Rational engineering of esterases for improved amidase...

Rational engineering of esterases for

improved amidase specificity in amide

synthesis and hydrolysis

Peter Hendil-Forssell

KTH Royal Institute of Technology

School of Biotechnology

Stockholm 2016

© Peter Hendil-Forssell

Stockholm 2016

KTH Royal Institute of Technology

School of Biotechnology

Division of Industrial Biotechnology

AlbaNova University Center

SE-106 91 Stockholm

Sweden

Printed by Universitetsservice US-AB

Drottning Kristinas väg 53B

SE-114 28 Stockholm

Sweden

ISBN 978-91-7729-210-4

TRITA-BIO Report 2016:21

ISSN 1654-2312

Front Cover: Candida antartica lipase B variant I189A in the modelled transition state for

nitrogen inversion stabilized by a water bridge connecting the scissile NH-group of the

substrate to the backbone of the enzyme.

Till Ann-Sofie och Stella

Abstract

Enzymes are nature’s own powerful catalysts that provide great reaction rate enhancements and make life on earth possible. Biocatalysis is an ever evolving field that uses enzymes or microorganisms for chemical synthesis. By utilizing enzymes that generally have evolved for specific reactions under mild conditions and temperatures, biocatalysis can be a more environmentally friendly option compared to traditional chemistry.

Amide-type chemistries are important and bond formation avoiding poor atom economy is one of the most prioritized research areas in organic chemistry. Biocatalysis could potentially be a solution but restricted substrate scope is a limitation. Esterases/lipases usually display a broad substrate scope and catalytic promiscuity but are generally poor at hydrolyzing amides compared to amidases/proteases. The difference between the two enzyme classes is hypothesized to reside in one key hydrogen bond present in amidases/proteases, which facilitates the transition state for nitrogen inversion during catalysis.

In this thesis the work has been focused on introducing a stabilizing hydrogen bond acceptor in esterases, mimicking that found in amidases/proteases, to develop better enzymatic catalysts for amide-based chemistries.

By two strategies, side-chain or water interaction, variants were created in three different esterases that displayed up to 210-times increased relative amidase specificity compared to the wild type enzyme. The reduction in activation enthalpy for amide hydrolysis was 3.3 kcal/mol, corresponding to a weak hydrogen bond. However, the engineered variant was also disfavored entropically by freezing molecules of water resulting in a free energy of activation of 2.1 kcal/mol. The results show an estimated lower limit on how much the hydrogen bond can be worth to catalysis.

Synthesis of amides in water is problematic due to unfavorable kinetic reaction control. However, with Mycobacterium smegmatis acyltransferase (MsAcT) kinetically controlled N-acylations in water could be performed. An enzymatic one-pot one-step cascade was developed for the formation of amides from aldehydes in water that after optimization gave 97% conversion of the aldehyde to the corresponding amide. In addition, engineered variants of MsAcT with increased substrate scope could catalyze the formation of an amide in water with 81% conversion of the amine, where the wild type gave no conversion. Moreover engineered variants of MsAcT displayed up to 32-fold change in specificity towards amide synthesis and a switch in reaction preference favoring amide over ester synthesis. These engineered variants have potential as catalysts for amide-based chemistries, for example in the cascade presented.

Keywords: Amidase, Biocatalysis, Enzyme, Esterase, Enzyme engineering, Lipase, Substrate

specificity

Sammanfattning

Enzymer är naturens egna kraftfulla katalysatorer som ger stora hastighetsökningar av kemiska reaktioner och gör livet på jorden möjligt. Biokatalys är ett ständigt växande område som använder enzymer eller mikroorganismer som katalysatorer för kemisk syntes. Genom att använda enzymer som generellt har utvecklats för specifika reaktioner under milda förhållanden och temperaturer kan biokatalys vara ett miljövänligt alternativ till traditionell kemi.

Syntes av amider, genom att undvika dålig atomekonomi, är ett av de mest prioriterade forskningsområdena. Biokatalys skulle kunna vara en lösning, men de befintliga katalysatorerna har ett dåligt substratomfång vilket begränsar möjligheterna. Esteraser/lipaser uppvisar vanligtvis brett substratomfång och katalytisk promiskuitet, men är i allmänhet dåliga på hydrolys av amider jämfört med amidaser/proteaser. En hypotes till skillnaden mellan de två enzymklasserna är att en viktig vätebindning kan bildas i amidaser/proteaser, vilket underlättar övergångstillståndet för inversionen av kvävet i det aktiverade komplexet under katalys.

I denna avhandling har arbetet fokuserats på att införa en stabiliserande vätebindningsacceptor i esteraser som liknar den som återfinns i amidaser/proteaser, i syfte att utveckla bättre enzymatiska katalysatorer för amidbaserade kemi.

Genom två strategier, antingen med sidokedja eller vatten interaktioner, har varianter skapats i tre olika esteraser, vilka visade som mest upp till 210 gånger ökad relativ amidasspecificitet jämfört med vildtyps-enzymet. Minskningen i aktiveringsentalpi för amidhydrolysen var 3,3 kcal/mol, vilket motsvarar en svag vätebindning. Dock missgynnades varianten entropiskt genom att frysa vattenmolekyler, vilket resulterade i att skillnaden i Gibbs fria aktiveringsenergi blev 2,1 kcal/mol. Resultaten visar en uppskattad undre gräns för hur mycket vätebindning kan vara värt i enzymkatalyserad amidkemi.

Syntes av amider i vatten är problematiskt på grund av ogynnsam kinetisk kontroll. Med Mycobacterium smegmatis acyltransferas (MsAcT) som uppvisar kinetiskt reglerad N-acylering vid syntes i vatten är det dock möjligt. En enzymatisk kaskadreaktion i vatten utvecklades för syntes av amider från aldehyder, utförd i samma reaktionskärl i ett steg, och uppnådde efter optimering 97% omsättning. Utöver detta skapades varianter av MsAcT med utökat substratomfång som kunde syntetisera en amid i vatten med 81% omsättning där vildtyps-enzymet inte gav någon omsättning. Dessutom konstruerades varianter av MsAcT som hade upp till 32 gångers förändrad specificitet mot amidsyntes och en ändrad reaktionspreferens där amidsyntesen gick fortare än estersyntesen. Dessa modifierade varianter har potential som katalysatorer för amidbaserad kemi exempelvis i den presenterade reaktionskaskaden.

Nyckelord: Amidas, Biokatalys, Enzym, Esteras, Enzymmodifiering, Lipas, Substratspecificitet

List of appended papers

Paper I

Syrén,‡ P.-O., Hendil-Forssell,

‡ P., Aumailley, L., Besenmatter, W.,

Gounine, F., Svendsen, A., Martinelle, M. and Hult, K. (2012), Esterases

with an introduced amidase-like hydrogen bond in the transition state

have increased amidase specificity. ChemBioChem, 13: 645–648.

Paper II

Hendil-Forssell, P., Martinelle, M. and Syrén, P.-O. (2015), Exploring

water as building bricks in enzyme engineering, Chem. Commun., 51:

17221-17224.

Paper III

Land, H., Hendil-Forssell, P., Martinelle, M. and Berglund. P. (2016),

One-pot biocatalytic amine transaminase/acyl transferase cascade for

aqueous formation of amides from aldehydes or ketones. Catal. Sci.

Technol., 6: 2897-2900.

Paper IV

Hendil-Forssell,‡ P., Semlitsch,

‡ S. and Martinelle, M. Engineering the

esterase/acyltransferase from Mycobacterium smegmatis: extended

substrate scope for amide synthesis in water. Manuscript

Paper V

Hendil-Forssell, P., Semlitsch, S. and Martinelle, M. Rational

engineering of an esterase/acyltransferase for improved amidase

specificity in amide synthesis and hydrolysis. Manuscript

‡ Shared first authorship

Contributions to appended papers

Paper I

Contribution to planning, execution and analysis of experiments for

Candida antartica lipase B. Minor contribution to the writing.

Paper II

Major contribution to planning, execution and analysis of experiments.

Contribution to the writing.

Paper III

Contribution to planning and part of execution and analysis of

experiments. Minor contribution to the writing.

Paper IV


Performed a majority of the writing.

Paper V


Performed a majority of the writing.

Papers not included in this thesis

Rüdiger, A., Hendil-Forssell, P., Hedfors, C., Martinell, M., Trey, S. and

Johansson, M. (2013), Chemo-enzymatic route to renewable thermosets

based on a suberin monomer. J. Renew. Mater., 2: 124-140

List of abbreviations

aw Water activity

A/Ala Alanine

ATA Amine transaminase

BS2 Bacillus subtilis esterase

CalA Candida antarctica lipase A

CalB Candida antarctica lipase B

D/Asp Aspartic acid

E Enzyme

EC Enzyme Commission

ES Enzyme-substrate

F/Phe Phenylalanine

G/Gly Glycine

Hic Humicola insolens cutinase

H/His Histidine

I/Ile Isoleucine

K/Lys Lysine

L/Leu Leucine

MsAcT Mycobacterium smegmatis acyltransferase

mut Mutant

P Product

PDB Protein Data Bank

QM Quantum mechanics

S Substrate

S/Ser Serine

Sp-ATA Silicibacter pomeroyi amine transaminase

T/Thr Threonine

TS Transition state

W/Trp Tryptophan

wt Wild type

“We don’t have time for discussions; we have to make more mutants”

Anonymous researcher, somewhere in the Netherlands (Nov-2010)

Table of Contents

1. Introduction ................................................................... 1

1.1 Enzymes ................................................................................................ 1

1.2 Biocatalysis .......................................................................................... 3

1.3 Enzyme engineering ............................................................................ 3

2. Catalytic principles of enzymes .................................. 7

2.1 The transition state .............................................................................. 7

2.2 Enzyme kinetics .................................................................................. 8

2.3 Transition state theory ...................................................................... 12

2.4 Selectivity and promiscuity ............................................................... 14

3. α/β hydrolases ............................................................ 17

3.1 The α/β hydrolase fold - One fold many activities ........................... 17

3.2 Candida antarctica lipase B ............................................................ 20

3.3 Mycobacterium smegmatis acyltransferase (MsAcT) .................... 22

3.4 Reaction mechanism ........................................................................ 23

4. Aim and motivation .................................................... 27

5. Rational engineering for increased amidase

specificity ........................................................................... 29

5.1 Amidases and the amide bond .......................................................... 29

5.2 Amide hydrolysis ............................................................................... 34

5.3 Amide synthesis ................................................................................ 49

6. Concluding remarks ................................................... 61

7. Acknowledgements .................................................... 63

8. Bibliography ................................................................ 67

I n t r o d u c t i o n | 1

1. Introduction

1.1 Enzymes Enzymes are catalytic proteins and constitute the cornerstone of all life.

They are biological catalysts for all organisms and speed up the

chemistries essential for survival. The word enzyme was first termed by

Wilhelm Kühne in the end of the 19th century and translates to ‘in leaven’

and referred to the unformed ferments, something in yeast that could

cause fermentation.1,2 Eduard Buchner later showed that cell free extracts

from yeast could ferment sugars.3,4 Three decades later it was shown by

James Sumner that an isolated crystallized protein could contain catalytic

power,5 a notion that got increasing recognition and made the link

between enzymes and proteins. The crystallization of proteins laid the

basis for modern enzymology since it promoted the possibility of solving

the structures of the enzymes by x-ray crystallography,6 a key method for

understanding mechanistic details of enzymes.

Enzymes are made out of combinations of amino acids into polymers

called polypeptides. There are 20 common amino acids that differ in

properties and the order and type of amino acid is determined by the

gene encoding that particular enzyme.7 Depending on the order of the

amino acids in the primary structure they fold into local structural

elements such as α-helices and β-sheets called secondary structures

driven by hydrogen bonding. The hydrophobic effect causes the

secondary structural elements to further pack in space into a stabile three

dimensional structure, the tertiary structure, which can be a functional

unit on its own or further interact with other polypeptide chains forming

a quaternary structure (Figure 1).8,9 The common 20 amino acids that are

coded for by the genome offers an inconceivable quantity of possible

combinations in one polypeptide. As an example a polypeptide of only 10

amino acids in length can be made in 2010 (≈1013) possible combinations,

which is more than the number of galaxies in the observable universe.10,11

2 | I n t r o d u c t i o n

Mind that enzymes constitutes of approximately 100-1000 amino acids,

which offers a potentially immense diversity.

Figure 1. Representative examples of the different levels to describe enzyme structures. From the primary structure of a single peptide chain, forming local secondary structural elements that are further folded into a three dimensional tertiary structure, which can form a complex with other folded peptide chains in a quaternary structure. The enzyme depicted is 4-oxalocrotonate tautomerase (Protein Data Bank code 4OTB), which forms a hexamer with six identical subunits.

12

Enzymes possess some key features that make them very attractive

catalysts. Firstly they accelerate the rate of reaction compared to the

uncatalyzed reaction. The most efficient known example is arginine

decarboxylase with a rate enhancement of 1019 compared to the

uncatalyzed reaction, or in other words a reaction with a half-life of 1.1

billion years uncatalyzedi has a turnover number of 1000 s-1 when

catalyzed by the enzyme.13 Furthermore, many enzymes also work at or

close to the diffusion limit, meaning that they will make a chemical

transformation as soon as they encounter the compound in solution.14

Secondly, enzymes are often selective towards which substrate they use

and what type of reaction they catalyze. The rate enhancements and

selectivities of enzymes make it possible for life to regulate functions

without detrimental side reactions.

i a halftime of 1.1 billion years equals a turn over rate of 2*10-17 s-1 for a first order rate equation: [A]t/[A]0=e-kt -> t½=ln(0.5)/-k


1.2 Biocatalysis Biocatalysis is an ever evolving field that is a mixture of chemistry and

biotechnology: the use of enzymes or microorganisms for chemical

synthesis.15 Prominent examples from industry are the conversion of

glucose to fructose for the production of high fructose corn syrup

catalyzed by glucose isomerase16 and cellulases for the use in textile

industry to produce stone washed appearances of jeans.17 Enzymes can

either be formulated as a product for the public, for example as a

component in detergents, or used as a catalyst by chemists for synthesis.18

Traditionally, synthetic chemistry uses organic solvents, high

temperature, protecting/deprotecting steps and toxic reagents, which add

up to an undesirable environmental impact.19 Enzymes, which can

catalyze reactions at conditions that differ from the ones in vivo, can be

implemented in these synthetic processes.20,21 The rate enhancement and

selectivity displayed by enzymes enable them to work at lower

temperatures under milder conditions and allow them to circumvent

protection chemistry and thus reduce the number of catalytic steps

needed and waste generated.19 However, enzymes have evolved for

reaction conditions within living organisms and not for producing

chemicals wanted by humans. Consequently, enzymes do not always work

with the compounds or in the conditions desired and engineering to

adapt the enzymes is needed.

1.3 Enzyme engineering The advancements in molecular biology have made it possible to alter the

genetic code and thereby change the amino acid sequence of enzymes.

This enables scientists to tailor the catalytic22-24 and physical

properties23,25-27 of the enzymes to fit the process at hand. Moreover,

enzyme engineering is an important technique for elucidating the

mechanism of enzymes.28-30

Traditionally, two routes for enzyme engineering have been employed:

rational design and random mutagenesis (Figure 2).15 Rational design is

preferable used when the structure and mechanism of the enzyme is

known and the amino acid substitutions can be made on a rational

notion, that an alteration will give a desired effect based on previous

knowledge. The type of alterations could be single changes in amino acids


to replacing or removing whole segments. Numerous examples where

rational engineering has been employed can be found in literature23,31-33

and Paper I, II, IV and V in this thesis.

When the structure of an enzyme is not known or rational engineering

has not been successful, the alternative method of random mutagenesis

can be used. By this strategy the gene is altered randomly, either by gene-

shuffling or error-prone polymerase chain reaction, creating a library

containing a large amount of different variants. The library is then tested

towards the characteristic of interest, for example improved activity

towards a specific compound. With directed evolution the improved

variant is subjugated to another round of randomization in an iterative

process and continues until a variant with the desired characteristics is

found.23,26,32,34 Directed evolution makes it possible to generate variants

that would otherwise not have been thought of or created by a rational

strategy. However, the method requires the design of good screening

methods that can handle the large libraries that will be generated.

Semi-rational approaches, a combination of the two main routes, have

emerged as an attractive mean since they provide the randomness of

directed evolution but the positions altered are chosen on rational basis

yielding smaller more focused libraries.15,23,34 The random combination of

several mutations at key positions can generate variants with synergistic

effects that would otherwise have been missed with rational design.

However, for each additional position examined the size of the library and

thus workload increases.35


Figure 2. The different routes for enzyme engineering. (A) Random mutagenesis generates mutations randomly across the enzyme coding gene creating a large library containing many different variants. With semi-rational approaches (B) the alterations are focused at specific positions or segments, which provide randomness at those positions but a much smaller library that needs to be screened compared to random mutagenesis. An iterative process can be employed in both (A) and (B) where the best variant(s) are further randomized. (C) By rational design the type of alteration and the position are decided based on knowledge about the mechanism and structure. Few variants are usually created that can be handled with ease.

C a t a l y t i c p r i n c i p l e s o f e n z y m e s | 7

2. Catalytic principles of enzymes

2.1 The transition state In a chemical reaction the transformation of a substrate (S) into product

(P) requires some kind of bond breaking. The transformation is

associated with a change in Gibbs free energy (ΔG) and is for example

negative for a favored reaction. Taking a transformation of S to P where a

new bond is formed and another one is broken, a species is formed during

the transformation where the new configuration partially exists together

with the old. This species is not stable and is associated with a large

amount of free energy but momentarily the species must exist before it

breaks down to P or S. The condition of this high energetic species is

known as the transition state (TS) and marks an energy barrier that needs

to be overcome for the reaction to proceed. The energy needed for the

system to go from S to the TS is called the activation energy and the larger

the difference is between S and TS the slower the reaction will be.36,37

Enzymes accelerate the rate of reaction compared to the uncatalyzed

reaction and do so by stabilizing the TS and thereby lowering the energy

barrier needed to be overcome. It was proposed by Linus Pauling that rate

enhancement provided by enzymes is achieved by being complementary

to the strained substrates, the transition states and not the ground states,

and in that way promotes TS formation after binding of the substrate.38

The view of enzymes being complementary to the TS of the reaction and

stabilizing it compared to the uncatalyzed reaction is still valid 70 years

later, although the understanding of the catalytic principles applied by

enzymes to stabilize the TS has broadened. The mechanisms employed by

enzymes to afford TS stabilization can be grouped into five categories and

an enzyme might employ several of them: 1) approximation of reactants,

2) general acid-base catalysis, 3) conformational distortion, 4)

preorganization of the active site complementary to the TS and 5)

covalent catalysis.39-41

8 | C a t a l y t i c p r i n c i p l e s o f e n z y m e s

For the simplest enzyme catalyzed reaction, where one substrate is

converted to one product, an intermediate is first formed where the

enzyme (E) binds the substrate non-covalently forming an enzyme-

substrate (ES) complex. The ES-complex can either dissociate or cross the

transition state to form a new non-covalent intermediate of enzyme and

product (EP). Depending on the type of catalysis several novel species,

both intermediates and transition states, can be formed along the

reaction coordinate that are unattainable for the uncatalyzed reaction.39

2.2 Enzyme kinetics

The initial rate (v) of the chemical transformation of substrate S to

product P by enzymes depends on the constants kcat and KM and the

concentrations of S and enzyme as given by the Henri-Michaelis-Menten

(Equation 1, where KM is the Michaelis constant and Vmax is the maximum

reaction velocity given by kcat (the turnover number) and the total enzyme

concentration [E]).42-44

𝑣 = 𝑘𝑐𝑎𝑡[𝐸] [𝑆]

𝐾𝑀+[𝑆]=

𝑉𝑚𝑎𝑥 [𝑆]

𝐾𝑀+[𝑆] (Equation 1)

The model is valid for one substrate when the enzyme is present in a

concentration much lower than the substrate and when there is negligible

accumulation of product and intermediates except a constant

concentration of ES-complex. This period of the reaction is called the

steady state and the rate equation can be derived from Scheme 1.ii The

conditions during the steady state make it possible to account for all

enzyme molecules as either being free or in an ES-complex as well as

approximating the free substrate concentration [S]free as the total

substrate concentration [S].

ii With the assumptions at steady state, the expression for ES-concentration:

𝑘1[𝐸]𝑓𝑟𝑒𝑒[𝑆]𝑓𝑟𝑒𝑒 = (𝑘−1 + 𝑘2)[𝐸𝑆] → [𝐸𝑆] =[𝐸]𝑓𝑟𝑒𝑒[𝑆]𝑓𝑟𝑒𝑒

𝑘−1+𝑘2𝑘1

=[𝐸]𝑓𝑟𝑒𝑒[𝑆]𝑓𝑟𝑒𝑒

𝐾𝑀≈

[𝐸][𝑆]

𝐾𝑀+[𝑆], combined with the expression for rate product formation: 𝑣 = 𝑘2[𝐸𝑆],

gives Equation 1.


Scheme 1. Reaction scheme of an one-substrate enzyme-catalyzed reaction. The enzyme and substrate form an ES-complex determined by the second order rate constant k1 and the concentrations of free E and S. The ES-complex can either form product or dissociate back to free enzyme and substrate, given by the ES concentration and the rate constants k2 or k-1.

Depending on the complexity of the enzyme catalyzed reaction, the

formation of E+P from the ES-complex can comprise of several chemical

transitions, each with its own rate constant. The rate constant k2 is then

replaced with the turn over number kcat. The turn over number usually

represents the slowest reaction step, but is from the definition in Scheme

1 a combination of all first order rate constants following to the formation

of the ES-complex (Equation 2). It should be noted that kcat is a measure

that gives the maximal rate of an enzymatic reaction at unlimited

substrate concentration and a fixed enzyme concentration (i.e. Vmax,

Equation 1).

1

𝑘𝑐𝑎𝑡=

1

𝑘2+

1

𝑘3+ ⋯ +

1

𝑘𝑛 (Equation 2)

The rate constant for formation of ES-complex is generally very large and

not rate limiting for catalysis in vitro where substrate concentrations

usually exceed the KM values.45 Thus k1 is normally excluded from kcat.

Enzymes that are rate limited by the diffusion are said to have reach

kinetic perfection, as a catalytic event occurs as soon as enzyme and

substrate meet. The highest values of k1 displayed by enzymes in solution

are 108-109 M-1 s-1.46

The Michaelis constant, KM, is viewed as a relative value for the affinity

for a substrate but can during certain circumstances, where k2<<k-1, be

equivalent to the equilibrium dissociation constant. The KM value controls

the rate of a reaction by determining how large fraction of the enzyme

that will be in an ES-complex at a specific substrate concentration. For

example, at a substrate concentration equal to KM the rate will be half that

of the theoretical possible as determined by kcat and the enzyme

concentration (Equation 3).

𝑣 = 𝑉𝑚𝑎𝑥 [𝑆]

[𝑆]+[𝑆]=

𝑉𝑚𝑎𝑥

2 (Equation 3)


A low KM value means that a low substrate concentration is required to

have a large fraction of the enzyme in an ES-complex. However, a low KM

value does not necessarily equal a better substrate. As long as the enzyme

can be saturated, a higher kcat is more desirable, from a biocatalytic point

of view, since reactions are often performed at high substrate

concentrations to optimize productivity (Figure 3).47,48

Figure 3. Comparison of two enzyme variants (X and Y) highlighting how different kinetic constants affects the initial rate at different substrate concentrations. For a hypothetical enzymatic reaction, enzyme variant X has a Vmax of 10 arbitrary units and KM of 0.1 mM, while variant Y has a Vmax of 100 arbitrary units and a KM of 10 mM. At a substrate concentration of 1 mM the enzyme variants will display equal initial rates.

The most commonly used method to measure the catalytic efficiency of an

enzyme is by the specificity constant, kcat/KM. The usefulness of the ratio

of the kinetic constants is that the value entails information both on

substrate affinity and transition state stabilization and thus provides a

lower limit of the second order rate constant for productive substrate

binding. The specificity constant can be determined by taking the ratio of

kcat/KM or estimated by measurement of the rate when substrate

concentration is much lower than KM (Equation 4).

0

5

10

15

20

25

0 1 2 3

Init

ial r

ate

[ar

b. u

nit

]

Substrate concentration [mM]

Variant X

Variant Y


𝑣 =𝑘𝑐𝑎𝑡[𝐸]𝑓𝑟𝑒𝑒[𝑆]𝑓𝑟𝑒𝑒

𝐾𝑀 , 𝑤ℎ𝑒𝑟𝑒 [𝐸]𝑓𝑟𝑒𝑒 ≈ [𝐸] → 𝑣 =

𝑘𝑐𝑎𝑡[𝐸][𝑆]𝑓𝑟𝑒𝑒

𝐾𝑀 (Equation 4)

The initial rate at low substrate concentrations ([S]<<KM) depends on the

free substrate concentration since the kcat/KM ratio is constant. The use of

the specificity constant as a suitable measure of enzyme catalytic

efficiency has been questioned,47-49 since it does not account for changes

in substrate and product concentrations during the time course of a

reaction. Taking Figure 3 as an example, enzyme variant X has ten times

higher substrate specificity compared to enzyme variant Y, but at high

susbtrate concentration (> 1 mM) variant Y has a higher productivity.

However, the consensus is that the specificity constant is adequate to

determine the selectivity (specificity ratios) of one enzyme towards

different substrates, A and B, and can be done so by comparing the

individually determined specificity constants or by comparing the rates

for the different products in competition (Equation 5).

𝑣𝐴[𝐵]

𝑣𝐵[𝐴]=

(𝑘𝑐𝑎𝑡

𝐾𝑀⁄ )𝐴

(𝑘𝑐𝑎𝑡

𝐾𝑀⁄ )𝐵

(Equation 5)

The discussion above is valid for one-substrate kinetic models. For more

complex reactions involving, for example, several substrates and/or

inhibition, other forms of the Henri-Michaelis-Menten equation need to

be used. Such examples are transacylations in organic media catalyzed by

lipases (Equation 6, where A and B are the two substrates and KMA and

KMB are their respective KM-value).50

𝑣 =𝑘cat[E][A][B]

𝐾MA [B]+𝐾M

B [A]+[A][B] (Equation 6)

In many two-substrate reactions the kinetic model can be simplified by

keeping one substrate at a high concentration well above its KM. The rate

of the reaction seems to depend on one substrate only,51 called pseudo

first order reactions, and the kinetic constants for the other substrate can

be estimated with good accuracy.


2.3 Transition state theory With the Eyring equation52,53 (Equation 7, where 𝜅 is the transmission

coefficient, kB is the Boltzmann constant, T is the temperature in Kelvin, h

is Planck’s constant and R is the ideal gas constant) and the transition

state theory, the energy barrier of the reaction can be estimated, if kcat and

KM or kcat/KM have been determined. The change in Gibbs free energy of

activation (ΔG‡) can further be divided into free enthalpy and entropy of

activation (ΔH‡ and ΔS‡)

𝑘 = 𝜅 𝑘𝐵𝑇

ℎ𝑒

−∆𝐺‡

𝑅𝑇 = 𝜅 𝑘𝐵𝑇

ℎ𝑒

−∆𝐻‡

𝑅𝑇 𝑒∆𝑆‡

𝑅 (Equation 7)

With the help of kinetic measurements and free energy calculations it is

possible to estimate the TS stabilization provided by enzymes compared

to the uncatalyzed reaction and relate the contributions from kcat and KM

to catalysis. The calculated relative energy levels can be visualized in an

energy profile diagram as shown in Figure 4, which helps in

comprehending the contributions from different factors affecting the

enzymatic reaction.

From the energy profile diagram it is clear that an enzyme becomes more

efficient by lowering the transition state energy level, ES‡, relative the free

enzyme and substrate. The enzyme can do so by increasing the specificity,

kcat/KM. The increased specificity could come from an alteration in kcat, KM

or both but to have an overall effect the TS energy level needs to be

affected. The change in Gibbs free energy can be divided into enthalpy

and entropy of activation as mentioned earlier with a temperature

dependent investigation, which provides further valuable information

about the reaction mechanism. In general a large positive ΔH‡ signifies

that for attaining the TS structure a large amount of chemistry in form of

bond deformation like stretching is needed. In contrast, if TΔS‡ is large

and negative, the formation of the TS requires the reacting substrate to be

constrained in precise orientations. Uncatalyzed reactions generally

display large enthalpies of activation and for the enzymatic counterpart a

major contribution to catalysis comes from reduced activation enthalpies

rather than increased entropies of activation.46,54


Figure 4. Energy profile diagram for an enzyme catalyzed reaction and the corresponding uncatalyzed reaction where one substrate (S) is converted into one product (P). The overall change in free energy of the reaction ΔGrxn is path independent and is the same whether a catalyst is used or not. In the enzyme catalyzed reaction (black line) going from S to P an enzyme-substrate complex is formed (ES) before the formation of the transition state ES

‡.

The change in free energy going from free enzyme and free substrate (ground state) to reach the transition state can be calculated by the Eyring equation ΔGkcat/KM = RT ln (kBT/h – kcat/KM) or by combining the individual changes in free energy of going from E+S to ES and going from ES to ES

‡. For the uncatalyzed reaction (grey line) the substrate directly

goes from S to P by overcoming the TS barrier (S‡). The change in free energy between

ground state and TS is dependent on the rate of the uncatalyzed reaction knon. The difference in free energy between the two transition states (ΔΔG

‡) is equal to the

stabilization of the TS brought upon by the enzyme.

The energy profile diagrams are primarily constructed from steady-state

kinetics meaning that an enzymatic reaction entailing several enzyme-

substate intermediates upon substrate binding is masked by the one

containing the largest barrier. It is important to remember that kcat often

represents several chemical steps and sets a lower limit of the slowest of

them.


2.4 Selectivity and promiscuity The selectivity of enzymes is what makes them the powerful catalysts for

life, since without it there would be no control for the cells. The selectivity

is what makes the enzyme distinguish different substrates from each

other and it can for example be divided into four categories: chemo-,

enantio-, regio-, and substrate selectivity. The selectivity is determined

by the ratios of the enzyme specificities towards the two substrates

(Equation 5) and can also be converted into free energies and viewed in a

reaction profile diagram with the Eyring equation (Equation 7). Hence,

the selectivity can be simplified into different effects of an enzymes

affinity and TS stabilization towards different compounds and the

absence of an enzymatic reaction is merely that one energy barrier is too

high for the enzyme to traverse.

Chemoselectivity is the possibility of an enzyme to distinguish between

two different functional groups and using one over the other (Figure 5A).

For example two lipases, Rhizomucor miehei lipase and Candida

antartica lipase B (CalB) have been shown to be selective towards

alcohols over thiols ranging from 10-730 over the uncatalyzed reaction, a

selectivity originating from KM.55 Enantioselectivity is the possibility of an

enzyme to discriminate two enantiomers from each other (Figure 5B).

Enantiomers are chemically identical except in the three dimensional

space, where they are the mirror image of each other. As an example CalB

is very selective towards one enantiomer over the other of 2-octanol due

to a substrate pocket that fits the medium substituent of the chiral center

for one of the enantiomers.56,57 Regioselectivity is an enzyme’s

discrimination of one functional group over another depending on the

structural position of the groups (Figure 5C). Taking CalB as an example

again, the enzyme is selective towards the primary hydroxyl of

carbohydrates compared to the other secondary hydroxyl groups.58,59

Substrate selectivity is often used as a general collective term for all kinds

of selectivities. It is a reasonable practice since a discrimination of

different chemical groups such as an alcohol from an amine is the

(chemo-) selectivity towards one substrate over the other. In this thesis

the term substrate selectivity is added for situations, when the other types

are not applicable. Such an example is the selectivity of CalB towards

ethyl pentanoates over other ethyl esters (Figure 5D).60 The chemical


group is the same, they are all esters, and what differs are the chain

length on the acyl part and consequently they are different substrates.

Figure 5. Schematic overview of the types of selectivities displayed by enzymes. In the middle of the figure CalB is depicted that displays chemoselectivity (A) towards 1-hexanol over 1-hexanethiol in the transacylation of ethyl octanoate.

55 In the transacylation of S-ethyl

thiooctanoate with racemic 2-octanol (B) CalB displays enantioselectivity towards the R-enantiomer.

56 In the acylation of α-D-glucopyranose with octanoic acid (C) CalB was

regioselective towards the C-6 hydroxyl group.59

Furthermore, CalB displays substrate selectivity (D) towards ethyl pentanoate compared to other ethyl esters in the transacylation with methanol.

60

The selectivities of enzymes towards their substrates and catalyzed

reactions are not always absolute and they are known to even catalyze

different reactions. This phenomenon is known as promiscuity and can be

categorized into three types: condition-, substrate- and catalytic

promiscuity.61 The promiscuity should be regarded as a selectivity feature

that is uncovered because the enzyme is placed in a condition with

substrates that it normally does not encounter. Promiscuity is considered


as a left over from evolution where the ancestral enzymes diverged and

became more specific towards certain type of chemistries. If the

promiscuity does not affect the native function of the enzyme in the cell

there is no selective pressure to remove it. Promiscuous activities could

be good starting points for engineering catalytic functions in stable

enzyme scaffolds.61-64

.

α / β h y d r o l a s e s | 17

3. α/β hydrolases

3.1 The α/β hydrolase fold - One fold many activities The α/β hydrolase fold is one of the biggest enzyme superfamiliesiii.66 The

family is called so because the structure has an open-twisted β-sheet,

typically consisting of 8 strands, making up the core and it is surrounded

by α-helices (Figure 6).

Figure 6. The α/β hydrolase fold diversity represented by (A) Haloalkane dehalogenase (PDB code 1B6G) from Xanthobacter autotrophicus and (B) Lipase A from Bacillus subtilis (PDB code 1R50). The haloalkane dehalogenase is a 310 amino acid long enzyme with a molecular weight of 35 kDa while the lipase contains 181 amino acids with a molecular weight of 19 kDa. Although different in size and catalytic preference the enzyme share the same fold with a central β-sheet (red) surrounded by α-helices (blue). The black arrows indicate the nucleophiles in the respective active sites situated at the end of a β-strand linked to an α-helix with a sharp turn, called the nucleophilic elbow.

iii Superfamily is the highest order of grouping related proteins together that are descendant from a common ancestor.65

18 | α / β h y d r o l a s e s

The same fold comprises enzymes with a wide catalytic preference such as

lipases/esterases,67 amidases,67,68 haloalkane dehalogenases,67,69 epoxide

hydrolases70 and hydroxynitrile lyases.71 Although different in both

primary structure and reaction mechanism the catalytic machinery is

similar in addition to the fold. They all have a catalytic triad with a

nucleophile, serine, cysteine or aspartate, and a charge-relay system

encompassing a histidine and an aspartate/glutamate, functioning as an

acid and base and stabilizing the charges (Scheme 2).67 The nucleophile is

situated on a sharp turn, named the nucleophile elbow, which is highly

conserved and exposes the nucleophile to more easily connect with the

substrate and the catalytic histidine.67

The α/β hydrolase fold is very malleable and can tolerate a wide range of

alterations without compromising the catalytic machinery demonstrated

by the vast structural diversity within the family,72 exemplified by the

large inserts in Geotrichum candidum lipase67,73 of the large esterase

family72 and the migration of the catalytic aspartate in human pancreatic

lipase.67,74 Moreover, the enzymes within the α/β hydrolase fold often

display catalytic promiscuity for example direct epoxidation and

conjugate additions catalyzed by lipases61,75 and esterase activity

displayed by hydroxynitrile lyases.76 These features make the α/β

hydrolase fold enzymes interesting in a biocatalytic perspective75,77,78 and

in particular lipase B from C. antarctica.58


Scheme 2. The first step in the reaction mechanism following the ES-complex formation in enzyme classes belonging to the α/β-hydrolase fold superfamily. A) The nucleophilic attack of the catalytic serine on the carbonyl carbon of the ester in lipases/esterases forming a tetrahedral intermediate that is stabilized by the oxyanion hole B) In hydroxynitrile lyases the catalytic serine abstracts a proton from the substrate instead of forming a covalent intermediate. The negative charge formed on the cyanide is stabilized by a positively charged lysine. C) In haloalkane dehalogenases the nucleophilic amino acid is an aspartate. The aspartate attacks the carbon in α-position to the halogen, which is displaced in a Sn2 type reaction stabilized by a tryptophan. The catalytic cycle is completed through an attack by water on the catalytic aspartate of the ester intermediate releasing the product alcohol. D) Epoxide hydrolases are similar to haloalkane dehalogenases. The epoxide is attacked by the catalytic aspartate and the formed oxyanion abstracts a proton from a tyrosine. The product diol is released through an attack by water on the aspartate of the ester intermediate. Scheme adapted from Jochens et al.

66


3.2 Candida antarctica lipase B Lipases (EC 3.1.1.3), or triacylglycerol lipases, are important for the

metabolism of organisms, since they hydrolyze triglycerides, an energy

storage molecule primarily used by eukaryotes,79,80 into free fatty acids

and glycerol.

The organism C. antarctica was first isolated in the middle of the 1960s

from a lake in the Antarctic81 and the same organism was in the end of the

1980s identified as a lipase producing organisms in a screening of

samples from Japan82-84. One of the lipases this organism produces was

designated lipase B and the crystal structure was solved in 1994 as the

sixth structure of a lipase.85

CalB belongs to the α/β hydrolase fold superfamily and the protein

consists of 317 amino acids and has a molecular weight of 33 kDa (Figure

7).85 The enzyme has unique features with a for lipases unusual

pentapeptide sequence in the nucleophilic elbow (TWSQG compared to

the consensus sequence GXSXG)67,85 and no pronounced lid and the

absence of interfacial activation.86,87 The enzyme has attracted much

interest from academia and industry due to its high chemo-,55,88,89

enantio-56,90-92 and regioselectivity59,93-95 as well as stability58,93,95,96 in

organic solvents as immobilized preparations.


Figure 7. The structure of CalB (PDB code 1LBS). The catalytic triad is shown in ball&stick in the center and extends parallel to the β-sheet towards the left.


3.3 Mycobacterium smegmatis acyltransferase Mycobacterium smegmatis acyltransferase (MsAcT) (EC 3.1.1.x) belongs

to the SGNH-hydrolase superfamily, named after the conserved amino

acids (serine, glycine, asparagine and histidine) important for catalysis.97-

99 The SGNH-hydrolases have been described as the “α/β hydrolase fold

in-laws”72 and are also known as GDSL-esterases named after the

different motif around the catalytic serine compared to lipases.98,100,101

The SGNH-hydrolases have a catalytic triad but differ from α/β

hydrolases by having the catalytic serine very close to the N-terminal on a

loop region and not on a tight turn and the catalytic histidine and

aspartate are only separated by two amino acids in the primary

structure.99,102 Within the SGNH-hydrolase family many types of activities

can be found including thioesterase,103-105 esterase,102,106-109

lipase,102,106,107,109 rhamnogalacturonan acetylesterase99,108 and

lysophospholipase.102,109,110 MsAcT is rare among the SGNH-hydrolases

since it forms an octamer in vitro (Figure 8) that buries the active site in a

hydrophobic channel and as a consequence promotes unique acyl transfer

capabilities.97 Each subunit consists of 216 amino acids with a molecular

weight of 23 kDa. Only few esterases have been reported to catalyse acyl

transfer reactions in water, often belonging to the Candida antarctica

lipase A (CalA) superfamily.111-114 Since the first report of the crystal

structure of MsAcT, the enzyme has been used in water systems for the

synthesis of peracetic acid97 for surface decontamination,115,116 in situ

oxidations of aldehyde to acid117 or ketones to lactones,118 and synthesis of

neopentyl glycol diacetate97,117,119 and amides (Paper III-V).


Figure 8. Structure of MsAcT (PDB code 2Q0S). (A) Quaternary structure of the eight subunits folding into one large complex. Each subunit (color coded distinctly) have the active site covered by two other subunits restricting the access. (B) The tertiary structure of one subunit of MsAcT. The catalytic triad is shown in ball&stick and elongates anti-parallel to the central β-sheet. The catalytic serine is situated on an α-helix connecting to a loop region without the constraint of a typical nucleophilic elbow.

3.4 Reaction mechanism The reaction kinetics of CalB and MsAcT follow a ping-pong bi bi

mechanism (Scheme 3)50 that is common for lipases120,121 and serine

proteases.122-124 The catalytic triad of CalB consists of Ser105, His224 and

Asp187 and the oxyanion hole constitute of Thr40 and Gln106. For

MsAcT the catalytic triad consists of Ser11, His195, Asp192 and the

oxyanion hole from Ala55, Asn95 and Ser11 (backbone NH). The

importance of the catalytic triad have for example been investigated by a

mutational study in subtilisin, where it accounted for 106 in kcat/KM,

predominantly due to reduced kcat.28 Moreover, the value of the hydrogen

bonds in the oxyanion hole have been studied in subtilisin,125 papain,126

cutinase127 and CalB128 where the loss of one hydrogen bond decreased

the specificity by a factor up to 103, mainly as a result of decreased kcat.

The studies indicate the significance of the catalytic machinery for the

rate accelerations displayed by these enzymes.


Scheme 3. Reaction mechanism for the transacylation of an acyl donor to an acyl acceptor catalyzed by lipases/esterases (amino acid numberings are according to CalB). The reaction is divided into two halves, acylation and deacylation, with the acyl-enzyme (bottom right) as an uniting intermediate. X = NH, O or S and R

2 and R

3 = H, CnHm.

The catalytic cycle can be divided into two half-reactions, acylation and

deacylation. During acylation, the first step of catalysis following the

formation of the ES-complex is the nucleophilic attack on the carbonyl

carbon by the catalytic serine. The catalytic histidine is promoting the

attack by acting as a base withdrawing the hydrogen from the serine. The

catalytic aspartate helps by stabilizing the positive charge formed on the

histidine through hydrogen bonding. In the tetrahedral intermediate (TI1)

formed upon the nucleophilic attack by the serine the negative charge on

the former carbonyl oxygen is stabilized by hydrogen bonds donated by

the oxyanion hole. The acylation is completed by the breakdown of TI1 to

the acyl-enzyme and leaving of the first product, an alcohol, amine, thiol

or water that accepts the hydrogen from the positively charged catalytic

histidine now acting as an acid. The deacylation, which is the reverse of

the acylation, starts with the binding of a new nucleophile, the acyl


acceptor, to the acyl-enzyme intermediate. Thereafter, the nucleophile is

activated by the catalytic histidine and attacks the carbonyl carbon

forming a new tetrahedral intermediate (TI2). The charges are again

stabilized by the oxyanion hole and the catalytic aspartate. The

deacylation is completed by the breakdown of TI2 to the second product

and free enzyme by the reformation of the carbon-oxygen double bond of

the product and the transfer of a proton from the catalytic histidine to the

serine.

Depending on the nature of the acyl acceptors the products will differ. For

example, during hydrolysis (X2 = O, R3 = H) the product will be a

carboxylic acid, which is the usual case when the reactions are run in

water. However, if another nucleophile is present in competition with

water the outcome could be a transacylation, meaning that the acyl donor

is transferred to another acyl acceptor than water. The common

denominator is the acyl-enzyme and the outcome depends on the

selectivity displayed by the acyl-enzyme towards the different

nucleophiles and their concentrations.

A i m a n d m o t i v a t i o n | 27

4. Aim and motivation

The aim of this study was to increase the understanding of what separates

an esterase (including lipases) from an amidase (including proteases) and

to use that knowledge to develop better enzymatic catalysts for amide-

based chemistries. The rationale behind the study consists of two parts.

First, amide-type chemistries are important for the industry. As an

example, over 90% of all drug candidate molecules from AstraZeneca,

GlaxoSmithKline and Pfizer in 2006 contained a nitrogen atom.

Moreover, 12% of all chemical transformations during the synthesis of

these drugs candidates were acylations and a majority of them were the

transformation of amine to amide.129 For that reason amide bond

formation avoiding poor atom economy has been identified as one of the

most important research areas by American Chemical Society, Green

Chemistry Institute and several global pharmaceutical corporations.130

Proteases are also important for virulence factors of viruses such as

human cytomegalovirus131 and perhaps more importantly human

immunodeficiency virus132 and understanding the mechanism of the

enzymes could help in development of efficient therapies.

The second part of the motivation is more of a basic scientific nature;

esterases have the same catalytic triad121 as serine-proteases and can even

share the same fold,68 but the enzyme classes display different

specificities towards esters and amides. At a first glance the ester and

amide substrates are similar (Figure 9) and the only difference is that the

oxygen is replaced by a nitrogen and hydrogen (NH-group). So why can

one class of enzymes efficiently catalyze the hydrolysis of amides and the

other cannot? One hypothesis is that the difference belongs to a hydrogen

bond donated by the NH-group of the amide substrate in TS that can be

accepted by amidases and not by esterases. The hydrogen bond facilitates

nitrogen inversion and lowers the free energy of activation for the

transition state.133

28 | A i m a n d m o t i v a t i o n

Figure 9. The structural similarity between an ester and an amide. In an amide the oxygen of the ester-functionality is exchanged to a nitrogen and a hydrogen.

According to the hypothesis that nitrogen inversion represents the

highest energy barrier, variants of CalB and MsAcT were created where a

hydrogen bond acceptor was introduced and evaluated in the hydrolysis

and with MsAcT also synthesis of amides. Synthesis of amides in water is

not easily attained due to unfavorable kinetic reaction control. However,

with the kinetically controlled N-acylations catalyzed by MsAcT the

synthesis of amides in water could be performed.

In paper I, II and V two engineering approaches were applied to afford

TS stabilization for nitrogen inversion, either by interaction from an

amino acid side chain or molecules of water. The variants were tested in

the hydrolysis of amides. In paper III a cascade was developed for the

synthesis of amides in water from aldehydes with the wild-type enzyme of

MsAcT coupled to an amine transaminase (EC 2.6.1.x). In paper IV and

V engineered variants of MsAcT were constructed to either afford a

broader scope of amide products or improved specificity for the synthesis

of amides.

R a t i o n a l e n g i n e e r i n g f o r i n c r e a s e d a m i d a s e s p e c i f i c i t y | 29

5. Rational engineering for increased

amidase specificity

5.1 Amidases and the amide bond Amides are more stable than esters due to a partial double bond character

of the nitrogen-carbonyl carbon bond. The free π-electrons of the

nitrogen are able to overlap with the π-orbital of the carbonyl carbon and

thus forming a more stable conjugated system (Scheme 4).134

Scheme 4. The electronic structures and resonance stabilization of amides. Other bonds than the N-C-O are omitted for clarity.

The resonance stabilization makes the carbonyl carbon of amides less

electron-deficient and more unreactive towards nucleophiles compared to

for example esters. Thus, amides are very stable at neutral conditions and

as an example peptides have been reported to have half-lives of 7-600

years.135-138

Amidases (the term also including proteases) are good catalysts for the

hydrolysis of both amides and esters.139-144 Amidases are potent catalysts

of both amide and ester hydrolysis due to the general

acylation/deacylation-mechanism depicted in Scheme 3. Although the

catalytic machinery and folding class can differ among the different

amidases they all promote hydrolysis by two common strategies: (1)

interaction with the carbonyl oxygen by hydrogen bonds making the

carbonyl carbon more electron-deficient and prone to attack by (2) an

activated nucleophile (side chain or water).145,146 However, these types of

interactions are general for both amide and ester hydrolysis catalyzed by

30 | R a t i o n a l e n g i n e e r i n g f o r i n c r e a s e d a m i d a s e s p e c i f i c i t y

both amidases and esterases (term including lipases) and do not explain

the differences in specificities. The relatively high amide hydrolytic

activities in amidases compared to esterases is hypothesized to be due to

a hydrogen bond stabilization that lowers the activation energy for

nitrogen inversion in TS.133 This hydrogen bond can be accepted in

amidases but not esterases and seems to be a universal solution employed

by amidases. An investigation of 15 different amidases representing 11

folding families with 9 different reaction mechanisms revealed that the

hydrogen bond interaction for nitrogen inversion was formed or

improved during construction of the structure representing the TS.133 The

study was later extended by nine additional amidases with identical

outcome.147

The stereoelectronic effect

Due to a stereoelectronic effect the free lone pair on the nitrogen will

form anti-periplanar to the bond formed between the catalytic

nucleophile and the carbonyl carbon of the amide functionality during

acylation.148,149 The former carbonyl oxygen will at the same time orient

one lone pair orbital in the same direction. The reason for this is a

favorable interaction between the orbitals of the non-bonding lone pairs

on the nitrogen and oxygen and the anti-bonding sigma orbital of the

bond being formed, which causes the lone pairs to be oriented anti-

periplanar to the bond formed (Figure 10). This linear combination of

atomic orbitals is worth around 5 kcal/mol in transition state

stabilization.148,149


Figure 10. Lone pair orientation on the amide nitrogen during acylation/deacylation of Nu-

according to the stereoelectronic effect. The lone pairs on the heteroatoms next to the bond formed/broken between Nu

- and the carbonyl carbon will be oriented anti-periplanar for a

stabilization by mixing of non-bonding lone pair orbitals (n) with anti-bonding sigma orbital (σ*). Figure adapted from the work by Syrén.

147

In accordance to the principle of microscopic reversibility150 the bond

broken needs two orbitals on the heteroatoms that are anti-periplanar to

lower the transition state energy. To be precise, in the tetrahedral

intermediate, the bond that is broken requires anti-periplanar lone pairs

on the other two heteroatoms. In an amidase the attack of the nucleophile

will cause the lone pair to point away from the catalytic base and can

therefore not abstract a proton in that conformation.133,151 The nitrogen

either has to invert or rotate for the reaction to proceed. It has been

proposed that the nitrogen inversion/rotation for amines is an inversion-

dominated process with TS energies ranging from 4-16 kcal/mol.152 It is

likely that the inversion/rotation of a tetrahedral intermediate with a

(large) substrate in the active site of an enzyme is an inversion-dominated

process as a result of steric clashes and favorable binding interactions (for

example substrate peptide side chains with protease active site). Using

quantum mechanics (QM) calculations the total barrier for nitrogen

inversion going from the ES-complex was estimated to 19.6 kcal/mol.89 In

addition, the TS for nitrogen inversion was calculated to be around 4.5

kcal/mol higher than TI1 (Scheme 5).147 With a stabilizing hydrogen bond

acceptor added to the model, the TS was lowered by 4 kcal/mol with a

difference between amide and ester transition states of 3 kcal/mol.133 A

difference is expected between an ester and an amide due to the ground

state resonance stabilization of the amide characterized by the high

rotational barrier of amides.153 Nitrogen inversion (or rotation) is

essential for the hydrolysis of amides by amidases/esterases. The reaction

mechanism for amide hydrolysis by α/β-hydrolases is as a result altered


compared to the mechanism shown in Scheme 3. A new step with a new

transition state for nitrogen inversion is added going from TI1 with the

nitrogen lone pair pointing away from the catalytic histidine to a new

tetrahedral intermediate (TI1-2) where the nitrogen has inverted (Scheme

5). Since the oxygen of the ester functionality will have two lone pair

orbitals after an attack by a nucleophile during ester hydrolysis, one of

them will point towards the catalytic base/acid (histidine) and there is no

need for inversion.


Scheme 5. Reaction mechanism for transacylation of an amide catalyzed by CalB. A new step, nitrogen inversion, is needed for the reaction to proceed towards products since the lone pair orbital on the scissile NH-group of the amide will point away from the catalytic histidine. X

1 = NH,

O or S and R2 and R

3 = H, CnHm


Types of hydrogen bond stabilization for nitrogen inversion

The hydrogen bond acceptor observed in amidases can be divided in two

groups: (1) substrate-assisted or (2) enzyme-assisted. The enzyme

assisted hydrogen bond acceptors are predominantly from a backbone

carbonyl.133,147 The only exception known to enzyme-assisted amidases,

where the hydrogen bond acceptor comes from a side chain instead of a

backbone carbonyl, is carboxypeptidase A. In a study with

carboxypeptidase A from bovine the side chain responsible for the

stabilizing hydrogen bond, Y248, was exchanged into a phenylalanine

resulting in a 50-fold decrease in amidase specificity while the esterase

specificity was decreased seven times. The altered specificities were

primarily due to an increased kcat for the amide and decreased KM for the

ester.154 Further studies highlighting the importance of nitrogen inversion

for amide hydrolysis and synthesis are the examples of substrate-assisted

catalysis where the acyl donor mimics the spatial arrangement found in

amidases.133,155 The use of ethyl acetamidoacetate or methyl

methoxyacetate shifted the amide over ester synthesis reaction specificity

70-280 times towards amide synthesis compared to a substrate that

cannot accept a hydrogen bond from the NH-group. Substrate-assisted

catalysis is one approach for increased amidase specificity. However, the

use of substrate-assisted catalysis is not a general solution since the

product range is limited although successfully employed for the synthesis

of various amides.133,155,156 In addition, there has been one example of

substrate-assisted catalysis for the formation of amides where the

reaction is suggested to proceed by a proton shuttle mechanism that

circumvents the need for nitrogen inversion.89

5.2 Amide hydrolysis From an enzyme engineering perspective there are three viable

possibilities to introduce increased amidase specificity addressing the

inversion of nitrogen:

1) H-bond interaction with a residue side chain

2) H-bond interaction with water residues

3) H-bond interaction with enzyme backbone


Introduction of a hydrogen bond acceptor for nitrogen inversion

In paper I, II and V rational enzyme engineering was used to introduce

a hydrogen bond acceptor for TS stabilization mimicking that discovered

in amidases, by either an amino acid side chain or by water residues in 3

different esterases. The key positions in CalB and Humicola insolens

cutinase (Hic) were chosen by overlapping the active site residues with

porcine prolyl oligopeptidase. Prolyl oligopeptidase has a substrate

assisted hydrogen bond acceptor from the carbonyl carbon of the P2-

carbonyliv of the substrate peptide.133 In CalB residue I189 was found to

be in a position that an engineered side chain could occupy the same

region as the P2-carbonyl oxygen of the substrate in prolyl oligopeptidase

(Figure 11).

Figure 11. Active site of CalB overlaid with the equivalent from prolyl oligopeptidase (PDB code 1E8N). The catalytic serine and histidine are shown from both prolyl oligopeptidase (yellow) and CalB. Amino acid numbers are according to the active site of CalB. The carbonyl oxygen of the P2 amino acid (magenta) of the peptide substrate in prolyl oligopeptidase accepts a hydrogen bond donated from the scissile NH-group highlighted with a red arrow. The side chain of I189Q can accept a hydrogen bond mimicking that of prolyl oligopeptidase.

iv P2-position is the second amino acid residue in the direction towards the N-terminal from the cleaved bond.


For Hic two positions were identified I167, which corresponds to I189 in

CalB, and L64. As for MsAcT, I194 corresponds to I189 in CalB and was

chosen for that reason. Further investigation of the active site of MsAcT

identified a phenylalanine (residue 154), which is oriented with its phenyl

ring pointing towards the NH-group of the substrate with a distance of

approximately 5 Å from the para-position on the ring to the nitrogen.

F154 was for MsAcT chosen as a second hot-spot position. All variants

were evaluated in the hydrolysis of both amide and ester (Table 1).

In Paper I, the best variants with an engineered side chain acceptor in

CalB and Hic displayed 50-fold change in relative amidase over esterase

specificity towards amide hydrolysis with an absolute increase of amidase

specificity of 5-8 times compared to wild type. Interestingly, the variant

with the best relative reaction specificity was the CalB double mutant

L140Q/I189Q. The L140Q mutation was introduced aiming to hold the

side chain of Q189 in place to increase the probability of forming the

stabilizing hydrogen bond in TS. The increase in absolute amidase

specificity corresponds to approximately -1.0 kcal/mol and -2.4 kcal/mol

for the relative reaction specificity (Δmut-wtΔamide-esterΔG‡), which reflects a

weak hydrogen bond.157 From Figure 11 it is noticeable that the

introduced side chains at position 189 do not have an optimal distance

and angle as the substrate P2-carbonyl oxygen found in prolyl

oligopeptidase, which might explain the results.

All variants in Paper I that potentially could form a stabilizing hydrogen

bond between the side chain of residue 189 and scissile NH-group

displayed improved amidase specificity. The exceptions were I189S/T,

which were almost inactive. Initially, the low activity was believed to be

due to a deactivation of the catalytic histidine (Paper I) but later

presumed to rise from a well ordered TS that was entropically penalized

(Paper II). The decrease in amidase specificity of CalB I189S and I189T

compared to wild type corresponds to an increase in Gibbs free energy of

activation (Δmut-wtΔG‡) of approximately 4 kcal/mol. The entropic cost for

freezing one water molecule from solution in ice or crystals is about 2

kcal/mol at 27 °C,158 and additional water molecules could be tightly

bound in these variants.


Table 1. Substrate specificities of wild type enzymes and engineered variants towards amide and ester hydrolysis.

Variant Relative reaction specificity

(kcat/KM)amide (kcat/KM)ester ΔΔG‡

amide[d]

ΔΔΔG‡[e]

(kcat/KM)amide/ (kcat/KM)ester

[M-1

s-1

] [M-1

s-1

] [kcal mol-1

] [kcal mol-1

]

CalBa

wild type 1 0.08 32 000 0 0

I189S 0.06 0.0002 1 700 3.6 1.8

I189V 0.97 0.05 22 000 0.24 0.2

L140Q 1.1 0.03 13 000 0.48 -0.06

I189N 21 0.11 2 500 -0.21 -1.8

I189Q 38 0.49 5 300 -1.1 -2.2

L140Q/I189Q 52 0.43 3 300 -1.0 -2.4

I189G 77 0.41 2 200 -1.0 -2.6

I189Ab 210 0.75 6 100 -2.1 -3.2


Hica

wild type 1 0.09 770 000 0 0

I167Q 47 0.69 120 000 -1.3 -2.4

MsAcTc

wild type 1 3.3 8 700 000 0 0

I194V 1.3 3.1 6 300 000 0.04 0.04

F154A 5.8 38 17 000 000 -1.5 -1.1

F154Y 14 14 2 700 000 -0.9 -1.6

F154Y/I194Q 82 8.7 280 000 -0.60 -2.7

I194Q 160 3.0 51 000 0.06 -3.2 a Determined at 37 °C.

b Determined at 26 °C.

c Determined at 21 °C.

d Δmut-wtΔG

‡.

e Δmut-wtΔamide-esterΔG

‡.


Water as hydrogen bond acceptor for nitrogen inversion

In Paper II an alternative engineering strategy was explored in an

attempt to explain the improved TS stabilization of nitrogen inversion

observed with a CalB variant, I189A, which cannot form a direct

hydrogen bond interaction to the scissile NH-group of the substrate

amide. With QM calculations, molecular dynamics (MD) and

thermodynamic investigation it was demonstrated that the increased

amidase activity most likely resides from a deshielding of the enzyme

backbone that increased the water affinity and probability of water

forming a stabilizing hydrogen bond through a short water bridge

anchored to the enzyme (Figure 12).

Figure 12. MD simulation snapshot of CalB I189A with a short water bridge consisting of two water molecules between the TS structure and the carbonyl backbone of A189. The red arrow highlights the stabilizing hydrogen bond for nitrogen inversion between the scissile NH-group and water.

The I189A variant displayed a 34-fold increase in absolute amidase

specificity and 210-fold increase in relative reaction specificity (amidase

over esterase) compared to wild type at 26 °C. The result in Paper II is to

the best of my knowledge the largest increase in promiscuous amidase

activity achieved in any engineering study so far. The increase in amidase


specificity of I189A was due to a decrease in activation enthalpy (Δmut-

wtΔH‡) of -3.3 kcal/mol and a decrease in activation entropy compared to

wild type (T*Δmut-wtΔS‡) of -1.2 kcal/mol at 26 °C. The result is in

accordance to an enhanced stabilization of TS for nitrogen inversion by

hydrogen bond formation and an entropic penalty for “freezing” water

molecules in TS.158,159 The entropy of activation was for CalB I189A -3.7

kcal/mol at 26 °C, which indicates that there are some ordering of the

system going from reactants to the TS. Interestingly, the wild type also

displayed negative entropy of activation for amide hydrolysis, which

correlates with the findings in the MD investigation that there is a low

probability of forming a favorable water interaction and that amide

hydrolysis probably is aided by this interaction. In D2O the results of

amide hydrolysis further supported the presence of a fixed water network

with a difference in enthalpy of activation (Δmut-wt,D2OΔH‡) of -7.8

kcal/mol and entropy of activation (T*Δmut-wt,D2OΔS‡) of -5.3 kcal/mol at

26 °C, which correlates well with the stronger hydrogen bonds in D2O

compared to H2O and the higher entropic cost of freezing D2O compared

to H2O.160 In a previous study by Kourist et al.,161 the removal of a

stabilizing water network by rational engineering in an esterase from

Bacillus subtilis (BS2) decreased the promiscuous amidase specificity

100-fold but only reduced the esterase specificity 7-fold. The

corresponding energy differences for the amidase specificity (Δmut-

wtΔG‡amide) was 2.8 kcal/mol, which was similar to the 2.1 kcal/mol

reported for I189A in Paper II. However, the water network in wild type

BS2 was beneficial also for the ester hydrolysis whereas in CalB I189A it

was unfavorable for ester hydrolysis.

It has been shown that water is significant in enzyme catalysis for

example for binding162-164 and specificity.165,166 The study in Paper II

further demonstrates the importance of water in enzyme catalysis and

how it can be used in enzyme engineering for hydrogen bond interactions.

It could provide a viable option for favorable interactions when TS

stabilization is not easily attained through traditional engineering of side

chains. Computational tools can aid in the design and the results from the

MD simulations agree well with the experimental work. For example,

I189A displayed a 7-fold increase in probability of forming a short water

bridge of two water molecules to the backbone carbonyl oxygen from the

NH-group of the substrate (Table 2). Furthermore, I189A displayed an


order of magnitude relative increase of productive TS structures (O-C-N-

H dihedral angle of the former carbonyl oxygen and scissile NH-group >

130°)133 compared to wild type. Both I189S and I189N displayed high

probability of forming a short water bridge consisting of two water

molecules between the substrate NH-group and the backbone carbonyl at

position 189. However, the functionalities provided by the side chains

added more complexity to the water patterns and large clusters of three-

four water molecules linking the side chain functionalities, carbonyl

oxygen and substrate NH-group were simultaneously observed. Fixing

such a large cluster of water would be associated with an entropic penalty

and the variants were evaluated from their ability to form a short water

bridge from the substrate to the side chain without the formation of the

unfavorable large water bridges.

Table 2. Probability of finding a hydrogen bond between the scissile NH-group and water in CalB variants and the probability for anchoring of that water to the enzyme at position 189 in the TS via a water network. Up to four water molecules in bridge length were considered.

Variant Probability of H-

bond from scissile

NH to a water

molecule [%]a

Probability to form a water

bridge between scissile NH and

backbone 189 (NH---[H2O]n---

O=C(189)) [%]a

n=1 n=2 Total

wild type 8.9 0 0.045 0.045

I189A 9.4 0.002 0.34 0.34

I189G 0.7 0.004 0.0003 0.004

I189N 12.7 0.025 0.24 0.27c

I189S 12.1 0 0.024 0.024c

I189Vb 5.2 0 0.004 0.004

a Based on 100 ns MD simulations in duplicate.

b Based on 50 ns MD simulations in duplicate.

c Probability is for water bridge to side chain functionality at residue 189 estimated from 60

even distributed snapshots.

The work in Paper II supports a notion that water should not be

neglected in MD simulations of enzymes. In the light of the results from

Paper II it is probable that the results from Paper I are not from a

direct bond interaction but stem from an increased probability of forming


a favorable water network. Likewise, water could explain the results from

Nakagawa et al.167 where random mutagenesis of a lipase from

Pseudomonas aeruginosa afforded a variant with 20-fold increase in

reaction specificity favoring amide over ester hydrolysis compared to the

wild type. The three mutations were positioned more than 6 Å away from

the scissile NH-group of the substrate and thus unlikely to directly

interact with a hydrogen bond.

Amidase activity in MsAcT

In Paper V the promiscuous amidase activity was engineered in MsAcT.

The best variant, I194Q, displayed a 160-fold increase in relative amidase

over esterase specificity compared to the wild type (Table 1).

Interestingly, I194Q displayed virtually unchanged absolute amidase

specificity but a large decrease in esterase specificity almost exclusively

due to increased KM (Table 3). Moreover, the F154A variant displayed 12

times increase in amidase specificity but also a two-fold increase in

esterase activity. From Table 1 it is clear that MsAcT has a very high

amidase specificity compared to the two other enzymes tested but also a

very high esterase activity. Generally, the high ester specificities displayed

by the MsAcT variants towards p-nitrophenyl acetate are due to low KM

values. Some of the MsAcT variants, such as F154A, displayed substrate

specificities towards the tested ester close to the diffusion limit. The high

specificities indicate that the formation of the ES-complex, determined by

the k1 rate constant, likely had an impact on the determined kcat value.

Interestingly, the KM values for MsAcT wt, F154A and I194V towards the

amide seemed to be at least three orders of magnitude greater than the

ester values, since no variant showed any tendency to substrate saturation

at the highest amide concentration of 1 mM. As a consequence, the

apparently high KM values of the amide imply that the ratio in kcat

between amide and ester hydrolysis are in a range of 10-3-10-4, which is

close to the difference of the inherent reactivities.139


Table 3. Kinetic parameters for p-nitrophenyl acetate hydrolysis by MsAcT wild type and variants.

Variant kcat/KM [s

-1 M

-1] kcat

[s

-1] KM [µM]

wild type 8 700 000 32 3.7

I194V 6 300 000 25 3.9

F154A 17 000 000 57 3.3

F154Y 2 700 000 57 21

F154Y/I194Q 280 000 33 120

I194Q 51 000 85 1 700a

a Up to 3.6 mM ester was used.

Knowing that water can play a crucial part in substrate specificity as in

Paper I and II selected variants were investigated in MD simulations to

obtain a better molecular understanding of the experimental results

(Table 4 and Figure 13).

Table 4. Probability to form a short water bridge between the scissile NH-group of the substrate and MsAcT variant (NHsubstrate---(H2O)n---Xenzyme).

Variant Probability to form a short water bridge between the scissile

NH-group of the substrate and the enzyme [%]a

n=0 n=1 n=2 Total

wild type 0.2 0.2 0 0.4

I194V 0.1 0 0 0.1

F154A 0 33 0.9 34

F154Y 0.3 1.2 4.3 5.8

I194Q 0 74 17 91 a Based on 20 ns MD simulation.


Figure 13. MD simulation snapshots of MsAcT variants during simulation performed in waterbox. The amide substrate, p-nitrophenyl acetamide, is shown in ball&stick as well as the water residues bridging the NH-group to the enzyme and the mutated amino acids. The hydrogen bond donated by the scissile NH-group of the substrate is highlighted by a black arrow. H195 and D192 of the catalytic triad are shown to the left and A55 and N94 of the oxyanion hole are situated to the right of the substrate. A) In F154A a hydrogen bond accepting water molecule was often found (P>0.3) linked between the side chain of T93 and N94. The smaller side chain at A154 made room for additional waters further into the active site above I194.B) For F154Y the water accepting a hydrogen bond from the NH-group of the amide substrate is linked to the enzyme at the side chains of F154Y and the catalytic D192. The anchoring point to D192 could be a deactivating interaction as the amino acid side chain becomes less electronegative and disrupts the charge stabilization of H195. C) A representative snapshot (P>0.7) from the active site of I194Q. A stabilizing hydrogen bond is formed between the NH-group of the substrate and a water molecule that is held in place by the backbone carbonyl oxygen of I189Q and the introduced side chain.


From the MD study it became evident that water most likely has an

important role during hydrolysis of amides catalyzed by the investigated

MsAcT variants. The hydrogen bond formation between the scissile NH-

group of the substrate and the enzyme are almost exclusively linked by a

short water bridge consisting of no more than two molecules. The

exceptions to this were MsAcT wild type and the “control” variant, I194V,

where a similar amino acid to the isoleucine present in wild type was

introduced. In these two examples few interactions with the NH-group of

the substrate were observed and water left the active site within 2 ns after

the simulation started. The water molecules left the active site through

what seems like a water-tunnel with entrance behind the active site. The

expelling of water molecules from the substrate/ligand binding site to

make room for the substrate/ligand and use the entropy gained by

releasing water from the enzyme to the bulk solution is a known strategy

applied by enzymes.162-165 However, even without the observation of a

hydrogen bond accepted from the NH-group of the substrate the enzyme

display a relatively high amidase activity compared to other esterases.168

An examination of the O-C-N-H dihedral angle of the substrate during

MD simulations (Figure 14) revealed that the MsAcT variants, compared

to CalB, have a greater population of structures that would correspond to

productive TS (O-C-N-H dihedral angle ≥ 130° as found in amidases).133


Figure 14. Dihedral angle distributions (O-C-N-H) for the amide substrate from the total population during 20 ns MD simulation of MsAcT wild type and variants. The threshold dihedral angle for a productive nitrogen inversion transition states is ≥130°.

133 Red: wild

type, Purple: F154A, Bright red: I194V, Green: F154Y, Blue: I194Q.

The dihedral angle could be one explanation to the relatively high

amidase specificity of MsAcT. With a relatively low barrier for nitrogen

inversion (calculated hydrogen bond stabilization of 4 kcal/mol in

CalB)133 an increased probability of forming transition state structure

could increase the rate of catalysis, even though the extra energy required

for nitrogen inversion is added to the activation energy for acylation.

However, the MsAcT variants F154A and I194Q are opposites regarding

observed probability of hydrogen bond formation (Table 4), dihedral

angle distribution (Figure 14) and determined amidase specificity (Table

1). A complete representation does not seem to be covered by only these

two parameters considered during MD simulations. Taking into account

the probability of forming a productive TS structure with a stabilizing

hydrogen bond acceptor (Table 5), the experimentally determined

amidase specificity is still not fully explained, although the results follow

a similar trend to the observed relative reaction specificity (amidase over

esterase). The high amidase performance of I194Q in silico but unaltered

specificity in vitro, makes it an interesting candidate for a

0

0.05

0.1

0.15

0.2

0.25

85 95 105 115 125 135 145 155

Pro

bab

ility

Dihedral angle

WT

F154A

I194V

F154Y

I194Q


thermodynamic investigation to establish what kind of enthalpic and

entropic contributions are present. To put the probability of 51% for a TS

structure for nitrogen inversion stabilized by a hydrogen bond displayed

by MsAcT I194Q in comparison to other enzymes, CalB I189A had a

probability of 0.3% and Solanum cardiophyllum lipid acyl hydrolase

displayed a probability of 2.1%.168 A well-ordered TS as found during MD

simulation of MsAcT I194Q would be more severely hampered by entropy

than CalB I189A in Paper II.

Table 5. Probability for MsAcT wild type and variants of forming a productive TS structure for nitrogen inversion stabilized by hydrogen bond acceptor.

Variant Probabilitya to form a

productive TS structure with hydrogen bond stabilization and O-C-N-H dihedral angle

≥130° [%]

Relative reaction specificity

(kcat/KM)amide/(kcat/KM)ester

wild type 0 1

I194V 0 1.3

F154A 1.8 5.8

F154Y 4.5 14

I194Q 51 160 a The probability is calculated from the total population of simulation snapshots that fulfill the

criteria of a hydrogen bond interaction to the enzyme (either direct or by a water bridge

consisting of no more than two water molecules) and a dihedral angle ≥130°. Based on 20

ns MD simulation.

Conclusion

The outcome of Paper I, II and V provide results that reinforce the

hypothesis that nitrogen inversion is the highest TS for amide hydrolysis

and that amidases reduce the activation energy by hydrogen bond

stabilization. Although with results signifying weak to moderate hydrogen

bonds in stabilization of TS for nitrogen inversion, the results estimate a

lower limit on how much the hydrogen bond can be worth to catalysis.

Moreover, as a strategy for increased amidase specificity in esterases, the

introduction of a hydrogen bond acceptor by either side chain

engineering or water network is a viable option. As demonstrated in three

different esterases the approach is general and straightforward. The

disadvantages with the strategy result from non-optimal distances and

angles and from the negative entropy contribution of freezing side chains


or water molecules in a precise spatial arrangement. An approach

circumventing the entropic penalty of freezing side chains or water would

be to use a hydrogen bond acceptor from the enzyme backbone as found

in many proteases.147 However, engineering an enzyme backbone to adopt

a required conformation is not an easy task and a good solution could be

to search other enzyme scaffolds for the right design of a direct hydrogen

bond interaction as demonstrated by Kürten et al.168

5.3 Amide synthesis Esterases have successfully been employed in the synthesis (and

hydrolysis) of amides to produce enantiomerically enriched amines that

are industrially relevant.92,156,169-171 Alternatives to enzymatic catalysis for

amide synthesis are for example boron and (transition) metal catalysts.

Generally, these catalysts require higher reaction temperatures compared

to enzymes, which can cause racemization or discoloration, and they are

not as selective as enzymes.171 Most studies of amide synthesis have been

limited to organic solvents to circumvent unfavorable kinetic reaction

control and the thermodynamic equilibrium towards hydrolysis in water.

The use of enzymes in organic solvents provide some advantages over

water such as higher solubility of most organic compounds and easier

product and enzyme recovery.20 However, if solvent is necessary water is

regarded as a “greener” alternative, as it is non-toxic, non-flammable and

highly abundant,19 which is in good agreement to the twelve principles of

green chemistry.172 Moreover, in organic solvents the activity of enzymes

are typically dependent on the water activity (aw) and polar solvents can

strip the enzymes of water rendering them inactive.173 In addition,

organic solvents can disrupt hydrogen bonds and hydrophobic

interactions, which could cause inactivation.173 Enzymes have evolved to

function in vivo, which for many enzymes are more analogous to water

systems than organic solvents. Performing reactions in water will have

higher probability of success compared to when performed in organic

solvents, since usually little considerations are needed on aw and stability.

The high acyltransfer over hydrolysis ratio displayed by MsAcT in water,

especially demonstrated for hydrogen peroxide,97 makes it a prime

candidate as a catalyst in environmentally benign systems for the

synthesis of high value chemicals such as amides.


Cascade reaction for the formation of amides

In Paper III a biocatalytic cascade174 was developed for the synthesis of

amides from aldehydes by coupling MsAcT to an amine transaminase

from Silicibacter pomeroyi (Sp-ATA)(Table 6). Amine transaminases are

good catalysts for the synthesis of amines and have been used extensively

as reviewed by Guo and Berglund.175 However, there are few examples

where the amine formation has been coupled to an amide formation

step.175,176 Moreover, the amidation has not been performed in one-pot

one-operational-step without any intermediate work-up.

Table 6. Conversions achieved by the developed amine transaminase/acyltransferase cascade

a.

Aldehyde Acyl donor Buffer Conversionb

R1 R

2 Concentration [M] [M] [%]

phenyl acetate 0.13 0.20 25(15)c

phenyl methyl methoxy 0.10 0.20 71



benzyl methyl methoxy 0.20 0.40 59

propyl methyl methoxy 0.20 0.40 31

hexyld methyl methoxy 0.20 0.40 55

a General conditions for all entries: CHES, pH 10 with 20 mM aldehyde, 0.5 M L-alanine and

enzymes (3 U/mL Sp-ATA and 1.6 U/mL MsAcT). b Conversion based on formed amide from aldehyde.

c 100 times lower MsAcT concentration.

d Supplemented with 5% DMSO.

The difference in conversions to amide between methyl acetate and

methyl methoxyaceatate (entry 1 and 2 in Table 6) demonstrates the value

of substrate assisted catalysis giving a hydrogen bond acceptor for the

stabilization of nitrogen inversion and what it can do for the selectivity

between transacylation over hydrolysis. MsAcT has 40 times lower


specificity towards methyl butyrate compared to acetate due to a limited

space in the active site for the acyl donor (Table 7, Figure 15). However,

reducing the MsAcT concentration 100-fold to allow for the amine

concentration to build up in the first enzymatic step catalyzed by the

amine transaminase, did not give a higher conversion. During the

optimization of the cascade the concentration of the acyl donor was

doubled in order to prevent the acyl donor to be hydrolyzed before the

amide had been synthesized. The ratios of the two enzymes were in this

sense important to balance, so that the formation of the amine from

aldehyde and the deacylation of the acyl donor with the amine instead of

water were optimal. A too high concentration of MsAcT would hydrolyze

a larger portion of the acyl donor before a sufficient conversion of the

amine was achieved. On the contrary, a too low concentration of MsAcT

would make the overall reaction proceed at a low rate. The buffer

concentration was also increased to prevent an early pH drop due to

hydrolysis and thus making the amine charged (pKabenzylamine=9.3) and

inactive as nucleophile for the deacylation. The increase in buffer and acyl

donor concentration improved the conversion from 71 to 97%. For the

other aldehydes (entry 4-6, Table 6) lower conversions were achieved, but

an optimization of the enzyme ratios to balance amine formation and

amine over water selectivity would potentially increase the conversion.

Extending the acyl donor specificity in MsAcT

In Paper IV the active site of MsAcT was engineered to extend the acyl

donor substrate scope by opening up a presumed acyl donor channel

analogous to the one found in CalB.57 An extended substrate scope would

provide a general catalyst for synthesis of a broader range of amides. The

best variant, T93A/F154A, could accept acyl donors with up to five

carbons longer alkyl chain-length compared to wild type with at least

400-fold increase in specificity (Table 7, Figure 15). In addition, the

T93A/F154A exhibited broad substrate specificity with similar

specificities for the tested methyl esters.


Table 7. Substrate specificities towards different methyl esters in hydrolysis catalyzed by MsAcT wild type and variants.

Methyl ester n wild type

L12A T93A F154A L12A/ T93A

L12A/ F154A

T93A/ F154A

kcat/KM [s-1

M-1

]a

acetate 0 3200 64 460 3300 44 300 530

propionate 1 630 27 170 690 40 68 160

butyrate 2 84 16 160 130 39 77 110

hexanoate 4 b 40

b

b 11 80 64

heptanoate 5 b 10

c c 8.4 47 180

nonanoate 7 b

b

c c b

b 81

a The substrate specificities were determined by linear regression of initial rates at low

substrate concentrations. b No activity detected, <0.2 s

-1 M

-1.

c Not determined.

The mutations made in the active site of MsAcT did not significantly alter

the acyl transfer to hydrolysis ratio with F154A being an exception (Figure

16) as tested according to the reaction system by Wittrup Larsen et al.177

The F154 position might be a hotspot in the enzyme regulating the acyl

transfer to hydrolysis ratio. The reduced ratio found in F154A could be

due to increased access of water in the active site making water a better

substrate.


Figure 15. Simulation snapshots of the active site of MsAcT viewed from the oxyanion hole. The entrance of the active site is to the left and is marked by a black arrow in Figure 15A. The models were made with butyl hexanoate (ball&stick) after 50 ps MD simulations. The acyl donor colored in blue, acyl acceptor in yellow and mutated amino acids in red (native) and purple (alanine mutation). A) The active site of MsAcT wild type. The limitation of the acyl donor pocket is clearly visible and steric hindrance forces the acyl donor chain to bend outwards at C3 and adopt a more high-energetic dihedral angle. B) T93A/F154A, the mutation opens up a big hole in the active site that can easily fit longer acyl donor chains and allow them to continue outward towards the active site entrance.


Figure 16. Specific activities (U/mg) of MsAcT wild type, variants and CalA for hydrolysis and transacylation. Transacylation over hydrolysis ratios are given on top of each column. Reactions were performed in 5 mM potassium phosphate, pH 7.2 with 5% acetonitrile, 8 mM vinyl acetate and 50 mM 1-butanol.

Acyl transfers in water catalyzed by MsAcT

To demonstrate the applicability of the generated variants, MsAcT

T93A/F154A was used for the synthesis of N-benzylhexanamide in water

(Table 8).

Table 8. Synthesis of N-benzylhexanamide in water: Conversion of methyl hexanoate and benzylamine for MsAcT wild type and T93A/F154A variant.

Varianta Conversion [%]

Benzylamine

Methyl hexanoate

wild type <2 32

T93A/F154A 81 ≥99

Controlb <2 46

a Reactions performed in Davies buffer,

178 pH 10, supplemented with 10% acetonitrile, 10

mM benzylamine and 30 mM methyl hexanoate. Conversion determined after 20 h. b Without enzyme.


The T93A/F154A variant reached 81% conversion of the amine, while the

wild type did not display any activity above background. The background

reaction showed amide product corresponding to <1% conversion

compared to that of MsAcT T93A/F154A. The extended substrate scope of

the T93A/F154A variant might potentially be used for the synthesis of a

variety of amides in water. Moreover, the T93A/F154A variant could be

combined with a transaminase in a cascade as in Paper III for the

synthesis of amides from aldehydes. The conversion of amine achieved in

Paper III and IV are in the same range although there is a difference in

acyl donor concentration. MsAcT displays selectivity towards the

competing acyl acceptor over water in the deacylation of the acyl-enzyme

and it is evident that the substrates play a crucial role for the

transacylation to hydrolysis ratio. In Paper IV the transacylation to

hydrolysis ratio was determined to 5:1 (4:1 with T93A/F154A) by initial

rate measurements for the deacylation of the acyl-enzyme with butanol

using 50 mM 1-butanol and 8 mM vinyl acetate. In the synthesis of N-

benzylhexanamide from 10 mM benzylamine and 30 mM methyl

hexanoate the amine reached 81% conversion. The 81% in total

conversion of the amine suggests that the initial ratio of benzylamine over

water in the deacylation of the acyl-enzyme, when the concentration of

the amine was the highest, was greater than the 4:1 ratio displayed with

butanol and water. Furthermore, Mathews et al. reported a transacylation

to hydrolysis ratio of 12:1 for a reaction performed in water with 30 mM

hydrogen peroxide and 10 mM tributyrin.97 Interestingly, CalA did in

Paper IV perform similar with the soluble vinyl acetate/butanol assay as

the previously reported transacylation over hydrolysis ratio with 30 mM

butanol and 10 mM acyl donor in emulsion.112 The transacylation to

hydrolysis ratios can clearly be altered depending on the substrates and

further investigation is needed to deduce what factors are important for

the selectivity towards the competing nucleophile over water.

Amide synthesis in water catalyzed by MsAcT

In addition to Paper I, II and V there are many reports on esterases

with engineered amidase specificity tested for hydrolysis of

amides.161,167,179-183 However, there are scarce with examples when the

engineered variants have been employed in the synthetic direction. One

reason for this one-sided distribution of reports could be that the

engineered variants have been successful in the hydrolytic direction but

unsuccessful for the synthesis of amides. The source for this behavior


might well be water. The synthesis of amides in water requires a catalyst

that is kinetically controlled for the N-acylation, which is not the case for

many esterases. A method to circumvent hydrolysis is to remove the

water and perform the reactions in organic solvents. However, changing

solvent alters the reaction system and water seems to be important for the

effect of the mutations as indicated in the hydrolysis of amides in Paper

I, II and V. For CalB the best engineered variants for amide hydrolysis

displayed minor contribution to amide synthesis when used in organic

solvent (Table 9).

Table 9. Relative reaction specificity for CalB wild type and variants towards hydrolysis and synthesis of amides and esters.

Variant Relative reaction specificity

Hydrolysisa Synthesis

b

wild type 1 1

I189A 210 1.3

L140Q/I189Q 52 1.5

a [(kcat/KM)amide/(kcat/KM)ester]variant/[(kcat/KM)amide/(kcat/KM)ester]wt. Data from Table 1.

b [(kcat/KM)amine/(kcat/KM)alcohol]variant/[(kcat/KM)amine/(kcat/KM)alcohol]wt. Reactions with acyl acceptors

in competition towards the acyl-enzyme. Performed in toluene with 100 mM butyl butyrate and 8 mM 2-phenylethylamine and 2-phenylethanol. Unpublished results.

MsAcT with its transacylation capabilities in water offers the possibility to

perform hydrolysis as well as synthesis in the same water based system.

In Paper V MsAcT wild type and variants were investigated for the

synthesis of amide and ester in water with the amine and alcohol in

competition towards deacylation of the acyl-enzyme (Scheme 6).


Scheme 6. Reaction scheme for the synthesis of ester and amide and the hydrolysis thereof catalyzed by MsAcT. Independent of the reaction direction the acyl-enzyme and free enzyme will be the same and the selectivities will be determined by the differences between the alcohol and amine or ester and amide and their interactions with the enzyme.

The selectivities displayed by the MsAcT variants in synthesis followed a

similar trend as the results from the hydrolysis (Table 10). Water seems

to be an important component in the engineered variants for amide

hydrolysis and synthesis as the relative reaction specificity for MsAcT

followed the same trend when the reactions were performed in water

(Table 10), contrary to when the CalB variants were used in organic

solvent for the synthesis (Table 9).


Table 10. Selectivity displayed by MsAcT wild type and variants for the amine over alcohol in the synthesis of amide and ester. Relative reaction specificity for the synthesis and hydrolysis of amide over ester compared to MsAcT wild type.

Variant Selectivity(amine/alcohol) Relative reaction

specificity

(kcat/KM)amine/(kcat/KM)alcohol Synthesis

a Hydrolysisb

wild type 0.2 1 1

F154A 0.5 2.8 5.8

F154Y 2.8 15 14

F154A/I194Q 1.7 9.0 32

F154Y/I194Q 1.7 9.2 82

I194Q 6.0 32 160 a [(kcat/KM)amine/(kcat/KM)alcohol]variant/[(kcat/KM)amine/(kcat/KM)alcohol]wt. Reactions performed in Davies

buffer, pH 11, containing 20 mM benzylacetate and 50 mM 1-butanol and 1-butylamine. b [(kcat/KM)amide/(kcat/KM)ester]variant/[(kcat/KM)amide/(kcat/KM)ester]wt. Data from Table 1.

All variants with improved relative reaction specificity in hydrolysis

compared to wild type displayed increased preference towards amide

formation in synthesis. Interestingly, four of the variants did even show a

change in synthetic preference and catalyzed amide formation faster than

the ester. Thus effectively pointing at the large differences in specificity

ratios between amide/ester and amine/alcohol with selectivities of about

10-6 in hydrolysis and 100 in synthesis and highlighting the challenge of

amide hydrolysis.

The best variant for hydrolysis of amides, I194Q, was also the best variant

for the synthesis. The variant did not display an increase in absolute

amidase specificity but instead decreased ester specificity. In synthesis

MsAcT I194Q displayed a 32-fold change in reaction preference favoring

amide formation compared to wild type. The relative reduced ester


specificity made I194Q less likely to use an alcohol for the deacylation of

the acyl-enzyme and thus promoting amine selectivity.

One drawback with the reaction systems in Paper V are that different

substrates were used in the hydrolysis and synthesis direction. Different

substrate pairs can interact with the enzyme differently and thus

contribute to the relative selectivity. As an example the more rigid p-

nitrophenyl ester/amide could have sterical interactions with the enzyme

that the butyl alcohol/amine do not. By choosing matching substrates the

free enzyme and acyl-enzyme will be the same, independent on the

direction of the reaction (hydrolysis contra synthesis). Consequently the

selectivity will be governed by the differences between alcohol and amine

during synthesis since the path leading to the acyl-enzyme is the same.

Likewise, the differences between ester and amide will determine the

selectivity during hydrolysis. In such cases it is anticipated that in a

reaction catalyzed by an enzyme the same contribution is given in the

forward and reverse direction.

C o n c l u d i n g r e m a r k s | 61

6. Concluding remarks

The reason why amidases proficiently can hydrolyze both amides and

esters whereas esterases cannot, although sharing the same catalytic

machinery, has been debated. A newly found hydrogen bond interaction

present in amidases but not esterases has been hypothesized to be an

important factor for the amidase specificity.133 In this thesis that

hydrogen bond interaction was investigated as a potential mean for better

catalysts.

The hydrogen bond stabilization for nitrogen inversion is important for

amidase specificity as demonstrated in this thesis with three different

esterases. An approach using molecular modeling for the design of an

engineered hydrogen bond interaction is introduced that utilizes either

amino acid side chains or water to achieve the stabilization. The

stabilization achieved is non-optimal due to imperfect distances and

angles but gives an estimate on what a hydrogen bond can be worth

towards changed amidase specificity. For the relative reaction specificity

the small free energy differences between amine and alcohol compared to

amide and ester makes it possible with the stabilization of the amine TS

to change reaction preference favoring amide synthesis.

In the design of the engineered variants water played a crucial role in the

catalysis beyond the role of being a solvent or substrate (in hydrolysis)

and should be taken into consideration for the design of the interaction

sought for. A variant with good performance in water could potentially be

ineffective when water is no longer there to assist in the stabilization.

Engineering an enzyme to lock side chains or water in a fixed spatial

arrangement might be associated with an entropic penalty, which reduces

the overall effect gained by the stabilization. A potentially advantageous

approach circumventing the entropic penalty of freezing side chains or

water would be to use a hydrogen bond acceptor from the enzyme

backbone as found in many proteases, which might be found in other

enzyme scaffolds.

62 | C o n c l u d i n g r e m a r k s

A c k n o w l e d g e m e n t s | 63

7. Acknowledgements

Time flies when you are having fun and during my studies, both Master

and PhD, I had some great times and met a lot of interesting people. ‘I

don't know half of you half as well as I should like; and I like less than

half of you half as well as you deserve’.

Some of you do deserve an extra honorable mention:

Mats, thank you for accepting me as diploma worker for my Master

degree project and then as a PhD student. I have valued the free reigns I

have been given in the lab for my projects and that you always have had

time for discussions when needed.

Thank you Kalle for your genuine interest for scientific discussions and

for always asking the right questions. Many things have become clearer

after talking with you. You are an inspiration to all of us.

Per-Olof, thank you for being a great mentor and even better friend. I

have enjoyed every time I gotten the possibility to work with you and I

feel privileged to have you as a friend. Thank you for all the fun times in

the lab and outside (did someone say Moscow? )!

Thank you Pål and Per, for briefly being my co-supervisors and having

an open door policy (when you are here ).

Thank you all past and present members of the Biocatalysis group. To

Cecilia, Magnus, Mohamad, Anders, Maria and Marianne for

being there when I started and (in)directly sharing your knowledge and

know-how. Karim, for being the last link between the old and the new

and giving me the window seat (you were probably too lazy to clean it ).

I’m looking forward to work with you. My old office mates: Henrik, for

being my partner in (not so many) crimes. Thank you for sharing this

PhD journey with me and all the good times along the way! Mattias, I

64 | A c k n o w l e d g e m e n t s

really admire your upright character. Thank you for discovering (the pubs

of) Manchester with me. The newer office additions: Fabian and Lisa

(that already left) and Shan and Federica for being nice and helpful

every day. Last but not least, thank you Fei (our office neighbor) for

being a kind colleague.

Stefan and Maja, the ones that came after me. How will you manage with

a tidy lab and things not getting turned up-side down? Stefan, it has

been a pleasure getting to know you and you have become a friend in the

lab as well as outside. Thank you for the work in our “recent”

collaborations and the careful reading of all my writing. Maja, you are

really something extra. Thank you for all the positive energy that you

bring to work every day. I leave you with a nod of the head

Mathilda, thank you sharing mutual appreciation for “proper” TV and

for contributing to a much nicer working environment with fredagsgodis

and fredagsöl and so forth.

To past and present tree people of Glycoscience (mostly past since I’m

getting “old”). Thank you Ela, Chris, Nuria, Lauren, Fran, Toni,

Sara, Erik, Felicia, Jens, Johan, Lana, Hu, Stefan, Paul, Andrea,

Gustav, Johanna, Ana Catarina, Oliver, Chunlin and Hugo for all

the fredagsöl any day of the week! Oh, and we also had morning coffee on

some occasions

Cortwa, I’m not forgetting about you and how could I? There is just

something special about some people. All I can say is that you are truly

missed and that I hope to see you soon (in your newly renovated house).

Thank you for all the great times!

The visiting scientists from around the world: Shu, Wang, Na Wang

and Doris. Thank you for everything you have taught me, big and small,

about science and everything else.

Thank you to all other co-workers at the divisions of Industrial

Biotechnology and Glycoscience for contributing to a friendly

atmosphere. Special thanks to Nina, Caroline and Ela N for taking care

of and making work easier for the rest of us!

A c k n o w l e d g e m e n t s | 65

Kaj, you have my gratitude for taking care of my computer related issues

and for all the small chats about everything.

The BioPol group is acknowledged for interesting discussions and

inputs. Special thanks to Mats J for being so enthusiastic about all the

results, big and small, and for being my co-supervisor. I would also like to

thank Mauro for teaching me some of the “strange” techniques in

polymer science.

To my new found friends at EnginZyme, Samuel, Robin and Linda,

“Det är här framtiden börjar”.

Tack till släkt och vänner utanför ”bubblan” för att ni finns! Peter och

ITG-gänget (Anne-Lill, Daniel, Jonas, Mikael och Veronica), nu

kanske det blir fler tillfällen att ses oftare?

Ett jättetack till Mamma, Pappa, Björn och Rolf samt Susanne och

Mikael för allt stöd och omtanke! Nu äntligen kanske det är någorlunda

klart vad jag, i alla fall bitvis, har sysslat de senaste åren?

Slutligen skulle jag vilja tacka min familj: Älskade Ann-Sofie och Stella.

Även fast arbetet med denna avhandling har tagit tid, mycket tid, är ni de

viktigaste personerna i mitt liv. Tack för att ni finns där och skänker mig

så otroligt mycket glädje! Nu ser jag fram emot nya äventyr tillsammans!

66 | A c k n o w l e d g e m e n t s

B i b l i o g r a p h y | 67

8. Bibliography

1. Kühne, W. Ueber das Verhalten verschiedener organisirter und sog. ungeformter Fermente. Verhandlungen des Naturhistorisch‐ medicinischen Vereins zu Heidelberg. Neue Folge 1877, 1: 190-193.

2. Gutfreund, H. Wilhelm Friedrich Kühne; An appreciation. FEBS Lett. 1976, 62: E1-E2.

3. Buchner, E. Alkoholische Gärung ohne Hefezellen. Ber. Dtsch. Chem. Ges. 1897, 30: 1110-1113.

4. Kohler, R. The background to Eduard Buchner's discovery of cell-free fermentation. J. Hist. Biol. 1971, 4: 35-61.

5. Sumner, J.B. The isolation and crystallization of the enzyme urease: preliminary paper J. Biol. Chem. 1926, 435-441.

6. Kendrew, J.C., Bodo, G., Dintzis, H.M., Parrish, R.G., Wyckoff, H. and Phillips, D.C. A three-dimensional model of the myoglobin molecule obtained by X-ray analysis. Nature 1958, 181: 662-666.

7. Berg, J.M., Tymoczko, J.L. and Stryer, L. (2006) Biochemistry. (6th Ed., pp. 34-

36). W. H. Freeman and Company. 8. Fromm, H.J. and Hargrove, M.S. (2012) Essentials of biochemistry. (pp. 35-44).

Springer Science+Business Media 9. Stenesh, J. (1998) Biochemistry. (1

st Ed., pp. 58-67). Springer Science+Business

Media 10. Gott III, J.R., Jurić, M., Schlegel, D., Hoyle, F., Vogeley, M., Tegmark, M.,

Bahcall, N. and Brinkmann, J. A map of the universe. Astrophys. J. 2005, 624: 463-484.

11. Conselice, C.J., Wilkinson, A., Duncan, K. and Mortlock, A. The evolution of galaxy number density at z < 8 and its implications. Astrophys. J. 2016, 830: 17 pp.

12. Subramanya, H.S., Roper, D.I., Dauter, Z., Dodson, E.J., Davies, G.J., Wilson, K.S. and Wigley, D.B. Enzymatic ketonization of 2-hydroxymuconate: specificity and mechanism investigated by the crystal structures of two isomerases. Biochemistry 1996, 35: 792-802.

13. Snider, M.J. and Wolfenden, R. The rate of spontaneous decarboxylation of amino acids. J. Am. Chem. Soc. 2000, 122: 11507-11508.

14. Miller, B.G. and Wolfenden, R. Catalytic proficiency: The unusual case of OMP decarboxylase. Annu. Rev. Biochem. 2002, 71: 847-885.

15. Bornscheuer, U.T., Huisman, G.W., Kazlauskas, R.J., Lutz, S., Moore, J.C. and Robins, K. Engineering the third wave of biocatalysis. Nature 2012, 485: 185-194.

16. Bhosale, S.H., Rao, M.B. and Deshpande, V.V. Molecular and industrial aspects of glucose isomerase. Microbiol. Rev. 1996, 60: 280-300.

17. Arja, M.-O. (2007) Cellulases in the Textile Industry. In J. Polaina and A.P. MacCabe (Eds.), Cellulases in the Textile Industry. (pp. 51-63). Springer Science+Business Media.

18. Houde, A., Kademi, A. and Leblanc, D. Lipases and their industrial applications: an overview. Appl. Biochem. Biotechnol. 2004, 118: 155-170.

19. Sheldon, R.A. The E factor: fifteen years on. Green Chem. 2007, 9: 1273-1283.

68 | B i b l i o g r a p h y

20. Zaks, A. and Klibanov, A.M. Enzyme-catalyzed processes in organic solvents. Proc. Natl. Acad. Sci. U.S.A. 1985, 82: 3192-3196.

21. Zaks, A. and Russell, A.J. Enzymes in organic solvents: properties and applications. J. Biotechnol. 1988, 8: 259-269.

22. Chen, K. and Arnold, F.H. Tuning the activity of an enzyme for unusual environments: Sequential random mutagenesis of subtilisin E for catalysis in dimethylformamide. Proc. Natl. Acad. Sci. U.S.A. 1993, 90: 5618-5622.

23. Savile, C.K., Janey, J.M., Mundorff, E.C., Moore, J.C., Tam, S., Jarvis, W.R., Colbeck, J.C., Krebber, A., Fleitz, F.J., Brands, J., Devine, P.N., Huisman, G.W. and Hughes, G.J. Biocatalytic asymmetric synthesis of chiral amines from ketones applied to sitagliptin manufacture. Science 2010, 329: 305-309.

24. Liu, D., Trodler, P., Eiben, S., Koschorreck, K., Muller, M., Pleiss, J., Maurer, S.C., Branneby, C., Schmid, R.D. and Hauer, B. Rational design of Pseudozyma antarctica lipase B yielding a general esterification catalyst. ChemBioChem 2010, 11: 789-795.

25. Arnold, F.H. Engineering enzymes for non-aqueous solvents. Trends Biotechnol. 1990, 8: 244-249.

26. Ness, J.E., Welch, M., Giver, L., Bueno, M., Cherry, J.R., Borchert, T.V., Stemmer, W.P. and Minshull, J. DNA shuffling of subgenomic sequences of subtilisin. Nat. Biotechnol. 1999, 17: 893-896.

27. Eijsink, V.G., Bjork, A., Gaseidnes, S., Sirevag, R., Synstad, B., van den Burg, B. and Vriend, G. Rational engineering of enzyme stability. J. Biotechnol. 2004, 113: 105-120.

28. Carter, P. and Wells, J.A. Dissecting the catalytic triad of a serine protease. Nature 1988, 332: 564-568.

29. Winter, G., Fersht, A.R., Wilkinson, A.J., Zoller, M. and Smith, M. Redesigning enzyme structure by site-directed mutagenesis: tyrosyl tRNA synthetase and ATP binding. Nature 1982, 299: 756-758.

30. Wilkinson, A.J., Fersht, A.R., Blow, D.M. and Winter, G. Site-directed mutagenesis as a probe of enzyme structure and catalysis: tyrosyl-tRNA synthetase cysteine-35 to glycine-35 mutation. Biochemistry 1983, 22: 3581-3586.

31. Cedrone, F., Ménez, A. and Quéméneur, E. Tailoring new enzyme functions by rational redesign. Curr. Opin. Struct. Biol. 2000, 10: 405-410.

32. Chen, R. Enzyme engineering: Rational redesign versus directed evolution. Trends Biotechnol. 2001, 19: 13-14.

33. Nestl, B.M., Nebel, B.A. and Hauer, B. Recent progress in industrial biocatalysis. Curr. Opin. Chem. Biol. 2011, 15: 187-193.

34. Otten, L.G., Hollmann, F. and Arends, I.W.C.E. Enzyme engineering for enantioselectivity: from trial-and-error to rational design? Trends Biotechnol. 2010, 28: 46-54.

35. Sandström, A.G., Wikmark, Y., Engström, K., Nyhlén, J. and Bäckvall, J.E. Combinatorial reshaping of the Candida antarctica lipase A substrate pocket for enantioselectivity using an extremely condensed library. Proc. Natl. Acad. Sci. U.S.A. 2012, 109: 78-83.

36. Ouellette, R.J. and Rawn, J.D. (2014) 3 – Introduction to organic reaction mechanisms. In Organic chemistry. (pp. 75-110). Elsevier.

37. Copeland, R.A. (2000) Enzymes: A practical introduction to structure, mechanism and data analysis. (2

nd Ed., pp. 26-29). Wiley-VCH.

38. Pauling, L. Molecular architecture and biological reactions. Chem. Eng. News 1946, 24: 1375-1377.

39. Copeland, R.A. (2000) Enzymes: a practical introduction to structure, mechanism and data analysis. (2

nd Ed., pp. 151-155). Wiley-VCH.

40. Cannon, W.R. and Benkovic, S.J. Solvation, reorganization energy, and biological catalysis. J. Biol. Chem. 1998, 273: 26257-26260.


41. Ringe, D. and Petsko, G.A. Biochemistry. How enzymes work. Science 2008, 320: 1428-1429.

42. Michaelis, L. and Menten, M.L. Die Kinetik der Invertinwirkung. Biochem. Z. 1913, 49: 333-369.

43. Briggs, G.E. and Haldane, J.B.S. A note on the kinetics of enzyme action. Biochem. J. 1925, 19: 338-339.

44. Johnson, K.A. and Goody, R.S. The original Michaelis constant: translation of the 1913 Michaelis-Menten paper. Biochemistry 2011, 50: 8264-8269.

45. Fersht, A. (1999) Structure and mechanism in protein science: A guide to enzyme catalysis and protein folding. (3

rd Ed., pp. 164-167). W. H Freeman and Company.

46. Wolfenden, R. and Snider, M.J. The depth of chemical time and the power of enzymes as catalysts. Acc. Chem. Res. 2001, 34: 938-945.

47. Bommarius, A.S. Biocatalysis: A status report. Annu. Rev. Chem. Biomol. Eng. 2015, 6: 319-345.

48. Fox, R.J. and Clay, M.D. Catalytic effectiveness, a measure of enzyme proficiency for industrial applications. Trends Biotechnol. 2009, 27: 137-140.

49. Eisenthal, R., Danson, M.J. and Hough, D.W. Catalytic efficiency and kcat/KM: a useful comparator? Trends Biotechnol. 2007, 25: 247-249.

50. Martinelle, M. and Hult, K. Kinetics of acyl transfer reactions in organic media catalysed by Candida antarctica lipase B. Biochim. Biophys. Acta 1995, 1251: 191-197.

51. Cornish-Bowden, A. (1979) Chapter 6 - Two-substrate reactions. In Fundamentals of enzyme kinetics. (pp. 99-129). Butterworth-Heinemann.

52. Eyring, H. The activated complex in chemical reactions. J. Chem. Phys. 1935, 3: 107.

53. Eyring, H. and Stearn, A.E. The application of the theory of absolute reacton rates to proteins. Chem. Rev. 1939, 24: 253-270.

54. Wolfenden, R., Snider, M., Ridgway, C. and Miller, B. The temperature dependence of enzyme rate enhancements. J. Am. Chem. Soc. 1999, 121: 7419-7420.

55. Hedfors, C., Hult, K. and Martinelle, M. Lipase chemoselectivity towards alcohol and thiol acyl acceptors in a transacylation reaction. J. Mol. Catal. B: Enzym. 2010, 66: 120-123.

56. Orrenius, C., Öhrner, N., Rotticci, D., Mattson, A., Hult, K. and Norin, T. Candida antarctica lipase B catalysed kinetic resolutions: Substrate structure requirements for the preparation of enantiomerically enriched secondary alcanols. Tetrahedron: Asymmetry 1995, 6: 1217-1220.

57. Uppenberg, J., Öhrner, N., Norin, M., Hult, K., Kleywegt, G.J., Patkar, S., Waagen, V., Anthonsen, T. and Jones, T.A. Crystallographic and molecular-modeling studies of lipase B from Candida antarctica reveal a stereospecificity pocket for secondary alcohols. Biochemistry 1995, 34: 16838-16851.

58. Anderson, E.M., Larsson, K.M. and Kirk, O. One Biocatalyst–Many Applications: The Use of Candida Antarctica B-Lipase in Organic Synthesis. Biocatal. Biotransformation 2009, 16: 181-204.

59. Ljunger, G., Adlercreutz, P. and Mattiasson, B. Lipase catalyzed acylation of glucose. Biotechnol. Lett. 1994, 16: 1167-1172.

60. Fransson, L. (2008) Enzyme substrate solvent interactions : a case study on serine hydrolases. Stockholm, KTH.

61. Hult, K. and Berglund, P. Enzyme promiscuity: mechanism and applications. Trends Biotechnol. 2007, 25: 231-238.

62. O'Brien, P.J. and Herschlag, D. Catalytic promiscuity and the evolution of new enzymatic activities. Chem. Biol. 1999, 6: R91-R105.

63. Copley, S. Enzymes with extra talents: Moonlighting functions and catalytic promiscuity. Curr. Opin. Chem. Biol. 2003, 7: 265-272.


64. Devamani, T., Rauwerdink, A.M., Lunzer, M., Jones, B.J., Mooney, J.L., Tan, M.A., Zhang, Z.J., Xu, J.H., Dean, A.M. and Kazlauskas, R.J. Catalytic promiscuity of ancestral esterases and hydroxynitrile lyases. J. Am. Chem. Soc. 2016, 138: 1046-1056.

65. Dayhoff, M.O., McLaughlin, P.J., Barker, W.C. and Hunt, L.T. Evolution of sequences within protein superfamilies. Naturwissenschaften 1975, 62: 154-161.

66. Jochens, H., Hesseler, M., Stiba, K., Padhi, S.K., Kazlauskas, R.J. and Bornscheuer, U.T. Protein engineering of α/β-hydrolase fold enzymes. ChemBioChem 2011, 12: 1508-1517.

67. Ollis, D.L., Cheah, E., Cygler, M., Dijkstra, B., Frolow, F., Franken, S.M., Harel, M., Remington, S.J., Silman, I., Schrag, J., Sussman, J.L., Verschueren, K.H.G. and Goldman, A. The α/β hydrolase fold. Protein Eng. 1992, 5: 197-211.

68. Lao, D.-I. and Remington, S.J. Structure of wheat serine carboxypeptidase II at 3.5-angstrom resolution. A new class of serine proteinase. J. Biol. Chem. 1990, 265: 6528-6531.

69. Pries, F., Kingma, J., Pentenga, M., van Pouderoyen, G., Jeronimus-Stratingh, C.M., Bruins, A.P. and Janssen, D.B. Site-directed mutagenesis and oxygen isotope incorporation studies of the nucleophilic aspartate of haloalkane dehalogenase. Biochemistry 1994, 33: 1242-1247.

70. Arand, M., Cronin, A., Adamska, M. and Oesch, F. Epoxide hydrolases: structure, function, mechanism, and assay. Methods Enzymol. 2005, 400: 569-588.

71. Gruber, K., Gartler, G., Krammer, B., Schwab, H. and Kratky, C. Reaction mechanism of hydroxynitrile lyases of the α/β-hydrolase superfamily: The three-dimensional structure of the transient enzyme-substrate complex certifies the crucial role of Lys236. J. Biol. Chem. 2004, 279: 20501-20510.

72. Heikinheimo, P., Goldman, A., Jeffries, C. and Ollis, D.L. Of barn owls and bankers: A lush variety of α/β hydrolases. Structure 1999, 7: R141-R146.

73. Schrag, J.D. and Cygler, M. 1.8 Å refined structure of the lipase from Geotrichum candidum. J. Mol. Biol. 1993, 230: 575-591.

74. Schrag, J.D., Winkler, F.K. and Cygler, M. Pancreatic lipases: evolutionary intermediates in a positional change of catalytic carboxylates? . J. Biol. Chem. 1992, 267: 4300-4303.

75. Rauwerdink, A. and Kazlauskas, R.J. How the same core catalytic machinery catalyzes 17 different reactions: the Ser-His-Asp catalytic triad of α/β-hydrolase fold enzymes. ACS Catal. 2015, 5: 6153-6176.

76. Nedrud, D.M., Lin, H., Lopez, G., Padhi, S.K., Legatt, G.A. and Kazlauskas, R.J. Uncovering divergent evolution of α/β-hydrolases: a surprising residue substitution needed to convert Hevea brasiliensis hydroxynitrile lyase into an esterase. Chem. Sci. 2014, 5: 4265-4277.

77. Schmid, A., Dordick, J.S., Hauer, B., Kiener, A., Wubbolts, M. and Witholt, B. Industrial biocatalysis today and tomorrow. Nature 2001, 409: 258-268.

78. Bornscheuer, U.T. and Kazlauskas, R.J. (2006) Hydrolases in organic synthesis: Regio- and stereoselective biotransformations. (2

nd Ed.). Wiley-VCH.

79. Berg, J.M., Tymoczko, J.L. and Stryer, L. (2006) Biochemistry. (6th Ed., p. 617).

W. H. Freeman and Company. 80. Alvarez, H.M. and Steinbuchel, A. Triacylglycerols in prokaryotic microorganisms.

Appl. Microbiol. Biotechnol. 2002, 60: 367-376. 81. Goto, S., Sugiyama, J. and Iizuka, H. A taxonomic study of antarctic yeasts.

Mycologia 1969, 61: 748-774. 82. Ishii, M. (1987) Positionally non-specific lipase from candida sp.; a method for

producing it and its use. WO8802775. 83. Heldt-Hansen, H.P., Ishii, M., Patkar, S.A., Hansen, T.T. and Eigtved, P. A new

immobilized positional nonspecific lipase for fat modification and ester synthesis. ACS Symp. Ser. 1989, 389: 158-172.


84. Nielsen, T.B., Ishii, M. and Kirk, O. (1999) Lipases A and B from the yeast Candida antarctica. In R. Margesin and F. Schinner (Eds.), Lipases A and B from the yeast Candida antarctica. (pp. 49-61). Springer Science+Business Media.

85. Uppenberg, J., Hansen, M.T., Patkar, S. and Jones, T.A. The sequence, crystal structure determination and refinement of two crystal forms of lipase B from Candida antarctica. Structure 1994, 2: 293-308.

86. Martinelle, M., Holmquist, M. and Hult, K. On the interfacial activation of Candida antarctica lipase A and B as compared with Humicola lanuginosa lipase. Biochim. Biophys. Acta 1995, 1258: 272-276.

87. Reis, P., Holmberg, K., Watzke, H., Leser, M.E. and Miller, R. Lipases at interfaces: A review. Adv. Colloid Interface Sci. 2009, 147-148: 237-250.

88. Takwa, M., Hult, K. and Martinelle, M. Single-step, solvent-free enzymatic route to α,ω-functionalized polypentadecalactone macromonomers. Macromolecules 2008, 41: 5230-5236.

89. Syrén, P.-O., Le Joubioux, F., Ben Henda, Y., Maugard, T., Hult, K. and Graber, M. Proton shuttle mechanism in the transition state of lipase-catalyzed N-acylation of amino alcohols. ChemCatChem 2013, 5: 1842-1853.

90. Frykman, H., Öhrner, N., Norin, T. and Hult, K. S-ethyl thiooctanoate as acyl donor in lipase catalysed resolution of secondary alcohols. Tetrahedron Lett. 1993, 34: 1367-1370.

91. Waagen, V., Hollingsæter, I., Partali, V., Thorstad, O. and Anthonsen, T. Enzymatic resolution of butanoic esters of 1-phenyl, 1-phenylmethyl, 1-[2-phenylethyl] and 1-12-phenoxyethyl] ethers of 3-methoxy-1,2-propanediol. Tetrahedron: Asymmetry 1993, 4: 2265-2274.

92. Reetz, M.T. and Dreisbach, C. Highly efficient lipase-catalyzed kinetic resolution of chiral amines. Chimia 1994, 48: 570.

93. Björkling, F., Godtfredsen, S.E. and Kirk, O. A highly selective enzyme-catalysed esterification of simple glucosides. J. Chem. Soc., Chem. Commun. 1989, 10.1039/c39890000934934-935.

94. Bertinotti, A., Carrea, G., Ottolina, G. and Riva, S. Regioselective esterification of polyhydroxylated steroids by Candida antarctica lipase B. Tetrahedron 1994, 50: 13165-13172.

95. Cao, L., Bornscheuer, U.T. and Schmid, R.D. Lipase-catalyzed solid phase synthesis of sugar esters. Lipid/Fett 1996, 98: 332-335.

96. Arroyo, M. and Sinisterra, J.V. High enantioselective esterification of 2-arylpropionic acids catalyzed by immobilized lipase from Candida antarctica: A mechanistic approach. J. Org. Chem. 1994, 59: 4410-4417.

97. Mathews, I., Soltis, M., Saldajeno, M., Ganshaw, G., Sala, R., Weyler, W., Cervin, M.A., Whited, G. and Bott, R. Structure of a novel enzyme that catalyzes acyl transfer to alcohols in aqueous conditions. Biochemistry 2007, 46: 8969-8979.

98. Upton, C. and Buckley, J.T. A new family of lipolytic enzymes? Trends Biochem. Sci. 1995, 20: 178-179.

99. Mølgaard, A., Kauppinen, S. and Larsen, S. Rhamnogalacturonan acetylesterase elucidates the structure and function of a new family of hydrolases. Structure 2000, 8: 373-383.

100. Akoh, C.C., Lee, G.C., Liaw, Y.C., Huang, T.H. and Shaw, J.F. GDSL family of serine esterases/lipases. Prog. Lipid Res. 2004, 43: 534-552.

101. Arpigny, J.L. and Jaeger, K.E. Bacterial lipolytic enzymes: classification and properties. Biochem. J. 1999, 343: 177-183.

102. Lo, Y.-C., Lin, S.-C., Shaw, J.-F. and Liaw, Y.-C. Crystal structure of Escherichia coli thioesterase I/protease I/lysophospholipase L1: Consensus sequence blocks constitute the catalytic center of SGNH-hydrolases through a conserved hydrogen bond network. J. Mol. Biol. 2003, 330: 539-551.


103. Barnes Jr., E.M. and Wakil, S.J. Studies on the mechanism of fatty acid synthesis. XIX. Preparation and general properties of palmityl thioesterase. J. Biol. Chem. 1968, 243: 2955-2962.

104. Cho, H. and Cronan, J.E. Escherichia coli thioesterase I, molecular cloning and sequencing of the structural gene and identification as a periplasmic enzyme. J. Biol. Chem. 1993, 268: 9238-9245.

105. Chang, R.C., Chien, J.C. and Shaw, J.F. Vibrio mimicus arylesterase has thioesterase and chymotrypsin-like activity. Biochem. Biophys. Res. Commun. 1995, 213: 475-483.

106. Shaw, J.F., Chang, R.C., Chuang, K.H., Yen, Y.T., Wang, Y.J. and Wang, F.G. Nucleotide sequence of a novel arylesterase gene from Vibro mimicus and characterization of the enzyme expressed in Escherichia coli. Biochem. J. 1994, 298: 675-680.

107. Lee, Y.L., Chen, J.C. and Shaw, J.F. The thioesterase I of Escherichia coli has arylesterase activity and shows stereospecificity for protease substrates. Biochem. Biophys. Res. Commun. 1997, 231: 452-456.

108. Kauppinen, S., Christgau, S., Kofod, L.V., Halkier, T., Dörreich, K. and Dalbøge, H. Molecular cloning and characterization of a rhamnogalacturonan acetylesterase from Aspergillus aculeatus. Synergism between rhamnogalacturonan degrading enzymes. J. Biol. Chem. 1995, 270: 27172-27178.

109. Lescic Asler, I., Ivic, N., Kovacic, F., Schell, S., Knorr, J., Krauss, U., Wilhelm, S., Kojic-Prodic, B. and Jaeger, K.E. Probing enzyme promiscuity of SGNH hydrolases. ChemBioChem 2010, 11: 2158-2167.

110. Karasawa, K., Yokoyama, K., Setaka, M. and Nojima, S. The Escherichia coli pldC gene encoding lysophospholipase L1 is identical to the apeA and tesA genes encoding protease I and thioesterase I, respectively. J. Biol. Chem. 1999, 126: 445-448.

111. Roustan, J.L., Chu, A.R., Moulin, G. and Bigey, F. A novel lipase/acyltransferase from the yeast Candida albicans: expression and characterisation of the recombinant enzyme. Appl. Microbiol. Biotechnol. 2005, 68: 203-212.

112. Neang, P.M., Subileau, M., Perrier, V. and Dubreucq, E. Peculiar features of four enzymes of the CaLA superfamily in aqueous media: Differences in substrate specificities and abilities to catalyze alcoholysis. J. Mol. Catal. B: Enzym. 2013, 94: 36-46.

113. Muller, J., Sowa, M.A., Fredrich, B., Brundiek, H. and Bornscheuer, U.T. Enhancing the acyltransferase activity of Candida antarctica Lipase A by rational design. ChemBioChem 2015, 16: 1791-1796.

114. Jan, A.H., Subileau, M., Deyrieux, C., Perrier, V. and Dubreucq, E. Elucidation of a key position for acyltransfer activity in Candida parapsilosis lipase/acyltransferase (CpLIP2) and in Pseudozyma antarctica lipase A (CAL-A) by rational design. Biochim. Biophys. Acta 2016, 1864: 187-194.

115. Dinu, C.Z., Zhu, G., Bale, S.S., Anand, G., Reeder, P.J., Sanford, K., Whited, G., Kane, R.S. and Dordick, J.S. Enzyme-based nanoscale composites for use as active decontamination surfaces. Adv. Funct. Mater. 2010, 20: 392-398.

116. Grover, N., Douaisi, M.P., Borkar, I.V., Lee, L., Dinu, C.Z., Kane, R.S. and Dordick, J.S. Perhydrolase-nanotube paint composites with sporicidal and antiviral activity. Appl. Microbiol. Biotechnol. 2013, 97: 8813-8821.

117. Wiermans, L., Hofzumahaus, S., Schotten, C., Weigand, L., Schallmey, M., Schallmey, A. and Domínguez de María, P. Transesterifications and peracid-assisted oxidations in aqueous media catalyzed by Mycobacterium smegmatis acyl transferase. ChemCatChem 2013, 5: 3719-3724.

118. Drożdż, A., Hanefeld, U., Szymańska, K., Jarzębski, A. and Chrobok, A. A robust chemo-enzymatic lactone synthesis using acyltransferase from Mycobacterium smegmatis. Catal. Commun. 2016, 81: 37-40.


119. Szymańska, K., Odrozek, K., Zniszczoł, A., Torrelo, G., Resch, V., Hanefeld, U. and Jarzębski, A.B. MsAcT in siliceous monolithic microreactors enables quantitative ester synthesis in water. Catal. Sci. Technol. 2016, 6: 4882-4888.

120. Chapus, C., Semeriva, M., Bovier-Lapierre, C. and Desnuelle, P. Mechanism of pancreatic lipase action. 1. Interfacial activation of pancreatic lipase. Biochemistry 1976, 15: 4980-4987.

121. Brady, L., Brzozowski, A.M., Derewenda, Z.S., Dodson, E., Dodson, G., Tolley, S., Turkenburg, J.P., Christiansen, L., Huge-Jensen, B., Norskov, L. and et al. A serine protease triad forms the catalytic centre of a triacylglycerol lipase. Nature 1990, 343: 767-770.

122. Wright, C.S. Comparison of the active site stereochemistry and substrate conformation in α-chymotrypsin and subtilisin BPN′. J. Mol. Biol. 1972, 67: 151-163.

123. Kraut, J. Serine proteases: Structure and mechanism of catalysis. Annu. Rev. Biochem. 1977, 46: 331-358.

124. Liao, D.I., Breddam, K., Sweet, R.M., Bullock, T. and Remington, S.J. Refined atomic model of wheat serine carboxypeptidase II at 2.2-Å resolution. Biochemistry 1992, 31: 9796-9812.

125. Wells, J.A., Cunningham, B.C., Graycar, T.P. and Estell, D.A. Importance of hydrogen-bond formation in stabilizing the transition state of subtilisin. Philos. Trans. A Math. Phys. Eng. Sci. 1986, 317: 415-423.

126. Menard, R., Carriere, J., Laflamme, P., Plouffe, C., Khouri, H.E., Vernet, T., Tessier, D.C., Thomas, D.Y. and Storer, A.C. Contribution of the glutamine 19 side chain to transition-state stabilization in the oxyanion hole of papain. Biochemistry 1991, 30: 8924-8928.

127. Nicolas, A., Egmond, M., Verrips, C.T., de Vlieg, J., Longhi, S., Cambillau, C. and Martinez, C. Contribution of cutinase serine 42 side chain to the stabilization of the oxyanion transition state. Biochemistry 1996, 35: 398-410.

128. Magnusson, A., Hult, K. and Holmquist, M. Creation of an enantioselective hydrolase by engineered substrate-assisted catalysis. J. Am. Chem. Soc. 2001, 123: 4354-4355.

129. Carey, J.S., Laffan, D., Thomson, C. and Williams, M.T. Analysis of the reactions used for the preparation of drug candidate molecules. Org. Biomol. Chem. 2006, 4: 2337-2347.

130. Constable, D.J.C., Dunn, P.J., Hayler, J.D., Humphrey, G.R., Leazer, J.J.L., Linderman, R.J., Lorenz, K., Manley, J., Pearlman, B.A., Wells, A., Zaks, A. and Zhang, T.Y. Key green chemistry research areas—a perspective from pharmaceutical manufacturers. Green Chem. 2007, 9: 411-420.

131. Chen, P., Tsuge, H., Almassy, R.J., Gribskov, C.L., Katoh, S., Vanderpool, D.L., Margosiak, S.A., Pinko, C., Matthews, D.A. and Kan, C.-C. Structure of the human cytomegalovirus protease catalytic domain reveals a novel serine protease fold and catalytic triad. Cell 1996, 86: 835-843.

132. Brik, A. and Wong, C.-H. HIV-1 protease: mechanism and drug discovery. Org. Biomol. Chem. 2003, 1: 5-14.

133. Syrén, P.-O. and Hult, K. Amidases have a hydrogen bond that facilitates nitrogen inversion, but esterases have not. ChemCatChem 2011, 3: 853-860.

134. Clayden, J., Greeves, N., Warren, S. and Wothers, P. (2001) Organic chemistry. (1

st Ed., p. 287). Oxford University Press.

135. Kahne, D. and Still, W.C. Hydrolysis of a peptide bond in neutral water. J. Am. Chem. Soc. 1988, 110: 7529-7534.

136. Radzicka, A. and Wolfenden, R. Rates of uncatalyzed peptide bond hydrolysis in neutral solution and the transition state affinities of proteases. J. Am. Chem. Soc. 1996, 118: 6105-6109.

137. Smith, R.M. and Hansen, D.E. The pH-rate profile for the hydrolysis of a peptide bond. J. Am. Chem. Soc. 1998, 120: 8910-8913.


138. Bryant, R.A.R. and Hansen, D.E. Direct measurement of the uncatalyzed rate of hydrolysis of a peptide bond. J. Am. Chem. Soc. 1996, 118: 5498-5499.

139. Bender, M.L., Schonbaum, G.R. and Zerner, B. Spectrophotometric investigations of the mechanism of α-chymotrypsin-catalyzed hydrolyses. Detection of the acyl-enzyme intermediate. J. Am. Chem. Soc. 1962, 84: 2540-2550.

140. Zerner, B., Bond, R.P.M. and Bender, M.L. Kinetic evidence for the formation of acyl-enzyme intermediates in the α-chymotrypsin-catalyzed hydrolyses of specific substrates. J. Am. Chem. Soc. 1964, 86: 3674-3679.

141. Abrahmsen, L., Tom, J., Burnier, J., Butcher, K.A., Kossiakoff, A. and Wells, J.A. Engineering subtilisin and its substrates for efficient ligation of peptide bonds in aqueous solution. Biochemistry 1991, 30: 4151-4159.

142. Hedstrom, L., Perona, J.J. and Rutter, W.J. Converting trypsin to chymotrypsin: Residue 172 is a substrate specificity determinant. Biochemistry 1994, 33: 8757-8763.

143. Patricelli, M.P. and Cravatt, B.F. Fatty acid amide hydrolase competitively degrades bioactive amides and esters through a nonconventional catalytic mechanism. Biochemistry 1999, 38: 14125-14130.

144. Wang, Q., Yang, G., Liu, Y. and Feng, Y. Discrimination of esterase and peptidase activities of acylaminoacyl peptidase from hyperthermophilic Aeropyrum pernix K1 by a single mutation. J. Biol. Chem. 2006, 281: 18618-18625.

145. Ekici, O.D., Paetzel, M. and Dalbey, R.E. Unconventional serine proteases: variations on the catalytic Ser/His/Asp triad configuration. Protein Sci. 2008, 17: 2023-2037.

146. Phillips, M.A. and Fletterick, R.J. Proteases. Curr. Opin. Struct. Biol. 1992, 2: 713-720.

147. Syrén, P.O. The solution of nitrogen inversion in amidases. FEBS J. 2013, 280: 3069-3083.

148. Deslongchamps, P. Stereoelectronic control in the cleavage of tetrahedral intermediates in the hydrolysis of esters and amides. Tetrahedron 1975, 31: 2463-2490.

149. Gorenstein, D.G. Stereoelectronic effects in biomolecules. Chem. Rev. 1987, 87: 1047-1077.

150. Burwell, R.L. and Pearson, R.G. The principle of microscopic reversibility. J. Phys. Chem. 1966, 70: 300-302.

151. Bizzozero, S.A. and Dutler, H. Stereochemical aspects of peptide hydrolysis catalyzed by serine proteases of the chymotrypsin type. Bioorg. Chem. 1981, 10: 46-62.

152. Belostotskii, A.M., Aped, P. and Hassner, A. MM3 force field as a tool in mechanistic studies of nitrogen inversion processes for alkylamines. Theochem. 1997, 398-399: 427-434.

153. Stewart, W.E. and Siddall III, T.H. Nuclear magnetic resonance studies of amides. Chem. Rev. 1970, 70: 517-551.

154. Cho, J.H., Kim, D.H., Kim, D.-H., Lee, K.J. and Choi, K.Y. The role of Tyr248 probed by mutant bovine carboxypeptidase A: Insight into the catalytic mechanism of carboxypeptidase A. Biochemistry 2001, 40: 10197-10203.

155. Cammenberg, M., Hult, K. and Park, S. Molecular basis for the enhanced lipase-catalyzed N-acylation of 1-phenylethanamine with methoxyacetate. ChemBioChem 2006, 7: 1745-1749.

156. Balkenhohl, F., Ditrich, K., Hauer, B. and Ladner, W. Optisch aktive Amine durch Lipase-katalysierte Methoxyacetylierung. J. prakt. Chem.-Chem. Ztg. 1997, 339: 381-384.

157. Fersht, A. (1999) Structure and mechanism in protein science: A guide to enzyme catalysis and protein folding. (3

rd Ed., p. 331). W. H Freeman and Company.


158. Dunitz, J.D. The entropic cost of bound water in crystals and biomolecules. Science 1994, 264: 670.

159. Young, T., Abel, R., Kim, B., Berne, B.J. and Friesner, R.A. Motifs for molecular recognition exploiting hydrophobic enclosure in protein-ligand binding. Proc. Natl. Acad. Sci. U.S.A. 2007, 104: 808-813.

160. Gardner Swain, C. and Bader, R.F.W. The nature of the structure difference between light and heavy water and the origin of the solvent isotope effect—I. Tetrahedron 1960, 10: 182-199.

161. Kourist, R., Bartsch, S., Fransson, L., Hult, K. and Bornscheuer, U.T. Understanding promiscuous amidase activity of an esterase from Bacillus subtilis. ChemBioChem 2008, 9: 67-69.

162. Abel, R., Young, T., Farid, R., Berne, B.J. and Friesner, R.A. Role of the active-site solvent in the thermodynamics of factor Xa ligand binding. J. Am. Chem. Soc. 2008, 130: 2817-2831.

163. Breiten, B., Lockett, M.R., Sherman, W., Fujita, S., Al-Sayah, M., Lange, H., Bowers, C.M., Heroux, A., Krilov, G. and Whitesides, G.M. Water networks contribute to enthalpy/entropy compensation in protein-ligand binding. J. Am. Chem. Soc. 2013, 135: 15579-15584.

164. Matsuoka, S., Sugiyama, S., Matsuoka, D., Hirose, M., Lethu, S., Ano, H., Hara, T., Ichihara, O., Kimura, S.R., Murakami, S., Ishida, H., Mizohata, E., Inoue, T. and Murata, M. Water-mediated recognition of simple alkyl chains by heart-type fatty-acid-binding protein. Angew. Chem. Int. Ed. 2015, 54: 1508-1511.

165. Syrén, P.O., Hammer, S.C., Claasen, B. and Hauer, B. Entropy is key to the formation of pentacyclic terpenoids by enzyme-catalyzed polycyclization. Angew. Chem. Int. Ed. 2014, 53: 4845-4849.

166. Leonard, V., Fransson, L., Lamare, S., Hult, K. and Graber, M. A water molecule in the stereospecificity pocket of Candida antarctica lipase B enhances enantioselectivity towards pentan-2-ol. ChemBioChem 2007, 8: 662-667.

167. Nakagawa, Y., Hasegawa, A., Hiratake, J. and Sakata, K. Engineering of Pseudomonas aeruginosa lipase by directed evolution for enhanced amidase activity: mechanistic implication for amide hydrolysis by serine hydrolases. Protein Eng. Des. Sel. 2007, 20: 339-346.

168. Kürten, C., Carlberg, B. and Syrén, P.-O. Mechanism-guided discovery of an esterase scaffold with promiscuous amidase activity. Catalysts 2016, 6: 90.

169. Veld, M.A.J., Hult, K., Palmans, A.R.A. and Meijer, E.W. Fast DKR of amines using isopropyl 2-methoxyacetate as acyl donor. Eur. J. Org. Chem. 2007, 2007: 5416-5421.

170. Kim, Y., Park, J. and Kim, M.-J. Dynamic kinetic resolution of amines and amino acids by enzyme-metal cocatalysis. ChemCatChem 2011, 3: 271-277.

171. Lundberg, H., Tinnis, F., Selander, N. and Adolfsson, H. Catalytic amide formation from non-activated carboxylic acids and amines. Chem. Soc. Rev. 2014, 43: 2714-2742.

172. Tang, S.L.Y., Smith, R.L. and Poliakoff, M. Principles of green chemistry: PRODUCTIVELY. Green Chem. 2005, 7: 761-762.

173. Dordick, J.S. Enzymatic catalysis in monophasic organic solvents. Enzyme Microb. Technol. 1989, 11: 194-211.

174. Ricca, E., Brucher, B. and Schrittwieser, J.H. Multi-enzymatic cascade reactions: Overview and perspectives. Adv. Synth. Catal. 2011, 353: 2239-2262.

175. Guo, F. and Berglund, P. Transaminase biocatalysis: optimization and application. Green Chem. 2016, 10.1039/c6gc02328b

176. Anderson, M., Afewerki, S., Berglund, P. and Córdova, A. Total Synthesis of Capsaicin Analogues from Lignin-Derived Compounds by Combined Heterogeneous Metal, Organocatalytic and Enzymatic Cascades in One Pot. Adv. Synth. Catal. 2014, 356: 2113-2118.


177. Larsen, M.W., Zielinska, D.F., Martinelle, M., Hidalgo, A., Jensen, L.J., Bornscheuer, U.T. and Hult, K. Suppression of water as a nucleophile in Candida antarctica lipase B catalysis. ChemBioChem 2010, 11: 796-801.

178. Davies, M.T. A universal buffer solution for use in ultra-violet spectrophotometry. Analyst 1959, 84: 248.

179. Jung, S., Kim, J. and Park, S. Rational design for enhancing promiscuous activity of Candida antarctica lipase B: A clue for the molecular basis of dissimilar activities between lipase and serine-protease. RSC Adv. 2013, 3: 2590.

180. Hackenschmidt, S., Moldenhauer, E.J., Behrens, G.A., Gand, M., Pavlidis, I.V. and Bornscheuer, U.T. Enhancement of promiscuous amidase activity of a Bacillus subtilis esterase by formation of a π-π network. ChemCatChem 2014, 6: 1015-1020.

181. Suplatov, D.A., Besenmatter, W., Švedas, V.K. and Svendsen, A. Bioinformatic analysis of alpha/beta-hydrolase fold enzymes reveals subfamily-specific positions responsible for discrimination of amidase and lipase activities. Protein Eng. Des. Sel. 2012, 25: 689-697.

182. Biundo, A., Ribitsch, D., Syren, P.-O. and Guebitz, G.M. Increasing amide acceptance on a polyester-hydrolyzing enzyme. N. Biotechnol. 2016, 33: S105.

183. Hediger, M.R., De Vico, L., Rannes, J.B., Jackel, C., Besenmatter, W., Svendsen, A. and Jensen, J.H. In silico screening of 393 mutants facilitates enzyme engineering of amidase activity in CalB. PeerJ 2013, 1:e145: DOI 10.7717/peerj.7145.

Rational engineering of esterases for improved amidase...

Documents

Transcript of Rational engineering of esterases for improved amidase...