ISSN 20799780, Review Journal of Chemistry, 2011, Vol. 1, No. 4, pp. 385–402. © Pleiades Publishing, Ltd., 2011. Original Russian Text © V.G. Debabov, A.S. Yanenko, 2011, published in Obzornyi Zhurnal po Khimii, 2011, Vol. 1, No. 4, pp. 376–394.
Biocatalytic Hydrolysis of Nitriles V. G. Debabov# and A. S. Yanenko State Research Institute for Genetics and Selection of Industrial Microorganisms, 1st Dorozhny proezd 1, Moscow, 117545 Russia Received November 10, 2010; in final form, February 3, 2011
Abstract—Two pathways of enzymatic hydrolysis of nitriles to carboxylic acids are known today. Under the action of nitrilases, nitriles turn into carboxylic acids in a single step via the addition of two water molecules. Under the action of nitrile hydratases, nitriles turn into amides, which are then hydrolyzed by amidase to carboxylic acids. This review deals with the structure, substrate specificity, mechanisms of action, and industrial potential of these three enzymes. Examples of successful use of the nitrilehydrolyzing enzymes in the largescale manufacture of acrylamide and nicotinamide in Russia and abroad and in the industrial synthesis of αhydroxy acids (glycolic and Rmandelic acids) are presented. The stereoselectivity and regioselectivity of the enzymes make them usable in the syn thesis of chiral synthons for the production of important pharmaceuticals (statins, antimitotic agents, and enzyme inhibitors). Keywords: enzymatic hydrolysis of nitriles, nitrile hydratases, amidases, nitrilases, use in organic syn thesis. DOI: 10.1134/S2079978011030010
CONTENTS 1. Introduction 2. Enzymes Catalyzing Nitrile Conversion 2.1. Nitrile Hydratases 2.2. Nitrilases 2.3. Amidases 3. Catalytic Potential of Enzymes for Nitrile Conversion 3.1. Chemoselectivity 3.2. Regioselectivity (E/Z Selectivity) 3.3. Enantioselectivity 4. LargeScale Application of the Enzymatic Hydrolysis of Nitriles for the Synthesis of Amides and Acids 4.1. BioAcrylamide 4.2. Ammonium Acrylate 4.3. αHydroxycarboxylic Acids 4.4. Optically Active 2Methyl and 2,2Dimethylcyclopropanecarboxylic Acids and Their Amides 5. Conclusions and Outlook 1. INTRODUCTION Nitriles, or organocyanides (R–CN), are widely spread in nature. They are synthesized by plants and serve as precursors of hormones (phenylacetonitrile), storage compounds (cyanoglycosides and cyanolip ids), etc [1]. Synthetic nitriles are widely employed in organic synthesis as precursors of various amides and acids. The conventional chemical hydrolysis of nitriles requires severe reaction conditions (acidic or alkaline pH and temperature above 100°C) and is accompanied by the formation of undesired byproducts and large amounts of waste. An alternative to this process is biocatalytic nitrile hydrolysis, which takes place under # Corresponding
author (tel.: +7 (495) 315 37 47, +7 (495) 315 12 47; fax: +7 (495) 315 05 01; email:
[email protected],
[email protected],
[email protected]).
385
386
DEBABOV, YANENKO
mild conditions (neutral pH and room temperature), selectively involves only nitrile groups, and is in some cases stereoselective and/or regioselective, which is particularly significant in the synthesis of bio logically active compounds. Two natural pathways are presently known for nitrile hydrolysis to carboxylic acids. The first pathway includes the successive action of two enzymes, namely, nitrile hydratase (EC 4.2.1.84) and amidase (EC 3.5.1.4). Nitrile hydratase adds one water molecule to the nitrile, converting it into an amide, and amidase hydrolyzes the amide to an acid (Scheme 1). The second pathway is the singlestep conversion of the nitrile into an organic acid via the addition of two water molecules catalyzed by nitrilase (EC 3.5.5.1), as is shown in Scheme 1. All of these enzymes—nitrile hydratases, nitrilases, and amidases—have recently attracted considerable attention from researchers and synthetic chemists as biocatalysts for organic syntheses. Over ten catalytic processes involving nitrilehydrolyzing enzymes have already been commercialized to date.
RCN nitrile
RCONH2 amide
nitrile hydratase + H2O
+H am 2 O ida se
O H 2 se +2 trila ni
RCOOH acid Scheme 1. Two pathways of the enzymatic hydrolysis of nitriles to carboxylic acids.
The purpose of this review is to consider the current data on the mechanism of action of nitrile metab olism enzymes, their properties, and their use in the chemical and pharmaceutical industries. 2. ENZYMES CATALYZNG NITRILE CONVERSION 2.1. Nitrile Hydratases Nitrile hydratase was discovered for the first time in cells of Rhodococcus rhodochrous J1 bacteria (for merly called Arthrobacter sp. J1) in 1980 [2]. Although dozens of enzymes from various bacteria have been studied to date, Rhodococcus and other Actinomycetales are still the most widespread sources of new nitrile hydratases [3]. On the whole, the main reservoir of nitrile hydratases are prokaryotes, primarily bac teria. A computeraided search for new nitrile hydratases in the genome sequences represented in public databases permitted to identify this enzyme for the first time in a eukaryotic organism, specifically, Monosiga brevicollis (UniRef database, UniProt identifier A9V2C1). However, analysis of the gene sequence suggests that there could be a horizontal transfer of nitrile hydratase genes from proteobacteria [4]. All hitherto studied nitrile hydratases have a common structure. They are heterodimeric proteins con sisting of α and β subunits with a molar mass of 22 and 28 kDa, respectively. With increasing concentra tion, the αβ dimers often form tetramers and higher oligomers. Nitrile hydratases are metalloenzymes containing nonheme iron (Fe3+) or cobalt (Co3+) in their active sites. Accordingly, the nitrile hydratases are divided into two groups: ironcontaining (Fetype) and cobaltcontaining (Cotype) enzymes. The nitrile hydratase types differ in substrate specificity and activity, although they are highly homologous in their amino acid sequences, particularly in the regions belonging to the active site. There are several enzymes containing other metals (Zn or Cu); however, there is no evidence that these metals are involved in catalysis [5, 6]. Fetype nitrile hydratases have an interesting specific feature. The enzymes synthesized in bacteria grown in the dark are inactive and can be activated by visible light [7–9]. It was demonstrated by optical spectroscopy, chromatography, and mass spectrometry that nitrile hydratases in the inactive state contain a nitrogen monoxide molecule bound to the iron atom [10]. When exposed to visible light, nitrogen mon oxide leaves the active site, and this is accompanied by local conformational changes and by the activation of the enzyme [11]. This is the earliest known case of regulation of a bacterial enzyme by nitrogen mon REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
BIOCATALYTIC HYDROLYSIS OF NITRILES
387
oxide; however, the physiological role of this regulation has not been elucidated as yet. It was long believed that the Cotype nitrile hydratases were not photoregulatable; however, it was later established that nitro gen monoxide hampers the activity of these enzymes as well, but at a concentration 10 times higher than in the case of the Fetype nitrile hydratases [12]. It is interesting that carbon monoxide exerts the same effect on the Cotype nitrile hydratases as nitro gen monoxide; that is, it inhibits the enzyme, and the inactive complex is activated by visible light. At the same time, carbon monoxide at any concentration does not inhibit Fetype nitrile hydratases [12]. In 1997, Huang et al. [13] determined the crystal structure of photoactivated nitrile hydratase from Rhodococcus sp. R312 with 2.3 Å resolution. In the following year, the structure of Rhodococcus sp. N771 nitrile hydratase was solved with 1.7 Å resolution [11]. Both are Fetype enzymes. The structures of two Cotype nitrile hydratases were soon established [14, 15]. All proteins in these enzymes have similar struc tures. They are asymmetric units consisting of (αβ)2 heterodimers. The metal atom is located in the cen tral cavity formed by the surfaces of two different subunits. The channel between the solvent (water) and the active site is formed by the surfaces of two subunits including five amino acid residues from the α sub unit and four from the β subunit. In Rhodoccocus sp. N771 nitrile hydratase, the length of this channel is 10 Å and its width is 4 Å, which is obviously insufficient for the substrate transport. Apparently, substrate access to the active site is ensured by dynamic oscillations of the protein structure during catalysis. All of the protein ligands coordinated to the metal atom are in the α subunit. These ligands are three cysteine thiolates (αCys109, αCys112, αCys114) and two nitrogen atoms of the main polypeptide chain (αSer113, αCys114). The sixth coordination site of the metal atom in the inactive form of a Fetype nitrile hydratase is occupied by a nitrogen monoxide molecule. The αCys112 and αСys114 residues from Rhodococcus sp. N771 are posttranslationally oxidized to cysteine sulfinic acid (–SO2H) and cysteine sulfenic acid (⎯SOH), respectively [14]. The metal atom is located in the center of an octahedron in such a way that its bonds to ligands are directed to the octahedron corners. Four ligands (αCys112, αCys114 and two amide nitrogen atoms of the main chain–αSer113 and αCys109) are nearly in one plane. The bond between the metal atom and the sulfur atom of the Cys109 residue is directed axially, perpendicular to the other ligands’ plane (Scheme 2). The sixth bond (also planeaxial) is either occupied by a carbon monooxide molecule (when a Fetype enzyme is inactive) or is vacant and serves to binding with substrate in an active enzyme. The structure of the active site of the Cotype nitrile hydratases is very similar to that of the Fe type enzymes. The only slight difference is that the αSer113 in the latter is replaced by a threonine residue. ser113 HO
O N
O
N
S cys114
C O O N
O
S cys112
C
O S cys109cys109
Scheme 2. Active site of a Fetype nitrile hydratase in inactive form (with a nitrogen monoxide molecule attached to the metal atom). Activation eliminates the nitrogen monoxide molecule, thus vacating a coordination site for the interaction with the substrate. Four ligating atoms—Cys114 and Cys112 sulfur atoms and two amide nitrogen atoms of Cys109 and Ser113 of the main polypeptide chain—are in one plane.
While these enzymes are very similar, it is still unclear why some of them have an iron atom in their active site and the others have a cobalt atom, all the more so as there were successful experiments in which cobalt was substituted for iron (e.g., in nitrile hydratase from Rhodococcus sp. N771) and the enzyme par tially retained its activity [16]. This suggests that the Fe and Cocontaining active sites have the same structure and that the mechanisms of their catalytic action are likely identical. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
388
DEBABOV, YANENKO
The mechanisms of the posttranslational modification of nitrile hydratases, specifically, the inclusion of metal atoms and cysteine oxidation, have long been investigated. The structural study of the genes encoding bacterial nitrile hydratases demonstrated that the genes for the α and β subunits are located in operons containing an additional gene whose function is necessary for the successful protein assembling. The product of this gene was given the name of an activator [17]. For the ironcontaining nitrile hydratases from Rhodococcus sp. N771 [18], Pseudomonas chloraraphis B32 [19], Pseudomonas putida 5B [20], and Rhodococcus sp. N774 [21], it was demonstrated that these genes are indispensable for functional expression. The presence of such genes was also demonstrated for cobalt containing nitrile hydratases [2224]. Recently, a detailed study of the enzyme maturation mechanism called selfsubunit swapping using lowmolecularweight nitrile hydratase from Rhodococcus rhodochrous J1 has been performed [25]. It was demonstrated that expression in the heterological system of only the structural genes for the α and β sub units of nitrile hydratase yields an inactive devoid of cobalt enzyme. However, if the complete operon con taining the activator gene (nhlE) along with the genes encoding the α and β subunits is expressed in the same system, the product is a active enzyme with a cobalt content of 0.88 ± 0.04 mol/(mol αβ dimer). During the enzyme purification from cells bearing the whole operon, the preparations were found to con tain αβ, α2β, and αe2 proteins. The αe2 protein shows no nitrile hydratase activity, but it contains cobalt (0.87 ± 0.03 mol/(mol αe2)). This protein was given the name of holoαe2. If αe2 is combined in vitro with an inactive cobaltfree nitrile hydratase and the mixture is incubated, the enzymatic activity in the mixture grows to reach its maximum in 12 h. These results could be explained in two ways: either cobalt is trans ferred from αe2 to the α subunit of α2β2, or the two proteins exchange their subunits. Experiments on labeled proteins demonstrated that the second mechanism, referred to as selfsubunit swapping, actually takes place [25] (Scheme 3). I
II αI
e
Co+2
αI e
αI
e
αIβ
IV
III
αIIβ αIe (αIIβ)2
αIe2
αIβ
αIβ αIβ
αIIe2
Scheme 3. Posttranslational maturation of cobaltcontaining nitrile hydratase from Rhodococcus rhodochrous J1 [25]: I— synthesis of separate polypeptide units; II—formation of subunit complexes, where α I is an unmodified subunit contain ing no metal; III—inclusion of Co+2 in the α subunit of the αe2 complex, Co+2 oxidation to Co+3, and the oxidation of two cysteines to sulfenic and sulfinic acids (the modified α subunit is designated αII); IV—exchange of αI and αII sub units between αIβ and αIIe2 and the formation of a an active enzyme (αIIβ)2.
It was also demonstrated that cysteine oxidation occurs exactly in the protein αIe2, not in αβ or α2β2 synthesized in the absence of the NhlE protein. The driving force of subunit exchange is likely the two salt bridges formed between the deprotonated cysteine sulfinic and cysteine sulfenic residues of the α subunit (CysSO2 and CysSO–) [26] and the two arginine residues of the β subunit, which are in conservative positions in all nitrile hydratases (R52 and REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
BIOCATALYTIC HYDROLYSIS OF NITRILES
389
R157). It was demonstrated in later works that the NhlE protein shows two activities: it can insert cobalt and oxidize cysteines; that is, it acts as a metallochaperone, including the oxidation function, and as a chaperone ensuring subunit exchange. The oxygen atoms involved in cysteine oxidation come from water, not from dioxygen [27]. It has recently been demonstrated that the maturation of highmolecularweight nitrile hydratase from R. rhodochrous J1 proceeds via the same mechanism. The chaperone protein in this process is NhhG [28]. For practical purposes, it proved to be possible to assemble a fullvalue protein in heterological expres sion systems, such as in E. coli, without employing an activator protein. The assemblage and modification of the protein are ensured by, e.g., GroESL chaperones of E. coli [29]. The specific activity of the protein assembled in the heterological system does not differ from that of the natural protein. There are several hypotheses as to the mechanism of nitrile hydrolysis by nitrile hydratases. Nearly all of them consider the metal ion in the active site of an enzyme as a Lewis acid. Huang et al. [13] suggested three schemes according to which functional groups from “three spheres” of the active site are involved in the nitrile activation. In the first scheme, the nitrile binds directly to the metal ion and thereby activates it for hydration (first sphere—the substrate is one of the ligands for the metal). This is followed by a nucleo philic attack of water on the activated carbon atom of the nitrile. In the second scheme, the carbon atom experiences a nucleophilic attack by the hydroxide ion bound to the metal (second sphere). In the third scheme, the metalbound hydroxide ion activates a water molecule (third sphere), which then acts as a nucleophile, attacking the nitrile carbon atom. The last mechanism does not involve the intermediates bound to the enzyme, ruling out ligand exchange around the metal ion. There is an extensive evidence in favor of the first mechanism, namely, the direct interaction between the nitrile and the metal ion of the active site. It was demonstrated that iodoacetonitrile, an analog of the substrate, is involved in the straight interaction with the metal ion [30]. Xray crystallographic studies of the P. thermophila nitrile hydratase and butyric acid complex (a weak competitive inhibitor) showed that the carboxyl oxygen atom binds directly to the metal ion [31]. The direct nitrile–metal binding is also indicated by adsorption and EPR spectroscopic data [32]. The catalytic process undoubtedly involves modified thiol residues. The nitrile hydratase reconstructed under anaerobic conditions (in an argon atmosphere) is catalytically inactive and acquires activity upon aerobic incubation. This is accompanied by the oxidation of the cysteines αС112SO2H and αС114SOH [33]. It is interesting that the oxidation of these cystenes causes the total inactivation of the enzyme, as was demonstrated by the irreversible inhibition of Rhodococcus sp. N771 nitrile hydratase with 2cyano2pro pyl hydroperoxide [34]. The best substantiated nitrile hydratase action mechanism is that suggested by Hashimoto et al. [35], for it relies on direct Xray structure data relevant to the mechanism of isonitrile conversion into amines catalyzed by the same nitrile hydratase. These authors discovered that Rhodococcus sp. N771 nitrile hydratase can convert isobutyl isonitrile into isobutylamine [36]. It turned out that isonitrile–enzyme binding is almost the same as in the case of the corresponding nitrile (as was indicated by comparable Km values), but the isonitriletoamine conversion rate (Vmax) is 10000 times lower. Making use of this cir cumstance, the authors carried out an elegant study including a series of consecutive Xray diffraction characterizations of the intermediates of the enzymatic reaction [35]. The results of this study provided direct evidence that the isonitrile is coordinated to the metal of the enzyme active site. With the similarity of the Km values for the nitrile and isonitrile taken into account, these results provide another argument in favor of the direct interaction of nitriles with the metal atom. According to the mechanism suggested in that study, the carbon atom of the nitrile (or isonitrile) coordinated to the metal atom experiences nucleo philic attack by a water molecule activated by oxygen of the cysteine 114 sulfenyl group. The Cys114SO– residue serves as the catalytic base. In the case of nitriles, the intermediate imidate compound rearranges into an amide; in the case of isonitriles, into an amine and carbon monoxide (Scheme 4). Experiments and calculations on model complexes imitating the active site of nitrile hydratase demonstrated that only sulfenyl oxygen can act as a nucleophile [37]. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
390
DEBABOV, YANENKO A
R
R +
H O
N
N
N H
C
C O−
Fe+3
Fe+2
R
O
R H H
Fe+2
S
S
C
H O
H N H C
O
Fe+2
O S
O O−
Fe+3
S cys114
cys114
cys114
H N
O
C
R
S cys114
cys114
B
R
R
C
C
H O
O−
Fe+3
Fe+2
O S
S
C H
N
N
R
R
H O
C
N
H
Fe+2
O S
H O
R C O H N H
H N Fe+2
O
O−
Fe+3
S
S
cys114 Scheme 4. Mechanism of the nitrile hydratasecatalyzed conversion of isonitriles into amines (A) and of nitriles into amides (B) [35]. S–O is cysteine sulfenic acid in the α114 position, which is deprotonated in the free enzyme.
The necessity of converting the second cysteine residue into sulfinic is likely due to the fact that this residue along with other cysteines forms a “claw setting” conformation favorable for substrate capture and coordination to the metal atom [11]. On the whole, the above mechanism is supported by theoretical cal culations, which confirm that sulfenyl oxygen acts as a base removing a proton from the water molecule attacking the carbon atom of the nitrile. These theoretical calculations attribute to the metal atom the function of a stabilizer of the anionic imidate intermediate rather than the properties of the nitrileacti vating Lewis acid [38]. 2.2. Nitrilases Nitrilases (EC 3.5.5.1) are enzymes capable of hydrolyzing organic nitriles to carboxylic acids and ammonium. The first nitrilase was described in 1962 [39]. This enzyme, isolated from barley, was reported to convert indole3acetonitrile into the plant hormone auxin (indole3acetic acid). Unlike nitrile hydratases, nitrilases are observed in all kinds of organisms, including bacteria, fungi, archaea, plants, and mammals [40–42]. According to their amino acid homology, the nitrilases are assigned to the protein superfamily, also called the nitrilase superfamily. Along with the nitrilases, other enzymes hydrolyzing the C–N bond, including amidases, carbamoylases, and Nacetyltransferases, are placed in this superfamily [43]. The members of the nitrilase superfamily have a specific spatial structure. The monomers of these proteins have a fourlayer sandwich structure (αββα, where α is a helix and β is a sheet). The structure of none of the nitrilases has been determined by Xray crystallography; threfore this spatial structure was deduced for nitrilases based on their homology with other members of the superfamily [44]. Nitrilases are homooligo meric proteins [45]. The majority of nitrilases in solution exist as inactive dimers that can associate into active oligomers [46]. This oligomerization can be initiated by a substrate (substrate activation); by a high enzyme concentration; and by changes in physicochemical conditions, such as heating, high concentra tions of salts (KCl, MgSO4, NH4Cl, (NH4)2SO4), and addition of an organic solvent [47–50]. It was demonstrated by electron microscopy that some microbial nitrilases can form regular structures, specifically, homooligomeric helices of various lengths (tens of monomers) [46, 51, 52]. One of the most comprehensively studied nitrilases from Rhodococcus rhodochrous J1 exists as three homooligomeric structures, namely, a dimer, a complex 480 kDa in size, and regular helices of various lengths. In solution, this nitrilase occurs as an inactive dimer. In the presence of a substrate, the diner turns into an active complex with a mass of 480 kDa [50]. Active helical oligomers were observed for Rhodococ cus rhodochrous J1 recombinant nitrilase isolated from E. coli [52]. It was shown that the recombinant enzyme undergoes posttranslational proteolysis in which 39 amino acids are cleaved from the N terminus of the protein. This is likely necessary for the formation of the active helical oligomers [52]. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
BIOCATALYTIC HYDROLYSIS OF NITRILES
391
The physiological role of the formation of the helical oligomeric structures is unclear, but they may be of interest to biotechnology due to a high density of catalytic sites in them. The enzymatic activity of nitrilases is due to the glutamic acid–lysine–cysteine (Glu–Lys–Cys) cata lytic triad [43]. It is believed that the reaction begins with a nucleophilic attack by the thiol group of cys teine on the nitrile carbon atom, and this yields a thioimidate covalently bound to the enzyme. Next, the thioimidate forms a tetrahedral intermediate complex by adding a water molecule (Scheme 5). In this pro cess, the glutamate acts as a base and the lysine residue stabilizes the tetrahedral intermediate [53]. Lys N
Enz SH + RCN
NH Enz S C R
H
H H2O
O
R
H O
N H
C
I
Glu O
H
S O R C NH2 + Enz
H
O II
Cys SH
H2O
IV
Lys N
H H
O H
R C O + Enz SH + R C O−NH4 V
S
H H
NH3
Cys
O
+
N H
−
Glu
O
O H
III
Scheme 5. Hypothetical mechanism of the catalytic action of nitrilases [54]. The sulfhydryl group of cysteine nucleophil ically attacks the nitrile carbon atom to yield product I. Adding a water molecule, I forms tetrahedral intermediate com plex II. Glutamic acid acts here as a base, and lysine stabilizes complex II. Complex II can decompose to amides and free enzyme, particularly when it has electronwithdrawing or bulky substituents. Normally, the addition of the second water molecule and the stabilization of the positive charge on the nitrogen atom of the substrate result in the formation of com plex III, which is broken up to release a carboxylic acid and ammonia. Eventually, the reaction yields the ammonium salt of the carboxylic acid and free enzyme.
It is hypothesized that the intermediate complex can breaks up via two routes leading to different prod ucts [54]. The first route includes NH3 elimination from the complex, resulting in the formation of a thioester, which then reacts with a second water molecule to yield a carboxylic acid and the initial form of the enzyme. This is the normal reaction route. The second route leads to thiol elimination and amide for mation. Thus, in some cases, nitrilase can also act as nitrile hydratase. For some substrates, the main reac tion product may be an amide. For example, the purified recombinant nitrilase NIT4 from the Arabidopsis thaliana plant produces more than 60% amide during the hydrolysis of 2cyanoLalanine [55]. Note that, the nitrilasecatalyzed amide formation is an enantioselective process, whereas the nitrile hydrolysis by nitrile hydratses is usually not [56]. According to their substrate specificity, the nitrilases are divided into three groups: (1) aromatic nit rilases, e.g., Rhodococcus rhodochrous J1 and fungal enzymes, which primarily hydrolyze aromatic nitriles [57]; (2) aliphatic nitrilases, e.g., R. rhodohrous K22 and Acidovorax faecalis 72W nitrilases, which mainly hydrolyze aliphatic nitriles [58]; and (3) nitrilases mainly hydrolyzing arylacetonitriles, e.g., arylacetoni trilase from Alcaligenes faecalis JM3 [59]. This classification is still extensively used, even though some nit rilases show wide substrate specificity and can be simultaneously assigned to several classes. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
392
DEBABOV, YANENKO
2.3. Amidases Amidases (EC 3.6.1.4) are enzymes hydrolyzing amides of carboxylic acids to acids and ammonium. The amidases are classified into two groups: aliphatic amidases, which belong to the nitrilase superfamily, and the AS (amidase signature) family. This classification is based on amino acid sequences and spatial structure. The amidases belonging to the nitrilase superfamily contain the Glu–Cys–Lys catalytic triad [60] and exist in solution as homotetrameric and homohexameric structures [61]. The ASfamily amidases employ the Ser–Ser–Lys catalytic triad and contain a very conservative sequence of 130 amino acids (AS sequence) [62]. In solution, the ASfamily amidases exist as homodimers and homooctamers [63, 64]. Amidases are widespread and occur in bacteria, yeasts, fungi, plants, and animals [40]. The mechanism of the catalytic action of amidases is as follows: the enzyme nucleophilically attacks the carbonyl group of the amide, which yields a tetrahedral intermediate complex; this complex releases ammonium, turning into an acyl–enzyme intermediate complex. The latter undergoes hydrolysis to a carboxylic acid and free enzyme (Scheme 6) [54]. Like nitrilases, amidases form a covalent bond with the substrate; the nitrilasesuperfamily amidases bind through a Cys residue, and ASfamily ami dases, through a Ser residue. The existence of this covalent bond was confirmed by Xray crystallogra phy [63–65]. O R C NH2
EXH
H O R C NH2 XE
H2O
R C N
EXH
NH R C XE
NH3
O R C XE H2O
O C
NH2OH H2O
R
+ EXH OH
R C NH OH O Scheme 6. Mechanisms of the reactions catalyzed by amidases [66].
Amidases show acyltransferase activity as well. The transfer of the acyl residue to hydroxylamine (Scheme 6) occurs particularly efficiently [66]. Hydroxamic acids are readily detectable in the presence of iron chloride, forming a bright red complex. This reaction is often used to determine the amidase activ ity [67]. It is interesting that amidases can display nitrilase activity. For instance, highpurity recombinant Rhodococcus rhodochrorus J1 amidase expressed in E. coli can hydrolyze benzonitrile to benzoic acid and ammonium [68]. However, the benzonitrile hydrolysis proceeds 6000 times less rapidly than benzamide hydrolysis. Replacement of Ser195 with Ala195 causes the total loss of amidase and nitrilase activities. This is evidence that both reactions take place in the same active site of the enzyme. Amidases display rather high stereospecificity. Most often, the Sisomer of the amide undergoes hydrolysis [69–71]. Search for Rspecific amidases is of considerable interest in the context of the synthe sis of starting compounds for the pharmaceutical industry. Such amidases have already been found, one of which is thermally stable amidase from Klebsiella oxytoca bacteria [72, 73]. 3. CATALYTIC POTENTIAL OF ENZYMES FOR NITRILE CONVERSION Nitrilehydrolyzing enzymes have great potential for producing of intermediates that are used for the synthesis of drugs, pesticides, and odorants. In these applications, the specific activity of the biocatalyst, its cost and resistance to high substrate and product concentrations are highly significant; however, chemoselectivity, regioselectivity, and enantioselectivity are crucially important. 3.1. Chemoselectivity Enzymes provide means to hydrolyze nitrile groups in molecules containing functional groups, that are unstable under chemical hydrolysis conditions. For example, nitrilase from Pseudomonas catalyzes the hydrolysis of aliphatic 2acetoxynitriles, leaving intact the ester bond [74]. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
BIOCATALYTIC HYDROLYSIS OF NITRILES
393
An interesting and practically important case of chemoselective nitrilase activity is the hydrolysis of one of the two nitrile groups in dinitriles, which yields cyanocarboxylic acids. This type of selectivity was observed for aromatic nitrilases [75], arylacetonitrilases [76], and aliphatic nitrilases [40, 77–79]. Nit rilase from Pseudomonas fluorescens Pf5 has recently been described. This enzyme exhibits a very high spe cific activity just toward dinitriles [48]. Among the aliphatic dinitriles, the nitrilase selectivity for the hydrolysis of one cyano group decreases with increasing chain length, but it is still observable up to six carbon atoms [40]. Dinitrile monohydrolysis was used to synthesize lactams in two stages with 80–90% yield [40]. The first step produces the ammonium salt of a cyanocarboxylic acid via dinitrile hydrolysis with of Acidovorax facilic 72W whole cells. At the second stage, this intermediate compound is hydrogenated without isola tion (Scheme 7) [40]. R
CN [CH2]n CN
COO−NH4 [CH2]n
+
nitrilase Acidovorax facilis 72W
R
H2 Raney nickel
R
[CH2]n O
CN
NH
Scheme 7. Twostep synthesis of lactams from dinitriles.
Among the nitrilehydrolyzing enzymes, nitrilases show the highest monohydrolysis selectivity toward dinitriles. However, some selectivity is also observed for nitrile hydratases. For instance, the hydrolysis of 2,6 and 2,4dicyanopyridine catalyzed by nitrile hydratase from Rhodococcus erythropolis produces amides of the corresponding cyanocarboxylic acids with the yields of 83 and 97% for 10 min; after 2 h and 118 h, the main products are diamides of pyridinedicarboxylic acids and pyridinedicarboxylic acids them selves, respectively (Scheme 8) [76]. CN nitrile hydratase/amidase from
CONH2
Rhodococcus erythropolis 10 min
N
N
CN 118 h
CN
nitrilase from Fusarium solani
2h
CONH2
COOH
N
N
CONH2
CN
118 h
COOH N COOH Scheme 8. Biocatalytic conversions of 2,6dicyanopyridine.
3.2. Regioselectivity (E/Z Selectivity) Many nitrilases are capable of hydrolyzing α,βunsaturated nitriles. When the substrate is a mixture of E and Zisomers, the enzymes generally prefer the Eisomer; that is, the main hydrolysis product is Ecarboxylic acid [48, 80, 81]. Hann et al. [81] carried out a detailed study of the enzymatic hydrolysis of (E,Z)2methylbut2enen itrile. Nitrilase from Acidovorax facilis 72W hydrolyzes only the Eisomer even when the enzyme is in large excess and the hydrolysis time is long. The resulting (E)2methylbut2enoic acid as its ammonium salt is readily separable from (Z)2methylbut2enenitrile. (E)2Methylbut2enoic acid is also called tiglic acid, and its Zisomer is called angelic acid. Both compounds are important intermediates in the synthesis of medicines and fragrances. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
394
DEBABOV, YANENKO
It is interesting that nitrile hydratases and amidases are both regioselective. (E,Z)2Methylbut2 enenitrile was also hydrolyzed with Comamonas testosterone cells, which show nitrile hydratase and ami dase activities [81]. Both nitrile hydratase and amidase preferentially hydrolyze the Eisomer. Upon pro longed hydrolysis, the Enitrile turns quantitatively into the Ecarboxylic acid, while the Znitrile only undergoes partial hydrolysis to an amide (approximately 15% conversion) [81]. 3.3. Enantioselectivity Among the enzymes examined, the amidases are the most and the nitrile hydratases are the least enan tioselective. Nitrilases usually perform the enantioselective nitrile hydrolysis, but only for certain substrates and under certain reaction conditions does the enantiomer excess (ee) exceed 90%. (S)Ibuprofen can be obtained from racemic 2(4isobutylphenyl)propionitrile with ee > 90% using Acinobacter AK226 nit rilase [82]. (S)Ibuprofen is a popular nonsteroid antiinflammatory drug, with only the (S)isomer being biologically active. The enantioselectivity of nitrilases depends on the amino acids adjacent to their active site. For exam ple, the Trp164Ala replacement in nitrilase from Alcaligenes faecalis ATCC 8750 converts this highly Rspecific nitrilase into an enzyme hydrolyzing the (R,S)nitrile of mandelic acid mainly into an Samide [83]. Amino acid susbstitutions affecting the amidetoacid ratio during the nitrile hydrolysis have been found. The replacement of Cys163 with asparagine or glutamine in nitrilase from Pseudomonas fluorescens EBC191 markedly increases the proportion of amide, and the replacement of Cys163 with alanine or serine reduces this proportion [84]. In nature, both R and Sspecific nitrilases occur. Diversa Corp. carried out a screening of DNA of over 600 biotopes and discovered more than 137 unique nitrilases [85]. These include nitrilases hydrolyzing racemic 3hydroxyglutaronitrile to optically active (S)4cyano3hydroxyglutaric acid (enzyme 4A2) and (R)4cyano3hydroxyglutaric acid (enzymes 1A8 and 1A9) with ee > 90% and ee > 95%, respec tively. These compounds are important intermediates in the synthesis of Lipitor, a drug inhibiting choles terol synthesis, which belongs to the class of statins [85]. The Diversa’s researchers also found out nit rilases that are (S) and (R)selective toward mandelonitrile and phenylacetaldehyde cyanohydrin. Nitrile hydratases. About 20 enantioselective nitrile hydratases have been described in the literature [86]. As a rule, the enantioselectivity of these enzymes is low (ee 5–50%). An optically pure amide can be obtained only in rare cases and only from certain substrates. For example, the amide of γphenoxyβ hydroxybutyric acid with ee > 99% was obtained using nitrile hydratase from Rhodococcus rhodochrous ATCC BAA870 [90]. Mandelamide with ee > 95% and 98.7% conversion was obtained using nitrile hydratase from Rhodococcus sp. HTUO6 [91]. Nitrile hydratases in bacteria exist together with amidases, which are highly enantioselective. This makes it possible to use whole cells as catalysts for obtaining optical isomers of carboxylic acids and their amides from nitriles. 4. LARGESCALE APPLICATION OF THE ENZYMATIC HYDROLYSIS OF NITRILES FOR THE SYNTHESIS OF AMIDES AND ACIDS Owing to their unique properties—high chemoselectivity, regioselectivity, and enantioselectivity and high specific activity and stability—nitrile hydratases, nitrilases, and amidases are being used increasingly widely in organic syntheses. In this section, we will provide several examples illustrating the commercial use of these enzymes in the production of amides and acids. 4.1. BioAcrylamide The industrial biocatalytic synthesis of acrylamide is an impressive example of how biotechnology can change conventional chemical processes. Acrylamide is among the most important products of the chemical industry. Acrylamidebased poly mers are widely used in various applications, including oil recovery, improving soil structure, enhancing soil waterholding capacity, water purification (as a flocculant), paper production, and others. Acrylamide gels are widely known to be used in molecular biology and medicine for electrophoretic separation of pro teins, determination of DNA nucleotide sequences, and enzyme immobilization. The annual world out put of acrylamide exceeded 500000 t in 2008. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
BIOCATALYTIC HYDROLYSIS OF NITRILES
395
Comparison of industrial biocatalysts for acrylamide synthesis Biocatalyst* Characteristic B23
J1
M33
RS1
Developer
University of Kyoto, Japan
University of State Research Institute Institute of Bio Kyoto, Japan for Genetics and Selec chemical Engi tion of Industrial Mic neering, China roorganisms, Russia
Bacteria producing nitrile hydratase
Pseudomonas chlororaphis
Rhodococus rhodochrous
Rhodococus rhodochrous
Nocardia sp.
Inducible
Inducible
Constitutive
Inducible
85
76
250
No data
1400
2100
>5000
>10000 (laboratory scale)
4
4
18–20
4
Low
0.967
0.015
No data
27
45
50
45
Nitrile hydratase synthesis Specific activity, units/(mg cells) Activity of culture, units/ml (20°C) Acrylamide synthesis temperature, °C Amidase activity, units/mg Final acrylamide concentration, %
* The data for B23, J1, M33, and RS1 were taken from [93], [96], and [102], respectively.
Three industrial acrylamide synthesis technologies—sulfuric, catalytic, and biocatalytic—have been implemented to date [92, 93]. In all of them, acrylamide results from acrylonitrile hydration. Both the sulfuric technology (outdated) and the Cucatalyzed process (which is the most widespread in the world) have serious drawbacks. Acrylamide obtained by acrylonitrile hydration over copper–nickel catalysts contains, as main impurities, variablevalence metal ions (due to the catalyst dissolving), acrylic acid, and byproducts resulting from the reactions occurring at the double bond of acrylonitrile and acry lamide. In order to obtain highquality acrylamide solutions, it is necessary to use a multistage purification system for removing the residual acrylonitrile, metal ions, and byproducts. The common disadvantage of the sulfuric and catalytic technologies is that they yield diluted (6–20%) aqueous solutions of acrylamide, which need expensive vacuum concentration to convert to the globally accepted standardized commer cialgrade product with 40–50% concentration. Furthermore, the formation of significant amounts of wastewater due to regeneration of ionexchangers used for the acrylamide purification is characteristic of the conventional processes. In addition, they require large investments for the creation of new, or renova tion of existing acrylamide facilities which, in sum, makes them complex processes. The biocatalytic technology employs microbial cells with high nitrile hydratase activity as acrylonitrile hydrolysis catalysts. The prerequisite for the development of this technology appeared in the 1970s, with the isolation of the Brevibacterium sp. R312 strain, capable of converting nitriles into amides [94, 95]. Very soon, in the mid1980s, the first industrial bioacrylamide plant, constructed by the Japanese company Nitto Chemical Ltd. [93] was put into operation. Since then, marked advances have been made in the bio acrylamide technology. At present, over 25% of the global acrylamide output (>40% in China) is produced biocatalytically. All of the world industrial biotechnologies are based on Japanese, Russian, or Chinese biocatalysts (table). Industrial biocatalysts include cells of actinomycetales (rhodococci or nocardia) containing Co type nitrile hydratases. Unlike the B23, J1, and RS1 strains, the M33 strain produces nitrile hydratase con stitutively and needs no inducer [96]. The M33 strain is a mutant of the natural strain of M8 obtained by genetic manipulations [96]. The synthesis of nitrile hydratases in the M8 and M33 strains is subject to complex regulation, including carbon and nitrogen catabolite repression [97, 98]. A detailed investigation of the effect of cobalt on nitrile hydratase in the M8 and M33 strains revealed a radically new mechanism of metaldependent enzymes regulation. The cobalt ions that are the part of enzyme prosthetic group enhance the transcription of the nitrile hydratase gene [99]. This regulation mechanism allows to block the nitrile hydratase formation at its early stages, if there are no conditions for enzyme activation in the medium. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
396
DEBABOV, YANENKO
Genetic modification of the M33 strain, optimization of the medium composition and culturing con ditions allowed the highest possible activity level to be attained in industrial scale (table). The twostep sys tem of culturing on glucose and acetate afforded cultures with an activity of 5000 units/ml [100]. Another advantage of the M33 biocatalyst is its low amidase activity (which is almost 100 times lower than that of J1), which inhibits the further hydrolysis of acrylamide [96]. For the J1 and RS1 biocatalysts, the problem of residual amidase activity is solved by lowering the pro cess temperature to 4°C [101, 102]. However, this simultaneously reduces the nitrile hydratase activity of the biocatalyst (by a factor of 2.3 per 10 deg) and raises the amount of energy spent on maintaining the reaction temperature, taking into account the exothermic character of the reaction. The low amidase activity and high thermal stability of M33 allow the process to be conducted at 20°C. These performance characteristics demonstrate that the M33 biocatalyst designed at the State Research Institute for Genetics and Selection of Industrial Microorganisms is now the best among the industrial biocatalysts. The M33 strain and the amide production process based on this strain have been patented in Russia and many for eign countries [103, 104]. Acrylamide plants using the M33 biocatalyst operate in the Russian Federation (Perm) and in the Republic of Korea (Ulsan). The unique properties of the M33 biocatalyst have made it possible to greatly simplify acrylamide pro duction. The new industrial bioacrylamide production technology developed by the State Research Institute for Genetics and Selection of Industrial Microorganisms in collaboration with its Saratov Branch (now Biamid Closed Joint Stock Company) and with the Saratov branch of the Polymer Chemistry and Technology Research Institute is as follows. A temperaturecontrolled reactor fitted with a stirrer is charged with purified, deionized water and with an appropriate amount of the catalyst (polyacrylamide particles containing a few M33 cells). Next, acrylonitrile is introduced in the reactor so that its concen tration does not exceed 0.5%. The reactor is maintained at 18–20°C. In 5–6 h, the acrylamide concen tration reaches 50%. The biocatalyst is separated from the solution by filtration. The end product is high quality, commercialgrade acrylamide (>99% purity, 50% solution) suitable for polymerization without further purification [105]. Thus, the biocatalytic acrylamide synthesis technology is superior to the conventional chemical (sul furic and catalytic) methods, for it ensures a higher acrylonitrile conversion and hydration selectivity and milder reaction conditions, affords a purer product, consumes less energy, and is environmentally friend lier. The trend of replacing of the conventional acrylamide production technologies by the biocatalytic technology has become steady and irreversible. The biocatalytic process has already been implemented in four countries: Japan, Russia, China, and South Korea. Biocatalysts designed for acrylamide production have found application in the synthesis of other amides owing to the wide specificity of the involved enzymes. Lonza Guangzhou Fine Chemicals Company (Guangzhou, China) has employed the J1 biocatalyst in the synthesis of nicotinamide (vitamin В3) from 3cyanopyridine [106]. This biocatalyst ensures high product selectivity at 100% conversion, whereas the conventional 3cyanopyridine hydrolysis in an alkali medium yields only about 4% nicotinic acid as a byproduct. The global annual nicotinamide output has exceeded 3000 t. The M33 biocatalyst is also efficient in the nicotinamide production. Using a batch reac tor replenished with 3cyanopyridine, it is possible to obtain 450 g/l of nicotinamide containing less than 0.05% nicotinic acid for 5–6 h [96]. Another biocatalyst, Pseudomonas chlororaphis B23, which is used by Nitto Chemical Industry Co. in acrylamide production, proved very efficient in the production of 5cyanovaleroamide (5CVAM) from adiponitrile. 5CVAM is a starting chemical in the synthesis of an azafenidin herbicide. The productivity of the biocatalyst (B23 cells immobilized in alginate gel) exceeds 3000 kg of 5CVAM per kilogram of bio catalyst at 97% conversion and 96% selectivity [107]. With manganese dioxide as a catalyst, the produc tivity is no higher than 1 kg per kilogram of catalyst at 35% conversion. 4.2. Ammonium Acrylate Acrylic acid and its salts are the most demanded acrylic monomers. Their global annual output is over 3 million t [92]. Currently, the main industrial method for acrylic acid production is vaporphase propy lene oxidation [92]. At moderate output rates, acrylic acid production by acrylonitrile hydrolysis on microbial catalysts is efficient as well. The biocatalytic production of acrylic acid can be carried out at the consumer’s facilities. This approach was used by the Ashland MSP Co. (Perm, Russia), which is now manufacturing hundreds of tons of ammonium acrylate via biocatalysis for its own needs (flocculant pro duction). An Alcaligenes sp. C32 strain and its mutant [108] serve as catalysts in this process. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
BIOCATALYTIC HYDROLYSIS OF NITRILES
397
4.3. αHydroxycarboxylic Acids The simplest route to αhydroxy acids is the hydrolysis of cyanohydrins (αhydroxynitriles). The obstacle to use of cyanohydrins is that they are unstable in aqueous solution, releasing cyanides that inhibit nitrilehydrolyzing enzymes. In the case of nitrile hydratase, cyanides likely form stable complexes with the metal atom of the active site. Most of the known nitrile hydratases are very sensitive to cyanides (the inhibition constants for the cyanide ion vary between 0.01 and 20 mM) [109, 110]. In the presence of cyanohydrins, nitrile hydratases lose activity very rapidly. Due to the high sensitivity of nitrile hydratases to cyanides, the enzymatic hydrolysis of cyanohydrins has not yet become industrially important. Nitrile hydratasecontaining strains resistant to 50 mM cyanide have been found out in recent years [111]. These enzymes have been employed in the hydrolysis of acetone cyanohydrin to αhydroxyisobutyramide, a syn thetic precursor of methacrylamide. Under the optimal reaction conditions (4°С, stepwise addition of the substrate), it is possible to obtain 140 g of αhydroxyisobutyramide per liter of the reaction mixture within 5 h [112]. The use of cyanideresistant nitrile hydratases in combination with stereoselective amidases provides a means to obtain enantiomers of αhydroxy acids [113]. Nitrilases are more resistant to the products of the cyanohydrins spontaneous hydrolysis. An industrial process for obtaining glycolic acid from glycolonitrile using a nitrilase has been designed (see below). Nit rilases also allow pure isomers of αhydroxy acids to be obtained. There are two approaches to the synthe sis of these isomers. The first approach is based on the application of enantioselective nitrilases [114], which make it possible to exclusively obtain a pure target isomer, e.g., (R)mandelic acid from racemic nitriles (see below). The second approach is based on the action of enantioconservative nitrilases [115], which are capable of converting the optically pure cyanohydrin (R) or (S)isomer to the chiral (R) or (S)isomers of the corresponding acid without changing the optical activity. At present, this approach is widely used in organic synthesis owing to the fact that (R) or (S)oxynitrilasebased biocatalysts for the stereoselective synthesis of (R) or (S)isomers of cyanohydrin from an aldehyde or ketone and a cyanide have been developed [116]. Glycolic acid. This compound has the simplest structure among the αhydroxycarboxylic acids. Gly colic acid is used as a monomer in the production of biocompatible and biodegradable polymers (polyg lycolic acid, polymers obtained by copolymerization of lactic and glycolic acids, etc.). Due to its unique skin penetration capacity, it is used as a component of various skin protectors. At present, glycolic acid is mainly produced from formaldehyde and carbon monoxide in the presence of an acid at a high temperature and pressure [117]. This process is complicated by an undesirable by product formation, glycolic acid decomposition, and spontaneous polymerization. A hybrid, chemical–enzymatic process for the glycolic acid synthesis from formaldehyde and hydro gen cyanide has been suggested (Scheme 9) [118]. The interaction of these compounds in an aqueous medium yields glycolonitrile, which is hydrolyzed to ammonium glycolate by the action of nitrilase. The glycolate is converted into free glycolic acid by ionexchange chromatography. HCHO + HCN
>99%
HO H2C C N glycolonitrile
nitrilase >99%
HO O + H2C C O−NH4 glycolate
HO O H2C C OH glycolic acid
Scheme 9. Hybrid, chemical–enzymatic synthesis of glycolic acid.
The biocatalyst designed for this process (immobilized E. coli cells with the recombinant nitrilase) affords 3.2 M ammonium glycolate solutions in continuous reactors and, under these conditions, shows a productivity exceeding 1000 g of glycolic acid per gram of dry cell mass [119]. The biocatalytic synthesis of glycolic acid has been implemented by CrossChem Co. [120]. The final product of this process is either crystalline glycolic acid or its 70% aqueous solution. (R)Mandelic acid. Over many decades, the greatest attention has been focused on the hydrolysis of mandelonitrile to (R)mandelic acid under the action of Rspecific nitrilases, because this acid is widely used in enantiomer separation. Mandelonitrile undergoes racemization in aqueous solution at neutral pH due to the reversible elimination of hydrogen cyanide. Therefore, the yield of the (R)isomer of the acid in the nitrilasecatalyzed hydrolysis of the racemic nitrile can theoretically be 100% (Scheme 10). O C
OH H
CN
+ HCN
OH
OH CN
+
nitrilase
Scheme 10. Biocatalytic synthesis of (R)mandelic acid. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
COOH
398
DEBABOV, YANENKO
This process was described for the first time by Yamamoto et al. [121], who used nitrilase from Alcali genes feacalis ATCC 8750. Since then, different laboratories have published a number of works aimed at attaining of a high specific activity and stability of nitrilases and, more importantly, their resistance to high substrate concentrations. The best results were obtained by cloning the nitrilase gene of Alcaligenes ECU0401 in E. coli cells. The recombinant cells are stable in the presence of 200 mM mandelonitrile and can be used repeatedly, retain ing 40% of their initial activity after ten cycles. By continuously feeding the nitrile into reactor during the process, it is possible to obtain up to 80 g/l of (R)mandelic acid with ee > 99% and nearly quantitative yield [122]. Concentrated solutions of (R)mandelic acid (12–14%) were obtained by using new nitrilases and by optimizing the conversion conditions, including maintaining a low aldehyde concentration in the reaction mixture [123]. Mitsubishi Rayon and BASF announced the production of hundreds of tons of (R)mandelic acid per year. The optical purity of their product is higher than 97%. Unfortunately, only the (R)isomer of mandelic acid can be obtained by enantioselective hydrolysis. No industrially significant (S)selective nitrilase has been discovered to date. 4.4. Optically Active 2Methyl and 2,2Dimethylcyclopropanecarboxylic Acids and Their Amides An impressive example of the application of nitrile hydratase in combination with amidase is the syn thesis of optically active 2methyl and 2,2dimethylcyclopropanecarboxylic acids and their derivatives. These compounds are key intermediates in the synthesis of curacin A, which is a potent antimitotic agent, and cilastatin, an inhibitor of kidney dehydropeptidase. The latter is usually introduced in organism in combination with antibiotics, penem, and carbopenem, in order to prevent their degradation in kidneys. Wide screening of bacterial strains revealed that Rhodococcus erythropolis ATCC2544 bacteria can con vert 2,2dimethylcyclopropanecarbonitrile into (S)2,2dimethylcyclopropanecarboxylic acid. However, the optical purity of the product was only 81.8% [124]. In 2007, Zheng et al. [125] reported that Delftia tsuruhatensis ZJB05174 bacteria are capable of (R)enantioselectively hydrolyzing 2.2dimethylcyclo propanecarboxamide. Amidase from the same strain was cloned in E. coli and was expressed under the control of the phage T7 RNA polymerase promoter. The recombinant strain synthesizes large amounts of amidase (5255 units/l in a 100l fermenter). This catalyst provided the basis for the industrial synthesis of (S)dimethylcyclopropanecarboxamide with ee of 99.32 (Scheme 11) [126]. (R)stereoselective amidase
CONH2
+ COOH (R)acid
CONH2 (S)amide
Scheme 11. (S)Dimethylcyclopropanecarboxamide synthesis catalyzed by (R)specific amidase.
5. CONCLUSIONS AND OUTLOOK Over ten of the presentday industrial chemical processes are based on the action of nitrilehydrolyzing enzymes. These enzymes are used to obtain both largescale products (hundreds or thousands of tons) and stereoisomers for drug synthesis (grams or kilograms). Owing to the chemoselectivity, regioselectivity, and enantioselectivity of the nitrilehydrolyzing enzymes, it has become possible to obtain compounds that are hard to access by means of conventional organic synthesis. The basic limitations in the use of these enzymes arise from their narrow substrate specificity, instability (low resistance to temperature, pH, and organic solvents), and expensiveness. The majority of these limitations can be eliminated by using modern methods, such as rapid screening of unique natural enzymes, genetic engineering of strains overproducing the desired enzymes, improve ment of the enzyme properties by sitespecific mutagenesis or directed evolution methods, and cell and enzyme immobilization. Thus, there is every reason to believe that, in the nearest future, the enzymatic hydrolysis of nitriles will become a standard method for obtaining a wide diversity of amides and organic acids. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
BIOCATALYTIC HYDROLYSIS OF NITRILES
399
REFERENCES 1. Conn, E.E., in Cyanide in Biology, Vennesland, B., Conn, E.E., Knowles, C.J., Westly, Y., and Wissing, F., Eds, London: Academic, 1981, p. 183. 2. Asano, Y., Tani, Y., and Yamada, H., Agric. Biol. Chem., 1980, vol. 44, p. 2251. 3. Cowan, D.A., Cameron, R.A., and Tsokoa, T.L., Adv. Appl. Microbiol., 2003, vol. 52, p. 123. 4. Foertner, K.U., Doerks, T., Muller, J., Raes, J., and Bork, P., PLoS ONE, 2008, no. 12, p. 3. 5. MaierGreiner, U.H., ObermaierSkobranek, B.M.M., Estemaier, L.M., Kammerloher, W., Freund, C., Wulfing, C., Burkert, U.J., Matern, D.H., Brener, M., Eulitz, M., Kufrevioglu, O.J., and Hartman, G.R., Proc. Natl. Acad. Sci. USA, 1991, vol. 88, p. 4260. 6. Okamoto, S. and Eltis, L.D., Mol. Microbiol., 2007, vol. 65, no. 3, p. 828. 7. Nakajima, Y., Doi, T., Saton, Y., Fujiwara, A., and Watanabe, I., Chem. Lett., 1987, vol. 9, p. 1767. 8. Bonnet, D., Artaud, I., Moali, C., Petre, D., and Mansuy, D., FEBS Lett., 1997, vol. 409, p. 216. 9. Endo, I., Yohda, M., and Okada, M., Trends Biotechnol., 1999, vol. 17, p. 244. 10. Tsujimura, M., Dohmae, N., Odaka, M., Chijimatsu, M., Takio, K., Yohda, M., Hoshino, M., Nagashima, S., and Endo, I., J. Biol. Chem., 1997, vol. 272, p. 29454. 11. Nagashima, S., Nakasako, M., Dohmae, N., Tsujimura, M., Takio, K., Odaka, M., Yohda, M., Kamiya, N., and Endo, I., Nat. Struct. Biol., 1998, vol. 5, p. 347. 12. Sari, M.A., Jaonen, M., Saroja, N.R., and Artaud, I., J. Inorg. Biochem., 2007, vol. 101, p. 614. 13. Huang, W.J., Jia, J., Cummings, J.G., Nelson, M.J., Schneider, G., and Lindqvist, Y., Structure, 1997, vol. 5, p. 691. 14. Miyanaga, A., Fushinobu, S., Ito, K., and Wakagi, T., Biochem. Biophys. Res. Commun., 2001, vol. 288, p. 1169. 15. Hourai, S., Miki, M., Takashima, Y., Mitsuda, S., and Yanagi, K., Biochem. Biophys. Res. Commun., 2003, vol. 312, p. 340. 16. Nojiri, M., Nakayama, H., Odaka, M., Yohada, M., Takio, K., and Endo, I., FEBS Lett., 2000, vol. 465, p. 173. 17. Endo, I., Nojiri, M., Tsujimura, M., Nakasako, M., Nagashima, S., Yohada, M., and Okada, M., J. Inorg. Bio chem., 2001, vol. 83, p. 247. 18. Nojiri, M., Yohada, M., Okada, M., Matsushita, Y., Tsujimura, M., Yoshida, T., Dohmae, N., Takio, K., and Endo, I., J. Biochem., 1999, vol. 125, p. 696. 19. Nishiama, M., Horinonchi, S., Kobayashi, M., Nagasawa, T., Yamada, H., and Berru, T., J. Bacteriol., 1991, vol. 173, p. 2465. 20. Wu, S., Fallon, R.D., and Payne, M.S., Appl. Microbiol. Biotechnol., 1997, vol. 48, p. 704. 21. Hashimoto, Y., Nishiyama, M., Horinonchi, S., and Berru, T., Biosci. Biotechnol. Biochem., 1994, vol. 58, p. 1859. 22. Komeda, H., Kabayashi, M., and Shimizu, S., J. Biol. Chem., 1996, vol. 271, p. 15796. 23. Komeda, H., Kabayashi, M., and Shimizu, S., Proc. Nat. Acad. Sci. U.S.A., 1996, vol. 93, p. 4267. 24. Yanenko, A.S., Doctoral (Biol.) Dissertation, Moscow: GosNIIgenetika, 2001. 25. Zhou, Z., Hashimoto, Y., Shira, K., and Kobayashi, M., Proc. Nat. Acad. Sci. U.S.A., 2008, vol. 105, p. 14849. 26. Nohuchi, T., Nojiri, M., Takei, K., Odaka, M., and Kamiya, N., Biochemistry, 2003, vol. 42, p. 11642. 27. Zhou, Z., Hashimoto, Y., and Kobayashi, M., J. Biol. Chem., 2009, vol. 284, p. 14930. 28. Zhou, Z., Hashimoto, Y., Cui, T., Washizawa, Y., Mino, H., and Kobayashi, M., Biochemistry, 2010, vol. 49, p. 9638. 29. Stevens, J.M., Saroja, R., Jaonen, M., Belghazi, M., Schmiter, D., Mansuy, I., Artand, I., and Sari, M.A., Pro tein Expr. Purif., 2003, vol. 29, p. 70. 30. Doan, P.E., Gurbiel, R.J., Cummins, J.C., Nelson, M.J., and Hoffman, B.M., J. Inorg. Biochem., 1999, vol. 74, p. 116. 31. Miyanaga, A., Fushinoba, S., Ito, K., Shoun, H., and Wakagi, T., Eur. J. Biochem., 2004, vol. 271, p. 429. 32. Sugiura, Y., Kuwahara, J., Nagasawa, T., and Yamada, H., J. Am. Chem. Soc., 1987, vol. 109, p. 5848. 33. Murakami, T., Nojiri, M., Nakayama, H., Odaka, M., Yohda, M., Dohmae, N., Takio, K., Nagamune, T., and Endo, I., Protein Sci., 2000, vol. 9, p. 1024. 34. Tsujimura, M., Odaka, M., Nakayama, H., Dohmae, N., Koshino, H., Asami, T., Hosgino, M., Takio, K., Yoshida, S., Maeda, M., and Endo, I., J. Am. Chem. Soc., 2003, vol. 125, p. 11532. 35. Hashimoto, K., Suzuki, H., Taniguchi, K., Noguchi, T., Yohda, M., and Odaka, M., J. Biol. Chem., 2008, vol. 283, p. 36617. 36. Taniguchi, K., Murata, K., Murakami, Y., Takahashi, S., Nakayama, H., Hashimoto, K., Koshino, H., Dohame, N., Yohda, M., Hirose, T., Maeda, M., and Odaka, M., J. Biosci. Bioeng., 2008, vol. 106, p. 174. 37. Yono, T., WasadaTsatsui, Y., Ari, H., Yamaguchi, S., Funahashi, Y., Osawa, T., and Masada, H., Inorg. Chem., 2007, vol. 46, p. 10345. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
400
DEBABOV, YANENKO
38. Hopmann, K.H., Guo, I.D., and Himo, F., Inorg. Chem., 2007, vol. 46, p. 4850. 39. Thimann, K.V. and Mahadevan, S., Arch. Biochem. Biophys., 1964, vol. 107, p. 62. 40. Chen, J., Zheng RenChao, Zheng YuGuo, and Shen YinChu, Adv. Biochem. Eng. Biotechnol., 2009, vol. 113, p. 37. 41. Martinkova, L. and Kzen, V., Curr. Opin. Chem., 2010, vol. 14, p. 130. 42. De Santis, G., Zhu, Z., Greenberg, W.A., Wong, K., Chaplin, J., Hanson, S.R., Farwell, B., Nicholson, L.W., Rand, C.L., and Weiner, D.P., J. Am. Chem. Soc., 2002, vol. 124, p. 9024. 43. Brenner, C., Curr. Opin. Struct. Biol., 2002, vol. 12, p. 775. 44. Kim, S., Tiwari, M.K., Moon, H.J., Yeya, M., Ram, H.T., Oh, D.K., Kim, J.W., and Lee, J.K., Appl. Microbiol. Biotechnol., 2009, vol. 83, p. 272. 45. Kobayashi, M. and Shimizu, Y., FEBS Microbiol. Lett., 1994, vol. 120, p. 217. 46. Thuku, R.N., Brady, D., Benedik, M.J., and Sewell, B.T., J. Appl. Microbiol., 2009, vol. 106, p. 703. 47. Harper, D.B., J. Biochem., 1977, vol. 165, p. 309. 48. Stevenson, D.E., Feng, R., Dumas, F., Groleau, D., Mihoc, A., and Storer, A.C., Biotechnol. Appl. Biochem., 1992, vol. 15, p. 283. 49. Hoyle, A.J., Bunch, A.W., and Knowles, C.J., Enzyme Microbiol Technol., 1998, vol. 23, p. 475. 50. Nagasawa, T., Weiser, M., Nakamura, T., Iwahara, H., Yoshida, T., and Gekko, K., Eur. J. Biochem., 2000, vol. 267, p. 138. 51. Sewell, B.T., Berman, M.N., Meyers, P.R., Jandhyala, D., and Benedik, M.J., Structure, 2003, vol. 11, p. 1. 52. Thuku, R.N., Weber, B.W., Varsani, A., and Sewell, B.T., FEBS J., 2007, vol. 274, p. 2099. 53. Yeom, S.J., Kim, H.J., Lee, J.K., Kim, D.E., and Oh, D.K., J. Biochem., 2008, vol. 415, p. 401. 54. Kobayashi, M., Goda, M., and Shimizu, S., Biochem. Biophys. Res. Commun., 1998, vol. 253, p. 662. 55. Piotrowski, M., Schonefelder, S., and Weiler, E.W., J. Biol. Chem., 2001, vol. 276, p. 2616. 56. Fernandes, B.C.M., Mateo, C., Kiziak, C., Chmura, A., Wacker, J., van Rantwijk, StolzA., and Sheldon, R.A., Adv. Synth. Catal., 2006, vol. 348, p. 2597. 57. Martinkova, L., Vejvoda, V., Kaplan, O., Kubac, D., Malandra, A., Cantarella, M., Bezonska, K., and Kren, V., Biotechnol. Adv., 2009, vol. 27, p. 661. 58. Kobayashi, M., Yanaka, N., Nagasawa, T., and Yamada, H., J. Bacteriol., 1990, vol. 172, p. 4807. 59. O’Relly, C. and Turner, P.D., J. Appl. Microbiol., 2003, vol. 95, p. 1161. 60. Pace, H. and Brenner, C., Genome Biol., 2001, vol. 2, p. 1. 61. Pertsovich, S.I., Guranda, D.T., Podchernyaev, D.A., Kotlova, E.K., Yanenko, A.S., and Shvyadas, V.K., Biokhimiya, 2005, vol. 70, no. 11, p. 1556 [Biochemistry (Moscow) (Engl. Transl.), vol. 70, no. 11, p. 1280]. 62. Chebron, H., Bigey, F., Arnand, A., and Galzy, P., Biochem. Biophys. Acta, 1996, vol. 1298, p. 285. 63. Ohtaki, A., Murata, K., Sato, Y., Noguchi, K., Miyatake, H., Dohmae, N., Yamada, K., Yohda, M., and Odaka, M., Biochim. Biophys. Acta, 2010, vol. 1804, p. 184. 64. Shin, S., Lee, T.H., Na, N.C., Koo, H.M., Kim, Y.S., Lee, H.S., Kim, Y.S., and Oh, B.H., EMBO J., 2002, vol. 21, p. 2509. 65. Shin, S., Yan, Y.S., Koo, H.M., Kim, Y.S., Choi, K.Y., and Oh, B.H., J. Biol. Chem., 2003, vol. 278, p. 24937. 66. Fournand, D. and Arnand, A., J. Appl. Microbiol., 2001, vol. 91, p. 381. 67. Backles, R.E. and Thelen, C.J., Anal. Chem., 1950, vol. 22, p. 676. 68. Kobayashi, M., Goda, M., and Shimizu, S., FEBS Lett., 1998, vol. 439, p. 325. 69. Snell, D. and Colby, I., Enzyme, 1999, vol. 24, p. 160. 70. Wang, M.X. and Feng, G.Q., Tetrahedron Lett., 2000, vol. 41, p. 650. 71. Doran, J.P., Duggan, P., Masterson, M., Turner, P.D., and O’Relly, C., Protein Expr. Purif., 2005, vol. 40, p. 190. 72. Shaw, N.M. and Nanghton, A.B., Tetrahedron Lett., 2004, vol. 60, p. 747. 73. WO Patent 9801568. 74. Heinemann, U., Kiziak, C., Zibek, S., Layh, N., Schmidt, M., Griengel, H., and Stolz, A., Appl. Microbiol. Biotechnol., 2003, vol. 63, p. 274. 75. Kobayashi, M., Yanaka, N., Nagasawa, T., and Yamada, H., Appl. Microbiol. Biotechnol., 1988, vol. 29, p. 231. 76. Nakai, T., Hasegawa, T., Yamashita, E., Yamamoto, M., Kumasaka, T., Ueki, T., Nanbu, H., Ikenaka, Y., Taka hashi, S., and Sato, M., Structure, 2000, vol. 8, p. 729. 77. BengisGarber, C. and Gutman, A., Appl. Microbiol. Biotechnol., 1989, vol. 32, p. 11. 78. Gavagan, J., Fager, S., Fallon, R., Folsom, P., Herkes, F., Eisenberg, A., Hann, E., and Di Casimo, R., J. Org. Chem., 1998, vol. 63, p. 4792. 79. Effenberger, F. and Osswald, S., Synthesis, 2001, vol. 12, p. 1866. 80. Effenberger, F. and Osswald, S., Tetrahedron: Asymmetry, 2001, vol. 12, p. 2581. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
BIOCATALYTIC HYDROLYSIS OF NITRILES
401
81. Hann, E., Sigmund, A., Fager, S., Cooling, F., Gavagan, J., Bramucci, M., Chanhan, S., Payen, M., and Di Cosimo, R., Tetrahedron: Asymmetry, 2004, vol. 60, p. 577. 82. Yamamoto, K., Ueno, Y., Otsubo, K., Kawakami, K., and Komatsu, K.I., Appl. Environ. Microbiol., 1990, vol. 56, p. 3125. 83. Kiziak, C., Klein, J., and Stolz, A., Protein Eng., 2007, vol. 20, p. 385. 84. Kiziak, C. and Stolz, A., Appl. Environ. Microbiol., 2009, vol. 75, p. 5592. 85. Robertson, D.E., Chaplin, J.A., De Santis, G., Podar, M., Madden, M., Chi, E., Richardson, T., Milan, A., Miller, M., Weiner, D.P., Wong, K., McQuaid, J., Farwell, B., Preston, L.A., Tan, X., Snead, M.A., Keller, M., Mathur, E., Kretz, P.L.., Bark, M.J., and Short, J.M., Appl. Environ. Microbiol., 2004, vol. 70, p. 2429. 86. Sosedov, O., Baum, S., Burger, S., Matzer, K., Kiziak, C., and Stolz, A., Appl. Environ. Microbiol., 2010, vol. 76, p. 3668. 87. Prasad, S. and Bhalla, T.C., Biotechnol. Adv., 2010, vol. 28, p. 725. 88. Fallon, R.D., Steiglitz, B., and Tumer, I., Appl. Microbiol. Biotechnol., 1997, vol. 47, p. 156. 89. Baner, R., Knackmuss, HJ., and Stolz, A., Appl. Microbiol. Biotechnol., 1998, vol. 49, p. 89. 90. Prepechalova, I., Martinkova, L., Stolz, A., Ovesna, M., Bezonska, K., and Kopecky, J., Appl. Microbiol. Bio technol., 2001, vol. 55, p. 150. 91. Knife, H.H., Chiba, V., Frederick, J., Bode, M.L., Mathiba, K., and Steenkamp, P.A., J. Mol. Catal., 2009, vol. 59, p. 231. 92. Plate, N.A. and Slivinskii, E.V., Osnovy khimii i tekhnologii monomerov (Principles of the Chemistry and Tech nology of Monomers), Moscow: Nauka, 2002. 93. Kobayashi, M., Nagasawa, T., and Yamada, H., Trends Biotechnol., 1992, vol. 10, p. 402. 94. US Patent 400081. 95. Bui, K., Arnaud, A., and Galzy, P., Enzyme Microbiol. Technol., 1982, vol. 4, p. 195. 96. Janenko, A.S., Astaurova, O.B., Voronin, S.P., Gerasimova, T.V., Kirsanov, N.B., Paukov, V.N., Poljakova, I.N., and Debabov, V.G., RF Patent 2053300, 1996. 97. Leonova (Pogorelova), T.E., Astaurova, O.B., Ryabchenko, L.E., and Yanenko, A.S., Appl. Biochem. Biotech nol., 2000, vol. 88, p. 231. 98. Astaurova, O.B., Leonova, T.E., Polyakova, I.N., Sineokaya, I.V., Gordeev, V.K., and Yanenko, A.S., Prikl. Biokhim. Mikrobiol., 2000, vol. 36, no. 1, p. 21 [Appl. Biochem. Microbiol. (Engl. Transl.), vol. 36, no. 1, p. 15]. 99. Pogorelova, T., Ryabchenko, L., Sunzov, N., and Yanenko, A., FEMS Microbiol. Lett., 1996, vol. 144, p. 191. 100. FRG Patent 10315376. 101. Yamada, H.. and Kobayashi, M., Biosci. Biotech. Biochem., 1996, vol. 60, p. 1391. 102. Liu, M., Li, C., Huang, Y., and Gao, Y., Chin. J. Process. Eng., 2004, vol. 4, p. 250. 103. Yanenko, A.S., Astaurova, O.B., Voronin, S.P., Gerasimova, T.V., Kirsanov, N.B., Paukov, V.N., Polyakova, I.N., and Debabov, V.G., US Patent 5827699, 1998. 104. FRG Patent 4480132. 105. Debabov, V.G., Voronin, S.P., Kozulin, S.V., Sinolitskij, M.K., Kozulina, T.N., Poljanskij, A.B., Sintin, A.A., Janenko, A.S., Bajburdov, T.A., Khorkin, A.A., Lujksaar, I.V., Reshetnikova, L.V., and Fedchenko, N.N., RF Patent 2077588, 1997. 106. Chassin, C., Spec. Chem., 1996, vol. 16, p. 102. 107. Hann, E.C., Eisenberg, A., Fager, S.K., Perkins, N.E., Gallagher, F.G., Cooper, S.M., Gavagan, J.E., Stieglitz, B., Hennessey, S.M., and DiCosimo, R., Bioorg. Med. Chem., 1999, vol. 7, p. 2239. 108. Kozulin, S.V., Voronin, S.P., Glinskij, S.A., Kozulina, T.N., Leonova, T.E., Novikov, A.D., Poltavskaja, S.V., Rjabchenko, L.E., Singirtsev, I.N., and Janenko, A.S., RF Patent 2337954, 2008. 109. Nagasawa, T., Nanba, H., Ryuno, K., Takeuchi, K., and Yamada, H., Eur. J. Biochem., 1987, vol. 162, p. 691. 110. Nagasawa, T., Takeuchi, K., and Yamada, H., Eur. J. Biochem., 1991, vol. 196, p. 581. 111. Gerasimova, T., Novikov, A., Osswald, S., and Yanenko, A., Eng. Life Sci., 2004, vol. 4, p. 543. 112. EEC Patent 1730177. 113. Osprian, I., Fechter, M.H., and Griengl, H., J. Mol. Catal., B, 2003, vol. 2425, p. 89. 114. Yamamoto, K., Fujimatsu, I., and Komatsu, K.I., J. Ferment. Bioeng., 1992, vol. 73, p. 425. 115. Kiziak, C., Conradt, D., Stolz, A., Mattes, R., and Klein, J., Arch. Microbiol., 2006, vol. 151, p. 3639. 116. Fechter, M.H. and Griengl, H., in Enzyme Catalysis in Organic Synthesis, Drauz K., Waldmann H., Eds., Wein heim: Wiley, 2002, p. 974. 117. DiCosimo, R., Payne, M.S., Panova, A., Thompson, J., and O’Keefe, D.P., US Patent 7198927, 2007. 118. Panova, A., Mersinger, L.J., Liu, Q., Foo, T., Roe, D.C., Spillan, W.L., Sigmund, A.E., BenBassat, A., Wag ner, A.W., O’Keefe, D.P., Wu, S., Petrillo, K.L., Payne, M.S., Breske, S.T., Gallagher, F.G., and DiCosimo, R., Adv. Synth. Catal., 2007, vol. 349, p. 1462. REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
402
DEBABOV, YANENKO
119. Wu, S., Fogiel, A.J., Petrillo, K.L., Jackson, R.E., Parker, K.N., Dicosimo, R., BenBassat, A., O’Keefe, D.P., and Payne, M.S., Biotechnol. Bioeng., 2008, vol. 99, p. 717. 120. CrossChem Company Website. http://www.crosschem.net/glycolic_acid.php 121. Yamamoto, K., Oishi, K., Fujimatsu, I., and Komatsu, K., Appl. Environ. Microbiol., 1991, vol. 57, p. 3028. 122. Zhang, Z.J., Xu, J.H., He, Y.C., Ouyang, L.M., and Liu, Y.Y., Bioprocess. Biosyst. Eng., 2010 (in press). doi10.1007/s004490100473z 123. EEC Patent 0773297. 124. Yeom, S.J., Kim, H.J., and Oh, D.K., Enzyme Microbiol. Technol., 2007, vol. 41, p. 842. 125. Zheng, R.C., Wang, Y.S., Lin, Z.Q., Zheng, Y.G., and Shen, Y.C., Res. Microbiol., 2007, vol. 158, p. 258. 126. Yang, Z.Y., Ni, Y., Lu, Z.Y., Liao, X.R., Zheng, Y.G., and Sun, ZH., Proc. Biochem. Soc., 2010 (in press). doi:1016/j.procbio.2010.08.005
REVIEW JOURNAL OF CHEMISTRY
Vol. 1
No. 4
2011
