Full text loading...
Chapter 36 : Enzyme Promiscuity and Evolution of New Protein Functions
Ebook: Choose a downloadable PDF or ePub file. Chapter is a downloadable PDF file. File must be downloaded within 48 hours of purchase
This chapter introduces the various categories of promiscuity and their possible role in enzyme evolution. It also reviews some of the well-known examples of enzymes for which the promiscuous function is industrially relevant, leading to conversions of nonnatural compounds. Enzyme promiscuity is becoming an increasingly important phenomenon in the field of biocatalysis, both directly and indirectly: directly, because many desirable conversions are “unnatural,” so expansion of the existing functions of protein catalysts is necessary to convert these “promiscuous” substrates; indirectly, through the observation that promiscuous enzymes may be particularly evolvable, opening a wide range of possibilities for engineering these potentially valuable side activities. Candida antarctica lipase B (CALB) is a typical representative of the α/β-hydrolase fold family of enzymes comprising not only a number of other esterases, but also haloalkane dehalogenases and epoxide hydrolases. Many of these natural products have unique bioactivity and are thus of great interest to the pharmaceutical industry. Halohydrin lyases or haloalcohol dehalogenases are found in various bacteria that are able to use halogenated hydrocarbons as a sole source of carbon and energy. The improvement of several of the promiscuous reactions of CALB was obtained by removing the catalytic nucleophile, naturally resulting in decreased performance for the native activity.
Different types of enzyme promiscuity. Promiscuous molecular recognition can take place at different stages of a reaction cycle. (A) In substrate promiscuity, the enzyme is able to recognize substrates with different structural features, but the chemical transformation performed is identical despite different substituents. (B) In product promiscuity, one substrate can be converted to many alternative products. The molecular recognition of the intermediate is conserved, but the folding of the intermediate or rearrangements of the reactive centers on the intermediate, e.g., by steric complementarity of the active-site template, lead to a variety of products onward from this reactive intermediate. (C) To achieve catalytic promiscuity, the tightest interactions—those in the transition state—have to be tolerant enough to recognize different reaction centers and bond-making and -breaking processes around them.
Mechanism of the native esterolytic activity of CALB. Upon substrate binding, the carbonyl functionality becomes positioned and polarized by the oxyanion hole (step a). The activated Asp-His pair activates the serine nucleophile, which performs a nucleophilic attack on the carbonyl carbon (b), ultimately resulting in the release of an alcohol and binding of a water molecule (c). The Asp-His pair now activates a water molecule for nucleophilic attack on the carbonyl carbon by general base catalysis, and the oxyanion hole performs the same role as it did for the first step, resulting in hydrolysis of the acyl-enzyme complex and release of the carboxylic acid (d).
Catalytic promiscuity of CALB. The oxyanion hole is a key catalytic feature, polarizing the carbonyl bond of the electrophile in each reaction and offsetting the development of negative charge at oxygen. The reactions catalyzed by CALB are: (A) native carboxylic ester hydrolysis ( 112 ); (B) aldol condensation ( 10 ); (C) Michael-type addition ( 13 , 105 ); (D) epoxidation ( 104 ); and (E) the predicted Baeyer-Villiger reaction ( 12 ). For the promiscuous functions, the catalytic nucleophile does not take part in catalysis. The other parts of the active-site residues play a similar role, and the serine has been replaced by an external nucleophile that is activated by the Asp-His pair, which acts as a general base.
Substrate promiscuity allows the combinatorial biosynthesis of avermectin analogs. The R2 group in the avermectin structure originates from either 2-methylbutyryl-coenzyme A or isobutyryl-coenzyme A, which are generated from l-leucine and l-valine, respectively. In S. avermitilis ATCC 53569, a branched-chain α-keto acid dehydrogenase that is essential for the in vivo production of the branched-chain fatty acids is absent ( 27 , 39 ), rendering this strain unable to produce avermectins. Feeding up 40 different fatty acid compounds (with a variety of substitutions in R2) to this mutant strain restores the biosynthetic machinery, resulting, in each case, in a different avermectin analog ( 28 ). For instance, in doramectin biosynthesis, cyclohexane carboxylic acid is incorporated ( 102 ), replacing either methyl-butyric acid or isobutyric acid in the wild-type strain ( 39 ).
Biodegradation routes for halogenated hydrocarbons involving haloalcohol dehaloge-nases. (A) Degradation of 1,3-dichloro-2-propanol to glycerol in A. radiobacter AD1, catalyzed by a haloalcohol dehalogenase (HheC) and an epoxide hydrolase (EchA) ( 113 ). (B) Degradation of 1,2-dibromoethane to ethylene oxide in Mycobacterium sp. GP1, catalyzed by a haloalkane deha-logenase (DhaAf) and a haloalcohol dehalogenase (HheB) ( 94 ).
Mechanism of haloalcohol dehalogenases and their likely precursors, the short-chain dehydrogenases/reductases (SDRs). (A) Proposed mechanism of A. radiobacter AD1 haloalcohol dehalogenase (HheC)-catalyzed ring closure of a vicinal haloalcohol. The Tyr145/Arg149 pair acts as a general base to abstract a proton from the hydroxyl group, resulting in an intramolecular nucleo-philic displacement of the halide, aided by leaving group stabilization by the halide binding site ( 25 ). (B) Proposed mechanism of SDR enzymes, in which the proton abstraction from the hydroxyl group is followed by hydrogen abstraction by the NAD+, resulting in the corresponding ketone.
Nucleophile promiscuity for the halohydrin dehalogenase-catalyzed epoxide ring opening. The position of the equilibrium is dependent on the nucleophile.
Schematic representation of (A) (R)- and (B) (S)-pNSO in the active site of haloalcohol dehalogenase from A. radiobacter AD1 (HheC) ( 26 ). The proposed difference in binding modes is based on X-ray structures of HheC in complex with (R)- and (S)-pNSO, respectively. Both structures were refined with p-nitrophenylcyclopropane as the substrate to obtain an unbiased view of the interactions of the Ser132/Tyr145 pair with the epoxide ring. These data show that the refinement data for both structures can be completely superimposed, apart from the epoxide ring, in which the (R)-enantiomer comes close enough to the Ser/Tyr pair to form a productive hydrogen bond. The calculated difference in Gibbs binding energy for both structures is also in agreement with the loss of one hydrogen bond.
One-pot synthesis of (S)-4-cyano-3-hydroxybutanoate methyl ester from racemic 4-chloro-3-hydroxybutanoate methylester using a mutant haloalcohol dehalogenase (HheC W249F [ 107 ]). This method makes use of both the native activity (step a) and one of the promiscuous activities (step b) of HheC ( 72 ).
Catalytic promiscuity of BVMOs. (A) Proposed reaction cycle for each of the different conversions ( 57 , 109 ). The reaction cycle starts with reduction of the oxidized flavin (I) by NADPH, resulting in reduced flavin (II) and NADP+, which is released from the enzyme only at the end of the catalytic cycle. The reduced flavin subsequently reacts with molecular oxygen to form the peroxyflavin intermediate (IIIa), which has been suggested to exist also in the protonated state (IIIb). This highly reactive intermediate is stabilized by various interactions with residues in the active site and can react with a variety of different substrates, both electron-rich and electron-deficient substrates (B) ( 119 ). The resulting hydroxyflavin (IV) subsequently decomposes into oxidized flavin (I) by eliminating water. Finally, NADP+ is released from the enzyme, allowing the enzyme to start another catalytic cycle.
Mechanisms of the various hydrolases that belong to the alkaline phosphatase su-perfamily. (A) Escherichia coli AP ( 99 ). (B) Xanthomonas axonopodis NPP ( 124 ). (C) P. aeruginosa arylsulfatase ( 7 ). (D) Mechanism proposed for the PMH from Rhizobium leguminosarum (RlPMH), based on X-ray structural data and the kinetic data of various active-site mutants ( 56 ). The active-site residues of PMH from Burkholderia caryophylli (BcPMH) superimpose ideally on RlPMH (86% amino acid identity), suggesting that BcPMH uses the same mechanism ( 115a ).
Catalytic promiscuity in the alkaline phosphatase superfamily a