide and the proline esters, in addition, the selectivity with respect to the respective L-enantiomer was also studied, and some strains, eg Klebsiella pneumoniae, Aeromonas sobria, Serratia plymuthica, Pseudomonas fluorescens and Aureobacterium sp. (DSM 10203) having a high enantioselectivity were found.
Example 3:
Growth and enzymatic activity of Aureobacterium sp.

a) N sources Aureobacterium sp. LS10 was incubated in the minimal medium described in Table 1 with glucose (10 g/1) as C source and a little yeast extract (0.2 g/1). Highly varied N sources (2 g/1; except for peptone: 10 g/1) , such as amino acids and acid amides, were added in this example. Cultures which reached a good cell density (OD6so > 0.5 in 48 h) were analysed for their enzymatic activity (hydrolysis of L-proline butyl ester). The activity after culture with L-prolinamide as N source was used as 1000.
Table 2: Enzymatic activity of Aureobacterium sp. LS10 on culture with various N sources.
N source Relative specific activity L~1 Peptone 37 Ammonium sulphate 13 L-arginine 43 L-asparagine 51 L-glutamine 28 L-prolinamide 100 L-alaninamide 38 Glycinamide 101 L-serinamide 246 L-valinamide 25 Nicotinamide 29 DL-pipecolinamide 44 L-lactamide 477 Numerous N sources could be utilized for growth by Aureobacterium sp. LS10, but the enzymatic activity was greatly influenced both negatively and positively (cf. Table 2).
b) C sources Aureobacterium sp. LS10 was incubated in the minimal medium described in Table 1 with L-lactamide (2 g/1) as N-source and a little yeast extract (0.2 g/1). Highly varied C sources (10 g/1) were added in this example. Cultures whilch reached a good cell density OD6so > 0.5 in 48 h) were analysed for their enzymatic activity (hydrolysis of L-proline butyl ester). The activity of cells after growth on glucose as C source was used as 100%.
Table 3: Enzymatic activity of Aureobacterium sp. LS10 on culture with various C sources C source Relative specific activity ~~l Glucose 100 D-fructose 87 D-xylose 115 Maltose 120 Sucrose 105 Glycerol 103 D-mannitol 98 Fumarate 56 L-aspartate 15 L-glutamate 22 Peptone 17 Aureobacterium sp. LS10 could utilize highly varied C sources, but prefers sugars and sugar alcohols. The enzymatic activity was only affected when amino acids were used as C sources (cf. Table 3).
Example 4:
Inducers of Aureobacterium sp. LS10 L-proline-ester-hydrolase a) Inducer activity of various molecules Aureobacterium sp. LS10 was incubated in a minimal medium described in Table 1 with yeast extract ( 2 g/1 ) as N source and glucose ( 10 g/1 ) as C source .
In addition, L-lactamide or other compounds structurally related to L-lactamide were added (each as 1 g/1) to the cultures which, after they reached an adequate cell density, were analysed for their enzymatic activity (hydrolysis of L-proline butyl ester), in order to test the potency of the compounds as inducers of L-proline-ester-hydrolase. In this example, in addition to the lactamide racemate, an inducing effect was only found for the N-methylated derivatives of L-lactamide (cf. Table 4).
Table 4: Enzymatic activity of Aureobacterium sp. LS10 on culture with various potential inducers Inducer Relative specific activity L~1 L-lactamide 100 DL-lactamide 113 N-methyl-L-lactamide 74 N,N-dimethyl-L-lactamide 27 b) Optimum inducer concentration Aureobacterium sp. LS10 was incubated as described under 4a). As inducer, use was made of L-lactamide, DL-lactamide or N-methyl-L-lactamide at various concentrations, and after a good cell density had been reached, the enzymatic activity (hydrolysis of L-proline butyl ester) was determined. The enzymatic activity at 1 g/1 of L-lactamide was used as 1000.
Fig. 1 shows the enzymatic activity of Aureobacterium sp. LS10 during culture with various inducers at variable concentrations.
It can be seen from the results that only the L-enantiomer of lactamide has inducing activity, the optimum concentration being from 0.5 to 1.5 g/1. With increasing inducer concentration, the enzymatic activity decreased again, which is probably due to the inhibitory action of the ammonium released in the hydrolysis of the amide.
In contrast thereto, N-methyl-L-lactamide achieves an activity plateau, probably since the rate of hydrolysis is markedly ~decreased. N-Methyl-L-lactamide is however only active at higher concentrations than L-lactamide, but at the optimum (3.0 g/1) makes higher enzymatic activity possible.
Example 5:
Effect of various parameters on activity and stereo-selectivity of L-proline-ester-hydrolase from Aureobacterium sp. LS10 Aureobacterium sp. LS10 was incubated in the minimal medium described in Table 1 with glucose (10 g/1) as C source, L-lactamide (2 g/1) as N source and a little yeast extract (0.2 g/1). After an adequate cell density had been reached, the culture was harvested, the cells were washed and resuspended in common salt solution (0.90). Aliquots of this bacterial suspension were tested for their hydrolysis activity to L- or D-proline butyl ester, with various parameters being varied.
a) Activity and stereoselectivity For optimum racemate separation of a proline ester, in addition to an enzymatic activity as high as possible, a stereoselectivity as high as possible is also critical. Studies were therefore made of which parameters are of importance in this case. The experiments in which mean values were used for all parameters was taken to establish the 1000 value for enzymatic activity.
Table 5: Effect of pH, temperature and cell density on activity and stereoselectivity of L-proline-ester-hydrolase (E value - quotient of the activity towards the L-proline esters and the corresponding D-proline esters).
pH Temp. Cell Relative volumetric E value [C] density activity with [OD6sol L-ester L$1 - -5..5 25 15 9 15 -..

5. 5 25 75 39 25 5.5 40 15 42 18 5.5 40 75 166 21 6.3 35 45 100 9 7.0 25 15 24 8 7.0 25 75 97 g 7.0 40 15 62 9 7.0 j 40 75 237 9 The enzymatic activity of L-proline-ester-hydrolase is substantially temperature-dependent. A
temperature increase of 15°C leads to about a quadrupling of activity. Increasing pH likewise leads to increased activity, but only to a relative small extent. Increasing the cell density achieves a virtually proportional increase in activity. The stereoselectivity, in contrast, is substantially temperature-independent. The most important parameter found in this case is the pH, which should be as low as possible. A high cell density likewise contributes slightly to enhanced selectivity. This experiment implies that at low pH, high cell density and high temperature, the best compromise of enzymatic activity and stereoselectivity may be achieved.
b) Product inhibition Since in the hydrolysis of L-proline butyl ester by Aureobacterium sp. LS10 L-proline-ester hydrolase, indications of inhibition of enzyme activity by the butanol released in the reaction were found, experiments were carried out to study the effect of various parameters on this inhibition. As reference, an experiment was carried out without butanol and the enzymatic activity determined here was used as 1000.
Table 6: Effect of temperature, cell density and butanol concentration on the activity of L-proline ester-hydrolase.

Temp. Cell Butanol Relative volumetric [C] density concen- activity with L-ester [OD65o] tration [$]
]

.

The cell density proved to be the most important parameter for butanol inhibition of L-proline-ester-hydrolase. At a low cell density, even 100 mM butanol leads to complete inactivation of the enzyme, whereas at high cell density, higher butanol concentrations are also less harmful. With increasing butanol concentration the inhibitory activity increases, but only to a relatively small extent.
Likewise, a temperature increase leads to an increased inactivation of the enzyme. To ensure complete conversion of L-proline ester in the specific hydrolysis with activity as high as possible, the pH
must therefore be low, the cell density high and the temperature low.
Example 6:
Effect of various alcohols on the activity of Aureobacterium sp. LS10 L-proline-ester-hydrolase Since butanol inhibits the enzymatic activity of L-proline-ester-hydrolase (cf. Example 5), the inhibitory action of various alcohols was tested. For this purpose, Aureobacterium sp. LS10 was incubated in the minimal medium described in Table 1 with glucose (10 g/1) as C source, L-lactamide (2 g/1) as N source and a little yeast extract (0.2 g/1). After an adequate cell density had been reached, the culture was harvested. After washing and resuspending the cells in common salt solution (0.90), aliquots of the bacterial suspension were incubated at room temperature with various alcohols at various concentrations for 30 minutes. The enzymatic activity was then determined.
Fig. 2 shows the effect of various alcohols on the activity of L-proline-ester-hydrolase.
This experiment shows that, except for methanol, all alcohols tested have an adverse effect on the enzymatic activity. With increasing concentration, the enzyme activity decreases continuously, in the extreme case even as far as zero. In this case, in addition, a trend is observable, that this adverse effect becomes more intense with increasing hydrophobicity of the alcohols.
Example 7:
Hydrolysis of various proline esters by L-proline-ester-hydrolase from Aureobacterium sp. LS10 The activity and stereoselectivity of the L-proline-ester-hydrolase towards various proline esters was investigated as a function of pH.
In this example, esters were selected whose stability in water is sufficiently high so that the ee value is not markedly impaired by non-specific chemical hydrolysis, and which are hydrolysed to form alcohols which have as low as possible an adverse effect on the enzymatic activity (cf. Example 6). For this purpose, Aureobacterium sp. LS10 was incubated in the minimal medium described in Table 1 with glucose (10 g/1) as C
source, L-lactamide (2 g/1) as N source and a little yeast extract (0.2 g/1). After an adequate cell density had been reached, the culture was harvested, the cells were washed and resuspended in common salt solution (0.90). The enzymatic activity towards various L- and D-proline esters was then determined. The 1000 value used was the hydrolysis of L-proline butyl ester at pH
7Ø The stereoselectivity of the enzyme was determined on the basis of the quotient of the activity towards the L-proline esters and the corresponding D-proline esters (E value).
Fig. 3 shows the enzymatic activity of the L-proline-ester-hydrolase with various L-proline esters as a function of pH.
Table 7: Stereoselectivity (E value) of the L-proline ester-hydrolase with various proline esters as a function of pH.
pH
5.4 6.2 7.0 Proline ethyl ester - 41 60 Proline propyl ester 31 46 56 Proline isopropyl ester >100 >100 >100 Proline butyl ester 81 68 33 With all four proline esters tested, the enzymatic activity increases with increasing pH. The hydrolysis of L-proline butyl ester is markedly slower than the hydrolysis of the three remaining esters. For all four esters, the L-enantiomer is hydrolysed markedly more rapidly than the D-enantiomer, so that good to very good E values are observable, the selectivity being able to change slightly as a function of pH, depending on the ester used.
Example 8:
Hydrolysis of proline isopropyl ester at varying substrate concentration A study was to be made of which product concentrations could be achieved with proline isopropyl ester, which had proven to be an expedient substrate.
For this purpose, Aureobacterium sp. LS10 was incubated with the minimal medium described in Table 1, with glucose (10 g/1) as C source, L-lactamide (2 g/1) as N

source and a little yeast extract (0.2 g/1). After an adequate cell density had been reached, the culture was harvested. The cells were then washed in common salt solution (0.90) and finally resuspended therein also.
The test was carried out with DL-proline isopropyl ester as substrate at various concentrations, while the other parameters were kept constant (25°C, pH 6.0, OD6so approximately 15). At various time points, aliquots were taken out and analysed by HPLC for their content of D- or L-proline isopropyl ester.
Fig. 4 shows the biotransformation by Aureobacterium sp. LS10 with variation in substrate concentration.
At 60 or 120 g/1 of DL-proline isopropyl ester, a very rapid hydrolysis of the L-ester was observed, which resulted in the formation of D ester at an ee value of greater than 980. At higher substrate concentrations (240 g/1 and 300 g/1), the rate of hydrolysis decreased markedly. With increasing concentration of isopropanol released, the reaction became increasingly slower, so that complete hydrolysis of the L-proline isopropyl ester became more difficult.
High ee values were only observed after a relatively long incubation time, in which case the yield of D-ester slightly decreased, however. As an alternative, the cell density was doubled in the assay batch (OD6so approximately 30), which, at 240 g/1 of DL-proline isopropyl ester, caused complete conversion of the L ester after about 4 h and gave an ee value of greater than 980 of the remaining D-proline isopropyl ester.
Example 9:
Preparation of D-proline using the L-proline-ester-hydrolase from Aureobacterium sp. LS10 a) Culture of Aureobacterium sp. LS10 on minimal medium Aureobacterium sp. LS10 was incubated in a Chemap fermenter (working volume 5 1) in minimal medium (cf. Table 1) with glucose (30 g/1) and L-lactamide (4 g/1) as C and N sources at 30°C. To facilitate cell growth, the medium also contained a small amount of yeast extract (0.7 g/1). During culturing, in accordance with the cell requirement, further L-lactamide was fed, until a sufficiently high cell density (OD6so approximately 23) and a high specific enzyme activity (>1.2 ~.mol of L-proline butyl ester hydrolysed/min x OD6so) were reached. The cells were separated off by centrifugation, washed in common salt solution (0.90) and finally resuspended therein.
401.4 g of DL-proline isopropyl ester were added to the bacterial suspension and the volume was set to 4.5 1 by adding water. This biotransformation assay batch was then incubated for 160 minutes at 25°C and pH 6.0 with stirring. The cells were then separated off by centrifugation and the residue was worked up. By repeated electrodialysis at pH 6.3, the D-proline isopropyl ester was first separated off here from L-proline. In the next step, D-proline was released from the ester by alkaline hydrolysis using sodium hydroxide solution and the common salt formed was separated off by electrodialysis (pH 6.3). The resultant D-proline solution was concentrated to dryness and an impurity due to a yellow dye was removed by washing the D-proline crystals with ethanolabsoiute (30 minutes at 60°C).
In this manner, ultimately, 64.0 g of D-proline (53.70 of theory) were isolated, which had a very high content (>99o by titrimetry) and also an excellent optical purity (ee > 99o by HPLC, [a]p2° - 83.7 for c = 4 in acetic acid).
b) Culture of Aureobacterium sp. LS10 on complete medium Aureobacterium sp. LS10 was incubated in a Chemap fermenter (working volume 2 1) in mineral medium (cf. Table 1) at 30°C with glucose (20 g/1) as C source and yeast extract (15 g/1) as N source. In addition, the medium comprised N-methyl-L-lactamide (3 g/1) as inducer. After an adequate cell density had been reached (OD6so approximately 30), the cells were harvested by centrifugation.
After washing and resuspension in common salt solution, the cells showed high activity (approximately 1.1 ~mol of L-proline butyl ester hydrolysed/min x OD6so) and some of them (OD6so - 21 for V = 1.15 1) were used for racemate separation of 228.2 g of DL-proline isopropyl ester. After 5.5 h, the hydrolysis of the L-proline ester was complete and the cells were separated off by centrifugation. 900 of the D-proline ester used was still detectable in solution, a very high optical purity (ee value - 99o by HPLC) being observed.
Example 10 Chemical syntheses 10.1 Racemization of L-proline To produce racemic proline, 160 g (1.39 mol) of L-proline were dissolved in 600 ml (10.49 mol) of acetic acid and 20 g (277 mmol) of butanal were added.
This solution was heated to 100°C in an autoclave and kept at this temperature for two hours. After cooling, the solution was concentrated as far as possible and racemic proline was crystallized by adding 700 ml of acetone. After drying the crystals, 144.4 g of proline were obtained here (90.30 of theory), the identity of which was confirmed by NMR spectroscopy and was virtually racemic ([a]D24 - -0.5°C for c - 5 in 1N
HC1).
As an alternative to this, proline could also be racemized in alkaline aqueous solution (cf. as part of the one-pot reaction in 10.3).
10.2 Esterification of DL-prloline To prepare racemic proline isopropyl ester, 300 g (2.61 mol) of DL-proline and 1000 ml (13.06 mol) of isopropanol were placed in a three-neck flask. This mixture was cooled to 10°C and 380 ml (5.22 mol) of thionyl chloride were then slowly added dropwise, keeping the temperature constant. After completion of the addition, the mixture was firstly stirred for 90 min at room temperature and then the solution was heated under reflux for some hours. After cooling and distilling off the remaining solvent, 521.9 g of a brown oil were obtained, HPLC analysis of which showed that a conversion rate of 95o was achieved.
Using the same procedure, other proline esters such as proline methyl ester, proline ethyl ester and proline butyl ester, as well as the corresponding esters N-methylproline, were also prepared.
10.3 One-pot reaction for preparing DL-proline ester Using the racemization of L-proline in aqueous alkaline solution, DL-proline isopropyl ester can also be prepared from L-proline in a one-pot process. For this purpose, 80 g (695 mmol) of L-proline were dissolved in 26 ml of 4N NaOH (104 mmol) and 148.5 g of water. This solution was then heated to 160°C in an autoclave, where a pressure of about 5 bar was reached, and was kept at this temperature for 14 hours. The water was then removed as far as possible by distillation, with use being made of the azeotropic distillation effect after addition of 250 ml of isopropanol, in order to remove residual amounts of water. After isopropanol had also been removed as well as possible by distillation, 78.5 g (2.43 mol) of isopropanol were added and the mixture was cooled to 10°C. Maintaining a temperature of 5-10°C, 103.5 g (0.87 mol) of thionyl chloride were then added dropwise. The assay batch was then slowly heated to 84°C in the course of 2.5 hours. After filtering off a white precipitate (NaCl) and distilling off the solvent, 138.4 g of a yellowish oil were obtained.
According to HPLC analysis, about 700 of the L-proline used were then in the form of DL-proline isopropyl ester.
10.4 Preparation of N-methylproline To prepare N-methylproline, 115.1 g (1.0 mol) of L-proline were dissolved in 375.8 g of 98o formic acid (8.0 mol) and then 329.1 g of 36.50 formaldehyde (4.0 mol) were then slowly added dropwise.
The reaction took place thereafter for four hours at 60°C. After separating off the solvent by distillation, there remained 184.5 g of an orange oil. This was dissolved in 100 ml of water and the pH adjusted to 6.3 (pI) by NaOH, so that N-methyl-L-proline was present as a zwitterion. The water was removed by distillation and the residue was taken up in 180 ml of ethanol. Water residues were removed by azeotropic distillation. The residue was then slurried in 500 ml of ethanol and kept for some hours at 8°C. Insoluble components were then filtered off and the filtrate was concentrated to dryness.
After redissolution in 120 ml of ethanol, N-methyl-L-proline was crystallized out by adding 95 ml of ether. After separating off and drying the crystals, 74.6 g of N-methyl-L-proline remained, the identity of which was confirmed by GC analysis.
10.5 Synthesis of N-methylprolineamide In order to prepare the corresponding amide from N-methylproline, 12.9 g (100 mmol) of N-methyl-L-proline and 100 ml (2.47 mol) of methanol were cooled to 0°C and 14.3 g (120 mmol) of thionyl chloride were slowly added dropwise, with the temperature being maintained. After the mixture had been stirred at room temperature for about 90 min, the assay batch was heated to reflux for 2.5 hours. After filtering off an insoluble impurity and concentration to dryness, there remained 15.1 g of an oily residue. Half of this was added to 20 ml of 50o K2C03 ~(0°C) and the N-methyl-L-proline methyl ester, which is uncharged under these conditions, was extracted with a total of 60 ml of ether. The residue, 4.1 g, remaining after concentration to dryness, was dissolved in a total of 60 ml of ammoniacal methanol and heated (40-80°C) in an autoclave for seven days. The residue obtained after the concentration to dryness was then dissolved in 20 ml of methanol, an insoluble impurity was separated off by filtration and the filtrate was finally concentrated to dryness. In the course of this, 3.0 g of N-methyl-L-prolinamide were produced, the identity of which was confirmed by GC analysis.

Claims (8)

claims:

1. Microorganisms of the genera Aureobacterium, Aeromonas or Serratia, characterized in that they are able to utilize proline derivatives, in the form of the racemate or optically active isomers, selected from the compounds of the general formula in which R1 is alkyl, acyl or hydrogen, R2 is hydrogen or hydroxyl and R3 is -NH2, aryloxy or alkoxy as sole source of nitrogen, as sole source of carbon or as sole source of carbon and nitrogen, and enzyme extracts thereof.

2. Microorganisms according to Claim 1 of the species Aureobacterium sp. LS10 (DSM 10203) and its functionally equivalent variants and mutants.

3. Enzyme having L-proline ester hydrolase and/or L-prolinamide hydrolase activity, obtainable from microorganisms according to one of Claims 1 or 2 and capable of hydrolysing proline derivatives, in the form of the racemate or optically active isomers, selected from the compounds of the general formula in which R1, R2 and R3 are as defined, and functionally equivalent variants and mutants thereof.

4. Process for preparing D-proline derivatives of the general formulae and/or in which R1, R2 and R3 are as defined in Claim 1, comprising the reaction of a proline derivative of the general formula in which R1, R2 and R3 are as defined, by means of a microorganism and/or enzyme preparation according to Claim 1 or of an enzyme according to Claim 3, to give an L-proline derivative of the general formula III, and, if desired, isolation of this compound and/or of the D-proline derivative of the formula II obtained in this reaction.

5. Process according to Claim 4, characterized in that the proline derivative of the general formula in which R1, R2 and R3 are as defined is prepared by, in a first stage, racemizing an L-proline derivative of the general formula in which R2 is as defined, to give a proline derivative of the general formula in which R2 is as defined and, in the second stage, converting this derivative into a proline derivative of the general formula in which R1, R2 and R3 are as defined.

6. Process according to Claim 4 or 5, characterized in that the reaction of the proline derivative of the general formula I is carried out by means of microorganisms of the species Aureobacterium sp. LS10 (DSM 10203) or with its functionally equivalent variants and mutants.

7. Process according to Claims 4 to 6, characterized in that the biotransformation is conducted at a pH of from 4 to 10 and at a temperature of from 10 to 50°C.

8. Process for preparing a D-proline derivative of the general formula in which R1 and R2 are as defined in Claim 1, characterized in that, in a first stage, an L-proline derivative of the general formula in which R2 is as defined is racemized to the corresponding proline derivative of the general formula in which R2 is as defined, this derivative in a second stage is converted into a proline derivative of the general formula in which R1, R2 and R3 are as defined and this latter derivative, in a third stage, is converted by means of a microorganism and/or enzyme preparation according to Claim 1 or an enzyme according to Claim 3 into an L-proline derivative of the general formula the biotransformation producing not only the L-proline derivative of the general formula III but also the D-proline derivative of the general formula which in a fourth stage is hydrolysed to give the product of the formula VI.