KTH Biotechnology
Enzyme selectivity as a tool in analytical chemistry
Licenciate thesis by
Anders Hamberg
Department of Biochemistry School of Biotechnology
Royal Institute of Technology (KTH)
Stockholm 2007
©Anders Hamberg School of Biotechnology Royal Institute of Technology (KTH) AlbaNova University Centre 106 91 Stockholm ISBN 978-91-7178-674-6 TRITA-BIO-Report 2007:5 ISSN 1654-2312
ii
Abstract Enzymes are useful tools as specific analytical reagents. Two different analysis methods were
developed for use in the separate fields of protein science and organic synthesis. Both
methods rely on the substrate specificity of enzymes. Enzyme catalysis and substrate
specificity is described and put in context with each of the two developed methods.
In paper I a method for C-terminal peptide sequencing was developed based on
conventional Carboxypeptidase Y digestion combined with matrix assisted laser
desorption/ionization mass spectrometry. An alternative nucleophile was used to obtain a
stable peptide ladder and improve sequence coverage.
In paper II and III, three different enzymes were used for rapid analysis of enantiomeric
excess and conversion of O-acylated cyanohydrins synthesized by a defined protocol. Horse
liver alcohol dehydrogenase, Candida antarctica lipase B and pig liver esterase were
sequentially added to a solution containing the O-acylated cyanohydrin. Each enzyme caused
a drop in absorbance from oxidation of NADH to NAD+. The conversion and enantiomeric
excess of the sample could be calculated from the relative differences in absorbance.
iii
Sammanfattning Enzymer kan med fördel användas som specifika reagens inom analytisk kemi. Två olika
analytiskkemiska metoder har utvecklats inom skilda användningsområden: proteinvetenskap
och organkemisk syntes. I båda metoder spelar enzymers substratspecificitet en avgörande
roll. Enzymkatalys och substratspecificitet sätts beskrivs i samband med de nyutvecklade
metoderna.
I artikel I beskrivs en metod för C-terminalsekvensering av peptider genom konventionell
trunkering med hjälp av karboxypeptidas Y. Fragmenterade peptider analyserades med matrix
assisted laser desorption/ionization-masspektrometri. För att skapa en stabil peptidstege och
därmed åstadkomma en förbättrad sekvenstäckning så tillsattes en alternativ nukleofil till
trunkeringslösningen.
I artikel II och III beskrivs en metod för snabbanalys av enantiomert överskott och
omsättning hos O-acylerade cyanohydriner, vilka syntetiserats genom ett definierat protokoll.
Tre olika enzymer, alkoholdehydrogenas från hästlever, lipas B från Candida antarctica och
grisleveresteras sattes i nämnd ordning en lösning innehållande den O-acylerade
cyanohydrinen. Varje enzym gav upphov till en oxidering av NADH till NAD+, vilket
resulterade i stegvis minskad absorbans. Provets omsättning och enantiomert överskott kunde
beräknas från de relativa absorbansskillnaderna.
iv
List of Publications
I Hamberg A., M. Kempka, J. Sjödal, J. Roerade, K. Hult, 2006, C-terminal
ladder sequencing of peptides using an alternative nucleophile in
carboxypeptidase Y digests, Analytical Biochemistry 357, 167-172.
II Hamberg A., S. Lundgren, M. Penhoat, C. Moberg, K. Hult, 2006, High-
throughput method for enantiomeric excess determination of O-acetylated
cyanohydrins, Journal of the American Chemical Society, 128, 2234-2235.
III A. Hamberg, S. Lundgren, E. Wingstrand, C. Moberg, K. Hult, High
Throughput Synthesis and Analysis of Acylated Cyanohydrins, in press 2007
(DOI: 10.1002/chem..200601638), Chemistry: a European Journal.
v
vi
Table of contents 1 Introduction ............................................................................................................................. 1
1.1 Biocatalysis .................................................................................................................. 1
1.2 Substrate specificity ..................................................................................................... 2
2 Amino acid sequence determination ....................................................................................... 5
2.1 Peptide sequencing....................................................................................................... 5
2.2 Carboxypeptidase mediated sequencing and competing nucleophiles ........................ 7
2.3 Substrate specificity ..................................................................................................... 8
2.4 Identification frequency ............................................................................................... 9
2.5 Results ........................................................................................................................ 10
3 Determination of Enantiomeric excess ................................................................................. 13
3.1 Screening for enantioselectivity ................................................................................. 13
3.2 Enzymatic determination of enantiomeric excess...................................................... 15
3.3 Enzymatic analysis of O-acylated cyanohydrins ....................................................... 15
3.4 Substrate specificity ................................................................................................... 17
3.5 Results and discussion................................................................................................ 18
4 Conclusions ........................................................................................................................... 23
5. List of Abbreviations............................................................................................................ 25
6. Acknowledgements .............................................................................................................. 27
7. References ............................................................................................................................ 29
vii
viii
Anders Hamberg
1 Introduction Enzymes are proteins developed in nature for catalyzing reactions required for survival of
different organisms. The high effectiveness and specificity of enzymes have made them
suitable for use in several different areas. In analytical chemistry, the use of enzymes is
extensive. Enzymatic methods in analytical chemistry have been described in several books
and reviews [1-3].
1.1 Biocatalysis A catalyst is generally defined as a substance which speeds up a chemical reaction without
being consumed in the process. In turn, the definition of an enzyme is a biological catalyst.
An example commonly used to illustrate the catalytic power of enzymes is the
decarboxylation of orotidine 5’-phosphate [4]. Spontaneous decarboxylation of the substance
occurs at a rate giving a half-time of 78 million years. When the reaction is catalyzed by the
enzyme orotidine 5’-phosphate decarboxylase from yeast the rate is enhanced by a factor of
1.4×1017. Catalysis is achieved by lowering the energy required for a reaction to take place,
the activation energy. Enzymes reduce the activation energy by stabilizing the transition-state
after binding a substrate (figure 1) [5].
1
Introduction
G
Reaction Coordinate
E + S
ES
ES#
E + P
E + S ES ES# E + P Figure 1. The energy profile diagram for an enzyme catalyzed reaction. The substrate and enzyme complex (ES) has a lower energy than the free components (E and S). The complex then passes a transition-state (ES#) before releasing the product (P).
The efficiency of catalysis for an enzyme catalyzed reaction is determined by catalytic
constants kcat and KM. kcat is a turnover number and proportional to the maximum reaction
rate, Vmax, for a specific enzyme concentration. Vmax is obtained at substrate saturation of the
enzyme. KM is the Michaelis-Menten constant, which is defined as the substrate concentration
needed to obtain ½ Vmax. KM is often described as a measure of the stability of the enzyme-
substrate complex (ES, figure 1), which is true for single substrate kinetics with very low kcat.
For the reaction to take place, the ES complex has to pass transition-state (ES#, figure 1). The
rate for this step is determined by kcat.
1.2 Substrate specificity A single enzyme may catalyze conversion of various substrates at different rates. For
example, isomerisation of glucose into fructose can be catalyzed by the enzyme xylose
isomerase. However, the natural reaction for xylose isomerase is isomerisation of D-xylose
into D-xylulose [6]. Still, the activity towards glucose is sufficient for industrial production of
high-fructose corn syrup, which is one of the largest industrial processes involving enzymes
[7]. The efficiency of an enzyme catalyzed reaction is often described as substrate specificity,
defined by the specificity constant, kcat/KM, for an enzyme towards a substrate. The energy
required for a substrate to reach transition state (from E + S to ES#, figure 1) can be calculated
by equation 1. Hence, the substrate specificity for an enzyme towards a substrate can be
2
Anders Hamberg
related to free energy. The case with orotidine 5’-phosphate decarboxylase, mentioned in
section 1.1, corresponds to a decrease in activation energy by 23 kcal/mol.
hTkRT
Kk
RTG B
M
cat lnln# +−=Δ (Equation 1)
Many enzymes catalyze reactions with a considerable enantioselectivity, which plays an
important part in nature as well as in modern chemistry. Enantioselectivity is derived from
differences in substrate specificity towards the enantiomers of a certain substrate. This means
that one enantiomer reacts faster than the other when a racemic mixture is subjected to an
enantioselective enzyme. The enantiomeric ratio, E, is a measure of the enantioselectivity for
an enzyme towards a substrate. E is defined by equation 2 where A denotes the fast reacting
enantiomer and B denotes the slow reacting enantiomer. At racemic conditions, E is equal to
the rate ratio of A over B (E = vA/vB).
BM
cat
AM
cat
Kk
Kk
E⎟⎠⎞⎜
⎝⎛
⎟⎠⎞⎜
⎝⎛
= (Equation 2)
Two different enzyme mediated analysis methods are described in the following chapters of
this thesis. Substrate specificity for the involved enzymes plays an important part in both
cases. In chapter 2 (based on paper I), a peptide sequencing method was developed.
Differences in substrate specificity towards various substrates posed a problem in this case,
which was decreased by adding a competing nucleophile. Chapter 3 describes the
development of a high-throughput screening method for library derived cyanohydrins (based
on paper II and III). In this case differences in substrate specificity, namely high
enantioselectivity, were needed for determination enantiomeric excess.
3
4
Anders Hamberg
2 Amino acid sequence determination Protein sciences include research fields of great interest. Proteomics can provide basic
knowledge for understanding of living organisms. Synthesis of novel proteins is attractive in
drug development. An important part in such research is amino acid determination of
expressed proteins and peptides [8-12]. Amino acid sequence determination of a full length
protein was first accomplished in 1953, revealing the sequence of bovine hormone insuline
[13]. Today, most sequencing is performed on peptides or protein fragments afforded by site
specific proteases as peptides and protein fragments are easier to handle than full length
proteins. Tools for amino acid sequence determination of peptides are therefore valuable in
both proteomics and protein synthesis.
2.1 Peptide sequencing Modern amino acid sequence determination of peptides or proteins often involves ladder
sequencing in mass spectrometry (MS) [8-10]. Mass analysis of a fragmented peptide allows
sequence determination from the mass differences between peptide fragments detected as
peaks in mass spectra with specific mass to charge ratios (m/z) as illustrated in figure 2. The
fragmentation can be accomplished by sequential truncations or by random peptide cleavage;
examples of both approaches are presented in the following text. Due to limitations in mass
accuracy when using MS for large macromolecules, ladder sequencing is mainly applicable to
sequencing peptides and small proteins [8]. If a large protein is to be sequenced, it has to be
digested into smaller fragments prior to sequencing.
The most commonly used method for ladder sequencing is tandem-mass spectrometry
(MS/MS) [8-10]. Peptide sequencing by MS/MS can be divided in two steps; the first step
involves isolation of a peptide of a certain mass, which is then fragmented in the second step
[14-16]. The sequence obtained in MS/MS can be read from both the N- and C-terminal of the
peptide, which is mostly an advantage. On the other hand, the double set of ladders displayed
in a mass spectrum may give peptide fragments with overlapping masses [17]. Furthermore,
depending on the fragmentation technique, all peptide bonds do not have the same tendency
to dissociate in MS/MS [17, 18]. Incomplete fragmentation combined with the risk of
obtaining peptide fragments with overlapping masses often lead to incomplete sequence
readout. Complementary methods are therefore often used to obtain sufficient sequence
coverage.
5
Amino acid sequence determination
A-B – C – D – E – F
A-B – C – D – E
A-B – C – D
A-B – C
m/z
A – B – C – D – E – F
A – B – C – D – E F
A – B – C – D E – F
A – B – C D – E – F
A – B C – D – E – F
A B – C – D – E – F
m/z
A B
Figure 2. Ladder sequencing of a peptide with the sequence ABCDEF. Peptide fragments are displayed as peaks in a mass spectrum. The amino acid sequence can be elucidated from mass differences between peaks in the mass spectrum. In A the ladder is created from truncation of the peptide from the C-terminus. The colours match C-terminal amino acids of the peptide fragment with the peaks in the corresponding mass spectrum. In B two ladders, C-terminal and the N-terminal, are created by random cleavage of the peptide bonds. The colour of the peaks in the mass spectrum is correlated to the colour of the amino acid adjacent to the cleavage site. Both the C- and N-terminal amino acids in the starting peptide have the same colour seen as a single peak with the highest m/z in the mass spectrum.
A relatively new alternative to MS/MS-sequencing is microwave assisted digestion of
proteins followed by mass analysis [19]. Fragmentation is accomplished by subjecting the
peptide to acid hydrolysis assisted by microwave irradiation prior to analysis. The amino acid
sequence can then be determined by simple MS. The microwave based methods have the
same advantage as MS/MS-sequencing concerning the ability to determine the amino acid
sequence from both ends of the peptide. This also results in the same risk as in MS/MS-
sequencing to obtain overlapping masses.
6
Anders Hamberg
Selective N-terminal sequencing of peptides and proteins is routinely performed by Edman
degradation [20]. Peptide degradation is carried out in cycles using phenyl isothiocyanate for
selective cleavage of the N-terminal amino acid, which can then be analyzed by
chromatography or capillary electrophoresis. As analysis is carried out on the amino acids one
by one, Edman degradation can be applied to full length proteins as well as peptides. Over the
decades, Edman degradation has matured into a reliable and automated method for peptide-
and protein sequencing. However, Edman degradation can be regarded as a laborious and
slow technique exhibiting poor sensitivity. Nonetheless, it is often used as a complement to
MS/MS-sequencing [10]. A protocol for peptide ladder sequencing resembling that of Edman
degradation is also commonly used as a complement to MS/MS-sequencing [21, 22]. Phenyl
isothiocyanate is used together with a terminating agent to prepare a peptide ladder which can
be analyzed in MS. As in Edman degradation, the amino acid sequence is determined from
the N-terminus.
If the amino acid sequence information is desired from the C-terminus, other methods have
to be employed. C-terminal specific degradation of peptides carried out in cycles can be
accomplished by alkylation chemistry [23, 24]. However, low yields in the degradation
reactions often make this method insufficient for sequencing, especially when peptide
samples are limited. Another C-terminal specific alternative is carboxypeptidase mediated
peptide ladder sequencing [25-31]. Mass analysis of a carboxypeptidase digestion of a peptide
can provide information about the C-terminal amino acid sequence. Carboxypeptidase
mediated amino acid determination is described in detail in the following section (2.2).
2.2 Carboxypeptidase mediated sequencing and competing nucleophiles Carboxypeptidases are proteolytic enzymes with selectivity towards the C-terminal amino
acid [32], and are frequently used for C-terminal specific peptide sequencing [25-31].
Aliquots are taken from a carboxypeptidase digest over a period of time. C-terminally
truncated peptide fragments are derived as digestion proceeds, and a peptide ladder is
generated. The C-terminal amino acid sequence can thereby be determined by mass
spectrometry. However, variations in substrate specificity combined with the generation of
numerous peptide fragments may cause insufficient sequence readout (explained in detail in
section 2.3).
A commonly used enzyme for amino acid sequence determination of peptides is
Carboxypeptidase Y (CPY) [25, 28-31]. It has been used both as a single enzyme and together
with another carboxypeptidase exhibiting complementary substrate specificity. CPY have also
7
Amino acid sequence determination
been studied for the use as a catalyst in transpeptidation reactions to make alterations at the C-
terminus of peptides [33-36]. In this case, an amine is used as an alternative nucleophile
which competes with water in a CPY digest. The result is that a fraction of the digested
peptides undergoes aminolysis while the rest of the peptides are hydrolysed.
An alternative nucleophile could be introduced to compete with water in CPY mediated
peptide sequencing (figure 1, paper I). The altered end group could slow down further
degradation and thereby stabilize the peptide ladder. The stabilized peptide fragments
obtained from aminolysis should give an increased number of peptide fragments of different
lengths. Another effect of the presence of a competing nucleophile is that amine-capped
peptide fragments can function as a buffer to refill the amount of peptide fragments which are
rapidly digested by the enzyme (figure 3, paper I). Altogether these effects may contribute to
increased sequence readout. This method would resemble the DNA sequencing method
developed by Sanger, where ddNTP competes with ordinary NTP in PCR (polymerase chain
reaction) [13]. The random incorporation of ddNTP stops further elongation and DNA
fragments of different lengths are created.
2.3 Substrate specificity Peptides bind to CPY by interactions to six different subsites named S1-S5 and S1’[37]. Each
site binds an amino acid in the peptide analogously named P1-P5 and P1’. P1’ is the C-terminal
amino acid, which is hydrolyzed off by the enzyme after cleavage between P1’ and P1. The C-
terminus of a peptide is recognized by interactions between the carboxylic end and protein
residues in the S1’-pocket of the enzyme [34, 38, 39]. The C-terminal amino acid is thereby
selectively cleaved off. The newly formed peptide fragment can undergo further truncations at
its C-terminus. In this way new substrates are created as degradation of the peptide proceeds.
The subsites of CPY vary in specificity. For example, the S1 preferentially accommodates
large hydrophobic amino acids [40]. As a result, the kcat/KM for CPY towards a peptide is
dependent of its amino acid sequence, and may vary more than 1000 times solely by
variations in the P1’- and P1-positions [32]. This corresponds to possible differences in
activation energy larger than 4 kcal/mol (room temperature).
The large variation in substrate specificity dependent on the amino acid sequence of the
digested peptide pose a problem in carboxypeptidase mediated ladder sequencing. As
fragments with various C-terminal amino acid sequences are generated throughout the
carboxypeptidase digestion, concentrations of the different peptide fragments may not always
be sufficient for analysis. The aim of incorporating an alternative nucleophile at the C-
8
Anders Hamberg
terminus of a fraction of the digested peptides is to cancel out some of the variations in
substrate specificity. An ultimate goal would be to obtain totally stable peptide fragments
from the alternative reaction. Although, such goal demands that the big differences in
substrate specificity towards the variety of peptide fragments derived in the digest have to be
overcome.
In order to effectively incorporate an alternative nucleophile at the C-terminus of the
digested peptide it should fit in the S1’-subsite. This also means that the enzyme will have
affinity towards the peptide with a bound alternative end group and will subject it to further
degradation. However, KM towards the free nucleophile does not have to be in the same range
as towards the digested peptide if the frequency of incorporation can be adjusted by varying
the concentration of free nucleophile. In the experiments presented in paper I the
concentration of the free alternative nucleophile is 4500 times higher than the starting
concentration of the peptide. The large difference in concentrations is combined with the
possibility for the alternative nucleophile to bind in any of the different subsites. Thus, the
concentration of the alternative nucleophile has to be balanced to avoid competitive inhibition
from the alternative nucleophile.
2.4 Identification frequency Carboxypeptidase mediated peptide ladder sequencing protocols can be compared using
identification frequency as a parameter of efficiency (paper I). The identification frequency is
based on the amount of information per mass spectrum of aliquots taken from digests of
different peptides at set reaction times. The calculation of identification frequencies is fully
explained in paper I. Determining identification frequencies can be advantageous when
comparing sequencing results over a defined range of sequences obtained by different digest
methods. The use of various peptides and several aliquots for determination of an
identification frequency can cancel out accidental deviations between measurements.
Variations in the success rate of a sequence experiment may be caused by many factors in the
sequencing method other than the enzymatic step; such as sample handling or discriminations
between different fragments in MS. The identification frequency can also reveal trends
derived from different digest protocols for ladder sequencing. For example, the probability to
identify an amino acid at a certain position in a peptide is dependent on the reaction times
used for taking aliquots from a carboxypeptidase digest. Substrate specificity of the
carboxypeptidase and concentrations of enzyme and substrate can also cause variations which
will be displayed in the identification frequency.
9
Amino acid sequence determination
An alternative to identification frequency is sequence coverage, which also can be
calculated per aliquot. A difference is that the sequence coverage is peptide specific whereas
results from sequencing of more than one peptide can be included when determining the
identification frequency. The sequence coverage is related to the identification frequencies by
their corresponding averages over a given sequence interval for one or several different
peptides. An average of the sequence coverage per aliquot taken from digests of various
peptides will be identical to the average identification frequency calculated from the same
experiments. However, the average of either coverage or identification frequency may not be
satisfactory when comparing two different digest methods as it will not show any trends
dependent on the amino acid sequence. Such trends can be displayed by calculating the
identification frequency for each amino acid position in the sequence interval.
2.5 Results The identification frequencies in carboxypeptidase Y digests of six different peptides were
determined for the amino acids at positions 1-10 from the C-terminus (paper I). Mass spectra
of aliquots taken at 1, 10, 100 and 1000 minutes of reaction time were used in the
calculations. The experiments were carried out using a buffer containing 2-
pyridylmethylamine (2-PMA) as an alternative nucleophile. A digest carried out in a buffer
without any alternative nucleophile was used as a reference. The average detection frequency
in the interval was 29% when 2-PMA was present compared to 19% in the reference. An
improvement of the identification frequency of around 50% was thereby accomplished. As
previously discussed (section 2.5) these values are identical to the average sequence coverage
per aliquot, calculated for the interval 1-10 from the C-terminus.
Initially, the identification frequency for each position of the amino acid sequence was
calculated (paper I, figure 5). Variations in identification frequencies dependent on the amino
acid position were revealed for both the 2-PMA containing digest and the reference. Both
digest methods resulted in similar trends with the highest identification frequency around
position 7, and higher values for the digest containing 2-PMA. These trends can be caused by
differences in substrate specificities combined with a relatively limited number of peptides.
Another reason for the observed trend is the reaction times. Changing the time protocol for
aliquot sampling could result in a shifted peak of identification frequency and a different
trend. It is apparent that both time protocol and variations in substrate specificities have the
same effect regardless of the presence of 2-PMA. The amine-capping by 2-PMA stabilize the
peptide fragments and enhance the probability to identify amino acids in CPY mediated
10
Anders Hamberg
peptide sequencing. Consequently, a smaller amount of aliquots and less peptide specific
optimization of the digest protocol is necessary to obtain a sufficient sequence coverage. Both
time and valuable peptide samples can thereby be saved by using the presented method.
ValLeuLeuHisPhe
Phe His Leu
Phe His Leu
Tyr + Ser
-90 Da
* **
**
*
**
** *
*
A
B
* *
*
m/z
Arb
itrar
y un
its
m/z
Arb
itrar
y un
its
Figure 3. Mass spectra of two CPY digest of N-acetyl-renin-tetradecapeptide after 10 min reaction time. The upper mass spectrum (A) corresponds to CPY digest without any alternative nucleophile. The lower mass spectrum (B) corresponds to CPY digest with 2-PMA present as an alternative nucleophile. The amino acids identified from hydrolysis fragment series are indicated with drawn double-point arrows and amino acids identified from amine-capped fragments are indicated with dashed double-point arrows. A dotted double-point arrow in spectrum B indicates a difference in the peptide ladder corresponding to the additive mass of tyrosine and serine minus 90 Da, which is the mass of a peptide-bound 2-PMA. A signal to noise ratio of 3 was used as a limit for identification of peptide fragments. Peaks marked with an asterisk (*) correspond to Na or K adducts. A comparison between mass spectra of 100 minutes aliquots taken from CPY digests of
bombesin with and without the alternative nucleophile 2-PMA is done in paper I, figure 3. A
similar example with mass spectra of 10 min. aliquots taken from CPY digests of N-acetyl-
renin-tetradecapeptide shows an increased number identified amino acids when the alternative
nucleophile 2-PMA is used (figure 3). The example with bombesin discussed in paper I
11
Amino acid sequence determination
suggests that an improvement in identification frequency can be accomplished by a buffering
effect from the amine-capped peptides. The C-terminal bound 2-PMA can be hydrolyzed off
to yield peptide fragments which are by large extent degraded by the enzyme (paper I, figure
4). The example in figure 3 shows that an improvement also can be caused by amine-capped
peptide fragments remaining in the digest solution after hydrolysis of the peptide fragments.
A double set of peptide fragments created from hydrolysis and aminolysis are displayed as
two ladders in mass spectra, with a shift corresponding to the mass of a peptide bound 2-
PMA. A gap in the sequence corresponding to the mass of tyrosine and serine after addition
of the mass of peptide-bound 2-PMA (90 Da) provides information regarding the amino acid
composition without revealing the internal sequence. The double identifications of amino
acids from the double set of peptide fragments and mass differences correlated to more than
one amino acid were not included in the identification frequencies discussed above.
12
Anders Hamberg
3 Determination of Enantiomeric excess Molecular chirality plays an important role when designing compounds for use in various
fields such as drug development and agricultural chemistry. In order to meet a great demand
of enantiomerically pure or enriched compounds, efficient methods for enantioselective
synthesis are necessary. Hence, production of enantioselctive catalysts is a central field of
research in both organic and enzymatic synthesis.
High-throughput methods have frequently been used for finding enantioselective catalysts
for various reactions. These methods involve generation and evaluation of large libraries of
diverse catalysts, rather than single sample experiments. Catalyst libraries can be generated by
combinatorial methods, which are based on the use of metal complexes with chiral ligands
and activators [41, 42]. The various combinations of ligands and activators induce diversity in
the library which may lead to enantiomeric enrichment of the synthesized compound. An
alternative to combinatorial chemistry is directed evolution for an enzyme catalyzed reaction
[43]. Error prone PCR, gene shuffling and saturation mutagenesis are commonly used for
inducing enzyme diversity.
3.1 Screening for enantioselectivity A challenge associated with production of large compound libraries is the demand of
increased throughput in analysis. Screening for enantiomeric excess (ee) of compounds
created by a catalyst library has become a research field of great interest reviewed in several
publications [43-45]. As no single screening method has yet proven to be universal, new
methods have to be specifically developed for the synthesized substance when creating a
catalyst library. Hence, several methods for rapid ee-determinations based on various
techniques have been developed.
Conventional methods for ee-determination have been modified and optimized for increased
throughput. Parallel apparatus for gas chromatography (GC) equipped with chiral columns
were optimized for rapid ee-determination of the substance 2-phenylpropanol [43, 46].
Studies using modified capillary electrophoresis (CE) for high throughput screening of
secondary amines have shown promising results [43, 47]. Capillary array electrophoresis
(CAE), originally designed for DNA separations, and CE on micro-chips have a potential to
increase throughput significantly compared to conventional CE. Circular dichroism has been
successfully employed in combination with high performance liquid chromatography (HPLC)
for ee-determinations of 1-phenylethanol. The HPLC is used for separation of reaction
13
Determination of enantiomeric excess
product from the reactants but no chromatographic separation of the enantiomers is necessary,
which increases throughput significantly [43, 48]. Both chromatography and capillary
electrophoresis are universal methods in the sense that almost any compound can be analyzed
by separation of the enantiomers. However, the retention times needed for separation are
dependent on the compound and may not be feasible when catalyst libraries grow.
Mass spectrometry has been used for ee-determinations of pseudo-enantiomers differing in
absolute configuration and mass [43, 49-51]. A mass ratio obtained by different amounts of
the pseudo-enantiomers can be correlated to the ee of the product [43, 49, 50]. Analysis has
also been carried out directly on pseudo-enantiomeric reaction intermediates associated with
the catalyst [51]. Fourier transform infrared spectroscopy (FTIR) and flow-through nuclear
magnetic resonance spectroscopy (NMR) are other suitable methods for determining
enantiomeric excess of 13C-labeled pseudo-enantiomers as reviewed by Reetz et al. [43].
Using MS, FTIR or NMR for ee-determinations in catalysis requires pure forms of mass- or
isotope-tagged pseudo-enantiomers either as a reactant or product in the catalyzed reaction.
These methods are therefore limited to ee-determinations in kinetic resolution of racemates
and desymmetrization of prochiral compounds bearing enantiotopic groups.
Numerous methods have been developed for enabling ee-determinations by conventional
spectrophotometry. Many of these methods are based on using molecules which can give a
change in fluorescence [52-55], color [56-58] or UV-absorbance [59] by enantiospecific
interactions with a target substance. The ultra high-throughput potential of fluorescence was
demonstrated by ee-determinations of various amino acids in microarray format [43, 60].
Primarily, interaction induced spectrophotometric ee-determinations can be done on chiral
amines, amino alcohols, amino acids, α-hydroxycarboxylic acids and monosaccarides.
Compounds such as simple alcohols, esters and ketones may require alternative methods [55].
Highly substance-specific methods for ee-determinations have been developed by utilizing
various biochemical techniques such as immunochemistry, protein-substance interactions and
enantioselective enzyme catalysis. Immunochemical methods are based on antibodies with
affinity towards the synthesized product [61, 62]. An attractive system is to use competitive
enzyme immunoassays for ee-determination [61], which requires enantiospecific antibodies
against the target substance. Theoretically, the immune system can raise antibodies against
any compound. However, production of antibodies with the required affinity is often a
laborious task. Recently, enantiospecific interaction with an engineered transmembrane
protein (i.e. an α-hemolysin variant) was discovered useful for ee-determinations [63].
Enantiomeric excess was determined for the pharmaceuticals ibuprofen and thalidomide.
14
Anders Hamberg
Enzymatic methods for determining enantiomeric excess (EMDee) have been developed for a
few substances [64-68]. The principals of EMDee are described in detail in the following
section (3.2).
3.2 Enzymatic determination of enantiomeric excess Enantioselective enzymes can be used for determining enantiomeric excess of a substance
[64-68]. The enzyme catalyzes conversion of the substance into a chemically different
species, enabling quantification of the different enantiomers by absorbance measurement. The
reaction rate at high substrate concentration (Vmax) can be directly correlated to the ee-value if
the enantiomer which is not favored in catalysis by the enzyme inhibits the enzyme [64].
Otherwise, knowledge of the total concentration is required for calculation of the ee-value
from the reaction rate [65, 67]. Alternatively, two different enzymes catalyzing the same
reaction but with different enantiopreference can be used without knowledge of the substrate
concentration [66]. A change in absorbance can be obtained by a pH-indicator [65], a reaction
product [64, 66] or the yield of a coupled reaction [67].
The method presented herein relies on using relative endpoints rather than relative reaction
rates for determining enantiomeric excess. This gives the advantage that the method becomes
less sensitive to variations in concentration of substrate and added enzymes. Consequently,
the yield from the synthesis does not have to be known before screening of the product.
Furthermore, the system becomes insensitive to fluctuations in enzyme concentrations and is
therefore suitable for miniaturization. For successful ee determination using endpoints it is
important that the enantioselectivity for the enzyme is sufficiently high (i.e. E>100) towards
the screened substrate, enabling depletion of one of the enantiomers. In this way a stable
signal can be obtained over time without any interference from background reaction of the
remaining enantiomer, which can be quantified by using a complementary enzyme. Thereby
the total concentration of the synthesized product can be determined from the total change.
3.3 Enzymatic analysis of O-acylated cyanohydrins The enzymatic method for determination of enantiomeric excess relying on relative endpoints
was implemented on O-acylated cyanohydrins (papers II, III), synthesized according to a
previously defined method [69]. The analysis of ee and conversion is carried out by
measuring changes in absorbance caused by three consecutive reaction steps. Each step is
carried out by addition of an enzyme exhibiting a desired specificity (figure 4). The first
enzyme added is Horse liver alcohol dehydrogenase, HLADH. By adding HLADH reduction
of aldehyde to alcohol is catalyzed yielding NAD+ from NADH (nicotinamide adenine
15
Determination of enantiomeric excess
dinucleotide), which causes a drop in absorbance (step 1, figure 4). The next step of the
analysis is carried out by adding the enzyme Candida antarctica Lipase B, CALB. Thereby,
ester hydrolysis of the (S)-enantiomer of the acylated cyanohydrin is catalyzed (step 2, figure
4). After depletion of the (S)-enantiomer Pig liver esterase, PLE, is added for hydrolysis of the
(R)-enantiomer (step 3, figure 4). The products from hydrolysis of both enantiomers are
coupled to the oxidation of NADH to NAD+ and the decrease in absorbance can be used for
determining ee. Additionally, conversion can be calculated using the decrease in absorbance
of all three reaction steps.
time
abso
rban
ce
1. HLADHUnreactedsubstrate
2. CALB(S)-enantiomer
3. PLE(R)-enantiomer
R
O
R
OH
NADH + H+ NAD+
HLADH
R CN
O
O
R'
R CN
OH
H2O
HO
O
R'
R
O
R
OH
NADH + H+ NAD+HCN
CALB Spontanous HLADH
R CN
O
O
R'
R CN
OH
H2O
HO
O
R'
R
O
R
OH
NADH + H+ NAD+HCN
PLE Spontanous HLADH
Figure 4. Three different enzymes are added in successive steps: 1) Reduction of aldehyde to alcohol catalyzed by HLADH. 2) Selective ester hydrolysis of the (S)-enantiomer of the product. 3) Unselective ester hydrolysis of the remaining enantiomer, (R), of the product. Oxidation of NADH to NAD+ corresponding to the yield in each reaction step enables determination of ee and conversion by absorbance measurement.
The product from the ester hydrolysis is in spontaneous equilibrium with the corresponding
aldehyde, which is used in synthesis. Hydrolysis of acylated cyanohydrins can thereby be
quantified from the reduction of aldehyde, coupled to oxidation of NAD+ from NADH (step 2
16
Anders Hamberg
and 3, figure 4). The reaction is catalyzed by HLADH which is already prevalent in the
reaction buffer as it is the first enzyme added (step 1, figure 4). This color reaction enables
determination of ee from absorbance measurements. Additionally, the amount of aldehyde
remaining from the synthesis of the acylated cyanohydrin is assessed and can be used to
calculate the conversion. Again, absolute concentration does not have to be known as relative
measurements are used.
CALB is known to exhibit high enantioselectivity towards secondary alcohols and esters
and is therefore a suitable catalyst for depletion of the (S)-acylated cyanohydrins. PLE
catalyzes hydrolysis of both (R)- and (S)-enantiomers of the acylated cyanohydrins. However,
the amount of the (R)-enantiomer can be determined by assuming that the remaining substrate
is in this configuration when the enzyme is added.
3.4 Substrate specificity An advantage with using enzymes for analysis is the possibility to use the same method for
screening of a relatively wide variety of substances. Still, highly selective conversion of
specific components in a reaction mixture can be obtained. In order to achieve accurate
determinations of ee and conversion using the presented method certain requirements on the
substrate specificity for the involved enzymes towards their substrates have to be met.
As discussed earlier in section 3.2, high enantioselectivity for the enzyme used for selective
depletion of one enantiomer is crucial for successful determination of ee in the presented
method. In agreement with Kaslauska’s rule [70], the enantioselectivity for a lipase towards
the O-acylated cyanohydrins should be dependent on the difference in size between the R-
and CN-group of the cyanohydrin (figure 4). The enantioselectivity for CALB towards
secondary alcohols and esters is derived from a stereospecificity pocket in the active site [71],
which should fit the cyano group of the cyanohydrin. Therefore, successful ee-determination
using CALB in the presented method may be possible for O-acylated cyanohydrins with
substituents at the R-position large enough to obtain the required enantioselectivity. Since the
R-group points out towards the entrance of the active site, its size should not be limited by the
enzyme.
The active site of CALB also have a pocket for the acyl part of the substrate (R’-group,
figure 2) [71]. Linear aliphatic R’-substituents of various lengths should not have any
problems to fit in this pocket [72]. However, too large hydrophobic substituents on either the
R- or the R’-group of the substrate may affect the solubility in aqueous buffer of both
17
Determination of enantiomeric excess
substrate and product and thereby prevent analysis. Some nonlinear acyl chains may also fit
into this pocket, since it is not as confined as the stereospecificity pocket.
The dynamic range in absorbance measurements is limited due to Lambet-Beer’s law (0.2-
0.8 absorbance units). The enzymes in the assay therefore have to have kcat/KM-values towards
the tested substances which enable catalysis at low substrate concentrations (10-100 µM). As
an example HLADH has a KM-value towards benzaldehyde which makes the enzyme suitable
as a catalyst when analyzing mandelonitrile acetate with the presented method [73]. In
contrast, the specificity for HLADH towards smaller substrates such as acetaldehyde is
considerably lower, making the enzyme inappropriate for analysis of lactonitrile acetate.
As mentioned earlier (section 3.3), the enzyme used for complementary ester hydrolysis
does not have to be enantioselctive towards the analyzed substance. Commercially available
PLE is often a mixture of isozymes which have low enantioselectivity and may thereby be a
good catalyst for both enantiomers [74].
In most type of chemical analysis, impurities such as byproducts or remaining reactants and
reagents from the synthesis may affect the analysis. Using enzymes with adequate substrate
specificities towards their substrates can prevent such effects without purification of the
analyte.
3.5 Results and discussion The screening method described in section 3.3 was evaluated by enzymatic determination of
various ee of mandelonitrile acetate, published in paper II. This initial study demonstrated
that the design of the method was adequate and applicable in both cuvette- and microtiter
plate format. However, a more extensive evaluation was needed to define the scope and
limitations of the method concerning substrate diversity, robustness and applicability in high-
throughput screening. In paper III, the method was hence tested using O-acylated
cyanohydrins with various substitutions at the chiral center of the cyanohydrin and at the acyl
part of the compound (respectively defined as the R- and R’-group in figure 4). Additionally,
paper III includes a study using the method for high-throughput screening of a combinatorial
catalyst library model.
18
Anders Hamberg
Table 1. Substances subjected to enzymatic analysis Substancenumber*
Substituents** R R’
Enzymatic determination ee conversion
1a C6H5 CH3 yes yes 1b 4-CH3-C6H5 CH3 yes yes 1c 4-CH3-O-C6H5 CH3 yes yes 1d 4-Cl-C6H5 CH3 yes yes 1e 2-Furyl CH3 yes yes 1f 3-(PhO)C6H5 CH3 no yes 1g 3-Pyridyl CH3 yes yes 1h 2-Pyridyl CH3 no yes 1i (E)-CH=CHPh CH3 yes yes 1j C(CH3)3 CH3 no no 1k (CH2)4CH3 CH3 no no 1l C6H5 CH2CH3 yes yes 1m C6H5 (CH2)2CH3 yes yes 1n C6H5 (CH2)3CH3 no no 1o C6H5 CH(CH3)2 yes yes 1p C6H5 C(CH3)3 no no 1q C6H5 (E)-CH=CHPh no no 1r C6H5 C6H5 no no *Structures of substance 1a-1r can be found in paper III.
R CN
O R'
O **Substituents on an acylated cyanohydrin as defined:
The substrates listed in table 1 were tested using crude samples of the reaction mixture. A
racemic sample and samples containing an enantiomeric excess of the (S)- as well as the (R)-
enantiomer were used for assessment. Verification was done by plotting the enzymatically
determined ee to their corresponding value determined by GC (figure 5). Furthermore, each
sample was analyzed at two substrate concentrations differing by a factor of 2. Enzymatic
determination of ee was achieved for 10 and conversion was enzymatically determined for 12
of 18 tested substances (tab. 1). Possible explanations for the reason why some compounds
could not be enzymatically analyzed are presented in paper III. High accuracy of the
enzymatically determined ee-values and conversion was obtained regardless the variations in
substrate concentrations.
19
Determination of enantiomeric excess
-100
-80
-60
-40
-20
0
20
40
60
80
100
-100 -50 0 50 100
GC (% ee )
Enzy
mat
ic (%
ee)
1d1l
Figure 5. Example of enzymatic determination of ee plotted as a function of the corresponding values determined by GC for 1d ( ) and 1l ( ). Positive ee values correspond to an excess of the (R)-enantiomer, negative to an excess of the (S)-enantiomer. The dotted line corresponds to EMD value = GC value. The continuous line is a linear regression of results from 1d (EMDee = 0.97GCee -2.8, r2 = 0.9992). The dashed line is a linear regression of results from 1l (EMDee = 0.82GCee – 0.66, r2 = 0.996).
A small catalyst library was created for synthesis of mandelonitrile acetate in a micro reactor.
A salen-Ti complex were used as a catalyst activated by various Lewis bases and samples
were taken at two different reaction times to obtain samples with different ee and conversion
(tab. 2). Enzymatic determination of ee and conversion was applied as previously described in
96-well microtiter plate format for screening the library. No purification of the samples was
done prior to the screening. Quadruplicate measurements were made for each sample to
evaluate the reproducibility of the method.
The enzymatically determined values for ee and conversion were successfully expressed as
a function of the values obtained from GC (paper III). The experiments showed that the
different catalytic complexes and Lewis bases used for synthesis had no apparent effect on the
analysis. Standard deviations of the measured ee and conversion differed between the
analyzed samples. This can be explained by variations in dynamic range of absorbance caused
by differences in conversion and concentration.
20
Anders Hamberg
Table 2. Catalyst library created from various Lewis bases, catalysts
and reaction times*. Lewis base Catalyst Reaction time
(min) ee (%)
conversion (%)
Et3N salen-Ti2 20 -49 30 Et3N salen-Ti2 40 -62 70 DBU salen-Ti2 20 -20 96 DBU salen-Ti2 40 -7 97 cinchonidine salen-Ti2 20 -63 22 cinchonidine salen-Ti2 40 -74 66 quinine salen-Ti2 20 -42 23 quinine salen-Ti2 40 -61 64 DIPEA salen-Ti2 20 -59 68 DIPEA salen-Ti2 40 -55 73 DMAP salen-Ti2 20 -75 45 DMAP salen-Ti2 40 -75 74 DEA salen-Ti2 20 -59 24 DEA salen-Ti2 40 -64 65 cinchonidine ent-salen-Ti2 20 46 42 cinchonidine ent-salen-Ti2 40 87 80 quinine ent-salen-Ti2 20 75 94 quinine ent-salen-Ti2 40 83 94 DABCO salen-Ti2 20 -35 12 DABCO salen-Ti2 40 -79 63
*All reactions were carried out in dichloromethane at room temperature using 2 equivalents of acetyl cyanide, 5 mol% catalyst and 10 mol% Lewis base. Both ee and conversion were determined by GC. Positive ee denote an excess of the (R)-enantiomer whereas negative ee denote an excess of the (S)-enantiomer.
In an early stage of the development of the enzymatic method for determination of
enantiomeric excess para-nitrophenol was used as a pH indicator to yield a color reaction for
absorbance measurement. The enzymes CALB and PLE were used as described in section
3.3. A linear relationship between ee-values determined enzymatically and by GC was
obtained for purified samples in cuvettes (paper III). However, a weak buffer had to be used
in order to obtain a measurable change in pH making the method sensitive to contaminations
such as atmospheric carbon dioxide. Still, the pH-indicator can work as a tool for designing
similar assays for other substances. The system with CALB and PLE could for example be
useful for determination of ee of other types of secondary esters.
An interesting feature of the enzymatic method for determining enantiomeric excess by
relative endpoints is that it is insensitive to variations in concentrations of both substrate and
enzyme. The method could thereby be suitable for miniaturization into a “lab on a chip”. A
first step towards a miniaturized system could be to move from absorbance to fluorescence
spectroscopy for increased sensitivity.
21
22
Anders Hamberg
4 Conclusions In the work presented in chapter 2 (based on paper I), an alternative nucleophile was used for
improving the sequence coverage in carboxypeptidase Y mediated peptide sequencing. The
developed method was proven to be useful for C-terminal sequencing of peptides. The
method can be directly applied on any peptide sequencing experiment where
carboxypeptidase Y truncation is considered. Further studies could help additional
improvements of the sequence coverage. For example, other compounds than 2-PMA could
be used to increase the stability of the peptide fragments derived in the digest.
Chapter 3 (based on paper II and III) presents a method for determining enantiomeric
excess and yield. The method was applied for evaluation of a protocol for combinatorial
synthesis of acylated cyanohydrins. The substrate specificity for three different enzymes,
HLADH, CALB and PLE, were utilized to accomplish measurable changes in absorbance
from oxidation of NADH to NAD+. The high enantioselectivity for CALB towards acylated
secondary alcohols enabled determinations of enantiomeric excess from relative endpoints.
Evaluations of the robustness, substrate diversity and high-throughput applicability of the
method showed good and promising results. An alternative colour reaction based on titration
of formed acid with coloured pH-indicator can open up possibilities to use CALB and PLE
for analysis of the enantioselectivity in ester synthesis protocols other than the synthesis of
cyanohydrins evaluated herein.
23
24
Anders Hamberg
5. List of Abbreviations kcat Turnover number for an enzyme towards a substrate
KM Michaelis-Menten constant
Vmax Maximum reaction rate for an enzyme catalyzed reaction
E Enantiomeric ratio
MS Mass spectrometry
m/z Mass to charge ratio
MS/MS Tandem mass spectrometry
CPY Carboxypeptidase Y
Pn Amino acid in a digested peptide or protein at position n on N-terminal side
from the cleavage site
Pn’ Amino acid in a digested peptide or protein at position n on C-terminal side
from the cleavage site
Sn Subsite in a peptidase or protease which accommodates the amino acid n in a
peptide or protein
2-PMA 2-pyridylmethylamine
PCR Polymerase chain reaction
ee Enantiomeric excess
GC Gas chromatography
CE Capillary electrophoresis
CAE Capillary array electrophoresis
HPLC High Performance Liquid Chromatography
FTIR Fourier transform infrared spectroscopy
NMR Nuclear magnetic resonance spectroscopy
UV Ultra violet
EMDee Enzymatic method for determining enantiomeric excess
HLADH Horse liver alcohol dehydrogenase
NAD+ Nicotinamide adenine dinucleotide, oxidized form
NADH Nicotinamide adenine dinucleotide, reduced form
CALB Candida antarctica Lipase B
PLE Pig liver esterase
25
26
Anders Hamberg
6. Acknowledgements This thesis has been influenced by a number of people to whom I am most grateful.
I would like to thank my supervisor, Professor Karl Hult, for accepting me as a student and
guiding me along the way. I look forward to continue working with you.
To Professor Johan Roerade, Martin Kempka and Dr Johan Sjödahl at the Department of
Analytical Chemistry, KTH School of Chemical Science and Engineering: I thank you for
providing expertise in mass spectrometry essential for the work published in paper I.
To Professor Christina Moberg, Stina Lundgren, Erica Wingstrand and Maël Penhoat at
Organic Chemistry, KTH School of Chemical Science and Engineering: Thank you for an
excellent collaboration with the work published in paper II and III.
I would also like to thank my present and previous colleagues in the Biocat group for help,
interesting discussions and for just being nice: Per Berglund, Cecilia Branneby, Karim
Engelmark Cassimje, Magnus Eriksson, Linda Fransson, Cecilia Hedfors, Marianne Larsen,
Anders Magnusson, Mats Martinelle Per-Olof Syrén, Maria Svedendahl and Mohamad Takwa
(presented in alphabetical order).
The rest of the floor 2, including administrators Lotta and Ela, the Wood Biotechnology group
and IT-coordinators Erik and Kaj, thank you for giving a helping hand once in a while and
providing a nice and stimulating working environment.
For financial support I thank The Nanochemistry program, funded by the Swedish Foundation
for Strategic Research.
On a personal level I want to thank my father Percy, my mother Sonja and her partner Arnulf.
My deepest gratitude goes out to my family, my fiancé Newroz and our son Robin. You are
far more important to me than anything else, I love you.
27
28
Anders Hamberg
7. References 1. Bergmeyer, H., Methods of Enzymatic Analysis, 3rd Edition, Vol 1-4. 1983.
2. Devlin, T.M., 1959, Enzymatic Methods in Analytical Chemistry. Analytical
Chemistry, 31, 977-981.
3. Fishman, M.M., 1980, Enzymes in Analytical-Chemistry. Analytical Chemistry, 52,
R185-R199.
4. Radzicka, A. and R. Wolfenden, 1995, A Proficient Enzyme. Science, 267, 90-93.
5. Giraldo, J., D. Roche, X. Rovira, and J. Serra, 2006, The catalytic power of enzymes:
Conformational selection or transition state stabilization? Febs Letters, 580, 2170-
2177.
6. Danno, G., 1970, D-glucose-isomerizing enzyme from Bacillus coagulans, strain HN-
68. V. Comparative study on the three activities of D-glucose, D-xylose, and D-
ribose isomerization of the crystalline enzyme. Agricultural and Biological Chemistry,
34, 10.
7. Bhosale, S.H., M.B. Rao, and V.V. Deshpande, 1996, Molecular and industrial
aspects of glucose isomerase. Microbiological Reviews, 60, 280-&.
8. Steen, H. and M. Mann, 2004, The ABC's (and XYZ's) of peptide sequencing. Nature
Reviews Molecular Cell Biology, 5, 699-711.
9. Patterson, S.D. and R.H. Aebersold, 2003, Proteomics: the first decade and beyond.
Nature Genetics, 33, 311-323.
10. Agrawal, G.K. and R. Rakwal, 2006, Rice proteomics: A cornerstone for cereal food
crop proteomes. Mass Spectrometry Reviews, 25, 1-53.
11. Wang, P., G. Arabaci, and D. Pei, 2001, Rapid sequencing of library-derived peptides
by partial Edman degradation and mass spectrometry. Journal of Combinatorial
Chemistry, 3, 251-254.
12. Joo, S.H., Q. Xiao, Y. Ling, B. Gopishetty, and D.H. Pei, 2006, High-throughput
sequence determination of cyclic peptide library members by partial Edman
degradation/mass spectrometry. Journal of the American Chemical Society, 128,
13000-13009.
13. Sanger, F., 1988, Sequences, Sequences, and Sequences. Annual Review of
Biochemistry, 57, 1-28.
29
References
14. Hunt, D.F., J.R. Yates, J. Shabanowitz, S. Winston, and C.R. Hauer, 1986, Protein
Sequencing by Tandem Mass-Spectrometry. Proceedings of the National Academy of
Sciences of the United States of America, 83, 6233-6237.
15. Biemann, K. and H.A. Scoble, 1987, Characterization by Tandem Mass-Spectrometry
of Structural Modifications in Proteins. Science, 237, 992-998.
16. Wilm, M., A. Shevchenko, T. Houthaeve, S. Breit, L. Schweigerer, T. Fotsis, and M.
Mann, 1996, Femtomole sequencing of proteins from polyacrylamide gels by nano-
electrospray mass spectrometry. Nature, 379, 466-469.
17. Budnik, B.A., M.L. Nielsen, J.V. Olsen, K.F. Haselmann, P. Horth, W. Haehnel, and
R.A. Zubarev, 2002, Can relative cleavage frequencies in peptides provide additional
sequence information? International Journal of Mass Spectrometry, 219, 283-294.
18. Horn, D.M., R.A. Zubarev, and F.W. McLafferty, 2000, Automated de novo
sequencing of proteins by tandem high-resolution mass spectrometry. Proceedings of
the National Academy of Sciences of the United States of America, 97, 10313-10317.
19. Zhong, H.Y., Y. Zhang, Z.H. Wen, and L. Li, 2004, Protein sequencing by mass
analysis of polypeptide ladders after controlled protein hydrolysis. Nature
Biotechnology, 22, 1291-1296.
20. Edman, P. and G. Begg, 1967, A Protein Sequenator. European Journal of
Biochemistry, 1, 80-91.
21. Chait, B.T., R. Wang, R.C. Beavis, and S.B.H. Kent, 1993, Protein Ladder
Sequencing. Science, 262, 89-92.
22. Suzuki, Y., M. Suzuki, Y. Nakahara, Y. Ito, E. Ito, N. Goto, K. Miseki, J. Iida, and A.
Suzuki, 2006, Structural characterization of glycopeptides by N-terminal protein
ladder sequencing. Analytical Chemistry, 78, 2239-2243.
23. Boyd, V.L., M. Bozzini, G. Zon, R.L. Noble, and R.J. Mattaliano, 1992, Sequencing of
Peptides and Proteins from the Carboxy Terminus. Analytical Biochemistry, 206,
344-352.
24. Samyn, B., K. Hardeman, J. Van der Eycken, and J. Van Beeumen, 2000,
Applicability of the alkylation chemistry for chemical C-terminal protein sequence
analysis. Analytical Chemistry, 72, 1389-1399.
25. Patterson, D.H., G.E. Tarr, F.E. Regnier, and S.A. Martin, 1995, C-Terminal Ladder
Sequencing Via Matrix-Assisted Laser-Desorption Mass-Spectrometry Coupled with
Carboxypeptidase-Y Time-Dependent and Concentration-Dependent Digestions.
Analytical Chemistry, 67, 3971-3978.
30
Anders Hamberg
26. Klarskov, K., K. Breddam, and P. Roepstorff, 1989, C-Terminal Sequence
Determination of Peptides Degraded with Carboxypeptidases of Different Specificities
and Analyzed by 252-Cf Plasma Desorption Mass-Spectrometry. Analytical
Biochemistry, 180, 28-37.
27. Daldegan, F., B. Ribadeaudumas, and K. Breddam, 1992, Purification and
Characterization of 2 Serine Carboxypeptidases from Aspergillus-Niger and Their
Use in C-Terminal Sequencing of Proteins and Peptide-Synthesis. Applied and
Environmental Microbiology, 58, 2144-2152.
28. Thiede, B., B. Wittmannliebold, M. Bienert, and E. Krause, 1995, Maldi-Ms for C-
Terminal Sequence Determination of Peptides and Proteins Degraded by
Carboxypeptidase-Y and Carboxypeptidase-P. Febs Letters, 357, 65-69.
29. Cool, D.R. and A. Hardiman, 2004, C-Terminal sequencing of peptide hormones using
carboxypeptidase Y and SELDI-TOF mass spectrometry. Biotechniques, 36, 32-34.
30. Bonetto, V., A.C. Bergman, H. Jornvall, and R. Sillard, 1997, C-terminal sequence
analysis of peptides and proteins using carboxypeptidases and mass spectrometry
after derivatization of Lys and Cys residues. Analytical Chemistry, 69, 1315-1319.
31. Huang, R.H., Y. Zhang, and D.C. Wang, 2003, Primary structural determination of N-
terminally blocked peptides from the bark of Eucommia ulmoides Oliv by mass
spectrometric analysis. Rapid Communications in Mass Spectrometry, 17, 903-908.
32. Hayashi, R., Y. Bai, and T. Hata, 1975, Kinetic Studies of Carboxypeptidase Y .1.
Kinetic-Parameters for Hydrolysis of Synthetic Substrates. Journal of Biochemistry,
77, 69-79.
33. Christensen, U., H.B. Drohse, and L. Molgaard, 1992, Mechanism of
Carboxypeptidase-Y-Catalyzed Peptide Semisynthesis. European Journal of
Biochemistry, 210, 467-473.
34. Mortensen, U.H., H.R. Stennicke, M. Raaschounielsen, and K. Breddam, 1994,
Mechanistic Study on Carboxypeptidase Y-Catalyzed Transacylation Reactions -
Mutationally Altered Enzymes for Peptide-Synthesis. Journal of the American
Chemical Society, 116, 34-41.
35. Stennicke, H.R., K. Olesen, S.B. Sorensen, and K. Breddam, 1997, C-terminal
incorporation of fluorogenic and affinity labels using wild-type and mutagenized
carboxypeptidase Y. Analytical Biochemistry, 248, 141-148.
36. Sorensen, S.B., M. Raaschounielsen, U.H. Mortensen, S.J. Remington, and K.
Breddam, 1995, Site-Directed Mutagenesis on (Serine) Carboxypeptidase-Y from
31
References
Yeast - the Significance of Thr60 and Met398 in Hydrolysis and Aminolysis Reactions.
Journal of the American Chemical Society, 117, 5944-5950.
37. Nakase, H., S. Murata, H. Ueno, and R. Hayashi, 2001, Substrate recognition
mechanism of carboxypeptidase Y. Bioscience Biotechnology and Biochemistry, 65,
2465-2471.
38. Mortensen, U.H., S.J. Remington, and K. Breddam, 1994, Site-Directed Mutagenesis
on (Serine) Carboxypeptidase-Y - a Hydrogen-Bond Network Stabilizes the
Transition-State by Interaction with the C-Terminal Carboxylate Group of the
Substrate. Biochemistry, 33, 508-517.
39. Bullock, T.L., B. Branchaud, and S.J. Remington, 1994, Structure of the Complex of
L-Benzylsuccinate with Wheat Serine Carboxypeptidase-Ii at 2.0-Angstrom
Resolution. Biochemistry, 33, 11127-11134.
40. Olesen, K., U.H. Mortensen, S. Aasmulolsen, M.C. Kiellandbrandt, S.J. Remington,
and K. Breddam, 1994, The Activity of Carboxypeptidase-Y toward Substrates with
Basic P-1 Amino-Acid-Residues Is Drastically Increased by Mutational Replacement
of Leucine-178. Biochemistry, 33, 11121-11126.
41. Ding, K.L., H.F. Du, Y. Yuan, and J. Long, 2004, Combinatorial approach to chiral
catalyst engineering and screening: Rational design and serendipity. Chemistry-a
European Journal, 10, 2872-2884.
42. Shimizu, K.D., M.L. Snapper, and A.H. Hoveyda, 1998, High-throughput strategies
for the discovery of catalysts. Chemistry-a European Journal, 4, 1885-1889.
43. Reetz, M.T., 2006, Directed evolution of enantioselective enzymes as catalysts for
organic synthesis. Advances in Catalysis, 49, 1-69.
44. Traverse, J.F. and M.L. Snapper, 2002, High-throughput methods for the development
of new catalytic asymmetric reactions. Drug Discovery Today, 7, 1002-1012.
45. Tsukamoto, M. and H.B. Kagan, 2002, Recent advances in the measurement of
enantiomeric excesses. Advanced Synthesis & Catalysis, 344, 453-463.
46. Reetz, M.T., K.M. Kuhling, S. Wilensek, H. Husmann, U.W. Hausig, and M. Hermes,
2001, A GC-based method for high-throughput screening of enantio selective
catalysts. Catalysis Today, 67, 389-396.
47. Reetz, M.T., K.M. Kuhling, A. Deege, H. Hinrichs, and D. Belder, 2000, Super-high-
throughput screening of enantioselective catalysts by using capillary array
electrophoresis. Angewandte Chemie-International Edition, 39, 3891-3893.
32
Anders Hamberg
48. Reetz, M.T., K.M. Kuhling, H. Hinrichs, and A. Deege, 2000, Circular dichroism as a
detection method in the screening of enantioselective catalysts. Chirality, 12, 479-482.
49. Guo, J.H., J.Y. Wu, G. Siuzdak, and M.G. Finn, 1999, Measurement of enantiomeric
excess by kinetic resolution and mass spectrometry. Angewandte Chemie-
International Edition, 38, 1755-1758.
50. Reetz, M.T., M.H. Becker, H.W. Klein, and D. Stockigt, 1999, A method for high-
throughput screening of enantioselective catalysts. Angewandte Chemie-International
Edition, 38, 1758-1761.
51. Markert, C. and A. Pfaltz, 2004, Screening of chiral catalysts and catalyst mixtures by
mass spectrometric monitoring of catalytic intermediates. Angewandte Chemie-
International Edition, 43, 2498-2500.
52. James, T.D., K.R.A.S. Sandanayake, and S. Shinkai, 1995, Chiral Discrimination of
Monosaccharides Using a Fluorescent Molecular Sensor. Nature, 374, 345-347.
53. Li, Z.B., J. Lin, and L. Pu, 2005, A cyclohexyl-1,2-diamine-derived bis(binaphthyl)
macrocycle: Enhanced sensitivity and enantioselectivity in the fluorescent recognition
of mandelic acid. Angewandte Chemie-International Edition, 44, 1690-1693.
54. Yan, Y. and M.L. Myrick, 1999, Simultaneous enantiomeric determination of dansyl-
D,L-phenylalanine by fluorescence spectroscopy in the presence of alpha-acid
glycoprotein. Analytical Chemistry, 71, 1958-1962.
55. Pu, L., 2004, Fluorescence of organic molecules in chiral recognition. Chemical
Reviews, 104, 1687-1716.
56. Eelkema, R., R.A. van Delden, and B.L. Feringa, 2004, Direct visual detection of the
stereoselectivity of a catalytic reaction. Angewandte Chemie-International Edition,
43, 5013-5016.
57. Zhu, L. and E.V. Anslyn, 2004, Facile quantification of enantiomeric excess and
concentration with indicator-displacement assays: An example in the analyses of
alpha-hydroxyacids. Journal of the American Chemical Society, 126, 3676-3677.
58. Folmer-Andersen, J.F., V.M. Lynch, and E.V. Anslyn, 2005, Colorimetric
enantiodiscrimination of alpha-amino acids in protic media. Journal of the American
Chemical Society, 127, 7986-7987.
59. Mei, X.F. and C. Wolf, 2006, Determination of enantiomeric excess and
concentration of unprotected amino acids, amines, amino alcohols, and carboxylic
acids by competitive binding assays with a chiral scandium complex. Journal of the
American Chemical Society, 128, 13326-13327.
33
References
60. Korbel, G.A., G. Lalic, and M.D. Shair, 2001, Reaction microarrays: A method for
rapidly determining the enantiomeric excess of thousands of samples. Journal of the
American Chemical Society, 123, 361-362.
61. Taran, F., C. Gauchet, B. Mohar, S. Meunier, A. Valleix, P.Y. Renard, C. Creminon,
J. Grassi, A. Wagner, and C. Mioskowski, 2002, High-throughput screening of
enantioselective catalysts by immunoassay. Angewandte Chemie-International
Edition, 41, 124-127.
62. Matsushita, M., K. Yoshida, N. Yamamoto, P. Wirsching, R.A. Lerner, and K.D.
Janda, 2003, High-throughput screening by using a blue-fluorescent antibody sensor.
Angewandte Chemie-International Edition, 42, 5984-5987.
63. Kang, X.F., S. Cheley, X.Y. Guan, and H. Bayley, 2006, Stochastic detection of
enantiomers. Journal of the American Chemical Society, 128, 10684-10685.
64. Abato, P. and C.T. Seto, 2001, EMDee: An enzymatic method for determining
enantiomeric excess. Journal of the American Chemical Society, 123, 9206-9207.
65. Onaran, M.B. and C.T. Seto, 2003, Using a lipase as a high-throughput screening
method for measuring the enantiomeric excess of allylic acetates. Journal of Organic
Chemistry, 68, 8136-8141.
66. Li, Z., L. Butikofer, and B. Witholt, 2004, High-throughput measurement of the
enantiomeric excess of chiral alcohols by using two enzymes. Angewandte Chemie-
International Edition, 43, 1698-1702.
67. Baumann, M., R. Sturmer, and U.T. Bornscheuer, 2001, A high-throughput-screening
method for the identification of active and enantioselective hydrolases. Angewandte
Chemie-International Edition, 40, 4201-4204.
68. Sprout, C.M. and C.T. Seto, 2005, Using enzyme inhibition as a high throughput
method to measure the enantiomeric excess of a chiral sulfoxide. Organic Letters, 7,
5099-5102.
69. Lundgren, S., E. Wingstrand, M. Penhoat, and C. Moberg, 2005, Dual Lewis acid-
Lewis base activation in enantioselective cyanation of aldehydes using acetyl cyanide
and cyanoformate as cyanide sources. Journal of the American Chemical Society,
127, 11592-11593.
70. Kazlauskas, R.J., A.N.E. Weissfloch, A.T. Rappaport, and L.A. Cuccia, 1991, A Rule
to Predict Which Enantiomer of a Secondary Alcohol Reacts Faster in Reactions
Catalyzed by Cholesterol Esterase, Lipase from Pseudomonas-Cepacia, and Lipase
from Candida-Rugosa. Journal of Organic Chemistry, 56, 2656-2665.
34
Anders Hamberg
71. Uppenberg, J., N. Ohrner, M. Norin, K. Hult, G.J. Kleywegt, S. Patkar, V. Waagen, T.
Anthonsen, and T.A. Jones, 1995, Crystallographic and molecular-modeling studies
of lipase B from Candida antarctica reveal a stereospecificity pocket for secondary
alcohols. Biochemistry, 34, 16838-16851.
72. Ottosson, J. and K. Hult, 2001, Influence of acyl chain length on the enantioselectivity
of Candida antarctica lipase B and its thermodynamic components in kinetic
resolution of sec-alcohols. Journal of Molecular Catalysis B-Enzymatic, 11, 1025-
1028.
73. Ryzewski, C.N. and R. Pietruszko, 1977, Horse Liver Alcohol-Dehydrogenase Ss -
Purification and Characterization of Homogeneous Isoenzyme. Archives of
Biochemistry and Biophysics, 183, 73-82.
74. Musidlowska-Persson, A. and U.T. Bornscheuer, 2002, Substrate specificity of the
gamma-isoenzyme of recombinant pig liver esterase towards acetates of secondary
alcohols. Journal of Molecular Catalysis B-Enzymatic, 19, 129-133.
35
