CA2254902A1 - Preparation of lactams from aliphatic .alpha.,.omega.-dinitriles - Google Patents

Abstract

A process for the preparation of five-membered or six-membered ring lactams from aliphatic .alpha.,.omega.-dinitriles has been developed. In the process an aliphatic .alpha.,.omega.-dinitrile is first converted to an ammonium salt of an .omega.-nitrilecarboxylic acid in aqueous solution using a catalyst having an aliphatic nitrilase (EC 3.5.5.7) activity, or a combination of nitrile hydratase (EC 4.2.1.84) and amidase (EC 3.5.1.4) activities. The ammonium salt of the .omega.-nitrilecarboxylic acid is then converted directly to the corresponding lactam by hydrogenation in aqueous solution, without isolation of the intermediate .omega.-nitrilecarboxylic acid or .omega.-aminocarboxylic acid. When the aliphatic .alpha.,.omega.-dinitrile is also unsymmetrically substituted at the .alpha.-carbon atom, the nitrilase produces the .omega.-nitrilecarboxylic acid ammonium salt resulting from hydrolysis of the .omega.-nitrile group with greater than 98 % regioselectivity, thereby producing only one of the two possible lactam products during the subsequent hydrogenation. A heat-treatment process to select for desirable regioselective nitrilase or nitrile hydratase activities while destroying undesirable activities is also provided.

Description

CA 022~4902 1998-11-13 W O 97144318 PCTrUS97/07796 TITLE
PREPARATION OF LACTAMS FROM ALIPHATIC a.~-D[NITRILES
l . Field of the Invention:
This invention relates to a process for the preparation of five-membered or 5 six-membered ring lactams from aliphatic a,~-dinitriles by a combination of biological and chemical techniques. More particularlv. an aliphatic (x.c~-dinitrile is first converted to an ammonium salt of an ~-nitrilecarboxylic acid in aqueoussolution using a catalyst having an aliphatic nitrilase (EC 3.5.5.7) activity, or a combination of nitrile hydratase (EC 4.2.1.84) and amidase (EC 3.5.1.4) activities.
10 The arnmonium salt of the c~)-nitrilecarboxylic acid is then converted directly to the corresponding lactam by hydrogenation in aqueous solution, without isolationof the intermediate c3-nitrilecarboxylic acid or cl)-aminocarboxylic acid. When the aliphatic a,cd-dinitrile is also unsymmetrically substituted at the a-carbonatom, the nitrilase produces the ~-nitrilecarboxylic acid ammonium salt res~lting 15 from hydrolysis of the c~)-nitrile group with greater than 98% regioselectivity, thereby producing only one of the two possible lactam products during the subsequent hydrogenation.

2. Description of the Related Art:
Nitriles are readily converted to the corresponding carboxylic acids by a 20 variety of chemical processes, but these processes typically require strongly acidic or basic reaction conditions and high reaction temperatures, and usually produceunwanted byproducts and/or large arnounts of inorganic salts as unwanted byproducts. Processes in which enzyme-catalyzed hydrolysis convert nitrile substrates to the corresponding carboxylic acids are often preferred to chemical25 methods, since these processes are often run at ambient temperature, do not require the use of strongly acidic or basic reaction conditions, and do not produce large amounts of unwanted byproducts. An additional advantage of the enzyme-catalyzed hydrolysis of nitriles over chemical hydrolysis is that, for the hydrolysis of a variety of aliphatic or aromatic dinitriles, the enzyme-catalyzed reaction can 30 be highly regioselective, where only one of the two nitrile groups is hydrolyzed to the corresponding carboxylic acid ammonium salt.
A nitrilase enzyme directly converts a nitrile to the corresponding carboxylic acid ammonium salt in aqueous solution without the interme~ te formation of an amide. The use of aromatic nitrilases for the hydrolysis of 35 aromatic nitriles to the corresponding carboxylic acid ammonium salts has been kno~n for many years, but it is only recent}y that the use of aliphatic nitrilases have been reported. Kobayashi et al. (Tetrahed~ron, (1990) vol. 46, 5587-5590;
J. Bacteriology, (1990), vol. 172, 4807-4815) have described an aliphatic nitrilase isolated from Rhodococcus rhodochrous K22 which catalyzed the hydrolysis of .. . .. . .

CA 022~4902 l998-ll-l3 W O 97/44318 PCTrUS97/07796 aliphatic nitriles to the corresponding carboxylic acid ammonium salts; several aliphatic a,~-dinitriles were also hydrolyzed, and glutaronitrile was converted to 4-cyanobutyric acid ammoniurn salt with 100% molar conversion using resting cells as catalyst. A nitrilase from Comamonas testosteroni has been isolated S which can convert a range of aliphatic a,co-dinitriles to either the corresponding ~-nitrilecarboxylic acid ammonium salt or the dicarboxylic acid ~ mmnnium salt (C~n~ n patent application CA 2,103,616 (1994/02/11); S. Lévy-Schil, et al., Gene, (1995), vol. 161, 15-20); for the hydrolysis of adiponitrile, a maximum yield of 5-cyanovaleric acid arnmonium salt of ca. 88% was obtained prior to complete conversion o~the 5-cyanovaleric acid arnmonium salt to adipic acid diarnmoniurn salt.
M. L. Gradley and C. J. Knowles (Biotechnology Lett., ( 1994), vol. 16, 41 -46) have reported the use of suspensions of Rhodococcus rhodochrous NCIMB 11216 having an aliphatic nitril~e activity for the hydrolysis of several 2-methylalkylnitriles. Complete conversion of (+/-)-2-methylbutyronitrile to 2-methylbutyric acid amrnonium salt was obtained, while the hydrolysis of (+/-)-2-methylhexanenitrile appeared to be enantiospecific for the (+)-enantiomer.
C. Bengis-Garber and A. L. Gutman (AppL Microbiol. Biotechnol., (1989), vol. 32, 11-16) have used Rhodococcus rhodochrous NCIMB 11216 as catalyst for the hydrolysis of several dinitriles. In this work, furnaronitrile and succinonitrile were converted to the co~ onding ~-nitrilecarboxylic acid ammonium salts, while gluL~u,lil,ile, adiponitrile, and pirnelonitrile were converted to the corresponding dicarboxylic acid diammoniurn salts.
A combination of two enzymes, nitrile hydratase ~NHase) and ~mifl~e, can be also be used to convert aliphatic nitriles to the corresponding carboxylic acid ammonium salts in aqueous solution. Here the aliphatic nitrile is initiallyconverted to an amide by the nitrile hydratase and then the arnide is subsequently converted by the amidase to the co~ ",onding carboxylic acid ammoniurn salt. A
wide variety of bacterial genera are known to possess a diverse spectrum of nitrile hydratase and ~mi~ e activities, including Rhodococcus, Pseudomonas, Alcaligenes, Arthrobacter, Bacillus, Bacteridium, Brevihacterium, Corynehacterium, and Micrococcus. Both aqueous suspensions of these microorg~ni.cmc and the isolated enzymes have been used to convert nitriles to arnides and carboxylic acid ammonium salts.
P. Honicke-Schmidt and M. P. Schneider (J. Chem. Soc., Chem. Commun., (19~0), 648-650) have used imrnobilized Rhodococcus sp. strain CH 5 to convert nitriles and dinitriles to carboxylic acid ammonium salts and c,)-nitrilecarboxylic acid ammonium salts, respectively. The cells contain both a nitrile hydratase and amidase activity which converts glutaronitrile to 4-cyanobutyric acid arnmoniurn CA 022~4902 1998-11-13 W O 97144318 PCTrUS97/07796 salt in 79% isolated vield based on 92% conversion of substrate. A. J. Blakely et al. (F~MS Microbioiogy Lett.. ( 1995), vol. 129, S7-62) have used the nitrile hydratase and amidase activity of suspensions of Rhodococcus AJ270 to regiospecifically hydrolyze malononitrile and adiponitrile to produce only the 5 corresponding ~-nitrilecarboxylic acid ammonium salts. H. Yamada et al.
(J. Fermen~. Technol., (1980), vol. 58, 495-500) describe the hydrolysis of glutaronitrile to a mixture of 4-cyanobutyramide, 4-cyanobutyric acid, glutaric acid and ammonia using Pseudomonas sp. K9, which contains both a nitrile hydratase and amidase. K. Yamamoto et al. (J. Ferment. Bioengineering, 1992, vol. 73, 125-129) described the use of Corynebacterium sp. CH S cells cont~inin~both a nitrile hydratase and amidase activity to convert trans- l ,4-dicyanocyclohexane to trans-4-cyanocyclohexanecarboxylic acid ammoniurn salt in 99.4% yield.
J. L. Moreau et al. (Biocatalysis, (1994), vol 10. 325-340) describe the 15 hydrolysis of adiponitrile to adipic acid, adipamide, and adipamic acid through the intermediate formation of S-cyanovaleric acid using Brevibacterium sp. R312 (nitrile hydratase and amidase activity). A. Kerridge et al. (Biorg Medicinal Chem., (1994), vol. 2, 447-455) report the use of Brevibacterium sp. R312 (nitrile hydratase and amidase activity) to hydrolyze prochiral 3-hydroxyglutaronitrile 20 derivatives to the corresponding (S)-cyanoacid ammonium salts. Eulo,veall Patent 178,106 B 1 (March 31, 1993) discloses selective transforrnation of one of the cyano groups of an aliphatic dinitrile to the corresponding carboxylic acid, amide, ester or thioester using the mononitrilase activity (defined as either nitrilase or a combination of nitrile hydratase/amidase) derived from Bacillus, Bacteridium, 25 ll~icrococcus or Bre~ibacterium. In addition to the many examples of bacterial catalysts having nitrilase activity or nitrile hydratasei:lmid~e activity, Y. Asano et al. (Agric. Biol. Chem., (1980), vol. 44, 2497-2498) demonstrated that the fungus Fusarium merismoides TG-I hydrolyzed glul~onil.;le to 4-cyanobutyric acid ammonium salt, and 2-methylglutaronitrile to 4-cyanopentanoic acid arnmonium 30 salt.
No prior art has been found which describes the hydrogenation of ammoniurn salts of aliphatic a)-nitrilecarboxylic acids in aqueous solution to directly produce the corresponding l~ct~mc In closely related art, U.S. Patent 4,329,498 describes the hydrogenation of muconic acid mononitrile to 35 6-aminocaproic acid (6-ACA) in dry ethanol saturated with ammonia, using a Raney nickel catalyst #2. After removal of the hydrogenation catalyst, heating the ethanolic solution of 6-ACA to 170~C-200~C was expected to result in the cyclization of 6-ACA to caprolactam. The reductive cyclization of either ,B-quinoxalinylpropanoic acids (E. C. Taylor et al., J. Am. Chem. Soc., (1965), CA 022~4902 1998-11-13 WO 97/44318 PCTrUS97/07796 vol. 87~ 1984-1990), or the related 2-(2-carboxvethyl)-3(4H)-quinoxalone (E. C.
Taylor et al., J. Am. Chem. Soc~, (1965), vol. 87, 1990- 1995) by hydrogenation in I N sodium hydroxide solution using Raney nickel as the catalyst has been reported to produce the corresponding five-membered ring lactams. but only after5 removal of the catalyst from the product mixture and acidification of the resulting filtrate. The authors state that for any of these reductions, "lactam formation can only proceed in acidic solution" (page 1992, second paragraph), presumably requiring the presence of the protonated carboxylic acid and not the carboxylatesalt. U.S. Patent 4,730,040 discloses a process for the p.cpdldlion of caprolactam, 10 reacting an aqueous solution of S-formylvaleric acid with ammonia and hydrogen in the presence of a hydrogenation catalyst, following which ammonia is separated from the product mixture and the resulting solution of 6-ACA is heated to 300~C.Previous work has disclosed single cells cont~ining both nitrile hydrata~.e and amidase activities that have been used to convert nitriles and dinitriles to15 various acid ammonium salts. However, no prior art has been found which describes the cyclization of ammonium salts of aliphatic c~-aminocarboxylic acids under the hydrogenation reaction conditions of the present invention (i.e., in an aqueous solution cont~ining an excess of added ammonium hydroxide) to produce the corresponding lactams. In closely related art, the cyclization of aliphatic 20 ~3-aminocarboxylic acids (but not the ammonium salts) to the corresponding l~t~mc under a variety of reaction conditions has been reported. F. Mares and D.Sheehan (Ind. L:ng. Chem. Process Des. Dev., (1978), vol. 17, 9-16) have described the cyclization of 6-arninocaproic acid (6-ACA) to caprolactam using water or ethanol as solvent. In water, the cyclization reaction was reversible at 25 concentrations below 1 mol/kg (ca. I M), and the concentration of caprolactamincreased with increasing telllp~,.dLllre; at a total concentration of 6-ACA andcaprolactam of 0.85 mol/kg (ca. 0.85 M), the percentage of caprolactam was reported to increase from 38.7% at 180~C to 92.2% at 250~C. In ethanol, a 98%
yield of caprolactarn was obtained at 200~C, reportedly due to a shift in the 30 equilibrium which favors the free-acid/free-arnine form of 6-ACA in ethanol, rather than the intramolecular alkylammoniurn carboxylate form of 6-ACA which predomin~tes in water. A process for the production of caprolactam from 6-ACA
is also described in U.S. Patent 4,599,199, where 6-ACA is introduced into a fluidized alumina bed in the presence of steam at from 290~C to 400~C. The 35 synthesis of five-, six- and seven-membered ring lactams by cyclodehydration of aliphatic ~-aminoacids on alumina or silica ge~ in toluene, and with continuous removal of the water produced during the reaction, has been reported by A. Bladé-Font (Tetrahedron Letters, (1980), vol. 21, 2443-2446). A free amino CA 022~4902 1998-ll-13 W O 97/44318 PCT~US97/07796 group (unprotonated) was reported to be necessary for cyclodehydration to take place.
No prior art has been found which describes the hydrogenation of ammonium salts of aliphatic cl~-nitrilecarboxylic acids in aqueous solution cont~ining methylamine to directly produce the corresponding N-methyl l~ct~
In closely related art. 1.5-dimethyl-2-pyrrolidinone was prepared by the hydrogenation of an aqueous solution of levulinic acid and methylarnine in waterusing a Raney nickel catalyst at 140~C and 1000-2000 psig of hydrogen R. L.
Frank et al., Org Sytheses, (1954), Coll. Vol. 3,328-329). The resulting 4-N-methylaminopentanoic acid methylarnmonium salt was then cyclized to the corresponding lactam by filtration of the product mixture and distillation of the filtrate to remove water and methylarnine. N-alkyl lactams have also been produced by the direct hydrogenation of an aqueous mixture Cont~ining 2-methylglutaronitrile, a primary alkylamine, and a hydrogenation catalyst, the process yielding a mixture of 1,3- and 1,5-dialkylpiperidone-2 (U.S. Patent 5,449,780). N-Substituted 2-pyrrolidinones have been prepared by the reaction ofy-valerolactone with an alkyl amine at 110-130~C, then heating the resulting mixture to 250-270~C while distilling off water (F. B. Zienty and G. W. Steahly,J. Am. Chem. Soc., (1947), vol. 69, 715-716).
The above processes for the production of lactams or N-alkyll~ct~rn~
suffer from one or more of the following disadvantages: the use of ten~p~.~L~Ires in excess of 250~C to obtain high yields of lactams when using water as a solvent, the removal of water from the reaction mixture to drive the equilibrium toward lactam formation, the adjustment of the pH of the reaction mixture to an acidic value to favor lactam formation, or the use of an organic solvent in which the starting material is sparingly soluble. Many of these processes generate undesirable waste streams, or mixtures of products which are not easily separated.
A significant advance would be a process for the conversion of an aliphatic a,cD-dinitrile to the corresponding lactarn or N-methyllactam in aqueous solution, in high yield with high regioselectivity, with little byproduct or waste stream production, and with a facile method of product recovery.
SUMMARY OF THE INVENTION
A process for the prep~d1ion of five-membered ring lactams or six-~ membered ring lactams from aliphatic a,~-dinitriles, having the steps:
(a) contacting an aliphatic a,~-dinitrile in an aqueous reaction mixture with an enzyme catalyst characterized by either (I) an aliphatic nitrilase activity, or (2) a combination of nitrile hydratase and arnidase activities, CA 022~4902 l998-ll-l3 W O 97/44318 PCTrUS97/07796 whereby the dliphatic a,cl)-dinitrile is converted to an cl~-nitrilecarboxylic acid ammonium salt;
(b) contacting the aqueous product mi~ture resulting from step (a) with hydrogen and a hydrogenation catalyst, whereby the ~-nitrile carboxylic acid S ammonium salt is converted directly to the corresponding lactam without isolation of the intermediate c3-nitrilecarboxylic acid, (~)-nitrilecarboxylic acid. ammonium salt, ~-aminocarboxylic acid, or ~-aminocarboxylic acid ammonium salt; and (c) recovering the lactam from the aqueous product mixture resulting from step (b).
Prior to step (b), ammonium hydroxide, ammonia gas, or methylamine may be added to the aqueous product mixture of step (a). This addition may be from 0 to 4 molar equivalents relative to the arnount of ~-nitrilecarboxylic acid arnmonium salt present.
A further embodiment of the invention uses an aliphatic o~,~-dinitrile with the formula NCCXa(R)(CH2)nCN, where a=0 or 1, X=hydrogen when a= 1, and R=H, unsubstituted or substituted alkyl, or unsubstituted or subs~ 3d alkenyl, or unsubstituted or substituted alkylidene, and where n= 1 or 2.
A further embodiment of the invention is a method for treating a whole cell catalyst to select for a regioselective nitrilase activity or nitrile hydratase activity capable of catalyzing the conversion of aliphatic o~,o~-dinitriles to the co.l~sl,ollding c~)-cyanocarboxylic acid ammonium salt. The whole cell catalyst to be treated is characterized by two types of activities: (1) a desirable regioselective nitrilase activity or regioselective nitrile hydratase activity and (2) an undesirable non-regioselective nitrilase or nitrile hydratase activity.
TledLIne-ll of the cell involves heating the whole cell catalyst to a t~ Je~Ule of about 35~C to 70~C for between 10 and 120 minlltes wherein the undesirable non-regioselective nitrilase activity or nitrile hydratase activity is destroyed and the desirable regioselective nitrilase or nitrile hydratase activity is preserved.
Further embo~iiment~ of the invention use enzyme catalysts in the form of whole microbial cells, perrneabilized microbial cells, one or more cell colllponellts of a microbial cell extract, and partially purified enzyme(s), or purified enzyme(s).
These enzyme catalysts can be immobilized on a support. Microorg~ni~m~ which are ch~d-;leliGed by an aliphatic nitrilase activity and useful in the process are Acidovoraxfacilis 72-PF-15 (ATCC 55747), Acidovoraxfacilis 72-PF-17 (ATCC 55745), and Acidovorax facilis 72W (ACC 55746). A microorganism characterized by a combination of nitrile hydratase and amidase activities and useful in the process is Comomonas testosteroni 5-MGAM-4D (ATCC 55744).

W O 97/44318 PCTrUS97/07796 A fi~rther embodiment is a process for the t~lcpa.~ion of five-membered ring lactams or six-membered ring lactams from aliphatic a,c~-dinitriles, comprising:
(a) cont~ctin~ an aliphatic a,cl)-dinitrile of either the formula:

R3 R~ R3 R4 where Rl and R2 are both H, and; R3,R4, R5 and R~ are each indep~n~i~ntly selected from the group consisting of H, unsubstituted or Cubsti~-t~d alkyl, or 10 unsubstituted or substituted alkenyl, in an aqueous reaction rnixture with anenzyme catalyst selected from the group con~i.cting of: Acidovoraxfacilis 72W
(ATCC 55746),Acidovorarfacilis 72-PF-15 (ATCC 55747),Acidovoraxfacilis 72-PF-17 (ATCC 55745), and Comamonas testosteroni 5-MGAM-4D
(ATCC 55744), whereby the aliphatic a,o~-dinitrile is converted to an 15 c~)-nitrilecarboxylic acid arnmonium salt;
(b) cont~cting the aqueous product mixture resulting from step (a) with hydrogen and a hydrogenation catalyst, wllc.~lJy the ~-nitrilecarboxylic acid ammonium salt is converted directly to the col.~,;,po,lding lactam ~,vithout isolation of the intermediate a)-nitrilecarboxylic acid, ~-nitrilecarboxylic acid 20 arnmonium salt, c,~-arninocarboxylic acid or c~-aminocarboxylic acid ammonium salt; and (c) recovering the lactam from the aqueous product mixture resulting from step (b).
A further embodiment is a process for the plep~alion of five-membered 25 ring lactams or six-membered ring lactams from aliphatic a,c3-dinitriles, compnsing:
(a) contacting an aliphatic a,~-dinitrile of either formula n R7~R8 R7~R8 Rll N ~ NC ~ Rl2 Rg Rlo R9 Rlo W O 97144318 PCTrUS97/07796 where R7, R8, R9, Rlo, Rl I and R12 are each independently selected from the group consisting of H, unsubstituted or substituted alkyl, or unsubstituted or substituted -alkenyl, and where either or both R7 or R8 is not H;
S in an aqueous reaction mixture with Comamonas testosteroni S-MGAM-4D
(ATCC 55~44), whereby the aliphatic a,cl)-dinitrile is converted to an cl~-nitrilecarboxylic acid ammonium salt;
(b) cont~tin~ the aqueous product mixture resulting from step (a) with hydrogen and a hydrogenation catalyst, whereby the ~-nitrilecarboxylic 10 acid ammonium salt is converted directly to the corresponding lactarn withoutisolation of the intermediate o~-nitrilecarboxylic acid, c~-nitrilecarboxylic acid ammonium salt, ~d-aminocarboxylic acid or ~-arninocarboxylic acid ammonium salt; and c) recovering the lactam from the a~ueous product mixture 15 resulting from step (b).
Finally, novel compounds are provided as follows:
a compound of Formula III

N 3 C_f_cH2co2 M
H
m where M+ is either H+ or NH4+p; and a compound of Formula IV, N c-c-cH2cH2co2 M
lV
where M+ is either H+ or NH4+.
BRIEF DESCRIPTION OF THE BIOLOGICAL DEPOSITS
Applicants have made the following biological deposits under the terms of the Budapest Treaty:

W O97/44318 PCTrUS97107796 Depositor Identific~tion Int'l. Depositorv ReferenceDesi(~"~liol- Date of Deposit Comomonas testosteroni S-MGAM-4D ATCC 55744 8 March t996 Acidovora~t facilis 72-PF- 17 ATCC 55745 8 March 1996 Acidovoraxfacilis72WATCC 55746 8 March 1996 Acidovor~Lr facilis 72-PF- 15 ATCC 55747 8 March 1996 As used herein, "ATCC" refers to the American Type Culture Collection international depository located as 12301 Par~lawn Drive, Rockville, MD 20852 U.S.A. The "ATCC No." is the accession number to cultures on deposit with the S ATCC.
DETAILED DESCRIPTION OF THE INVENTION
A process to prepare lactarns from aliphatic ~ dinitriles in high yields has been developed which utilizes a combination of enzymatic and chemical reactions. In cases where the ~,~-dinitrile is unsymmetrically substit-lte~i, high 10 regioselectivity to one of two possible lactam products (Scheme I ) is seen.
Scheme 1.

Nccxa(cH2)ncN ~ NCab(CH2)ncO2 NH4 ~ o n~n~se NH ~1 NH~e/~dase CH3NH2 where a = 0 or 1;
X = hydrogen when a = I;
n= I or2 R = H, unsubstituted or substituted alkyl, or unsubstituted or substituted alkenyl, or unsubstituted or substituted alkylidene; and Rl = H, -CH3 The products of the present invention are useful as precursors for polymers, solvents, and chemicals of high value in the agricultural and pharm~eutical industries. The process uses temperatures less than 250~C to obtain a high yield of lactam when using water as a solvent. Relative to previously known ~hlomir~l lactam plocesses, the claimed invention genel~les little waste and permits a facile appluach to product recovery.
In the application, unless specifically stated otherwise. the following abbreviations and definitions apply:
"Enzyme catalyst" refers to a catalyst which is characterized by either a nitrilase activity or a combination of a nitrile hydratase activity and an ~mirl~e activity. The catalyst may be in the form of a whole microbial cell, CA 022~4902 1998-11-13 W O 97/44318 PCT~US97107796 permeabilized microbial cell(s), one or more cell component of a microbial cell extract, partially purified enzyme(s), or purified enzyme(s).
"Hydrogenation catalyst" refers to a material that accelerates hydrogenation without itself being con~-lmP~I or undergoing a chemical change.
S "Aqueous product mixture" is used to refer to an aqueous mixture cont:~ining a product resulting from the corresponding process step.
A. Conversion of an aliphatic a,cl)-dinitrile to the corresponding ~-nitrilecarboxylic acid ammonium salt in high yield and with high regioselectivity .
The first step of this process is the conversion of an aliphatic a,c~-dinitrile to the corresponding cl~-nitrilecarboxylic acid arnmonium salt, using an enzyme catalyst. The enzyme catalyst has either a nitrilase activity, where the nitrilase converts the a,~-dinitrile directly to a corresponding cl)-nitrilecarboxylic acid ammonium salt (eqn. I ), or a combination of two enzyme activities, nitrile hydratase (NHase) and amidase, where the aliphatic a,a~-dinitrile is initially converted to a co-nitrilealkylamide by the nitrile hydratase, and then the c~
-nitrilealkylamide is subsequently converted by the amidase to the correspondingcl)-nitrilecarboxylic acid ammonium salt (eqn. 2):
nh~e ~
(I) R - CN ~ R ~
2H2O O-NH4+

NH~e ~ ~n~e ~
(2) R - CN r R ~ ~ R ~
H2O \NH2 H2o O-NH4+
A novel microbe Acidovorax facilis 72W (ATCC 55746) has been isolated from soil sarnples which had been exposed to aliphatic nitriles or dinitriles, and which could utilize 2-ethylsuccinonitrile as a nitrogen source. When used as a microbial whole-cell catalyst for the hydrolysis of unsymmetrically substituted a-alkyl-a,~-dinitriles such as 2-methylglutaronitrile (2-MGN) or 2-ethyl-succinonitrile (2-ESN), a mixture of products is obtained. Over the course of the hydrolysis reactions, the corresponding dicarboxylic acid mono~mi~es and dicarboxylic acids are produced in addition to the desired c3-nitrilecarboxylic acid.
It was discovered that heating a suspension of Acidovorax facilis 72W
(ATCC 55746) in a suitable buffer at 50~C for a short period of time deactivatesan undesirable nitrile hydratase activity of the whole-cell catalyst which catalyzed the production of the undesirable dicarboxylic acid monoamides (and which were further converted by an amidase to the corresponding dicarboxylic acid). In thismanner, a whole-cell catalyst is prepared which contains a nitrilase activity which CA 022~4902 1998-11-13 W O 97/44318 PCT~US97/07796 converts an a-alkyl-a,~l)-dinitrile to only the cl)-nitrilecarboxylic acid ammonium salt resulting from hydrolysis of the ~-nitrile group.
Heat-treatment of suspensions of Acidovorax facilis 72W (Al~CC 55746) at 50~C for one hour produces a microbial whole-cell catalyst which hydrolyzes - 5 2-methylglutaronitrile (2-MGN) to 4-cyanopentanoic acid (4-CPA) ammonium salt, 2-methyleneglutaronitrile (2-MEGN) to 4-cyano-4-pentenoic acid (4-CPEA) arnmonium salt, or 2-ethylsuccinonitrile (2-ESN) to 3-cyanopentanoic acid (3-CPA) ammonium salt with extremely high regioselectivity, such that at complete conversion of the dinitrile, at least a 98% yield of the ~-nitrilecarboxylic acid arnmonium salt is produced by hydrolysis ofthe cl)-nitrile group (Table 1):

w-nitrile/acid concentration yield a,cl~-dinitrile ammoniumsalt (M) (%) 2-MGN 4-CPA(NH4+) 0.10 99.3 " " 0.40 99.4 " " 1.00 99.4 " " 1.85 98.8 ~' " 2.00 98.9 2-MEGN 4-CPEA(NH4+) 1.25 100 '~ " 2.00 100 2-ESN 3-CPA(NH4+) 0.10 100 " " 0.40 100 " " 1.00 100 " " 1.25 100 There are currently no non-enzymatic methods for the selective hydrolysis of only one nitrile group of an aliphatic dinitrile to either an amide group or a 15 carboxylic acid group at complete conversion of the dinitrile. If such a reaction is run to incomplete conversion (< 20% conversion) in order to obtain a high selectivity to a monoamide or monoacid hydrolysis product, a separation step is then required to isolate the product from unreacted dinitrile, and for recycle of dinitrile into a subsequent reaction. Non-enzymatic hydrolysis reactions also 20 typically involve heating solutions of the nitrile or dinitrile at elevated temperatures, often times in the presence of strong acid or base, while the enzyme-catalyzed reaction described above are carried out at arnbient temperature in aqueous solution and at neutral pH with no added acid or base.
Two mutants of the Acidovorax facilis 72W (ATCC 55746) strain have 25 been plep~d which produce only very low levels of the undesirable nitrile .

CA 022~4902 1998-ll-13 W O 97/44318 PCT~US97/07796 hydratase activity responsible for the forrnation of undesirable byproducts. These mutant strains. Acidovorax facilis 72-PF- 15 (ATCC 55747) and Acidovorax facilis72-PF- 17 (ATCC 55745), do not require heat-treatment of the cells prior to use as catalyst for the hydrolysis of an aliphatic a~-dinitrile to the corresponding S ammonium salt of a cl)-nitrilecarbo~ylic acid. A comparison of the yields of 4-CPA and 2-methylglutaric acid (2-MGA) produced by the hydrolysis of 2-MGN
using untreated and heat-treated Acidovorax facilis 72W (ATCC 55746), and the untreated Acidovoraxfacilis 72-PF-15 (ATCC 55747) mutant strain are shown in Table 2:

[2-MGN] 4-CPA 2-MGA
catalyst (M) (% yield) (% yield) A. facilis 72W, untreated 0.10 62.7 34.6 A. facilis 72W, heat-treated0.10 99.3 0.7 A. facilis 72-PF-15, untreated0.10 96.8 3.6 A. facilis 72W, heat-treated0.40 99.4 0.6 A. facilis 72-PF-15, untreated0.40 98.8 1.2 A. facilis 72W, heat-treated1.00 98.7 1.3 A. facilis 72-PF-15, untreated1.00 99.2 0.8 When heat-treated Acidovoraxfacilis 72W (ATCC SS746) is used as a catalyst for the hydrolysis of aqueous solutions of the unsubstituted aliphatic a,~-dinitriles succinonitrile (SCN, 1.25 M ) or glutaronitrile (GLN, 1.5 M), the15 corresponding c3-nitrilecarboxylic acid ammoniurn salts 3-cyanopropionic acid(3-CPRA) and 4-cyanobutyric acid (4-CBA) are produced in yields of 99.7% and 92.3%, respectively, and the col,es~onding dicarboxylic acids are the only observed byproducts. When this same catalyst is used to convert adiponitrile (ADN) to S-cyanopentanoic acid (S-CPA) ammoniurn salt, adipic acid (ADA) 20 rli~mmonium salt is the major product (> 50% yield). Neither ~cidovorax facilis 72W (ATCC 55746) nor Acidovoraxfacilis 72-PF-15 (ATCC 55747) are suitable as catalyst for the plep~dlion of S-CPA arnmoniurn salt in high yield.
More than 30 different microbial cultures isolated from soil samples which had been exposed to aliphatic nitriles or dinitriles, and which could grow on 25 various nitriles or amides as nitrogen source, were screened for high selectivity for S-CPA production. A second novel microbe, Comamonas testosteroni S-MGAM-4D (ATCC 55744), was isolated (using 2-methylglutaramide as nitrogen source) which contained several nitrile hydratase and amidase activities.
When used as a whole cell catalyst for the hydrolysis of ADN, the resulting CA 022~4902 1998-ll-13 W O 97/44318 PCTrUS97/07796 product mixture is composed primarily of ADA diammonium salt, adipamide (ADAM) and adipamic acid (ADMA), with only a minor yield of S-CPA
ammonium salt observed. It was again found that heating the microbe at 50~C for a short period of time deactivated an undesirable nitrile hydratase activity to a - 5 great extent~ leaving the microbial catalyst with a nitrile hydratase activity which converts ADN to S-cyanovaleramide, and an amidase which converts S-cyanovaleramide to 5-CPA ammonium salt. Heat-treatment of Comamonas testosteroni S-MGAM-4D at 50~C for one hour results in a microbial cell catalystwhich produces 5-CPA in yields as high as 97% at complete conversion of ADN.
The ability to elimin~te unwanted nitrile hydratase by heat trP~tmçnt, while at the same time leaving a relatively heat-stable nitrilase activity or nitrile hydratase activity for the conversion of dinitriles to the cyano acids was not previously known and could not have been predicted because the telllp~lalule stability of the nitrilase or nitrile hydratase enzyme was unkno~vn. It is expected that heat treatment at temperatures between 35~C and 70~C for between lO and 120 minutes will produce the described useful effect.
B. P,~;p~dtion of five- or six-membered ring lactarns.
A method has been discovered for the ple~,alation of five-membered ring or six-membered ring lactams in high yields by the direct hydrogenation of the ~-nitrilecarboxylic acid ammonium salt product mixture produced by the enzyme-catalyzed hydrolysis of aliphatic a,(o-dinitriles in aqueous solution. This method does not require the isolation of the c~-nitrilecarboxylic acid ammonium salt from the product mixture of the hydrolysis reaction prior to the hydrogenation step, nor does it require the conversion of the cl~-nitrilecarboxylic acid ammonium salt to the free acid (e.g., conversion of 4-CPA ammonium salt to 4-CPA) prior to hydrogenation, or isolation of the resulting ~-aminocarboxylic acid ammonium salt from the hydrogenation product mixture and conversion of the ammonium salt to the free carboxylic acid prior to cyclization.
After producing an aqueous product mixture containing the amrnonium salt of a ~-nitrilecarboxylic acid from an aliphatic a,~-dinitrile by using an enzyme catalyst (eqn. 3), removal of the enzyme catalyst and reaction of the resulting aqueous solution with hydrogen and a stoichiometric excess of added arnmonia (as ammonium hydroxide) in the presence of a suitable hydrogenation catalyst was expected to produce an aqueous solution containing an aliphatic o~-aminocarboxylic acid ammonium salt (eqn. 4):

.. . .

CA 022~4902 1998-11-13 W O 97/44318 PCT~US97107796 R R

(3) NCCH(CH~)nCN H20 . NCC!H(CH2)nCO~ N~l+
nltnlase or NHase/amidase R R

(4) NC(~H(CH2)nC02 NH4+ H2 . H~NCH~H(CH~)nCO2 NH4+
~6 NH~OH
The use of an excess of ammonia during the hydrogenation of a nitrile to the corresponding arnine is necessary to limit reductive alkylation of an imine intermediate (produced during the hydrogenation of the nitrile group to an amine) 5 by the product ~-aminocarboxylic acid, which results in dimer forrnation and yield loss. This technique is well-docl-m~nt~l (De Bellefon et al., Catal. Rev. Sci.
Eng., (1994), vol. 36, 459-506) and is commonly practiced by those skilled in the art of hydrogenation of nitriles.
Based on the prior art, it was expected that the c~-aminocarboxylic acid 10 ammonium salt produced by the hydrogenation of a cl~-nitrilecarboxylic acid arnmonium salt would have to be converted to the free acid and isolated (eqn. S)before a th~ ly induced cyclization reaction to produce the desired lactarn could be performed (eqn. 6). According to Mares et al. (supra), 6-aminocaproic acid (6-ACA) (eqn. 6, n=3, R=H) and caprolactarn exist as a reversible 15 equilibrium mixture at concentrations of less than l.0 mol/kg (ca. l .0 M) in water and the concentration of caprolactam increases with increasing ttl,lp~ld~u~e. At a total concentration of 6-ACA and caprolactam of 0.85 mol/kg (ca. 0.85 M), the percentage of caprolactam was reported by Mares et al. to increase from 38.7% atl 80~C to 92.2% at 250~C.
R R

I HCI

(5) H2NCH2CH(CH2~CO2-NH4+ ~ H2NCH2CH(CH2~cO2H

(-NH4C~

I ~at R ~ (CH2~

(6) H2NCH2CH(CH2~CO2H ~ ~ O

(-H2O) - N

H
In the present case, where an excess of added arnmonia (as amrnonium hydroxide) is present and the pH of the reaction mixture is between pH 9 and pH 10, it was not expected that the amrnoniurn salt of 6-ACA would cyclize to produce significant amounts of caprolactam at hydrogenation tempel~lures of less25 than 200~C. The pKa's of the carboxylic acid group and the protonated arnine CA 022~4902 1998-11-13 wo 97/44318 PCT/USg7/07796 group of 6-aminocaproic acid are 4.373 and 10.804, respectively (Lange's Handbook of Chemistry, J. A. Dean, ed., 14th edn., (1992), McGraw-Hill~ NY, p. 8.22 (as 6-aminohexanoic acid)), and the pKa of NH4+ is 9.25. At a reaction mixture pH of between 9 and 10, it can be calculated that at least 99.997% of the 5 6-ACA exists in solution as the amrnonium salt of the carboxylic acid, and additionally, approximately half of the arnine groups of the 6-arninocaproic acid ammonium salt are also protonated. Therefore, it was not expected that significant arnounts of the 6-ACA arnmoniurn salt would cyclize to produce caprolactam under the hydrogenation conditions of the present invention, where 10 the displacement of a hydroxyl anion (-OH) from the cyclic reaction interrnediate represented in Equation (7) is not favored:

R R ~ CH2)n + R ~ CH2~

(7) H2NcH2cH(cH2)nco2-NH4~ N k o- ~ ~

H2+ , H
When hydrogenations of aqueous solutions of 5-CPA arnrnonium salt (prepared by enzymatic hydrolysis of ADN) are performed in the presence of from 0 M to 2.0 M NH40H at temperatures of from 70~C to 160~C, complete conversion of 5-CPA to 6-ACA amrnonium salt was observed with little byproduct formation, and, as predicted, less than 3% yields of caprolactarn (theresulting seven-membered ring lactam) are obtained (Table 3):

Temp. [NH40H] tune 5-CPA caprolactam (~C)S-CPA ammonium salt(M) (M)(h) conv. (%) (%) 1 .0 0 2 1 00 0.7 1.0 1.0 2 100 0.7 1.0 1.5 2 94 0.8 1.0 2.0 2 84 0.8 120 1.0 0 2 99 0.9 120 1.0 1.0 2 100 0.9 120 1.0 I.S 2 97 1.0 120 1 .0 2.0 2 97 1 . I
160 1.0 0 2 97 2.5 160 1.0 1.0 2 84 2.3 160 1.0 1.5 2 97 3.0 160 1.0 2.0 2 95 2.7 Yields of caprolactarn of less than 3% are also obtained when aqueous solutions prepared by mixing ~lth~l~tic 6-ACA and ammonium hydroxide are CA 022~4902 1998-11-13 W O 97/44318 PCT~US97107796 treated under the same hydrogenation reaction conditions as for hydrogenation ofthe arnmonium salt of 5-CPA prel)aled by enzymatic hydrolysis of ADN. The cyclization of the ammonium salt of 6-ACA in aqueous solutions prepared by the hydrogenation of 5-CPA ammonium salt at 70~C was attempted at tempelat~.es 5 greater than 200~C by first removing the hydrogenation catalyst, then heating the ca. l.0 M 6-ACA amrnonium salt product mixtures at 280~C for 2 h. Yields of caprolactam of less than 18% were produced (Table 4). These results dernonstratethat only very low yields of caprolactam are obtained by the direct hydrogenation of aqueous solutions of 5-CPA ammonium salt in the presence of excess 10 ~mmoni~ This was expected in light of the predicted inability of the 6-ACA
ammonium salt to undergo cyclization, particularly in the presence of excess ammonium hydroxide.

Temp. 6-ACA ~~ lsalt ~IH4OH] time (~C) (M) (M) (h) 280 1.0 0 2 17.8 280 1.0 1.0 2 17.2 280 1.0 1.5 2 11.0 280 1.0 2.0 2 9.0 280 1.0 2.5 2 3.1 It was also expected that when aqueous mixtures of 3-cyanopentanoic acid (3-CPA) amrnonium salt (prepared by enzymatic hydrolysis of 2-ethylsuccinonitrile (2-ESN)) or 4-cyanovaleric acid (4-CPA) ammoniurn salt (prepared by enzymatic hydrolysis of 2-methylglutaronitrile (2-MGN)) were hydrogenated at from 70~C to 1 60~C in the presence of excess ammonia, the 20 corresponding cl)-arninocarboxylic acid ammonium salts would be produced, andthat these salts would have to be isolated as the free acids before the cyclization reaction to the corresponding lactarn could be perfor ned (as was the case for 5-CPA ammonium salt). Although the pKa's of the carboxylic acid and the protonated arnine functionalities of the product 4-amino-3-ethylbutyric acid or 25 5-amino-4-methylpentanoic acid have not been reported, it is reasonable to assume that these pKas are similar to those of the 6-ACA isomer.
Unexpectedly, hydrogenation of aqueous solutions of 3-CPA ammonium salt (produced by enzymatic hydrolysis of 2-ESN) in the presence of Raney nickeland excess ammonia at pH 9-10 and at temperatures of from 70~C to 1 80~C for 30 2 h produce the corresponding lactam 4-ethylpyrrolidin-2-one (4-EPRD) directly, and at yields of up to 91% (Table 5). The yield of 4-EPRD also increases with CA 022=,4902 1998-11-13 increasing concentration of added arnmonium hydroxide (added in addition to the arnrnonium ion concentration already present as the arnmonium salt of the carboxylic acid), which in(li~te~ the desirability of performing the hydrogenation of aqueous solutions of the ammoniurn salts of the mononitrile acids in the 5 presence of added arnmonia in order to limit well-kno~n reductive alkylation reactions which produce dimer and polymer (Table 6).

Temp.3-CPA ~mmoni--m salt [NH40H] wt. % time 3-CPA 4-EPRD
(~C) (M) (M)Raney Ni (h)(% conv.)(% yield) 1.0 2.0 5 2 7 1.5 120 1.0 2.0 5 2 55 22.7 140 1.0 2.0 5 2 71 55.4 140 1.0 2.0 10 2 100 89.9 160 1.0 2.0 5 2 100 90.1 160 1.0 2.0 10 2 100 91.3 180 1.0 2.0 5 2 100 86.1 180 1.0 2.0 10 2 100 90.0 Temp.3-CPA ~-nmonillm saltrNH4OH] wt. % time 3-CPA 4-EPRD
(~C) (M) (M)Raney Ni (h)(% conv.)(% yield) 160 1.0 0 5 2 99 80.1 160 1.0 1.0 5 2 99 87.6 160 1.0 2.0 5 2 100 90.1 160 1.0 3.0 5 2 100 85.4 180 1.0 0 5 2 100 75.1 180 1.0 1.0 5 2 100 85.8 180 1.0 2.0 5 2 100 88.5 180 1.0 3.0 5 2 100 90.0 The hydrogenation of aqueous solutions of 4-CPA arnmonium salt (produced by enzymatic hydrolysis of 2-MGN) in the presence of Raney nickel and excess ammonia at pH 9-10 and at 1 60~C for 2 h produces the corresponding lactam 5-methyl-2-piperidone (5-MPPD) directly, and at yields as high as 96%
(Table 7):

CA 022~4902 1998-11-13 W O 97144318 PCTrUS97/07796 TABLE~
Temp. ~-CPA ammonium salt [NHL~OH] wt. % time ~-CPA 5-MPPD
(~C) (M) (M)R~nev Ni (h)(~'0 conv.) (% yield) 1 60 1 .0 0 5 ' I 00 85.6 1 60 1 .0 2.0 5 3 1 00 96.4 . 1 80 1 .0 2.0 5 ~ 1 00 9 1 .4 1 80 1 .0 3.0 5 ~ 100 89.5 Hydrogenation of aqueous solutions of the arnrnonium salts of 3-cyano-propionic acid (3-CPRA) or 4-cyanobutyric acid (4-CBA) (produced by enzymatic 5 hydrolysis of the corresponding a,cD-dinitriles) produce the corresponding lactams 2-pyrrolidinone and 2-piperidone in 91.0% yield and 93.5% yield, respectively.
Hydrogenation of aqueous solutions of 4-cyano-4-pentenoic acid (4-CP~A) amrnonium salt (produced by enzymatic hydrolysis of 2-methvlenegluL~o~ ,;le) result in hydrogenation of both the nitrile and carbon-carbon double bond to 10 produce 5-methyl-2-piperidone in up to 85% yield.
By adding an excess of ammonia (as ammonium hydroxide) to the hydrogenation reactions in order to limit reductive alkylation during the hydrogenation of nitriles to arnines, several additional byproduct-forming reactions could also have occurred (De Bellefon et al., Catal. Rev. Sci. Eng., (1994), vol. 36, 459-506). It is well-known to those skilled in the art that a cornmon method for the prepa~ation of an amide or carboxylic acid from a nitrileis to heat an aqueous mixture of the nitrile in the presence of an acid or base catalyst. Therefore, an expected competing reaction of the ammonium salt of a mononitrile carboxylic acid under the hydrogenation conditions used in the present invention would be the base-catalyzed hydrolysis of the nitrile group toproduce either the dicarboxylic acid monoarnide ammonium salt or the dicarboxylic acid diammonium salt. These unwanted amide/acid and dicarboxylic acid ammoniurn salts byproducts are produced during the hydrogenations, but in very low yields compared to the yields of lactam.
A second byproduct-forming reaction between the excess arnmonia present and the product lactarn could have produced an equilibrium mixture of the lactarn with the expected amrnonolysis product, an c~-aminocarboxamide. The high yields of the five-membered and six-membered ring lactams ~ in~d under the present reaction conditions suggest that the ammonolysis (or base-catalyzed hydrolysis) of the product lactams is not significant.
In addition to producing lactams from aliphatic a,c~-dinitriles, ~ethyl-lactarns are prepared by the substitution of methylamine for amrnonia in the CA 022~4902 1998-11-13 hydrogenation of aqueous solutions of the arnmonium salts of 4-CPA or 3-CPA.
Addition of from one to tour equivalents of methylamine (pKa 10.62 for the protonated arnine) to an aqueous solution of 4-CPA containing one equivalent of ammonium ion (pKa 9.25) was expected to produce a significant amount of free a~r~nonia (due to the relative differences in pKa's of protonated methylamine and ammonium ions in water). This free arnmonia could then compete with unprotonated methylamine for reaction with the intermediate imine produced during the hydrogenation of the nitrile group, leading to the production of a mixture of the arnmonium salts of 5-amino-4-methylpentanoic acid and 5-N-methylarnino-4-methylpentanoic acid, respectively, which in turn cyclize to produce 5-MPPD and 1,5-dimethyl-2-piperidone (I,5-DMPD), respectively.
The relative yields of 5-MPPD and 1,5-DMPD produced by hydrogenation of 1.0 M aqueous solutions of the arnmonium salt of 4-CPA are found to be dependent on the choice of catalyst. Raney nickel and rutheniurn on alumina eachproduce 5-MPPD as the major lactam product, even in the presence of 3.0 M
methylamine, while l,5-DMPD is the ma~or product when using 5% Pd/C or 4.5%
Pd/0.5% Pt/C as catalyst at the sarne methylarnine concentration. When using 5%
Pd/C as catalyst, the yield of 1,5-DMPD increases with increasing concentration of methylarnine (Table 8).
The substitution of 2.0 M methylamine for arnrnonia in the hydrogenation of an aqueous solutions of 1.0 M 3-CPA ammonium salt at 140~C and using a Pd/C catalyst produces 4-ethyl- 1 -methylpyrrolidin-2-one (4-EMPRD) and 4-EPRD in 69.8% and 20.4% yields, respectively, at 96% conversion.

TABL~ 8 Temp.4-CPA(NH4+)CH3NH2 time4-CPA 1,5-DMPD S-MPPD
(~C) (M) (M)catalyst (h)(% conv) (%yield) (%yield) 160 1.0 3.0 Ra-Ni 2 100 19.268.5 160 1.0 3.0 RUAl2O3 2 100 19.063.5 140 1.0 3.0 PdtC 2 99 83.4 0 160 1.0 1.0 Pd/C 2 100 53.5 0 160 1.0 1.25 Pd/C 2 100 68.3 0 160 1.0 1.5 PdtC 2 100 76.4 0 160 1.0 2.0 Pd/C 2 100 81.5 0 160 1.0 3.0 Pd/C 2 100 81.7 7.8 160 1.0 4.0 Pd/C 2 100 88.4 1.9 180 1.0 3.0 Pd/C 2 97 73.0 6.6 160 1.4 2.3 PdtPttC 2 99 94.0 3.1 W O 97/44318 PCTrUS97/07796 DESCRIPTION OF THE PREFERRED EMBODIMENTS
Methods and Materials Microbial Catalysts for the Pl~alation of ~-Nitrilecarboxylic Acids Two microorg~ni~m~ have been isolated for use as a microbial catalyst for S the conversion of aliphatic a,c3-dinitriles to the corresponding c,~-nitrilecarboxylic acids Acidovoraxfacilis 72W (ATCC 55746) and Comamonas tes~osteroni 5-MGAM-4D (ATCC 55744) Acidovorax facilis 72W (ATCC 557463 was isolated from soil collected in Orange, Texas Standard enrichment procedures were used with the following medium (E2 Basal Medium, pH 7 2) E2 Basal Medium gL

NaH2Po4 0 69 Sodium citrate 0 1 CaCl2-2H2O 0 02S

NaCI 1 0 MgSO4-7H2O 0 5 FeSO4-7H20 CoCI2 6H2O
MnC12-4H20 0.001 ZnC12 0 0005 NaMoO4 2H2o 0 0025 NiCI2-6H20 0.01 CUso4-2H2o Biotin 0 0002 Folic Acid 0 0002 Pyridoxine-HCI 0 001 Riboflavine 0 0005 Thiamine-HCI 0 00005 Nicotinic Acid 0 0005 Pantothenic Acid 0 0005 Vitamin B12 0 00001 p-Aminobenzoic Acid 0 0005 The following supplementations were made to the E2 basal medium for the enrichments described above Strain Enrichment Nitro~en Source Other Supplements A. facilis 72W 0.2 % (v/v) ethylsuccinonitrile 0 3% (v/v) glycerol CA 022~4902 1998-ll-13 W O 97/44318 PCTrUS97/07796 Strains were originally selected based on growth and ammonia production on the enrichment nitrile. Isolates were purified by repeated passing on Bacto~
Brain Heart Infusion Agar (Difco, Detroit. Michigan) followed by screening for ammonia production from the enrichrnent nitrile. Purified strains were identified 5 based on their carbon source utilization profile on a Biolog~ test system (Hayward, CA, USA) using Gram negative test plates.
- For testing nitrile hydrolysis activity, E2 basal medium with 10 g~L
glucose was used to grow A. Sacilis 72W. The medium was supplern~nted with 25 mM (+)-2-methylglutaronitrile. A 10 mL volume of supplemented E2 medium 10 was inoculated with 0.1 mL of frozen stock culture. Following overnight growth at room temperature (22-25~C) on a shaker at 250 rpm, the 10 mL inoculum was added to 990 mL of fresh medium in a 2 L flask. The cells were grown overnight at room tel~lpcld~ with stirring at a rate high enough to cause bubble formationin the medium. Cells were harvested by centrifugation~ washed once with 50 rnM
15 phosphate buffer(pH 7.2)/15% glycerol and the concentrated cell paste was immediately frozen on dry ice and stored at -65~C. Adiponitrile, 10 mM, was alsoused in the l liter fermentations. Fermentations were stopped after 16-20 hours of growth. The cell suspension was chilled to 4~C, harvested by centrifugation and frozen at -60~C following one wash with 15% glycerol in 0.05 M phosphate 20 buffer, pH 7.2. Thawed cell pastes were used for testing nitrile hydrolysis activity. The desired property of the microorganism is a nitrile hydrolyzing activity capable of regiospecific attack of a dinitrile compound in the absence of interfering amidase activity. Microor~ni~m.c tend to undergo mutation. Some mutations may be favorable to the desired nitrile conversion. Thus, even mutants25 of the native strain may be used to carry out the process of the instant invention.
Standard enrichment procedures were also used for Comamonas testosteroni 5-MGAM-4D with E2 Basal Medium, pH 7.2 modified by having vitarnins at one tenth the concentration in the standard basal medium described above. The following supplement~tions were made to the modified E2 basal 30 medium for the enrichments:
Strain Enrichment Nitro~en Source Other Supplements C. testosteroni 2-Methylglutaramide glycerol (0.6%) 5-MGAM-4D (MGAM; 1.0% w/v) Strains were originally selected based on growth on the enrichment nitrile-35 arnide. Isolates were purified by repeated passing on agar plates using the above mediurn. Purified strains were identified based on their carbon source utilization profile on a Biolog~ test system (Hayward, CA, USA) using Gram negative test plates.

CA 022~4902 1998-11-13 For testing nitrile hydrolysis activity, modified E~ basal medium with 6.0 g/L of either glucose or glycerol was used to grow cell material. The mediumwas supplemented with 1.0% MGAM. A 250 mL unbaffled shake flask containing 50 mL of supplemented E2 medium was inoculated with 0.2 mL of frozen stock culture and grown for 72 h at 30~C on a shaker at ~00 rpm. The cells were harvested by centrifugation and washed with 10 mL of 20 mM
KH2PO4, pH 7Ø The cells were screened in 10 mL reactions cont~ining 20 mM
KH2PO4, pH 7.0 and 0.1 M of either methylglutaronitrile or methylgluL~dl.lide for regiospecific hydrolysis using HPLC. The desired property of the microorganism is a nitrile hydrolyzing activity capable of regiospecific attack of a dinitrile compound in the absence of interfering arnidase activitv.
Microorg~ni~m~ tend to undergo mutation. Some mutations may be favorable to the desired nitrile conversion. Thus, even mutants of the native strain may be used to carry out the process of the instant invention.
The present invention is not limited to the particular organi~m.c mentioned above, but includes the use of variants and mutants thereof that retain the desired property. Such variants and mutants can be produced from parent strains by various known means such as x-ray radiation, UV-radiation, and chemical mutagens.
To produce biocatalyst for process demonstration (Examples 1-26), the following media were used.
Strain Mediurn 72-PF- 15 Lauria-Bertani Medium(Bacto~ tryptone, l O g/L + Bacto~ yeast extract, 5 g/L + NaCl, l O g/L) + 0.5%(w/v) sodium succinate-6H2O
72 W E2 + I % (w/v) glucose + 0.4 % (w/v) adipamide 5-MGAM-4D E2 + 1 % (w/v) glucose + 0.2 % (w/v) propionamide To initiate growth, 10 mL of the a~plol..iate medium was inoculated with 0.1 mL of frozen stock culture. Following overnight growth at 28~C with ~h?kin at 250 rpm, the growing cell suspension was transferred to l L of the same medium in a 2 L flask and growth continued at 28~C with ~h~king. The 1 L
growing cell suspension was then added to 9 L of the same medium in a 10 L
fermentation vessel where growth continlle~l Nominal conditions in the fermPnt~rwere: >80% oxygen saturation, 25~C, pH 7.2, 300-1000 rpm. After 20-91 hours, the vessel was chilled to 8-12~C and glycerol added to 10% final concentration.
Cell material was harvested by centrifugation. The concentrated cell paste was im merli~tely frozen on dry ice and stored at -70~C until use. Numerous other supplementations which will serve as carbon and nitrogen sources for cell growthin E2 basal medium are known to those skilled in the art. These, as well as . . .

CA 022~4902 1998-11-13 W O 97144318 PCT~US97/07796 complex nutrient media. can be used to produce biocatalyst. The particular mediadescribed above should not be viewed as restrictive.
Selection of Mutant Strains of ,4cido~ora~facilis 72W Oeficient in NHase Activity Mutants of Acidovorax facilis 72W (ATCC 55746) with reduced capacity to produce the undesirable 2-methylglutaric acid by-product during hydrolysis of2-MGN to 4-CPA were selected based on their inability to use 2-MGN as a carbon and energy source. Specifically, an overnight culture of strain A. facilis 72W
grown on LB/succinate medium (1% (w/v) Bacto-tryptone (Difco, Detroit, Michigan, USA), 0.5% (w/v) Bacto-yeast extract (Difco), 1% (w/v) NaCl, 0.5%
(w/v) sodium succinate hexahydrate) was exposed to 100 llg/mL solution of N-methyl-N'-nitro-N-nitrosoguanidine, a mutagenic agent, for approximately 30 minllt~.s This resulted in a 99.9% reduction in viable cells in the culture.
Mutagenized cells were washed free of the mutagen by centrifugation in sterile, I M sodium phosphate buffer, pH 7.2. Washed cells were resuspended in LB/succinate medium and grown overnight at 30~C. Cells were then washed by centrifugation in sterile. 50 mM sodium phosphate buffer, pH 7.2, and resuspended in E2 minim~l medium(without glucose) cont~ining 0.2% (v/v) 2-methylglutaronitrile, and the antibiotics cycloserine, 0.2 mg/mL and piperacillin, 40 llg/mL. Cells were incubated overnight at 30~C and again washedin sterile, 50 mM sodium phosphate buffer, pH 7.2. Washed cells were spread on agar plates cont~ining a non-selective medium: E2 minim~l medium(without glucose) plus 0.2% (v/v) 2-methylglutaronitrile and 0.5% (w/v) sodium succinate hexahydrate, at a concentration of 40-100 colony-forming units per plate. Plateswere incubated for approximately 48 h at 30~C to allow colonies to develop.
Colonies which developed were replica plated onto agar plates Cont~ining selective medium: E2 minim~l medium(without glucose) plus 0.2% (v/v) 2-MGN. Plates were incubated 48 h at 30~C to allow colonies to develop.
Mutants with desirable qualities do not grow well on the selective medium.
Therefore, after 48 h, replicated plates were compared and strains showing growth only on non-selective medium were saved for further testing.
In total, approximately 5,120 colonies were ~hecl~Pd from 89 plates and 19 strains with the desirable qualities were identified. These mutant strains were further tested for growth in liquid, E2 minim~i medium(without glucose) plus 0.2% (v/v) 2-MGN. Strains which showed little or no growth in this medium were screened for their ability to produce 2-methylglutaric acid during growth in liquid mediurn concicSing of E2 minim~l medium (without glucose) plus 0.2%
{v/v) 2-MGN and 0.5% (w/v) sodium succinate hexahydrate. As a result of this CA 022~4902 1998-11-13 process, two mutant strains, identified as Acidovorax facilis 7~-PF- l S
(ATCC 55747) and ~lcidovoraxfacilis 72-PF17 (ATCC 55745) were chosen for further development due to their greatly ~iminiched capacity to produce 2-methylglutaric acid.
s Aliphatic a,~-Dinitrile ~Iydrolysis Reactions An aqueous solution cont~ininE the amrnonium salt of an aliphatic (,)-nitrilecarboxylic acid is pl~l,~ed by mixing the corresponding aliphatic a,c3-dinitrile with an a~ueous suspension of the app~ iate enzyme catalyst (as identified in part A above). Whole microbial cells can be used as catalyst without any pretreatment. Alternatively, they can be immobilized in a polymer matrix (e.g., alginate beads or polyacrylamide gel (PAG) particles) or on an insoluble solid support (e.g., celite) to facilitate recovery and reuse of the catalyst. Methods for the imrnobilization of cells in a polymer matrix or on an insoluble solid support have been widely reported and are well-known to those skilled-in-the-art.
The nitrilase enzyme, or nitrile hydratase and ~mitl~e enzymes, can also be isolated from the whole cells and used directly as catalyst, or the enzyme(s) can be immobilized in a polymer matrix or on an insoluble support. These methods have also been widely reported and are well-known to those skilled in the art.
Some of the aliphatic a,~D-dinitriles used as starting material in the present invention are only moderately water soluble. Their solubility is also dependent on the temperature of the solution and the salt concentration (buffer and/or ~-nitrilecarboxylic acid ammoniurn salt) in the aqueous phase. For example, adiponitrile was determined to have a solubility limit of ca. 0.60 M, (25~C, 20 mM phosphate buffer, pH 7) and under the same conditions, 2-methylglutaronitrile was determined to have a solubility limit of ca. 0.52 M. In this case, production of an aqueous solution of a ~-nitrilecarboxylic acid ammonium salt at a concentration greater than the solubilty limit of the starting a,o~-dinitrile is accomplished using a reaction mixture which is initially composed of two phases: an aqueous phase cont~ininE the enzyme catalyst and dissolved a,~D-dinitrile, and an organic phase (the undissolved a,c~-dinitrile). As the reaction progresses, the dinitrile dissolves into the a~ueous phase, and eventually a single phase product mixture is obtained.
The concentration of enzyme catalyst in the reaction mixture is dependent on the specific catalytic activity of the enzyme catalyst and is chosen to obtain the desired rate of reaction. The wet cell weight of the microbial cells used as catalyst in hydrolysis reactions typically ranges from 0.001 grarns to 0.100 grams of wetcells per mL of total reaction volume, preferably from 0.002 grams to 0.050 grams of wet cells per mL. The specific activity of the microbial cells (IU/gram wet cell CA 022~4902 1998-ll-13 wt.) is determined by measuring the rate of conversion o~ a O. l O M solution of a dinitrile substrate to the desired ~-nitrilecarboxylic acid product at 25~C using a known weight of microbial cell catalyst. An IU (International Unit) of enzyme activity is defined as the amount of enzyme activity required to convert one 5 micromole of substrate to product per minute.
The temperature of the hydrolysis reaction is chosen to both optimize both the reaction rate and the stability of the enzyme catalyst activity. The telllpGld~
of the reaction may range from just above the freezing point of the suspension (ca.
0~C) to 60~C, with a preferred range of reaction temperature of from 5~C to 35~C.
10 The microbial cell catalyst suspension may be prepared by suspending the cells in distilled water, or in a aqueous solution of a buffer which will m~int~in the initial pH of the reaction between 5.0 and l O.0, preferably between 6.0 and 8Ø As thereaction proceeds, the pH of the reaction mixture may change due to the formation of an ammonium salt of the carboxylic acid from the corresponding nitrile 15 functionality of the dinitrile. The reaction can be run to complete conversion of dinitrile with no pH control, or a suitable acid or base can be added over the course of the reaction to m~int~in the desired pH.
The final concentration of aliphatic cl)-nitrilecarboxylic acid ammonium salt in the product mixture at complete conversion of the a,cl~-dinitrile may range 20 from 0.001 M to the solubility limit of the aliphatic co-nitrilecarboxylic acid ammonium salt. Typically, the concentration of the cl)-nitrilecarboxylic acid amrnonium salt ranged from 0.10 M to 2.0 M. The product mixture of the hydrolysis reaction may be used directly in the subsequent hydrogenation reaction after recovery of the enzyme catalyst by centrifugation and/or filtration. The 25 cl~-nitrilecarboxylic acid may also be isolated from the product mixture (after removal of the catalyst) by adjusting the pH of the reaction mixture to between 2.0 and 2.5 with conc. HCl, saturation of the resulting solution with sodium chloride, and extraction of the to-nitrilecarboxylic acid with a suitable organic solvent such as ethyl acetate, ethyl ether, or dichloromethane. The combined organic extracts30 are then combined, stirred with a suitable drying agent (e.g., magnesium sulfate), filtered, and the solvent removed (e.g., by rotary evaporation) to produce the desired product in high yield and in high purity (typically 98-99% pure). If desired, the product can be further purified by recrystallization or (lictill~tion.
35 Hydrogenation/Cyclization of c~)-Nitrilecarboxylic Acid Ammonium Salts Catalytic hydrogenation is a preferred method for preparing an aliphatic amine from an aliphatic nitrile. In the present invention, the ~-aminocarboxylicacid produced during the hydrogenation cyclizes to the corresponding five-membered or six-membered ring lactam. An aqueous solution of an ammonium CA 022~4902 1998-11-13 W O 97/44318 PCTrUS97/07796 salt of an ~-nitrilecarboxylic acid (prepared by centrifugation and filtration of the aqueous product mixture produced by the enzymatic hydrolysis of the corresponding aliphatic a,~-dinitrile) is first mi~ed with concentrated ammoniumhydroxide and water to produce a solution which contains from one to four stoichiometric equivalents of added ammonium hydroxide. The ammonium hydroxide is added to limit the reductive alkylation of the ~-nitrilecarboxylic acid by the product c,~-aminocarboxylic acid during the course of the hydrogenation. A
two- to three-fold stoichiometric excess of the ammonium hydroxide relative to the amount of c~)-nitrilecarboxylic acid present in the reaction mixture is pl~erel.ed to achieve an optimum yield of the desired lactam. Optionally, amrnonia gas can be substituted for the amrnonium hydroxide added to the reaction mixture. The initial concentration of the ~-nitrilecarboxylic acid arnmonium salt in the hydrogenation reaction mixture may range from S weight percent to 20 weight percent of the solution, with a preferred range of from 7.5 weight percent to 12.5 weightpercent.
For the preparation of a N-methyllactam, methylamine is substituted for amrnonium hydroxide or ammonia, using one to four stoichiometric equivalents relative to the amount of ~-nitrilecarboxylic acid present in the reaction mixture.
A three-fold to four-fold stoichiometric excess of methylamine relative to the amount of c~-nitrilecarboxylic acid present in the reaction mixture is pl~f~ ;d to achieve an optimum yield of the desired N-methyll~ct~m To the hydrogenation reaction mixture described above is then added a suitable hydrogenation catalyst, and the resulting mixture heated under ~s~u.e with hydrogen gas to convert the ~o-nitrilecarboxylic acid ammonium salt to the corresponding five-membered or six-membered ring lactam. Hydrogenation catalysts suitable for this purpose include (but are not limited to) the variouspl~timlm metals, such as iridium, osmium, rhodium, ruthenium, pl~tinllm, and palladium; also various other transition metals such as cobalt, copper, nickel and zinc. The catalyst may be unsupported, (for exarnple as Raney nickel or platinumoxide), or it may be supported (for example, as palladium on carbon, platinum onalumina, or nickel on kieselguhr).
The hydrogenation catalyst is used at a minimum concentration sufficient to obtain the desired reaction rate and total conversion of starting m~tni~l~ under the chosen reaction conditions. This concentration is easily detennin~cl by trial.
The catalyst may be used in amounts of from 0.001 to 20 or more parts by weight of catalyst per 100 parts of c~-nitrilecarboxylic acid employed in the reaction. The catalyst loading in the reaction mixture is typically from 1% to 10% (weight catalyst/weight of a)-nitrilecarboxylic acid), with a 3% to 5% catalyst loading preferred. Raney nickel (e.g., Cr-promoted Raney nickel catalyst (Grace Davison CA 022~4902 l998-ll-l3 W O 97/44318 PCTrUS97/07796 Raney 2400 active metal catalyst)) is a preferred catalyst for reactions run in the presence of added ammonia to produce lactams, while 5% or 10% palladium on carbon, or 4.5% palladium/0.5% platinum on carbon are plefel~ed for reactions run in the presence of added methylamine to produce ~-methyllactarns.
The hydrogenation temperature and pressure can vary widely. The temperature may generally be in the range of from 45~C to 200~C~ preferably from70~C to 1 80~C. The hydrogen pressure is generally in the range of from about atmospheric to about 100 atmospheres, preferably from 30 to 60 atmospheres.
The hydrogenation is performed without any pH adj--ctmP~lt of the reaction mixture, which with the addition of an excess of ammonium hydroxide, ammonia, or methylamine is generally between a pH of from 9 to 12. Within this pH range, the exact value may be adjusted to obtain the desired pH by adding any compatible, non-interfering base or acid. Suitable bases include, but are not limited to, alkali metal hydroxides, carbonates, bicarbonates and phosph~tes, while suitable acids include, but are not limited to, hydrochloric, sulfuric, orphosphoric acid.
Lysis of the microbial cell catalysts during the hydrolysis reaction or cont~min~nt.c present from the catalyst pre~)~d1ion could have introduced compounds into the hydrogenation reaction mixture (e.g., thiols) which could have poisoned the catalyst activity. No poisoning or deactivation of the hydrogenation catalysts was observed when comparing the hydrogenation of c,)-nitrilecarboxylic acid ammonium salt mixtures produced via microbial hydrolysis with the hydrogenation of aqueous solutions of the same cl)-nitrilecarboxylic acid which was isolated from the hydrolysis product mixtures and purified prior to hydrogenation.
The lactam or N-methyllactam may be readily isolated from the hydrogenation product mixture by first filtering the mixture to recover the hydrogenation catalyst, and then the product can be distilled directly from the recl-lting aqueous filtrate. The ammonia produced during the cyclization reaction or added to the reaction mixture can also be recovered for recycling by this lli.ctill~tion process, and the generation of undesirable inorganic salts as waste products is avoided. The lactams or N-methyl lactams may also be recovered by filtering the hydrogenation mixture, adj--ctm~nt of the filtrate to a pH of ca. 7 with conc. HCI and saturation with sodium chloride, extraction of the lactam or N-methyllactarn (batch or continuous extraction) with an organic solvent such asethyl acetate, dichloromethane~ or ethyl ether, and recovery from the organic extract by distillation or crystallization. In the accompanying examples, the isolated yields reported for this method are unoptimi7~, and this method was CA 022~4902 1998-11-13 W O 97/44318 PCTrUS97/07796 used simply to obtain purified product for analysis and confirmation of chemicalidentity.
In the following exarnples, which serve to further illustrate the invention and not to limit it, the % recovery of aliphatic a.~-dinitriles and the % yields of the hydrolysis products formed during the microbial hydrolysis reactions were based on the initial arnount of ~ dinitrile present in the reaction mixture (unless otherwise noted), and determined by HPLC using a refractive index detector and either a Supelcosil LC-18-DB column (25 cm x 4.6 mm dia.) or a Bio-Rad HPX-87H column (30 cm x 7.8 mrn dia.). The yields of lactarns and N-methyl-lactams produced by the hydrogenation of aqueous solutions of ~-nitrilecarboxylic acid arnrnonium salts were based on the initial concentration of c~-nitrilecarboxylic acid ammonium salt present in the reaction mixture (unless otherwise noted), and determined by gas chromatography using a DB-1701 capillary colurnn (30 m x 0.53 mm ID, 1 micron film thickness).

4-Cyanopelllanoic acid (amrnoniurn salt) First, 0.60 grarns (wet cell weight) of frozen Acidovorax facilis 72W
(ATCC 55746) cells (previously heat-treated at 50~C for I h before freezing) were placed into a 15-mL polypropylene centrifuge tube and followed by addition of 12 mL of potassiurn phosphate buffer (20 mM, pH 7.0). After the cells were thawed and suspended, the resulting suspension was centrifuged and the supçrn~t~nt discarded. The resulting cell pellet was resuspended in a total volume of 12 mL of this same phosphate buffer. Into a second 15-mL polypropylene centrifuge tube was weighed 0.1081 g (0.114 mL, 1.00 mmol, 0.100 M) of 2-methylglutaronitrile, then 9.~9 mL of the A. facilis 72W (ATCC 55746) cell suspension (0.494 g wet cell weight) was added and the resulting suspension mixed on a rotating platform at 27~C. Samples (0.300 mL) were withdrawn and centrifuged, then 0.180 mL of the sup~rn~t~nt was placed in a Millipore Ultrafree-MC filter unit (10 K MWCO) and mixed with 0.020 mL of an aqueous solution of 0.750 M N-methylpropionamide (HPLC external standard solution). Sufficient 1.0 M HCl was added to lower the pH of the sample to ca. 2.5 and the resulting solution was filtered and analyzed by HPLC. After 1.0 h, the HPLC yields of 4-cyanopentanoic acid and 2-methylglutaric acid were 99.3% and 0.7%, respectively, with no 2-methylgl~ll~vnil,;le rem~ining EXAMPLE 2 (COMPARATIVE) 4-Cyanopentanoic acid (ammonium salt) The procedure described in Example I was repeated using Acidovorax facilis 72W (ATCC 55746) cells which had not been heat-treated at 50~C for 1 h before freezing. After 1.0 h, the HPLC yields of 4-cyanopentanoic acid and CA 022~4902 1998-ll-13 W O 97/44318 PCTrUS97/07796 2-methylglutaric acid were 62.7% and 34.6%, respectively. with no 2-methyl-glutaronitrile rem~ining.

4-Cyano~entanoic acid (ammonium salt) First. I .136 grams (wet cell weight) of frozen Acidovorax facilis 72W
(ATCC 55746) cells (previousJy heat-treated at 50~C for 1 h before freezing) were placed into a S0-mL polypropylene centrifuge tube, followed by 21.6 mL of potassium phosphate buffer (20 mM, pH 7.0). After the cells were thawed and suspended, the resulting suspension was centrifuged and the supernatant discarded. The resulting cell pellet was resuspended in a total volume of 22.7 mL
of this same phosphate buffer. Into a 15-mL polypropylene centrifuge tube was weighed 0.4355 g (0.458 mL, 4.00 rnmol, 0.403 M) of 2-methyl-glutaronitrile, then 9.54 mL of the A. facilis 72W (ATCC 55746) cell suspension (0.477 g wet cell weight) was added and the resulting suspension mixed on a rotating platformat 27~C. Samples (0.300 mL) were diluted 1 :4 with distilled water, then centrifuged, and 0.180 mL of the supernatant was placed in a Millipore Ultrafree-MC filter unit (10 K MWCO) and mixed with 0.020 mL of an aqueous solution of 0.750 M N-methylpropionamide (HPLC extemal standard solution). Sufficient 1.0 M HCI was added to lower the pH of the sample to ca. 2.5 and the resulting solution was filtered and analyzed by HPLC. After 4.0 h, the HPLC yields of 4-cyanopentanoic acid and 2-methylglutaric acid were 99.4% and 0.6%, respectively, with no 2-methylglularu~ ile rem~ining.

4-Cvanopentanoic acid (ammoniurn salt) The procedure described in Example 3 was repeated~ except that 1.086 g (1.143 mL, 10.04 mmol, two-phase reaction. 1.00 M product) of 2-methyl-glul~onillile was mixed with 8.86 mL of the heat-treated A. facilis 72W
(ATCC 55746) cell suspension (0.443 g wet cell weight) was mixed in a 15 mL
polypropylene centrifuge tube on a rotating platform at 27~C. Samples (0.300 mL) were diluted 1:10 with distilled water, then centrifuged, and 0.180 mL
of the supern~t~nt was placed in a Millipore Ultrafree-MC filter unit (10 K
MWCO) and mixed with 0.020 mL of an aqueous solution of 0.750 M
N-m~lyl~ .ionamide (HPLC external standard solution). Sufficient 1.0 M HCl was added to lower the pH of the sample to ca. 2.5, and the resulting solution was filtered and analyzed by HPLC. After 15.25 h, the HPLC yields of 4-cyanopentanoic acid and 2-methylglutaric acid were 98.7% and 1.3%, respectively, with no 2-methylglutaronitrile rem~inin~.

CA 022~4902 1998-ll-13 W O 97/44318 PCT~US97/07796 EXAMPLE S
4-Cvanopentanoic acid (ammonium salt~
The procedure described in Example 1 was repeated using a suspension of ~cidovoraYfacilis mutant strain 72-P~-15 (ATCC 55747) which had not been 5 heat-treated at 50~C for I h. After 3.0 h, the HPLC yields of 4-cyanopentanoicacid and 2-methylglutaric acid were 96.8% and 3.6%, respectively. ~vith no 2-methylglutaronitrile rem~ining.

4-Cvanopentanoic acid (ammonium salt) The procedure described in Example 3 was repeated using a suspension of ,4cidovorax facilis mutant strain 72PF- l S (ATCC 55747) which had not been heat-treated at 50~C for I h. A mixture of 0.4355 g (0.458 mL, 4.00 rnmol, 0.403 M) of 2-methylglul~unil-;le and 9.54 mL of the A. facilis mutant strain 72-PF- 15 cell suspension (0.477 g wet cell weight) was mixed in a l S mL
polypropylene centrifuge tube on a rotating platforrn at 27~C. After 6.0 h, the HPLC yields of 4-cyanopentanoic acid and 2-methylglutaric acid were 98.8% and 1.2%, respectively, with no 2-methylglu~onil,;le le.llAi~ e 4-C~anopentanoic acid (ammonium salt) The procedure described in Example 4 was repeated using a suspension of Acidovoraxfacilis mutant strain 72-PF-lS (ATCC 55747) which had not been heat-treated at 50~C for 1 h. A mixture of 1.086 g (1.143 rnL, 10.04 mmol, 1.00 M) of 2-methylglutaronitrile and 8.86 mL of the A. facilis mutant strain 72PF-lS (ATCC 55747) cell suspension (0.443 g wet cell weight) was mixed in a 15 mL polypropylene centrifuge tube on a rotating platform at 27~C. After 15.25 h, the HPLC yields of 4-cyanopentanoic acid and 2-methylglutaric acid were 99.2% and 0.8%, respectively, with no 2-methylglu~ol~.;le rçrn~inin~.

4-Cvano~el"~loic acid Isolation Into a 2-L erlenmeyer flask equipped with a magnetic stir bar was weighed lS0 g (wet cell weight) offrozen Acidovoraxfacilis 72W (ATCC 55746) (not previously heat-treated at 50~C for 1 h before freezing). Potassium phosphate buffer (20 mM~ pH 7.0) was then added to a total volume of 1.50 L. After the cells were thawed and suspended, the resulting suspension was heated in a water bath to 50~C for 1 h, then cooled to 10~C in an ice/water bath, centrifuged, and the supernatant discarded. The resulting cell pellet was washed once by resuspensionin 1.50 L of the same phosphate buffer, followed by centrifugation. The washed cell pellet was transferred to a 4-L erlenmeyer flask equipped with magnetic stir bar, then suspended in a total volume of 2.5 L of potassium phosphate buffer CA 022~4902 1998-ll-13 W O 97/44318 PCT~US97/07796 (20 mM. pH 7.0). With stirring, 129.6 g (136.4 mL, 1.20 mol. 0.400 M) of . -methylglutaronitrile was added, and the final volume adjusted to 3.00 L with the same phosphate buffer. The mixture was stirred at 25~Ct and samples were withdrawn at regular intervals and analyzed by HPLC. After 21.5 h, the HPLC
yields of 4-cyanopentarloic acid (4-CPA) and 2-methylglutaric acid were 99.5%
and 0.5%, respectively, with no 2-methylglutaronitrile rem~ining.
The reaction mixture was centrifuged, the cell pellet recovered for reuse, and the resulting supernatant dec~ntçd and filtered using an Amicon 2.5 L FilterUnit equipped with a YM-IO filter (IOK MWCO). The filtrate was placed in a 4.0 L flask, and the pH of the solution adjusted to 2.5 with 6 N HCI. To the solution was then added sodium chloride with stirring until saturated; then 1.0 L
portions of the resulting solution were extracted with 4 x 500 mL of ethyl ether.
The combined ether extracts were dried (MgSO4), filtered, and the solution concentrated to 1.0 L by rotary evaporation at reduced pressure. To the solutionwas then added 1.2 L of hexane and 0.200 mL of ethyl ether, and the resulting solution cooled to -78~C. The resulting white crystalline solid was isolated by rapid vacuum filtration and washing with 300 mL of cold (5~C) hexane. Residual solvent was removed under high vacuum (150 millitorr) to yield 120.3 g (79%
isolated yield) of 4-cyanopentanoic acid (mp. 31.9-32.6~C).

4-Cvanopentanoic acid (ammoniurn salt) The recovered cell pellet from the Example 8 was reused in several consecutive 3.0-L batch reactions for the hydrolysis of 2-methylglutaronitrile to 4-cyanopentanoic acid ammonium salt. At concentrations of 2-methylglutaronitrile greater than 0.400 M, the solubility of the dinitrile in the aqueous cell suspension was exceeded, and these reactions ran as two-phase aqueouslorganic mixtures until the rem~ining 2-methylglutaronitrile was soluble in the reaction mixture. The Table below lists the final concentration of 4-CPA
arnmoniurn salt produced, and the percent yields of 4-CPA and 2-methylglutaric acid. Based on dry weight of cell catalyst (1 gram wet weight = 0.20 gram dry weight), 86 g 4-CPA/g dry weight of Acidovoraxfacilis 72W (ATCC 55746) was produced.

CA 022~4902 1998-11-13 Time [4-CPA(NH4)] 4-CPA 2-MGA
r~n# (h) (M) (%yield)(%yield~
23 0.40 99.5 0.5 2 22 0.40 99.1 0.9 3 46 1.00 99.2 0.8 4 50 1.00 99.4 0.6 99 1.85 98.8 1.2 6 261 2.00 98.9 1.1 3-Cyanopentanoic acid (ammonium salt) First, 0.60 grams (wet cell weight) of frozen Acidovorax facilis 72W
(ATCC 55746) (previously heat-treated at 50~C for I h before freezing), was placed into a l S-mL polypropylene centrifuge tube, and then followed by the addition of 10 mL of potassium phosphate buffer (20 mM, pH 7.0). After the cellswere thawed and suspended, the resulting suspension was centrifuged, and the supem~t~nt discarded. The resulting cell pellet was resuspended in a total volurne of l O mL of this same phosphate buffer. Into a second 15-mL polypropylene centrifuge tube was weighed 0.1080 g (0.112 mL, l.00 mmol, 0.100 M) of 2-ethylsuccinonitrile, 8.00 mL of the A. facilis 72W cell suspension (0.5 g wet cell weight) was added, the total volume adjusted to 10.0 mL with potassium phosphate buffer (20 mM, pH 7.0), and the resulting suspension mixed on a rotating platform at 27~C. Sarnples (0.300 mL) were withdrawn and centrifuged, then 0.180 mL of the supern~t~nt was placed in a Millipore Ultrafree-MC filter unit (0.22 micron) and mixed with 0.020 mL of an aqueous solution of 0.750 M
N-methylpropionamide (HPLC external standard solution). The resulting solution was filtered and analyzed by HPLC. After 1.0 h, the HPLC yield of 3-cyano-pentanoic acid was 100%, with no 2-ethylsuccinonitrile rern~ining, and no other byproducts observed.

3-CYanopentanoic acid (ammonium salt) The procedure described in Example 10 was repeated, using 0.4337 g (0.449 rnL, 4.01 mmol, 0.401 M) of 2-ethylsuccinonitrile and 8.00 mL of the heat-treated A. facilis 72W (ATCC 55746) cell suspension (0.5 g wet cell weight), in a total volume of 10.0 mL (adjusted with potassium phosphate buffer (20 mM, pH 7.0)). Samples (0.100 mL) were withdrawn, diluted I :4 with water and centrifuged; 0.180 mL of the supernatant was then placed in a Millipore Ultrafree-MC filter unit (0.22 micron) and mixed with 0.020 mL of an a~ueous solution of 0.750 M N-methylpropionamide (HPLC external standard solution). The CA 022~4902 1998-11-13 W O 97/44318 PCT~US97/07796 resulting solution was filtered and analyzed by HPLC. After 8.0 h, the HPLC
yield of 3-cyanopentanoic acid was 100%, with no 2-ethvlsuccinonitrile rem~ining, and no other byproducts observed.
EXAMP~E 12 3-Cyanopentanoic acid (ammonium salt) The procedure described in Example lO was repeated, using 1.087 g (1.13 mL, 10. I mmol, two-phase reaction, l .01 M product) of 2-ethylsuccinonitrile and 8.00 mL of the heat-treated ~. facilis 72W
(ATCC 55746) cell suspension (0.5 g wet cell weight), in a total volume of l O.O mL (adjusted with potassium phosphate buffer (20 mM, pH 7.0)). Sarnples (O.100 mL) were withdrawn, diluted 1: 10 with water and centrifuged, then 0.180 mL of the supernatant was placed in a Millipore Ultrafree-MC filter unit (0.22 micron) and mixed with 0.020 mL of an aqueous solution of 0.750 M
N-methylpropionamide (HPLC external standard solution). The resulting solution was filtered and analyzed by HPLC. After 71 h, the HPLC yield of 3-cyanopentanoic acid was 100%, with no 2-ethylsuccinonitrile r~ g, and no other byproducts observed.

3-Cvanopentanoic acid Isolation Into a 4.0 L erlenrneyer flask equipped with magnetic stir bar was placed 161 g of frozen Acidovorar facilis 72W (ATCC 55746) cells (previously heat-treated at 50~C for 1 h before freezing) and 1.60 L of potassium phosphate buffer (20 mM, pH 7.0) at 27~C. With stirring, the cells were thawed and suspended, then 325 g (336.0 mL, 3.00 mole) of 2-ethylsuccinonitrile was added with stirnng, and the total volume of the mixture adjusted to 2.40 L with potassium phosphate buffer (20 mM, pH 7.0). The resulting mixture was stirred at 27~C, and samples (O.100 mL) were withdrawn, diluted 1: 10 with water and centrifuged, 0.180 mL ofthe supçrn~t~nt was then placed in a Millipore Ultrafree-MC filter unit (0.22 micron) and mixed with 0.020 mL of an aqueous solution of 0.750 M
N-methylpropionamide (HPLC external standard solution). The resulting solution was filtered and analyzed by HPLC. After 183 h. the HPLC yield of 3-cyano-pentanoic acid was 100%, with no 2-ethylsuccinonitrile renl~ining, and no other byproducts observed. The reaction mixture was centrifuged, the cell pellet recovered for reuse, and the resulting 2.3 L of supernatant decz~nted and filtered using an Amicon 2.5 L Filter Unit equipped with a YM-lO filter (lOK MWCO).
The concentration of 3-cyanopentanoic acid ammoniurn salt in the filtrate was 1.26 M.
A 300-mL portion of the filtrate described above, cont~ining 1.26 M
3-cyanopentanoic acid ammonium salt, was adjusted to pH 2.5 with ca. 60 mL of CA 022~4902 1998-ll-13 W O 97/44318 PCTrUS97/07796 6 N HCI, then saturated with sodium chloride and e:ctracted with ~ x 200 mL of ethyl ether. The combined organic e~ctracts were dried over magnesium sulfate, filtered, and the solvent removed by rotary evaporation at reduced pressure at 28~C. The resulting slightly-yellow viscous oil was stirred under high vacuum (50 millitorr) to remove residual solvent at 2~~C. then cooled to -20~C for 2-3 h to produce 3-cyanopentanoic acid as a crystalline white solid (45.9 g, 96% yield);
m.p. 33.0-34.0~C.

3-CvanoPentanoic acid (ammonium salt) The procedure described in Example 10 was repeated using a suspension of Acidovoraxfacilis mutant strain 72-PF-IS (ATCC 55747) which had not been heat-treated at 50~C. After 1.0 h, the HPLC yield of 3-cyanopentanoic acid was 100%, with no 2-ethylsuccinonitrile r~nn~ining, and no other byproducts observed.

3-Cvanopentanoic acid (am~nonium salt) The procedure described in Example 14 was repeated using 0.4325 g (0.455 mL, 4.00 mrnol, 0.400 M) of 2-ethylsuccinonitrile and 8.00 mL (0.5 g wet cell weight) of a suspension of ,4cidovorax facilis mutant strain 72-PF- 15 (ATCC 55747) which had not been heat-treated at 50~C, in a total volume of 10.0 mL (adjusted with potassiurn phosphate buffer (20 mM, pH 7.0)). After 25 h, the HPLC yield of 3-cyanopentanoic acid was 100%, with no 2-ethylsuccinonitrile rem~ining, and no other byproducts observed.

3-Cvanopentanoic acid (amrnonium salt) The procedure described in Exarnple 14 was repeated, using 1.087 g (1.14mL, 10.1 mmol,two-phasereaction, 1.01 Mproduct)of2-ethyl-succinonitrile and 8.00 mL (0.5 g wet cell weight) of a suspension of Acidovoraxfacilis mutant strain 72-PF-IS (ATCC 55747) which had not been heat-treated at 50~C, in a total volume of 10.0 mL (adjusted with potassium phosph~t~ buffer (20 mM, pH 7.0)). After 71 h, the HPLC yield of 3-cyanopentanoic acid was 76.6%, with 25.1 % 2-ethylsuccinonitrile rem~ining and no other byproducts observed.

4-Cyano~-pentenoic acid Isolation Into a 500 mL erlenmeyer flask equipped with m~gnPSic stir bar was placed 50.0 g of frozen Acidovorax facilis 72W (ATCC 55746) cells (previously heat-treated at 50~C for I h before freezing) and 450 mL of potassium phosphate buffer (20 mM, pH 7.0) at 27~C. With stirring, the cells were thawed and suspended, then centrifuged and the supernatant discarded. The cell pellet was CA 022~4902 1998-ll-13 W O 97/44318 PCTrUS97/07796 resuspended in a total volume of 863 mL of potassium phosphate buffer (20 mM, pH 7.0) in a 1.0 L flask~ then 133 g (137.0 mL, 1.25 mole) of 2-methyleneglutaronitrile was added with stirring at 27~C. At concentrations of 2-methyleneglutaronitrile greater than 0.400 M the solubility of the dinitrile in the aqueous cell suspension ~vas exceeded, and these reactions ran as two-phase aqueoustorganic mi.~ctures until the rem~ining 2-methyleneglutaronitrile was soluble in the reaction mixture. Samples (0.100 mL) were withdrawn, diluted 1: 10 with water and centrifuged, then 0.150 mL of the supernatant was placed in a Millipore Ultrafree-MC filter unit (0.22 micron) and mixed with 0.150 mL of an aqueous solution of 0.150 M N-methylpropionamide (HPLC external standard solution). The resulting solution was filtered and analyzed by HPLC. After 26 h,the HPLC yield of 4-cyano-4-pentenoic acid was 100%, with no 2-methyleneglutaronitrile rem~ining, and no other byproducts observed. The reaction mixture was centrifuged, the cell pellet was recovered for reuse, and the supernatant was filtered using an Amicon 2.5 L Filter Unit equipped with a YM- 10 filter (1 OK MWCO). The final concentration of 4-cyano-4-pentenoic acid ammonium salt in the filtrate was 1.298 M.
A 100-mL portion of the filtrate cont~ining the 4-cyano-4-pentenoic acid ammonium salt product mixture described above was adjusted to pH 2.7 with 6 N
HCI, then saturated with sodium chloride and extracted with 4 x 100 mL of ethyl ether. The combined organic extracts were dried over m~nesium sulfate, filtered,and the volume of the combined extracts reduced to 100 mL by rotary evaporation at reduced pressure at 28~C. To the ether solution was added 200 mL of hexane, and the resulting solution cooled to -78~C. The resulting white solid which cryst~lli7ed was isolated by rapid vacuum filtration and washing with 100 mL of cold (5~C) hexane. Residual solvent was removed under high vacuum (150 millitorr) to yield 9.80 g (60% isolated yield) of 4-cyano-4-pentenoic acid(m.p. 26.5-27.0~C, stored at -20~C).

4-Cvano-4-pentenoic (ammonium salt) The recovered cell pellet from the Example 17 (Acidovorax facilis 72W
(ATCC 55746) cells) was reused in a second consecutive l.O-L batch reactions forthe hydrolysis of 2-methyleneglutaronitrile to 4-cyano-4-pentenoic acid arnmonium salt. At concentrations of 2-methyleneglutaronitrile greater than 0.400 M, the solubility of the dinitrile in the aqueous cell suspension was excee~le~l and these reactions ran as two-phase aqueouslorganic mixtures until the r~m~ining 2-methyleneglutaronitrile was soluble in the reaction mixture. The cell pellet was resuspended in a total volume of 781 ml of potassium phosphate buffer(20 mM, pH 7.0) in a 1.0 I, flask, then 212 g (219.0 mL7 2.00 mole) of CA 022~4902 1998-11-13 WO 97/44318 PCT/USg7/07796 2-methyleneglutaronitrile was added with stirring at 27~C. Samples (0.100 mL) were withdrawn. diluted 1:20 with water and centrifuged. then 0.150 mL of the supernatant was placed in a Millipore Ultrafree-~lC filter unit (0.22 micron) and mixed with 0.150 mL of an aqueous solution of 0.150 M N-methylpropionamide 5 (HPLC external standard solution). The resulting solution was filtered and analyzed by HPLC. After 53 h, the HPLC yield of 4-cyano-4-pentenoic acid was 100%, with no 2-methyleneglutaronitrile rem~;ninE, and no other byproducts observed.

3-Cyanopropanoic acid (arnmonium salt) First, 0.50 grams (wet cell weight) of frozen Acidovorax facilis 72W
(ATCC 55746) (previously heat-treated at 50~C for 1 h before freezing) were placed into a 1 S-mL polypropylene centrifuge tube and followed by the addition of 10 mL of potassium phosphate buffer (20 mM, pH 7.0). After the cells were thawed and suspended, the resulting suspension was centrifuged, and the sup~rn~t~nt discarded. The resulting cell pellet was resuspended in a total volurne of 10 rnL ofthis same phosph~te buffer. Into a second 15-mL polypropylene centrifuge tube was placed 0.3236 g (4.00 rnmol, 0.400 M) of succinonitrile dissolved in a total volu ne of 8.0 rnI, of potassiurn phosphate buffer (20 mM, pH 7.0), then 2.00 rnL of the A. facilis 72W (A~CC 55746) cell suspension (0.10 g wet cell weight) was added, and the resulting suspension mixed on a rotating platforrn at 27~C. Samples (0.100 rnL) were withdrawn, diluted 1:4 withwater and centrifuged, then 0.150 mL of the sup~ . "~ l1 was placed in a Millipore Ultrafree-MC filter unit (0.22 micron) and mixed with 0.015 rnL of 0.1 M HCI
and 0.150 mL of an aqueous solution of 0.100 M propionic acid (HPLC external standard solution). The resulting solution was filtered and analyzed by HPLC.
After 3.0 h, the HPLC yield of 3-cyanopropanoic acid was 100%, with no succinonitrile rem~ining, and no other byproducts observed.

3-C~anol)roPalloic acid Isolation Into a 500 mL erlenmeyer flask equipped with m~gn~tic stir bar was placed 20.0 g of frozen Acidovorax facilis 72W (ATCC 55746) cells (heat-treated at 50~C for 1 h before freezing) and 200 mL of potassiurn phosphate buffer (20 mM, pH 7.0) at 27~C. With stirring, the cells were thawed and suspended, then centrifuged and the supçrn~t~nt discarded. This wash procedure was repeated. The resulting cell pellet was resuspended in a total volume of 1.00 L of potassium phosphate buffer (20 mM, pH 7.0) cont~ining 101.1 g (1.25 mole, 1.25 M) succinonitrile and the resulting mixture was stirred in a l-L flask placed in a water bath at 27~C. Samples (0.100 mL) were withdrawn, diluted 1 :10 with CA 022~4902 1998-11-13 W O 97/44318 PCTrUS97/07796 water and centrifuged, then 0.150 mL of the supernatant was placed in a Millipore Ultrafree-MC filter unit (0.22 micron) and mixed with 0.015 mL of 0. I M HCI
and 0.150 mL of an aqueous solution of 0.100 M propionic acid (HPLC external standard solution). The resulting solution was filtered and analyzed by HPLC.
After I .0 h, the HPLC yield of 3-cyanopropanoic acid and succinic acid were 99.7% and 0.3%, respectively, with no succinonitrile rem~inine. The reaction mixture was centrifuged, and the supern~t:~nt was filtered using an Amicon 2.5 LFilter Unit equipped with a YM-IO filter (IOK MWCO). The final concentration of 3-cyanol)ropanoic acid ammonium salt in the filtrate was 1.31 M.
A 200-mL portion of the filtrate containing the 3-cyanopropanoic acid ammonium salt product mixture described above was adjusted to pH 2.5 with 6 N
HCI, then saturated with sodium chloride and extracted with 4 x 200 mL of ethyl ether. The combined organic extracts were dried over magnesium sulfate, filtered, and the solvent removed by rotary evaporation at reduced pressure. The resultingcolorless oil was dissolved in 150 mL of ethyl ether, then 100 mL of hexane was added, and the resulting solution cooled to -78~C. The resulting white solid which crystallized was isolated by vacuum filtration, and residual solvent was removedunder high vacuum (150 millitorr) to yield 14.32 g (55% isolated yield) of 3-cyanopropanoic acid (m.p. 49.5-51.0~C).

4-Cvanobutyric acid (amrnonium salt) First, 0.50 grarns (wet cell weight) of frozen Acidovorax facilis 72W
(ATCC 55746) (previously heat-treated at 50~C for 1 h before freezing) were placed into a 15-mL polypropylene centrifuge tube and then followed by the addition of 10 mL of potassium phosphate buffer (20 mM, pH 7.0). After the cellswere thawed and suspended, the resulting suspension was centrifuged, and the supem~t~nt discarded. The res--ltine cell pellet was resuspended in a total volurne of 10 mL of this same phosphate buffer. Into a second 15-mL polypropylene centrifuge tube was placed 0.3830 g (4.03 rnmol, 0.400 M) of glutaronitrile dissolved in a total volume of 6.0 mL of potassiurn phosphate buffer (20 mM, pH 7.0), then 4.00 mL of the A. facilis 72W (ATCC 55746) cell suspension (0.20 g wet cell weight) was added, and the resulting suspension mixed on a rotating platform at 27~C. Samples (0.100 mL) were withdrawn, diluted 1 :4 with water and centrifuged, then 0.150 mL of the supernatant was placed in a Millipore Ultrafree-MC filter unit (0.22 micron) and mixed with 0.015 mL of 0.1 M HCI
and 0.150 mL of an aqueous solution of 0.150 M potassium acetate (HPLC
external standard solution). The resulting solution was filtered and analyzed byHPLC. After 4.0 h, the HPLC yield of 4-cyanobutyric acid and glutaric acid were 85.1% and 8.2%, respectively, with 5.7% glutaronitrile rern~inine.
3~

CA 022~4902 1998-11-13 W O 97/44318 PCTrUS97/07796 EXA MPLE ~
4-Cvanobutyric acid Isola~ion Into a 250 mL erlenrneyer flask equipped with magnetic stir bar was placed l S.O g of frozen Acidovorax facilis 72W (ATCC 55746) cells (heat-treatedS at 50~C for I h before freezing) and 135 mL of potassium phosphate buffer (20 mM, pH 7.0) at 27~C. With stirring, the cells were thawed and suspended, then centrifuged and the supernatant discarded. This wash procedure was repeated. The resulting cell pellet was resuspended in a total volume of l OO mL of potassium phosphate buffer (20 mM, pH 7.0) and this cell suspension was added to a SOO mL flask cont~ining a magentic stir bar, 42.78 g (0.450 mole) of glutaronitrile and 157.2 mL of potassium phosphate buffer (20 mM, pH 7.0). The resulting mixture, cont~iningl.S M gluLalonillile, was stirred at 27~C. ~amples (O.100 mL) were withdrawn, diluted l: 10 with water and centrifuged. Then, 0.200 rnL of the supernatant was placed in a Millipore Ultrafree-MC filter unit (0.22 micron) and mixed with 0.020 mL of 0. l M HCI and 0.200 mL of an aqueous solution of O. l SO M potassium acetate (HPLC external standard solution).
The resulting solution was filtered and analyzed by HPLC. After 4.0 h, the HPLC
yield of 4-cyanobutyric acid ammonium salt and glutaric acid diammonium salt were 92.3% and 7.7%, respectively, with no glutaronitrile rem~ining The reaction mixture was centrifuged and the supernatant was filtered using an Amicon 2.5 L Filter Unit e~uipped with a YM-10 filter (lOK MWCO). The filtrate con~ining the 4-cyanobutyric acid ammonium salt product mixture described above was adjusted to pH 3.5 with 6 N HCl, then saturated with sodium chloride and extracted with 4 x 300 rnL of ethyl ether. The combined organic extracts were dried over m~gne~jum sulfate, filtered, and the solvent removed byrotary evaporation at reduced pressure followed by stirring under vacuum ( l OO millitorr) to yield 35.3 g (62% yield) of 4-cyanobutyric acid as a pale yellow oil which cryst~lli7Pd upon st~n-ling. The solid was recrystallized from 1: l ethyl acetate/hexane at 5~C (mp. 39.6-40.2~C).

5-Cyanopentanoic acid (ammonium salt) First, 2.0 grarns (wet cell weight) of frozen Comamonas tes~osteroni S-MGAM-4D (ATCC 55744) cells (previously heat-treated at 50~C for l h before freezing) were placed into a SO-mL polypropylene centrifuge tube and followed by the addition of 30 mL of potassium phosphate buffer (20 mM, pH 7.0). After the cells were thawed and suspended, the resulting suspension was centrifuged, and the supernatant discarded. The resulting cell pellet was resuspended in a total volume of 30 mL of this same phosphate buffer. Into a second 15-mL
polypropylene centrifuge tube was weighed 0.1085 g (O.114 mL, 1.00 mmol, CA 022~4902 1998-11-13 0.100 M) of adiponitrile, then 9.29 mL of potassium phosphate buffer (20 mM, pH 7.0) and 0.60 mL of the ~'omamonas testosteroni 5-MGAM-4D
(ATCC 55744) cell suspension (0.0~0 g wet cell weight) was added, and the resulting suspension miYed on a rotating platforrn at 27~C. Samples (0.300 mL) were withdrawn and centrifuged, then 0.180 mL of the supernatant was placed in aMillipore Ultrafree-MC filter unit (10 K MWCO) and mixed with 0.020 mL of an aqueous solution of 0.750 M N-ethylacetamide (HPLC external standard solution) and sufficient L0 M HCL to lower the pH of the sample to ca. pH 2.5. The resulting solution was filtered and analyzed by HPLC. After 5.0 h, the HPLC
yield of 5-cyanopentanoic acid~ adipamic acid. adipamide, and adipic acid were 96.6%, 1.7%, 1.4% and 0.3%, with no adiponitrile rem~ining 5-Cyanopentanoic acid ~ammoniurn salt) The procedure described in Example 23 was repeated, using 0.4338 g (0.457 mL, 4.01 mmol, 0.401 M) of adiponitrile and 3.0 mL of the heat-treated Comamonas testosteroni 5-MGAM-4D (ATCC 55744) cell suspension (0.20 g wet cell weight), in a total volume of 10.0 mL (adjusted with potassium phosphate buffer (20 mM, pH 7.0)). Samples (0.100 mL) were withdrawn, diluted l :4 with water and centrifuged, then 0.180 mL of the supernatant was placed in a Millipore Ultrafree-MC filter unit (10 K MWCO) and mixed with 0.020 rnL of an aqueous solution of 0.750 M N-ethyl~cet~mid~ (HPLC external standard solution) and sufficient 1.0 M HCI to lower the pH of the sample to ca. pH 2.5. The resulting solution was filtered and analyzed by HPLC. After 5.0 h, the HPLC yield of 5-cyanopentanoic acid, adipamic acid, adipamide, and adipic acid were 94.0%, 4.0%, 0.6% and 1.4%, with no adiponitrile rem~ining.

5-Cyanopentanoic acid (ammonium salt) The procedure described in Example 23 was repeated, using 1.084 g (1.14 rnL, 10.0 mmol, two-phase reaction, 1.00 M product) of adiponitrile and 7.5 mL of the heat-treated Comamonas tesJosteroni 5-MGAM-4D (ATCC 55744) cell suspension (0.50 g wet cell weight), in a total volume of 10.0 mL (adjustedwith potassium phosphate buffer (20 mM, pH 7.0)). Samples (0.100 mL) were withdrawn, diluted l: I 0 with water and centrifuged, then 0.180 mL of the supernatant was placed in a Millipore Ultrafree-MC filter unit ( l O K MWCO) andmixed with 0.020 mL of an aqueous solution of 0.750 M N-ethylacetamide (HPLC external standard solution) and sufficient 1.0 M HCl to lower the pH of the sarnple to ca. pH 2.5. The resulting solution was filtered and analyzed by HPLC. After 24.0 h, the HPLC yield of 5-cyanopentanoic acid, adipamic acid, CA 022~4902 1998-11-13 W O 97144318 PCT~US97/07796 adipamide, and adipic acid were 90.0%~ 4.2%, 0.0% and 5.7%~ with no adiponitrile rem~ining 5-CYanopentanoic acid Isolation S Into a 2-~ erlenmeyer flask equipped with magnetic stir bar was placed a suspension of 4.0 g of Comamonas testosteroni 5-MGAM-4D (ATCC 55744) cells (previously heat-treated at 50~C for I h) in 1.72 L of potassium phosphatebuffer (20 mM, pH 7.0) at 27~C. To the suspension was then added with stirring 270.4 g (284.3 mL, 2.5 mole, ca. 1.25 M final product concentration) of adiponitrile, and the resulting mixture was stirred at 27~C. Samples (0.200 mL) were withdrawn, diluted 1 :5 with water and centrifuged, then 0.200 mL of the supernatant was placed in a Millipore Ultrafree-MC filter unit (10 K MWCO) and sufficient 1.0 M HCI added to lower the pH of the sample to ca. pH 2.5. The resulting solution was filtered and analyzed by HPLC. After 63 h, the hydrolysisreaction had slowed considerably, so an additional 10.0 g of the same microbial cell catalyst was added to the mixture. After 86 h, the HPLC yield of 5-cyanopentanoic acid, adipamic acid, adipamide, and adipic acid were 88.2%, 4.7%, 6.6% and 0.0%, with no adiponitrile rem~ining The reaction mixture was centrifuged, and the supernatant was filtered using an Amicon 2.5 L Filter Unit equipped with a YM- 10 filter ( l OK MWCO).
A 200-mL portion of the filtrate of the product mixture described above, cont~inin~ the 5-cyanopentanoic acid ammonium salt (1.13 M), was adjusted to pH 2.5 with 6 N HCl, then saturated with sodium chloride and extracted with 4 x 200 mL of ethyl ether. The combined ether extracts were dried over magnesium sulfate, filtered, and the solvent removed by rotary evaporation at reduced pressure. Rem~inin~ ether was removed by stirring the colorless liquid at room temperature under high vacuum (60 millitorr) for 5 h to yield 27.32 g (95%
isolated yield) of 5-cyanopentanoic acid. The 5-cyanopentanoic acid was then distilled under vacuum (75 millitorr) at 110-112~C without decomposition.

5-MethYI-2-Piperidone from 4-cvanoDentanoic acid ammonium salt Into a 100 mL gr~ te~l cylinder was placed 54.4 mL of an aqueous reaction mixture cont~inin~ 1.85 M 4-cyanopentanoic acid ammonium salt (0.1 mole 4-cyanopentanoic acid ammonium salt, produced by the enzymatic hydrolysis of 2-methylglutaronitrile; Example 9, filtered product mixture from reaction #5), then 12.9 mL o~concentrated ammonium hydroxide (29.3% NH3, 0.2 mole NH3) was added and the final volume adjusted to 100 mL with distilled water. The final concentrations of 4-cyanopentanoic acid ammonium salt and added ammonium hydroxide were 1.0 M and 2.0 M~ respectively. To the resulting CA 022~4902 1998-ll-13 W O 97/44318 PCTrUS97/07796 solution was added 0.631 g (5 wt. %/wt. of 4-cyanopentanoic acid) of ehromium-promoted Raney Nickel (Grace Davison Raney~ 2400 Active Metal Catalyst), and the resulting mixture charged to a 300-mL 314 SS Autoclave Engineers EZE-Seal stirred autoclave equipped with a Dispersimax~ turbine-tvpe impeller After 5 flushing the reactor with nitrogen, the contents of the reactor were stirred at 1000 rpm and heated at 160~C under 500 psig of hydrogen gas for 3 h. After cooling to room temperature. analysis of the final reaction mixture by gas chromatography indicated a 96.4% yield of 5-methyl-2-piperidone, with no 4-eyanopentanoic acid ammonium salt rem~ining.
The product mixture was filtered to remove the catalyst, then adjusted to pH 6.0 with 6 N HCI and saturated with sodium chloride. The resulting solution was extraeted five times with 100 mL of diehloro-methane, and the combined organic extracts dried over m~gnesium sulfate, filtered, and the solvent removedby rotary evaporation under reduced pressure to yield a colorless oil. After removal of the renl~ining solvent under vacuum (0. I rnm Hg), the oil crystallized to form a white solid, which was recrystallized from 150 mL of ethyl ether at -78~C to yield 6.69 g (59% isolated yield) of 5-methyl-2-piperidone (mp 55.5-56.2~C).

5-Methvl-2-Piperidone from 4-cvanopentanoic acid The reaction described in Exarnple 27 was repeated using 12.71 g (0.100 mole) of erystalline 4-cyanopentanoic acid (from an isolation deseribed in Example 8) and 19.34 mL of concentrated ammonium hvdroxide (29.3% NH3, 0.3 mole NH3) in a total volume of 100 mL. The final concentrations of 4-cyanopentanoic acid ammonium salt and ammonia were l.0 M and 2.0 M, respectively. Analysis of the final reaction mixture by gas chromatography indieated a 91.1 % yield of 5-methyl-2-piperidone, with no 4-cyanopentanoie aeidrem:~ining.

5-MethYI-2-Piperidone from 4-cvanopentanoic acid arnmonium salt Hydrogenations of 5-mL aqueous reaction mixtures cont~inin~ 1.0 M
4-eyanopentanoie acid ammonium salt (filtered product mixture from Example 9, reaction #5), 0 to 3.0 M ammonium hydroxide, and 5 wt. % or 10 wt. % (relative to weight of 4-cyanopentanoic acid) of a hydrogenation catalyst selected from a group consisting of Cr-promoted Raney nickel catalyst (Raney 2400), 5%
palladium on carbon, 10% palladium on carbon, 5% ruthenium on alumina, or 10% ruthenium on alumina were run in glass shaker tubes at 500 psig hydrogen gas and at either 160~C or 180~C were examined for the production of the corresponding lactam 5-methyl-2-piperidone (5-MPPD):

CA 022~4902 1998-11-13 W O 97/44318 PCT~US97/07796 Temp. LNH4oH] wt. %Time ~'0 '~-CPA S-MPPD
(~C)catalyst (M)catalyst (h)conversion (%) 160Raney 2400 0 5 2 100 85.6 160Raney 2400 ~.0 S 3 100 96.4 160 Raney 2400 3.0 5 ~ 100 86.5 160 5% Pd/C 2.0 5 2 10 0 160 10% Pd/C 2.0 ~ 2 14 0 160 5% Ru/A1203 2.0 5 2 15 0 160 10% Ru/A1203 2.0 5 2 20 0 180 Raney 2400 2.0 5 2 ~ 00 91.4 180 Raney 2400 3.0 5 2 100 89.5 4-EthylpYrrolidin-2-one from 3-cvanopentanoic acid amrnonium salt Into a 100 mL graduated cylinder was placed 79.4 mL of an aqueous reaction mixture cont~ining 1.26 M 3-cyanopentanoic acid ammonium salt (0.1 mole 3-cyanopentanoic acid ammonium salt, produced by the enzymatic hydrolysis of 2-ethylsuccinonitrile; Example 13 filtered product mixture), then 12.9 mL of concentrated arnmonium hydroxide (29.3% NH3, 0.2mole NH3) wa3 added and the final volume adjusted to 100 mL with distilled water. The final concentrations of 3-cyanopentanoic acid ammonium salt and added amrnonium hydroxide were 1.0 M and 2.0 M, respectively. To the resulting solution was added 0.631 g (5 wt. %/wt. of 3-cyano-pentanoic acid) of chromium-promoted Raney Nicke} (Grace Davison Raney~ 2400 Active Metal Catalyst), and the resulting mixture charged to a 300-mL 314 SS Autoclave Engineers EZE-Seal stirred autoclave equipped with a Dispersimax~ turbine-type impeller. After flushing the reactor with nitrogen, the contents of the reactor were stirred at 1000 rpm and heated at 1 60~C under 500 psig of hydrogen gas for 4 h. After cooling to room t~ cl~Lure, analysis ofthe final reaction mixture by gas chromatography indicated a 90.7% yield of 4-ethylpyrrolidin-2-one, with no 3-cyanopentanoic acid ammonium salt rem~ining 4-EthvlPvrrolidin-2-one from 3-CYanopentanoic acid The reaction described in Example 30 was repeated using 12.71 g (0.100 mole) of crystalline 3-cyanopentanoic acid (from isolation described in Example 13) and 19.34 mL of concentrated ammonium hydroxide (29.3% NH3, 0.3 mole NH3) in a total volume of 100 mL. The final concentrations of 3-cyanopentanoic acid ammonium salt and added arnmonium hydroxide were CA 022~4902 l998-ll-l3 W O 97/44318 PCTrUS97tO7796 1.0 M and 2.0 M, respectively. Analysis of the final reaction mixture (2 h reaction time) by gas chromatography indicated a 92.1 % yield of 4-ethylpyrrolidin-2-one,with no 3-cyanopentanoic acid rem:lining.
The product mixture was filtered to remove the catalyst, then adjusted to S pE~ 7.0 with 6 N HCI and saturated with sodium chloride. The resulting solution was extracted four times with 100 mL of dichloro-methane, and the combined organic extracts dried over m~gnesium sulfate, filtered, and the solvent removedby rotary evaporation under reduced ples~u,e to yield a colorless oil. After removal of the rem~ining solvent under vacuum (0.1 mm Hg), the oil was dissolved in 150 mL of ethyl ether, which was then cooled to -78~C. After 1 h, the white solid which had crystallized was collected by vacuum filtration to yield a total of 8.96 g (79% isolated yield) of 4-ethylpyrrolidin-2-one (mp 40.5-41.5~C).

4-Ethvlpvrrolidin-2-one from 3-Cvanopentanoic acid Ammonium Salt The reaction described in Example 30 was repeated exactly as described except that the temperature employed for the hydrogenation was 140~C. Analysis of the final reaction mixture (4 h reaction time) by gas chromatography indicated a 87.2% yield of 4-ethylpyrrolidin-2-one, with no 3-cyanopentanoic acid rem~ining. EXAMPLE 33 4-Ethvlpyrrolidin-2-one from 3-Cyanopentanoic acid Ammonium Salt The reaction described in Example 30 was repeated exactly as described except that 1.262 g (10 wt. %/wt of 3-cyano-pentanoic acid) of chromium-promoted Raney Nickel (Grace Davison Raney~ 2400 Active Metal Catalyst) was employed. Analysis of the final reaction mixture (1.5 h reaction time) by gas chromatography indicated a 91.0% yield of 4-ethylpyrrolidin-2-one, with no 3-cyanopentanoic acid rem~ining 4-EthvlpYrrolidin-2-one from 3-C~lanoPentanoic acid Ammonium Salt (Te~lycld~llre Dependence) Hydrogenations of 5-mL aqueous reaction mixtures cont~ining 1.0 M
3-cyanopentanoic acid (3-CPA) arnmonium salt (filtered product mixture from Example 13), 2.0 M ammonium hydroxide, and 5 wt. % or 10 wt. % of Cr-promoted Raney nickel catalyst (Raney 2400) (relative to weight of 3-cyanopentanoic acid) were run in glass shaker tubes at 500 psig hydrogen gas and at tell~p~,.dl~lres from 70~C to 180~C for 2 h, then analyzed by high pleJ~ e liquid chromatography for conversion of 3-CPA and by gas chromatography for the production of 4-ethylpyrrolidin-2-one:

Temp. 3-CPA :~-nmonillm salt [NH40H] wt.% Time 3-CPA 4-EPRD
(~C) (M) (M)Raney Ni (h)(% conv.)(% yield) 1.0 2.0 5 2 7 1 5 120 1.0 2.0 5 2 55 22.7 I ~0 1.0 2.0 5 2 71 55.4 140 1.0 2.0 10 2 100 89.9 160 1.0 2.0 5 2 100 90 1 160 1.0 2.0 10 2 100 91.3 180 1.0 2.0 5 2 100 86.1 180 1.0 2.0 10 2 100 90.0 4-EthylPvrrolidin-2-one from 3-Cyanopentanoic acid Ammonium Salt (NH40H Concentration Depçndence) 5Hydrogenations of 5-mL aqueous reaction mixtures cont~inin~ 1.0 M
3-cyanopentanoic acid (3-CPA) ammonium salt (filtered product mixture i~om Example 13), from 0 M to 3.0 M ammoniurn hydroxide, and S wt. % of Cr-promoted Raney nickel catalyst (Raney 2400) (relative to weight of 3-cyanopentanoic acid) were run in glass shaker tubes at 500 psig hydrogen gas 10and at te~l~y~ldL'lres of 160~C or 180~C for 2 h, then analyzed by high pressure liquid chroniatography for conversion of 3-CPA and by gas chromatography for the production of 4-ethylpyrrolidin-2-one:

Temp. 3-CPA ammonium salt [NH40H] wt. %Time 3-CPA 4-EPRD
(C) (M) (M)Raney Ni (h)(% conv.)(% yield) 160 1.0 0 5 2 99 80.1 160 1.0 1.0 5 2 99 87.6 160 1.0 2.0 5 2 100 90.1 160 1.0 3.0 5 2 100 85.4 180 1.0 0 5 2 100 75.1 180 1.0 1.0 5 2 100 85.8 180 1.0 2.0 5 2 100 88.5 180 1.0 3.0 5 2 100 90.0 4-Ethvlpvrrolidin-2-one from 3-Cyanopentanoic acid Ammonium Salt (3-CPA Concentration DePendence) Hydrogenations of 5-mL aqueous reaction mixtures containing 1.0 M, 1.5 M, or 2.0 M 3-cyanopentanoic acid (3-CPA) (isolated from filtered product 20mixture, Example 13), and 2.0 M, 3.0 M, or 4.0 M ammonium hydroxide, respectively, were run using as catalyst S vvt. % of Cr-promoted Raney nickel (Raney 2400) (relative to weight of 3-cyanopentanoic acid) in glass shaker tubesat 500 psig hydrogen gas and at temperatures of l 60~C or 1 80~C for 2 h, then analyzed by high pressure liquid chromatography for conversion of 3-CPA and by 5 gas chromatography for the production of 4-ethylpyrrolidin-2-one:

Temp.3-CPA ammonium salt [NH40H]wt. ~/0 Time3-CPA 4-EPRD
(~C) (M) (M)Raney Ni (h)(% conv.) (% yield) 160 1.0 2.0 5 2 100 90.1 160 1.5 3.0 5 ~ 99 76.1 1 60 2.0 4.0 5 ~ 99 80.0 1 80 1 .0 2.0 5 2 1 00 88.5 1 80 1 .5 3.0 5 2 99 77.3 1 80 2.0 4.0 5 2 99 84. 1 Caprolactam from 5-CYanopentanoic Acid Ammonium Salt Hydrogenations of 5-mL aqueous reaction mixtures cont~ining 1.0 M
5-cyanopentanoic acid (5-CPA) ammoniurn salt (prepared from enzymatic hydrolysis of adiponitrile; f1ltered product mixture from Exarnple 26), from 0 Mto 2.5 M ammonium hydroxide, and 5 wt. % of Cr-promoted Raney nickel catalyst (Raney 2400) (relative to weight of 5-cyanopentanoic acid) were run in glass shaker tubes at 500 psig hydrogen gas and at t~---p~ld~ lres of from 70~C to 160~C
for 2 h, then analyzed by high pressure liquid chromatography for conversion of 5-CPA and by gas chromatography for the production of caprolactam:

Temp.5-CPA arnmonium salt [NH40H] Time S-CPA caprolactam (~C) (M~ (M) ~) (% conv.)(% yield) 1 .0 0 2 1 00 0.7 1.0 1.0 2 100 0.7 1.0 1.5 2 94 0.8 1.0 2.0 2 84 0.8 1.0 2.5 2 99 0.8 120 1.0 0 2 99 0.9 120 1.0 1.0 2 100 0.9 120 1.0 1.5 2 97 1.0 120 1 .0 2.0 2 97 1 . I
I 60 1 .0 0 2 97 2.5 W O 97/44318 PCT~US97/07796 160 1.0 1.0 2 84 ~3 160 1.0 1.5 ~ 97 3.0 160 1.0 2.0 2 95 2.7 The cyclization of the ammonium salt of 6-aminocaproic acid (6-ACA) in the aqueous product mixtures prepared by the hydrogenation of 5-CPA
ammonium salt at 70~C (as described above) was attempted at a higher S temperature by first removing the hydrogenation catalyst by filtration, then heating the ca. l.0 M 6-ACA amrnonium salt reaction mixtures at 280~C for 2 h:

Temp. 6-ACA arnmonium salt (M) [NH40H] Tirne caprolactarn (~C) (preparedas describedabove) (M~ (h) (% yield) 280 1.0 0 2 17.8 280 1.0 1.0 2 17.2 280 1.0 1.5 2 11.0 280 1.0 2.0 2 9.0 280 1.0 2.5 2 3.1 1,5-DimethYI-~-Piperidone from 4-cyano~entanoic acid ammonium salt Hydrogenations of S-mL aqueous reaction mixtures containing 1.0 M
4-cyanopentanoic acid ammonium salt (filtered product mixture from Example 9, reaction #5), 3.0 M methylamine, and S wt. % or 10 wt. % (relative to weight of 4-cyanopentanoic acid) of a hydrogenation catalyst selected from a group 15 con.ci.C~ing of Cr-promoted Raney nickel catalyst (Raney 2400), 5% palladium on carbon, 5~/O palladium on alumina, 5% rutheniurn on alumina, or 4.5%
palladium/0.5% platinum on carbon were run in glass shaker tubes at 500 psig hydrogen gas and at 1 60~C for 2 h, then analyzed for the production of 1,5-dimethyl-2-piperidone (5-DMPD) and 5-methyl-2-piperidone (5-MPPD):

Temp. wt. % Time4-CPA S-DMPD 5-MPPD
(~C) catalystcatalyst (h)(% conv.)(% yield)(% yield) 160 Raney 2400 5 2 100 19.2 68.5 160 Raney 2400 10 2 100 21.7 69.1 160 5 % Ru/A1203 5 2 100 19.0 63.5 160 5 % Pd/C 5 2 99 81.7 7.8 160 5 % Pd/A1203 5 2 93 77.2 4.6 160 4.5 % Pd/0.5 %Pt/C 5 2 97 77.8 6.2 CA 02254902 1998-ll-13 W O 97/44318 PCTrUS97/07796 I~S-Dimethvl-2-Piperidone from 4-cvanopentanoic acid ammonium salt Hydrogenations of 5-mL aqueous reaction mixt~res cont~ininE 1.0 M
4-cyanopentanoic acid ammoniurn salt (filtered product mixture from Exarnple 9, reaction #5), from 1.0 M to 3.0 M methylamine. and 5 wt. % (relative to weight of 4-cyanopentanoic acid) of 5% palladium on carbon were run in glass shaker tubes '' at 500 psig hydrogen gas and at 140~C for 2 h, then analyzed for the production of 1,5-dimethyl-2-piperidone (S-DMPD), S-methyl-2-piperidone (S-MPPD)~ and 2-methylglutaric acid (2-MGA):

Temp. time 4-CPA S-DMPD 5-MPPD '-MGA
(~C)catalyst(h) [CH3NH~](M) (% conv.) (% yieldl (% yield) (% yield) 1405% Pd/C 2 1.0 100 53.3 0 0.7 1405% Pd/C 2 1.25 100 65.8 0 0.7 1405% PdtC 2 1.5 99 74.5 0 0.8 1405% PdlC 2 2.0 98 80.2 0 1.0 1405% PdlC 2 3.0 99 83.4 0 1.1 1~5-Dimethvl-2-Piperidone from 4-cyanopentanoic acid ammonium salt Hydrogenations of S-mL aqueous reaction mixtures cont~inin~ 1.0 M
15 4-cyanopc~ oic acid arnmonium salt (filtered product mixture from Exa~nple 9,reaction #5), from 1.0 M to 4.0 M methylamine, and 5 wt. % (relative to weight of 4-cyanopentanoic acid) of 5% palladiurn on carbon were run in glass shaker tubesat 500 psig hydrogen gas and at 1 60~C for 2 h, then analyzed for the production of 1,5-dimethyl-2-piperidone (5-DMPD), 5-methyl-2-piperidone (5-MPPD), and 20 2-methylglutaric acid (2-MGA):

Temp. time[CH3NH2] 4-CPA 5-DMPD S-MPPD 2-MGA
(~C)catalyst (h) (M) (% conv.)(% yield)(% yield)(% yield) 1605% Pd/C 2 1.0 100 53.5 0 0.7 1605% PdlC 2 1.25 100 68.3 0 0.8 1605% Pd/C 2 1.5 100 76.4 0 0.9 1605% PdlC _ 2.0 100 81.5 0 1.8 1605% PdlC 2 3.0 99 81.7 7.8 3.3 1605% Pd/C 2 4.0 99 88.4 1.9 2.6 W O 97/44318 PCTrUS97/07796 1.5-Dimethvl-2-Piperidone from ~-cvanopentanoic acid ammonium salt Hydrogenations of 5-mL aqueous reaction mixtures cont~3ining 1.0 M
4-cyanopentanoic acid arnmonium salt (filtered product mixture from Exarnple 9, 5 reaction #5), from 3.0 M to 4.0 M methylamine. and S wt. % (relative to weight of 4-cyanopentanoic acid) of either 5% palladium on carbon or 4.5% palladiurn/0.5%
platinum on carbon were run in glass shaker tubes at 500 psig hydrogen gas and at 1 80~C for 2 h, then analyzed for the production of I ,5-dimethyl-2-piperidone (5-DMPD), 5-methyl-2-piperidone (5-MPPD), and 2-methylglutaric acid 10 (2-MGA):

Temp [CH3NH2] 4-CPA 5-DMPD 5-MPPD 2-MGA
~~C) catalyst (M~ (%conv.) (% yield) (% yield) (% yield~
180 5 % P~C 3.0 97 73.0 6.6 180 4.5 % P~0.5% P~C 3.0 96 69.0 1.9 23.9 180 5 % P~C 4.0 95 66.6 2.6 33.5 1.5-Dimethvl-2-Piperidone from 4-Cvanopentanoic Acid Ammoniurn Salt Into a 100 mL gr~d1l~ted cylinder was placed 54.4 mL of an aqueous reaction mixture Cont~ining 1.84 M 4-cyanopentanoic acid ammonium salt (0.1 mole 4-cyanopentanoic acid ammonium salt, produced by the enzymatic hydrolysis of 2-methyl~luL~onitrile; Example 9, filtered product mixture from reaction #5), then 25.8 ml of 40 wt. % methylamine (9.31 g methylamine, 0.3 mole) was added and the final volume adjusted to 100 mL with distilled water.
The final concentrations of 4-cyanopentanoic acid ammonium salt and methylamine were 1.0 M and 3.0 M, respectively. To the resulting solution was added 0.636 g (~ wt. %/wt. of 4-cyanopentanoic acid) of 5% Pd on carbon powder, and the resulting mixnlre charged to a 300-mL 314 SS Autoclave Engineers EZE-Seal stirred autoclave e~uipped with a Dispersimax~ turbine-type impeller. After flushing the reactor with nitrogen, the contents of the reactor were stirred at 1000 rpm and heated at 1 60~C under 500 psig of hydrogen gas for 4 h.Samples (ca. 1.5 mL) were removed via a sampling tube over the course of the reaction for analysis. After cooling to room temperature, analysis of the final reaction mixture by gas chromatography indicated a 72.8% yield of 1,5-dimethyl-2-piperidone, 3.5% 5-methyl-2-piperidone, 19.9% 2-methylglutaric acid, and no 4-cyanopentanoic acid ammonium salt ~ ining The product mixture (61 mL after sampling) was filtered to remove the catalyst, then adjusted to pH 7.0 with 6 N HCI and saturated with sodium chloride.

CA 022~4902 1998-11-13 The resulting solution was extracted four times with l O0 mL of ethyl ether, andthe combined organic extracts dried over magnesium sulfate~ filtered. and the solvent removed by rotary evaporation under reduced pressure to yield a colorless liquid. This liquid was distilled at 3.5 Torr and the fraction boiling at 70.0-71 .5~C
collected to yield 4.65 g (60% isolated yield) of 1,5-dimethyl-~-piperidone (S-DMPD).

4-Ethvl-l-MethYlpYrrolidin-2-one from 3-Cvanopentanoic Acid Ammonium Salt Into a 100 mL gr~ ted cylinder was placed 79.4 mL of an aqueous reaction mixture containing 1.26 M 3-cyanopentanoic acid ammonium salt (0.1 mole 3-cyanopentanoic acid ammonium salt, produced by the enzymatic hydrolysis of 2-ethylsuccinonitrile; Example 13 filtered product mixture), then 17.2 mL of 40 wt. % methylamine (6.21 g methylamine, 0.2 mole) was added and the final volume adjusted to 100 mL with distilled water. The final concentrations of 3-cyanopentanoic acid ammonium salt and methylamine were 1.0 M and 2.0 M, respectively. To the resulting solution was added 0.636 g (5 wt. %/wt. of3-cyanopentanoic acid) of 5% Pd on carbon powder, and the resulting mixture charged to a 300-mL 314 SS Autoclave Engineers EZE-Seal stirred autoclave equipped with a Dispersimaxt turbine-type impeller. After flushing the reactor ~vith nitrogen, the contents of the reactor were stirred at 1000 rpm and heated at 140~C under 500 psig of hydrogen gas for 4 h. Samples (ca. 1.5 mL) were removed via a sampling tube over the course of the reaction for analysis. After cooling to room temperature, analysis of the final reaction mixture by gas chromatography indicated a 69.8% yield of 4-ethyl-1-methylpyrrolidin-2-one and a 20.4% yield of 4-ethylpyrrolidin-2-one, with no 3-cyanopentanoic acid ammonium salt renl~ining.
The product mixture (80 mL after sampling) was filtered to remove the catalyst, then adjusted to pH 7.0 with 6 N HCl and saturated with sodium chloride.
The resulting solution was extracted four times with 100 mL of dichloro-methane,and the combined organic extracts dried over m~gnesiurn sulfate, filtered, and the solvent removed by rotary evaporation under reduced pressure to yield a colorless liquid. This liquid was fractionally-distilled at 16 Torr, and the fraction boiling at 100~C was collected (5.15 g, 51% yield). The resulting 4-ethyl-1-methyl-pyrrolidin-2-one contained ~5% 4-ethylpyrrolidin-2-one as impurity, so the liquid was redistilled at 40 Torr and the fraction boiling at 128~C collected to yield 3.71 g (37% isolated yield) of 4-ethyl- 1-methylpyrrolidin-2-one (4-EMPRD).

CA 022~4902 1998-11-13 W O 97/44318 PCTrUS97/07796 5-MethYI-2-Pi~eridone from 4-Cyano-4-pentenoic Acid Ammonium Salt Into a l 00 mL graduated cylinder was placed 77.0 mL of an aqueous reaction mixture cont~inin~ 1.30 M 4-cyano-4-pentenoic acid ammonium salt 5 (0.1 mole 4-eyanopentanoic acid ammonium salt, produced by the enzymatic hydrolysis of 2-methyleneglutaronitrile; Example 17), then 12.9 mL of eoneentrated ammonium hydroxide (29.3% N~I3, 0.2 mole NH3) was added and the final volume adjusted to 100 mL with distilled water. The final concent~ations of 4-eyano-4-pentenoic acid ammonium salt and added ammonium hydroxide were 1.0 M and 2.0 M, respectively. To the resulting solution was added 0.626 g (5 wt. %/vrt. of 4-cyano-4-pentenoic acid) of ehromium-promoted Raney Niekel (Graee Davison Raney~ 2400 Aetive Metal Catalyst), and the resulting mixture eharged to a 300-mL 314 SS Autoelave F.ngineers LZE-Seal stirred autoelave equipped with a Dispersimax' 9 turbine-type impeller. After flushing the reactorlS with nitrogen, the contents of the reaetor were stirred at 1000 rpm and heated under 500 psig of hydrogen gas at 50~C for 5 h, then for an additional 3 h at 1 60~C. After cooling to room tenlp~Lure, analysis of the final reaction ~lliX~ e by gas ehromatography indieated a 85.0% yield of 5-methyl-2-piperidone, with no 4-eyano-4-pentenoie acid ammonium salt rem~ining.

2-PYrrolidinone from 3-Cvanopropionie Acid Ammonium Salt Into a l O0 mL graduated cylinder was plaeed 75.8 mL of an aqueous reaetion mixture eont~ining l .3 l M 3-cyanopropionic acid ammonium salt (0.1 mole 3-cyanopropionic aeid ammonium salt, produeed by the enzymatie hydrolysis of succinonitrile; Example 20), then l 9.4 mL of eoneentrated ammonium hydroxide (29.3% NH3, 0.3 mole NH3) was added and the final volume adjusted to l O0 mL with distilled water. The final eoneentrations of 3-eyanopropionie acid ammonium salt and added ammonium hydroxide were l.0 M and 3.0 M, respectively. To the resulting solution was added 0.99 g (lO wt.
%/wt. of 3-cyanopropionic aeid) of chromium-promoted Raney Nickel (Graee Davison Raney~' 2400 Active Metal Catalyst), and the resulting mixture charged to a 300-mL 314 SS Autoelave Fngineers EZE-Seal stirred autoclave equipped v~ith a Dispersimax~ turbine-type impeller. After flushing the reaetor with nitrogen, the eontents of the reactor were stirred at 1000 rpm and heated under 500 psig of hydrogen gas at 70~C for 4.5 h, then for an additional 5 h at l ~0~C.
Analysis by gas chromatography indieated a 9 l .0% yield of 2-pyrrolidinone, with no 3-cyanopropionic acid ammonium salt r~m~inin~

CA 022~4902 1998-ll-13 2-Piperidone from 4-CvanobutYric Acid Ammonium Salt The procedure described in Example 22 was repeated. After 4.0 h~ the HPLC yield of 4-cyanobutyric acid arnmonium salt and glutaric acid diammonium 5 salt was 91.7% and 7.5%, respectively~ with no glutaronitrile rem~ining The final concentration of 4-cyanobutyric acid ammoniurn salt in the centrifuged and filtered reaction mixture was 1.42 M. Into a 100 mL graduated cylinderwas placed 70.6 mL (0.100 mole of 4-cyanobutyric acid amrnonium salt) of the filtered aqueous reaction mixture, then 19.4 mL of concentrated ammonium hydroxide (29.3% NH3, 0.3 mole NH3) was added and the final volume adjusted to 100 mL
with distilled water. The final concentrations of 4-cyanopropionic acid ammonium salt and added ammonium hydroxide were 1.0 M and 3.0 M, lei,~e~ilively. To the resulting solution was added 1.13 g (10 wt. %/wt. of 4-cyanobutyric acid) of chromium-promoted Raney Nickel (Grace Davison Raney'~' 2400 Active Metal Catalyst), and the resulting mixture charged to a 300-mL 314 SS Autoclave Engineers EZE-Seal stirred autoclave equipped with a Dispersimax~' turbine-type impeller. After flushing the reactor with nitrogen, the contents of the reactor were stirred at 1000 rpm and heated under 500 psig of hydrogen gas at 70~C for 3.5 h. Analysis by gas chromatography indicated a 29.7% yield of 2-piperidone, with no 4-cyanobutyric acid ammonium salt rcn~ining. The l~nlpeldture was increased to 1 80~C for an additional 2 h, and subsequent analysis of a sample by gas chromatography indicated a 93.5% yield of 2-pyrrolidinone.

5-Methvl-2-PiPeridone from 4-CvanoPentanoic acid Ammonium Salt ~ifty milliliters of an aqueous mixture cont~ining 1.85 M
4-cyanopentanoic acid ammonium salt (0.1 mole 4-cyanopentanoic acid ammonium salt, produced by the enzymatic hydrolysis of 2-methylglutaronitrile;
Example 9, filtered product mixture from reaction #5), 5.6 g of 29% aqueous ammonium hydroxide and 44 mL D.I. water was charged to a 300 mL autoclave.
To this solution was added 0.73 g (3 wt% based on 4-cyanopentanoic acid) of 4.5% Pd/0.5% Pt on carbon catalyst. The autoclave was sealed and purged 3 times with hydrogen followed by heating to 160~C under lO0 psig hydrogen and slow stirring. At 1 60~C, the pressure was raised to 800 psig and maximum stirring commenced. After 3 hours, the reactor was cooled, vented and purged with nitrogen. Gas chromatographic analysis of the product mixture indicated a 96% conversion of 4-cyanopentanoic acid and a 95.5% yield (99.5% selectivity) to S-methyl-2-piperidone.

CA 022~4902 1998-ll-13 W O 97/44318 PCTrUS97/07796 1~5-Dimethvl-2-PiPeridone from 4-Cvanopentanoic acid Ammonium Salt One hundred milliliters of an aqueous mixture containing 1.85 M
4-cyanopentanoic acid ammonium salt (0.2 mole 4-cyanopentanoic acid 5 ammonium salt, produced by the enzymatic hydrolysis of 2-methylglutaronitrile;Example 9, filtered product mixture from reaction #5), and 14 g ( 0.2 mole) 40%
aqueous methylamine was charged to a 300 mL autoclave. To this solution was added 4.5% PdtO.5% Pt on carbon catalyst. The autoclave was sealed and purged 3 times with hydrogen followed by heating to reaction under 100 psig hydrogen 10 and slow stirring. At reaction temperature, the pressure was raised and maximum stirring comrnenced. After a given reaction time, the reactor was cooled, ventedand purged with nitrogen. Gas chromatographic analysis of the product mixture for several runs at different reaction conditions are summarized below:

Temp. H2catalyst loading time4-CPA 5-DMPD 5-MPPD
(~C~ (psig)(wt %) (h)(% conv.) (% yield! (% yield) 175 300 7.4 3 99 67.8 28.2 160 500 10 2 99 68.6 28.1 145 500 10 2 93 65.2 26.4 1.5-DimethYI-2-Piperidone from 4-Cvanopentanoic Acid Ammonium Salt ~ydrogenations of 4-cyanopentanoic acid amrnonium salt were performed in 5 mL glass shaker tubes at 800 psig using different Pd on carbon catalysts.
20 Four milliliters of 1.85 M (7 mmoles) aqueous 5-cyanopentanoic acid (filteredproduct mixture from Example 9, reaction #5), 0.88 g (11.4 mmoles) 40%
methylamine and catalyst from a group consisting of 5% Pd/C and 4.5% Pd/0.5%
Pt/C were charged to the tube and run for 3 hours. Gas chromatographic analysis of the product mixtures, after cooling to 25~, at different reaction conditions are 25 summarized below:

Temp. catalyst loading time4-CPA 5-DMPD 5-MPPD
(~C) catalyst(wt %) (h)(% conv.) (% yield) (% yield) 160 5% Pd/C 1.2 2 99 91.8 5.0 t 60 4.5% Pd/0.5% Pt/C 0.7 2 99 94.0 3 . I
1 50 5% Pd/C 0.7 2 99 92.5 2.2 . , , W O 97/44318 PCTrUS97/07796 INDICATIONS RELATING TO A DE:POSITED MICROORGANISM
(PCTRule 13bis) A. The ' made below relate to the ~h"uul_ referred to in lhe d.s.
on page 9 , line B. IDENTIFICATION OF DEPOSIT Furthcr dcposils are identirled oll an ad~ ll31 sl1ecl O
Name of deposi~ary instiîution AMERICAN TYPE CULTURE COLLECTION
Address of depositary institution (including postal code and country) 12301 Parklawn Drive Rockville, Maryland 20852 US

Dalc of deposil Accession Number 08 March 1996 (08.03.96) 55744 C. ADDITIONAL INDICATIONS ~leave blank If not applicable) This inlorm3~ion is eontinued on an additional sheet O
In respect of those designations in which a European patent is sought, a sample of the deposited microorganism will be made available until the publication of the mention of the grant of the European patent or until the date on which the application has been refused or withdrawn or is deemed to be withdrawn, only by the issue of such a sample to an expert nomillated by the person requesting the sample. (Rule 23(4) EPC) D. DESIGNATED STATES FOR WHICH INDICATIONS ARE MADE (if t/le indlcations are notfor all de.~iRnated StatesJ

E. SEPARATE FURNISHING OF INDICATIONS (lea~ blank if not applicable) Thc i ' listed below will be submitted lo lhe hlt~ utl3l Bureau later ~s/~ecilytheReneralnature ofthe uIdicatlotl. e.,e, "~Iccession Nlonber of Deposit J

For receiving Offlee use only For hlhludt~ondl ~ureau use only ~This sheet was reeeived with the ' -I ~,F ~ 3 This sheet was reccived by the Interna~ional aureau on:

AUthor~ ed ~~~f~ to Authorized orrloer ;.Jr.. ~?~ .n;3~ ~;vis;~n l orm PCT/RO/134 (luly 1992) 52/1 W O 97144318 PCTrUS97/07796 INDICATIONSRELATING TO A DEPOSITED MICROORGANISM
(PCTRulc 13bis) A. The ~ made below relate to the .,.,c.uu.,, rcferred to in the d~ liun onpage 9 , line B. IDENTIFICATION OF DEPOSITFurthcr dcposits arc idcntificd on an additional shcct O
Naine of dcpositary institution AMERICAN TYPE CULTURE COLLECTION
Address of depositary institution (including poslal code and coun~ry) 12301 Parklawn Drive Rockville, Maryland 20852 US

Datc of dcpositAccession Numher 08 March 1996 (08.03.96) 55745 C. ADDITIONAL INDICATIONS (leave olonk if not opplicable) This informalion is continued on an additional sheet O
In respect of those designations ln which a European patent is sought, a sample of the deposited microorganism will be made available until the pu~lication of the mention of the grant of the European patent or until the date on which the application has been refused or withdrawn or is deemed to be withdrawn, only by the issue of such a sample to an expert nominated by the person requesting the sample. (Rule 28(4) EPC) D. DESIGNATED STATES FOR WHICH INDICATIONS ARE MADE (If thc indlcations are notfor all ~s,~".nt~d States) E. SEPARATE FURNISHING OF INDICATIONS ~leove blank if not ~t~t I ' le) Thc: ' ' nnC listedbclowwillbesubmittedtotheI 'Burcaulatcr fspecify~hegeneralrlotLIreo/(heindicatiolLse.g~ "Accession Mlalber of Deposit'~

For receiving Office use only For Intcrnational Bureau use only ~his sheet was reccived with the; ~ ' ' ' O This shec~ was recei~ed by the I ~dli~ ' Bureau on:

Authonzed of ficer ~h't~ ' Authorized orficcr h~slon FormPCT~O/l34(Julyl992) W O 97/44318 PCT~US97/07796 INDICATIONS RELATING TO A DE:POSITED MICROORGANISM
(PCTRule 13bis) A. The " - made below rclale to the uu~ l rcfcrrcd tû in the d~ tiu on page 9 , linc B. IDENTIFICATION OF DEPOSIT Furthcr dcposils arc idcntificd on an no iilional shcct O
Narnc of dcpositary institution AMERICAN TYPE CULTURE COLLECTION
Addrcss of dcpositary institution (inC/Ydlt1g poslal code ond counlry)12301 Parklawn Drive Rockville, Maryland 20852 US

Datc of deposit Accession Numbcr 08 March 1996 (08.03.96) 55746 C. ADDITIONAL INDICATIONS ~leoYc blonk if not applicable) This information is continued on an additional shcet O
In respect of those designations in which a European patent is sought, a sample of the deposited microorganism will be made available until the publication of the ~ention of the grant of the European patent or until the date on which the application has been refused or withdrawn or is deemed to be withdrawn, only by the issue of such a sample to an expert nominated by the person requesting the sample. (Rule 28(4) EPC) D. DESIGNATED STATES FOR WHICH INDICATIONS ARE MADE (if tllc indications are nolfor oll desigrla~ed Sla~es) E. SEPARATE FURNISHING OF INDICATIONS (Icave blank if not applicablc) Thc~ clistedbelowwilibcsubmittcdtothel 'Burcaulatcr(s/~ecifylllegenerallla~ureoJ~llellldica~io/lseg."~lcccss~on Nl~nlber of Deposil") For receiving omcc usc only For I ~ ~ Burcau use only ~Ihis shcct was rcceived with the il~t~ dtiollal - r r ~ " O This sheel was reccivcd by the Intcrnafiollal [lurcau on:

Authorizcd o~lccr Authorized orficer ;~us~tn ~Jhit~
~&T !~ na! ~hJ~cian ForrnPCT/RO/134(July 1992) W O 97/44318 PCT~US97/07796 INDICATIONS RELATING TO A DE:POSITED MICROORGANISM
(PCTRulc 13bis) A. The ' ~ 'i made below relate to the ~.u.~, rcferred lo in Ihe d~ tiu~.
onpage 9 , line B. IDENTIFICATION OF DEPOSIT ~'urthcr dcposits are identificd on an adciiliollnl shcct O
Name of depositary institution AMERICAN TYPE CULTURE COLLECTION
Address of depositary institution (includ ng posial code and coun~ry) 12301 Parklawn Drive Rockville, Maryland 20852 US

Dalc of deposit Accession Numbcr 08 March 1996 (08~03.96) 55747 C. ADDITIONAL INDICATIONS (leave blank if noi orr~;C~ "e) This ;nfu~ tlull is continued on an additional sheet O
In respec~ of those designations ill which a European patent is sought,a sample of the deposited microorganism will be made available until the pu~lication of the mention of the grant of the European patent or until the date on which the application has been refused or withdrawn or is deemed to be withdrawn, only by the issue of such a sample to an expert nominated by the person requesting the sample. (Rule 28(4) EPC) D. DESIGNATED STATES FOR WHICH INDICATIONS ARE: MADE; (i~ C i . '' '~v are notfor oll designated ~ia~es) E. SEPARATE FURNISHING OF INDICATIONS (leave blank if no~ o ic ~) Thein~ p~innclistcdbclowwillbesubmittedtothcl ~ ' 'Burcaulatcr(spec,i~lhegeneralnatureofiheirdica~ionse.g~ ccession Noo~ber of Deposil ) For rcceiving Of ficc usc only For I ' '~ ~' Bureau use oniy ,5~This sheet was received with the ' ~ O This shect was received by the International Bureau on:

Authorizedofficer ~ n ~1~ Authorizedo~ficer i7t~ h~isTon Forrn PC,T/RO/134 (July 1992) ~;2/4

Claims (26)

Claims:

1. A process for the preparation of five-membered ring lactams or six-membered ring lactams from aliphatic .alpha.,.omega.-dinitriles, comprising:
(a) contacting an aliphatic .alpha.,.omega.-dinitrile in an aqueous reaction mixture with an enzyme catalyst characterized by either 1) an aliphatic nitrilase activity, or 2) a combination of nitrile hydratase and amidase activities, whereby the aliphatic .alpha.,.omega.-dinitrile is converted to an .omega.-nitrilecarboxylic acid ammonium salt;
(b) contacting the aqueous product mixture resulting from step (a) with hydrogen and a hydrogenation catalyst, whereby the .omega.-nitrilecarboxylic acid ammonium salt is converted directly to the corresponding lactam without isolation of the intermediate .omega.-nitrilecarboxylic acid, .omega.-nitrilecarboxylic acid ammonium salt, .omega.-aminocarboxylic acid or .omega.-aminocarboxylic acid ammonium salt; and (c) recovering the lactam from the aqueous product mixture resulting from step (b).

2. The process of Claim 1, wherein the aliphatic .alpha.,.omega.-dinitrile has the formula NCCX a (R)(CH2)n CN, where a = 1, where X = hydrogen, and R = H, alkyl or substituted alkyl, or alkenyl or unsubstituted alkenyl and where a = 0, R = alkylidene or substituted alkylidene, and where n = 1 or 2.

3. The process of Claim 1 wherein the enzyme catalyst is in the form of whole microbial cells, permeabilized microbial cells, one or more cell components of a microbial cell extract, partially purified enzyme(s), or purified enzyme(s).

4. The process of Claim 3 wherein the enzyme catalyst is characterized by an aliphatic nitrilase activity and is in the form of whole microbial cells of Acidovorax facilis 72-PF-15 (ATCC 55747), Acidovorax facilis 72-PF-17 (ATCC 55745), or Acidovorax facilis 72W (ATCC 55746).
\

5. The process of Claim 3 wherein the enzyme catalyst is characterized by a combination of nitrile hydratase and amidase activities and is in the form of whole microbial cells of Comamonas testosteroni 5-MGAM-4D (ATCC 55744).

6. The process of Claim 3 wherein the enzyme catalyst is in the form of whole microbial cells immobilized in or on an insoluble support.

7. The process of Claim 3 wherein the enzyme catalyst is partially-purified or purified enzyme(s) immobilized on a soluble or insoluble support.

8. The process of Claim 1 wherein the aliphatic .alpha.,.omega.-dinitrile is unsymmetrically substituted at the .alpha.-carbon atom, and the enzyme catalyst is characterized by aliphatic nitrilase activity that produces the .omega.-nitrilecarboxylic acid ammonium salt resulting from regioselective hydrolysis of the .omega.-nitrile group.

9. The process of Claim 1 further comprising adding ammonium hydroxide, ammonia gas or methylamine to the aqueous product mixture containing the .omega.-nitrilecarboxylic acid ammonium salt prior to the hydrogenation reaction of step (b).

10. The process of Claim 9 wherein the amount of ammonium hydroxide, ammonia gas or methylamine added to the aqueous product mixture is from 0 to 4 molar equivalents relative to the amount of .omega.-nitrilecarboxylic acid ammonium salt present.

11. The process of Claim 9 wherein methylamine is added to the aqueous product mixture containing the .omega.-nitrilecarboxylic acid ammonium salt prior to the hydrogenation reaction of step (b) and the product of the hydrogenation reaction is an N-methyllactam.

12. The process of Claim 1 wherein the temperature of the hydrogenation reaction is from 45 °C to 200 °C.

13. A process for the preparation of five-membered ring lactams or six-membered ring lactams from aliphatic .alpha.,.omega.-dinitriles, comprising:(a) contacting an aliphatic .alpha.,.omega.-dinitrile of either the formula:

or where R1 and R2 are both H, and;
R3, R4, R5 and R6 are each independently selected from the group consisting of H, alkyl or substituted alkyl, or alkenyl or substituted alkenyl, or R3 and R4 taken together are alkylidene or substituted alkylidene, or independently R5 and R6 taken together are alkylidene or substituted alkylidene, in an aqueous reaction mixture with an enzyme catalyst selected from the group consisting of:
whole cells of Acidovorax facilis 72W (ATCC 55746), whole cells of Acidovorax facilis 72-PF-15(ATCC 55747), whole cells of Acidovorax facilis 72-PF-17(ATCC 55745), and whole cells of Comamonas testosteroni 5-MGAM-4D
(ATCC 55744), whereby the aliphatic .alpha.,.omega.-dinitrile is converted to an .omega.-nitrilecarboxylic acid ammonium salt;
b) contacting the aqueous product mixture resulting from step (a) with hydrogen and a hydrogenation catalyst, whereby the .omega.-nitrilecarboxylic acid ammonium salt is converted directly to the corresponding lactam without isolation of the intermediate .omega.-nitrilecarboxylic acid, .omega.-nitrilecarboxylic acid ammonium salt, .omega.-aminocarboxylic acid or .omega.-aminocarboxylic acid ammonium salt; and c) recovering the lactam from the aqueous product mixture resulting from step (b).

14. A process for the preparation of five-membered ring lactams or six-membered ring lactams from aliphatic .alpha.,.omega.-dinitriles, comprising:(a) contacting an aliphatic .alpha.,.omega.-dinitrile of either formula or where R7, R8, R9, R10, R11 and R12 are each independently selected from the group consisting of H, alkyl or substituted alkyl, or alkenyl or substituted alkenyl, or R9 and R10 taken together are alkylidene or substituted alkylidene, or independently R11 and R12 taken together are alkylidene or substituted alkylidene, and where either or both R7 or R8 is not H;

in an aqueous reaction mixture with Comamonas teatosteroni 5-MGAM-4D (ATCC 55744), whereby the aliphatic .alpha.,.omega.-dinitrile is converted to an .omega.-nitrilecarboxylic acid ammonium salt;
b) contacting the aqueous product mixture resulting from step (a) with hydrogen and a hydrogenation catalyst, whereby the .omega.-nitrilecarboxylic acid ammonium salt is converted directly to the corresponding lactam without isolation of the intermediate .omega.-nitrilecarboxylic acid, .omega.-nitrilecarboxylic acid ammonium salt, .omega.-aminocarboxylic acid or .omega.-aminocarboxylic acid ammonium salt; and c) recovering the lactam from the aqueous product mixture resulting from step (b).

15. The process of Claim 13 wherein the enzyme catalyst is in the form of whole microbial cells, permeabilized microbial cells, one or more cell components of a microbial cell extract, and partially purified enzyme(s) or purified enzyme(s) derived from:
Acidovorax facilis 72W (ATCC 55746), Acidovorax facilis 72-PF-15 (ATCC 55747), Acidovorax facilis 72-PF-17 (ATCC 55745), and Comamonas testosteroni 5-MGAM-4D (ATCC 55744).

16. The process of Claim 14 wherein the enzyme catalyst is in the form of whole microbial cells, permeabilized microbial cells, one or more cell components of a microbial cell extract, and partially purified enzyme(s) or purified enzyme(s) derived from Comamonas testosteroni 5-MGAM-4D
(ATCC 55744).

17. The process of Claim 15 or 16 wherein the enzyme catalyst is in the form of whole microbial cells immobilized on an insoluble support.

18. The process of Claim 15 or 16 wherein the enzyme catalyst is partially purified or purified enzyme(s) immobilized on a soluble or insoluble support.

19. The process of Claim 13 or 14 wherein the aliphatic .alpha.,.omega.-dinitrile is unsymmetrically substituted at the .alpha.-carbon atom, and the enzyme catalyst is characterized by aliphatic nitrilase activity that produces the .omega.-nitrilecarboxylic acid ammonium salt resulting from regioselective hydrolysis of the .omega.-nitrile group.

20. Isolated microorganisms characterized by an aliphatic nitrilase activity and selected from the group consisting of Acidovorax facilis 72W
(ATCC 55746), Acidovorax facilis 72-PF-15 (ATCC 55747), Acidovorax facilis 72-PF-17 (ATCC 55745).

21. Isolated Comamonas testosteroni 5-MGAM-4D (ATCC 55744) characterized by a combination of nitrile hydratase and amidase activities.

22. A process for the preparation of five-membered ring lactams or six-membered ring lactams from an aqueous reaction mixture containing an .omega.-nitrilecarboxylic acid ammonium salt, comprising:
(a) contacting an aqueous product mixture containing an .omega.-nitrilecarboxylic acid ammonium salt of either the formula:

or where R1 and R2 are both H, and;
R3, R4, R5 and R6 are each independently selected from the group consisting of H, alkyl or substituted alkyl, or alkenyl or substituted alkenyl, or R3 and R4 taken together are alkylidene or substituted alkylidene, or independently R5 and R6 taken together are alkylidene or substituted alkylidene, with hydrogen and a hydrogenation catalyst, whereby the .omega.-nitrilecarboxylic acid ammonium salt is converted directly to the corresponding lactam without isolation of the intermediate .omega.-nitrilecarboxylic acid, .omega.-nitrilecarboxylic acid ammonium salt, .omega.-aminocarboxylic acid or .omega.-aminocarboxylic acid ammonium salt; and (b) recovering the lactam from the aqueous product mixture resulting from step (a).

23. A compound of Formula III, where M+ is either H+ or NH4+.

24. A compound of Formula IV, where M+ is either H+ or NH4+.

25. The process of Claim 1 wherein the aqueous reaction mixture of step (a) comprises:
(i) an aqueous phase containing:
(a) the enzyme catalyst characterized by either the aliphatic nitrilase activity or a combination of nitrile hydratase and amidase activities; and (b) dissolved .alpha.,.omega.-dinitrile; and (ii) an organic phase comprising undissolved .alpha.,.omega.-dinitrile.

26. A method for selecting for one of two enzyme activities selected from the group consisting of a regioselective nitrilase activity or a nitrile hydratase activity capable of catalyzing the converison of aliphatic .alpha.,.omega.-dinitriles to the corresponding .omega.-cyanocarboxylic acid ammonium salt, the method comprising:heating a whole cell catalyst, characterized by a desirable regioselective nitrilase activity or nitrile hydratase activity as well as an undesirable non-regioselective nitrilase or nitrile hydratase activity, to a temperature of about 35 °C to 70 °C for between 10 and 120 minutes wherein the undesirable non-regioselective nitrilase activity or nitrile hydratase activity is denatured and the desirable regioselective nitrilase or nitrile hydratase activity is preserved.