Stereochemistry of the reaction of the inhibitor β-chloroalanine with mercaptoethanol, a β-substitution reaction catalysed by an aminotransferase

Article abstract of DOI:10.1039/b000442l

L-Aspartate aminotransferase, a member of the α-family of PLP mediated enzymes, which normally catalyses transamination, has been used to catalyse the β-substitution reaction of stereospecifically labelled samples of the enzyme inhibitor β-chloro-L-alanin

Full text of DOI:10.1039/b000442l

Stereochemistry of the reaction of the inhibitor b-chloroalanine with
mercaptoethanol, a b-substitution reaction catalysed by an aminotransferase
Benjamin Adams, Kenneth J. M. Beresford, Sheena M. Whyte and Douglas W Young*
Sussex Centre for Biomolecular Design and Drug Development, University of Sussex, Falmer, Brighton,
UK BN1 9QJ. E-mail: d.w.young@sussex.ac.uk
Received (in Liverpool, UK) 17th January 2000, Accepted 12th February 2000
L
-Aspartate aminotransferase, a member of the a-family of
PLP mediated enzymes, which normally catalyses trans-
amination, has been used to catalyse the b-substitution
reaction of stereospecifically labelled samples of the enzyme
inhibitor b-chloro- -alanine with 2-mercaptoethanol; the
L
stereochemistry of the products was assigned by independ-
ent synthesis, showing that the abnormal substitution
reaction proceeds with overall retention of stereochemistry,
the usual stereochemical consequence of reactions catalysed
by enzymes of the b-family of PLP mediated enzymes which
have low homology with enzymes of the a-family.
Scheme 1
normal role this was.10 We therefore prepared samples of b-
Pyridoxal phosphate (PLP) dependent enzymes catalyse a wide
variety of different reactions, mainly involving amino acids.
Early theories1–3 to generalise and explain the chemical basis
for these widely differing reactions have largely been con-
firmed. Dunathan and Voet4 proposed that PLP-dependent
enzymes evolved from a common ancestor protein, and there is
evidence5 that this is true for one of the three families of PLP
mediated enzymes, the a-family, which contains a predominant
number of enzymes. Homology studies have recently suggested
that the a- and g-families of enzymes may be distantly related
but that enzymes of the b-family are not closely related to either
of these families.5
chloro- -alanine 1 which were stereospecifically labelled at C-
L
3. This was achieved by first converting the labelled samples of
methyl (2S)-N-tritylaziridine-2-carboxylate 7(HB
2H)11 and
7(HA
2H)11 to the corresponding urethanes 8† by reaction
=
=
with trifluoroacetic acid in chloroform to remove the trityl
group, followed immediately by reaction with benzyl chloro-
formate under Schotten Baumann conditions (Scheme 2). b-
Substitution with chloride ion was achieved by reaction with
TiCl4 in CH2Cl2–CHCl3, and deprotection in refluxing 4 M
1
2
H2SO4 then gave the enzyme inhibitors 1.† The H and H
NMR spectra indicated that the inhibitors 1 were unique
diastereoisomers and, since substitution is accompanied by
inversion of stereochemistry at the labelled b-position, these
were assigned as (2R,3S)-[3-2H1]- and (2R,3R)-[2,3-2H2]-b-
L
-Aspartate aminotransferase (EC 2.6.1.1) is an enzyme of
the a-family, catalysing the transamination of -aspartic acid to
a-ketoglutaric acid giving oxaloacetic and glutamic acids. b-
Chloro-
-alanine 16,7 has been shown not only to inhibit this
L
chloroalanine, 1(HB
=
2H) and 1(HA 2H), respectively.
=
L
enzyme, but also to be converted to pyruvic acid by it. It has also
been observed that introduction of thiosulfate during the
inactivation of
sulfate 2 reversed inhibition and caused production of
cysteine-S-sulfonate 3.8 Further, inhibition of
-amino acid
L
-aspartate aminotransferase by
L
-serine-O-
L
-
D
aminotransferase by b-bromo-
D
-alanine was reversed by addi-
Scheme 2
tion of thiols and, when b-mercaptoethylamine was used, S-(b-
1
aminoethyl)-
D-cysteine and D -thiomorpholine-2-carboxylic
The enzyme -aspartate aminotransferase (EC 2.6.1.1) was
L
acid were products.9 Thus these enzymes have catalysed b-
substitution reactions typical of enzymes of the b-family, rather
than transamination which is typical of a-family enzymes. This
would suggest that the quinonoid form 4, formed in the ‘normal’
reaction, instead of protonation at C-4A by Lys258, converted to
5 in the inhibition reaction. This may be processed to pyruvate
or an inhibitor complex in the absence of a thiol. In the presence
of a thiol, however, the reaction presumably proceeds as shown
in Scheme 1 eventually to give the final cysteine derivative 6.
The change in role from catalyst for transamination to
catalyst for b-substitution exhibited by aspartate aminotransfer-
isolated by a standard purification protocol12 from E. coli
TY10313 which was transformed with plasmid pKDHE19/
AspC.14 It had specific activity of 36.57 units mg21 and
appeared as one major band of molecular weight 43000 Da on
SDS-PAGE. Incubation with (2R,3R)-[2,3-2H2]-b-chloroala-
2
nine 1(HA = H) in the presence of a variety of thiols and under
various conditions was now undertaken. The best results were
obtained at pH 8.4 when mercaptoethanol was employed‡ and
samples of labelled 3-(2-hydroxyethyl)cysteine 9a were ob-
tained in which the a-deuterium atom had exchanged as
expected from the mechanism in Scheme 1. The b-deuterium
atom also exchanged on prolonged incubation, an effect which
had been previously noted in reactions with this enzyme.15,16
However, when the incubation was stopped after 3 h, a sample
of (2R)-3-(2-hydroxyethyl)cysteine 9a was obtained with a 1H
NMR spectrum which showed that there was stereospecific
ase on change of substrate from
L
-aspartate to b-chloro- -
L
alanine 1 was intriguing and we decided to examine the overall
stereochemistry of the b-substitution reaction in the presence of
thiol to see whether it was in keeping with the retention
generally shown by reactions catalysed by the enzymes whose
DOI: 10.1039/b000442l
Chem. Commun., 2000, 619–620
This journal is © The Royal Society of Chemistry 2000
619

ucts to be deduced. It was evident that the product from the
incubation using (2R,3S)-[3-2H1]-3-b-chloroalanine 1(HB
2H) was (2R,3S)-[3-2H1]-3-(2-hydroxyethyl)cysteine 9a(HB
=
=
2H) [Fig. 1(e)] and that the product when (2R,3R)-[2,3-2H2]-
3-b-chloroalanine 1(HB
=
2H) was used was (2R,3R)-[3-2H1]-
2H) [Fig. 1(e)].
3-(2-hydroxyethyl)cysteine 9a(HA
=
These results imply that the b-replacement reaction, which in
this case is catalysed by an enzyme of the a-family whose
normal function is transamination, occurs with overall retention
of stereochemistry. This is the general expectation10 for the
PLP-mediated enzymes of the b-family where b-replacement
reactions are the norm and may imply a closer relationship
between families than homology5 suggests. Although X-ray
structures are available for aspartate aminotransferase17 and
other a-family enzymes, tryptophan synthase (EC 4.2.1.20) is
one of the few enzymes of the b-family whose tertiary structure
has been defined by X-ray cystallography.18 It is interesting
that, in the presence of thiols, tryptophan synthase has been
shown to catalyse transamination and b-replacement of -serine
L
by mercaptoethanol.19 Further, the Lys87Thr mutant of trypto-
phan synthase will not turn over the natural substrate serine in
+
the absence of NH4 but it will turn over b-chloro-
L
-
alanine.20
We thank the BBSRC (K. J. M. B.) and the Leukaemia
Research Fund (B. A.) for fellowships and the EPSRC for a
studentship (S. M. W.), Mr C. Dadswell for NMR experiments,
the EPSRC National Mass Spectrometry Service, Swansea for
accurate mass measurements and Professor H. Kagamiyama for
E. coli TY103 and plasmid pKDHE19/AspC.
Notes and references
† These compounds had the expected analytical and spectroscopic prop-
erties.
‡ Incubation conditions: b-chloro-L-alanine: (0.08 mmol) and 2-mercapto-
ethanol (0.15 mmol) in sodium arsenate buffer (2 ml) containing 80 units of
enzyme.
Fig. 1 1H NMR spectra in 10% 2HCl–2H2O of (a) synthetic (2R)-
3-(2-hydroxyethyl)cysteine 9, (b) synthetic (2R,3S)-[3-2H1]-3-(2-hydroxy-
ethyl)cysteine 9b(HB
ethyl)cysteine 9b(HA
(2R,3R)-[2,3-2H2]-b-chloroalanine 1(HA
transferase and (e) product from incubation of (2R,3S)-[3-2H1]-b-chloro-
alanine 1(HB -aspartate aminotransferase.
2H) with
=
2H), (c) synthetic (2R,3R)-[2,3-2H2]-3-(2-hydroxy-
=
2H), (d) product from incubation of
2H) with
-aspartate amino-
=
L
1 A. E. Braunstein and M. M. Shemyakin, Biokhimya, 1953, 18, 393.
2 D. E. Metzler, M. Ikawa and E. E. Snell, J. Am. Chem. Soc., 1954, 76,
648.
3 H. C. Dunathan, Proc. Natl. Acad. Sci. USA, 1966, 55, 712.
4 H. C. Dunathan and J. G. Voet, Proc. Natl. Acad. Sci. USA, 1974, 71,
3888.
=
L
labelling at C-3 [Fig. 1(d)]. (2R,3S)-[3-2H1]-3-b-Chloroalanine
2
1(HB = H) was therefore incubated for the same time when the
1H NMR spectrum [Fig. 1(e)] of the product 9a indicated that it
was the C-3 epimer.
5 F. W. Alexander, E. Sandmeier, P. K. Mehta and P. Christen, Eur. J.
Biochem., 1994, 219, 953.
The trapping reaction was evidently stereospecific and it was
now necessary to determine the absolute stereochemistry of the
overall reaction. This was achieved by independent synthesis of
samples of (2R)-3-(2-hydroxyethyl)cysteine 9b which were
labelled stereospecifically with deuterium at C-3 in an unambi-
guous manner. This is shown in Scheme 3, the labelled
carbobenzyloxyaziridines 8 being reacted with mercaptoethanol
containing a catalytic quantity of boron trifluoride etherate.
Inversion of stereochemistry at the labelled atom, C-3, is
expected and 1H and 2H NMR spectra showed that the labelled
products† were single diastereoisomers. Hydrolysis in refluxing
4 M H2SO4 then gave the free amino acids 9b.
6 Y. Morino, A. M. Osman and M. Okamoto, J. Biol. Chem., 1974, 249,
6684.
7 Y. Morino and S. Tanase, J. Biol. Chem., 1978, 253, 252.
8 D. Cavallini, G. Federici, F. Bossa and F. Granata, Eur. J. Biochem.,
1973, 39, 301.
9 T. S. Soper and J. M. Manning, Biochemistry, 1978, 17, 3377.
10 D. W. Young, Top. Stereochem., 1994, 21, 381.
11 K. J. M. Beresford and D. W. Young, Tetrahedron, 1996, 52, 9891.
12 M. Herold and K. Kirschner, Biochemistry, 1990, 29, 1907.
13 T. Yano, S. Kuramitsu, S. Tanase, Y. Morino, K. Hiromi and H.
Kagamiyama, J. Biol. Chem., 1991, 266, 6079.
14 S. Kamitori, K. Hirotsu, T. Higuchi, K. Kondo, K. Inoue, S. Kuramitsu,
H. Kagamiyama, Y. Higuchi, N. Yasuoka, M. Kusunoki and Y.
Matsuura, J. Biochem., 1987, 101, 813.
15 U. Walter, H. Luthe, F. Gerhart and H.-D. Söling, Eur. J. Biochem.,
1975, 59, 395.
16 U. M. Babu and R. B. Johnston, Biochemistry, 1976, 15, 5671.
17 See for example: V. N. Malashkevich, M. D. Toney and J. N. Jansonius,
Biochemistry, 1993, 32, 13451.
18 T. R. Schneider, E. Gerhardt, M. Lee, P.-H. Liang, K. S. Anderson and
I. Schlichting, Biochemistry, 1998, 37, 5394.
19 E. W. Miles, M. Hatanaka and I. P. Crawford, Biochemistry, 1968, 7,
2742.
20 Z. Lu, S. Nagata, P. McPhie and E. W. Miles, J. Biol. Chem., 1993, 268,
8727.
Scheme 3
1
The H NMR spectra of the synthetic samples of (2R,3S)-
[3-2H1]-3-(2-hydroxyethyl)cysteine 9b(HB
=
2H)† [Fig. 1(b)]
and (2R,3R)-[2,3-2H2]-3-(2-hydroxyethyl)cysteine 9b(HA
=
2H)† [Fig. 1(c)] allowed the 3-pro-S and 3-pro-R protons in the
spectrum of 3-(2-hydroxyethyl)cysteine 9 to be assigned and
therefore the absolute stereochemistry of the incubation prod-
Communication b000442l
620
Chem. Commun., 2000, 619–620