Kinetic analysis of L-carnosine formation by β-aminopeptidases

Article abstract of DOI:10.1002/adsc.200900697

The β,α-dipeptide L-carnosine occurs in high concentrations in long-lived innervated mammalian tissues and is widely sold as a food additive. On a large scale L-carnosine is produced by chemical synthesis procedures. We have established two aqueous enzymatic reaction systems for the preparation of L-carnosine using the dissolved bacterial β-aminopeptidases DmpA from Ochrobactrum anthropi and BapA from Sphingosinicella xenopeptidilytica as catalysts and investigated the kinetics of the enzymecatalyzed peptide couplings. DmpA catalyzed the formation of L-carnosine from C-terminally activated β-alanine derivatives (acyl donor) and L-histidine (acyl acceptor) in an aqueous reaction mixture at pH 10 with high catalytic rates (V_max=19.2 mmol min^-1 per mg of protein, k _cat=12.9 s^-1), whereas Vmax in the BapA-catalyzed coupling reaction remained below 1.4 mmol min^-1 per mg of protein (k _cat=0.87 s^-1). Although the equilibrium of this reaction lies on the side of the hydrolysis products, the reaction is under kinetic control and L-carnosine temporarily accumulated to concentrations that correspond to yields of more than 50% with respect to the employed acyl donor. However, competing nucleophiles caused unwanted hydrolysis and coupling reactions that led to decreased product yield and to formation of various peptidic by-products. The substitution of l-histidine for L-histidine methyl ester as acyl acceptor shifted the pK_a of the amino functionality from 9.25 to 6.97, which caused a drastic reduction in the amount of coupling by-products in an aqueous reaction system at pH 8.

Full text of DOI:10.1002/adsc.200900697

FULL PAPERS
DOI: 10.1002/adsc.200900697
Kinetic Analysis of l-Carnosine Formation by b-Aminopeptidases
Tobias Heck,a,b Venkata S. Makam,a,c Jochen Lutz,d Lars M. Blank,d
Andreas Schmid,d Dieter Seebach,b Hans-Peter E. Kohler,a,* and Birgit Geuekea
a
Eawag, Swiss Federal Institute of Aquatic Science and Technology, ꢀberlandstrasse 133, 8600 Dꢁbendorf, Switzerland
Fax : (+41)-44-823-5547; e-mail: kohler@eawag.ch
Department of Chemistry and Applied Biosciences, ETH Zꢁrich, Wolfgang-Pauli-Strasse 10, 8093 Zꢁrich, Switzerland
Present address: Department of Pharmacy, National University of Singapore, 18 Science Drive 4, Singapore 117543
Laboratory of Chemical Biotechnology, TU Dortmund University, Emil-Figge-Strasse 66, 44221 Dortmund, Germany
b
c
d
Received: October 6, 2009; Published online: February 9, 2010
Abstract: The b,a-dipeptide l-carnosine occurs in 0.87 sÀ1). Although the equilibrium of this reaction
high concentrations in long-lived innervated mam- lies on the side of the hydrolysis products, the reac-
malian tissues and is widely sold as a food additive. tion is under kinetic control and l-carnosine tempo-
On a large scale l-carnosine is produced by chemical rarily accumulated to concentrations that correspond
synthesis procedures. We have established two aque- to yields of more than 50% with respect to the em-
ous enzymatic reaction systems for the preparation ployed acyl donor. However, competing nucleophiles
of l-carnosine using the dissolved bacterial b-amino- caused unwanted hydrolysis and coupling reactions
peptidases DmpA from Ochrobactrum anthropi and that led to decreased product yield and to formation
BapA from Sphingosinicella xenopeptidilytica as cat- of various peptidic by-products. The substitution of
alysts and investigated the kinetics of the enzyme- l-histidine for l-histidine methyl ester as acyl accept-
catalyzed peptide couplings. DmpA catalyzed the or shifted the pKa of the amino functionality from
formation of l-carnosine from C-terminally activated 9.25 to 6.97, which caused a drastic reduction in the
b-alanine derivatives (acyl donor) and l-histidine amount of coupling by-products in an aqueous reac-
(acyl acceptor) in an aqueous reaction mixture at tion system at pH 8.
pH 10 with high catalytic rates (Vmax =19.2 mmol
minÀ1 per mg of protein, kcat =12.9 sÀ1), whereas Vmax Keywords: b-aminopeptidases; l-carnosine; enzyme-
in the BapA-catalyzed coupling reaction remained catalyzed peptide synthesis; N-terminal nucleophile
below 1.4 mmol minÀ1 per mg of protein (kcat
=
(Ntn) hydrolases; nucleophile reactivity; b-peptides
Introduction
the translational machinery, but is formed by a specif-
ic enzyme, carnosine synthetase (EC 6.3.2.11),
b-Peptides and mixed b,a-peptides consisting of ho- through ATP-dependent coupling of b-alanine and l-
mologues of the 20 proteinogenic amino acids are histidine.[3] Levels of l-carnosine are controlled by
promising compounds for the design of novel peptido- tissue and serum carnosinases (EC 3.4.13.3 and
mimetics because the incorporation of b-amino acids 3.4.13.20, respectively), which hydrolyze l-carnosine
into peptides enhances proteolytic and metabolic sta- into its amino acid constituents, b-alanine and l-histi-
bility.[1] Naturally occurring peptides that exclusively dine.[4] l-Carnosine exhibits anti-oxidant, anti-glycat-
consist of b-amino acids are not known so far. How- ing and hydroxyl-radical scavenger properties[5] and
ever, compounds containing b-peptidic substructures has been implicated in neuroprotection and preven-
are common in nature. Prominent examples include tion of diabetes-induced nephropathy and artheroscle-
pantothenic acid, cyanobacterial microcystins, and the rosis.[6] Due to its proposed beneficial biological activ-
mammalian b,a-dipeptide l-carnosine (b-alanine-l- ities l-carnosine is widely sold as a food additive with
histidine) [b-alanine is called here b-homoglycine (H- anti-ageing potential and in complex with zinc as an
bhGly-OH) according to the nomenclature for b- anti-ulcer drug.[7]
amino acids with proteinogenic side chains[2]]. As one
The chemical bulk production of l-carnosine is well
of the few b-amino acid-containing peptides in the established and different chemical routes are avail-
mammalian metabolism, l-carnosine accumulates to able.[8] As alternatives to chemical peptide synthesis
high concentrations in innervated tissues, such as skel- in solution or on a solid-phase, enzymatic approaches
etal muscle and brain. l-Carnosine is not a product of with proteases as catalysts have gained increasing at-
Adv. Synth. Catal. 2010, 352, 407 – 415
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
407

Tobias Heck et al.
FULL PAPERS
Scheme 1. b-Aminopeptidase-catalyzed peptide formation under kinetic control. Side reactions caused by competing nucleo-
philes lead to dimerization and hydrolysis of the acyl donor (Scheme modified from Heck et al.).[14] Possible further coupling
reactions and secondary hydrolysis of the newly formed compounds are not depicted in the scheme.
tention. The use of enzymes as biocatalysts is often cally controlled reaction system,[16] the free amino
advantageous due to mild reaction conditions, high group of a second amino acid or a small peptide acts
stereoselectivity of the biocatalyst and low protective as a nucleophile (acyl acceptor) that accepts the acyl
group requirements.[9] However, most proteases and moiety from the active site leading to formation of
peptidases are limited to reactions with the 20 protei- the desired peptide product. In an aqueous reaction
nogenic a-amino acids. Thus, b-amino acid-containing mixture the acyl enzyme can undergo competing hy-
peptides are generally unusual targets for enzymatic drolysis resulting in the free amino acid, which is con-
conversions.
sequently unavailable for enzyme acylation. Since b-
Three members of the N-terminal hydrolase (Ntn) aminopeptidases require substrates with a free amino
superfamily,[10] namely the b-aminopeptidases BapA terminus,[11d] we used unprotected acyl donors for the
from Sphingosinicella xenopeptidilytica 3-2W4, BapA coupling reaction. Due to the presence of the free
from Sphingosinicella microcystinivorans Y2, and amino terminus the acyl donor itself can behave as a
DmpA from Ochrobactrum anthropi were previously nucleophilic acceptor for an acyl moiety bound to the
described to cleave N-terminal b-amino acid residues active site, which leads to dimerization of the acyl
from peptides that are otherwise resistant to proteo- donor. The dimer as well as the peptide product may
lytic breakdown.[11] In recent investigations, we char- undergo secondary hydrolysis or behave as further
acterized the biochemical properties of these enzymes nucleophiles that can react with the acyl enzyme. Due
regarding the hydrolysis of b- and mixed b,a-pep- to secondary hydrolysis of the newly formed peptide
tides[12] and reported the usefulness of the b-amino- bond the product accumulates to a clear-cut maxi-
peptidases as catalysts for the efficient kinetic resolu- mum in a kinetically controlled reaction system. To
tion of various aliphatic b-amino acid amides.[13] In obtain maximum product yields, a kinetic analysis of
addition to their hydrolytic properties, we showed the complex interplay between product formation and
that DmpA and BapA were able to catalyze the for- competing coupling and hydrolysis reactions is indis-
mation of b-amino acid oligomers and other b-amino pensable.
acid-containing peptides from C-terminally activated
Due to its biological and commercial relevance, we
b-amino acid building blocks.[14] In another study with chose l-carnosine (H-b-Ala-His-OH) as a target com-
the lipases CAL-A and CAL-B from Candida antarc- pound to investigate b-aminopeptidase-catalyzed pep-
tica and with Burkholderia cepacia lipase PS-D, sever- tide couplings in aqueous solutions in more detail. As
al b-dipeptides and derivatives of l-carnosine were reported earlier, l-carnosine was found to be formed
successfully synthesized in organic solvents.[15]
by the b-aminopeptidases DmpA from Ochrobactrum
For the peptide couplings catalyzed by the dis- anthropi and BapA from Sphingosinicella xenopeptidi-
solved b-aminopeptidases, we employed b-amino acid lytica 3-2W4 in solution[14] and in recombinant yeast
p-nitroanilides (H-bhXaa-pNA) or b-amino acid and bacterial whole cell systems producing these en-
amides (H-bhXaa-NH2) as mildly activated acyl zymes.[17] In the present investigation, we optimized
donors. Upon release of the leaving group (p-nitroani- the b-aminopeptidase-catalyzed formation of l-carno-
line or ammonia, respectively) they form a character- sine by applying the kinetic model presented in Eq.
istic acyl enzyme complex (Scheme 1). In this kineti- (2). We evaluated the synthesis reaction in terms of
408
asc.wiley-vch.de
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 407 – 415

Kinetic Analysis of l-Carnosine Formation by b-Aminopeptidases
product yields and kinetic parameters at different acyl concentration of acyl acceptor the initial synthesis/hy-
acceptor concentrations. In order to reduce the drolysis ratio reaches a maximum value of 1/g.[18b]
amount of peptidic by-products a second approach
was followed employing a C-terminally modified acyl
acceptor.
Results and Discussion
The determination of nucleophile reactivity in b-
aminopeptidase-catalyzed peptide couplings is more
complex, because besides the hydrolysis of the acyl
donor the dimerization of the acyl donor occurs as a
second undesired side reaction that has to be consid-
Nucleophile Reactivity in b-Aminopeptidase-
Catalyzed l-Carnosine Formation
To optimize the product yield of b-aminopeptidase- ered (Scheme 1). Hence, we define the nucleophile
catalyzed peptide formation and to compare the syn- reactivity of the acyl acceptor for b-aminopeptidase-
thetic performances of DmpA and BapA we adopted catalyzed peptide synthesis as the ratio between the
a kinetic model for acyl group transfer by proteases initial reaction rate of product formation (vS) and the
that form an acyl enzyme intermediate.[18] In previous sum of the initial reaction rates for acyl donor hydrol-
studies, this model was successfully applied for a ki- ysis (vH) and dimer formation (vD) [Eq. (2)].
netic evaluation of the biosynthesis of b-lactam antibi-
otics by penicillin acylase.[19] Accordingly, the synthet-
ic efficiency of peptide formation under kinetic con-
trol depends on the relative reactivity of the nucleo-
phile and is therefore referred to as “nucleophile re-
activity”. In the case of penicillin acylase-catalyzed
To compare the synthetic efficiency of l-carnosine
acyl transfer, the nucleophile reactivity of the acyl ac- formation by the b-aminopeptidases DmpA from
ceptor is defined as the ratio between the initial rate Ochrobactrum anthropi and BapA from Sphingosini-
of product synthesis (vS) and acyl donor hydrolysis cella xenopeptidilytica 3-2W4 we determined the nu-
(vH). Improved ratios of (vS/vH) can be directly corre- cleophile reactivity of the acyl acceptor l-histidine at
lated with increased product yields. The synthetic effi- two different pH values. Due to easy detectability of
ciency of the coupling reaction is quantitatively de- substrates and products we employed the chromogen-
scribed by the reaction parameters b0 and g, and hy- ic p-nitroanilide derivative of b-alanine (H-b-Ala-
perbolically depends on the initial concentration of pNA) as activated acyl donor (Scheme 2). We mea-
the employed nucleophile [Nu] [Eq. (1)]. The param- sured the initial rates of l-carnosine formation as well
eter b0 represents the preference of the acyl enzyme as the rates of acyl donor hydrolysis and dimer forma-
to react with the nucleophilic acyl acceptor instead of tion at different concentrations of l-histidine, which
water and should be high for good synthetic efficien- served as acyl acceptor. The synthetic efficiency of b-
cy.[18a] The introduction of the parameter g takes into aminopeptidase-catalyzed carnosine formation was
account that upon attack of the acyl acceptor on the quantified as the ratio [vS/ACHTUNGRTENUNG(vH+vD)]ini (Figure 1).
acyl enzyme an acyl enzyme-nucleophile complex is Table 1 summarizes the values for the reaction param-
formed, which is still prone to hydrolysis. g character- eters b0 and g that were obtained by fitting experi-
izes the intensity of hydrolysis of the acyl enzyme-nu- mental data to Eq. (2).
cleophile complex. In contrast to b0, g should be low
As expected from our previously published results,
for good synthetic efficiency as under a saturating DmpA catalyzed the formation of l-carnosine with
much higher initial rates than BapA.[14] Despite the
Scheme 2. Formation of l-carnosine from H-b-Ala-pNA or H-b-Ala-NH2 (acyl donor) and l-histidine (acyl acceptor) cata-
lyzed by b-aminopeptidases. Theoretical pKa values of the protonated amino functionalities are given in frames.
Adv. Synth. Catal. 2010, 352, 407 – 415
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
409

Tobias Heck et al.
FULL PAPERS
uisite for reaching high product titers and (ii) the acyl
acceptor has to be used in excess in order to favor
product formation over acyl donor hydrolysis and di-
merization.
l-Carnosine Formation from H-b-Ala-NH2 and l-
Histidine
In order to avoid the ecologically harmful leaving
group p-nitroaniline and to enhance the solubility of
the acyl donor we employed b-alanine amide (H-b-
Ala-NH2) instead of H-b-Ala-pNA for all further ex-
periments. As illustrated before, the synthetic per-
formances of DmpA and BapA could be greatly im-
proved by employing high nucleophile concentrations
in a basic reaction system. Hence, we investigated the
enzyme-catalyzed formation of l-carnosine from vary-
ing concentrations of H-b-Ala-NH2 and four concen-
trations of l-histidine (15, 30, 60 and 120 mM) in a re-
action system at pH 10. l-Histidine was always used
in an at least 1.25-fold molar excess over the activated
acyl donor and its concentration was assumed to be
constant during the initial phase of product formation.
We determined the initial synthetic rates as well as
the maximum accumulation of l-carnosine in the re-
action mixture (Figure 2). DmpA catalyzed the for-
mation of l-carnosine with high specific activities of
up to 16 mmol minÀ1 per mg of protein, whereas the
Figure 1. Nucleophile reactivity of l-histidine in the forma-
tion of l-carnosine from H-b-Ala-pNA (6 mM) and varying
concentrations of l-histidine by the b-aminopeptidases
DmpA (circles) and BapA (squares). Reactions were car-
ried out at 378C and the pH values of the reaction systems
were set to pH 8 (filled symbols) or pH 10 (open symbols).
Experimental data were fitted to Eq. (2).
Table 1. Reaction parameters for b-aminopeptidase-cata-
lyzed l-carnosine formation from H-b-Ala-pNA (6 mM) and
l-histidine (6 to 60 mM) obtained by fitting experimental
data to Eq. (2).
Enzyme
DmpA
pH
b0 [MÀ1]
g
catalytic
rates
of
BapA
remained
below
8
10
8
0.72
1.9
4.6
13
0.67
0.24
0.37
0.11
0.5 mmol minÀ1 per mg of protein. As expected in a ki-
netically controlled reaction, the coupling product ac-
cumulated to a maximum and was subsequently hy-
drolyzed by the enzymes. The maximum achieved
concentrations of l-carnosine strongly depended on
the employed excess of l-histidine over H-b-Ala-NH2.
BapA
10
low overall activities for l-carnosine formation, the We observed maximum l-carnosine yields of 40 and
synthetic performance [vS/A(vH+vD)]ini of the BapA-cat- 53% with respect to the initial concentration of the
CTHUNGTRENNUNG
alyzed coupling reaction was superior to that of reaction rate-limiting acyl donor with the enzymes
DmpA (Figure 1). For both enzymes we observed a DmpA and BapA, respectively. The yield was sub-
drastic improvement of the synthetic performance at stantially lower when the concentration of the nucleo-
elevated nucleophile concentrations. Furthermore, the phile was decreased. This result is in agreement with
choice of the pH value largely influenced the nucleo- our observations on the nucleophile reactivity of l-
phile reactivity of the acyl acceptor. An increase in histidine in the b-aminopeptidase-catalyzed coupling
pH of the reaction system from pH 8 to pH 10 kept reactions using the p-nitroanilide derivative of b-ala-
the amino group of l-histidine predominantly in its nine as acyl donor instead of b-alanine amide
deprotonated state and thus improved the synthetic (Figure 1). Although l-carnosine is formed by DmpA
performance of both catalysts. This led to a 2.6- and with much higher initial rates than it is formed by
2.8-fold increase of the reaction parameter b0 for BapA, the latter is the more efficient catalyst for the
DmpA and BapA, respectively, and to a 2.8- and 3.4- accumulation of higher product concentrations in the
fold decrease of g, which indicates a more efficient reaction mixtures.
formation of l-carnosine at pH 10 [Eq. (2) and
In order to estimate the kinetic reaction parameters
Table 1]. It is important to mention that the highly al- Km and Vmax of the enzymatic formation of l-carno-
kaline pH of the reaction system did not affect the sine, the data sets obtained for both enzymes in the
stability of the peptide product and the enzymes over presence of an excess of l-histidine were fitted to Mi-
the reaction period. From these experiments we con- chaelis–Menten models [Eq. (3) and Eq. (4)]. l-Car-
clude that (i) an alkaline reaction system is a prereq- nosine formation catalyzed by BapA at l-histidine
410
asc.wiley-vch.de
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 407 – 415

Kinetic Analysis of l-Carnosine Formation by b-Aminopeptidases
*
Figure 2. Initial rates of l-carnosine formation from H-b-Ala-NH2 in the presence of excess concentrations of l-histidine (
&
*
15 mM, 30 mM, & 60 mM, 120 mM) catalyzed by the b-aminopeptidases DmpA and BapA at 378C and pH 10. Experi-
mental data were fitted to Eq. (3) or Eq. (4), respectively (see also Table 2). Yields are given in percent of l-carnosine pro-
duced per initial H-b-Ala-NH2 supplied.
concentrations of 15 and 30 mM followed the classical kinetic reaction parameters for b-aminopeptidase-cat-
Michaelis–Menten model [Eq. (3)], whereas the other alyzed l-carnosine formation are summarized in
data sets showed decreased activities at elevated con- Table 2.
centrations of H-b-Ala-NH2 and hence were fitted to
The calculated Km values for the acyl donor dif-
the substrate surplus inhibition model [Eq. (4)]. The fered significantly among the enzymes. DmpA had a
high affinity for H-b-Ala-NH2 with Km values below
1 mM, whereas the Km values of BapA were
ꢀ11.6 mM (Table 2). The catalytic rates for the con-
version of H-b-Ala-NH2 and l-histidine to l-carnosine
strongly depended on the employed excess of l-histi-
dine. With DmpA the calculated Vmax values for l-car-
nosine formation ranged between 7.4 mmol minÀ1 per
mg of protein (kcat =5.0 sÀ1) and 19.2 mmol minÀ1 per
mg of protein (kcat =12.9 sÀ1), whereas the catalytic
Table 2. Kinetic parameters of l-carnosine formation by DmpA and BapA determined at varying concentrations of H-b-
Ala-NH2 and excess of l-histidine. Experimental data were fitted to the Michaelis–Menten equation with substrate surplus
inhibition [Eq. (4)], except curves for BapA at 15 and 30 mMl-histidine that followed the classical Michaelis–Menten model
[Eq. (3)]. n.d. not determined.
DmpA
BapA
l-Histidine Km
KS
[mM]
Vmax [mmol minÀ1 kcat
Km
[mM]
KS
[mM]
Vmax [mmol minÀ1 kcat
[mM]
[mM]
per mg of protein]
A
per mg of protein] ACTHNGUTERNNUG
[sÀ1]
15
30
60
120
0.98Æ0.13 15.4Æ2.9 7.4Æ0.48
0.42Æ0.06 28.7Æ4.9 10.6Æ0.50
0.45Æ0.07 29.4Æ4.0 16.7Æ0.72
0.48Æ0.05 56.4Æ7.0 19.2Æ0.59
5.0Æ0.32 11.6Æ2.9 n.d.
7.1Æ0.34 18.4Æ0.95 n.d.
0.086Æ0.01
0.27Æ0.01
0.055Æ0.01
0.17Æ0.01
0.45Æ0.12
0.87Æ0.23
11.2Æ0.48 19.9Æ7.7 45.8Æ25.1 0.70Æ0.19
12.9Æ0.40 23.0Æ8.0 32.2Æ14.4 1.4Æ0.35
Adv. Synth. Catal. 2010, 352, 407 – 415
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
411

Tobias Heck et al.
FULL PAPERS
rates of BapA remained below 1.4 mmol minÀ1 per mg protonated. The enzyme-catalyzed formation of l-car-
of protein (kcat =0.87 sÀ1). When we used initial l-his- nosine methyl ester was subsequently followed by hy-
tidine concentrations of 15 and 30 mM and acyl donor drolysis of the ester bond under alkaline conditions
concentrations below 12 and 24 mM, respectively, no and heating to yield the final product l-carnosine
substrate surplus inhibition was observed for BapA (Scheme 3). The stability of the peptide bond of l-car-
(Figure 2). Despite the high synthetic rates in the nosine was not affected under these conditions.
DmpA-catalyzed reaction, l-carnosine did not accu-
In a reaction mixture containing 6 mM H-b-Ala-
mulate to yields higher than 40% due to competing NH2 and 30 mM l-histidine methyl ester DmpA cata-
hydrolysis and coupling reactions. Higher product lyzed the formation of the methyl ester derivative of
titers could be reached in the BapA-catalyzed reac- l-carnosine, which was chemically hydrolyzed under
tions although the enzyme catalyzed the formation of alkaline conditions to the free peptide product l-car-
l-carnosine much more slowly than DmpA. This nosine. We observed a four-fold increase in the cata-
result is supported by the higher nucleophile reactivi- lytic rates of l-carnosine formation when the pH of
ty of l-histidine observed in the BapA-catalyzed reac- the reaction system was raised from pH 7 (2.8 mmol
tion (see Figure 1), which may be caused by better ac- minÀ1 per mg of protein) to pH 8 (11.9 mmol minÀ1
cessibility of the enzymeꢃs active site for the nucleo- per mg of protein). l-Carnosine methyl ester tempo-
phile.
rarily accumulated to maximum concentrations of
0.4 mM (7% yield of l-carnosie after alkaline hydrol-
ysis) in the reaction system at pH 7 and to 1.2 mM
(20% yield of l-carnosine) at pH 8 before enzymatic
hydrolysis of the product outbalanced its formation
(Figure 3). With l-histidine as acyl acceptor at pH 10
l-Carnosine Formation from H-b-Ala-NH2 and l-
Histidine Methyl Ester
In order to reduce the amount of by-products we fol- we obtained a specific activity for l-carnosine forma-
lowed a second approach for the DmpA-catalyzed tion of 7.8 mmol minÀ1 per mg of protein and a maxi-
formation of l-carnosine using l-histidine methyl mum accumulation of l-carnosine of 1.1 mM (18%
ester instead of free l-histidine as acyl acceptor. yield). Product titers could be further increased by
Esterification of the carboxyl terminus increased its feeding of an additional 6 mM H-b-Ala-NH2 to the
electron-attracting effect and reduced the pKa value reaction mixtures. The HPLC chromatograms of the
of the amino group by more than two log-scales analyzed samples reveal that several by-products
(from pKa 9.25 to 6.97). Hence, it was possible to were produced during the formation of l-carnosine
lower the pH of the reaction system from pH 10 to (Figure 3, bottom row). All observed by-products
pH 7 or 8, respectively. Under these conditions the were of peptidic nature and were formed by enzyme-
nucleophilic character of the acyl acceptor was main- catalyzed couplings between acyl donor molecules
tained while at the same time the amino termini of (oligomerization) and by the N-terminal extension of
the acyl donor and the peptide product were mostly the product l-carnosine leading to H-b-Ala-b-Ala-l-
Scheme 3. Formation of l-carnosine from H-b-Ala-NH2 (acyl donor) and l-histidine methyl ester (acyl acceptor) catalyzed
by DmpA. Theoretical pKa values of the protonated amino functionalities are given in frames.
412
asc.wiley-vch.de
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 407 – 415

Kinetic Analysis of l-Carnosine Formation by b-Aminopeptidases
*
Figure 3. DmpA-catalyzed formation of l-carnosine ( ) from H-b-Ala-NH2 (6 mM) and l-histidine methyl ester (30 mM) in
300 mM MOPS buffer at pH 7 (A) and in 300 mM Tris/HCl buffer at pH 8 (B) afteer subsequent hydrolysis of the ester
bond (see Scheme 3) and from H-b-Ala-NH2 (6 mM) and l-histidine (30 mM) in 100 mM sodium carbonate buffer at pH 10
*
(C) (see Scheme 3). After 1 hour an additional 6 mM H-b-Ala-NH2 was fed to the reaction mixtures ( ). The lower dia-
grams show the corresponding HPLC diagrams at the point of maximum product titer in a batch synthesis (A at 2 h, B at
1.5 h, C at 2 h). Peaks are normalized to l-carnosine (100%). The observed by-product peaks correspond to 1 H-(b-Ala)2-
NH2, 2 H-(b-Ala)2-OH, 3 H-(b-Ala)3-NH2, 4 H-(b-Ala)2-l-His-OH (see Table 3).
Table 3. HPLC-MS analysis of the by-products formed
during the formation of l-carnosine. Peak numbers corre-
spond to the labels in Figure 3.
around 9.1 (Scheme 2) and also act as potent nucleo-
philes at pH 10, which leads to formation of the unde-
sired coupling by-products. In contrast, the amino ter-
mini of H-b-Ala-NH2 and l-carnosine are mostly in
the protonated state when the pH of the reaction
system is lowered to pH 8 and their ability to react as
nucleophiles is drastically reduced. However, l-histi-
dine methyl ester (pKa 6.97), in contrast to l-histidine
(pKa 9.12), still functions as a nucleophile at pH 8 to
accept the acyl moiety from the active site of the
enzyme, which leads to formation of the peptide
product.
Peak
RT
[min]
Calcd.
mass
[M+H]+
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
l-carnosine
3.9
3.3
4.6
5.7
6.4
226.1
159.1
160.1
230.1
297.1
227.1
160.0
161.1
231.3
298.2
H-(b-Ala)2-NH2
H-(b-Ala)2-OH
H-(b-Ala)3-NH2
H-(b-Ala)2-l-His-OH
1
2
3
4
His-OH (Table 3). The hydrolysis of the acyl donor,
which leads to a significant loss of substrate during Conclusions
the reaction and hence to reduced product yields,
could not be monitored because H-b-Ala-OH was not In the present study, the kinetics of b-aminopepti-
detectable with the employed HPLC method. Al- dase-catalyzed l-carnosine formation from C-termi-
though similar yields of l-carnosine were obtained in nally modified b-alanine derivatives and l-histidine or
the reaction systems at pH 8 and 10, the accumulation l-histidine methyl ester were investigated in aqueous
of peptidic by-products was considerably reduced solution. In this kinetically controlled reaction the
when l-histidine methyl ester was used as the acyl ac- transfer of the acyl moiety from the acyl enzyme to
ceptor at pH 8 instead of l-histidine at pH 10 competing nucleophiles led to the formation of unde-
(Figure 3). This observation can be explained by the sired by-products. The efficiency of l-carnosine for-
fact that the free amino groups of H-b-Ala-NH2 and mation was strongly pH-dependent and the best re-
the product l-carnosine have similar pKa values sults were obtained at pH 10. DmpA catalyzed the
Adv. Synth. Catal. 2010, 352, 407 – 415
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
413

Tobias Heck et al.
FULL PAPERS
ments were performed at 595 nm with a Specord S 100 spec-
trophotometer (Analytik Jena, Jena, Germany). For the cal-
culation of pKa values we used the calculator plugins of the
software Marvin 5.2 (http://www. chemaxon.com).
formation of l-carnosine with synthetic rates. In con-
trast, BapA exhibited much lower rates, but it was the
catalyst with the higher synthetic performance, which
led to the temporary accumulation of product yields
above 50%. The formation of peptidic by-products
could be reduced when l-histidine methyl ester was
used as acyl acceptor in a reaction system at pH 8,
Enzyme Preparation
The enzymes DmpA from Ochrobactrum anthropi
rather than free l-histidine at pH 10. Our approach to LMG7991 and BapA from Sphingosinicella xenopeptidilytica
3-2W4 were recombinantly expressed in E. coli and purified
according to published procedures.[11c,12] The lyophilized
enzyme powders were dissolved in a 10 mM Tris/HCl buffer
at pH 8 and the protein concentrations of the enzyme stock
solutions were determined spectrophotometrically.
enhance the nucelophilic character of the acceptor
amino acid by modifying its caboxy terminus could
generally serve as a useful strategy to reduce the
amount of unwanted by-products in enzyme-catalyzed
peptide couplings.
The formation of l-carnosine can be regarded as a
model reaction for the b-aminopeptidase-catalyzed
synthesis of peptides with N-terminal b-amino acids.
Due to its broad substrate specificity for many N-ter-
minal b-amino acids,[12,14] we think that especially
BapA could become a versatile catalyst for the forma-
tion of various b-amino acid-containing peptides.
Standard Enzyme Assay for the Formation of l-
Carnosine from H-b-Ala-pNA or H-b-Ala-NH2 and
l-Histidine
The reaction mixtures contained H-b-Ala-pNA or H-b-Ala-
NH2 and l-histidine in different molar ratios in a 100 mM
sodium carbonate buffer at pH 10. Limiting amounts of
either DmpA or BapA were added to initiate the reactions
at 378C. Samples were withdrawn at intervals from the mix-
tures and the enzymatic reaction was quenched by the addi-
tion of 25% (v/v) 1M HCl. The formation of l-carnosine
was analyzed by HPLC, and by-products of the reaction
were identified by LC/MS.
Experimental Section
General Remarks
l-Histidine, l-histidine methyl ester (H-l-His-OMe·2HCl),
l-carnosine and p-nitroaniline were obtained from Sigma–
Aldrich (Buchs, Switzerland), H-b-Ala-pNA·HBr from
Bachem (Bubendorf, Switzerland) and H-b-Ala-NH2·HCl
from Advanced ChemTech (Louisville, KY, USA). The ref-
erence peptides H-(b-Ala)2-NH2, H-(b-Ala)2-OH, H-(b-
Ala)3-NH2 and H-(b-Ala)2-l-His-OH were synthesized by
solid-phase peptide synthesis and were gifts from EMC mi-
crocollections GmbH (Tꢁbingen, Germany). l-Carnosine
and peptide products were analyzed by reversed-phase chro-
matography on a Dionex HPLC system equipped with a
P680 pump, an ASI-100 automated sample injector, an Ulti-
Mate 3000 thermostated column compartment and a UVD
340U photodiode array detector (Dionex, Sunnyvale, CA,
USA). All compounds were separated on a Nucleodur C18
Pyramid stationary phase (250ꢄ4 mm, 5 mm particle size;
Macherey–Nagel, Dꢁren, Germany), which was equilibrated
with 0.1% trifluoroacetic acid (TFA) in water at a column
temperature of 108C. l-Carnosine and peptidic by-products
of the reaction were separated with an isocratic flow of
0.1% TFA at a constant flow rate of 1 mLminÀ1. Product
formation was detected at 205 nm and quantified according
to a standard curve of l-carnosine. All p-nitroanilide deriva-
tives and free p-nitroaniline were separated with a linear
gradient of 0 to 40% acetonitrile and detected at wave-
lengths of 318 and 383 nm, respectively. Quantification of p-
nitroanilide derivatives and of free p-nitroaniline was
achieved on the basis of standard curves of H-b-Ala-pNA
and p-nitroaniline, respectively. Mass spectra of all com-
pounds were recorded with an API 4000 liquid chromatog-
raphy/tandem MS system connected to an Agilent 1100 LC
system. For protein determination, we used five-fold concen-
trated Bradford reagent (Bio-Rad, CH-Reinach) and bovine
serum albumin (BSA) as a standard; absorbance measure-
Standard Enzyme Assay for the Formation of l-
Carnosine from H-b-Ala-NH2 and l-Histidine Methyl
Ester
The reaction mixtures contained H-b-Ala-NH2 and l-histi-
dine methyl ester in different molar ratios in a 300 mM Tris/
HCl buffer at pH 8. DmpA was added to initiate the reac-
tion at 378C and samples were withdrawn regularly from
the mixtures. The addition of NaOH to a final concentration
of 0.5M (pH 12) and heating of the samples at 958 for 7 min
stopped the enzymatic reaction and led to complete hydrol-
ysis of all ester bonds. Before HPLC and LC/MS analysis,
samples were acidified by adding HCl to a final concentra-
tion of 0.5M (pH 2).
Determination of Kinetic Parameters
The initial rates of DmpA- and BapA-catalyzed l-carnosine
formation at different ratios of H-b-Ala-NH2 and l-histidine
were calculated from the experimental data obtained by
HPLC measurements. Experimental data were fitted to Mi-
chaelis–Menten models without [Eq. (3)] and with [Eq. (4)]
substrate surplus inhibition by non-linear regression analysis
with the VisualEnzymics software (Softzymics, Princeton,
NJ, USA) for the program IGOR Pro (WaveMetrics,
Oswego, OR, USA).
Acknowledgements
The authors are grateful to Donald Hilvert (ETH Zꢀrich) for
very helpful advice and to Karl-Heinz Wiesmꢀller and
Renate Spohn (EMC microcollections GmbH) for providing
414
asc.wiley-vch.de
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 407 – 415

Kinetic Analysis of l-Carnosine Formation by b-Aminopeptidases
reference peptides for HPLC analysis. This work was finan-
cially supported by the Swiss National Science Foundation
(SNF Project No. 3152A0-100770) and the Deutsche Bundes-
stiftung Umwelt (DBU Project No. 13176–32).
Bartram, P. Nawroth, H.-P. Hammes, B. A. Yard, J.
Zschocke, F. J. van der Woude, Diabetes 2005, 54,
2320–2327; c) I. Rashid, D. M. van Reyk, M. J. Davies,
FEBS Lett. 2007, 581, 1067–1070.
[7] a) T. Matsukura, T. Takahashi, Y. Nishimura, M.
Takaya, H. Fujimura, (Hamari Chemicals Company),
US Patent 4,981,846, 1991; b) M. Hirano, S. Katayama,
(Zeria Pharmaceutical Company, Hamari Chemicals
Company), US Patent 6,169,083 B1, 2001.
[8] a) F. J. Vinick, (Pfizer Inc.), US Patent 4,359,416, 1982;
b) S. Hildbrand, T. Ruppen, D. Veghini, (Lonza AG),
US Patent 6,878,829 B2, 2005.
[9] a) F. Bordusa, Chem. Rev. 2002, 102, 4817–4867; b) K.
Faber, Biotransformations in organic chemistry: a text-
book, 5th edn., Springer, Berlin, 2004, pp 370–377;
c) C. Lombard, J. Saulnier, J. M. Wallach, Protein Pept.
Lett. 2005, 12, 621–629; d) D. Kumar, T. C. Bhalla,
Appl. Microbiol. Biotechnol. 2005, 68, 726–736; e) M.
Yagasaki, S. Hashimoto, Appl. Microbiol. Biotechnol.
2008, 81, 13–22.
References
[1] a) J. Frackenpohl, P. I. Arvidsson, J. V. Schreiber, D.
Seebach, ChemBioChem 2001, 2, 445–455; b) J. A.
Patch, A. E. Barron, Curr. Opin. Chem. Biol. 2002, 6,
872–877; c) D. L. Steer, R. A. Lew, P. Perlmutter, A. I.
Smith, M.-I. Aguilar, Curr. Med. Chem. 2002, 9, 811–
822; d) H. Wiegand, B. Wirz, A. Schweitzer, G. P. Ca-
menisch, M. I. Rodriguez Perez, G. Gross, R. Woessner,
R. Voges, P. I. Arvidsson, J. Frackenpohl, D. Seebach,
Biopharm. Drug Dispos. 2002, 23, 251–262; e) H. Wie-
gand, B. Wirz, A. Schweitzer, G. Gross, M. I. Rodriguez
Perez, H. Andres, T. Kimmerlin, M. Rueping, D. See-
bach, Chem. Biodiversity 2004, 1, 1812–1828; f) A. I.
Webb, M. A. Dunstone, N. A. Williamson, J. D. Price,
A. de Kauwe, W. Chen, A. Oakley, P. Perlmutter, J.
McCluskey, M.-I. Aguilar, J. Rossjohn, A. W. Purcell, J.
Immunol. 2005, 175, 3810–3818; g) M.-I. Aguilar,
A. W. Purcell, R. Devi, R. Lew, J. Rossjohn, A. I.
Smith, P. Perlmutter, Org. Biomol. Chem. 2007, 5,
2884–2890; h) D. Seebach, J. Gardiner, Acc. Chem.
Res. 2008, 41, 1366–1375.
[10] J. A. Brannigan, G. Dodson, H. J. Duggleby, P. C. E.
Moody, J. L. Smith, D. R. Tomchick, A. G. Murzin,
Nature 1995, 378, 416–419.
[11] a) L. Fanuel, C. Goffin, A. Cheggour, B. Devreese, G.
van Driessche, B. Joris, J. van Beeumen, J.-M. Frꢅre,
Biochem. J. 1999, 341, 147–155; b) B. Geueke, K.
Namoto, D. Seebach, H.-P. E. Kohler, J. Bacteriol. 2005,
187, 5910–5917; c) B. Geueke, T. Heck, M. Limbach,
V. Nesatyy, D. Seebach, H.-P. E. Kohler, FEBS J. 2006,
273, 5261–5272; d) B. Geueke, H.-P. E. Kohler, Appl.
Microbiol. Biotechnol. 2007, 74, 1197–1204.
[12] T. Heck. M. Limbach, B. Geueke, M. Zacharias, J.
Gardiner, H.-P. E. Kohler, D. Seebach, Chem. Biodi-
versity 2006, 3, 1325–1348.
[13] T. Heck, D. Seebach, S. Osswald, M. K. J. ter Wiel, H.-
P. E. Kohler, ChemBioChem 2009, 10, 1558–1561.
[14] T. Heck, H.-P. E. Kohler, M. Limbach, O. Flçgel, D.
Seebach, B. Geueke, Chem. Biodiversity 2007, 4, 2016–
2030.
[15] a) X.-G. Li, L. T. Kanerva, Org. Lett. 2006, 8, 5593–
5596; b) X.-G. Li, M. Lꢆhitie, L. T. Kanerva, Tetrahe-
dron: Asymmetry 2008, 19, 1857–1861; c) P. D’Arrigo,
L. T. Kanerva, X.-G. Li, C. Saraceno, S. Servi, D. Tes-
saro, Tetrahedron: Asymmetry 2009, 20, 1641–1645.
[16] a) V. Kasche, Enzyme Microb. Technol. 1986, 8, 4–16;
b) V. Schellenberger, H.-D. Jakubke, Angew. Chem.
1991, 103, 1440–1452; Angew. Chem. Int. Ed. Engl.
1991, 30, 1437–1449.
[2] D. Seebach, A. K. Beck, D. J. Bierbaum, Chem. Biodi-
versity 2004, 1, 1111–1239.
[3] a) G. D. Kalyankar, A. Meister, J. Biol. Chem. 1959,
234, 3210–3218; b) R. E. Winnick, T. Winnick, Bio-
chim. Biophys. Acta 1960, 37, 214–220; c) S. D. Skaper,
S. Das, F. D. Marshall, J. Neurochem. 1973, 21, 1429–
1445; d) H. Horinishi, M. Grillo, F. L. Margolis, J. Neu-
rochem. 1978, 31, 909–919.
[4] a) A. Pegova, H. Abe, A. Boldyrev, Comp. Biochem.
Physiol., Part B: Biochem. Mol. Biol. 2000, 127, 443–
446; b) M. Teufel, V. Saudek, J.-P. Ledig, A. Bernhardt,
S. Boularand, A. Carreau, N. J. Cairns, C. Carter, D. J.
Cowley, D. Duverger, A. J. Ganzhorn, C. Guenet, B.
Heintzelmann, V. Laucher, C. Sauvage, T. Smirnova, J.
Biol. Chem. 2003, 278, 6521–6531; c) H. Unno, T. Ya-
mashita, S. Ujita, N. Okumura, H. Otani, A. Okumura,
K. Nagai, M. Kusunoki, J. Biol. Chem. 2008, 283,
27289–27299.
[5] a) R. Kohen, Y. Yamamoto, K. C. Cundy, B. N. Ames,
Proc. Natl. Acad. Sci. USA 1988, 85, 3175–3179;
b) M. A. Babizhayev, M. C. Seguin, J. Gueyne, R. P. Ev-
stigneeva, E. A. Ageyeva, G. A. Zheltukhina, Biochem.
J. 1994, 304, 509–516; c) A. R. Hipkiss, Int. J. Biochem.
Cell Biol. 1998, 30, 863–868; d) K. Bauer, Neurochem.
Res. 2005, 30, 1339–1345; e) A. Guiotto, A. Calderan,
P. Ruzza, G. Borin, Curr. Med. Chem. 2005, 12, 2293–
2315.
[6] a) R. Tabakman, P. Lazarovici, R. Kohen, J. Neurosci.
Res. 2002, 68, 463–469; b) B. Janssen, D. Hohenadel, P.
Brinkkoetter, V. Peters, N. Rind, C. Fischer, I. Rychlik,
M. Cerna, M. Romzova, E. de Heer, H. Baelde, S. J. L.
Bakker, M. Zirie, E. Rondeau, P. Mathieson, M. A.
Saleem, J. Meyer, H. Kçppel, S. Sauerhoefer, C. R.
[17] J. Heyland, N. Antweiler, J. Lutz, T. Heck, B. Geueke,
H.-P. E. Kohler, L. M. Blank, A. Schmid, Microb. Bio-
technol. 2010, 3, 74–83.
[18] a) M. Y. Gololobov, I. L. Borisov, V. M. Belikov, V. K.
ˇ
Svedas, Biotechnol. Bioeng. 1988, 32, 866–872; b) M. Y.
ˇ
Gololobov, I. L. Borisov, V. K. Svedas, J. Theor. Biol.
1989, 140, 193–204.
[19] a) M. I. Youshko, G. G. Chilov, T. A. Shcherbakova,
ˇ
V. K. Svedas, Biochim. Biophys. Acta Proteins Proteo-
mics 2002, 1599, 134–140; b) S. A. W. Jager, I. V. Sha-
ˇ
povalova, P. A. Jekel, W. B. L. Alkema, V. K. Svedas,
D. B. Janssen, J. Biotechnol. 2008, 133, 18–26.
Adv. Synth. Catal. 2010, 352, 407 – 415
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
415