Unexpected stereoselective exchange of straight-chain fatty acyl-CoA α-protons by human α-methylacyl-CoA racemase 1A (P504S)

Article abstract of DOI:10.1039/c002509g

α-Methylacyl-CoA racemase (AMACR; P504S) catalysed exchange of straight-chain fatty acyl-CoA α-protons. One α-proton was removed in each catalytic cycle, with the pro-S proton preferred. This reaction was most efficient for straight-chain substrates with longer side-chains. 2-Methyldecanoyl-CoA underwent α-proton exchange 3× more efficiently (as judged by K_cat/K_m) than decanoyl-CoA.

Full text of DOI:10.1039/c002509g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Unexpected stereoselective exchange of straight-chain fatty acyl-CoA
a-protons by human a-methylacyl-CoA racemase 1A (P504S)w
Fouzia A. Sattar, Daniel J. Darley, Francesco Politano, Timothy J. Woodman,
Michael D. Threadgill and Matthew D. Lloyd*
Received (in Cambridge, UK) 5th February 2010, Accepted 9th March 2010
First published as an Advance Article on the web 30th March 2010
DOI: 10.1039/c002509g
a-Methylacyl-CoA racemase (AMACR;z P504S) catalysed
exchange of straight-chain fatty acyl-CoA a-protons. One
a-proton was removed in each catalytic cycle, with the pro-S
proton preferred. This reaction was most efficient for straight-
chain substrates with longer side-chains. 2-Methyldecanoyl-CoA
underwent a-proton exchange 3ꢀ more efficiently (as judged by
based on the structure of the Mycobacterium tuberculosis
homologue, MCR.8 The enzyme is an atypical two-base
racemase, in that it incorporates a high level of deuterium
into product in both directions.1 This type of behaviour is
more typical of one-base racemases. AMACR protein
levels are increased in prostate cancer and it has attracted
recent attention as a marker (P504S)9,10 and potential drug
target.11–13
K
cat/Km) than decanoyl-CoA.
Straight-chain fatty acyl-CoA esters are abundant in both
peroxisomes and mitochondria, the compartments in which
AMACR is localised. The strong resemblance to the natural
substrates suggests that they might be potential substrates of
AMACR. This communication reports that exchange of the
a-protons of straight-chain fatty acyl-CoA esters is indeed
catalysed by human AMACR 1A (Scheme 2). Specific
questions addressed in this communication are: (1) Is one or
are both a-protons exchanged during each catalytic event? (2)
If only one a-proton is exchanged, is the reaction stereo-
selective? (3) Is the presence of substrate straight-chain
fatty acyl-CoA esters likely to interfere with or modulate
‘‘conventional’’ racemase activity of AMACR in vivo?
Initially, decanoyl-CoA 2 was incubated with recombinant
human AMACR 1A in buffer containing 2H2O. Decanoyl-
CoA 2 was chosen as this is the straight-chain analogue of
the previously reported S- and R-2-methyldecanoyl-CoA
substrates (1S and 1R).1 Incubations of 2 with active enzyme
resulted in a reduction in the peak intensity of the triplet for
Branched-chain fatty acids, e.g. phytanic acid (3R/S,7R,11R,
15-tetramethylhexadecanoic acid), are important components
of the human diet and are also used as drugs e.g. ibuprofen.1,2
Phytanic acid is derived from the phytol side chain of
chlorophyll A and is abundant in red meat and dairy
products.3 High amounts of phytanic acid in the diet are
a risk factor for prostate cancer,4,5 the most common
male-specific cancer. The presence of the 3-methyl group in
phytanic acid prevents b-oxidation, and it is processed as its
CoA ester by peroxisomal a-oxidation to give pristanic acid
(2R/S,6R,10R,14-tetramethylpentadecanoic acid). Pristanic
acid is metabolised as its CoA ester by b-oxidation in
peroxisomes and subsequently in mitochondria for chain-
shortened derivatives.2,3,6 The b-oxidation pathway only oxidises
a-methyl fatty acyl-CoA esters with 2S configuration,2,7 but
2R-methylacyl-CoA esters are produced during the degradation
of phytanic acid and other endogenous fatty acids.
Chiral inversion of these R-2-methylacyl-CoA esters is
catalysed by the enzyme a-methylacyl-CoA racemase
(AMACR) (Scheme 1),1,2 and proceeds by removal of an
a-proton to give an enol/enolate intermediate followed by
non-stereoselective reprotonation.1 The enzyme has two bases
within the active site, the His-122/Glu-237 pair and Asp-156,
1
the a-protons at d 2.36–2.46 in the H NMR spectrum, along
with changes in the structure of the b-proton quintet at d 1.40.
Levels of conversion of ca. 35–40% were observed after
incubation for 16 h. Incubation of 2,20-[2H2]-decanoyl-CoA
3 (R1, R2 = 2H) with active AMACR in 1H2O buffer resulted
in the appearance of a triplet in the 1H NMR spectrum for the
a-protons and changes in the b-protons signal from a triplet
into a quintet. High levels of conversion were observed,
indicating that the a-1H 2 a-2H exchange reactions are
efficiently catalysed by AMACR in both directions. These
Scheme
1
Reaction catalysed by a-methylacyl-CoA racemase
(AMACR), using R- and S-2-methyldecanoyl-CoA 1R, 1S as example
substrates.
Medicinal Chemistry, Department of Pharmacy & Pharmacology,
University of Bath, Claverton Down, Bath BA2 7AY, UK.
E-mail: M.D.Lloyd@bath.ac.uk; Fax: +44 (0)1225-386114;
Tel: +44 (0)1225-386786
w Electronic supplementary information (ESI) available: Synthetic
methods, compound characterisation data and enzyme kinetic plots.
See DOI: 10.1039/c002509g
Scheme 2 Incorporation of deuterium during reaction of straight-
chain fatty acyl-CoA esters (decanoyl-CoA 2) with AMACR
1
2
(R1, R2 = H or H).
ꢁc
This journal is The Royal Society of Chemistry 2010
3348 | Chem. Commun., 2010, 46, 3348–3350

View Article Online
Fig. 2 1H NMR spectrum of monodeutero methyl O-decanoyl
R-mandelates 6.
Fig. 1 13C NMR spectrum of 2-[13C]-decanoyl-CoA 4 incubated with
AMACR for 1 h at 30 1C, showing conversion of the substrate
singlet at d 37.65 to a triplet at d 37.25, indicating replacement of
resulting acids were derivatised with R-2-mandelate methyl
ester. In unlabelled methyl O-decanoyl-S-mandelate, the
2-protons of the decanoyl unit are inequivalent and resonate
at different frequencies in the 1H NMR spectrum, with the
pro-S-proton signal centered at d 2.29 and the pro-R-proton at
d 2.33. These assignments were made by analogy with the
findings of Parker14 who showed that the pro-R 2-H of the
aliphatic acyl group of a range of methyl O-(fatty-acyl)-S-
mandelates always resonated at higher frequency than did the
corresponding pro-S 2-H. Assignments of the more upfield
and more downfield signals are reversed for chiral derivatization
with methyl R-mandelate. Schwab and Lin15 also used this
method to establish the absolute configuration of synthetic
2-2H1-decanoic acid. 1H NMR analysis of our derivatised
product mixture 6 showed that integral of the signal at d
2.33 (from the 2S-proton of the decanoic acid)14 was much
smaller than that of the signal at d 2.29 (from the 2R-proton)
(Fig. 2). Comparison of the integrals with those for the OMe
signal of the methyl mandelate unit showed that the 2-pro-S
proton had been replaced approximately five times more often
than the 2-pro-R proton in the decanoyl-CoA 2. It is not clear
whether this result arises from stereoselective deprotonation of
the acyl-CoA substrate or stereoselective reprotonation of the
enol/enolate intermediate.
1
2
one H with H.
changes were not observed in negative controls lacking active
enzyme.
To determine whether one or both a-protons could be
exchanged by AMACR in each catalytic cycle, incubations
were carried out with 2-[13C]-decanoyl-CoA 4 in 2H2O buffer.
This resulted in exchange of the a-1H for deuterium, as judged
by 13C NMR spectroscopy. At t = 0, the spectrum of the
incubation mixture showed only a strong singlet at d 37.65
1
arising from the 2-13C carrying only 2 ꢀ H. The 13C NMR
spectrum at early reaction time points also contained a triplet
at d 37.25, with peak-height ratio 1 : 1 : 1, indicating that
only one deuterium was directly attached to the 2-13C (Fig. 1).
Only at very high levels of conversion was a small quintet
(corresponding to 2-13C2H2) seen in this region of the
spectrum. These observations are only consistent with exchange
of one a-proton at each ‘‘visit’’ of the substrate to the active
site of the enzyme; the slow formation of 2-13C2H2-decanoyl-
CoA occurs from a second ‘‘visit’’ to the enzyme active site by
2-13C2H1-decanoyl-CoA 5. This result was initially surprising
since the rate of exchange of protons was expected to be much
faster than that of release of product, and hence exchange of
both a-protons was expected. However, the proposed catalytic
mechanism of AMACR1,8 is consistent with substitution of a
single proton. In this mechanism, one of the a-protons is
removed by either Asp-152 or Glu-237/His-122, depending
on whether the pro-R or pro-S proton is removed. The
deprotonated intermediate can then react either with the
proton on the initial base (to give back an unlabeled acyl-CoA)
or with a deuterium on the other base. Extraction of the
second a-proton can only take place if proton transfer occurs
between the two catalytic bases or between the bases and bulk
solvent before release of product.
The possibility that binding of straight-chain acyl-CoA
substrates could interfere with branched-chain fatty acid
metabolism was then investigated. Thus, the catalytic
efficiency of proton exchange by AMACR with S-2-methyl-
decanoyl-CoA 1S and decanoyl-CoA 2 was determined by
Michaelis–Menten kinetics. The known substrate S-2-methyl-
decanoyl-CoA 1S gave the following kinetic parameters:
Km = 614 mM; Vmax = 88.7 nmol minꢂ1 mgꢂ1; kcat
=
0.07 sꢂ1; kcat/Km = 114 Mꢂ1 sꢂ1 at 30 1C, consistent
with previous reports.1 Kinetic analysis of decanoyl-CoA 2
substrate with the AMACR showed non-competitive
substrate inhibition at higher substrate concentrations,
probably due to incomplete dissolution in the buffer or
formation of micelles. The following kinetic parameters were
The stereochemical course of this single substitution with
deuterium was then investigated. Decanoyl-CoA 2 was incubated
with active enzyme until ca. 60% exchange of the a-protons
with deuterium had occurred. Following quenching of the
reaction, the CoA ester mixture was hydrolysed and the
estimated: Km = 225 mM; Vmax = 10.6 nmol minꢂ1 mgꢂ1
;
ꢁc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 3348–3350 | 3349

View Article Online
increase the efficiency of catalytic conversion of longer chain
substrates (2-methyldecanoyl-CoA 1 vs. decanoyl-CoA 2)
but appears to have little effect on catalytic efficiency with
short-chain substrates.
In summary, this communication demonstrates that
straight-chain fatty acyl-CoA esters are able to bind to
recombinant human AMACR 1A and undergo deprotonation
and reprotonation events. Only one of the two a-protons is
exchanged in each AMACR catalytic cycle and the enzyme has
some degree of stereoselectivity. The origin of this stereo-
selectivity merits further investigation. These results also
reveal new details about the catalytic mechanism and substrate
binding characteristics of human AMACR 1A. Moreover,
this appears to be the only example of a racemase enzyme
catalysing proton-exchange in a non-racemisable substrate.
We thank Mr J. Crossman for purifying human recombinant
AMACR 1A. This work was funded by Cancer Research UK
and the Higher Education Commission of the Government of
Pakistan (support to FAS). Part of this work was undertaken
as an Erasmus Exchange Student Project by FP.
Fig. 3 Exchange of a-1H for 2H in straight-chain acyl-CoAs of
differing chain lengths. Error bars are ꢃ one standard deviation
(n = 2).
kcat = 0.0084 sꢂ1; kcat/Km = 37.4 Mꢂ1 ꢂ1. Therefore, the
s
S-2-methyldecanoyl-CoA 1S substrate appears to be
exchanged ca. 3-fold more efficiently than decanoyl-CoA 2,
as judged by kcat/Km. However, the non-Michaelis–Menten
behaviour of decanoyl-CoA 2 means this difference could be
significantly larger. It therefore seems likely that a-proton
exchange of straight-chain fatty acyl-CoA esters by AMACR
is not physiologically significant in the presence of branched-
chain substrates.
Notes and references
z Abbreviations used: AMACR, a-methylacyl-CoA racemase; MCR,
M. tuberculosis homologue of AMACR.
1 D. Darley, D. S. Butler, S. J. Prideaux, T. W. Thornton,
A. D. Wilson, T. J. Woodman, M. D. Threadgill and
M. D. Lloyd, Org. Biomol. Chem., 2009, 7, 543–552.
2 M. D. Lloyd, D. J. Darley, A. S. Wierzbicki and M. D. Threadgill,
FEBS J., 2008, 275, 1089–1102.
3 A. S. Wierzbicki, M. D. Lloyd, C. J. Schofield, M. D. Feher and
F. B. Gibberd, J. Neurochem., 2002, 80, 727–735.
4 J. A. Mobley, I. Leav, P. Zielie, C. Wotkowitz, J. Evans,
Y. W. Lam, B. S. L’Esperance, Z. Jiang and S. Ho, Cancer
Epidemiol. Biomarkers Prev., 2003, 12, 775–783.
5 J. F. Xu, T. Thornburg, A. R. Turner, M. Vitolins, D. Case,
J. Shadle, L. Hinson, J. L. Sun, W. N. Liu, B. L. Chang,
T. S. Adams, S. L. Zheng and F. M. Torti, Prostate, 2005, 63,
209–214.
6 M. Mukherji, C. J. Schofield, A. S. Wierzbicki, G. A. Jansen, R. J.
A. Wanders and M. D. Lloyd, Prog. Lipid Res., 2003, 42, 359–376.
7 W. Schmitz and E. Conzelmann, Eur. J. Biochem., 1997, 244,
434–440.
8 P. Bhaumik, W. Schmitz, A. Hassinen, J. K. Hiltunen,
E. Conzelmann and R. K. Wierenga, J. Mol. Biol., 2007, 367,
1145–1161.
9 J. Luo, S. Zha, W. R. Gage, T. A. Dunn, J. L. Hicks, C. J. Bennett,
C. N. Ewing, E. A. Platz, S. Ferdinandusse, R. J. Wanders,
J. M. Trent, W. B. Isaacs and A. M. De Marzo, Cancer Res.,
2002, 62, 2220–2226.
10 T. Thornburg, A. R. Turner, Y. Q. Chen, M. Vitolins, B. Chang
and J. Xu, Future Oncol., 2006, 2, 213–223.
11 S. Zha, S. Ferdinandusse, S. Denis, R. J. Wanders, C. M. Ewing,
J. Luo, A. M. De Marzo and W. B. Isaacs, Cancer Res., 2003, 63,
7365–7376.
The effect of chain-length on the rates of proton exchange of
acyl-CoA substrates by AMACR was then investigated by
incubation of acyl-CoAs of varying chain lengths with
the enzyme for 16 h at 30 1C in buffer containing 2H2O.
Acetyl-CoA was not significantly converted ({1% conver-
sion). The extents of exchange of the a-protons for deuterium
for C4–C10 straight-chain substrates were measured by 1H
NMR (Fig. 3). Increasing the length of the side-chain of
the acyl-CoA ester increased the extent of conversion.
Butanoyl-CoA was exchanged to o5%, whilst pentanoyl-
CoA, hexanoyl-CoA and heptanoyl-CoA were exchanged to
10–25%. The greatest levels of exchange (ca. 40%) were
observed for octanoyl-CoA and decanoyl-CoA 2. Conversion
of S-2-methyldecanoyl-CoA 1S was >95% under the same
conditions. This apparent dependence of the binding and
turnover of the substrates on the chain-length is consistent
with substrate binding by hydrophobic interactions with the
enzyme. Interestingly, the crystal structures of MCR8 with
acyl-CoA ligands bound show that the side-chain of the
substrate interacts with a methionine-rich hydrophobic region
at the entrance of the active site.
12 K. Takahara, H. Azuma, T. Sakamoto, S. Kiyama, T. Inamoto,
N. Ibuki, T. Nishida, H. Nomi, T. Ubai, N. Segawa and
Y. Katsuoka, Anticancer Res., 2009, 29, 2497–2505.
13 A. J. Carnell, I. Hale, S. Denis, R. J. A. Wanders, W. B. Isaacs,
B. A. Wilson and S. Ferdinandusse, J. Med. Chem., 2007, 50,
2700–2707.
14 D. Parker, J. Chem. Soc., Perkin Trans. 2, 1983, 83–88.
15 J. M. Schwab and D. C. T. Lin, J. Am. Chem. Soc., 1985, 107,
6046–6052.
Conversion of 2-methylpropanoyl-CoA (isobutyryl-CoA)
by AMACR was also investigated. Only ca. 1% exchange of
1H to 2H was observed under these assay conditions,
compared to o5% for butanoyl-CoA. These two substrates
have the same number of carbon atoms and differ only in
length of side-chain and in that the former is a branched-chain
substrate. Thus, the presence of a methyl group appears to
ꢁc
This journal is The Royal Society of Chemistry 2010
3350 | Chem. Commun., 2010, 46, 3348–3350